diff --git a/.gitattributes b/.gitattributes
index 7680d27e364f3d16e4dcfebbc72153262315be12..a6c4c904308038ff534fc2bd5f761cd699ab03f5 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -2020,3 +2020,12 @@ gemma-2-9b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-
 gemma-2-9b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.1-num-102-sd-1/checkpoint-73/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2-9b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.1-num-102-sd-1/checkpoint-80/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2-9b-it_int4_arc_challenge-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.1-num-102-sd-1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..57c0547536cf633208bd0753b6f45625ae2084ca
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:160428ee21539129b32c4f50e46b1c3c4e1b361d3d8a42304e1e889b4311be3f
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..4af5e8ac081e351427a7d80895162354afcd5c41
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b6415221e63bd9243653ca4db881c9a24e06529573526a57c54638a354ff704
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f9f37f0c6f36863a476bc5d782da1dc1227d4a3a
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c2dacc6026c12435306b5e224714e03840f94e8bcf4edaac51836364ce95581
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..4dbb86c96c25c37e0ee412e257b05f66ec5a0653
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e01ed117ebafd67aa3ac43f1079b6c9a2464882d1daaaf6d16e7f2436768cd91
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a950a8c6d57271ef31241e933d75f97ce5ae95f9
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2aa5fa807faf2a2d533b4e4cb39858015cf2b195082b515ecc420c95a95ec16
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..44546134701828ee7218a2fd8b705334b2e54d9e
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/trainer_state.json
@@ -0,0 +1,7804 @@
+{
+  "best_metric": 1.1086540222167969,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755",
+  "epoch": 7.997096188747731,
+  "eval_steps": 10,
+  "global_step": 11016,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007259528130671506,
+      "grad_norm": 0.46010470390319824,
+      "learning_rate": 0.0002,
+      "loss": 2.4936,
+      "step": 10
+    },
+    {
+      "epoch": 0.014519056261343012,
+      "grad_norm": 0.5103179216384888,
+      "learning_rate": 0.0002,
+      "loss": 1.8442,
+      "step": 20
+    },
+    {
+      "epoch": 0.021778584392014518,
+      "grad_norm": 0.6282716989517212,
+      "learning_rate": 0.0002,
+      "loss": 1.559,
+      "step": 30
+    },
+    {
+      "epoch": 0.029038112522686024,
+      "grad_norm": 1.2480497360229492,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 40
+    },
+    {
+      "epoch": 0.036297640653357534,
+      "grad_norm": 0.4114855229854584,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 50
+    },
+    {
+      "epoch": 0.043557168784029036,
+      "grad_norm": 0.49482840299606323,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 60
+    },
+    {
+      "epoch": 0.050816696914700546,
+      "grad_norm": 0.4536272883415222,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 70
+    },
+    {
+      "epoch": 0.05807622504537205,
+      "grad_norm": 0.32328274846076965,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 80
+    },
+    {
+      "epoch": 0.06533575317604355,
+      "grad_norm": 0.40990468859672546,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 90
+    },
+    {
+      "epoch": 0.07259528130671507,
+      "grad_norm": 0.37273502349853516,
+      "learning_rate": 0.0002,
+      "loss": 1.2216,
+      "step": 100
+    },
+    {
+      "epoch": 0.07985480943738657,
+      "grad_norm": 0.3903871476650238,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 110
+    },
+    {
+      "epoch": 0.08711433756805807,
+      "grad_norm": 0.3272787928581238,
+      "learning_rate": 0.0002,
+      "loss": 1.3718,
+      "step": 120
+    },
+    {
+      "epoch": 0.09437386569872959,
+      "grad_norm": 0.3622824251651764,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 130
+    },
+    {
+      "epoch": 0.10163339382940109,
+      "grad_norm": 0.3503916561603546,
+      "learning_rate": 0.0002,
+      "loss": 1.2966,
+      "step": 140
+    },
+    {
+      "epoch": 0.1088929219600726,
+      "grad_norm": 0.32787832617759705,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 150
+    },
+    {
+      "epoch": 0.1161524500907441,
+      "grad_norm": 0.5822657942771912,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 160
+    },
+    {
+      "epoch": 0.12341197822141561,
+      "grad_norm": 0.28028249740600586,
+      "learning_rate": 0.0002,
+      "loss": 1.2091,
+      "step": 170
+    },
+    {
+      "epoch": 0.1306715063520871,
+      "grad_norm": 0.33602750301361084,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 180
+    },
+    {
+      "epoch": 0.13793103448275862,
+      "grad_norm": 0.29106274247169495,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 190
+    },
+    {
+      "epoch": 0.14519056261343014,
+      "grad_norm": 0.38753363490104675,
+      "learning_rate": 0.0002,
+      "loss": 1.3261,
+      "step": 200
+    },
+    {
+      "epoch": 0.15245009074410162,
+      "grad_norm": 0.361009418964386,
+      "learning_rate": 0.0002,
+      "loss": 1.1536,
+      "step": 210
+    },
+    {
+      "epoch": 0.15970961887477314,
+      "grad_norm": 0.6743836402893066,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 220
+    },
+    {
+      "epoch": 0.16696914700544466,
+      "grad_norm": 0.3923613727092743,
+      "learning_rate": 0.0002,
+      "loss": 1.264,
+      "step": 230
+    },
+    {
+      "epoch": 0.17422867513611615,
+      "grad_norm": 0.2809699773788452,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 240
+    },
+    {
+      "epoch": 0.18148820326678766,
+      "grad_norm": 0.3631494641304016,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 250
+    },
+    {
+      "epoch": 0.18874773139745918,
+      "grad_norm": 0.24658171832561493,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 260
+    },
+    {
+      "epoch": 0.19600725952813067,
+      "grad_norm": 0.5780664682388306,
+      "learning_rate": 0.0002,
+      "loss": 1.0065,
+      "step": 270
+    },
+    {
+      "epoch": 0.20326678765880218,
+      "grad_norm": 0.3056720495223999,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 280
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 0.2857084572315216,
+      "learning_rate": 0.0002,
+      "loss": 1.1503,
+      "step": 290
+    },
+    {
+      "epoch": 0.2177858439201452,
+      "grad_norm": 0.6645345687866211,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 300
+    },
+    {
+      "epoch": 0.2250453720508167,
+      "grad_norm": 0.5966078639030457,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 310
+    },
+    {
+      "epoch": 0.2323049001814882,
+      "grad_norm": 0.40937140583992004,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 320
+    },
+    {
+      "epoch": 0.2395644283121597,
+      "grad_norm": 0.5642806887626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 330
+    },
+    {
+      "epoch": 0.24682395644283123,
+      "grad_norm": 0.2750748097896576,
+      "learning_rate": 0.0002,
+      "loss": 1.1703,
+      "step": 340
+    },
+    {
+      "epoch": 0.2540834845735027,
+      "grad_norm": 0.34350234270095825,
+      "learning_rate": 0.0002,
+      "loss": 1.2598,
+      "step": 350
+    },
+    {
+      "epoch": 0.2613430127041742,
+      "grad_norm": 0.6767239570617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1942,
+      "step": 360
+    },
+    {
+      "epoch": 0.26860254083484575,
+      "grad_norm": 0.31006959080696106,
+      "learning_rate": 0.0002,
+      "loss": 1.1436,
+      "step": 370
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "grad_norm": 0.3825474679470062,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 380
+    },
+    {
+      "epoch": 0.2831215970961887,
+      "grad_norm": 0.30241551995277405,
+      "learning_rate": 0.0002,
+      "loss": 1.1523,
+      "step": 390
+    },
+    {
+      "epoch": 0.29038112522686027,
+      "grad_norm": 0.2962397336959839,
+      "learning_rate": 0.0002,
+      "loss": 1.1298,
+      "step": 400
+    },
+    {
+      "epoch": 0.29764065335753176,
+      "grad_norm": 0.2600369155406952,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 410
+    },
+    {
+      "epoch": 0.30490018148820325,
+      "grad_norm": 0.3675060272216797,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 420
+    },
+    {
+      "epoch": 0.3121597096188748,
+      "grad_norm": 0.3429498076438904,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 430
+    },
+    {
+      "epoch": 0.3194192377495463,
+      "grad_norm": 0.34311825037002563,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 440
+    },
+    {
+      "epoch": 0.32667876588021777,
+      "grad_norm": 0.37872210144996643,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 450
+    },
+    {
+      "epoch": 0.3339382940108893,
+      "grad_norm": 0.33271121978759766,
+      "learning_rate": 0.0002,
+      "loss": 1.1365,
+      "step": 460
+    },
+    {
+      "epoch": 0.3411978221415608,
+      "grad_norm": 0.34605276584625244,
+      "learning_rate": 0.0002,
+      "loss": 1.1572,
+      "step": 470
+    },
+    {
+      "epoch": 0.3484573502722323,
+      "grad_norm": 0.41050562262535095,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 480
+    },
+    {
+      "epoch": 0.35571687840290384,
+      "grad_norm": 0.2066836953163147,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 490
+    },
+    {
+      "epoch": 0.3629764065335753,
+      "grad_norm": 0.2859014868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.2072,
+      "step": 500
+    },
+    {
+      "epoch": 0.3702359346642468,
+      "grad_norm": 0.28763777017593384,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 510
+    },
+    {
+      "epoch": 0.37749546279491836,
+      "grad_norm": 0.2730471193790436,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 520
+    },
+    {
+      "epoch": 0.38475499092558985,
+      "grad_norm": 0.3968936800956726,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 530
+    },
+    {
+      "epoch": 0.39201451905626133,
+      "grad_norm": 0.3624701201915741,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 540
+    },
+    {
+      "epoch": 0.3992740471869328,
+      "grad_norm": 0.3303608298301697,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 550
+    },
+    {
+      "epoch": 0.40653357531760437,
+      "grad_norm": 0.33507466316223145,
+      "learning_rate": 0.0002,
+      "loss": 1.203,
+      "step": 560
+    },
+    {
+      "epoch": 0.41379310344827586,
+      "grad_norm": 0.3297670781612396,
+      "learning_rate": 0.0002,
+      "loss": 1.2294,
+      "step": 570
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.32334890961647034,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 580
+    },
+    {
+      "epoch": 0.4283121597096189,
+      "grad_norm": 0.30281195044517517,
+      "learning_rate": 0.0002,
+      "loss": 1.112,
+      "step": 590
+    },
+    {
+      "epoch": 0.4355716878402904,
+      "grad_norm": 0.5900027751922607,
+      "learning_rate": 0.0002,
+      "loss": 1.1352,
+      "step": 600
+    },
+    {
+      "epoch": 0.44283121597096187,
+      "grad_norm": 0.28437477350234985,
+      "learning_rate": 0.0002,
+      "loss": 1.1575,
+      "step": 610
+    },
+    {
+      "epoch": 0.4500907441016334,
+      "grad_norm": 0.39601704478263855,
+      "learning_rate": 0.0002,
+      "loss": 1.2009,
+      "step": 620
+    },
+    {
+      "epoch": 0.4573502722323049,
+      "grad_norm": 0.41971510648727417,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 630
+    },
+    {
+      "epoch": 0.4646098003629764,
+      "grad_norm": 0.33814409375190735,
+      "learning_rate": 0.0002,
+      "loss": 1.236,
+      "step": 640
+    },
+    {
+      "epoch": 0.47186932849364793,
+      "grad_norm": 0.575718104839325,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 650
+    },
+    {
+      "epoch": 0.4791288566243194,
+      "grad_norm": 0.37927401065826416,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 660
+    },
+    {
+      "epoch": 0.4863883847549909,
+      "grad_norm": 0.3224332630634308,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 670
+    },
+    {
+      "epoch": 0.49364791288566245,
+      "grad_norm": 0.32683515548706055,
+      "learning_rate": 0.0002,
+      "loss": 1.1354,
+      "step": 680
+    },
+    {
+      "epoch": 0.5009074410163339,
+      "grad_norm": 0.4316163659095764,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 690
+    },
+    {
+      "epoch": 0.5081669691470054,
+      "grad_norm": 0.342602401971817,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 700
+    },
+    {
+      "epoch": 0.515426497277677,
+      "grad_norm": 0.2794898748397827,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 710
+    },
+    {
+      "epoch": 0.5226860254083484,
+      "grad_norm": 0.3322339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.2395,
+      "step": 720
+    },
+    {
+      "epoch": 0.52994555353902,
+      "grad_norm": 0.3088509142398834,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 730
+    },
+    {
+      "epoch": 0.5372050816696915,
+      "grad_norm": 0.24444378912448883,
+      "learning_rate": 0.0002,
+      "loss": 1.1495,
+      "step": 740
+    },
+    {
+      "epoch": 0.5444646098003629,
+      "grad_norm": 0.3483171761035919,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 750
+    },
+    {
+      "epoch": 0.5517241379310345,
+      "grad_norm": 0.3425690531730652,
+      "learning_rate": 0.0002,
+      "loss": 1.2248,
+      "step": 760
+    },
+    {
+      "epoch": 0.558983666061706,
+      "grad_norm": 0.31841927766799927,
+      "learning_rate": 0.0002,
+      "loss": 1.1127,
+      "step": 770
+    },
+    {
+      "epoch": 0.5662431941923775,
+      "grad_norm": 0.39423868060112,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 780
+    },
+    {
+      "epoch": 0.573502722323049,
+      "grad_norm": 0.30328479409217834,
+      "learning_rate": 0.0002,
+      "loss": 1.0583,
+      "step": 790
+    },
+    {
+      "epoch": 0.5807622504537205,
+      "grad_norm": 0.24475938081741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0718,
+      "step": 800
+    },
+    {
+      "epoch": 0.588021778584392,
+      "grad_norm": 0.37132805585861206,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 810
+    },
+    {
+      "epoch": 0.5952813067150635,
+      "grad_norm": 0.32195979356765747,
+      "learning_rate": 0.0002,
+      "loss": 1.1544,
+      "step": 820
+    },
+    {
+      "epoch": 0.6025408348457351,
+      "grad_norm": 0.2848738729953766,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 830
+    },
+    {
+      "epoch": 0.6098003629764065,
+      "grad_norm": 0.28015264868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.0937,
+      "step": 840
+    },
+    {
+      "epoch": 0.617059891107078,
+      "grad_norm": 0.37796008586883545,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 850
+    },
+    {
+      "epoch": 0.6243194192377496,
+      "grad_norm": 0.39311841130256653,
+      "learning_rate": 0.0002,
+      "loss": 1.1455,
+      "step": 860
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.2761685252189636,
+      "learning_rate": 0.0002,
+      "loss": 1.1066,
+      "step": 870
+    },
+    {
+      "epoch": 0.6388384754990926,
+      "grad_norm": 0.3826720118522644,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 880
+    },
+    {
+      "epoch": 0.6460980036297641,
+      "grad_norm": 0.30076679587364197,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 890
+    },
+    {
+      "epoch": 0.6533575317604355,
+      "grad_norm": 0.21997687220573425,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 900
+    },
+    {
+      "epoch": 0.6606170598911071,
+      "grad_norm": 0.32593777775764465,
+      "learning_rate": 0.0002,
+      "loss": 1.1603,
+      "step": 910
+    },
+    {
+      "epoch": 0.6678765880217786,
+      "grad_norm": 0.30347898602485657,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 920
+    },
+    {
+      "epoch": 0.6751361161524501,
+      "grad_norm": 0.44173774123191833,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 930
+    },
+    {
+      "epoch": 0.6823956442831216,
+      "grad_norm": 0.2507467269897461,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 940
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.29463833570480347,
+      "learning_rate": 0.0002,
+      "loss": 1.3001,
+      "step": 950
+    },
+    {
+      "epoch": 0.6969147005444646,
+      "grad_norm": 0.9363154172897339,
+      "learning_rate": 0.0002,
+      "loss": 1.1651,
+      "step": 960
+    },
+    {
+      "epoch": 0.7041742286751361,
+      "grad_norm": 0.3236212134361267,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 970
+    },
+    {
+      "epoch": 0.7114337568058077,
+      "grad_norm": 0.3123254179954529,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 980
+    },
+    {
+      "epoch": 0.7186932849364791,
+      "grad_norm": 0.3395805060863495,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 990
+    },
+    {
+      "epoch": 0.7259528130671506,
+      "grad_norm": 0.3240964412689209,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7332123411978222,
+      "grad_norm": 0.31902948021888733,
+      "learning_rate": 0.0002,
+      "loss": 1.0313,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7404718693284936,
+      "grad_norm": 0.4848408102989197,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7477313974591652,
+      "grad_norm": 0.33006033301353455,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7549909255898367,
+      "grad_norm": 0.2928730547428131,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7622504537205081,
+      "grad_norm": 0.3529164791107178,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7695099818511797,
+      "grad_norm": 0.2736213803291321,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7767695099818511,
+      "grad_norm": 0.7200686931610107,
+      "learning_rate": 0.0002,
+      "loss": 1.1169,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7840290381125227,
+      "grad_norm": 0.33396708965301514,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7912885662431942,
+      "grad_norm": 1.5760449171066284,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7985480943738656,
+      "grad_norm": 0.28138381242752075,
+      "learning_rate": 0.0002,
+      "loss": 1.1126,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8058076225045372,
+      "grad_norm": 0.2597472369670868,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8130671506352087,
+      "grad_norm": 0.3305445611476898,
+      "learning_rate": 0.0002,
+      "loss": 1.1177,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8203266787658802,
+      "grad_norm": 0.3934599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.0849,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8275862068965517,
+      "grad_norm": 0.3472191393375397,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8348457350272233,
+      "grad_norm": 0.2857365906238556,
+      "learning_rate": 0.0002,
+      "loss": 1.265,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.3207702934741974,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8493647912885662,
+      "grad_norm": 0.3176484704017639,
+      "learning_rate": 0.0002,
+      "loss": 1.1669,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8566243194192378,
+      "grad_norm": 0.40685558319091797,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8638838475499092,
+      "grad_norm": 0.31125199794769287,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8711433756805808,
+      "grad_norm": 0.7361181378364563,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8784029038112523,
+      "grad_norm": 0.33699527382850647,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8856624319419237,
+      "grad_norm": 0.3315220773220062,
+      "learning_rate": 0.0002,
+      "loss": 1.11,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8929219600725953,
+      "grad_norm": 0.6256054043769836,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9001814882032668,
+      "grad_norm": 0.3692137897014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1738,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9074410163339383,
+      "grad_norm": 0.3538484573364258,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9147005444646098,
+      "grad_norm": 0.27863040566444397,
+      "learning_rate": 0.0002,
+      "loss": 1.0878,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9219600725952813,
+      "grad_norm": 0.3322528302669525,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9292196007259528,
+      "grad_norm": 0.7553173303604126,
+      "learning_rate": 0.0002,
+      "loss": 1.0158,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9364791288566243,
+      "grad_norm": 0.4856191575527191,
+      "learning_rate": 0.0002,
+      "loss": 1.198,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9437386569872959,
+      "grad_norm": 0.3668074905872345,
+      "learning_rate": 0.0002,
+      "loss": 1.1678,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9509981851179673,
+      "grad_norm": 0.29851067066192627,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9582577132486388,
+      "grad_norm": 0.276664674282074,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9655172413793104,
+      "grad_norm": 0.2941018342971802,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9727767695099818,
+      "grad_norm": 0.3505859076976776,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9800362976406534,
+      "grad_norm": 0.3067687451839447,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9872958257713249,
+      "grad_norm": 0.27151066064834595,
+      "learning_rate": 0.0002,
+      "loss": 1.1511,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9945553539019963,
+      "grad_norm": 0.36370083689689636,
+      "learning_rate": 0.0002,
+      "loss": 1.1919,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9996370235934664,
+      "eval_loss": 1.1381088495254517,
+      "eval_runtime": 96.0848,
+      "eval_samples_per_second": 4.538,
+      "eval_steps_per_second": 0.572,
+      "step": 1377
+    },
+    {
+      "epoch": 1.0018148820326678,
+      "grad_norm": 0.27980583906173706,
+      "learning_rate": 0.0002,
+      "loss": 1.0082,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0090744101633393,
+      "grad_norm": 0.26713913679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0163339382940109,
+      "grad_norm": 0.3089541494846344,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0235934664246824,
+      "grad_norm": 0.4188242256641388,
+      "learning_rate": 0.0002,
+      "loss": 0.9649,
+      "step": 1410
+    },
+    {
+      "epoch": 1.030852994555354,
+      "grad_norm": 0.5246463418006897,
+      "learning_rate": 0.0002,
+      "loss": 0.9821,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0381125226860255,
+      "grad_norm": 0.2728777825832367,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0453720508166968,
+      "grad_norm": 0.38167616724967957,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.4439380168914795,
+      "learning_rate": 0.0002,
+      "loss": 1.0229,
+      "step": 1450
+    },
+    {
+      "epoch": 1.05989110707804,
+      "grad_norm": 0.30954182147979736,
+      "learning_rate": 0.0002,
+      "loss": 1.0451,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0671506352087115,
+      "grad_norm": 0.4022280275821686,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 1470
+    },
+    {
+      "epoch": 1.074410163339383,
+      "grad_norm": 0.7390811443328857,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0816696914700545,
+      "grad_norm": 0.3885486423969269,
+      "learning_rate": 0.0002,
+      "loss": 1.1378,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0889292196007259,
+      "grad_norm": 0.5275560617446899,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0961887477313974,
+      "grad_norm": 0.35112282633781433,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1510
+    },
+    {
+      "epoch": 1.103448275862069,
+      "grad_norm": 0.33714351058006287,
+      "learning_rate": 0.0002,
+      "loss": 1.0331,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1107078039927405,
+      "grad_norm": 0.31221693754196167,
+      "learning_rate": 0.0002,
+      "loss": 0.8723,
+      "step": 1530
+    },
+    {
+      "epoch": 1.117967332123412,
+      "grad_norm": 0.27549654245376587,
+      "learning_rate": 0.0002,
+      "loss": 1.0701,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1252268602540836,
+      "grad_norm": 0.8465521335601807,
+      "learning_rate": 0.0002,
+      "loss": 1.0114,
+      "step": 1550
+    },
+    {
+      "epoch": 1.132486388384755,
+      "grad_norm": 0.36125949025154114,
+      "learning_rate": 0.0002,
+      "loss": 0.9643,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1397459165154264,
+      "grad_norm": 0.37420371174812317,
+      "learning_rate": 0.0002,
+      "loss": 1.0744,
+      "step": 1570
+    },
+    {
+      "epoch": 1.147005444646098,
+      "grad_norm": 0.3294760584831238,
+      "learning_rate": 0.0002,
+      "loss": 1.165,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1542649727767695,
+      "grad_norm": 0.3881238102912903,
+      "learning_rate": 0.0002,
+      "loss": 0.9207,
+      "step": 1590
+    },
+    {
+      "epoch": 1.161524500907441,
+      "grad_norm": 0.4766491651535034,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1687840290381124,
+      "grad_norm": 0.395530104637146,
+      "learning_rate": 0.0002,
+      "loss": 1.0092,
+      "step": 1610
+    },
+    {
+      "epoch": 1.176043557168784,
+      "grad_norm": 0.3297106623649597,
+      "learning_rate": 0.0002,
+      "loss": 0.9895,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1833030852994555,
+      "grad_norm": 0.39528271555900574,
+      "learning_rate": 0.0002,
+      "loss": 1.1121,
+      "step": 1630
+    },
+    {
+      "epoch": 1.190562613430127,
+      "grad_norm": 0.3370221257209778,
+      "learning_rate": 0.0002,
+      "loss": 0.9202,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1978221415607986,
+      "grad_norm": 0.31922030448913574,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2050816696914701,
+      "grad_norm": 0.6142027378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2123411978221417,
+      "grad_norm": 0.44769710302352905,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 1670
+    },
+    {
+      "epoch": 1.219600725952813,
+      "grad_norm": 0.41233646869659424,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2268602540834845,
+      "grad_norm": 0.2928866147994995,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1690
+    },
+    {
+      "epoch": 1.234119782214156,
+      "grad_norm": 0.36913734674453735,
+      "learning_rate": 0.0002,
+      "loss": 0.9074,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2413793103448276,
+      "grad_norm": 0.5281891226768494,
+      "learning_rate": 0.0002,
+      "loss": 1.0847,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2486388384754992,
+      "grad_norm": 0.3374697268009186,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2558983666061705,
+      "grad_norm": 0.3802020847797394,
+      "learning_rate": 0.0002,
+      "loss": 0.9826,
+      "step": 1730
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.38048651814460754,
+      "learning_rate": 0.0002,
+      "loss": 1.0191,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2704174228675136,
+      "grad_norm": 0.6676169633865356,
+      "learning_rate": 0.0002,
+      "loss": 1.0142,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2776769509981851,
+      "grad_norm": 0.4075961410999298,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2849364791288567,
+      "grad_norm": 0.4374721348285675,
+      "learning_rate": 0.0002,
+      "loss": 1.0301,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2921960072595282,
+      "grad_norm": 0.4638824164867401,
+      "learning_rate": 0.0002,
+      "loss": 1.0601,
+      "step": 1780
+    },
+    {
+      "epoch": 1.2994555353901998,
+      "grad_norm": 0.38631564378738403,
+      "learning_rate": 0.0002,
+      "loss": 0.9593,
+      "step": 1790
+    },
+    {
+      "epoch": 1.306715063520871,
+      "grad_norm": 0.35873809456825256,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3139745916515426,
+      "grad_norm": 0.33602237701416016,
+      "learning_rate": 0.0002,
+      "loss": 0.9835,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3212341197822142,
+      "grad_norm": 0.46696463227272034,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3284936479128857,
+      "grad_norm": 0.368958979845047,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 1830
+    },
+    {
+      "epoch": 1.335753176043557,
+      "grad_norm": 0.3160957396030426,
+      "learning_rate": 0.0002,
+      "loss": 1.0456,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3430127041742286,
+      "grad_norm": 0.4511511027812958,
+      "learning_rate": 0.0002,
+      "loss": 1.0264,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3502722323049001,
+      "grad_norm": 0.6769845485687256,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3575317604355717,
+      "grad_norm": 0.3749309480190277,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3647912885662432,
+      "grad_norm": 0.553798496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.0223,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3720508166969148,
+      "grad_norm": 0.3538985550403595,
+      "learning_rate": 0.0002,
+      "loss": 0.9634,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.5501534938812256,
+      "learning_rate": 0.0002,
+      "loss": 1.0275,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3865698729582578,
+      "grad_norm": 0.4432051181793213,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3938294010889292,
+      "grad_norm": 0.41755786538124084,
+      "learning_rate": 0.0002,
+      "loss": 0.9082,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4010889292196007,
+      "grad_norm": 0.5626114010810852,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4083484573502723,
+      "grad_norm": 0.44038185477256775,
+      "learning_rate": 0.0002,
+      "loss": 1.0687,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4156079854809438,
+      "grad_norm": 0.3397001624107361,
+      "learning_rate": 0.0002,
+      "loss": 0.9454,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4228675136116151,
+      "grad_norm": 0.4325368106365204,
+      "learning_rate": 0.0002,
+      "loss": 0.943,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4301270417422867,
+      "grad_norm": 0.3900907039642334,
+      "learning_rate": 0.0002,
+      "loss": 1.0488,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4373865698729582,
+      "grad_norm": 0.369612455368042,
+      "learning_rate": 0.0002,
+      "loss": 0.9699,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4446460980036298,
+      "grad_norm": 0.4389338791370392,
+      "learning_rate": 0.0002,
+      "loss": 1.0609,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4519056261343013,
+      "grad_norm": 1.694450855255127,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4591651542649728,
+      "grad_norm": 0.516957700252533,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4664246823956444,
+      "grad_norm": 0.45515501499176025,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.4153139591217041,
+      "learning_rate": 0.0002,
+      "loss": 1.0188,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4809437386569873,
+      "grad_norm": 0.44353052973747253,
+      "learning_rate": 0.0002,
+      "loss": 1.1521,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4882032667876588,
+      "grad_norm": 0.570554256439209,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4954627949183303,
+      "grad_norm": 0.5742740035057068,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5027223230490017,
+      "grad_norm": 0.5890517830848694,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5099818511796732,
+      "grad_norm": 0.4162650406360626,
+      "learning_rate": 0.0002,
+      "loss": 1.0639,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5172413793103448,
+      "grad_norm": 0.7334717512130737,
+      "learning_rate": 0.0002,
+      "loss": 1.071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5245009074410163,
+      "grad_norm": 0.2963249683380127,
+      "learning_rate": 0.0002,
+      "loss": 0.8957,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5317604355716878,
+      "grad_norm": 0.30676454305648804,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5390199637023594,
+      "grad_norm": 0.35984641313552856,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 2120
+    },
+    {
+      "epoch": 1.546279491833031,
+      "grad_norm": 0.3384549617767334,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5535390199637025,
+      "grad_norm": 0.4725518226623535,
+      "learning_rate": 0.0002,
+      "loss": 0.9874,
+      "step": 2140
+    },
+    {
+      "epoch": 1.560798548094374,
+      "grad_norm": 0.3252685070037842,
+      "learning_rate": 0.0002,
+      "loss": 1.1166,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5680580762250453,
+      "grad_norm": 0.45043081045150757,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5753176043557169,
+      "grad_norm": 0.374208003282547,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5825771324863884,
+      "grad_norm": 0.5118404030799866,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5898366606170597,
+      "grad_norm": 0.482170969247818,
+      "learning_rate": 0.0002,
+      "loss": 1.0062,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5970961887477313,
+      "grad_norm": 0.5337533950805664,
+      "learning_rate": 0.0002,
+      "loss": 1.0872,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6043557168784028,
+      "grad_norm": 0.5195064544677734,
+      "learning_rate": 0.0002,
+      "loss": 1.0405,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6116152450090744,
+      "grad_norm": 0.30807098746299744,
+      "learning_rate": 0.0002,
+      "loss": 1.0454,
+      "step": 2220
+    },
+    {
+      "epoch": 1.618874773139746,
+      "grad_norm": 0.3962925672531128,
+      "learning_rate": 0.0002,
+      "loss": 1.0293,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6261343012704175,
+      "grad_norm": 0.7636962532997131,
+      "learning_rate": 0.0002,
+      "loss": 1.0137,
+      "step": 2240
+    },
+    {
+      "epoch": 1.633393829401089,
+      "grad_norm": 0.32380592823028564,
+      "learning_rate": 0.0002,
+      "loss": 0.991,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6406533575317606,
+      "grad_norm": 0.5767741799354553,
+      "learning_rate": 0.0002,
+      "loss": 0.9471,
+      "step": 2260
+    },
+    {
+      "epoch": 1.647912885662432,
+      "grad_norm": 0.39964812994003296,
+      "learning_rate": 0.0002,
+      "loss": 0.9524,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6551724137931034,
+      "grad_norm": 0.622629702091217,
+      "learning_rate": 0.0002,
+      "loss": 0.9866,
+      "step": 2280
+    },
+    {
+      "epoch": 1.662431941923775,
+      "grad_norm": 0.40202152729034424,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6696914700544465,
+      "grad_norm": 0.4467349052429199,
+      "learning_rate": 0.0002,
+      "loss": 0.9859,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6769509981851178,
+      "grad_norm": 0.5026949048042297,
+      "learning_rate": 0.0002,
+      "loss": 1.0312,
+      "step": 2310
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.43754541873931885,
+      "learning_rate": 0.0002,
+      "loss": 1.0776,
+      "step": 2320
+    },
+    {
+      "epoch": 1.691470054446461,
+      "grad_norm": 0.42869430780410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6987295825771325,
+      "grad_norm": 0.4192679524421692,
+      "learning_rate": 0.0002,
+      "loss": 1.0747,
+      "step": 2340
+    },
+    {
+      "epoch": 1.705989110707804,
+      "grad_norm": 0.3243155777454376,
+      "learning_rate": 0.0002,
+      "loss": 1.0054,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7132486388384756,
+      "grad_norm": 1.0514075756072998,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 2360
+    },
+    {
+      "epoch": 1.720508166969147,
+      "grad_norm": 0.4719122648239136,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7277676950998186,
+      "grad_norm": 0.3846144676208496,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 2380
+    },
+    {
+      "epoch": 1.73502722323049,
+      "grad_norm": 0.7266581058502197,
+      "learning_rate": 0.0002,
+      "loss": 0.9935,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7422867513611615,
+      "grad_norm": 0.6183241605758667,
+      "learning_rate": 0.0002,
+      "loss": 0.9509,
+      "step": 2400
+    },
+    {
+      "epoch": 1.749546279491833,
+      "grad_norm": 0.3658260405063629,
+      "learning_rate": 0.0002,
+      "loss": 0.9713,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7568058076225044,
+      "grad_norm": 0.6036322712898254,
+      "learning_rate": 0.0002,
+      "loss": 1.0033,
+      "step": 2420
+    },
+    {
+      "epoch": 1.764065335753176,
+      "grad_norm": 0.7872936129570007,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7713248638838475,
+      "grad_norm": 0.35946124792099,
+      "learning_rate": 0.0002,
+      "loss": 0.99,
+      "step": 2440
+    },
+    {
+      "epoch": 1.778584392014519,
+      "grad_norm": 0.3740338981151581,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2450
+    },
+    {
+      "epoch": 1.7858439201451906,
+      "grad_norm": 0.6150230169296265,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 2460
+    },
+    {
+      "epoch": 1.793103448275862,
+      "grad_norm": 0.4726075530052185,
+      "learning_rate": 0.0002,
+      "loss": 1.0373,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8003629764065336,
+      "grad_norm": 0.31292253732681274,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8076225045372052,
+      "grad_norm": 0.4463104009628296,
+      "learning_rate": 0.0002,
+      "loss": 1.0205,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8148820326678767,
+      "grad_norm": 0.7848200798034668,
+      "learning_rate": 0.0002,
+      "loss": 1.0771,
+      "step": 2500
+    },
+    {
+      "epoch": 1.822141560798548,
+      "grad_norm": 0.5562082529067993,
+      "learning_rate": 0.0002,
+      "loss": 0.9508,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8294010889292196,
+      "grad_norm": 0.39892756938934326,
+      "learning_rate": 0.0002,
+      "loss": 0.9343,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8366606170598911,
+      "grad_norm": 1.2923320531845093,
+      "learning_rate": 0.0002,
+      "loss": 1.0489,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8439201451905625,
+      "grad_norm": 0.6316490769386292,
+      "learning_rate": 0.0002,
+      "loss": 1.1203,
+      "step": 2540
+    },
+    {
+      "epoch": 1.851179673321234,
+      "grad_norm": 0.46100497245788574,
+      "learning_rate": 0.0002,
+      "loss": 0.9131,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8584392014519056,
+      "grad_norm": 0.7902987003326416,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 2560
+    },
+    {
+      "epoch": 1.865698729582577,
+      "grad_norm": 0.4596365690231323,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8729582577132486,
+      "grad_norm": 0.6592172384262085,
+      "learning_rate": 0.0002,
+      "loss": 0.9322,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8802177858439202,
+      "grad_norm": 0.5764662027359009,
+      "learning_rate": 0.0002,
+      "loss": 0.9549,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8874773139745917,
+      "grad_norm": 0.8421637415885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9485,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.5635305047035217,
+      "learning_rate": 0.0002,
+      "loss": 1.012,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9019963702359348,
+      "grad_norm": 0.46231237053871155,
+      "learning_rate": 0.0002,
+      "loss": 0.8907,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9092558983666061,
+      "grad_norm": 0.3944607079029083,
+      "learning_rate": 0.0002,
+      "loss": 0.9543,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9165154264972777,
+      "grad_norm": 0.4753907322883606,
+      "learning_rate": 0.0002,
+      "loss": 0.9964,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9237749546279492,
+      "grad_norm": 0.4151090979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.0217,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9310344827586206,
+      "grad_norm": 0.6793725490570068,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 2660
+    },
+    {
+      "epoch": 1.938294010889292,
+      "grad_norm": 0.339755117893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0401,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9455535390199636,
+      "grad_norm": 0.40789374709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.082,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9528130671506352,
+      "grad_norm": 0.3750005066394806,
+      "learning_rate": 0.0002,
+      "loss": 0.9973,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9600725952813067,
+      "grad_norm": 0.39684441685676575,
+      "learning_rate": 0.0002,
+      "loss": 0.9837,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9673321234119783,
+      "grad_norm": 0.378287672996521,
+      "learning_rate": 0.0002,
+      "loss": 1.1204,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9745916515426498,
+      "grad_norm": 0.3668482005596161,
+      "learning_rate": 0.0002,
+      "loss": 0.9284,
+      "step": 2720
+    },
+    {
+      "epoch": 1.9818511796733214,
+      "grad_norm": 0.49997565150260925,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 2730
+    },
+    {
+      "epoch": 1.989110707803993,
+      "grad_norm": 0.36852124333381653,
+      "learning_rate": 0.0002,
+      "loss": 0.9311,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9963702359346642,
+      "grad_norm": 0.5203380584716797,
+      "learning_rate": 0.0002,
+      "loss": 1.0898,
+      "step": 2750
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1086540222167969,
+      "eval_runtime": 95.6866,
+      "eval_samples_per_second": 4.557,
+      "eval_steps_per_second": 0.575,
+      "step": 2755
+    },
+    {
+      "epoch": 2.0036297640653356,
+      "grad_norm": 0.35921573638916016,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2760
+    },
+    {
+      "epoch": 2.010889292196007,
+      "grad_norm": 1.013861894607544,
+      "learning_rate": 0.0002,
+      "loss": 0.8968,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0181488203266786,
+      "grad_norm": 0.4425240159034729,
+      "learning_rate": 0.0002,
+      "loss": 0.7873,
+      "step": 2780
+    },
+    {
+      "epoch": 2.02540834845735,
+      "grad_norm": 0.60174161195755,
+      "learning_rate": 0.0002,
+      "loss": 0.8758,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0326678765880217,
+      "grad_norm": 0.47582098841667175,
+      "learning_rate": 0.0002,
+      "loss": 0.7989,
+      "step": 2800
+    },
+    {
+      "epoch": 2.0399274047186933,
+      "grad_norm": 0.6012811660766602,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 2810
+    },
+    {
+      "epoch": 2.047186932849365,
+      "grad_norm": 0.4444408118724823,
+      "learning_rate": 0.0002,
+      "loss": 0.8109,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0544464609800364,
+      "grad_norm": 0.4864003360271454,
+      "learning_rate": 0.0002,
+      "loss": 0.8621,
+      "step": 2830
+    },
+    {
+      "epoch": 2.061705989110708,
+      "grad_norm": 0.5104215741157532,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.6218489408493042,
+      "learning_rate": 0.0002,
+      "loss": 0.7789,
+      "step": 2850
+    },
+    {
+      "epoch": 2.076225045372051,
+      "grad_norm": 0.705784261226654,
+      "learning_rate": 0.0002,
+      "loss": 0.8778,
+      "step": 2860
+    },
+    {
+      "epoch": 2.0834845735027225,
+      "grad_norm": 0.48091503977775574,
+      "learning_rate": 0.0002,
+      "loss": 0.7851,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0907441016333936,
+      "grad_norm": 0.5062456727027893,
+      "learning_rate": 0.0002,
+      "loss": 0.8444,
+      "step": 2880
+    },
+    {
+      "epoch": 2.098003629764065,
+      "grad_norm": 0.4862022399902344,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 0.48264044523239136,
+      "learning_rate": 0.0002,
+      "loss": 0.8593,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1125226860254083,
+      "grad_norm": 0.43744346499443054,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2910
+    },
+    {
+      "epoch": 2.11978221415608,
+      "grad_norm": 0.5480492115020752,
+      "learning_rate": 0.0002,
+      "loss": 0.9099,
+      "step": 2920
+    },
+    {
+      "epoch": 2.1270417422867514,
+      "grad_norm": 0.5068560838699341,
+      "learning_rate": 0.0002,
+      "loss": 0.7727,
+      "step": 2930
+    },
+    {
+      "epoch": 2.134301270417423,
+      "grad_norm": 0.4650852680206299,
+      "learning_rate": 0.0002,
+      "loss": 0.8403,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1415607985480944,
+      "grad_norm": 0.4929981231689453,
+      "learning_rate": 0.0002,
+      "loss": 0.8467,
+      "step": 2950
+    },
+    {
+      "epoch": 2.148820326678766,
+      "grad_norm": 0.6537389755249023,
+      "learning_rate": 0.0002,
+      "loss": 0.8747,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1560798548094375,
+      "grad_norm": 0.8032940626144409,
+      "learning_rate": 0.0002,
+      "loss": 0.7663,
+      "step": 2970
+    },
+    {
+      "epoch": 2.163339382940109,
+      "grad_norm": 0.7131643891334534,
+      "learning_rate": 0.0002,
+      "loss": 0.7604,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1705989110707806,
+      "grad_norm": 0.6034275889396667,
+      "learning_rate": 0.0002,
+      "loss": 0.8424,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1778584392014517,
+      "grad_norm": 0.6081095933914185,
+      "learning_rate": 0.0002,
+      "loss": 0.8838,
+      "step": 3000
+    },
+    {
+      "epoch": 2.1851179673321233,
+      "grad_norm": 0.5706912875175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9013,
+      "step": 3010
+    },
+    {
+      "epoch": 2.192377495462795,
+      "grad_norm": 0.6742380261421204,
+      "learning_rate": 0.0002,
+      "loss": 0.8473,
+      "step": 3020
+    },
+    {
+      "epoch": 2.1996370235934664,
+      "grad_norm": 0.4847496449947357,
+      "learning_rate": 0.0002,
+      "loss": 0.8372,
+      "step": 3030
+    },
+    {
+      "epoch": 2.206896551724138,
+      "grad_norm": 0.5529342889785767,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2141560798548094,
+      "grad_norm": 0.6108783483505249,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3050
+    },
+    {
+      "epoch": 2.221415607985481,
+      "grad_norm": 0.8841571807861328,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2286751361161525,
+      "grad_norm": 0.4227530360221863,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 3070
+    },
+    {
+      "epoch": 2.235934664246824,
+      "grad_norm": 0.461935818195343,
+      "learning_rate": 0.0002,
+      "loss": 0.7925,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2431941923774956,
+      "grad_norm": 0.5407412648200989,
+      "learning_rate": 0.0002,
+      "loss": 0.8542,
+      "step": 3090
+    },
+    {
+      "epoch": 2.250453720508167,
+      "grad_norm": 0.5057998895645142,
+      "learning_rate": 0.0002,
+      "loss": 0.8392,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2577132486388383,
+      "grad_norm": 0.530057966709137,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 3110
+    },
+    {
+      "epoch": 2.26497277676951,
+      "grad_norm": 0.5066842436790466,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2722323049001814,
+      "grad_norm": 0.5069178342819214,
+      "learning_rate": 0.0002,
+      "loss": 0.7903,
+      "step": 3130
+    },
+    {
+      "epoch": 2.279491833030853,
+      "grad_norm": 0.6095499396324158,
+      "learning_rate": 0.0002,
+      "loss": 0.8458,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2867513611615244,
+      "grad_norm": 0.49524766206741333,
+      "learning_rate": 0.0002,
+      "loss": 0.8171,
+      "step": 3150
+    },
+    {
+      "epoch": 2.294010889292196,
+      "grad_norm": 0.5334409475326538,
+      "learning_rate": 0.0002,
+      "loss": 0.7568,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3012704174228675,
+      "grad_norm": 1.681748867034912,
+      "learning_rate": 0.0002,
+      "loss": 0.7533,
+      "step": 3170
+    },
+    {
+      "epoch": 2.308529945553539,
+      "grad_norm": 0.7225565314292908,
+      "learning_rate": 0.0002,
+      "loss": 0.8155,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3157894736842106,
+      "grad_norm": 0.5379388928413391,
+      "learning_rate": 0.0002,
+      "loss": 0.8034,
+      "step": 3190
+    },
+    {
+      "epoch": 2.323049001814882,
+      "grad_norm": 0.45770326256752014,
+      "learning_rate": 0.0002,
+      "loss": 0.8833,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3303085299455537,
+      "grad_norm": 0.6984533071517944,
+      "learning_rate": 0.0002,
+      "loss": 0.8419,
+      "step": 3210
+    },
+    {
+      "epoch": 2.337568058076225,
+      "grad_norm": 0.6725744605064392,
+      "learning_rate": 0.0002,
+      "loss": 0.7414,
+      "step": 3220
+    },
+    {
+      "epoch": 2.344827586206897,
+      "grad_norm": 1.1247471570968628,
+      "learning_rate": 0.0002,
+      "loss": 0.8657,
+      "step": 3230
+    },
+    {
+      "epoch": 2.352087114337568,
+      "grad_norm": 1.0240263938903809,
+      "learning_rate": 0.0002,
+      "loss": 0.7782,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3593466424682394,
+      "grad_norm": 0.5608096122741699,
+      "learning_rate": 0.0002,
+      "loss": 0.826,
+      "step": 3250
+    },
+    {
+      "epoch": 2.366606170598911,
+      "grad_norm": 0.8294990062713623,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3738656987295825,
+      "grad_norm": 0.6734224557876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9845,
+      "step": 3270
+    },
+    {
+      "epoch": 2.381125226860254,
+      "grad_norm": 0.6862800717353821,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3883847549909256,
+      "grad_norm": 0.5442930459976196,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 3290
+    },
+    {
+      "epoch": 2.395644283121597,
+      "grad_norm": 0.5745394229888916,
+      "learning_rate": 0.0002,
+      "loss": 0.8543,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4029038112522687,
+      "grad_norm": 0.6257799863815308,
+      "learning_rate": 0.0002,
+      "loss": 0.833,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4101633393829403,
+      "grad_norm": 0.5608420968055725,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3320
+    },
+    {
+      "epoch": 2.417422867513612,
+      "grad_norm": 0.5512017011642456,
+      "learning_rate": 0.0002,
+      "loss": 0.8745,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4246823956442833,
+      "grad_norm": 0.7452999353408813,
+      "learning_rate": 0.0002,
+      "loss": 0.8642,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4319419237749544,
+      "grad_norm": 0.4604301452636719,
+      "learning_rate": 0.0002,
+      "loss": 0.8417,
+      "step": 3350
+    },
+    {
+      "epoch": 2.439201451905626,
+      "grad_norm": 0.8225823640823364,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4464609800362975,
+      "grad_norm": 0.8604981303215027,
+      "learning_rate": 0.0002,
+      "loss": 0.8144,
+      "step": 3370
+    },
+    {
+      "epoch": 2.453720508166969,
+      "grad_norm": 0.6620925664901733,
+      "learning_rate": 0.0002,
+      "loss": 0.9276,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4609800362976406,
+      "grad_norm": 0.4750158488750458,
+      "learning_rate": 0.0002,
+      "loss": 0.8381,
+      "step": 3390
+    },
+    {
+      "epoch": 2.468239564428312,
+      "grad_norm": 0.6061418056488037,
+      "learning_rate": 0.0002,
+      "loss": 0.8249,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4754990925589837,
+      "grad_norm": 0.5977247357368469,
+      "learning_rate": 0.0002,
+      "loss": 0.8944,
+      "step": 3410
+    },
+    {
+      "epoch": 2.4827586206896552,
+      "grad_norm": 0.5004227757453918,
+      "learning_rate": 0.0002,
+      "loss": 0.8366,
+      "step": 3420
+    },
+    {
+      "epoch": 2.490018148820327,
+      "grad_norm": 0.46428972482681274,
+      "learning_rate": 0.0002,
+      "loss": 0.9133,
+      "step": 3430
+    },
+    {
+      "epoch": 2.4972776769509983,
+      "grad_norm": 0.7658395171165466,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 3440
+    },
+    {
+      "epoch": 2.50453720508167,
+      "grad_norm": 0.621819794178009,
+      "learning_rate": 0.0002,
+      "loss": 0.9066,
+      "step": 3450
+    },
+    {
+      "epoch": 2.511796733212341,
+      "grad_norm": 0.4556088149547577,
+      "learning_rate": 0.0002,
+      "loss": 0.8481,
+      "step": 3460
+    },
+    {
+      "epoch": 2.519056261343013,
+      "grad_norm": 0.6124140024185181,
+      "learning_rate": 0.0002,
+      "loss": 0.8522,
+      "step": 3470
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 0.6256231665611267,
+      "learning_rate": 0.0002,
+      "loss": 0.8158,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5335753176043556,
+      "grad_norm": 0.5464141964912415,
+      "learning_rate": 0.0002,
+      "loss": 0.9097,
+      "step": 3490
+    },
+    {
+      "epoch": 2.540834845735027,
+      "grad_norm": 0.51471346616745,
+      "learning_rate": 0.0002,
+      "loss": 0.8708,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5480943738656987,
+      "grad_norm": 0.5326165556907654,
+      "learning_rate": 0.0002,
+      "loss": 0.9042,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5553539019963702,
+      "grad_norm": 0.4750378429889679,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 3520
+    },
+    {
+      "epoch": 2.562613430127042,
+      "grad_norm": 0.5292279124259949,
+      "learning_rate": 0.0002,
+      "loss": 0.8248,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5698729582577133,
+      "grad_norm": 0.6145227551460266,
+      "learning_rate": 0.0002,
+      "loss": 0.8808,
+      "step": 3540
+    },
+    {
+      "epoch": 2.577132486388385,
+      "grad_norm": 0.8275189399719238,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5843920145190564,
+      "grad_norm": 0.5037438273429871,
+      "learning_rate": 0.0002,
+      "loss": 0.875,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5916515426497275,
+      "grad_norm": 0.5838707685470581,
+      "learning_rate": 0.0002,
+      "loss": 0.8324,
+      "step": 3570
+    },
+    {
+      "epoch": 2.5989110707803995,
+      "grad_norm": 0.5398710370063782,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6061705989110706,
+      "grad_norm": 0.6115376949310303,
+      "learning_rate": 0.0002,
+      "loss": 0.7843,
+      "step": 3590
+    },
+    {
+      "epoch": 2.613430127041742,
+      "grad_norm": 1.118809461593628,
+      "learning_rate": 0.0002,
+      "loss": 0.8142,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6206896551724137,
+      "grad_norm": 0.7811771631240845,
+      "learning_rate": 0.0002,
+      "loss": 0.8282,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6279491833030852,
+      "grad_norm": 0.5753175020217896,
+      "learning_rate": 0.0002,
+      "loss": 0.7997,
+      "step": 3620
+    },
+    {
+      "epoch": 2.635208711433757,
+      "grad_norm": 0.550829291343689,
+      "learning_rate": 0.0002,
+      "loss": 0.7705,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6424682395644283,
+      "grad_norm": 0.5360019207000732,
+      "learning_rate": 0.0002,
+      "loss": 0.8564,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6497277676951,
+      "grad_norm": 0.63050377368927,
+      "learning_rate": 0.0002,
+      "loss": 0.7648,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6569872958257714,
+      "grad_norm": 0.5833110213279724,
+      "learning_rate": 0.0002,
+      "loss": 0.9156,
+      "step": 3660
+    },
+    {
+      "epoch": 2.664246823956443,
+      "grad_norm": 0.5543047189712524,
+      "learning_rate": 0.0002,
+      "loss": 0.8033,
+      "step": 3670
+    },
+    {
+      "epoch": 2.671506352087114,
+      "grad_norm": 0.6842212080955505,
+      "learning_rate": 0.0002,
+      "loss": 0.8867,
+      "step": 3680
+    },
+    {
+      "epoch": 2.678765880217786,
+      "grad_norm": 0.6859333515167236,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 3690
+    },
+    {
+      "epoch": 2.686025408348457,
+      "grad_norm": 0.7038410902023315,
+      "learning_rate": 0.0002,
+      "loss": 0.8081,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6932849364791287,
+      "grad_norm": 0.447233647108078,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7005444646098002,
+      "grad_norm": 0.6453872919082642,
+      "learning_rate": 0.0002,
+      "loss": 0.8951,
+      "step": 3720
+    },
+    {
+      "epoch": 2.707803992740472,
+      "grad_norm": 0.8025672435760498,
+      "learning_rate": 0.0002,
+      "loss": 0.8292,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7150635208711433,
+      "grad_norm": 0.5997087955474854,
+      "learning_rate": 0.0002,
+      "loss": 0.8488,
+      "step": 3740
+    },
+    {
+      "epoch": 2.722323049001815,
+      "grad_norm": 0.6901142001152039,
+      "learning_rate": 0.0002,
+      "loss": 0.8308,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7295825771324864,
+      "grad_norm": 1.036145567893982,
+      "learning_rate": 0.0002,
+      "loss": 0.8517,
+      "step": 3760
+    },
+    {
+      "epoch": 2.736842105263158,
+      "grad_norm": 0.7207072377204895,
+      "learning_rate": 0.0002,
+      "loss": 0.8455,
+      "step": 3770
+    },
+    {
+      "epoch": 2.7441016333938295,
+      "grad_norm": 1.0452989339828491,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 3780
+    },
+    {
+      "epoch": 2.751361161524501,
+      "grad_norm": 0.5615278482437134,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.46439215540885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7658802177858437,
+      "grad_norm": 0.7134917974472046,
+      "learning_rate": 0.0002,
+      "loss": 0.7847,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7731397459165157,
+      "grad_norm": 0.5139115452766418,
+      "learning_rate": 0.0002,
+      "loss": 0.8315,
+      "step": 3820
+    },
+    {
+      "epoch": 2.780399274047187,
+      "grad_norm": 0.8595781326293945,
+      "learning_rate": 0.0002,
+      "loss": 0.8995,
+      "step": 3830
+    },
+    {
+      "epoch": 2.7876588021778583,
+      "grad_norm": 0.544614851474762,
+      "learning_rate": 0.0002,
+      "loss": 0.9079,
+      "step": 3840
+    },
+    {
+      "epoch": 2.79491833030853,
+      "grad_norm": 0.6073850393295288,
+      "learning_rate": 0.0002,
+      "loss": 0.8322,
+      "step": 3850
+    },
+    {
+      "epoch": 2.8021778584392014,
+      "grad_norm": 0.8224069476127625,
+      "learning_rate": 0.0002,
+      "loss": 0.8277,
+      "step": 3860
+    },
+    {
+      "epoch": 2.809437386569873,
+      "grad_norm": 0.5347970128059387,
+      "learning_rate": 0.0002,
+      "loss": 0.815,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8166969147005445,
+      "grad_norm": 0.6799601912498474,
+      "learning_rate": 0.0002,
+      "loss": 0.8219,
+      "step": 3880
+    },
+    {
+      "epoch": 2.823956442831216,
+      "grad_norm": 0.5219197869300842,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8312159709618876,
+      "grad_norm": 0.5710130333900452,
+      "learning_rate": 0.0002,
+      "loss": 0.8649,
+      "step": 3900
+    },
+    {
+      "epoch": 2.838475499092559,
+      "grad_norm": 0.5857213139533997,
+      "learning_rate": 0.0002,
+      "loss": 0.8741,
+      "step": 3910
+    },
+    {
+      "epoch": 2.8457350272232302,
+      "grad_norm": 0.5206644535064697,
+      "learning_rate": 0.0002,
+      "loss": 0.8154,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8529945553539022,
+      "grad_norm": 1.2902015447616577,
+      "learning_rate": 0.0002,
+      "loss": 0.9328,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8602540834845733,
+      "grad_norm": 0.5252797603607178,
+      "learning_rate": 0.0002,
+      "loss": 0.8192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.867513611615245,
+      "grad_norm": 0.5925108790397644,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8747731397459164,
+      "grad_norm": 0.9719610810279846,
+      "learning_rate": 0.0002,
+      "loss": 0.8611,
+      "step": 3960
+    },
+    {
+      "epoch": 2.882032667876588,
+      "grad_norm": 0.6834747195243835,
+      "learning_rate": 0.0002,
+      "loss": 0.8531,
+      "step": 3970
+    },
+    {
+      "epoch": 2.8892921960072595,
+      "grad_norm": 0.6952353119850159,
+      "learning_rate": 0.0002,
+      "loss": 0.8124,
+      "step": 3980
+    },
+    {
+      "epoch": 2.896551724137931,
+      "grad_norm": 0.49889910221099854,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9038112522686026,
+      "grad_norm": 0.5007026791572571,
+      "learning_rate": 0.0002,
+      "loss": 0.8197,
+      "step": 4000
+    },
+    {
+      "epoch": 2.911070780399274,
+      "grad_norm": 0.5474239587783813,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9183303085299457,
+      "grad_norm": 0.6617428660392761,
+      "learning_rate": 0.0002,
+      "loss": 0.8579,
+      "step": 4020
+    },
+    {
+      "epoch": 2.925589836660617,
+      "grad_norm": 0.6097776293754578,
+      "learning_rate": 0.0002,
+      "loss": 0.8583,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9328493647912888,
+      "grad_norm": 0.5985828638076782,
+      "learning_rate": 0.0002,
+      "loss": 0.8296,
+      "step": 4040
+    },
+    {
+      "epoch": 2.94010889292196,
+      "grad_norm": 0.769488513469696,
+      "learning_rate": 0.0002,
+      "loss": 0.9134,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 0.5167055130004883,
+      "learning_rate": 0.0002,
+      "loss": 0.8148,
+      "step": 4060
+    },
+    {
+      "epoch": 2.954627949183303,
+      "grad_norm": 0.6265496015548706,
+      "learning_rate": 0.0002,
+      "loss": 0.8665,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9618874773139745,
+      "grad_norm": 1.2644082307815552,
+      "learning_rate": 0.0002,
+      "loss": 0.9218,
+      "step": 4080
+    },
+    {
+      "epoch": 2.969147005444646,
+      "grad_norm": 0.6007561087608337,
+      "learning_rate": 0.0002,
+      "loss": 0.8026,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9764065335753176,
+      "grad_norm": 0.47984927892684937,
+      "learning_rate": 0.0002,
+      "loss": 0.8262,
+      "step": 4100
+    },
+    {
+      "epoch": 2.983666061705989,
+      "grad_norm": 1.128198504447937,
+      "learning_rate": 0.0002,
+      "loss": 0.7897,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9909255898366607,
+      "grad_norm": 0.526292085647583,
+      "learning_rate": 0.0002,
+      "loss": 0.7883,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9981851179673322,
+      "grad_norm": 0.5462674498558044,
+      "learning_rate": 0.0002,
+      "loss": 0.7801,
+      "step": 4130
+    },
+    {
+      "epoch": 2.9996370235934666,
+      "eval_loss": 1.1093357801437378,
+      "eval_runtime": 46.2498,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 4132
+    },
+    {
+      "epoch": 3.0054446460980038,
+      "grad_norm": 0.48288026452064514,
+      "learning_rate": 0.0002,
+      "loss": 0.721,
+      "step": 4140
+    },
+    {
+      "epoch": 3.0127041742286753,
+      "grad_norm": 1.0181782245635986,
+      "learning_rate": 0.0002,
+      "loss": 0.6769,
+      "step": 4150
+    },
+    {
+      "epoch": 3.019963702359347,
+      "grad_norm": 0.7718019485473633,
+      "learning_rate": 0.0002,
+      "loss": 0.7185,
+      "step": 4160
+    },
+    {
+      "epoch": 3.027223230490018,
+      "grad_norm": 0.7492219805717468,
+      "learning_rate": 0.0002,
+      "loss": 0.6552,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0344827586206895,
+      "grad_norm": 0.9363632798194885,
+      "learning_rate": 0.0002,
+      "loss": 0.6678,
+      "step": 4180
+    },
+    {
+      "epoch": 3.041742286751361,
+      "grad_norm": 0.6888533234596252,
+      "learning_rate": 0.0002,
+      "loss": 0.7187,
+      "step": 4190
+    },
+    {
+      "epoch": 3.0490018148820326,
+      "grad_norm": 0.7072834968566895,
+      "learning_rate": 0.0002,
+      "loss": 0.6469,
+      "step": 4200
+    },
+    {
+      "epoch": 3.056261343012704,
+      "grad_norm": 0.7182047963142395,
+      "learning_rate": 0.0002,
+      "loss": 0.6387,
+      "step": 4210
+    },
+    {
+      "epoch": 3.0635208711433757,
+      "grad_norm": 0.7194355130195618,
+      "learning_rate": 0.0002,
+      "loss": 0.6385,
+      "step": 4220
+    },
+    {
+      "epoch": 3.0707803992740472,
+      "grad_norm": 0.9454023838043213,
+      "learning_rate": 0.0002,
+      "loss": 0.5812,
+      "step": 4230
+    },
+    {
+      "epoch": 3.0780399274047188,
+      "grad_norm": 0.838657557964325,
+      "learning_rate": 0.0002,
+      "loss": 0.6036,
+      "step": 4240
+    },
+    {
+      "epoch": 3.0852994555353903,
+      "grad_norm": 0.740113377571106,
+      "learning_rate": 0.0002,
+      "loss": 0.646,
+      "step": 4250
+    },
+    {
+      "epoch": 3.092558983666062,
+      "grad_norm": 0.6616561412811279,
+      "learning_rate": 0.0002,
+      "loss": 0.604,
+      "step": 4260
+    },
+    {
+      "epoch": 3.0998185117967334,
+      "grad_norm": 0.8846506476402283,
+      "learning_rate": 0.0002,
+      "loss": 0.6462,
+      "step": 4270
+    },
+    {
+      "epoch": 3.107078039927405,
+      "grad_norm": 0.6322125792503357,
+      "learning_rate": 0.0002,
+      "loss": 0.6037,
+      "step": 4280
+    },
+    {
+      "epoch": 3.114337568058076,
+      "grad_norm": 0.7461467385292053,
+      "learning_rate": 0.0002,
+      "loss": 0.5953,
+      "step": 4290
+    },
+    {
+      "epoch": 3.1215970961887476,
+      "grad_norm": 0.8251287341117859,
+      "learning_rate": 0.0002,
+      "loss": 0.6099,
+      "step": 4300
+    },
+    {
+      "epoch": 3.128856624319419,
+      "grad_norm": 0.8767673373222351,
+      "learning_rate": 0.0002,
+      "loss": 0.6284,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1361161524500907,
+      "grad_norm": 0.7758759260177612,
+      "learning_rate": 0.0002,
+      "loss": 0.7535,
+      "step": 4320
+    },
+    {
+      "epoch": 3.143375680580762,
+      "grad_norm": 1.1056879758834839,
+      "learning_rate": 0.0002,
+      "loss": 0.6624,
+      "step": 4330
+    },
+    {
+      "epoch": 3.1506352087114338,
+      "grad_norm": 0.8259835243225098,
+      "learning_rate": 0.0002,
+      "loss": 0.691,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 0.6607027053833008,
+      "learning_rate": 0.0002,
+      "loss": 0.6635,
+      "step": 4350
+    },
+    {
+      "epoch": 3.165154264972777,
+      "grad_norm": 0.7983301281929016,
+      "learning_rate": 0.0002,
+      "loss": 0.5911,
+      "step": 4360
+    },
+    {
+      "epoch": 3.1724137931034484,
+      "grad_norm": 0.6725239157676697,
+      "learning_rate": 0.0002,
+      "loss": 0.6496,
+      "step": 4370
+    },
+    {
+      "epoch": 3.17967332123412,
+      "grad_norm": 0.9052095413208008,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 4380
+    },
+    {
+      "epoch": 3.1869328493647915,
+      "grad_norm": 0.8131307363510132,
+      "learning_rate": 0.0002,
+      "loss": 0.6877,
+      "step": 4390
+    },
+    {
+      "epoch": 3.1941923774954626,
+      "grad_norm": 0.6435626149177551,
+      "learning_rate": 0.0002,
+      "loss": 0.6384,
+      "step": 4400
+    },
+    {
+      "epoch": 3.201451905626134,
+      "grad_norm": 0.84367436170578,
+      "learning_rate": 0.0002,
+      "loss": 0.5819,
+      "step": 4410
+    },
+    {
+      "epoch": 3.2087114337568057,
+      "grad_norm": 1.5018867254257202,
+      "learning_rate": 0.0002,
+      "loss": 0.6104,
+      "step": 4420
+    },
+    {
+      "epoch": 3.215970961887477,
+      "grad_norm": 0.7019091844558716,
+      "learning_rate": 0.0002,
+      "loss": 0.6838,
+      "step": 4430
+    },
+    {
+      "epoch": 3.2232304900181488,
+      "grad_norm": 0.9164197444915771,
+      "learning_rate": 0.0002,
+      "loss": 0.6153,
+      "step": 4440
+    },
+    {
+      "epoch": 3.2304900181488203,
+      "grad_norm": 0.7890861630439758,
+      "learning_rate": 0.0002,
+      "loss": 0.6618,
+      "step": 4450
+    },
+    {
+      "epoch": 3.237749546279492,
+      "grad_norm": 0.6517660617828369,
+      "learning_rate": 0.0002,
+      "loss": 0.6401,
+      "step": 4460
+    },
+    {
+      "epoch": 3.2450090744101634,
+      "grad_norm": 1.10188889503479,
+      "learning_rate": 0.0002,
+      "loss": 0.6699,
+      "step": 4470
+    },
+    {
+      "epoch": 3.252268602540835,
+      "grad_norm": 0.8158330917358398,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 4480
+    },
+    {
+      "epoch": 3.2595281306715065,
+      "grad_norm": 0.7663109302520752,
+      "learning_rate": 0.0002,
+      "loss": 0.7757,
+      "step": 4490
+    },
+    {
+      "epoch": 3.266787658802178,
+      "grad_norm": 0.8473444581031799,
+      "learning_rate": 0.0002,
+      "loss": 0.6539,
+      "step": 4500
+    },
+    {
+      "epoch": 3.274047186932849,
+      "grad_norm": 0.9724768996238708,
+      "learning_rate": 0.0002,
+      "loss": 0.6511,
+      "step": 4510
+    },
+    {
+      "epoch": 3.281306715063521,
+      "grad_norm": 0.8516759276390076,
+      "learning_rate": 0.0002,
+      "loss": 0.5464,
+      "step": 4520
+    },
+    {
+      "epoch": 3.288566243194192,
+      "grad_norm": 0.7543437480926514,
+      "learning_rate": 0.0002,
+      "loss": 0.6534,
+      "step": 4530
+    },
+    {
+      "epoch": 3.2958257713248638,
+      "grad_norm": 1.0472029447555542,
+      "learning_rate": 0.0002,
+      "loss": 0.6095,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3030852994555353,
+      "grad_norm": 0.6240826845169067,
+      "learning_rate": 0.0002,
+      "loss": 0.6216,
+      "step": 4550
+    },
+    {
+      "epoch": 3.310344827586207,
+      "grad_norm": 0.9957774877548218,
+      "learning_rate": 0.0002,
+      "loss": 0.6223,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3176043557168784,
+      "grad_norm": 0.6448912620544434,
+      "learning_rate": 0.0002,
+      "loss": 0.618,
+      "step": 4570
+    },
+    {
+      "epoch": 3.32486388384755,
+      "grad_norm": 0.7519692778587341,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 4580
+    },
+    {
+      "epoch": 3.3321234119782215,
+      "grad_norm": 0.7367453575134277,
+      "learning_rate": 0.0002,
+      "loss": 0.6672,
+      "step": 4590
+    },
+    {
+      "epoch": 3.339382940108893,
+      "grad_norm": 0.8064960837364197,
+      "learning_rate": 0.0002,
+      "loss": 0.6517,
+      "step": 4600
+    },
+    {
+      "epoch": 3.3466424682395646,
+      "grad_norm": 0.7664631009101868,
+      "learning_rate": 0.0002,
+      "loss": 0.6062,
+      "step": 4610
+    },
+    {
+      "epoch": 3.353901996370236,
+      "grad_norm": 0.7803396582603455,
+      "learning_rate": 0.0002,
+      "loss": 0.6834,
+      "step": 4620
+    },
+    {
+      "epoch": 3.3611615245009077,
+      "grad_norm": 0.9141599535942078,
+      "learning_rate": 0.0002,
+      "loss": 0.6961,
+      "step": 4630
+    },
+    {
+      "epoch": 3.3684210526315788,
+      "grad_norm": 0.9719856381416321,
+      "learning_rate": 0.0002,
+      "loss": 0.6889,
+      "step": 4640
+    },
+    {
+      "epoch": 3.3756805807622503,
+      "grad_norm": 0.9223218560218811,
+      "learning_rate": 0.0002,
+      "loss": 0.6914,
+      "step": 4650
+    },
+    {
+      "epoch": 3.382940108892922,
+      "grad_norm": 0.7289277911186218,
+      "learning_rate": 0.0002,
+      "loss": 0.5981,
+      "step": 4660
+    },
+    {
+      "epoch": 3.3901996370235934,
+      "grad_norm": 1.039724349975586,
+      "learning_rate": 0.0002,
+      "loss": 0.595,
+      "step": 4670
+    },
+    {
+      "epoch": 3.397459165154265,
+      "grad_norm": 1.397438883781433,
+      "learning_rate": 0.0002,
+      "loss": 0.8121,
+      "step": 4680
+    },
+    {
+      "epoch": 3.4047186932849365,
+      "grad_norm": 1.0069999694824219,
+      "learning_rate": 0.0002,
+      "loss": 0.6334,
+      "step": 4690
+    },
+    {
+      "epoch": 3.411978221415608,
+      "grad_norm": 0.816291332244873,
+      "learning_rate": 0.0002,
+      "loss": 0.6598,
+      "step": 4700
+    },
+    {
+      "epoch": 3.4192377495462796,
+      "grad_norm": 1.2831530570983887,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 4710
+    },
+    {
+      "epoch": 3.426497277676951,
+      "grad_norm": 0.9573889970779419,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 4720
+    },
+    {
+      "epoch": 3.4337568058076227,
+      "grad_norm": 0.7685632705688477,
+      "learning_rate": 0.0002,
+      "loss": 0.7279,
+      "step": 4730
+    },
+    {
+      "epoch": 3.441016333938294,
+      "grad_norm": 0.7019195556640625,
+      "learning_rate": 0.0002,
+      "loss": 0.6104,
+      "step": 4740
+    },
+    {
+      "epoch": 3.4482758620689653,
+      "grad_norm": 0.7244833707809448,
+      "learning_rate": 0.0002,
+      "loss": 0.7606,
+      "step": 4750
+    },
+    {
+      "epoch": 3.455535390199637,
+      "grad_norm": 1.3468551635742188,
+      "learning_rate": 0.0002,
+      "loss": 0.6951,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4627949183303084,
+      "grad_norm": 0.822846531867981,
+      "learning_rate": 0.0002,
+      "loss": 0.6945,
+      "step": 4770
+    },
+    {
+      "epoch": 3.47005444646098,
+      "grad_norm": 0.7311608195304871,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 4780
+    },
+    {
+      "epoch": 3.4773139745916515,
+      "grad_norm": 0.9466770887374878,
+      "learning_rate": 0.0002,
+      "loss": 0.7019,
+      "step": 4790
+    },
+    {
+      "epoch": 3.484573502722323,
+      "grad_norm": 1.1527155637741089,
+      "learning_rate": 0.0002,
+      "loss": 0.7767,
+      "step": 4800
+    },
+    {
+      "epoch": 3.4918330308529946,
+      "grad_norm": 1.1288906335830688,
+      "learning_rate": 0.0002,
+      "loss": 0.6882,
+      "step": 4810
+    },
+    {
+      "epoch": 3.499092558983666,
+      "grad_norm": 0.9096164107322693,
+      "learning_rate": 0.0002,
+      "loss": 0.6564,
+      "step": 4820
+    },
+    {
+      "epoch": 3.5063520871143377,
+      "grad_norm": 0.7988565564155579,
+      "learning_rate": 0.0002,
+      "loss": 0.6127,
+      "step": 4830
+    },
+    {
+      "epoch": 3.513611615245009,
+      "grad_norm": 0.7183415293693542,
+      "learning_rate": 0.0002,
+      "loss": 0.7004,
+      "step": 4840
+    },
+    {
+      "epoch": 3.5208711433756807,
+      "grad_norm": 0.6614915132522583,
+      "learning_rate": 0.0002,
+      "loss": 0.74,
+      "step": 4850
+    },
+    {
+      "epoch": 3.528130671506352,
+      "grad_norm": 0.8609521985054016,
+      "learning_rate": 0.0002,
+      "loss": 0.7271,
+      "step": 4860
+    },
+    {
+      "epoch": 3.535390199637024,
+      "grad_norm": 0.86552894115448,
+      "learning_rate": 0.0002,
+      "loss": 0.6664,
+      "step": 4870
+    },
+    {
+      "epoch": 3.542649727767695,
+      "grad_norm": 0.6926496028900146,
+      "learning_rate": 0.0002,
+      "loss": 0.6432,
+      "step": 4880
+    },
+    {
+      "epoch": 3.5499092558983665,
+      "grad_norm": 0.8157467246055603,
+      "learning_rate": 0.0002,
+      "loss": 0.7117,
+      "step": 4890
+    },
+    {
+      "epoch": 3.557168784029038,
+      "grad_norm": 0.9085357189178467,
+      "learning_rate": 0.0002,
+      "loss": 0.6201,
+      "step": 4900
+    },
+    {
+      "epoch": 3.5644283121597096,
+      "grad_norm": 0.6322644948959351,
+      "learning_rate": 0.0002,
+      "loss": 0.6521,
+      "step": 4910
+    },
+    {
+      "epoch": 3.571687840290381,
+      "grad_norm": 1.263205885887146,
+      "learning_rate": 0.0002,
+      "loss": 0.6607,
+      "step": 4920
+    },
+    {
+      "epoch": 3.5789473684210527,
+      "grad_norm": 0.8901070356369019,
+      "learning_rate": 0.0002,
+      "loss": 0.6657,
+      "step": 4930
+    },
+    {
+      "epoch": 3.586206896551724,
+      "grad_norm": 0.7983952164649963,
+      "learning_rate": 0.0002,
+      "loss": 0.6434,
+      "step": 4940
+    },
+    {
+      "epoch": 3.5934664246823957,
+      "grad_norm": 0.9887813925743103,
+      "learning_rate": 0.0002,
+      "loss": 0.6861,
+      "step": 4950
+    },
+    {
+      "epoch": 3.6007259528130673,
+      "grad_norm": 0.7895187735557556,
+      "learning_rate": 0.0002,
+      "loss": 0.6502,
+      "step": 4960
+    },
+    {
+      "epoch": 3.6079854809437384,
+      "grad_norm": 0.9685819745063782,
+      "learning_rate": 0.0002,
+      "loss": 0.7111,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6152450090744104,
+      "grad_norm": 0.6576591730117798,
+      "learning_rate": 0.0002,
+      "loss": 0.6915,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6225045372050815,
+      "grad_norm": 0.856985330581665,
+      "learning_rate": 0.0002,
+      "loss": 0.6195,
+      "step": 4990
+    },
+    {
+      "epoch": 3.629764065335753,
+      "grad_norm": 0.7230252623558044,
+      "learning_rate": 0.0002,
+      "loss": 0.6318,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6370235934664246,
+      "grad_norm": 0.8260893821716309,
+      "learning_rate": 0.0002,
+      "loss": 0.742,
+      "step": 5010
+    },
+    {
+      "epoch": 3.644283121597096,
+      "grad_norm": 0.7635950446128845,
+      "learning_rate": 0.0002,
+      "loss": 0.7223,
+      "step": 5020
+    },
+    {
+      "epoch": 3.6515426497277677,
+      "grad_norm": 0.7060768604278564,
+      "learning_rate": 0.0002,
+      "loss": 0.6837,
+      "step": 5030
+    },
+    {
+      "epoch": 3.658802177858439,
+      "grad_norm": 0.8020303249359131,
+      "learning_rate": 0.0002,
+      "loss": 0.6921,
+      "step": 5040
+    },
+    {
+      "epoch": 3.6660617059891107,
+      "grad_norm": 0.8530341386795044,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 5050
+    },
+    {
+      "epoch": 3.6733212341197823,
+      "grad_norm": 0.6667101979255676,
+      "learning_rate": 0.0002,
+      "loss": 0.7222,
+      "step": 5060
+    },
+    {
+      "epoch": 3.680580762250454,
+      "grad_norm": 0.7385406494140625,
+      "learning_rate": 0.0002,
+      "loss": 0.7081,
+      "step": 5070
+    },
+    {
+      "epoch": 3.6878402903811254,
+      "grad_norm": 0.7753380537033081,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 5080
+    },
+    {
+      "epoch": 3.695099818511797,
+      "grad_norm": 0.7516207098960876,
+      "learning_rate": 0.0002,
+      "loss": 0.6491,
+      "step": 5090
+    },
+    {
+      "epoch": 3.702359346642468,
+      "grad_norm": 0.8171586394309998,
+      "learning_rate": 0.0002,
+      "loss": 0.672,
+      "step": 5100
+    },
+    {
+      "epoch": 3.70961887477314,
+      "grad_norm": 1.0796279907226562,
+      "learning_rate": 0.0002,
+      "loss": 0.7459,
+      "step": 5110
+    },
+    {
+      "epoch": 3.716878402903811,
+      "grad_norm": 0.6957688927650452,
+      "learning_rate": 0.0002,
+      "loss": 0.5948,
+      "step": 5120
+    },
+    {
+      "epoch": 3.7241379310344827,
+      "grad_norm": 0.8550161719322205,
+      "learning_rate": 0.0002,
+      "loss": 0.7515,
+      "step": 5130
+    },
+    {
+      "epoch": 3.731397459165154,
+      "grad_norm": 0.9396728277206421,
+      "learning_rate": 0.0002,
+      "loss": 0.7286,
+      "step": 5140
+    },
+    {
+      "epoch": 3.7386569872958257,
+      "grad_norm": 1.4264805316925049,
+      "learning_rate": 0.0002,
+      "loss": 0.7594,
+      "step": 5150
+    },
+    {
+      "epoch": 3.7459165154264973,
+      "grad_norm": 0.8725108504295349,
+      "learning_rate": 0.0002,
+      "loss": 0.6575,
+      "step": 5160
+    },
+    {
+      "epoch": 3.753176043557169,
+      "grad_norm": 1.0346195697784424,
+      "learning_rate": 0.0002,
+      "loss": 0.6778,
+      "step": 5170
+    },
+    {
+      "epoch": 3.7604355716878404,
+      "grad_norm": 0.5395554304122925,
+      "learning_rate": 0.0002,
+      "loss": 0.6371,
+      "step": 5180
+    },
+    {
+      "epoch": 3.767695099818512,
+      "grad_norm": 1.3153616189956665,
+      "learning_rate": 0.0002,
+      "loss": 0.7308,
+      "step": 5190
+    },
+    {
+      "epoch": 3.7749546279491835,
+      "grad_norm": 0.9879828691482544,
+      "learning_rate": 0.0002,
+      "loss": 0.78,
+      "step": 5200
+    },
+    {
+      "epoch": 3.7822141560798546,
+      "grad_norm": 0.8876672983169556,
+      "learning_rate": 0.0002,
+      "loss": 0.7068,
+      "step": 5210
+    },
+    {
+      "epoch": 3.7894736842105265,
+      "grad_norm": 0.8363267779350281,
+      "learning_rate": 0.0002,
+      "loss": 0.6283,
+      "step": 5220
+    },
+    {
+      "epoch": 3.7967332123411976,
+      "grad_norm": 0.637294590473175,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 5230
+    },
+    {
+      "epoch": 3.803992740471869,
+      "grad_norm": 1.1408970355987549,
+      "learning_rate": 0.0002,
+      "loss": 0.6685,
+      "step": 5240
+    },
+    {
+      "epoch": 3.8112522686025407,
+      "grad_norm": 1.0128360986709595,
+      "learning_rate": 0.0002,
+      "loss": 0.6761,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8185117967332123,
+      "grad_norm": 0.8061144351959229,
+      "learning_rate": 0.0002,
+      "loss": 0.6764,
+      "step": 5260
+    },
+    {
+      "epoch": 3.825771324863884,
+      "grad_norm": 0.9626626968383789,
+      "learning_rate": 0.0002,
+      "loss": 0.7254,
+      "step": 5270
+    },
+    {
+      "epoch": 3.8330308529945554,
+      "grad_norm": 0.9013627171516418,
+      "learning_rate": 0.0002,
+      "loss": 0.7367,
+      "step": 5280
+    },
+    {
+      "epoch": 3.840290381125227,
+      "grad_norm": 0.8411344289779663,
+      "learning_rate": 0.0002,
+      "loss": 0.6806,
+      "step": 5290
+    },
+    {
+      "epoch": 3.8475499092558985,
+      "grad_norm": 0.7426059246063232,
+      "learning_rate": 0.0002,
+      "loss": 0.6818,
+      "step": 5300
+    },
+    {
+      "epoch": 3.85480943738657,
+      "grad_norm": 1.003413438796997,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 5310
+    },
+    {
+      "epoch": 3.862068965517241,
+      "grad_norm": 0.7527840733528137,
+      "learning_rate": 0.0002,
+      "loss": 0.8554,
+      "step": 5320
+    },
+    {
+      "epoch": 3.869328493647913,
+      "grad_norm": 0.738610565662384,
+      "learning_rate": 0.0002,
+      "loss": 0.7521,
+      "step": 5330
+    },
+    {
+      "epoch": 3.876588021778584,
+      "grad_norm": 0.7277999520301819,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 5340
+    },
+    {
+      "epoch": 3.8838475499092557,
+      "grad_norm": 0.5951359272003174,
+      "learning_rate": 0.0002,
+      "loss": 0.7503,
+      "step": 5350
+    },
+    {
+      "epoch": 3.8911070780399273,
+      "grad_norm": 1.043884038925171,
+      "learning_rate": 0.0002,
+      "loss": 0.7447,
+      "step": 5360
+    },
+    {
+      "epoch": 3.898366606170599,
+      "grad_norm": 0.8436498045921326,
+      "learning_rate": 0.0002,
+      "loss": 0.6862,
+      "step": 5370
+    },
+    {
+      "epoch": 3.9056261343012704,
+      "grad_norm": 0.5603365302085876,
+      "learning_rate": 0.0002,
+      "loss": 0.665,
+      "step": 5380
+    },
+    {
+      "epoch": 3.912885662431942,
+      "grad_norm": 1.0128886699676514,
+      "learning_rate": 0.0002,
+      "loss": 0.7098,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9201451905626135,
+      "grad_norm": 0.7970930337905884,
+      "learning_rate": 0.0002,
+      "loss": 0.6707,
+      "step": 5400
+    },
+    {
+      "epoch": 3.927404718693285,
+      "grad_norm": 0.7699369192123413,
+      "learning_rate": 0.0002,
+      "loss": 0.637,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9346642468239565,
+      "grad_norm": 0.800561249256134,
+      "learning_rate": 0.0002,
+      "loss": 0.6742,
+      "step": 5420
+    },
+    {
+      "epoch": 3.941923774954628,
+      "grad_norm": 0.8020331859588623,
+      "learning_rate": 0.0002,
+      "loss": 0.7208,
+      "step": 5430
+    },
+    {
+      "epoch": 3.9491833030852996,
+      "grad_norm": 0.7461140155792236,
+      "learning_rate": 0.0002,
+      "loss": 0.7294,
+      "step": 5440
+    },
+    {
+      "epoch": 3.9564428312159707,
+      "grad_norm": 0.8346918821334839,
+      "learning_rate": 0.0002,
+      "loss": 0.7013,
+      "step": 5450
+    },
+    {
+      "epoch": 3.9637023593466427,
+      "grad_norm": 0.9723302125930786,
+      "learning_rate": 0.0002,
+      "loss": 0.6289,
+      "step": 5460
+    },
+    {
+      "epoch": 3.970961887477314,
+      "grad_norm": 0.6809740662574768,
+      "learning_rate": 0.0002,
+      "loss": 0.8029,
+      "step": 5470
+    },
+    {
+      "epoch": 3.9782214156079854,
+      "grad_norm": 0.7353498339653015,
+      "learning_rate": 0.0002,
+      "loss": 0.6896,
+      "step": 5480
+    },
+    {
+      "epoch": 3.985480943738657,
+      "grad_norm": 0.748009443283081,
+      "learning_rate": 0.0002,
+      "loss": 0.6722,
+      "step": 5490
+    },
+    {
+      "epoch": 3.9927404718693285,
+      "grad_norm": 1.3656195402145386,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 5500
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.8402108550071716,
+      "learning_rate": 0.0002,
+      "loss": 0.7368,
+      "step": 5510
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.17229425907135,
+      "eval_runtime": 46.2554,
+      "eval_samples_per_second": 9.426,
+      "eval_steps_per_second": 1.189,
+      "step": 5510
+    },
+    {
+      "epoch": 4.007259528130671,
+      "grad_norm": 0.8601235747337341,
+      "learning_rate": 0.0002,
+      "loss": 0.4637,
+      "step": 5520
+    },
+    {
+      "epoch": 4.014519056261343,
+      "grad_norm": 1.2635200023651123,
+      "learning_rate": 0.0002,
+      "loss": 0.4717,
+      "step": 5530
+    },
+    {
+      "epoch": 4.021778584392014,
+      "grad_norm": 1.0257477760314941,
+      "learning_rate": 0.0002,
+      "loss": 0.503,
+      "step": 5540
+    },
+    {
+      "epoch": 4.029038112522686,
+      "grad_norm": 0.9436745047569275,
+      "learning_rate": 0.0002,
+      "loss": 0.4547,
+      "step": 5550
+    },
+    {
+      "epoch": 4.036297640653357,
+      "grad_norm": 0.9443606734275818,
+      "learning_rate": 0.0002,
+      "loss": 0.459,
+      "step": 5560
+    },
+    {
+      "epoch": 4.043557168784029,
+      "grad_norm": 1.3965742588043213,
+      "learning_rate": 0.0002,
+      "loss": 0.5386,
+      "step": 5570
+    },
+    {
+      "epoch": 4.0508166969147,
+      "grad_norm": 0.8973520398139954,
+      "learning_rate": 0.0002,
+      "loss": 0.4248,
+      "step": 5580
+    },
+    {
+      "epoch": 4.058076225045372,
+      "grad_norm": 0.9998409748077393,
+      "learning_rate": 0.0002,
+      "loss": 0.4111,
+      "step": 5590
+    },
+    {
+      "epoch": 4.0653357531760435,
+      "grad_norm": 1.1213387250900269,
+      "learning_rate": 0.0002,
+      "loss": 0.4828,
+      "step": 5600
+    },
+    {
+      "epoch": 4.072595281306715,
+      "grad_norm": 0.7064558863639832,
+      "learning_rate": 0.0002,
+      "loss": 0.439,
+      "step": 5610
+    },
+    {
+      "epoch": 4.0798548094373865,
+      "grad_norm": 1.2390803098678589,
+      "learning_rate": 0.0002,
+      "loss": 0.4607,
+      "step": 5620
+    },
+    {
+      "epoch": 4.087114337568058,
+      "grad_norm": 1.123469591140747,
+      "learning_rate": 0.0002,
+      "loss": 0.5014,
+      "step": 5630
+    },
+    {
+      "epoch": 4.09437386569873,
+      "grad_norm": 1.229573369026184,
+      "learning_rate": 0.0002,
+      "loss": 0.513,
+      "step": 5640
+    },
+    {
+      "epoch": 4.101633393829401,
+      "grad_norm": 1.7182831764221191,
+      "learning_rate": 0.0002,
+      "loss": 0.5258,
+      "step": 5650
+    },
+    {
+      "epoch": 4.108892921960073,
+      "grad_norm": 0.894903302192688,
+      "learning_rate": 0.0002,
+      "loss": 0.5371,
+      "step": 5660
+    },
+    {
+      "epoch": 4.116152450090744,
+      "grad_norm": 0.8754552006721497,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 5670
+    },
+    {
+      "epoch": 4.123411978221416,
+      "grad_norm": 1.2401553392410278,
+      "learning_rate": 0.0002,
+      "loss": 0.491,
+      "step": 5680
+    },
+    {
+      "epoch": 4.130671506352087,
+      "grad_norm": 0.8631148934364319,
+      "learning_rate": 0.0002,
+      "loss": 0.4549,
+      "step": 5690
+    },
+    {
+      "epoch": 4.137931034482759,
+      "grad_norm": 1.1798022985458374,
+      "learning_rate": 0.0002,
+      "loss": 0.487,
+      "step": 5700
+    },
+    {
+      "epoch": 4.14519056261343,
+      "grad_norm": 0.8344549536705017,
+      "learning_rate": 0.0002,
+      "loss": 0.4522,
+      "step": 5710
+    },
+    {
+      "epoch": 4.152450090744102,
+      "grad_norm": 1.2342697381973267,
+      "learning_rate": 0.0002,
+      "loss": 0.4559,
+      "step": 5720
+    },
+    {
+      "epoch": 4.159709618874773,
+      "grad_norm": 1.1601094007492065,
+      "learning_rate": 0.0002,
+      "loss": 0.53,
+      "step": 5730
+    },
+    {
+      "epoch": 4.166969147005445,
+      "grad_norm": 1.2925703525543213,
+      "learning_rate": 0.0002,
+      "loss": 0.4755,
+      "step": 5740
+    },
+    {
+      "epoch": 4.174228675136116,
+      "grad_norm": 1.0870997905731201,
+      "learning_rate": 0.0002,
+      "loss": 0.4973,
+      "step": 5750
+    },
+    {
+      "epoch": 4.181488203266787,
+      "grad_norm": 0.9077792763710022,
+      "learning_rate": 0.0002,
+      "loss": 0.5184,
+      "step": 5760
+    },
+    {
+      "epoch": 4.188747731397459,
+      "grad_norm": 1.009273886680603,
+      "learning_rate": 0.0002,
+      "loss": 0.4606,
+      "step": 5770
+    },
+    {
+      "epoch": 4.19600725952813,
+      "grad_norm": 1.2465530633926392,
+      "learning_rate": 0.0002,
+      "loss": 0.5383,
+      "step": 5780
+    },
+    {
+      "epoch": 4.203266787658802,
+      "grad_norm": 1.2261253595352173,
+      "learning_rate": 0.0002,
+      "loss": 0.4938,
+      "step": 5790
+    },
+    {
+      "epoch": 4.2105263157894735,
+      "grad_norm": 1.1498041152954102,
+      "learning_rate": 0.0002,
+      "loss": 0.5255,
+      "step": 5800
+    },
+    {
+      "epoch": 4.217785843920145,
+      "grad_norm": 1.1966725587844849,
+      "learning_rate": 0.0002,
+      "loss": 0.5269,
+      "step": 5810
+    },
+    {
+      "epoch": 4.2250453720508165,
+      "grad_norm": 1.2651296854019165,
+      "learning_rate": 0.0002,
+      "loss": 0.5626,
+      "step": 5820
+    },
+    {
+      "epoch": 4.2323049001814885,
+      "grad_norm": 1.0388574600219727,
+      "learning_rate": 0.0002,
+      "loss": 0.5213,
+      "step": 5830
+    },
+    {
+      "epoch": 4.23956442831216,
+      "grad_norm": 1.3042771816253662,
+      "learning_rate": 0.0002,
+      "loss": 0.4965,
+      "step": 5840
+    },
+    {
+      "epoch": 4.246823956442832,
+      "grad_norm": 1.1127727031707764,
+      "learning_rate": 0.0002,
+      "loss": 0.5116,
+      "step": 5850
+    },
+    {
+      "epoch": 4.254083484573503,
+      "grad_norm": 0.9653958082199097,
+      "learning_rate": 0.0002,
+      "loss": 0.5197,
+      "step": 5860
+    },
+    {
+      "epoch": 4.261343012704174,
+      "grad_norm": 1.0500504970550537,
+      "learning_rate": 0.0002,
+      "loss": 0.4414,
+      "step": 5870
+    },
+    {
+      "epoch": 4.268602540834846,
+      "grad_norm": 1.1476165056228638,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 5880
+    },
+    {
+      "epoch": 4.275862068965517,
+      "grad_norm": 0.9424414038658142,
+      "learning_rate": 0.0002,
+      "loss": 0.4667,
+      "step": 5890
+    },
+    {
+      "epoch": 4.283121597096189,
+      "grad_norm": 1.3309166431427002,
+      "learning_rate": 0.0002,
+      "loss": 0.5039,
+      "step": 5900
+    },
+    {
+      "epoch": 4.29038112522686,
+      "grad_norm": 1.3025873899459839,
+      "learning_rate": 0.0002,
+      "loss": 0.5472,
+      "step": 5910
+    },
+    {
+      "epoch": 4.297640653357532,
+      "grad_norm": 1.1442325115203857,
+      "learning_rate": 0.0002,
+      "loss": 0.4644,
+      "step": 5920
+    },
+    {
+      "epoch": 4.304900181488203,
+      "grad_norm": 0.9820859432220459,
+      "learning_rate": 0.0002,
+      "loss": 0.5066,
+      "step": 5930
+    },
+    {
+      "epoch": 4.312159709618875,
+      "grad_norm": 0.9615740180015564,
+      "learning_rate": 0.0002,
+      "loss": 0.5224,
+      "step": 5940
+    },
+    {
+      "epoch": 4.319419237749546,
+      "grad_norm": 1.1627109050750732,
+      "learning_rate": 0.0002,
+      "loss": 0.5665,
+      "step": 5950
+    },
+    {
+      "epoch": 4.326678765880218,
+      "grad_norm": 0.9381322860717773,
+      "learning_rate": 0.0002,
+      "loss": 0.4852,
+      "step": 5960
+    },
+    {
+      "epoch": 4.333938294010889,
+      "grad_norm": 0.8154335618019104,
+      "learning_rate": 0.0002,
+      "loss": 0.4532,
+      "step": 5970
+    },
+    {
+      "epoch": 4.341197822141561,
+      "grad_norm": 0.877671480178833,
+      "learning_rate": 0.0002,
+      "loss": 0.5237,
+      "step": 5980
+    },
+    {
+      "epoch": 4.348457350272232,
+      "grad_norm": 1.1742031574249268,
+      "learning_rate": 0.0002,
+      "loss": 0.6113,
+      "step": 5990
+    },
+    {
+      "epoch": 4.3557168784029034,
+      "grad_norm": 1.0352917909622192,
+      "learning_rate": 0.0002,
+      "loss": 0.5704,
+      "step": 6000
+    },
+    {
+      "epoch": 4.362976406533575,
+      "grad_norm": 0.9963878989219666,
+      "learning_rate": 0.0002,
+      "loss": 0.4996,
+      "step": 6010
+    },
+    {
+      "epoch": 4.3702359346642465,
+      "grad_norm": 1.1892237663269043,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 6020
+    },
+    {
+      "epoch": 4.3774954627949185,
+      "grad_norm": 1.2516111135482788,
+      "learning_rate": 0.0002,
+      "loss": 0.5224,
+      "step": 6030
+    },
+    {
+      "epoch": 4.38475499092559,
+      "grad_norm": 1.2111951112747192,
+      "learning_rate": 0.0002,
+      "loss": 0.5952,
+      "step": 6040
+    },
+    {
+      "epoch": 4.392014519056262,
+      "grad_norm": 1.0820083618164062,
+      "learning_rate": 0.0002,
+      "loss": 0.4275,
+      "step": 6050
+    },
+    {
+      "epoch": 4.399274047186933,
+      "grad_norm": 1.033915638923645,
+      "learning_rate": 0.0002,
+      "loss": 0.5117,
+      "step": 6060
+    },
+    {
+      "epoch": 4.406533575317605,
+      "grad_norm": 1.0635870695114136,
+      "learning_rate": 0.0002,
+      "loss": 0.5431,
+      "step": 6070
+    },
+    {
+      "epoch": 4.413793103448276,
+      "grad_norm": 1.0520414113998413,
+      "learning_rate": 0.0002,
+      "loss": 0.5341,
+      "step": 6080
+    },
+    {
+      "epoch": 4.421052631578947,
+      "grad_norm": 1.0821926593780518,
+      "learning_rate": 0.0002,
+      "loss": 0.512,
+      "step": 6090
+    },
+    {
+      "epoch": 4.428312159709619,
+      "grad_norm": 1.0533246994018555,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 6100
+    },
+    {
+      "epoch": 4.43557168784029,
+      "grad_norm": 0.9231932759284973,
+      "learning_rate": 0.0002,
+      "loss": 0.4577,
+      "step": 6110
+    },
+    {
+      "epoch": 4.442831215970962,
+      "grad_norm": 0.9910260438919067,
+      "learning_rate": 0.0002,
+      "loss": 0.583,
+      "step": 6120
+    },
+    {
+      "epoch": 4.450090744101633,
+      "grad_norm": 1.061949372291565,
+      "learning_rate": 0.0002,
+      "loss": 0.4717,
+      "step": 6130
+    },
+    {
+      "epoch": 4.457350272232305,
+      "grad_norm": 1.2927039861679077,
+      "learning_rate": 0.0002,
+      "loss": 0.5893,
+      "step": 6140
+    },
+    {
+      "epoch": 4.464609800362976,
+      "grad_norm": 1.3966081142425537,
+      "learning_rate": 0.0002,
+      "loss": 0.4684,
+      "step": 6150
+    },
+    {
+      "epoch": 4.471869328493648,
+      "grad_norm": 1.3835992813110352,
+      "learning_rate": 0.0002,
+      "loss": 0.5507,
+      "step": 6160
+    },
+    {
+      "epoch": 4.479128856624319,
+      "grad_norm": 1.0892692804336548,
+      "learning_rate": 0.0002,
+      "loss": 0.5911,
+      "step": 6170
+    },
+    {
+      "epoch": 4.486388384754991,
+      "grad_norm": 1.0318800210952759,
+      "learning_rate": 0.0002,
+      "loss": 0.478,
+      "step": 6180
+    },
+    {
+      "epoch": 4.493647912885662,
+      "grad_norm": 0.8174677491188049,
+      "learning_rate": 0.0002,
+      "loss": 0.5198,
+      "step": 6190
+    },
+    {
+      "epoch": 4.500907441016334,
+      "grad_norm": 1.4157509803771973,
+      "learning_rate": 0.0002,
+      "loss": 0.5387,
+      "step": 6200
+    },
+    {
+      "epoch": 4.508166969147005,
+      "grad_norm": 1.5244114398956299,
+      "learning_rate": 0.0002,
+      "loss": 0.5868,
+      "step": 6210
+    },
+    {
+      "epoch": 4.5154264972776765,
+      "grad_norm": 0.8164850473403931,
+      "learning_rate": 0.0002,
+      "loss": 0.4642,
+      "step": 6220
+    },
+    {
+      "epoch": 4.5226860254083485,
+      "grad_norm": 1.2904746532440186,
+      "learning_rate": 0.0002,
+      "loss": 0.522,
+      "step": 6230
+    },
+    {
+      "epoch": 4.52994555353902,
+      "grad_norm": 0.7987732887268066,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 6240
+    },
+    {
+      "epoch": 4.537205081669692,
+      "grad_norm": 0.831040620803833,
+      "learning_rate": 0.0002,
+      "loss": 0.4615,
+      "step": 6250
+    },
+    {
+      "epoch": 4.544464609800363,
+      "grad_norm": 0.9545485973358154,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 6260
+    },
+    {
+      "epoch": 4.551724137931035,
+      "grad_norm": 0.9291793704032898,
+      "learning_rate": 0.0002,
+      "loss": 0.5515,
+      "step": 6270
+    },
+    {
+      "epoch": 4.558983666061706,
+      "grad_norm": 0.8977208733558655,
+      "learning_rate": 0.0002,
+      "loss": 0.4535,
+      "step": 6280
+    },
+    {
+      "epoch": 4.566243194192378,
+      "grad_norm": 1.1768537759780884,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 6290
+    },
+    {
+      "epoch": 4.573502722323049,
+      "grad_norm": 1.0688952207565308,
+      "learning_rate": 0.0002,
+      "loss": 0.5925,
+      "step": 6300
+    },
+    {
+      "epoch": 4.580762250453721,
+      "grad_norm": 0.8800966739654541,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 6310
+    },
+    {
+      "epoch": 4.588021778584392,
+      "grad_norm": 1.0911834239959717,
+      "learning_rate": 0.0002,
+      "loss": 0.6106,
+      "step": 6320
+    },
+    {
+      "epoch": 4.595281306715064,
+      "grad_norm": 1.1420872211456299,
+      "learning_rate": 0.0002,
+      "loss": 0.5109,
+      "step": 6330
+    },
+    {
+      "epoch": 4.602540834845735,
+      "grad_norm": 1.0215224027633667,
+      "learning_rate": 0.0002,
+      "loss": 0.5147,
+      "step": 6340
+    },
+    {
+      "epoch": 4.609800362976406,
+      "grad_norm": 0.9685489535331726,
+      "learning_rate": 0.0002,
+      "loss": 0.592,
+      "step": 6350
+    },
+    {
+      "epoch": 4.617059891107078,
+      "grad_norm": 1.12773597240448,
+      "learning_rate": 0.0002,
+      "loss": 0.5775,
+      "step": 6360
+    },
+    {
+      "epoch": 4.624319419237749,
+      "grad_norm": 1.0663973093032837,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 6370
+    },
+    {
+      "epoch": 4.631578947368421,
+      "grad_norm": 1.1707262992858887,
+      "learning_rate": 0.0002,
+      "loss": 0.512,
+      "step": 6380
+    },
+    {
+      "epoch": 4.638838475499092,
+      "grad_norm": 1.0672980546951294,
+      "learning_rate": 0.0002,
+      "loss": 0.5497,
+      "step": 6390
+    },
+    {
+      "epoch": 4.646098003629764,
+      "grad_norm": 1.1464333534240723,
+      "learning_rate": 0.0002,
+      "loss": 0.5699,
+      "step": 6400
+    },
+    {
+      "epoch": 4.653357531760435,
+      "grad_norm": 1.070230484008789,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 6410
+    },
+    {
+      "epoch": 4.660617059891107,
+      "grad_norm": 0.9673764109611511,
+      "learning_rate": 0.0002,
+      "loss": 0.5013,
+      "step": 6420
+    },
+    {
+      "epoch": 4.6678765880217785,
+      "grad_norm": 1.0189043283462524,
+      "learning_rate": 0.0002,
+      "loss": 0.5901,
+      "step": 6430
+    },
+    {
+      "epoch": 4.67513611615245,
+      "grad_norm": 1.185896396636963,
+      "learning_rate": 0.0002,
+      "loss": 0.5193,
+      "step": 6440
+    },
+    {
+      "epoch": 4.682395644283122,
+      "grad_norm": 1.0682812929153442,
+      "learning_rate": 0.0002,
+      "loss": 0.5318,
+      "step": 6450
+    },
+    {
+      "epoch": 4.689655172413794,
+      "grad_norm": 1.3586071729660034,
+      "learning_rate": 0.0002,
+      "loss": 0.5773,
+      "step": 6460
+    },
+    {
+      "epoch": 4.696914700544465,
+      "grad_norm": 0.6561792492866516,
+      "learning_rate": 0.0002,
+      "loss": 0.5482,
+      "step": 6470
+    },
+    {
+      "epoch": 4.704174228675136,
+      "grad_norm": 1.1394113302230835,
+      "learning_rate": 0.0002,
+      "loss": 0.5711,
+      "step": 6480
+    },
+    {
+      "epoch": 4.711433756805808,
+      "grad_norm": 0.9683151245117188,
+      "learning_rate": 0.0002,
+      "loss": 0.5325,
+      "step": 6490
+    },
+    {
+      "epoch": 4.718693284936479,
+      "grad_norm": 1.0247553586959839,
+      "learning_rate": 0.0002,
+      "loss": 0.5139,
+      "step": 6500
+    },
+    {
+      "epoch": 4.725952813067151,
+      "grad_norm": 0.8046169281005859,
+      "learning_rate": 0.0002,
+      "loss": 0.5794,
+      "step": 6510
+    },
+    {
+      "epoch": 4.733212341197822,
+      "grad_norm": 1.0710240602493286,
+      "learning_rate": 0.0002,
+      "loss": 0.5471,
+      "step": 6520
+    },
+    {
+      "epoch": 4.740471869328494,
+      "grad_norm": 0.9438924193382263,
+      "learning_rate": 0.0002,
+      "loss": 0.4805,
+      "step": 6530
+    },
+    {
+      "epoch": 4.747731397459165,
+      "grad_norm": 0.869162380695343,
+      "learning_rate": 0.0002,
+      "loss": 0.5404,
+      "step": 6540
+    },
+    {
+      "epoch": 4.754990925589837,
+      "grad_norm": 0.9776787161827087,
+      "learning_rate": 0.0002,
+      "loss": 0.6379,
+      "step": 6550
+    },
+    {
+      "epoch": 4.762250453720508,
+      "grad_norm": 1.1990505456924438,
+      "learning_rate": 0.0002,
+      "loss": 0.5288,
+      "step": 6560
+    },
+    {
+      "epoch": 4.769509981851179,
+      "grad_norm": 1.0582209825515747,
+      "learning_rate": 0.0002,
+      "loss": 0.5539,
+      "step": 6570
+    },
+    {
+      "epoch": 4.776769509981851,
+      "grad_norm": 0.9966367483139038,
+      "learning_rate": 0.0002,
+      "loss": 0.489,
+      "step": 6580
+    },
+    {
+      "epoch": 4.784029038112522,
+      "grad_norm": 0.9130612015724182,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 6590
+    },
+    {
+      "epoch": 4.791288566243194,
+      "grad_norm": 1.0950500965118408,
+      "learning_rate": 0.0002,
+      "loss": 0.5864,
+      "step": 6600
+    },
+    {
+      "epoch": 4.798548094373865,
+      "grad_norm": 1.108681321144104,
+      "learning_rate": 0.0002,
+      "loss": 0.5266,
+      "step": 6610
+    },
+    {
+      "epoch": 4.805807622504537,
+      "grad_norm": 1.1873763799667358,
+      "learning_rate": 0.0002,
+      "loss": 0.5875,
+      "step": 6620
+    },
+    {
+      "epoch": 4.8130671506352085,
+      "grad_norm": 1.305367112159729,
+      "learning_rate": 0.0002,
+      "loss": 0.5736,
+      "step": 6630
+    },
+    {
+      "epoch": 4.8203266787658805,
+      "grad_norm": 1.2801482677459717,
+      "learning_rate": 0.0002,
+      "loss": 0.5636,
+      "step": 6640
+    },
+    {
+      "epoch": 4.827586206896552,
+      "grad_norm": 1.26764976978302,
+      "learning_rate": 0.0002,
+      "loss": 0.582,
+      "step": 6650
+    },
+    {
+      "epoch": 4.834845735027224,
+      "grad_norm": 1.0018208026885986,
+      "learning_rate": 0.0002,
+      "loss": 0.5259,
+      "step": 6660
+    },
+    {
+      "epoch": 4.842105263157895,
+      "grad_norm": 1.2326326370239258,
+      "learning_rate": 0.0002,
+      "loss": 0.548,
+      "step": 6670
+    },
+    {
+      "epoch": 4.849364791288567,
+      "grad_norm": 0.9707282781600952,
+      "learning_rate": 0.0002,
+      "loss": 0.5933,
+      "step": 6680
+    },
+    {
+      "epoch": 4.856624319419238,
+      "grad_norm": 1.2772048711776733,
+      "learning_rate": 0.0002,
+      "loss": 0.5612,
+      "step": 6690
+    },
+    {
+      "epoch": 4.863883847549909,
+      "grad_norm": 2.6652262210845947,
+      "learning_rate": 0.0002,
+      "loss": 0.5346,
+      "step": 6700
+    },
+    {
+      "epoch": 4.871143375680581,
+      "grad_norm": 1.215828537940979,
+      "learning_rate": 0.0002,
+      "loss": 0.5428,
+      "step": 6710
+    },
+    {
+      "epoch": 4.878402903811252,
+      "grad_norm": 1.3704510927200317,
+      "learning_rate": 0.0002,
+      "loss": 0.6571,
+      "step": 6720
+    },
+    {
+      "epoch": 4.885662431941924,
+      "grad_norm": 0.7781757116317749,
+      "learning_rate": 0.0002,
+      "loss": 0.4963,
+      "step": 6730
+    },
+    {
+      "epoch": 4.892921960072595,
+      "grad_norm": 1.1883646249771118,
+      "learning_rate": 0.0002,
+      "loss": 0.5989,
+      "step": 6740
+    },
+    {
+      "epoch": 4.900181488203267,
+      "grad_norm": 0.9216066002845764,
+      "learning_rate": 0.0002,
+      "loss": 0.6067,
+      "step": 6750
+    },
+    {
+      "epoch": 4.907441016333938,
+      "grad_norm": 1.0558464527130127,
+      "learning_rate": 0.0002,
+      "loss": 0.5085,
+      "step": 6760
+    },
+    {
+      "epoch": 4.91470054446461,
+      "grad_norm": 1.032656192779541,
+      "learning_rate": 0.0002,
+      "loss": 0.5216,
+      "step": 6770
+    },
+    {
+      "epoch": 4.921960072595281,
+      "grad_norm": 1.1261441707611084,
+      "learning_rate": 0.0002,
+      "loss": 0.5426,
+      "step": 6780
+    },
+    {
+      "epoch": 4.929219600725952,
+      "grad_norm": 1.2178640365600586,
+      "learning_rate": 0.0002,
+      "loss": 0.5295,
+      "step": 6790
+    },
+    {
+      "epoch": 4.936479128856624,
+      "grad_norm": 1.5369361639022827,
+      "learning_rate": 0.0002,
+      "loss": 0.5476,
+      "step": 6800
+    },
+    {
+      "epoch": 4.943738656987296,
+      "grad_norm": 1.1188377141952515,
+      "learning_rate": 0.0002,
+      "loss": 0.5358,
+      "step": 6810
+    },
+    {
+      "epoch": 4.950998185117967,
+      "grad_norm": 1.2506113052368164,
+      "learning_rate": 0.0002,
+      "loss": 0.5483,
+      "step": 6820
+    },
+    {
+      "epoch": 4.9582577132486385,
+      "grad_norm": 0.8776047825813293,
+      "learning_rate": 0.0002,
+      "loss": 0.567,
+      "step": 6830
+    },
+    {
+      "epoch": 4.9655172413793105,
+      "grad_norm": 0.9700555205345154,
+      "learning_rate": 0.0002,
+      "loss": 0.5764,
+      "step": 6840
+    },
+    {
+      "epoch": 4.972776769509982,
+      "grad_norm": 1.2713534832000732,
+      "learning_rate": 0.0002,
+      "loss": 0.5396,
+      "step": 6850
+    },
+    {
+      "epoch": 4.980036297640654,
+      "grad_norm": 0.9855955243110657,
+      "learning_rate": 0.0002,
+      "loss": 0.5451,
+      "step": 6860
+    },
+    {
+      "epoch": 4.987295825771325,
+      "grad_norm": 0.8734853863716125,
+      "learning_rate": 0.0002,
+      "loss": 0.5884,
+      "step": 6870
+    },
+    {
+      "epoch": 4.994555353901997,
+      "grad_norm": 0.8065403699874878,
+      "learning_rate": 0.0002,
+      "loss": 0.5189,
+      "step": 6880
+    },
+    {
+      "epoch": 4.999637023593467,
+      "eval_loss": 1.3302682638168335,
+      "eval_runtime": 46.2496,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 6887
+    },
+    {
+      "epoch": 5.001814882032668,
+      "grad_norm": 0.5163813829421997,
+      "learning_rate": 0.0002,
+      "loss": 0.4889,
+      "step": 6890
+    },
+    {
+      "epoch": 5.00907441016334,
+      "grad_norm": 1.1496137380599976,
+      "learning_rate": 0.0002,
+      "loss": 0.3545,
+      "step": 6900
+    },
+    {
+      "epoch": 5.016333938294011,
+      "grad_norm": 1.0133885145187378,
+      "learning_rate": 0.0002,
+      "loss": 0.39,
+      "step": 6910
+    },
+    {
+      "epoch": 5.023593466424682,
+      "grad_norm": 0.9479621052742004,
+      "learning_rate": 0.0002,
+      "loss": 0.3693,
+      "step": 6920
+    },
+    {
+      "epoch": 5.030852994555354,
+      "grad_norm": 0.8587583303451538,
+      "learning_rate": 0.0002,
+      "loss": 0.4012,
+      "step": 6930
+    },
+    {
+      "epoch": 5.038112522686025,
+      "grad_norm": 1.3314697742462158,
+      "learning_rate": 0.0002,
+      "loss": 0.3428,
+      "step": 6940
+    },
+    {
+      "epoch": 5.045372050816697,
+      "grad_norm": 1.195448875427246,
+      "learning_rate": 0.0002,
+      "loss": 0.3909,
+      "step": 6950
+    },
+    {
+      "epoch": 5.052631578947368,
+      "grad_norm": 1.2482256889343262,
+      "learning_rate": 0.0002,
+      "loss": 0.3322,
+      "step": 6960
+    },
+    {
+      "epoch": 5.05989110707804,
+      "grad_norm": 1.2011528015136719,
+      "learning_rate": 0.0002,
+      "loss": 0.3893,
+      "step": 6970
+    },
+    {
+      "epoch": 5.067150635208711,
+      "grad_norm": 1.3997188806533813,
+      "learning_rate": 0.0002,
+      "loss": 0.3265,
+      "step": 6980
+    },
+    {
+      "epoch": 5.074410163339383,
+      "grad_norm": 1.2147513628005981,
+      "learning_rate": 0.0002,
+      "loss": 0.3716,
+      "step": 6990
+    },
+    {
+      "epoch": 5.081669691470054,
+      "grad_norm": 1.6030137538909912,
+      "learning_rate": 0.0002,
+      "loss": 0.4053,
+      "step": 7000
+    },
+    {
+      "epoch": 5.088929219600726,
+      "grad_norm": 0.9466970562934875,
+      "learning_rate": 0.0002,
+      "loss": 0.3665,
+      "step": 7010
+    },
+    {
+      "epoch": 5.096188747731397,
+      "grad_norm": 1.4593411684036255,
+      "learning_rate": 0.0002,
+      "loss": 0.3451,
+      "step": 7020
+    },
+    {
+      "epoch": 5.103448275862069,
+      "grad_norm": 1.2196033000946045,
+      "learning_rate": 0.0002,
+      "loss": 0.3843,
+      "step": 7030
+    },
+    {
+      "epoch": 5.1107078039927405,
+      "grad_norm": 1.1341328620910645,
+      "learning_rate": 0.0002,
+      "loss": 0.3896,
+      "step": 7040
+    },
+    {
+      "epoch": 5.117967332123412,
+      "grad_norm": 1.2248976230621338,
+      "learning_rate": 0.0002,
+      "loss": 0.3627,
+      "step": 7050
+    },
+    {
+      "epoch": 5.125226860254084,
+      "grad_norm": 1.1620593070983887,
+      "learning_rate": 0.0002,
+      "loss": 0.3784,
+      "step": 7060
+    },
+    {
+      "epoch": 5.132486388384755,
+      "grad_norm": 0.9300723671913147,
+      "learning_rate": 0.0002,
+      "loss": 0.3678,
+      "step": 7070
+    },
+    {
+      "epoch": 5.139745916515427,
+      "grad_norm": 1.2265169620513916,
+      "learning_rate": 0.0002,
+      "loss": 0.3756,
+      "step": 7080
+    },
+    {
+      "epoch": 5.147005444646098,
+      "grad_norm": 1.4430373907089233,
+      "learning_rate": 0.0002,
+      "loss": 0.3595,
+      "step": 7090
+    },
+    {
+      "epoch": 5.15426497277677,
+      "grad_norm": 1.0821576118469238,
+      "learning_rate": 0.0002,
+      "loss": 0.3788,
+      "step": 7100
+    },
+    {
+      "epoch": 5.161524500907441,
+      "grad_norm": 1.2574739456176758,
+      "learning_rate": 0.0002,
+      "loss": 0.383,
+      "step": 7110
+    },
+    {
+      "epoch": 5.168784029038113,
+      "grad_norm": 1.1806069612503052,
+      "learning_rate": 0.0002,
+      "loss": 0.3692,
+      "step": 7120
+    },
+    {
+      "epoch": 5.176043557168784,
+      "grad_norm": 0.9900956153869629,
+      "learning_rate": 0.0002,
+      "loss": 0.3978,
+      "step": 7130
+    },
+    {
+      "epoch": 5.183303085299456,
+      "grad_norm": 1.2414425611495972,
+      "learning_rate": 0.0002,
+      "loss": 0.4358,
+      "step": 7140
+    },
+    {
+      "epoch": 5.190562613430127,
+      "grad_norm": 0.8220699429512024,
+      "learning_rate": 0.0002,
+      "loss": 0.3485,
+      "step": 7150
+    },
+    {
+      "epoch": 5.197822141560798,
+      "grad_norm": 1.29408860206604,
+      "learning_rate": 0.0002,
+      "loss": 0.3517,
+      "step": 7160
+    },
+    {
+      "epoch": 5.20508166969147,
+      "grad_norm": 0.8510639071464539,
+      "learning_rate": 0.0002,
+      "loss": 0.3405,
+      "step": 7170
+    },
+    {
+      "epoch": 5.212341197822141,
+      "grad_norm": 1.3448902368545532,
+      "learning_rate": 0.0002,
+      "loss": 0.4233,
+      "step": 7180
+    },
+    {
+      "epoch": 5.219600725952813,
+      "grad_norm": 1.054451584815979,
+      "learning_rate": 0.0002,
+      "loss": 0.3808,
+      "step": 7190
+    },
+    {
+      "epoch": 5.226860254083484,
+      "grad_norm": 1.3752713203430176,
+      "learning_rate": 0.0002,
+      "loss": 0.368,
+      "step": 7200
+    },
+    {
+      "epoch": 5.234119782214156,
+      "grad_norm": 1.4848095178604126,
+      "learning_rate": 0.0002,
+      "loss": 0.3844,
+      "step": 7210
+    },
+    {
+      "epoch": 5.241379310344827,
+      "grad_norm": 1.428842544555664,
+      "learning_rate": 0.0002,
+      "loss": 0.4187,
+      "step": 7220
+    },
+    {
+      "epoch": 5.248638838475499,
+      "grad_norm": 1.1703591346740723,
+      "learning_rate": 0.0002,
+      "loss": 0.3778,
+      "step": 7230
+    },
+    {
+      "epoch": 5.2558983666061705,
+      "grad_norm": 1.2180451154708862,
+      "learning_rate": 0.0002,
+      "loss": 0.417,
+      "step": 7240
+    },
+    {
+      "epoch": 5.2631578947368425,
+      "grad_norm": 1.094045877456665,
+      "learning_rate": 0.0002,
+      "loss": 0.3656,
+      "step": 7250
+    },
+    {
+      "epoch": 5.270417422867514,
+      "grad_norm": 0.9545766115188599,
+      "learning_rate": 0.0002,
+      "loss": 0.4331,
+      "step": 7260
+    },
+    {
+      "epoch": 5.277676950998185,
+      "grad_norm": 0.8356652855873108,
+      "learning_rate": 0.0002,
+      "loss": 0.3642,
+      "step": 7270
+    },
+    {
+      "epoch": 5.284936479128857,
+      "grad_norm": 1.148160457611084,
+      "learning_rate": 0.0002,
+      "loss": 0.3576,
+      "step": 7280
+    },
+    {
+      "epoch": 5.292196007259528,
+      "grad_norm": 1.2009977102279663,
+      "learning_rate": 0.0002,
+      "loss": 0.4178,
+      "step": 7290
+    },
+    {
+      "epoch": 5.2994555353902,
+      "grad_norm": 1.3283873796463013,
+      "learning_rate": 0.0002,
+      "loss": 0.3977,
+      "step": 7300
+    },
+    {
+      "epoch": 5.306715063520871,
+      "grad_norm": 0.9850481748580933,
+      "learning_rate": 0.0002,
+      "loss": 0.3853,
+      "step": 7310
+    },
+    {
+      "epoch": 5.313974591651543,
+      "grad_norm": 1.367550015449524,
+      "learning_rate": 0.0002,
+      "loss": 0.3645,
+      "step": 7320
+    },
+    {
+      "epoch": 5.321234119782214,
+      "grad_norm": 0.8602936863899231,
+      "learning_rate": 0.0002,
+      "loss": 0.3898,
+      "step": 7330
+    },
+    {
+      "epoch": 5.328493647912886,
+      "grad_norm": 1.1130679845809937,
+      "learning_rate": 0.0002,
+      "loss": 0.4173,
+      "step": 7340
+    },
+    {
+      "epoch": 5.335753176043557,
+      "grad_norm": 1.3002253770828247,
+      "learning_rate": 0.0002,
+      "loss": 0.3642,
+      "step": 7350
+    },
+    {
+      "epoch": 5.343012704174229,
+      "grad_norm": 1.6235289573669434,
+      "learning_rate": 0.0002,
+      "loss": 0.4138,
+      "step": 7360
+    },
+    {
+      "epoch": 5.3502722323049,
+      "grad_norm": 1.156379222869873,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 7370
+    },
+    {
+      "epoch": 5.357531760435572,
+      "grad_norm": 1.0569308996200562,
+      "learning_rate": 0.0002,
+      "loss": 0.3222,
+      "step": 7380
+    },
+    {
+      "epoch": 5.364791288566243,
+      "grad_norm": 1.6674021482467651,
+      "learning_rate": 0.0002,
+      "loss": 0.3573,
+      "step": 7390
+    },
+    {
+      "epoch": 5.372050816696914,
+      "grad_norm": 1.2962018251419067,
+      "learning_rate": 0.0002,
+      "loss": 0.4325,
+      "step": 7400
+    },
+    {
+      "epoch": 5.379310344827586,
+      "grad_norm": 1.1904195547103882,
+      "learning_rate": 0.0002,
+      "loss": 0.3809,
+      "step": 7410
+    },
+    {
+      "epoch": 5.386569872958257,
+      "grad_norm": 1.316245675086975,
+      "learning_rate": 0.0002,
+      "loss": 0.3728,
+      "step": 7420
+    },
+    {
+      "epoch": 5.393829401088929,
+      "grad_norm": 1.127570390701294,
+      "learning_rate": 0.0002,
+      "loss": 0.4096,
+      "step": 7430
+    },
+    {
+      "epoch": 5.4010889292196005,
+      "grad_norm": 1.3895777463912964,
+      "learning_rate": 0.0002,
+      "loss": 0.3933,
+      "step": 7440
+    },
+    {
+      "epoch": 5.4083484573502725,
+      "grad_norm": 1.626830816268921,
+      "learning_rate": 0.0002,
+      "loss": 0.4085,
+      "step": 7450
+    },
+    {
+      "epoch": 5.415607985480944,
+      "grad_norm": 1.3703926801681519,
+      "learning_rate": 0.0002,
+      "loss": 0.4186,
+      "step": 7460
+    },
+    {
+      "epoch": 5.422867513611616,
+      "grad_norm": 1.3854840993881226,
+      "learning_rate": 0.0002,
+      "loss": 0.3517,
+      "step": 7470
+    },
+    {
+      "epoch": 5.430127041742287,
+      "grad_norm": 1.107065200805664,
+      "learning_rate": 0.0002,
+      "loss": 0.3714,
+      "step": 7480
+    },
+    {
+      "epoch": 5.437386569872959,
+      "grad_norm": 0.7843456268310547,
+      "learning_rate": 0.0002,
+      "loss": 0.3855,
+      "step": 7490
+    },
+    {
+      "epoch": 5.44464609800363,
+      "grad_norm": 1.6692372560501099,
+      "learning_rate": 0.0002,
+      "loss": 0.4159,
+      "step": 7500
+    },
+    {
+      "epoch": 5.451905626134302,
+      "grad_norm": 1.2583858966827393,
+      "learning_rate": 0.0002,
+      "loss": 0.4185,
+      "step": 7510
+    },
+    {
+      "epoch": 5.459165154264973,
+      "grad_norm": 1.6827000379562378,
+      "learning_rate": 0.0002,
+      "loss": 0.4401,
+      "step": 7520
+    },
+    {
+      "epoch": 5.466424682395644,
+      "grad_norm": 1.6680560111999512,
+      "learning_rate": 0.0002,
+      "loss": 0.397,
+      "step": 7530
+    },
+    {
+      "epoch": 5.473684210526316,
+      "grad_norm": 1.3696072101593018,
+      "learning_rate": 0.0002,
+      "loss": 0.4193,
+      "step": 7540
+    },
+    {
+      "epoch": 5.480943738656987,
+      "grad_norm": 1.4523496627807617,
+      "learning_rate": 0.0002,
+      "loss": 0.4244,
+      "step": 7550
+    },
+    {
+      "epoch": 5.488203266787659,
+      "grad_norm": 1.3432692289352417,
+      "learning_rate": 0.0002,
+      "loss": 0.3609,
+      "step": 7560
+    },
+    {
+      "epoch": 5.49546279491833,
+      "grad_norm": 1.363818645477295,
+      "learning_rate": 0.0002,
+      "loss": 0.3675,
+      "step": 7570
+    },
+    {
+      "epoch": 5.502722323049002,
+      "grad_norm": 1.0176721811294556,
+      "learning_rate": 0.0002,
+      "loss": 0.3726,
+      "step": 7580
+    },
+    {
+      "epoch": 5.509981851179673,
+      "grad_norm": 1.1625547409057617,
+      "learning_rate": 0.0002,
+      "loss": 0.3751,
+      "step": 7590
+    },
+    {
+      "epoch": 5.517241379310345,
+      "grad_norm": 1.2480388879776,
+      "learning_rate": 0.0002,
+      "loss": 0.433,
+      "step": 7600
+    },
+    {
+      "epoch": 5.524500907441016,
+      "grad_norm": 1.341509222984314,
+      "learning_rate": 0.0002,
+      "loss": 0.4511,
+      "step": 7610
+    },
+    {
+      "epoch": 5.531760435571687,
+      "grad_norm": 1.7048436403274536,
+      "learning_rate": 0.0002,
+      "loss": 0.4642,
+      "step": 7620
+    },
+    {
+      "epoch": 5.539019963702359,
+      "grad_norm": 1.1435480117797852,
+      "learning_rate": 0.0002,
+      "loss": 0.4509,
+      "step": 7630
+    },
+    {
+      "epoch": 5.5462794918330305,
+      "grad_norm": 1.2381842136383057,
+      "learning_rate": 0.0002,
+      "loss": 0.4528,
+      "step": 7640
+    },
+    {
+      "epoch": 5.5535390199637025,
+      "grad_norm": 1.50786292552948,
+      "learning_rate": 0.0002,
+      "loss": 0.4496,
+      "step": 7650
+    },
+    {
+      "epoch": 5.560798548094374,
+      "grad_norm": 1.2263519763946533,
+      "learning_rate": 0.0002,
+      "loss": 0.4242,
+      "step": 7660
+    },
+    {
+      "epoch": 5.568058076225046,
+      "grad_norm": 1.2864696979522705,
+      "learning_rate": 0.0002,
+      "loss": 0.418,
+      "step": 7670
+    },
+    {
+      "epoch": 5.575317604355717,
+      "grad_norm": 1.4443191289901733,
+      "learning_rate": 0.0002,
+      "loss": 0.3832,
+      "step": 7680
+    },
+    {
+      "epoch": 5.582577132486389,
+      "grad_norm": 1.3360971212387085,
+      "learning_rate": 0.0002,
+      "loss": 0.3964,
+      "step": 7690
+    },
+    {
+      "epoch": 5.58983666061706,
+      "grad_norm": 1.391828179359436,
+      "learning_rate": 0.0002,
+      "loss": 0.4639,
+      "step": 7700
+    },
+    {
+      "epoch": 5.597096188747732,
+      "grad_norm": 1.3699384927749634,
+      "learning_rate": 0.0002,
+      "loss": 0.4722,
+      "step": 7710
+    },
+    {
+      "epoch": 5.604355716878403,
+      "grad_norm": 1.3778468370437622,
+      "learning_rate": 0.0002,
+      "loss": 0.4302,
+      "step": 7720
+    },
+    {
+      "epoch": 5.611615245009075,
+      "grad_norm": 1.1009501218795776,
+      "learning_rate": 0.0002,
+      "loss": 0.4179,
+      "step": 7730
+    },
+    {
+      "epoch": 5.618874773139746,
+      "grad_norm": 1.0410021543502808,
+      "learning_rate": 0.0002,
+      "loss": 0.4104,
+      "step": 7740
+    },
+    {
+      "epoch": 5.626134301270417,
+      "grad_norm": 1.1012226343154907,
+      "learning_rate": 0.0002,
+      "loss": 0.4489,
+      "step": 7750
+    },
+    {
+      "epoch": 5.633393829401089,
+      "grad_norm": 1.3246384859085083,
+      "learning_rate": 0.0002,
+      "loss": 0.4544,
+      "step": 7760
+    },
+    {
+      "epoch": 5.64065335753176,
+      "grad_norm": 1.4301716089248657,
+      "learning_rate": 0.0002,
+      "loss": 0.4381,
+      "step": 7770
+    },
+    {
+      "epoch": 5.647912885662432,
+      "grad_norm": 1.1368978023529053,
+      "learning_rate": 0.0002,
+      "loss": 0.4297,
+      "step": 7780
+    },
+    {
+      "epoch": 5.655172413793103,
+      "grad_norm": 1.3493064641952515,
+      "learning_rate": 0.0002,
+      "loss": 0.4063,
+      "step": 7790
+    },
+    {
+      "epoch": 5.662431941923775,
+      "grad_norm": 1.3328721523284912,
+      "learning_rate": 0.0002,
+      "loss": 0.4562,
+      "step": 7800
+    },
+    {
+      "epoch": 5.669691470054446,
+      "grad_norm": 1.3235671520233154,
+      "learning_rate": 0.0002,
+      "loss": 0.4075,
+      "step": 7810
+    },
+    {
+      "epoch": 5.676950998185118,
+      "grad_norm": 1.1961841583251953,
+      "learning_rate": 0.0002,
+      "loss": 0.4589,
+      "step": 7820
+    },
+    {
+      "epoch": 5.684210526315789,
+      "grad_norm": 1.4189636707305908,
+      "learning_rate": 0.0002,
+      "loss": 0.4503,
+      "step": 7830
+    },
+    {
+      "epoch": 5.691470054446461,
+      "grad_norm": 1.3551312685012817,
+      "learning_rate": 0.0002,
+      "loss": 0.4452,
+      "step": 7840
+    },
+    {
+      "epoch": 5.6987295825771325,
+      "grad_norm": 1.449987769126892,
+      "learning_rate": 0.0002,
+      "loss": 0.4268,
+      "step": 7850
+    },
+    {
+      "epoch": 5.7059891107078045,
+      "grad_norm": 1.1225156784057617,
+      "learning_rate": 0.0002,
+      "loss": 0.4141,
+      "step": 7860
+    },
+    {
+      "epoch": 5.713248638838476,
+      "grad_norm": 1.4734594821929932,
+      "learning_rate": 0.0002,
+      "loss": 0.41,
+      "step": 7870
+    },
+    {
+      "epoch": 5.720508166969147,
+      "grad_norm": 1.3793359994888306,
+      "learning_rate": 0.0002,
+      "loss": 0.4013,
+      "step": 7880
+    },
+    {
+      "epoch": 5.727767695099819,
+      "grad_norm": 1.2431834936141968,
+      "learning_rate": 0.0002,
+      "loss": 0.4065,
+      "step": 7890
+    },
+    {
+      "epoch": 5.73502722323049,
+      "grad_norm": 1.1158313751220703,
+      "learning_rate": 0.0002,
+      "loss": 0.4595,
+      "step": 7900
+    },
+    {
+      "epoch": 5.742286751361162,
+      "grad_norm": 1.212248682975769,
+      "learning_rate": 0.0002,
+      "loss": 0.4342,
+      "step": 7910
+    },
+    {
+      "epoch": 5.749546279491833,
+      "grad_norm": 1.5259995460510254,
+      "learning_rate": 0.0002,
+      "loss": 0.4611,
+      "step": 7920
+    },
+    {
+      "epoch": 5.756805807622505,
+      "grad_norm": 1.3909121751785278,
+      "learning_rate": 0.0002,
+      "loss": 0.4483,
+      "step": 7930
+    },
+    {
+      "epoch": 5.764065335753176,
+      "grad_norm": 1.2511249780654907,
+      "learning_rate": 0.0002,
+      "loss": 0.4325,
+      "step": 7940
+    },
+    {
+      "epoch": 5.771324863883848,
+      "grad_norm": 1.2511906623840332,
+      "learning_rate": 0.0002,
+      "loss": 0.4048,
+      "step": 7950
+    },
+    {
+      "epoch": 5.778584392014519,
+      "grad_norm": 1.1489921808242798,
+      "learning_rate": 0.0002,
+      "loss": 0.3715,
+      "step": 7960
+    },
+    {
+      "epoch": 5.78584392014519,
+      "grad_norm": 1.028943419456482,
+      "learning_rate": 0.0002,
+      "loss": 0.4196,
+      "step": 7970
+    },
+    {
+      "epoch": 5.793103448275862,
+      "grad_norm": 1.0820423364639282,
+      "learning_rate": 0.0002,
+      "loss": 0.4334,
+      "step": 7980
+    },
+    {
+      "epoch": 5.800362976406533,
+      "grad_norm": 1.296520471572876,
+      "learning_rate": 0.0002,
+      "loss": 0.3917,
+      "step": 7990
+    },
+    {
+      "epoch": 5.807622504537205,
+      "grad_norm": 1.3597749471664429,
+      "learning_rate": 0.0002,
+      "loss": 0.4509,
+      "step": 8000
+    },
+    {
+      "epoch": 5.814882032667876,
+      "grad_norm": 0.8741790652275085,
+      "learning_rate": 0.0002,
+      "loss": 0.4535,
+      "step": 8010
+    },
+    {
+      "epoch": 5.822141560798548,
+      "grad_norm": 1.1471822261810303,
+      "learning_rate": 0.0002,
+      "loss": 0.4239,
+      "step": 8020
+    },
+    {
+      "epoch": 5.829401088929219,
+      "grad_norm": 1.2997334003448486,
+      "learning_rate": 0.0002,
+      "loss": 0.5042,
+      "step": 8030
+    },
+    {
+      "epoch": 5.836660617059891,
+      "grad_norm": 1.1027175188064575,
+      "learning_rate": 0.0002,
+      "loss": 0.4758,
+      "step": 8040
+    },
+    {
+      "epoch": 5.8439201451905625,
+      "grad_norm": 1.2695307731628418,
+      "learning_rate": 0.0002,
+      "loss": 0.4192,
+      "step": 8050
+    },
+    {
+      "epoch": 5.8511796733212345,
+      "grad_norm": 1.5275461673736572,
+      "learning_rate": 0.0002,
+      "loss": 0.5173,
+      "step": 8060
+    },
+    {
+      "epoch": 5.8584392014519056,
+      "grad_norm": 1.3059501647949219,
+      "learning_rate": 0.0002,
+      "loss": 0.5012,
+      "step": 8070
+    },
+    {
+      "epoch": 5.8656987295825775,
+      "grad_norm": 1.57442045211792,
+      "learning_rate": 0.0002,
+      "loss": 0.4425,
+      "step": 8080
+    },
+    {
+      "epoch": 5.872958257713249,
+      "grad_norm": 1.119564414024353,
+      "learning_rate": 0.0002,
+      "loss": 0.4261,
+      "step": 8090
+    },
+    {
+      "epoch": 5.88021778584392,
+      "grad_norm": 1.6517373323440552,
+      "learning_rate": 0.0002,
+      "loss": 0.465,
+      "step": 8100
+    },
+    {
+      "epoch": 5.887477313974592,
+      "grad_norm": 1.4093554019927979,
+      "learning_rate": 0.0002,
+      "loss": 0.4406,
+      "step": 8110
+    },
+    {
+      "epoch": 5.894736842105263,
+      "grad_norm": 1.278843641281128,
+      "learning_rate": 0.0002,
+      "loss": 0.4433,
+      "step": 8120
+    },
+    {
+      "epoch": 5.901996370235935,
+      "grad_norm": 1.2042944431304932,
+      "learning_rate": 0.0002,
+      "loss": 0.4007,
+      "step": 8130
+    },
+    {
+      "epoch": 5.909255898366606,
+      "grad_norm": 1.1788326501846313,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 8140
+    },
+    {
+      "epoch": 5.916515426497278,
+      "grad_norm": 1.4364569187164307,
+      "learning_rate": 0.0002,
+      "loss": 0.4506,
+      "step": 8150
+    },
+    {
+      "epoch": 5.923774954627949,
+      "grad_norm": 1.1704283952713013,
+      "learning_rate": 0.0002,
+      "loss": 0.4651,
+      "step": 8160
+    },
+    {
+      "epoch": 5.931034482758621,
+      "grad_norm": 1.040814995765686,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 8170
+    },
+    {
+      "epoch": 5.938294010889292,
+      "grad_norm": 1.1367416381835938,
+      "learning_rate": 0.0002,
+      "loss": 0.4038,
+      "step": 8180
+    },
+    {
+      "epoch": 5.945553539019964,
+      "grad_norm": 1.3401511907577515,
+      "learning_rate": 0.0002,
+      "loss": 0.4387,
+      "step": 8190
+    },
+    {
+      "epoch": 5.952813067150635,
+      "grad_norm": 1.1154041290283203,
+      "learning_rate": 0.0002,
+      "loss": 0.4396,
+      "step": 8200
+    },
+    {
+      "epoch": 5.960072595281307,
+      "grad_norm": 1.426089882850647,
+      "learning_rate": 0.0002,
+      "loss": 0.4744,
+      "step": 8210
+    },
+    {
+      "epoch": 5.967332123411978,
+      "grad_norm": 1.3170222043991089,
+      "learning_rate": 0.0002,
+      "loss": 0.4105,
+      "step": 8220
+    },
+    {
+      "epoch": 5.974591651542649,
+      "grad_norm": 1.1960029602050781,
+      "learning_rate": 0.0002,
+      "loss": 0.4137,
+      "step": 8230
+    },
+    {
+      "epoch": 5.981851179673321,
+      "grad_norm": 1.0843931436538696,
+      "learning_rate": 0.0002,
+      "loss": 0.423,
+      "step": 8240
+    },
+    {
+      "epoch": 5.9891107078039925,
+      "grad_norm": 1.050421118736267,
+      "learning_rate": 0.0002,
+      "loss": 0.459,
+      "step": 8250
+    },
+    {
+      "epoch": 5.9963702359346644,
+      "grad_norm": 1.0183138847351074,
+      "learning_rate": 0.0002,
+      "loss": 0.3993,
+      "step": 8260
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.4677470922470093,
+      "eval_runtime": 46.2504,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 8265
+    },
+    {
+      "epoch": 6.0036297640653356,
+      "grad_norm": 1.1702998876571655,
+      "learning_rate": 0.0002,
+      "loss": 0.3947,
+      "step": 8270
+    },
+    {
+      "epoch": 6.0108892921960075,
+      "grad_norm": 1.5389727354049683,
+      "learning_rate": 0.0002,
+      "loss": 0.2854,
+      "step": 8280
+    },
+    {
+      "epoch": 6.018148820326679,
+      "grad_norm": 1.502568244934082,
+      "learning_rate": 0.0002,
+      "loss": 0.2603,
+      "step": 8290
+    },
+    {
+      "epoch": 6.025408348457351,
+      "grad_norm": 1.3846043348312378,
+      "learning_rate": 0.0002,
+      "loss": 0.3329,
+      "step": 8300
+    },
+    {
+      "epoch": 6.032667876588022,
+      "grad_norm": 1.173553228378296,
+      "learning_rate": 0.0002,
+      "loss": 0.2651,
+      "step": 8310
+    },
+    {
+      "epoch": 6.039927404718694,
+      "grad_norm": 1.5325932502746582,
+      "learning_rate": 0.0002,
+      "loss": 0.3142,
+      "step": 8320
+    },
+    {
+      "epoch": 6.047186932849365,
+      "grad_norm": 1.303783655166626,
+      "learning_rate": 0.0002,
+      "loss": 0.2511,
+      "step": 8330
+    },
+    {
+      "epoch": 6.054446460980036,
+      "grad_norm": 0.9408994913101196,
+      "learning_rate": 0.0002,
+      "loss": 0.2352,
+      "step": 8340
+    },
+    {
+      "epoch": 6.061705989110708,
+      "grad_norm": 1.5430388450622559,
+      "learning_rate": 0.0002,
+      "loss": 0.2548,
+      "step": 8350
+    },
+    {
+      "epoch": 6.068965517241379,
+      "grad_norm": 0.8765342235565186,
+      "learning_rate": 0.0002,
+      "loss": 0.2682,
+      "step": 8360
+    },
+    {
+      "epoch": 6.076225045372051,
+      "grad_norm": 1.2363157272338867,
+      "learning_rate": 0.0002,
+      "loss": 0.2614,
+      "step": 8370
+    },
+    {
+      "epoch": 6.083484573502722,
+      "grad_norm": 1.21284818649292,
+      "learning_rate": 0.0002,
+      "loss": 0.294,
+      "step": 8380
+    },
+    {
+      "epoch": 6.090744101633394,
+      "grad_norm": 1.3261712789535522,
+      "learning_rate": 0.0002,
+      "loss": 0.2498,
+      "step": 8390
+    },
+    {
+      "epoch": 6.098003629764065,
+      "grad_norm": 1.077317714691162,
+      "learning_rate": 0.0002,
+      "loss": 0.2649,
+      "step": 8400
+    },
+    {
+      "epoch": 6.105263157894737,
+      "grad_norm": 0.9873808026313782,
+      "learning_rate": 0.0002,
+      "loss": 0.269,
+      "step": 8410
+    },
+    {
+      "epoch": 6.112522686025408,
+      "grad_norm": 1.032258152961731,
+      "learning_rate": 0.0002,
+      "loss": 0.2736,
+      "step": 8420
+    },
+    {
+      "epoch": 6.11978221415608,
+      "grad_norm": 1.1014811992645264,
+      "learning_rate": 0.0002,
+      "loss": 0.2854,
+      "step": 8430
+    },
+    {
+      "epoch": 6.127041742286751,
+      "grad_norm": 1.4264203310012817,
+      "learning_rate": 0.0002,
+      "loss": 0.2924,
+      "step": 8440
+    },
+    {
+      "epoch": 6.1343012704174225,
+      "grad_norm": 1.4086531400680542,
+      "learning_rate": 0.0002,
+      "loss": 0.3388,
+      "step": 8450
+    },
+    {
+      "epoch": 6.1415607985480944,
+      "grad_norm": 1.3842453956604004,
+      "learning_rate": 0.0002,
+      "loss": 0.2786,
+      "step": 8460
+    },
+    {
+      "epoch": 6.1488203266787655,
+      "grad_norm": 1.4356757402420044,
+      "learning_rate": 0.0002,
+      "loss": 0.3201,
+      "step": 8470
+    },
+    {
+      "epoch": 6.1560798548094375,
+      "grad_norm": 1.193315029144287,
+      "learning_rate": 0.0002,
+      "loss": 0.2908,
+      "step": 8480
+    },
+    {
+      "epoch": 6.163339382940109,
+      "grad_norm": 1.0623924732208252,
+      "learning_rate": 0.0002,
+      "loss": 0.342,
+      "step": 8490
+    },
+    {
+      "epoch": 6.170598911070781,
+      "grad_norm": 1.5484434366226196,
+      "learning_rate": 0.0002,
+      "loss": 0.3257,
+      "step": 8500
+    },
+    {
+      "epoch": 6.177858439201452,
+      "grad_norm": 1.3520029783248901,
+      "learning_rate": 0.0002,
+      "loss": 0.2861,
+      "step": 8510
+    },
+    {
+      "epoch": 6.185117967332124,
+      "grad_norm": 1.2773103713989258,
+      "learning_rate": 0.0002,
+      "loss": 0.3242,
+      "step": 8520
+    },
+    {
+      "epoch": 6.192377495462795,
+      "grad_norm": 1.4675105810165405,
+      "learning_rate": 0.0002,
+      "loss": 0.3108,
+      "step": 8530
+    },
+    {
+      "epoch": 6.199637023593467,
+      "grad_norm": 1.2118732929229736,
+      "learning_rate": 0.0002,
+      "loss": 0.3044,
+      "step": 8540
+    },
+    {
+      "epoch": 6.206896551724138,
+      "grad_norm": 1.264024257659912,
+      "learning_rate": 0.0002,
+      "loss": 0.2726,
+      "step": 8550
+    },
+    {
+      "epoch": 6.21415607985481,
+      "grad_norm": 1.406931757926941,
+      "learning_rate": 0.0002,
+      "loss": 0.306,
+      "step": 8560
+    },
+    {
+      "epoch": 6.221415607985481,
+      "grad_norm": 1.385459542274475,
+      "learning_rate": 0.0002,
+      "loss": 0.2904,
+      "step": 8570
+    },
+    {
+      "epoch": 6.228675136116152,
+      "grad_norm": 1.9336168766021729,
+      "learning_rate": 0.0002,
+      "loss": 0.3413,
+      "step": 8580
+    },
+    {
+      "epoch": 6.235934664246824,
+      "grad_norm": 0.9880136847496033,
+      "learning_rate": 0.0002,
+      "loss": 0.2769,
+      "step": 8590
+    },
+    {
+      "epoch": 6.243194192377495,
+      "grad_norm": 1.3870339393615723,
+      "learning_rate": 0.0002,
+      "loss": 0.3035,
+      "step": 8600
+    },
+    {
+      "epoch": 6.250453720508167,
+      "grad_norm": 1.2303647994995117,
+      "learning_rate": 0.0002,
+      "loss": 0.286,
+      "step": 8610
+    },
+    {
+      "epoch": 6.257713248638838,
+      "grad_norm": 1.5406211614608765,
+      "learning_rate": 0.0002,
+      "loss": 0.3113,
+      "step": 8620
+    },
+    {
+      "epoch": 6.26497277676951,
+      "grad_norm": 1.2436790466308594,
+      "learning_rate": 0.0002,
+      "loss": 0.292,
+      "step": 8630
+    },
+    {
+      "epoch": 6.272232304900181,
+      "grad_norm": 0.8844212293624878,
+      "learning_rate": 0.0002,
+      "loss": 0.3102,
+      "step": 8640
+    },
+    {
+      "epoch": 6.279491833030853,
+      "grad_norm": 1.2846336364746094,
+      "learning_rate": 0.0002,
+      "loss": 0.3373,
+      "step": 8650
+    },
+    {
+      "epoch": 6.286751361161524,
+      "grad_norm": 1.593814730644226,
+      "learning_rate": 0.0002,
+      "loss": 0.3535,
+      "step": 8660
+    },
+    {
+      "epoch": 6.2940108892921955,
+      "grad_norm": 1.2277469635009766,
+      "learning_rate": 0.0002,
+      "loss": 0.3413,
+      "step": 8670
+    },
+    {
+      "epoch": 6.3012704174228675,
+      "grad_norm": 1.2574384212493896,
+      "learning_rate": 0.0002,
+      "loss": 0.2958,
+      "step": 8680
+    },
+    {
+      "epoch": 6.308529945553539,
+      "grad_norm": 1.335150957107544,
+      "learning_rate": 0.0002,
+      "loss": 0.3251,
+      "step": 8690
+    },
+    {
+      "epoch": 6.315789473684211,
+      "grad_norm": 1.3140437602996826,
+      "learning_rate": 0.0002,
+      "loss": 0.3009,
+      "step": 8700
+    },
+    {
+      "epoch": 6.323049001814882,
+      "grad_norm": 1.1689209938049316,
+      "learning_rate": 0.0002,
+      "loss": 0.2783,
+      "step": 8710
+    },
+    {
+      "epoch": 6.330308529945554,
+      "grad_norm": 1.6448503732681274,
+      "learning_rate": 0.0002,
+      "loss": 0.3476,
+      "step": 8720
+    },
+    {
+      "epoch": 6.337568058076225,
+      "grad_norm": 0.9944145679473877,
+      "learning_rate": 0.0002,
+      "loss": 0.2934,
+      "step": 8730
+    },
+    {
+      "epoch": 6.344827586206897,
+      "grad_norm": 1.1775634288787842,
+      "learning_rate": 0.0002,
+      "loss": 0.3315,
+      "step": 8740
+    },
+    {
+      "epoch": 6.352087114337568,
+      "grad_norm": 1.8438639640808105,
+      "learning_rate": 0.0002,
+      "loss": 0.3514,
+      "step": 8750
+    },
+    {
+      "epoch": 6.35934664246824,
+      "grad_norm": 1.062495470046997,
+      "learning_rate": 0.0002,
+      "loss": 0.303,
+      "step": 8760
+    },
+    {
+      "epoch": 6.366606170598911,
+      "grad_norm": 1.3224315643310547,
+      "learning_rate": 0.0002,
+      "loss": 0.2737,
+      "step": 8770
+    },
+    {
+      "epoch": 6.373865698729583,
+      "grad_norm": 1.399844765663147,
+      "learning_rate": 0.0002,
+      "loss": 0.3445,
+      "step": 8780
+    },
+    {
+      "epoch": 6.381125226860254,
+      "grad_norm": 1.0409915447235107,
+      "learning_rate": 0.0002,
+      "loss": 0.3277,
+      "step": 8790
+    },
+    {
+      "epoch": 6.388384754990925,
+      "grad_norm": 1.5657726526260376,
+      "learning_rate": 0.0002,
+      "loss": 0.3218,
+      "step": 8800
+    },
+    {
+      "epoch": 6.395644283121597,
+      "grad_norm": 1.4098644256591797,
+      "learning_rate": 0.0002,
+      "loss": 0.3031,
+      "step": 8810
+    },
+    {
+      "epoch": 6.402903811252268,
+      "grad_norm": 1.5154732465744019,
+      "learning_rate": 0.0002,
+      "loss": 0.3133,
+      "step": 8820
+    },
+    {
+      "epoch": 6.41016333938294,
+      "grad_norm": 1.1139698028564453,
+      "learning_rate": 0.0002,
+      "loss": 0.3111,
+      "step": 8830
+    },
+    {
+      "epoch": 6.417422867513611,
+      "grad_norm": 1.4149729013442993,
+      "learning_rate": 0.0002,
+      "loss": 0.3553,
+      "step": 8840
+    },
+    {
+      "epoch": 6.424682395644283,
+      "grad_norm": 1.2632299661636353,
+      "learning_rate": 0.0002,
+      "loss": 0.287,
+      "step": 8850
+    },
+    {
+      "epoch": 6.431941923774954,
+      "grad_norm": 1.6636109352111816,
+      "learning_rate": 0.0002,
+      "loss": 0.3198,
+      "step": 8860
+    },
+    {
+      "epoch": 6.439201451905626,
+      "grad_norm": 1.4149386882781982,
+      "learning_rate": 0.0002,
+      "loss": 0.3749,
+      "step": 8870
+    },
+    {
+      "epoch": 6.4464609800362975,
+      "grad_norm": 1.1396206617355347,
+      "learning_rate": 0.0002,
+      "loss": 0.3504,
+      "step": 8880
+    },
+    {
+      "epoch": 6.4537205081669695,
+      "grad_norm": 1.2188775539398193,
+      "learning_rate": 0.0002,
+      "loss": 0.3328,
+      "step": 8890
+    },
+    {
+      "epoch": 6.460980036297641,
+      "grad_norm": 0.9740369319915771,
+      "learning_rate": 0.0002,
+      "loss": 0.3427,
+      "step": 8900
+    },
+    {
+      "epoch": 6.468239564428313,
+      "grad_norm": 1.228569746017456,
+      "learning_rate": 0.0002,
+      "loss": 0.3223,
+      "step": 8910
+    },
+    {
+      "epoch": 6.475499092558984,
+      "grad_norm": 1.5019789934158325,
+      "learning_rate": 0.0002,
+      "loss": 0.3151,
+      "step": 8920
+    },
+    {
+      "epoch": 6.482758620689655,
+      "grad_norm": 1.3320101499557495,
+      "learning_rate": 0.0002,
+      "loss": 0.2916,
+      "step": 8930
+    },
+    {
+      "epoch": 6.490018148820327,
+      "grad_norm": 1.5551502704620361,
+      "learning_rate": 0.0002,
+      "loss": 0.298,
+      "step": 8940
+    },
+    {
+      "epoch": 6.497277676950998,
+      "grad_norm": 1.470131754875183,
+      "learning_rate": 0.0002,
+      "loss": 0.3238,
+      "step": 8950
+    },
+    {
+      "epoch": 6.50453720508167,
+      "grad_norm": 1.1803025007247925,
+      "learning_rate": 0.0002,
+      "loss": 0.2808,
+      "step": 8960
+    },
+    {
+      "epoch": 6.511796733212341,
+      "grad_norm": 1.3505640029907227,
+      "learning_rate": 0.0002,
+      "loss": 0.3025,
+      "step": 8970
+    },
+    {
+      "epoch": 6.519056261343013,
+      "grad_norm": 1.13093900680542,
+      "learning_rate": 0.0002,
+      "loss": 0.3124,
+      "step": 8980
+    },
+    {
+      "epoch": 6.526315789473684,
+      "grad_norm": 1.347386121749878,
+      "learning_rate": 0.0002,
+      "loss": 0.3454,
+      "step": 8990
+    },
+    {
+      "epoch": 6.533575317604356,
+      "grad_norm": 1.7879165410995483,
+      "learning_rate": 0.0002,
+      "loss": 0.3532,
+      "step": 9000
+    },
+    {
+      "epoch": 6.540834845735027,
+      "grad_norm": 1.2168169021606445,
+      "learning_rate": 0.0002,
+      "loss": 0.3382,
+      "step": 9010
+    },
+    {
+      "epoch": 6.548094373865698,
+      "grad_norm": 1.1758877038955688,
+      "learning_rate": 0.0002,
+      "loss": 0.3413,
+      "step": 9020
+    },
+    {
+      "epoch": 6.55535390199637,
+      "grad_norm": 1.7366445064544678,
+      "learning_rate": 0.0002,
+      "loss": 0.2806,
+      "step": 9030
+    },
+    {
+      "epoch": 6.562613430127042,
+      "grad_norm": 1.5919222831726074,
+      "learning_rate": 0.0002,
+      "loss": 0.3437,
+      "step": 9040
+    },
+    {
+      "epoch": 6.569872958257713,
+      "grad_norm": 1.336863398551941,
+      "learning_rate": 0.0002,
+      "loss": 0.3261,
+      "step": 9050
+    },
+    {
+      "epoch": 6.577132486388384,
+      "grad_norm": 1.1769421100616455,
+      "learning_rate": 0.0002,
+      "loss": 0.3103,
+      "step": 9060
+    },
+    {
+      "epoch": 6.584392014519056,
+      "grad_norm": 1.0048751831054688,
+      "learning_rate": 0.0002,
+      "loss": 0.3295,
+      "step": 9070
+    },
+    {
+      "epoch": 6.5916515426497275,
+      "grad_norm": 1.5268515348434448,
+      "learning_rate": 0.0002,
+      "loss": 0.3156,
+      "step": 9080
+    },
+    {
+      "epoch": 6.5989110707803995,
+      "grad_norm": 1.434610366821289,
+      "learning_rate": 0.0002,
+      "loss": 0.3752,
+      "step": 9090
+    },
+    {
+      "epoch": 6.606170598911071,
+      "grad_norm": 1.1151410341262817,
+      "learning_rate": 0.0002,
+      "loss": 0.3375,
+      "step": 9100
+    },
+    {
+      "epoch": 6.613430127041743,
+      "grad_norm": 1.6690642833709717,
+      "learning_rate": 0.0002,
+      "loss": 0.363,
+      "step": 9110
+    },
+    {
+      "epoch": 6.620689655172414,
+      "grad_norm": 1.4495552778244019,
+      "learning_rate": 0.0002,
+      "loss": 0.3703,
+      "step": 9120
+    },
+    {
+      "epoch": 6.627949183303086,
+      "grad_norm": 1.377621054649353,
+      "learning_rate": 0.0002,
+      "loss": 0.3648,
+      "step": 9130
+    },
+    {
+      "epoch": 6.635208711433757,
+      "grad_norm": 1.5459434986114502,
+      "learning_rate": 0.0002,
+      "loss": 0.3766,
+      "step": 9140
+    },
+    {
+      "epoch": 6.642468239564428,
+      "grad_norm": 1.0920850038528442,
+      "learning_rate": 0.0002,
+      "loss": 0.3196,
+      "step": 9150
+    },
+    {
+      "epoch": 6.6497277676951,
+      "grad_norm": 1.6708381175994873,
+      "learning_rate": 0.0002,
+      "loss": 0.3505,
+      "step": 9160
+    },
+    {
+      "epoch": 6.656987295825771,
+      "grad_norm": 1.747514009475708,
+      "learning_rate": 0.0002,
+      "loss": 0.368,
+      "step": 9170
+    },
+    {
+      "epoch": 6.664246823956443,
+      "grad_norm": 1.133466362953186,
+      "learning_rate": 0.0002,
+      "loss": 0.3099,
+      "step": 9180
+    },
+    {
+      "epoch": 6.671506352087114,
+      "grad_norm": 1.394358515739441,
+      "learning_rate": 0.0002,
+      "loss": 0.3175,
+      "step": 9190
+    },
+    {
+      "epoch": 6.678765880217786,
+      "grad_norm": 0.9258374571800232,
+      "learning_rate": 0.0002,
+      "loss": 0.2981,
+      "step": 9200
+    },
+    {
+      "epoch": 6.686025408348457,
+      "grad_norm": 1.3750739097595215,
+      "learning_rate": 0.0002,
+      "loss": 0.3723,
+      "step": 9210
+    },
+    {
+      "epoch": 6.693284936479129,
+      "grad_norm": 0.8604967594146729,
+      "learning_rate": 0.0002,
+      "loss": 0.3441,
+      "step": 9220
+    },
+    {
+      "epoch": 6.7005444646098,
+      "grad_norm": 1.6074559688568115,
+      "learning_rate": 0.0002,
+      "loss": 0.3775,
+      "step": 9230
+    },
+    {
+      "epoch": 6.707803992740472,
+      "grad_norm": 0.9576877355575562,
+      "learning_rate": 0.0002,
+      "loss": 0.3139,
+      "step": 9240
+    },
+    {
+      "epoch": 6.715063520871143,
+      "grad_norm": 1.7193048000335693,
+      "learning_rate": 0.0002,
+      "loss": 0.3633,
+      "step": 9250
+    },
+    {
+      "epoch": 6.722323049001815,
+      "grad_norm": 1.3131844997406006,
+      "learning_rate": 0.0002,
+      "loss": 0.3139,
+      "step": 9260
+    },
+    {
+      "epoch": 6.729582577132486,
+      "grad_norm": 1.2978184223175049,
+      "learning_rate": 0.0002,
+      "loss": 0.3121,
+      "step": 9270
+    },
+    {
+      "epoch": 6.7368421052631575,
+      "grad_norm": 1.4792617559432983,
+      "learning_rate": 0.0002,
+      "loss": 0.3534,
+      "step": 9280
+    },
+    {
+      "epoch": 6.7441016333938295,
+      "grad_norm": 1.1265567541122437,
+      "learning_rate": 0.0002,
+      "loss": 0.3429,
+      "step": 9290
+    },
+    {
+      "epoch": 6.751361161524501,
+      "grad_norm": 1.8553377389907837,
+      "learning_rate": 0.0002,
+      "loss": 0.3526,
+      "step": 9300
+    },
+    {
+      "epoch": 6.758620689655173,
+      "grad_norm": 1.3602519035339355,
+      "learning_rate": 0.0002,
+      "loss": 0.3666,
+      "step": 9310
+    },
+    {
+      "epoch": 6.765880217785844,
+      "grad_norm": 1.2874794006347656,
+      "learning_rate": 0.0002,
+      "loss": 0.2922,
+      "step": 9320
+    },
+    {
+      "epoch": 6.773139745916516,
+      "grad_norm": 1.4834712743759155,
+      "learning_rate": 0.0002,
+      "loss": 0.3816,
+      "step": 9330
+    },
+    {
+      "epoch": 6.780399274047187,
+      "grad_norm": 2.0824034214019775,
+      "learning_rate": 0.0002,
+      "loss": 0.3557,
+      "step": 9340
+    },
+    {
+      "epoch": 6.787658802177859,
+      "grad_norm": 1.2267698049545288,
+      "learning_rate": 0.0002,
+      "loss": 0.3174,
+      "step": 9350
+    },
+    {
+      "epoch": 6.79491833030853,
+      "grad_norm": 1.4485498666763306,
+      "learning_rate": 0.0002,
+      "loss": 0.3665,
+      "step": 9360
+    },
+    {
+      "epoch": 6.802177858439201,
+      "grad_norm": 1.3199396133422852,
+      "learning_rate": 0.0002,
+      "loss": 0.3676,
+      "step": 9370
+    },
+    {
+      "epoch": 6.809437386569873,
+      "grad_norm": 1.2552456855773926,
+      "learning_rate": 0.0002,
+      "loss": 0.298,
+      "step": 9380
+    },
+    {
+      "epoch": 6.816696914700545,
+      "grad_norm": 1.3895127773284912,
+      "learning_rate": 0.0002,
+      "loss": 0.3152,
+      "step": 9390
+    },
+    {
+      "epoch": 6.823956442831216,
+      "grad_norm": 1.7637823820114136,
+      "learning_rate": 0.0002,
+      "loss": 0.3375,
+      "step": 9400
+    },
+    {
+      "epoch": 6.831215970961887,
+      "grad_norm": 1.6004475355148315,
+      "learning_rate": 0.0002,
+      "loss": 0.3234,
+      "step": 9410
+    },
+    {
+      "epoch": 6.838475499092559,
+      "grad_norm": 1.4133695363998413,
+      "learning_rate": 0.0002,
+      "loss": 0.3364,
+      "step": 9420
+    },
+    {
+      "epoch": 6.84573502722323,
+      "grad_norm": 1.1583502292633057,
+      "learning_rate": 0.0002,
+      "loss": 0.3656,
+      "step": 9430
+    },
+    {
+      "epoch": 6.852994555353902,
+      "grad_norm": 1.3769075870513916,
+      "learning_rate": 0.0002,
+      "loss": 0.3499,
+      "step": 9440
+    },
+    {
+      "epoch": 6.860254083484573,
+      "grad_norm": 1.1831218004226685,
+      "learning_rate": 0.0002,
+      "loss": 0.3333,
+      "step": 9450
+    },
+    {
+      "epoch": 6.867513611615245,
+      "grad_norm": 1.6092621088027954,
+      "learning_rate": 0.0002,
+      "loss": 0.3501,
+      "step": 9460
+    },
+    {
+      "epoch": 6.874773139745916,
+      "grad_norm": 1.3850210905075073,
+      "learning_rate": 0.0002,
+      "loss": 0.3933,
+      "step": 9470
+    },
+    {
+      "epoch": 6.882032667876588,
+      "grad_norm": 1.4119619131088257,
+      "learning_rate": 0.0002,
+      "loss": 0.3868,
+      "step": 9480
+    },
+    {
+      "epoch": 6.8892921960072595,
+      "grad_norm": 1.3494242429733276,
+      "learning_rate": 0.0002,
+      "loss": 0.3939,
+      "step": 9490
+    },
+    {
+      "epoch": 6.896551724137931,
+      "grad_norm": 1.3130041360855103,
+      "learning_rate": 0.0002,
+      "loss": 0.3217,
+      "step": 9500
+    },
+    {
+      "epoch": 6.903811252268603,
+      "grad_norm": 1.169256329536438,
+      "learning_rate": 0.0002,
+      "loss": 0.3738,
+      "step": 9510
+    },
+    {
+      "epoch": 6.911070780399274,
+      "grad_norm": 1.7475035190582275,
+      "learning_rate": 0.0002,
+      "loss": 0.408,
+      "step": 9520
+    },
+    {
+      "epoch": 6.918330308529946,
+      "grad_norm": 1.440434217453003,
+      "learning_rate": 0.0002,
+      "loss": 0.3407,
+      "step": 9530
+    },
+    {
+      "epoch": 6.925589836660617,
+      "grad_norm": 1.6768704652786255,
+      "learning_rate": 0.0002,
+      "loss": 0.3707,
+      "step": 9540
+    },
+    {
+      "epoch": 6.932849364791289,
+      "grad_norm": 1.3720577955245972,
+      "learning_rate": 0.0002,
+      "loss": 0.3283,
+      "step": 9550
+    },
+    {
+      "epoch": 6.94010889292196,
+      "grad_norm": 1.8140523433685303,
+      "learning_rate": 0.0002,
+      "loss": 0.3257,
+      "step": 9560
+    },
+    {
+      "epoch": 6.947368421052632,
+      "grad_norm": 1.1828241348266602,
+      "learning_rate": 0.0002,
+      "loss": 0.3308,
+      "step": 9570
+    },
+    {
+      "epoch": 6.954627949183303,
+      "grad_norm": 1.2755135297775269,
+      "learning_rate": 0.0002,
+      "loss": 0.3536,
+      "step": 9580
+    },
+    {
+      "epoch": 6.961887477313975,
+      "grad_norm": 1.622009038925171,
+      "learning_rate": 0.0002,
+      "loss": 0.3711,
+      "step": 9590
+    },
+    {
+      "epoch": 6.969147005444646,
+      "grad_norm": 1.1543664932250977,
+      "learning_rate": 0.0002,
+      "loss": 0.3529,
+      "step": 9600
+    },
+    {
+      "epoch": 6.976406533575318,
+      "grad_norm": 1.6755319833755493,
+      "learning_rate": 0.0002,
+      "loss": 0.416,
+      "step": 9610
+    },
+    {
+      "epoch": 6.983666061705989,
+      "grad_norm": 1.3726437091827393,
+      "learning_rate": 0.0002,
+      "loss": 0.339,
+      "step": 9620
+    },
+    {
+      "epoch": 6.99092558983666,
+      "grad_norm": 1.1605958938598633,
+      "learning_rate": 0.0002,
+      "loss": 0.3709,
+      "step": 9630
+    },
+    {
+      "epoch": 6.998185117967332,
+      "grad_norm": 1.5371781587600708,
+      "learning_rate": 0.0002,
+      "loss": 0.371,
+      "step": 9640
+    },
+    {
+      "epoch": 6.999637023593467,
+      "eval_loss": 1.6280181407928467,
+      "eval_runtime": 46.1964,
+      "eval_samples_per_second": 9.438,
+      "eval_steps_per_second": 1.191,
+      "step": 9642
+    },
+    {
+      "epoch": 7.005444646098003,
+      "grad_norm": 1.1645569801330566,
+      "learning_rate": 0.0002,
+      "loss": 0.2449,
+      "step": 9650
+    },
+    {
+      "epoch": 7.012704174228675,
+      "grad_norm": 0.7663792967796326,
+      "learning_rate": 0.0002,
+      "loss": 0.251,
+      "step": 9660
+    },
+    {
+      "epoch": 7.019963702359346,
+      "grad_norm": 1.5808782577514648,
+      "learning_rate": 0.0002,
+      "loss": 0.2553,
+      "step": 9670
+    },
+    {
+      "epoch": 7.027223230490018,
+      "grad_norm": 1.046607255935669,
+      "learning_rate": 0.0002,
+      "loss": 0.2349,
+      "step": 9680
+    },
+    {
+      "epoch": 7.0344827586206895,
+      "grad_norm": 1.2008668184280396,
+      "learning_rate": 0.0002,
+      "loss": 0.2078,
+      "step": 9690
+    },
+    {
+      "epoch": 7.0417422867513615,
+      "grad_norm": 1.9596126079559326,
+      "learning_rate": 0.0002,
+      "loss": 0.2519,
+      "step": 9700
+    },
+    {
+      "epoch": 7.049001814882033,
+      "grad_norm": 1.0400182008743286,
+      "learning_rate": 0.0002,
+      "loss": 0.2275,
+      "step": 9710
+    },
+    {
+      "epoch": 7.056261343012705,
+      "grad_norm": 1.3162504434585571,
+      "learning_rate": 0.0002,
+      "loss": 0.2136,
+      "step": 9720
+    },
+    {
+      "epoch": 7.063520871143376,
+      "grad_norm": 1.180074691772461,
+      "learning_rate": 0.0002,
+      "loss": 0.2227,
+      "step": 9730
+    },
+    {
+      "epoch": 7.070780399274047,
+      "grad_norm": 1.2093719244003296,
+      "learning_rate": 0.0002,
+      "loss": 0.2206,
+      "step": 9740
+    },
+    {
+      "epoch": 7.078039927404719,
+      "grad_norm": 1.4838900566101074,
+      "learning_rate": 0.0002,
+      "loss": 0.2387,
+      "step": 9750
+    },
+    {
+      "epoch": 7.08529945553539,
+      "grad_norm": 1.2319235801696777,
+      "learning_rate": 0.0002,
+      "loss": 0.2391,
+      "step": 9760
+    },
+    {
+      "epoch": 7.092558983666062,
+      "grad_norm": 1.2346558570861816,
+      "learning_rate": 0.0002,
+      "loss": 0.2624,
+      "step": 9770
+    },
+    {
+      "epoch": 7.099818511796733,
+      "grad_norm": 1.0748975276947021,
+      "learning_rate": 0.0002,
+      "loss": 0.2321,
+      "step": 9780
+    },
+    {
+      "epoch": 7.107078039927405,
+      "grad_norm": 1.0162630081176758,
+      "learning_rate": 0.0002,
+      "loss": 0.2751,
+      "step": 9790
+    },
+    {
+      "epoch": 7.114337568058076,
+      "grad_norm": 1.0014166831970215,
+      "learning_rate": 0.0002,
+      "loss": 0.2142,
+      "step": 9800
+    },
+    {
+      "epoch": 7.121597096188748,
+      "grad_norm": 1.0928411483764648,
+      "learning_rate": 0.0002,
+      "loss": 0.2439,
+      "step": 9810
+    },
+    {
+      "epoch": 7.128856624319419,
+      "grad_norm": 1.181496500968933,
+      "learning_rate": 0.0002,
+      "loss": 0.2261,
+      "step": 9820
+    },
+    {
+      "epoch": 7.136116152450091,
+      "grad_norm": 1.5846176147460938,
+      "learning_rate": 0.0002,
+      "loss": 0.2345,
+      "step": 9830
+    },
+    {
+      "epoch": 7.143375680580762,
+      "grad_norm": 0.8734912872314453,
+      "learning_rate": 0.0002,
+      "loss": 0.2282,
+      "step": 9840
+    },
+    {
+      "epoch": 7.150635208711433,
+      "grad_norm": 1.1599528789520264,
+      "learning_rate": 0.0002,
+      "loss": 0.2127,
+      "step": 9850
+    },
+    {
+      "epoch": 7.157894736842105,
+      "grad_norm": 1.168256402015686,
+      "learning_rate": 0.0002,
+      "loss": 0.2614,
+      "step": 9860
+    },
+    {
+      "epoch": 7.165154264972776,
+      "grad_norm": 1.4439860582351685,
+      "learning_rate": 0.0002,
+      "loss": 0.2418,
+      "step": 9870
+    },
+    {
+      "epoch": 7.172413793103448,
+      "grad_norm": 1.3615007400512695,
+      "learning_rate": 0.0002,
+      "loss": 0.2183,
+      "step": 9880
+    },
+    {
+      "epoch": 7.1796733212341195,
+      "grad_norm": 1.1908115148544312,
+      "learning_rate": 0.0002,
+      "loss": 0.2366,
+      "step": 9890
+    },
+    {
+      "epoch": 7.1869328493647915,
+      "grad_norm": 1.452515959739685,
+      "learning_rate": 0.0002,
+      "loss": 0.2338,
+      "step": 9900
+    },
+    {
+      "epoch": 7.194192377495463,
+      "grad_norm": 0.8387667536735535,
+      "learning_rate": 0.0002,
+      "loss": 0.2203,
+      "step": 9910
+    },
+    {
+      "epoch": 7.201451905626135,
+      "grad_norm": 1.3990435600280762,
+      "learning_rate": 0.0002,
+      "loss": 0.2117,
+      "step": 9920
+    },
+    {
+      "epoch": 7.208711433756806,
+      "grad_norm": 1.057800531387329,
+      "learning_rate": 0.0002,
+      "loss": 0.2188,
+      "step": 9930
+    },
+    {
+      "epoch": 7.215970961887478,
+      "grad_norm": 1.3718253374099731,
+      "learning_rate": 0.0002,
+      "loss": 0.2516,
+      "step": 9940
+    },
+    {
+      "epoch": 7.223230490018149,
+      "grad_norm": 1.2011432647705078,
+      "learning_rate": 0.0002,
+      "loss": 0.2084,
+      "step": 9950
+    },
+    {
+      "epoch": 7.230490018148821,
+      "grad_norm": 1.1608737707138062,
+      "learning_rate": 0.0002,
+      "loss": 0.2063,
+      "step": 9960
+    },
+    {
+      "epoch": 7.237749546279492,
+      "grad_norm": 1.7522791624069214,
+      "learning_rate": 0.0002,
+      "loss": 0.2275,
+      "step": 9970
+    },
+    {
+      "epoch": 7.245009074410163,
+      "grad_norm": 1.0787912607192993,
+      "learning_rate": 0.0002,
+      "loss": 0.2358,
+      "step": 9980
+    },
+    {
+      "epoch": 7.252268602540835,
+      "grad_norm": 1.8227689266204834,
+      "learning_rate": 0.0002,
+      "loss": 0.2361,
+      "step": 9990
+    },
+    {
+      "epoch": 7.259528130671506,
+      "grad_norm": 1.1438913345336914,
+      "learning_rate": 0.0002,
+      "loss": 0.2217,
+      "step": 10000
+    },
+    {
+      "epoch": 7.266787658802178,
+      "grad_norm": 1.331770420074463,
+      "learning_rate": 0.0002,
+      "loss": 0.2438,
+      "step": 10010
+    },
+    {
+      "epoch": 7.274047186932849,
+      "grad_norm": 1.2809056043624878,
+      "learning_rate": 0.0002,
+      "loss": 0.2622,
+      "step": 10020
+    },
+    {
+      "epoch": 7.281306715063521,
+      "grad_norm": 1.2245303392410278,
+      "learning_rate": 0.0002,
+      "loss": 0.2634,
+      "step": 10030
+    },
+    {
+      "epoch": 7.288566243194192,
+      "grad_norm": 1.2359435558319092,
+      "learning_rate": 0.0002,
+      "loss": 0.2706,
+      "step": 10040
+    },
+    {
+      "epoch": 7.295825771324864,
+      "grad_norm": 1.3707170486450195,
+      "learning_rate": 0.0002,
+      "loss": 0.2427,
+      "step": 10050
+    },
+    {
+      "epoch": 7.303085299455535,
+      "grad_norm": 1.7405836582183838,
+      "learning_rate": 0.0002,
+      "loss": 0.2497,
+      "step": 10060
+    },
+    {
+      "epoch": 7.310344827586207,
+      "grad_norm": 1.446069359779358,
+      "learning_rate": 0.0002,
+      "loss": 0.2628,
+      "step": 10070
+    },
+    {
+      "epoch": 7.317604355716878,
+      "grad_norm": 1.48823082447052,
+      "learning_rate": 0.0002,
+      "loss": 0.2457,
+      "step": 10080
+    },
+    {
+      "epoch": 7.3248638838475495,
+      "grad_norm": 1.1720311641693115,
+      "learning_rate": 0.0002,
+      "loss": 0.2391,
+      "step": 10090
+    },
+    {
+      "epoch": 7.3321234119782215,
+      "grad_norm": 1.5485225915908813,
+      "learning_rate": 0.0002,
+      "loss": 0.2701,
+      "step": 10100
+    },
+    {
+      "epoch": 7.339382940108893,
+      "grad_norm": 1.6018894910812378,
+      "learning_rate": 0.0002,
+      "loss": 0.2855,
+      "step": 10110
+    },
+    {
+      "epoch": 7.346642468239565,
+      "grad_norm": 1.4753694534301758,
+      "learning_rate": 0.0002,
+      "loss": 0.2662,
+      "step": 10120
+    },
+    {
+      "epoch": 7.353901996370236,
+      "grad_norm": 1.3604710102081299,
+      "learning_rate": 0.0002,
+      "loss": 0.2323,
+      "step": 10130
+    },
+    {
+      "epoch": 7.361161524500908,
+      "grad_norm": 1.5755873918533325,
+      "learning_rate": 0.0002,
+      "loss": 0.2664,
+      "step": 10140
+    },
+    {
+      "epoch": 7.368421052631579,
+      "grad_norm": 0.9421692490577698,
+      "learning_rate": 0.0002,
+      "loss": 0.215,
+      "step": 10150
+    },
+    {
+      "epoch": 7.375680580762251,
+      "grad_norm": 1.3055956363677979,
+      "learning_rate": 0.0002,
+      "loss": 0.297,
+      "step": 10160
+    },
+    {
+      "epoch": 7.382940108892922,
+      "grad_norm": 1.4764302968978882,
+      "learning_rate": 0.0002,
+      "loss": 0.2385,
+      "step": 10170
+    },
+    {
+      "epoch": 7.390199637023594,
+      "grad_norm": 1.3726946115493774,
+      "learning_rate": 0.0002,
+      "loss": 0.2724,
+      "step": 10180
+    },
+    {
+      "epoch": 7.397459165154265,
+      "grad_norm": 1.446473240852356,
+      "learning_rate": 0.0002,
+      "loss": 0.2599,
+      "step": 10190
+    },
+    {
+      "epoch": 7.404718693284936,
+      "grad_norm": 1.489094614982605,
+      "learning_rate": 0.0002,
+      "loss": 0.2837,
+      "step": 10200
+    },
+    {
+      "epoch": 7.411978221415608,
+      "grad_norm": 1.247572898864746,
+      "learning_rate": 0.0002,
+      "loss": 0.2786,
+      "step": 10210
+    },
+    {
+      "epoch": 7.419237749546279,
+      "grad_norm": 1.2741918563842773,
+      "learning_rate": 0.0002,
+      "loss": 0.2498,
+      "step": 10220
+    },
+    {
+      "epoch": 7.426497277676951,
+      "grad_norm": 1.0347636938095093,
+      "learning_rate": 0.0002,
+      "loss": 0.2649,
+      "step": 10230
+    },
+    {
+      "epoch": 7.433756805807622,
+      "grad_norm": 1.3295499086380005,
+      "learning_rate": 0.0002,
+      "loss": 0.2419,
+      "step": 10240
+    },
+    {
+      "epoch": 7.441016333938294,
+      "grad_norm": 1.6056840419769287,
+      "learning_rate": 0.0002,
+      "loss": 0.226,
+      "step": 10250
+    },
+    {
+      "epoch": 7.448275862068965,
+      "grad_norm": 1.4824398756027222,
+      "learning_rate": 0.0002,
+      "loss": 0.2571,
+      "step": 10260
+    },
+    {
+      "epoch": 7.455535390199637,
+      "grad_norm": 1.6259359121322632,
+      "learning_rate": 0.0002,
+      "loss": 0.2939,
+      "step": 10270
+    },
+    {
+      "epoch": 7.462794918330308,
+      "grad_norm": 1.5065499544143677,
+      "learning_rate": 0.0002,
+      "loss": 0.2873,
+      "step": 10280
+    },
+    {
+      "epoch": 7.47005444646098,
+      "grad_norm": 1.3505364656448364,
+      "learning_rate": 0.0002,
+      "loss": 0.2614,
+      "step": 10290
+    },
+    {
+      "epoch": 7.4773139745916515,
+      "grad_norm": 1.4457359313964844,
+      "learning_rate": 0.0002,
+      "loss": 0.2862,
+      "step": 10300
+    },
+    {
+      "epoch": 7.4845735027223235,
+      "grad_norm": 1.0782662630081177,
+      "learning_rate": 0.0002,
+      "loss": 0.233,
+      "step": 10310
+    },
+    {
+      "epoch": 7.491833030852995,
+      "grad_norm": 1.4209016561508179,
+      "learning_rate": 0.0002,
+      "loss": 0.2506,
+      "step": 10320
+    },
+    {
+      "epoch": 7.499092558983666,
+      "grad_norm": 1.4511336088180542,
+      "learning_rate": 0.0002,
+      "loss": 0.2741,
+      "step": 10330
+    },
+    {
+      "epoch": 7.506352087114338,
+      "grad_norm": 1.306691288948059,
+      "learning_rate": 0.0002,
+      "loss": 0.2736,
+      "step": 10340
+    },
+    {
+      "epoch": 7.513611615245009,
+      "grad_norm": 1.0647870302200317,
+      "learning_rate": 0.0002,
+      "loss": 0.2528,
+      "step": 10350
+    },
+    {
+      "epoch": 7.520871143375681,
+      "grad_norm": 1.0374330282211304,
+      "learning_rate": 0.0002,
+      "loss": 0.2402,
+      "step": 10360
+    },
+    {
+      "epoch": 7.528130671506352,
+      "grad_norm": 0.8428803086280823,
+      "learning_rate": 0.0002,
+      "loss": 0.3057,
+      "step": 10370
+    },
+    {
+      "epoch": 7.535390199637024,
+      "grad_norm": 1.3868707418441772,
+      "learning_rate": 0.0002,
+      "loss": 0.2446,
+      "step": 10380
+    },
+    {
+      "epoch": 7.542649727767695,
+      "grad_norm": 1.4324088096618652,
+      "learning_rate": 0.0002,
+      "loss": 0.2851,
+      "step": 10390
+    },
+    {
+      "epoch": 7.549909255898367,
+      "grad_norm": 1.6413776874542236,
+      "learning_rate": 0.0002,
+      "loss": 0.292,
+      "step": 10400
+    },
+    {
+      "epoch": 7.557168784029038,
+      "grad_norm": 1.4302188158035278,
+      "learning_rate": 0.0002,
+      "loss": 0.2585,
+      "step": 10410
+    },
+    {
+      "epoch": 7.564428312159709,
+      "grad_norm": 1.3648983240127563,
+      "learning_rate": 0.0002,
+      "loss": 0.306,
+      "step": 10420
+    },
+    {
+      "epoch": 7.571687840290381,
+      "grad_norm": 1.4480061531066895,
+      "learning_rate": 0.0002,
+      "loss": 0.2927,
+      "step": 10430
+    },
+    {
+      "epoch": 7.578947368421053,
+      "grad_norm": 1.0944541692733765,
+      "learning_rate": 0.0002,
+      "loss": 0.2711,
+      "step": 10440
+    },
+    {
+      "epoch": 7.586206896551724,
+      "grad_norm": 1.4632091522216797,
+      "learning_rate": 0.0002,
+      "loss": 0.2698,
+      "step": 10450
+    },
+    {
+      "epoch": 7.593466424682395,
+      "grad_norm": 1.562364935874939,
+      "learning_rate": 0.0002,
+      "loss": 0.2994,
+      "step": 10460
+    },
+    {
+      "epoch": 7.600725952813067,
+      "grad_norm": 1.2138582468032837,
+      "learning_rate": 0.0002,
+      "loss": 0.271,
+      "step": 10470
+    },
+    {
+      "epoch": 7.607985480943738,
+      "grad_norm": 1.467578411102295,
+      "learning_rate": 0.0002,
+      "loss": 0.2638,
+      "step": 10480
+    },
+    {
+      "epoch": 7.61524500907441,
+      "grad_norm": 1.3470213413238525,
+      "learning_rate": 0.0002,
+      "loss": 0.2949,
+      "step": 10490
+    },
+    {
+      "epoch": 7.6225045372050815,
+      "grad_norm": 1.5385268926620483,
+      "learning_rate": 0.0002,
+      "loss": 0.2518,
+      "step": 10500
+    },
+    {
+      "epoch": 7.6297640653357535,
+      "grad_norm": 1.1245018243789673,
+      "learning_rate": 0.0002,
+      "loss": 0.2527,
+      "step": 10510
+    },
+    {
+      "epoch": 7.637023593466425,
+      "grad_norm": 1.3161317110061646,
+      "learning_rate": 0.0002,
+      "loss": 0.267,
+      "step": 10520
+    },
+    {
+      "epoch": 7.6442831215970966,
+      "grad_norm": 1.0402427911758423,
+      "learning_rate": 0.0002,
+      "loss": 0.2803,
+      "step": 10530
+    },
+    {
+      "epoch": 7.651542649727768,
+      "grad_norm": 1.2699987888336182,
+      "learning_rate": 0.0002,
+      "loss": 0.2676,
+      "step": 10540
+    },
+    {
+      "epoch": 7.658802177858439,
+      "grad_norm": 1.47243332862854,
+      "learning_rate": 0.0002,
+      "loss": 0.2835,
+      "step": 10550
+    },
+    {
+      "epoch": 7.666061705989111,
+      "grad_norm": 1.1261144876480103,
+      "learning_rate": 0.0002,
+      "loss": 0.2624,
+      "step": 10560
+    },
+    {
+      "epoch": 7.673321234119782,
+      "grad_norm": 1.5402237176895142,
+      "learning_rate": 0.0002,
+      "loss": 0.282,
+      "step": 10570
+    },
+    {
+      "epoch": 7.680580762250454,
+      "grad_norm": 1.1316986083984375,
+      "learning_rate": 0.0002,
+      "loss": 0.2866,
+      "step": 10580
+    },
+    {
+      "epoch": 7.687840290381125,
+      "grad_norm": 1.2155439853668213,
+      "learning_rate": 0.0002,
+      "loss": 0.2593,
+      "step": 10590
+    },
+    {
+      "epoch": 7.695099818511797,
+      "grad_norm": 1.566380500793457,
+      "learning_rate": 0.0002,
+      "loss": 0.274,
+      "step": 10600
+    },
+    {
+      "epoch": 7.702359346642468,
+      "grad_norm": 1.7367318868637085,
+      "learning_rate": 0.0002,
+      "loss": 0.2664,
+      "step": 10610
+    },
+    {
+      "epoch": 7.70961887477314,
+      "grad_norm": 1.5213567018508911,
+      "learning_rate": 0.0002,
+      "loss": 0.3115,
+      "step": 10620
+    },
+    {
+      "epoch": 7.716878402903811,
+      "grad_norm": 1.3955585956573486,
+      "learning_rate": 0.0002,
+      "loss": 0.2863,
+      "step": 10630
+    },
+    {
+      "epoch": 7.724137931034483,
+      "grad_norm": 1.321916937828064,
+      "learning_rate": 0.0002,
+      "loss": 0.2683,
+      "step": 10640
+    },
+    {
+      "epoch": 7.731397459165154,
+      "grad_norm": 1.8494919538497925,
+      "learning_rate": 0.0002,
+      "loss": 0.306,
+      "step": 10650
+    },
+    {
+      "epoch": 7.738656987295826,
+      "grad_norm": 1.5309828519821167,
+      "learning_rate": 0.0002,
+      "loss": 0.2544,
+      "step": 10660
+    },
+    {
+      "epoch": 7.745916515426497,
+      "grad_norm": 1.3796069622039795,
+      "learning_rate": 0.0002,
+      "loss": 0.2693,
+      "step": 10670
+    },
+    {
+      "epoch": 7.753176043557168,
+      "grad_norm": 1.2416858673095703,
+      "learning_rate": 0.0002,
+      "loss": 0.2872,
+      "step": 10680
+    },
+    {
+      "epoch": 7.76043557168784,
+      "grad_norm": 1.4447332620620728,
+      "learning_rate": 0.0002,
+      "loss": 0.2729,
+      "step": 10690
+    },
+    {
+      "epoch": 7.7676950998185115,
+      "grad_norm": 1.2003352642059326,
+      "learning_rate": 0.0002,
+      "loss": 0.2736,
+      "step": 10700
+    },
+    {
+      "epoch": 7.7749546279491835,
+      "grad_norm": 1.3607908487319946,
+      "learning_rate": 0.0002,
+      "loss": 0.2771,
+      "step": 10710
+    },
+    {
+      "epoch": 7.782214156079855,
+      "grad_norm": 1.1789227724075317,
+      "learning_rate": 0.0002,
+      "loss": 0.2739,
+      "step": 10720
+    },
+    {
+      "epoch": 7.7894736842105265,
+      "grad_norm": 1.2998148202896118,
+      "learning_rate": 0.0002,
+      "loss": 0.2927,
+      "step": 10730
+    },
+    {
+      "epoch": 7.796733212341198,
+      "grad_norm": 1.8224656581878662,
+      "learning_rate": 0.0002,
+      "loss": 0.2825,
+      "step": 10740
+    },
+    {
+      "epoch": 7.80399274047187,
+      "grad_norm": 1.2510570287704468,
+      "learning_rate": 0.0002,
+      "loss": 0.2864,
+      "step": 10750
+    },
+    {
+      "epoch": 7.811252268602541,
+      "grad_norm": 1.065926194190979,
+      "learning_rate": 0.0002,
+      "loss": 0.3007,
+      "step": 10760
+    },
+    {
+      "epoch": 7.818511796733213,
+      "grad_norm": 1.0313589572906494,
+      "learning_rate": 0.0002,
+      "loss": 0.277,
+      "step": 10770
+    },
+    {
+      "epoch": 7.825771324863884,
+      "grad_norm": 1.10769784450531,
+      "learning_rate": 0.0002,
+      "loss": 0.2954,
+      "step": 10780
+    },
+    {
+      "epoch": 7.833030852994556,
+      "grad_norm": 1.4168727397918701,
+      "learning_rate": 0.0002,
+      "loss": 0.2893,
+      "step": 10790
+    },
+    {
+      "epoch": 7.840290381125227,
+      "grad_norm": 1.8239266872406006,
+      "learning_rate": 0.0002,
+      "loss": 0.2903,
+      "step": 10800
+    },
+    {
+      "epoch": 7.847549909255898,
+      "grad_norm": 1.5748721361160278,
+      "learning_rate": 0.0002,
+      "loss": 0.2573,
+      "step": 10810
+    },
+    {
+      "epoch": 7.85480943738657,
+      "grad_norm": 1.5762766599655151,
+      "learning_rate": 0.0002,
+      "loss": 0.3282,
+      "step": 10820
+    },
+    {
+      "epoch": 7.862068965517241,
+      "grad_norm": 1.1119135618209839,
+      "learning_rate": 0.0002,
+      "loss": 0.2981,
+      "step": 10830
+    },
+    {
+      "epoch": 7.869328493647913,
+      "grad_norm": 1.478314995765686,
+      "learning_rate": 0.0002,
+      "loss": 0.3037,
+      "step": 10840
+    },
+    {
+      "epoch": 7.876588021778584,
+      "grad_norm": 1.2225514650344849,
+      "learning_rate": 0.0002,
+      "loss": 0.2866,
+      "step": 10850
+    },
+    {
+      "epoch": 7.883847549909256,
+      "grad_norm": 1.503473162651062,
+      "learning_rate": 0.0002,
+      "loss": 0.2795,
+      "step": 10860
+    },
+    {
+      "epoch": 7.891107078039927,
+      "grad_norm": 1.0334484577178955,
+      "learning_rate": 0.0002,
+      "loss": 0.2732,
+      "step": 10870
+    },
+    {
+      "epoch": 7.898366606170599,
+      "grad_norm": 1.2068367004394531,
+      "learning_rate": 0.0002,
+      "loss": 0.3206,
+      "step": 10880
+    },
+    {
+      "epoch": 7.90562613430127,
+      "grad_norm": 1.3105504512786865,
+      "learning_rate": 0.0002,
+      "loss": 0.2936,
+      "step": 10890
+    },
+    {
+      "epoch": 7.9128856624319415,
+      "grad_norm": 1.2941272258758545,
+      "learning_rate": 0.0002,
+      "loss": 0.3063,
+      "step": 10900
+    },
+    {
+      "epoch": 7.9201451905626135,
+      "grad_norm": 1.2809823751449585,
+      "learning_rate": 0.0002,
+      "loss": 0.2862,
+      "step": 10910
+    },
+    {
+      "epoch": 7.927404718693285,
+      "grad_norm": 1.5362727642059326,
+      "learning_rate": 0.0002,
+      "loss": 0.3202,
+      "step": 10920
+    },
+    {
+      "epoch": 7.9346642468239565,
+      "grad_norm": 1.5019413232803345,
+      "learning_rate": 0.0002,
+      "loss": 0.3005,
+      "step": 10930
+    },
+    {
+      "epoch": 7.941923774954628,
+      "grad_norm": 1.5947920083999634,
+      "learning_rate": 0.0002,
+      "loss": 0.2976,
+      "step": 10940
+    },
+    {
+      "epoch": 7.9491833030853,
+      "grad_norm": 1.9482372999191284,
+      "learning_rate": 0.0002,
+      "loss": 0.2976,
+      "step": 10950
+    },
+    {
+      "epoch": 7.956442831215971,
+      "grad_norm": 1.8445630073547363,
+      "learning_rate": 0.0002,
+      "loss": 0.3297,
+      "step": 10960
+    },
+    {
+      "epoch": 7.963702359346643,
+      "grad_norm": 1.4342153072357178,
+      "learning_rate": 0.0002,
+      "loss": 0.2843,
+      "step": 10970
+    },
+    {
+      "epoch": 7.970961887477314,
+      "grad_norm": 1.3202505111694336,
+      "learning_rate": 0.0002,
+      "loss": 0.3066,
+      "step": 10980
+    },
+    {
+      "epoch": 7.978221415607986,
+      "grad_norm": 1.186015009880066,
+      "learning_rate": 0.0002,
+      "loss": 0.2785,
+      "step": 10990
+    },
+    {
+      "epoch": 7.985480943738657,
+      "grad_norm": 1.2714571952819824,
+      "learning_rate": 0.0002,
+      "loss": 0.324,
+      "step": 11000
+    },
+    {
+      "epoch": 7.992740471869329,
+      "grad_norm": 0.9723673462867737,
+      "learning_rate": 0.0002,
+      "loss": 0.2795,
+      "step": 11010
+    },
+    {
+      "epoch": 7.997096188747731,
+      "eval_loss": 1.7563573122024536,
+      "eval_runtime": 46.2097,
+      "eval_samples_per_second": 9.435,
+      "eval_steps_per_second": 1.19,
+      "step": 11016
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11016,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.6582302183981056e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..140bceabdec44fc473842cb9ace4a62a4dc9fe89
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-11016/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4abec5480aff960b5a7b03a48be1cdb1ae29ba8792469c451f5c2f269e4be93b
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..50be09609fd83716946490443d34010d2770b093
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94721f474e700335b4e7d287e4ace795275fe6ce10243090004d47de0d72cc6f
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6b21fe839855af6807a65cfe584b243175fdf6e2
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff779cf41c9d8ba5d9c69b4f9dab3d83650430b82da4bc81d777594ea2625626
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a1dd6057ad2a35105ac4f8a2d7298a68767ebd70
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c0c855d1585015f23e10dfd2b87b895f7082d0deb7de58a335a2c4c760766ac
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7e248b8433174e9b822c1813f26e4a98349a0b55
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fc8e94bd1f293c364bb03428d8b8836c3baea7e161d96b587666edd284ae6bd
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..44bc92c6a39237e8538a7b3d9491d5cd66170d7f
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/trainer_state.json
@@ -0,0 +1,1000 @@
+{
+  "best_metric": 1.1381088495254517,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377",
+  "epoch": 0.9996370235934664,
+  "eval_steps": 10,
+  "global_step": 1377,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007259528130671506,
+      "grad_norm": 0.46010470390319824,
+      "learning_rate": 0.0002,
+      "loss": 2.4936,
+      "step": 10
+    },
+    {
+      "epoch": 0.014519056261343012,
+      "grad_norm": 0.5103179216384888,
+      "learning_rate": 0.0002,
+      "loss": 1.8442,
+      "step": 20
+    },
+    {
+      "epoch": 0.021778584392014518,
+      "grad_norm": 0.6282716989517212,
+      "learning_rate": 0.0002,
+      "loss": 1.559,
+      "step": 30
+    },
+    {
+      "epoch": 0.029038112522686024,
+      "grad_norm": 1.2480497360229492,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 40
+    },
+    {
+      "epoch": 0.036297640653357534,
+      "grad_norm": 0.4114855229854584,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 50
+    },
+    {
+      "epoch": 0.043557168784029036,
+      "grad_norm": 0.49482840299606323,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 60
+    },
+    {
+      "epoch": 0.050816696914700546,
+      "grad_norm": 0.4536272883415222,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 70
+    },
+    {
+      "epoch": 0.05807622504537205,
+      "grad_norm": 0.32328274846076965,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 80
+    },
+    {
+      "epoch": 0.06533575317604355,
+      "grad_norm": 0.40990468859672546,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 90
+    },
+    {
+      "epoch": 0.07259528130671507,
+      "grad_norm": 0.37273502349853516,
+      "learning_rate": 0.0002,
+      "loss": 1.2216,
+      "step": 100
+    },
+    {
+      "epoch": 0.07985480943738657,
+      "grad_norm": 0.3903871476650238,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 110
+    },
+    {
+      "epoch": 0.08711433756805807,
+      "grad_norm": 0.3272787928581238,
+      "learning_rate": 0.0002,
+      "loss": 1.3718,
+      "step": 120
+    },
+    {
+      "epoch": 0.09437386569872959,
+      "grad_norm": 0.3622824251651764,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 130
+    },
+    {
+      "epoch": 0.10163339382940109,
+      "grad_norm": 0.3503916561603546,
+      "learning_rate": 0.0002,
+      "loss": 1.2966,
+      "step": 140
+    },
+    {
+      "epoch": 0.1088929219600726,
+      "grad_norm": 0.32787832617759705,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 150
+    },
+    {
+      "epoch": 0.1161524500907441,
+      "grad_norm": 0.5822657942771912,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 160
+    },
+    {
+      "epoch": 0.12341197822141561,
+      "grad_norm": 0.28028249740600586,
+      "learning_rate": 0.0002,
+      "loss": 1.2091,
+      "step": 170
+    },
+    {
+      "epoch": 0.1306715063520871,
+      "grad_norm": 0.33602750301361084,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 180
+    },
+    {
+      "epoch": 0.13793103448275862,
+      "grad_norm": 0.29106274247169495,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 190
+    },
+    {
+      "epoch": 0.14519056261343014,
+      "grad_norm": 0.38753363490104675,
+      "learning_rate": 0.0002,
+      "loss": 1.3261,
+      "step": 200
+    },
+    {
+      "epoch": 0.15245009074410162,
+      "grad_norm": 0.361009418964386,
+      "learning_rate": 0.0002,
+      "loss": 1.1536,
+      "step": 210
+    },
+    {
+      "epoch": 0.15970961887477314,
+      "grad_norm": 0.6743836402893066,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 220
+    },
+    {
+      "epoch": 0.16696914700544466,
+      "grad_norm": 0.3923613727092743,
+      "learning_rate": 0.0002,
+      "loss": 1.264,
+      "step": 230
+    },
+    {
+      "epoch": 0.17422867513611615,
+      "grad_norm": 0.2809699773788452,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 240
+    },
+    {
+      "epoch": 0.18148820326678766,
+      "grad_norm": 0.3631494641304016,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 250
+    },
+    {
+      "epoch": 0.18874773139745918,
+      "grad_norm": 0.24658171832561493,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 260
+    },
+    {
+      "epoch": 0.19600725952813067,
+      "grad_norm": 0.5780664682388306,
+      "learning_rate": 0.0002,
+      "loss": 1.0065,
+      "step": 270
+    },
+    {
+      "epoch": 0.20326678765880218,
+      "grad_norm": 0.3056720495223999,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 280
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 0.2857084572315216,
+      "learning_rate": 0.0002,
+      "loss": 1.1503,
+      "step": 290
+    },
+    {
+      "epoch": 0.2177858439201452,
+      "grad_norm": 0.6645345687866211,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 300
+    },
+    {
+      "epoch": 0.2250453720508167,
+      "grad_norm": 0.5966078639030457,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 310
+    },
+    {
+      "epoch": 0.2323049001814882,
+      "grad_norm": 0.40937140583992004,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 320
+    },
+    {
+      "epoch": 0.2395644283121597,
+      "grad_norm": 0.5642806887626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 330
+    },
+    {
+      "epoch": 0.24682395644283123,
+      "grad_norm": 0.2750748097896576,
+      "learning_rate": 0.0002,
+      "loss": 1.1703,
+      "step": 340
+    },
+    {
+      "epoch": 0.2540834845735027,
+      "grad_norm": 0.34350234270095825,
+      "learning_rate": 0.0002,
+      "loss": 1.2598,
+      "step": 350
+    },
+    {
+      "epoch": 0.2613430127041742,
+      "grad_norm": 0.6767239570617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1942,
+      "step": 360
+    },
+    {
+      "epoch": 0.26860254083484575,
+      "grad_norm": 0.31006959080696106,
+      "learning_rate": 0.0002,
+      "loss": 1.1436,
+      "step": 370
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "grad_norm": 0.3825474679470062,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 380
+    },
+    {
+      "epoch": 0.2831215970961887,
+      "grad_norm": 0.30241551995277405,
+      "learning_rate": 0.0002,
+      "loss": 1.1523,
+      "step": 390
+    },
+    {
+      "epoch": 0.29038112522686027,
+      "grad_norm": 0.2962397336959839,
+      "learning_rate": 0.0002,
+      "loss": 1.1298,
+      "step": 400
+    },
+    {
+      "epoch": 0.29764065335753176,
+      "grad_norm": 0.2600369155406952,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 410
+    },
+    {
+      "epoch": 0.30490018148820325,
+      "grad_norm": 0.3675060272216797,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 420
+    },
+    {
+      "epoch": 0.3121597096188748,
+      "grad_norm": 0.3429498076438904,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 430
+    },
+    {
+      "epoch": 0.3194192377495463,
+      "grad_norm": 0.34311825037002563,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 440
+    },
+    {
+      "epoch": 0.32667876588021777,
+      "grad_norm": 0.37872210144996643,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 450
+    },
+    {
+      "epoch": 0.3339382940108893,
+      "grad_norm": 0.33271121978759766,
+      "learning_rate": 0.0002,
+      "loss": 1.1365,
+      "step": 460
+    },
+    {
+      "epoch": 0.3411978221415608,
+      "grad_norm": 0.34605276584625244,
+      "learning_rate": 0.0002,
+      "loss": 1.1572,
+      "step": 470
+    },
+    {
+      "epoch": 0.3484573502722323,
+      "grad_norm": 0.41050562262535095,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 480
+    },
+    {
+      "epoch": 0.35571687840290384,
+      "grad_norm": 0.2066836953163147,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 490
+    },
+    {
+      "epoch": 0.3629764065335753,
+      "grad_norm": 0.2859014868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.2072,
+      "step": 500
+    },
+    {
+      "epoch": 0.3702359346642468,
+      "grad_norm": 0.28763777017593384,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 510
+    },
+    {
+      "epoch": 0.37749546279491836,
+      "grad_norm": 0.2730471193790436,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 520
+    },
+    {
+      "epoch": 0.38475499092558985,
+      "grad_norm": 0.3968936800956726,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 530
+    },
+    {
+      "epoch": 0.39201451905626133,
+      "grad_norm": 0.3624701201915741,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 540
+    },
+    {
+      "epoch": 0.3992740471869328,
+      "grad_norm": 0.3303608298301697,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 550
+    },
+    {
+      "epoch": 0.40653357531760437,
+      "grad_norm": 0.33507466316223145,
+      "learning_rate": 0.0002,
+      "loss": 1.203,
+      "step": 560
+    },
+    {
+      "epoch": 0.41379310344827586,
+      "grad_norm": 0.3297670781612396,
+      "learning_rate": 0.0002,
+      "loss": 1.2294,
+      "step": 570
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.32334890961647034,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 580
+    },
+    {
+      "epoch": 0.4283121597096189,
+      "grad_norm": 0.30281195044517517,
+      "learning_rate": 0.0002,
+      "loss": 1.112,
+      "step": 590
+    },
+    {
+      "epoch": 0.4355716878402904,
+      "grad_norm": 0.5900027751922607,
+      "learning_rate": 0.0002,
+      "loss": 1.1352,
+      "step": 600
+    },
+    {
+      "epoch": 0.44283121597096187,
+      "grad_norm": 0.28437477350234985,
+      "learning_rate": 0.0002,
+      "loss": 1.1575,
+      "step": 610
+    },
+    {
+      "epoch": 0.4500907441016334,
+      "grad_norm": 0.39601704478263855,
+      "learning_rate": 0.0002,
+      "loss": 1.2009,
+      "step": 620
+    },
+    {
+      "epoch": 0.4573502722323049,
+      "grad_norm": 0.41971510648727417,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 630
+    },
+    {
+      "epoch": 0.4646098003629764,
+      "grad_norm": 0.33814409375190735,
+      "learning_rate": 0.0002,
+      "loss": 1.236,
+      "step": 640
+    },
+    {
+      "epoch": 0.47186932849364793,
+      "grad_norm": 0.575718104839325,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 650
+    },
+    {
+      "epoch": 0.4791288566243194,
+      "grad_norm": 0.37927401065826416,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 660
+    },
+    {
+      "epoch": 0.4863883847549909,
+      "grad_norm": 0.3224332630634308,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 670
+    },
+    {
+      "epoch": 0.49364791288566245,
+      "grad_norm": 0.32683515548706055,
+      "learning_rate": 0.0002,
+      "loss": 1.1354,
+      "step": 680
+    },
+    {
+      "epoch": 0.5009074410163339,
+      "grad_norm": 0.4316163659095764,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 690
+    },
+    {
+      "epoch": 0.5081669691470054,
+      "grad_norm": 0.342602401971817,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 700
+    },
+    {
+      "epoch": 0.515426497277677,
+      "grad_norm": 0.2794898748397827,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 710
+    },
+    {
+      "epoch": 0.5226860254083484,
+      "grad_norm": 0.3322339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.2395,
+      "step": 720
+    },
+    {
+      "epoch": 0.52994555353902,
+      "grad_norm": 0.3088509142398834,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 730
+    },
+    {
+      "epoch": 0.5372050816696915,
+      "grad_norm": 0.24444378912448883,
+      "learning_rate": 0.0002,
+      "loss": 1.1495,
+      "step": 740
+    },
+    {
+      "epoch": 0.5444646098003629,
+      "grad_norm": 0.3483171761035919,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 750
+    },
+    {
+      "epoch": 0.5517241379310345,
+      "grad_norm": 0.3425690531730652,
+      "learning_rate": 0.0002,
+      "loss": 1.2248,
+      "step": 760
+    },
+    {
+      "epoch": 0.558983666061706,
+      "grad_norm": 0.31841927766799927,
+      "learning_rate": 0.0002,
+      "loss": 1.1127,
+      "step": 770
+    },
+    {
+      "epoch": 0.5662431941923775,
+      "grad_norm": 0.39423868060112,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 780
+    },
+    {
+      "epoch": 0.573502722323049,
+      "grad_norm": 0.30328479409217834,
+      "learning_rate": 0.0002,
+      "loss": 1.0583,
+      "step": 790
+    },
+    {
+      "epoch": 0.5807622504537205,
+      "grad_norm": 0.24475938081741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0718,
+      "step": 800
+    },
+    {
+      "epoch": 0.588021778584392,
+      "grad_norm": 0.37132805585861206,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 810
+    },
+    {
+      "epoch": 0.5952813067150635,
+      "grad_norm": 0.32195979356765747,
+      "learning_rate": 0.0002,
+      "loss": 1.1544,
+      "step": 820
+    },
+    {
+      "epoch": 0.6025408348457351,
+      "grad_norm": 0.2848738729953766,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 830
+    },
+    {
+      "epoch": 0.6098003629764065,
+      "grad_norm": 0.28015264868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.0937,
+      "step": 840
+    },
+    {
+      "epoch": 0.617059891107078,
+      "grad_norm": 0.37796008586883545,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 850
+    },
+    {
+      "epoch": 0.6243194192377496,
+      "grad_norm": 0.39311841130256653,
+      "learning_rate": 0.0002,
+      "loss": 1.1455,
+      "step": 860
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.2761685252189636,
+      "learning_rate": 0.0002,
+      "loss": 1.1066,
+      "step": 870
+    },
+    {
+      "epoch": 0.6388384754990926,
+      "grad_norm": 0.3826720118522644,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 880
+    },
+    {
+      "epoch": 0.6460980036297641,
+      "grad_norm": 0.30076679587364197,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 890
+    },
+    {
+      "epoch": 0.6533575317604355,
+      "grad_norm": 0.21997687220573425,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 900
+    },
+    {
+      "epoch": 0.6606170598911071,
+      "grad_norm": 0.32593777775764465,
+      "learning_rate": 0.0002,
+      "loss": 1.1603,
+      "step": 910
+    },
+    {
+      "epoch": 0.6678765880217786,
+      "grad_norm": 0.30347898602485657,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 920
+    },
+    {
+      "epoch": 0.6751361161524501,
+      "grad_norm": 0.44173774123191833,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 930
+    },
+    {
+      "epoch": 0.6823956442831216,
+      "grad_norm": 0.2507467269897461,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 940
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.29463833570480347,
+      "learning_rate": 0.0002,
+      "loss": 1.3001,
+      "step": 950
+    },
+    {
+      "epoch": 0.6969147005444646,
+      "grad_norm": 0.9363154172897339,
+      "learning_rate": 0.0002,
+      "loss": 1.1651,
+      "step": 960
+    },
+    {
+      "epoch": 0.7041742286751361,
+      "grad_norm": 0.3236212134361267,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 970
+    },
+    {
+      "epoch": 0.7114337568058077,
+      "grad_norm": 0.3123254179954529,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 980
+    },
+    {
+      "epoch": 0.7186932849364791,
+      "grad_norm": 0.3395805060863495,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 990
+    },
+    {
+      "epoch": 0.7259528130671506,
+      "grad_norm": 0.3240964412689209,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7332123411978222,
+      "grad_norm": 0.31902948021888733,
+      "learning_rate": 0.0002,
+      "loss": 1.0313,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7404718693284936,
+      "grad_norm": 0.4848408102989197,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7477313974591652,
+      "grad_norm": 0.33006033301353455,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7549909255898367,
+      "grad_norm": 0.2928730547428131,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7622504537205081,
+      "grad_norm": 0.3529164791107178,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7695099818511797,
+      "grad_norm": 0.2736213803291321,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7767695099818511,
+      "grad_norm": 0.7200686931610107,
+      "learning_rate": 0.0002,
+      "loss": 1.1169,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7840290381125227,
+      "grad_norm": 0.33396708965301514,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7912885662431942,
+      "grad_norm": 1.5760449171066284,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7985480943738656,
+      "grad_norm": 0.28138381242752075,
+      "learning_rate": 0.0002,
+      "loss": 1.1126,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8058076225045372,
+      "grad_norm": 0.2597472369670868,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8130671506352087,
+      "grad_norm": 0.3305445611476898,
+      "learning_rate": 0.0002,
+      "loss": 1.1177,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8203266787658802,
+      "grad_norm": 0.3934599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.0849,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8275862068965517,
+      "grad_norm": 0.3472191393375397,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8348457350272233,
+      "grad_norm": 0.2857365906238556,
+      "learning_rate": 0.0002,
+      "loss": 1.265,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.3207702934741974,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8493647912885662,
+      "grad_norm": 0.3176484704017639,
+      "learning_rate": 0.0002,
+      "loss": 1.1669,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8566243194192378,
+      "grad_norm": 0.40685558319091797,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8638838475499092,
+      "grad_norm": 0.31125199794769287,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8711433756805808,
+      "grad_norm": 0.7361181378364563,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8784029038112523,
+      "grad_norm": 0.33699527382850647,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8856624319419237,
+      "grad_norm": 0.3315220773220062,
+      "learning_rate": 0.0002,
+      "loss": 1.11,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8929219600725953,
+      "grad_norm": 0.6256054043769836,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9001814882032668,
+      "grad_norm": 0.3692137897014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1738,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9074410163339383,
+      "grad_norm": 0.3538484573364258,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9147005444646098,
+      "grad_norm": 0.27863040566444397,
+      "learning_rate": 0.0002,
+      "loss": 1.0878,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9219600725952813,
+      "grad_norm": 0.3322528302669525,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9292196007259528,
+      "grad_norm": 0.7553173303604126,
+      "learning_rate": 0.0002,
+      "loss": 1.0158,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9364791288566243,
+      "grad_norm": 0.4856191575527191,
+      "learning_rate": 0.0002,
+      "loss": 1.198,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9437386569872959,
+      "grad_norm": 0.3668074905872345,
+      "learning_rate": 0.0002,
+      "loss": 1.1678,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9509981851179673,
+      "grad_norm": 0.29851067066192627,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9582577132486388,
+      "grad_norm": 0.276664674282074,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9655172413793104,
+      "grad_norm": 0.2941018342971802,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9727767695099818,
+      "grad_norm": 0.3505859076976776,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9800362976406534,
+      "grad_norm": 0.3067687451839447,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9872958257713249,
+      "grad_norm": 0.27151066064834595,
+      "learning_rate": 0.0002,
+      "loss": 1.1511,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9945553539019963,
+      "grad_norm": 0.36370083689689636,
+      "learning_rate": 0.0002,
+      "loss": 1.1919,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9996370235934664,
+      "eval_loss": 1.1381088495254517,
+      "eval_runtime": 96.0848,
+      "eval_samples_per_second": 4.538,
+      "eval_steps_per_second": 0.572,
+      "step": 1377
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11016,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 7.07535596027904e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..140bceabdec44fc473842cb9ace4a62a4dc9fe89
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4abec5480aff960b5a7b03a48be1cdb1ae29ba8792469c451f5c2f269e4be93b
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..57c0547536cf633208bd0753b6f45625ae2084ca
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:160428ee21539129b32c4f50e46b1c3c4e1b361d3d8a42304e1e889b4311be3f
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2ec8fef898c8a9d69fd77cbd4c39bcaab6095ea7
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6219d7e87d25414277667ab600b4628a4df624a52ae01b039c4d2501eb65a758
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..11de8bffda431939f1acb049557abb89aa5f64c3
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b622e5fc2fcbc23cb0255cd1db1954304c4d3fd922f5914715c3f7cbf19176a
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..fb63a029f86fa9cb62df58e019414d8be905de14
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:736497d6fe23b956ec574624f614b5057faee2badff1718a5f7fc886a837e97e
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..e11b3edc57e74e2ec6e2734caa22f66ef6e95f84
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/trainer_state.json
@@ -0,0 +1,1974 @@
+{
+  "best_metric": 1.1086540222167969,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 2755,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007259528130671506,
+      "grad_norm": 0.46010470390319824,
+      "learning_rate": 0.0002,
+      "loss": 2.4936,
+      "step": 10
+    },
+    {
+      "epoch": 0.014519056261343012,
+      "grad_norm": 0.5103179216384888,
+      "learning_rate": 0.0002,
+      "loss": 1.8442,
+      "step": 20
+    },
+    {
+      "epoch": 0.021778584392014518,
+      "grad_norm": 0.6282716989517212,
+      "learning_rate": 0.0002,
+      "loss": 1.559,
+      "step": 30
+    },
+    {
+      "epoch": 0.029038112522686024,
+      "grad_norm": 1.2480497360229492,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 40
+    },
+    {
+      "epoch": 0.036297640653357534,
+      "grad_norm": 0.4114855229854584,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 50
+    },
+    {
+      "epoch": 0.043557168784029036,
+      "grad_norm": 0.49482840299606323,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 60
+    },
+    {
+      "epoch": 0.050816696914700546,
+      "grad_norm": 0.4536272883415222,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 70
+    },
+    {
+      "epoch": 0.05807622504537205,
+      "grad_norm": 0.32328274846076965,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 80
+    },
+    {
+      "epoch": 0.06533575317604355,
+      "grad_norm": 0.40990468859672546,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 90
+    },
+    {
+      "epoch": 0.07259528130671507,
+      "grad_norm": 0.37273502349853516,
+      "learning_rate": 0.0002,
+      "loss": 1.2216,
+      "step": 100
+    },
+    {
+      "epoch": 0.07985480943738657,
+      "grad_norm": 0.3903871476650238,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 110
+    },
+    {
+      "epoch": 0.08711433756805807,
+      "grad_norm": 0.3272787928581238,
+      "learning_rate": 0.0002,
+      "loss": 1.3718,
+      "step": 120
+    },
+    {
+      "epoch": 0.09437386569872959,
+      "grad_norm": 0.3622824251651764,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 130
+    },
+    {
+      "epoch": 0.10163339382940109,
+      "grad_norm": 0.3503916561603546,
+      "learning_rate": 0.0002,
+      "loss": 1.2966,
+      "step": 140
+    },
+    {
+      "epoch": 0.1088929219600726,
+      "grad_norm": 0.32787832617759705,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 150
+    },
+    {
+      "epoch": 0.1161524500907441,
+      "grad_norm": 0.5822657942771912,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 160
+    },
+    {
+      "epoch": 0.12341197822141561,
+      "grad_norm": 0.28028249740600586,
+      "learning_rate": 0.0002,
+      "loss": 1.2091,
+      "step": 170
+    },
+    {
+      "epoch": 0.1306715063520871,
+      "grad_norm": 0.33602750301361084,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 180
+    },
+    {
+      "epoch": 0.13793103448275862,
+      "grad_norm": 0.29106274247169495,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 190
+    },
+    {
+      "epoch": 0.14519056261343014,
+      "grad_norm": 0.38753363490104675,
+      "learning_rate": 0.0002,
+      "loss": 1.3261,
+      "step": 200
+    },
+    {
+      "epoch": 0.15245009074410162,
+      "grad_norm": 0.361009418964386,
+      "learning_rate": 0.0002,
+      "loss": 1.1536,
+      "step": 210
+    },
+    {
+      "epoch": 0.15970961887477314,
+      "grad_norm": 0.6743836402893066,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 220
+    },
+    {
+      "epoch": 0.16696914700544466,
+      "grad_norm": 0.3923613727092743,
+      "learning_rate": 0.0002,
+      "loss": 1.264,
+      "step": 230
+    },
+    {
+      "epoch": 0.17422867513611615,
+      "grad_norm": 0.2809699773788452,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 240
+    },
+    {
+      "epoch": 0.18148820326678766,
+      "grad_norm": 0.3631494641304016,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 250
+    },
+    {
+      "epoch": 0.18874773139745918,
+      "grad_norm": 0.24658171832561493,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 260
+    },
+    {
+      "epoch": 0.19600725952813067,
+      "grad_norm": 0.5780664682388306,
+      "learning_rate": 0.0002,
+      "loss": 1.0065,
+      "step": 270
+    },
+    {
+      "epoch": 0.20326678765880218,
+      "grad_norm": 0.3056720495223999,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 280
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 0.2857084572315216,
+      "learning_rate": 0.0002,
+      "loss": 1.1503,
+      "step": 290
+    },
+    {
+      "epoch": 0.2177858439201452,
+      "grad_norm": 0.6645345687866211,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 300
+    },
+    {
+      "epoch": 0.2250453720508167,
+      "grad_norm": 0.5966078639030457,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 310
+    },
+    {
+      "epoch": 0.2323049001814882,
+      "grad_norm": 0.40937140583992004,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 320
+    },
+    {
+      "epoch": 0.2395644283121597,
+      "grad_norm": 0.5642806887626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 330
+    },
+    {
+      "epoch": 0.24682395644283123,
+      "grad_norm": 0.2750748097896576,
+      "learning_rate": 0.0002,
+      "loss": 1.1703,
+      "step": 340
+    },
+    {
+      "epoch": 0.2540834845735027,
+      "grad_norm": 0.34350234270095825,
+      "learning_rate": 0.0002,
+      "loss": 1.2598,
+      "step": 350
+    },
+    {
+      "epoch": 0.2613430127041742,
+      "grad_norm": 0.6767239570617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1942,
+      "step": 360
+    },
+    {
+      "epoch": 0.26860254083484575,
+      "grad_norm": 0.31006959080696106,
+      "learning_rate": 0.0002,
+      "loss": 1.1436,
+      "step": 370
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "grad_norm": 0.3825474679470062,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 380
+    },
+    {
+      "epoch": 0.2831215970961887,
+      "grad_norm": 0.30241551995277405,
+      "learning_rate": 0.0002,
+      "loss": 1.1523,
+      "step": 390
+    },
+    {
+      "epoch": 0.29038112522686027,
+      "grad_norm": 0.2962397336959839,
+      "learning_rate": 0.0002,
+      "loss": 1.1298,
+      "step": 400
+    },
+    {
+      "epoch": 0.29764065335753176,
+      "grad_norm": 0.2600369155406952,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 410
+    },
+    {
+      "epoch": 0.30490018148820325,
+      "grad_norm": 0.3675060272216797,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 420
+    },
+    {
+      "epoch": 0.3121597096188748,
+      "grad_norm": 0.3429498076438904,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 430
+    },
+    {
+      "epoch": 0.3194192377495463,
+      "grad_norm": 0.34311825037002563,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 440
+    },
+    {
+      "epoch": 0.32667876588021777,
+      "grad_norm": 0.37872210144996643,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 450
+    },
+    {
+      "epoch": 0.3339382940108893,
+      "grad_norm": 0.33271121978759766,
+      "learning_rate": 0.0002,
+      "loss": 1.1365,
+      "step": 460
+    },
+    {
+      "epoch": 0.3411978221415608,
+      "grad_norm": 0.34605276584625244,
+      "learning_rate": 0.0002,
+      "loss": 1.1572,
+      "step": 470
+    },
+    {
+      "epoch": 0.3484573502722323,
+      "grad_norm": 0.41050562262535095,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 480
+    },
+    {
+      "epoch": 0.35571687840290384,
+      "grad_norm": 0.2066836953163147,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 490
+    },
+    {
+      "epoch": 0.3629764065335753,
+      "grad_norm": 0.2859014868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.2072,
+      "step": 500
+    },
+    {
+      "epoch": 0.3702359346642468,
+      "grad_norm": 0.28763777017593384,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 510
+    },
+    {
+      "epoch": 0.37749546279491836,
+      "grad_norm": 0.2730471193790436,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 520
+    },
+    {
+      "epoch": 0.38475499092558985,
+      "grad_norm": 0.3968936800956726,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 530
+    },
+    {
+      "epoch": 0.39201451905626133,
+      "grad_norm": 0.3624701201915741,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 540
+    },
+    {
+      "epoch": 0.3992740471869328,
+      "grad_norm": 0.3303608298301697,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 550
+    },
+    {
+      "epoch": 0.40653357531760437,
+      "grad_norm": 0.33507466316223145,
+      "learning_rate": 0.0002,
+      "loss": 1.203,
+      "step": 560
+    },
+    {
+      "epoch": 0.41379310344827586,
+      "grad_norm": 0.3297670781612396,
+      "learning_rate": 0.0002,
+      "loss": 1.2294,
+      "step": 570
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.32334890961647034,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 580
+    },
+    {
+      "epoch": 0.4283121597096189,
+      "grad_norm": 0.30281195044517517,
+      "learning_rate": 0.0002,
+      "loss": 1.112,
+      "step": 590
+    },
+    {
+      "epoch": 0.4355716878402904,
+      "grad_norm": 0.5900027751922607,
+      "learning_rate": 0.0002,
+      "loss": 1.1352,
+      "step": 600
+    },
+    {
+      "epoch": 0.44283121597096187,
+      "grad_norm": 0.28437477350234985,
+      "learning_rate": 0.0002,
+      "loss": 1.1575,
+      "step": 610
+    },
+    {
+      "epoch": 0.4500907441016334,
+      "grad_norm": 0.39601704478263855,
+      "learning_rate": 0.0002,
+      "loss": 1.2009,
+      "step": 620
+    },
+    {
+      "epoch": 0.4573502722323049,
+      "grad_norm": 0.41971510648727417,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 630
+    },
+    {
+      "epoch": 0.4646098003629764,
+      "grad_norm": 0.33814409375190735,
+      "learning_rate": 0.0002,
+      "loss": 1.236,
+      "step": 640
+    },
+    {
+      "epoch": 0.47186932849364793,
+      "grad_norm": 0.575718104839325,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 650
+    },
+    {
+      "epoch": 0.4791288566243194,
+      "grad_norm": 0.37927401065826416,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 660
+    },
+    {
+      "epoch": 0.4863883847549909,
+      "grad_norm": 0.3224332630634308,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 670
+    },
+    {
+      "epoch": 0.49364791288566245,
+      "grad_norm": 0.32683515548706055,
+      "learning_rate": 0.0002,
+      "loss": 1.1354,
+      "step": 680
+    },
+    {
+      "epoch": 0.5009074410163339,
+      "grad_norm": 0.4316163659095764,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 690
+    },
+    {
+      "epoch": 0.5081669691470054,
+      "grad_norm": 0.342602401971817,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 700
+    },
+    {
+      "epoch": 0.515426497277677,
+      "grad_norm": 0.2794898748397827,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 710
+    },
+    {
+      "epoch": 0.5226860254083484,
+      "grad_norm": 0.3322339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.2395,
+      "step": 720
+    },
+    {
+      "epoch": 0.52994555353902,
+      "grad_norm": 0.3088509142398834,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 730
+    },
+    {
+      "epoch": 0.5372050816696915,
+      "grad_norm": 0.24444378912448883,
+      "learning_rate": 0.0002,
+      "loss": 1.1495,
+      "step": 740
+    },
+    {
+      "epoch": 0.5444646098003629,
+      "grad_norm": 0.3483171761035919,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 750
+    },
+    {
+      "epoch": 0.5517241379310345,
+      "grad_norm": 0.3425690531730652,
+      "learning_rate": 0.0002,
+      "loss": 1.2248,
+      "step": 760
+    },
+    {
+      "epoch": 0.558983666061706,
+      "grad_norm": 0.31841927766799927,
+      "learning_rate": 0.0002,
+      "loss": 1.1127,
+      "step": 770
+    },
+    {
+      "epoch": 0.5662431941923775,
+      "grad_norm": 0.39423868060112,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 780
+    },
+    {
+      "epoch": 0.573502722323049,
+      "grad_norm": 0.30328479409217834,
+      "learning_rate": 0.0002,
+      "loss": 1.0583,
+      "step": 790
+    },
+    {
+      "epoch": 0.5807622504537205,
+      "grad_norm": 0.24475938081741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0718,
+      "step": 800
+    },
+    {
+      "epoch": 0.588021778584392,
+      "grad_norm": 0.37132805585861206,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 810
+    },
+    {
+      "epoch": 0.5952813067150635,
+      "grad_norm": 0.32195979356765747,
+      "learning_rate": 0.0002,
+      "loss": 1.1544,
+      "step": 820
+    },
+    {
+      "epoch": 0.6025408348457351,
+      "grad_norm": 0.2848738729953766,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 830
+    },
+    {
+      "epoch": 0.6098003629764065,
+      "grad_norm": 0.28015264868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.0937,
+      "step": 840
+    },
+    {
+      "epoch": 0.617059891107078,
+      "grad_norm": 0.37796008586883545,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 850
+    },
+    {
+      "epoch": 0.6243194192377496,
+      "grad_norm": 0.39311841130256653,
+      "learning_rate": 0.0002,
+      "loss": 1.1455,
+      "step": 860
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.2761685252189636,
+      "learning_rate": 0.0002,
+      "loss": 1.1066,
+      "step": 870
+    },
+    {
+      "epoch": 0.6388384754990926,
+      "grad_norm": 0.3826720118522644,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 880
+    },
+    {
+      "epoch": 0.6460980036297641,
+      "grad_norm": 0.30076679587364197,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 890
+    },
+    {
+      "epoch": 0.6533575317604355,
+      "grad_norm": 0.21997687220573425,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 900
+    },
+    {
+      "epoch": 0.6606170598911071,
+      "grad_norm": 0.32593777775764465,
+      "learning_rate": 0.0002,
+      "loss": 1.1603,
+      "step": 910
+    },
+    {
+      "epoch": 0.6678765880217786,
+      "grad_norm": 0.30347898602485657,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 920
+    },
+    {
+      "epoch": 0.6751361161524501,
+      "grad_norm": 0.44173774123191833,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 930
+    },
+    {
+      "epoch": 0.6823956442831216,
+      "grad_norm": 0.2507467269897461,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 940
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.29463833570480347,
+      "learning_rate": 0.0002,
+      "loss": 1.3001,
+      "step": 950
+    },
+    {
+      "epoch": 0.6969147005444646,
+      "grad_norm": 0.9363154172897339,
+      "learning_rate": 0.0002,
+      "loss": 1.1651,
+      "step": 960
+    },
+    {
+      "epoch": 0.7041742286751361,
+      "grad_norm": 0.3236212134361267,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 970
+    },
+    {
+      "epoch": 0.7114337568058077,
+      "grad_norm": 0.3123254179954529,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 980
+    },
+    {
+      "epoch": 0.7186932849364791,
+      "grad_norm": 0.3395805060863495,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 990
+    },
+    {
+      "epoch": 0.7259528130671506,
+      "grad_norm": 0.3240964412689209,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7332123411978222,
+      "grad_norm": 0.31902948021888733,
+      "learning_rate": 0.0002,
+      "loss": 1.0313,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7404718693284936,
+      "grad_norm": 0.4848408102989197,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7477313974591652,
+      "grad_norm": 0.33006033301353455,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7549909255898367,
+      "grad_norm": 0.2928730547428131,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7622504537205081,
+      "grad_norm": 0.3529164791107178,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7695099818511797,
+      "grad_norm": 0.2736213803291321,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7767695099818511,
+      "grad_norm": 0.7200686931610107,
+      "learning_rate": 0.0002,
+      "loss": 1.1169,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7840290381125227,
+      "grad_norm": 0.33396708965301514,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7912885662431942,
+      "grad_norm": 1.5760449171066284,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7985480943738656,
+      "grad_norm": 0.28138381242752075,
+      "learning_rate": 0.0002,
+      "loss": 1.1126,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8058076225045372,
+      "grad_norm": 0.2597472369670868,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8130671506352087,
+      "grad_norm": 0.3305445611476898,
+      "learning_rate": 0.0002,
+      "loss": 1.1177,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8203266787658802,
+      "grad_norm": 0.3934599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.0849,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8275862068965517,
+      "grad_norm": 0.3472191393375397,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8348457350272233,
+      "grad_norm": 0.2857365906238556,
+      "learning_rate": 0.0002,
+      "loss": 1.265,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.3207702934741974,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8493647912885662,
+      "grad_norm": 0.3176484704017639,
+      "learning_rate": 0.0002,
+      "loss": 1.1669,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8566243194192378,
+      "grad_norm": 0.40685558319091797,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8638838475499092,
+      "grad_norm": 0.31125199794769287,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8711433756805808,
+      "grad_norm": 0.7361181378364563,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8784029038112523,
+      "grad_norm": 0.33699527382850647,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8856624319419237,
+      "grad_norm": 0.3315220773220062,
+      "learning_rate": 0.0002,
+      "loss": 1.11,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8929219600725953,
+      "grad_norm": 0.6256054043769836,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9001814882032668,
+      "grad_norm": 0.3692137897014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1738,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9074410163339383,
+      "grad_norm": 0.3538484573364258,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9147005444646098,
+      "grad_norm": 0.27863040566444397,
+      "learning_rate": 0.0002,
+      "loss": 1.0878,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9219600725952813,
+      "grad_norm": 0.3322528302669525,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9292196007259528,
+      "grad_norm": 0.7553173303604126,
+      "learning_rate": 0.0002,
+      "loss": 1.0158,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9364791288566243,
+      "grad_norm": 0.4856191575527191,
+      "learning_rate": 0.0002,
+      "loss": 1.198,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9437386569872959,
+      "grad_norm": 0.3668074905872345,
+      "learning_rate": 0.0002,
+      "loss": 1.1678,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9509981851179673,
+      "grad_norm": 0.29851067066192627,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9582577132486388,
+      "grad_norm": 0.276664674282074,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9655172413793104,
+      "grad_norm": 0.2941018342971802,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9727767695099818,
+      "grad_norm": 0.3505859076976776,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9800362976406534,
+      "grad_norm": 0.3067687451839447,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9872958257713249,
+      "grad_norm": 0.27151066064834595,
+      "learning_rate": 0.0002,
+      "loss": 1.1511,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9945553539019963,
+      "grad_norm": 0.36370083689689636,
+      "learning_rate": 0.0002,
+      "loss": 1.1919,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9996370235934664,
+      "eval_loss": 1.1381088495254517,
+      "eval_runtime": 96.0848,
+      "eval_samples_per_second": 4.538,
+      "eval_steps_per_second": 0.572,
+      "step": 1377
+    },
+    {
+      "epoch": 1.0018148820326678,
+      "grad_norm": 0.27980583906173706,
+      "learning_rate": 0.0002,
+      "loss": 1.0082,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0090744101633393,
+      "grad_norm": 0.26713913679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0163339382940109,
+      "grad_norm": 0.3089541494846344,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0235934664246824,
+      "grad_norm": 0.4188242256641388,
+      "learning_rate": 0.0002,
+      "loss": 0.9649,
+      "step": 1410
+    },
+    {
+      "epoch": 1.030852994555354,
+      "grad_norm": 0.5246463418006897,
+      "learning_rate": 0.0002,
+      "loss": 0.9821,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0381125226860255,
+      "grad_norm": 0.2728777825832367,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0453720508166968,
+      "grad_norm": 0.38167616724967957,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.4439380168914795,
+      "learning_rate": 0.0002,
+      "loss": 1.0229,
+      "step": 1450
+    },
+    {
+      "epoch": 1.05989110707804,
+      "grad_norm": 0.30954182147979736,
+      "learning_rate": 0.0002,
+      "loss": 1.0451,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0671506352087115,
+      "grad_norm": 0.4022280275821686,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 1470
+    },
+    {
+      "epoch": 1.074410163339383,
+      "grad_norm": 0.7390811443328857,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0816696914700545,
+      "grad_norm": 0.3885486423969269,
+      "learning_rate": 0.0002,
+      "loss": 1.1378,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0889292196007259,
+      "grad_norm": 0.5275560617446899,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0961887477313974,
+      "grad_norm": 0.35112282633781433,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1510
+    },
+    {
+      "epoch": 1.103448275862069,
+      "grad_norm": 0.33714351058006287,
+      "learning_rate": 0.0002,
+      "loss": 1.0331,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1107078039927405,
+      "grad_norm": 0.31221693754196167,
+      "learning_rate": 0.0002,
+      "loss": 0.8723,
+      "step": 1530
+    },
+    {
+      "epoch": 1.117967332123412,
+      "grad_norm": 0.27549654245376587,
+      "learning_rate": 0.0002,
+      "loss": 1.0701,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1252268602540836,
+      "grad_norm": 0.8465521335601807,
+      "learning_rate": 0.0002,
+      "loss": 1.0114,
+      "step": 1550
+    },
+    {
+      "epoch": 1.132486388384755,
+      "grad_norm": 0.36125949025154114,
+      "learning_rate": 0.0002,
+      "loss": 0.9643,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1397459165154264,
+      "grad_norm": 0.37420371174812317,
+      "learning_rate": 0.0002,
+      "loss": 1.0744,
+      "step": 1570
+    },
+    {
+      "epoch": 1.147005444646098,
+      "grad_norm": 0.3294760584831238,
+      "learning_rate": 0.0002,
+      "loss": 1.165,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1542649727767695,
+      "grad_norm": 0.3881238102912903,
+      "learning_rate": 0.0002,
+      "loss": 0.9207,
+      "step": 1590
+    },
+    {
+      "epoch": 1.161524500907441,
+      "grad_norm": 0.4766491651535034,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1687840290381124,
+      "grad_norm": 0.395530104637146,
+      "learning_rate": 0.0002,
+      "loss": 1.0092,
+      "step": 1610
+    },
+    {
+      "epoch": 1.176043557168784,
+      "grad_norm": 0.3297106623649597,
+      "learning_rate": 0.0002,
+      "loss": 0.9895,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1833030852994555,
+      "grad_norm": 0.39528271555900574,
+      "learning_rate": 0.0002,
+      "loss": 1.1121,
+      "step": 1630
+    },
+    {
+      "epoch": 1.190562613430127,
+      "grad_norm": 0.3370221257209778,
+      "learning_rate": 0.0002,
+      "loss": 0.9202,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1978221415607986,
+      "grad_norm": 0.31922030448913574,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2050816696914701,
+      "grad_norm": 0.6142027378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2123411978221417,
+      "grad_norm": 0.44769710302352905,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 1670
+    },
+    {
+      "epoch": 1.219600725952813,
+      "grad_norm": 0.41233646869659424,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2268602540834845,
+      "grad_norm": 0.2928866147994995,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1690
+    },
+    {
+      "epoch": 1.234119782214156,
+      "grad_norm": 0.36913734674453735,
+      "learning_rate": 0.0002,
+      "loss": 0.9074,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2413793103448276,
+      "grad_norm": 0.5281891226768494,
+      "learning_rate": 0.0002,
+      "loss": 1.0847,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2486388384754992,
+      "grad_norm": 0.3374697268009186,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2558983666061705,
+      "grad_norm": 0.3802020847797394,
+      "learning_rate": 0.0002,
+      "loss": 0.9826,
+      "step": 1730
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.38048651814460754,
+      "learning_rate": 0.0002,
+      "loss": 1.0191,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2704174228675136,
+      "grad_norm": 0.6676169633865356,
+      "learning_rate": 0.0002,
+      "loss": 1.0142,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2776769509981851,
+      "grad_norm": 0.4075961410999298,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2849364791288567,
+      "grad_norm": 0.4374721348285675,
+      "learning_rate": 0.0002,
+      "loss": 1.0301,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2921960072595282,
+      "grad_norm": 0.4638824164867401,
+      "learning_rate": 0.0002,
+      "loss": 1.0601,
+      "step": 1780
+    },
+    {
+      "epoch": 1.2994555353901998,
+      "grad_norm": 0.38631564378738403,
+      "learning_rate": 0.0002,
+      "loss": 0.9593,
+      "step": 1790
+    },
+    {
+      "epoch": 1.306715063520871,
+      "grad_norm": 0.35873809456825256,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3139745916515426,
+      "grad_norm": 0.33602237701416016,
+      "learning_rate": 0.0002,
+      "loss": 0.9835,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3212341197822142,
+      "grad_norm": 0.46696463227272034,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3284936479128857,
+      "grad_norm": 0.368958979845047,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 1830
+    },
+    {
+      "epoch": 1.335753176043557,
+      "grad_norm": 0.3160957396030426,
+      "learning_rate": 0.0002,
+      "loss": 1.0456,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3430127041742286,
+      "grad_norm": 0.4511511027812958,
+      "learning_rate": 0.0002,
+      "loss": 1.0264,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3502722323049001,
+      "grad_norm": 0.6769845485687256,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3575317604355717,
+      "grad_norm": 0.3749309480190277,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3647912885662432,
+      "grad_norm": 0.553798496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.0223,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3720508166969148,
+      "grad_norm": 0.3538985550403595,
+      "learning_rate": 0.0002,
+      "loss": 0.9634,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.5501534938812256,
+      "learning_rate": 0.0002,
+      "loss": 1.0275,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3865698729582578,
+      "grad_norm": 0.4432051181793213,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3938294010889292,
+      "grad_norm": 0.41755786538124084,
+      "learning_rate": 0.0002,
+      "loss": 0.9082,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4010889292196007,
+      "grad_norm": 0.5626114010810852,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4083484573502723,
+      "grad_norm": 0.44038185477256775,
+      "learning_rate": 0.0002,
+      "loss": 1.0687,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4156079854809438,
+      "grad_norm": 0.3397001624107361,
+      "learning_rate": 0.0002,
+      "loss": 0.9454,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4228675136116151,
+      "grad_norm": 0.4325368106365204,
+      "learning_rate": 0.0002,
+      "loss": 0.943,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4301270417422867,
+      "grad_norm": 0.3900907039642334,
+      "learning_rate": 0.0002,
+      "loss": 1.0488,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4373865698729582,
+      "grad_norm": 0.369612455368042,
+      "learning_rate": 0.0002,
+      "loss": 0.9699,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4446460980036298,
+      "grad_norm": 0.4389338791370392,
+      "learning_rate": 0.0002,
+      "loss": 1.0609,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4519056261343013,
+      "grad_norm": 1.694450855255127,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4591651542649728,
+      "grad_norm": 0.516957700252533,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4664246823956444,
+      "grad_norm": 0.45515501499176025,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.4153139591217041,
+      "learning_rate": 0.0002,
+      "loss": 1.0188,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4809437386569873,
+      "grad_norm": 0.44353052973747253,
+      "learning_rate": 0.0002,
+      "loss": 1.1521,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4882032667876588,
+      "grad_norm": 0.570554256439209,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4954627949183303,
+      "grad_norm": 0.5742740035057068,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5027223230490017,
+      "grad_norm": 0.5890517830848694,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5099818511796732,
+      "grad_norm": 0.4162650406360626,
+      "learning_rate": 0.0002,
+      "loss": 1.0639,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5172413793103448,
+      "grad_norm": 0.7334717512130737,
+      "learning_rate": 0.0002,
+      "loss": 1.071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5245009074410163,
+      "grad_norm": 0.2963249683380127,
+      "learning_rate": 0.0002,
+      "loss": 0.8957,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5317604355716878,
+      "grad_norm": 0.30676454305648804,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5390199637023594,
+      "grad_norm": 0.35984641313552856,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 2120
+    },
+    {
+      "epoch": 1.546279491833031,
+      "grad_norm": 0.3384549617767334,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5535390199637025,
+      "grad_norm": 0.4725518226623535,
+      "learning_rate": 0.0002,
+      "loss": 0.9874,
+      "step": 2140
+    },
+    {
+      "epoch": 1.560798548094374,
+      "grad_norm": 0.3252685070037842,
+      "learning_rate": 0.0002,
+      "loss": 1.1166,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5680580762250453,
+      "grad_norm": 0.45043081045150757,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5753176043557169,
+      "grad_norm": 0.374208003282547,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5825771324863884,
+      "grad_norm": 0.5118404030799866,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5898366606170597,
+      "grad_norm": 0.482170969247818,
+      "learning_rate": 0.0002,
+      "loss": 1.0062,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5970961887477313,
+      "grad_norm": 0.5337533950805664,
+      "learning_rate": 0.0002,
+      "loss": 1.0872,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6043557168784028,
+      "grad_norm": 0.5195064544677734,
+      "learning_rate": 0.0002,
+      "loss": 1.0405,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6116152450090744,
+      "grad_norm": 0.30807098746299744,
+      "learning_rate": 0.0002,
+      "loss": 1.0454,
+      "step": 2220
+    },
+    {
+      "epoch": 1.618874773139746,
+      "grad_norm": 0.3962925672531128,
+      "learning_rate": 0.0002,
+      "loss": 1.0293,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6261343012704175,
+      "grad_norm": 0.7636962532997131,
+      "learning_rate": 0.0002,
+      "loss": 1.0137,
+      "step": 2240
+    },
+    {
+      "epoch": 1.633393829401089,
+      "grad_norm": 0.32380592823028564,
+      "learning_rate": 0.0002,
+      "loss": 0.991,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6406533575317606,
+      "grad_norm": 0.5767741799354553,
+      "learning_rate": 0.0002,
+      "loss": 0.9471,
+      "step": 2260
+    },
+    {
+      "epoch": 1.647912885662432,
+      "grad_norm": 0.39964812994003296,
+      "learning_rate": 0.0002,
+      "loss": 0.9524,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6551724137931034,
+      "grad_norm": 0.622629702091217,
+      "learning_rate": 0.0002,
+      "loss": 0.9866,
+      "step": 2280
+    },
+    {
+      "epoch": 1.662431941923775,
+      "grad_norm": 0.40202152729034424,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6696914700544465,
+      "grad_norm": 0.4467349052429199,
+      "learning_rate": 0.0002,
+      "loss": 0.9859,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6769509981851178,
+      "grad_norm": 0.5026949048042297,
+      "learning_rate": 0.0002,
+      "loss": 1.0312,
+      "step": 2310
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.43754541873931885,
+      "learning_rate": 0.0002,
+      "loss": 1.0776,
+      "step": 2320
+    },
+    {
+      "epoch": 1.691470054446461,
+      "grad_norm": 0.42869430780410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6987295825771325,
+      "grad_norm": 0.4192679524421692,
+      "learning_rate": 0.0002,
+      "loss": 1.0747,
+      "step": 2340
+    },
+    {
+      "epoch": 1.705989110707804,
+      "grad_norm": 0.3243155777454376,
+      "learning_rate": 0.0002,
+      "loss": 1.0054,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7132486388384756,
+      "grad_norm": 1.0514075756072998,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 2360
+    },
+    {
+      "epoch": 1.720508166969147,
+      "grad_norm": 0.4719122648239136,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7277676950998186,
+      "grad_norm": 0.3846144676208496,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 2380
+    },
+    {
+      "epoch": 1.73502722323049,
+      "grad_norm": 0.7266581058502197,
+      "learning_rate": 0.0002,
+      "loss": 0.9935,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7422867513611615,
+      "grad_norm": 0.6183241605758667,
+      "learning_rate": 0.0002,
+      "loss": 0.9509,
+      "step": 2400
+    },
+    {
+      "epoch": 1.749546279491833,
+      "grad_norm": 0.3658260405063629,
+      "learning_rate": 0.0002,
+      "loss": 0.9713,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7568058076225044,
+      "grad_norm": 0.6036322712898254,
+      "learning_rate": 0.0002,
+      "loss": 1.0033,
+      "step": 2420
+    },
+    {
+      "epoch": 1.764065335753176,
+      "grad_norm": 0.7872936129570007,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7713248638838475,
+      "grad_norm": 0.35946124792099,
+      "learning_rate": 0.0002,
+      "loss": 0.99,
+      "step": 2440
+    },
+    {
+      "epoch": 1.778584392014519,
+      "grad_norm": 0.3740338981151581,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2450
+    },
+    {
+      "epoch": 1.7858439201451906,
+      "grad_norm": 0.6150230169296265,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 2460
+    },
+    {
+      "epoch": 1.793103448275862,
+      "grad_norm": 0.4726075530052185,
+      "learning_rate": 0.0002,
+      "loss": 1.0373,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8003629764065336,
+      "grad_norm": 0.31292253732681274,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8076225045372052,
+      "grad_norm": 0.4463104009628296,
+      "learning_rate": 0.0002,
+      "loss": 1.0205,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8148820326678767,
+      "grad_norm": 0.7848200798034668,
+      "learning_rate": 0.0002,
+      "loss": 1.0771,
+      "step": 2500
+    },
+    {
+      "epoch": 1.822141560798548,
+      "grad_norm": 0.5562082529067993,
+      "learning_rate": 0.0002,
+      "loss": 0.9508,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8294010889292196,
+      "grad_norm": 0.39892756938934326,
+      "learning_rate": 0.0002,
+      "loss": 0.9343,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8366606170598911,
+      "grad_norm": 1.2923320531845093,
+      "learning_rate": 0.0002,
+      "loss": 1.0489,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8439201451905625,
+      "grad_norm": 0.6316490769386292,
+      "learning_rate": 0.0002,
+      "loss": 1.1203,
+      "step": 2540
+    },
+    {
+      "epoch": 1.851179673321234,
+      "grad_norm": 0.46100497245788574,
+      "learning_rate": 0.0002,
+      "loss": 0.9131,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8584392014519056,
+      "grad_norm": 0.7902987003326416,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 2560
+    },
+    {
+      "epoch": 1.865698729582577,
+      "grad_norm": 0.4596365690231323,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8729582577132486,
+      "grad_norm": 0.6592172384262085,
+      "learning_rate": 0.0002,
+      "loss": 0.9322,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8802177858439202,
+      "grad_norm": 0.5764662027359009,
+      "learning_rate": 0.0002,
+      "loss": 0.9549,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8874773139745917,
+      "grad_norm": 0.8421637415885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9485,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.5635305047035217,
+      "learning_rate": 0.0002,
+      "loss": 1.012,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9019963702359348,
+      "grad_norm": 0.46231237053871155,
+      "learning_rate": 0.0002,
+      "loss": 0.8907,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9092558983666061,
+      "grad_norm": 0.3944607079029083,
+      "learning_rate": 0.0002,
+      "loss": 0.9543,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9165154264972777,
+      "grad_norm": 0.4753907322883606,
+      "learning_rate": 0.0002,
+      "loss": 0.9964,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9237749546279492,
+      "grad_norm": 0.4151090979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.0217,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9310344827586206,
+      "grad_norm": 0.6793725490570068,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 2660
+    },
+    {
+      "epoch": 1.938294010889292,
+      "grad_norm": 0.339755117893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0401,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9455535390199636,
+      "grad_norm": 0.40789374709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.082,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9528130671506352,
+      "grad_norm": 0.3750005066394806,
+      "learning_rate": 0.0002,
+      "loss": 0.9973,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9600725952813067,
+      "grad_norm": 0.39684441685676575,
+      "learning_rate": 0.0002,
+      "loss": 0.9837,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9673321234119783,
+      "grad_norm": 0.378287672996521,
+      "learning_rate": 0.0002,
+      "loss": 1.1204,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9745916515426498,
+      "grad_norm": 0.3668482005596161,
+      "learning_rate": 0.0002,
+      "loss": 0.9284,
+      "step": 2720
+    },
+    {
+      "epoch": 1.9818511796733214,
+      "grad_norm": 0.49997565150260925,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 2730
+    },
+    {
+      "epoch": 1.989110707803993,
+      "grad_norm": 0.36852124333381653,
+      "learning_rate": 0.0002,
+      "loss": 0.9311,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9963702359346642,
+      "grad_norm": 0.5203380584716797,
+      "learning_rate": 0.0002,
+      "loss": 1.0898,
+      "step": 2750
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1086540222167969,
+      "eval_runtime": 95.6866,
+      "eval_samples_per_second": 4.557,
+      "eval_steps_per_second": 0.575,
+      "step": 2755
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11016,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.415071192055808e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..140bceabdec44fc473842cb9ace4a62a4dc9fe89
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4abec5480aff960b5a7b03a48be1cdb1ae29ba8792469c451f5c2f269e4be93b
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..f86d1076a959ad7c340c5f949c6b4b4b5c701e81
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec236235430d88b0ad03c049b7c5fb8da1beded051d5a85162681953de990a28
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..254c107dc7d441bd7e87c04f7cd394d5cc445643
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:983c73e9e8d3c89c114e978e78a75ed19d4826eeb7fb5778ed383f7d848f3d28
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..2e974d8deb466a24553e90c17446b11000cf19e9
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d996d97ac7036342ea56710f25417defb3ec407c108a59efb25cf82648499bd1
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e4000c9c3ae30d23b98f9d56765f7e8540cc8a8d
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f54153354bf146aa410b1e9645486e50d7214025e24901ae1b6c5a987a1a46e9
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..7434b2f03b41ca20d41d38a0fd1103894b284f71
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/trainer_state.json
@@ -0,0 +1,2948 @@
+{
+  "best_metric": 1.1086540222167969,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755",
+  "epoch": 2.9996370235934666,
+  "eval_steps": 10,
+  "global_step": 4132,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007259528130671506,
+      "grad_norm": 0.46010470390319824,
+      "learning_rate": 0.0002,
+      "loss": 2.4936,
+      "step": 10
+    },
+    {
+      "epoch": 0.014519056261343012,
+      "grad_norm": 0.5103179216384888,
+      "learning_rate": 0.0002,
+      "loss": 1.8442,
+      "step": 20
+    },
+    {
+      "epoch": 0.021778584392014518,
+      "grad_norm": 0.6282716989517212,
+      "learning_rate": 0.0002,
+      "loss": 1.559,
+      "step": 30
+    },
+    {
+      "epoch": 0.029038112522686024,
+      "grad_norm": 1.2480497360229492,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 40
+    },
+    {
+      "epoch": 0.036297640653357534,
+      "grad_norm": 0.4114855229854584,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 50
+    },
+    {
+      "epoch": 0.043557168784029036,
+      "grad_norm": 0.49482840299606323,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 60
+    },
+    {
+      "epoch": 0.050816696914700546,
+      "grad_norm": 0.4536272883415222,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 70
+    },
+    {
+      "epoch": 0.05807622504537205,
+      "grad_norm": 0.32328274846076965,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 80
+    },
+    {
+      "epoch": 0.06533575317604355,
+      "grad_norm": 0.40990468859672546,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 90
+    },
+    {
+      "epoch": 0.07259528130671507,
+      "grad_norm": 0.37273502349853516,
+      "learning_rate": 0.0002,
+      "loss": 1.2216,
+      "step": 100
+    },
+    {
+      "epoch": 0.07985480943738657,
+      "grad_norm": 0.3903871476650238,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 110
+    },
+    {
+      "epoch": 0.08711433756805807,
+      "grad_norm": 0.3272787928581238,
+      "learning_rate": 0.0002,
+      "loss": 1.3718,
+      "step": 120
+    },
+    {
+      "epoch": 0.09437386569872959,
+      "grad_norm": 0.3622824251651764,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 130
+    },
+    {
+      "epoch": 0.10163339382940109,
+      "grad_norm": 0.3503916561603546,
+      "learning_rate": 0.0002,
+      "loss": 1.2966,
+      "step": 140
+    },
+    {
+      "epoch": 0.1088929219600726,
+      "grad_norm": 0.32787832617759705,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 150
+    },
+    {
+      "epoch": 0.1161524500907441,
+      "grad_norm": 0.5822657942771912,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 160
+    },
+    {
+      "epoch": 0.12341197822141561,
+      "grad_norm": 0.28028249740600586,
+      "learning_rate": 0.0002,
+      "loss": 1.2091,
+      "step": 170
+    },
+    {
+      "epoch": 0.1306715063520871,
+      "grad_norm": 0.33602750301361084,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 180
+    },
+    {
+      "epoch": 0.13793103448275862,
+      "grad_norm": 0.29106274247169495,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 190
+    },
+    {
+      "epoch": 0.14519056261343014,
+      "grad_norm": 0.38753363490104675,
+      "learning_rate": 0.0002,
+      "loss": 1.3261,
+      "step": 200
+    },
+    {
+      "epoch": 0.15245009074410162,
+      "grad_norm": 0.361009418964386,
+      "learning_rate": 0.0002,
+      "loss": 1.1536,
+      "step": 210
+    },
+    {
+      "epoch": 0.15970961887477314,
+      "grad_norm": 0.6743836402893066,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 220
+    },
+    {
+      "epoch": 0.16696914700544466,
+      "grad_norm": 0.3923613727092743,
+      "learning_rate": 0.0002,
+      "loss": 1.264,
+      "step": 230
+    },
+    {
+      "epoch": 0.17422867513611615,
+      "grad_norm": 0.2809699773788452,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 240
+    },
+    {
+      "epoch": 0.18148820326678766,
+      "grad_norm": 0.3631494641304016,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 250
+    },
+    {
+      "epoch": 0.18874773139745918,
+      "grad_norm": 0.24658171832561493,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 260
+    },
+    {
+      "epoch": 0.19600725952813067,
+      "grad_norm": 0.5780664682388306,
+      "learning_rate": 0.0002,
+      "loss": 1.0065,
+      "step": 270
+    },
+    {
+      "epoch": 0.20326678765880218,
+      "grad_norm": 0.3056720495223999,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 280
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 0.2857084572315216,
+      "learning_rate": 0.0002,
+      "loss": 1.1503,
+      "step": 290
+    },
+    {
+      "epoch": 0.2177858439201452,
+      "grad_norm": 0.6645345687866211,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 300
+    },
+    {
+      "epoch": 0.2250453720508167,
+      "grad_norm": 0.5966078639030457,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 310
+    },
+    {
+      "epoch": 0.2323049001814882,
+      "grad_norm": 0.40937140583992004,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 320
+    },
+    {
+      "epoch": 0.2395644283121597,
+      "grad_norm": 0.5642806887626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 330
+    },
+    {
+      "epoch": 0.24682395644283123,
+      "grad_norm": 0.2750748097896576,
+      "learning_rate": 0.0002,
+      "loss": 1.1703,
+      "step": 340
+    },
+    {
+      "epoch": 0.2540834845735027,
+      "grad_norm": 0.34350234270095825,
+      "learning_rate": 0.0002,
+      "loss": 1.2598,
+      "step": 350
+    },
+    {
+      "epoch": 0.2613430127041742,
+      "grad_norm": 0.6767239570617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1942,
+      "step": 360
+    },
+    {
+      "epoch": 0.26860254083484575,
+      "grad_norm": 0.31006959080696106,
+      "learning_rate": 0.0002,
+      "loss": 1.1436,
+      "step": 370
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "grad_norm": 0.3825474679470062,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 380
+    },
+    {
+      "epoch": 0.2831215970961887,
+      "grad_norm": 0.30241551995277405,
+      "learning_rate": 0.0002,
+      "loss": 1.1523,
+      "step": 390
+    },
+    {
+      "epoch": 0.29038112522686027,
+      "grad_norm": 0.2962397336959839,
+      "learning_rate": 0.0002,
+      "loss": 1.1298,
+      "step": 400
+    },
+    {
+      "epoch": 0.29764065335753176,
+      "grad_norm": 0.2600369155406952,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 410
+    },
+    {
+      "epoch": 0.30490018148820325,
+      "grad_norm": 0.3675060272216797,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 420
+    },
+    {
+      "epoch": 0.3121597096188748,
+      "grad_norm": 0.3429498076438904,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 430
+    },
+    {
+      "epoch": 0.3194192377495463,
+      "grad_norm": 0.34311825037002563,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 440
+    },
+    {
+      "epoch": 0.32667876588021777,
+      "grad_norm": 0.37872210144996643,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 450
+    },
+    {
+      "epoch": 0.3339382940108893,
+      "grad_norm": 0.33271121978759766,
+      "learning_rate": 0.0002,
+      "loss": 1.1365,
+      "step": 460
+    },
+    {
+      "epoch": 0.3411978221415608,
+      "grad_norm": 0.34605276584625244,
+      "learning_rate": 0.0002,
+      "loss": 1.1572,
+      "step": 470
+    },
+    {
+      "epoch": 0.3484573502722323,
+      "grad_norm": 0.41050562262535095,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 480
+    },
+    {
+      "epoch": 0.35571687840290384,
+      "grad_norm": 0.2066836953163147,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 490
+    },
+    {
+      "epoch": 0.3629764065335753,
+      "grad_norm": 0.2859014868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.2072,
+      "step": 500
+    },
+    {
+      "epoch": 0.3702359346642468,
+      "grad_norm": 0.28763777017593384,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 510
+    },
+    {
+      "epoch": 0.37749546279491836,
+      "grad_norm": 0.2730471193790436,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 520
+    },
+    {
+      "epoch": 0.38475499092558985,
+      "grad_norm": 0.3968936800956726,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 530
+    },
+    {
+      "epoch": 0.39201451905626133,
+      "grad_norm": 0.3624701201915741,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 540
+    },
+    {
+      "epoch": 0.3992740471869328,
+      "grad_norm": 0.3303608298301697,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 550
+    },
+    {
+      "epoch": 0.40653357531760437,
+      "grad_norm": 0.33507466316223145,
+      "learning_rate": 0.0002,
+      "loss": 1.203,
+      "step": 560
+    },
+    {
+      "epoch": 0.41379310344827586,
+      "grad_norm": 0.3297670781612396,
+      "learning_rate": 0.0002,
+      "loss": 1.2294,
+      "step": 570
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.32334890961647034,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 580
+    },
+    {
+      "epoch": 0.4283121597096189,
+      "grad_norm": 0.30281195044517517,
+      "learning_rate": 0.0002,
+      "loss": 1.112,
+      "step": 590
+    },
+    {
+      "epoch": 0.4355716878402904,
+      "grad_norm": 0.5900027751922607,
+      "learning_rate": 0.0002,
+      "loss": 1.1352,
+      "step": 600
+    },
+    {
+      "epoch": 0.44283121597096187,
+      "grad_norm": 0.28437477350234985,
+      "learning_rate": 0.0002,
+      "loss": 1.1575,
+      "step": 610
+    },
+    {
+      "epoch": 0.4500907441016334,
+      "grad_norm": 0.39601704478263855,
+      "learning_rate": 0.0002,
+      "loss": 1.2009,
+      "step": 620
+    },
+    {
+      "epoch": 0.4573502722323049,
+      "grad_norm": 0.41971510648727417,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 630
+    },
+    {
+      "epoch": 0.4646098003629764,
+      "grad_norm": 0.33814409375190735,
+      "learning_rate": 0.0002,
+      "loss": 1.236,
+      "step": 640
+    },
+    {
+      "epoch": 0.47186932849364793,
+      "grad_norm": 0.575718104839325,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 650
+    },
+    {
+      "epoch": 0.4791288566243194,
+      "grad_norm": 0.37927401065826416,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 660
+    },
+    {
+      "epoch": 0.4863883847549909,
+      "grad_norm": 0.3224332630634308,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 670
+    },
+    {
+      "epoch": 0.49364791288566245,
+      "grad_norm": 0.32683515548706055,
+      "learning_rate": 0.0002,
+      "loss": 1.1354,
+      "step": 680
+    },
+    {
+      "epoch": 0.5009074410163339,
+      "grad_norm": 0.4316163659095764,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 690
+    },
+    {
+      "epoch": 0.5081669691470054,
+      "grad_norm": 0.342602401971817,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 700
+    },
+    {
+      "epoch": 0.515426497277677,
+      "grad_norm": 0.2794898748397827,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 710
+    },
+    {
+      "epoch": 0.5226860254083484,
+      "grad_norm": 0.3322339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.2395,
+      "step": 720
+    },
+    {
+      "epoch": 0.52994555353902,
+      "grad_norm": 0.3088509142398834,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 730
+    },
+    {
+      "epoch": 0.5372050816696915,
+      "grad_norm": 0.24444378912448883,
+      "learning_rate": 0.0002,
+      "loss": 1.1495,
+      "step": 740
+    },
+    {
+      "epoch": 0.5444646098003629,
+      "grad_norm": 0.3483171761035919,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 750
+    },
+    {
+      "epoch": 0.5517241379310345,
+      "grad_norm": 0.3425690531730652,
+      "learning_rate": 0.0002,
+      "loss": 1.2248,
+      "step": 760
+    },
+    {
+      "epoch": 0.558983666061706,
+      "grad_norm": 0.31841927766799927,
+      "learning_rate": 0.0002,
+      "loss": 1.1127,
+      "step": 770
+    },
+    {
+      "epoch": 0.5662431941923775,
+      "grad_norm": 0.39423868060112,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 780
+    },
+    {
+      "epoch": 0.573502722323049,
+      "grad_norm": 0.30328479409217834,
+      "learning_rate": 0.0002,
+      "loss": 1.0583,
+      "step": 790
+    },
+    {
+      "epoch": 0.5807622504537205,
+      "grad_norm": 0.24475938081741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0718,
+      "step": 800
+    },
+    {
+      "epoch": 0.588021778584392,
+      "grad_norm": 0.37132805585861206,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 810
+    },
+    {
+      "epoch": 0.5952813067150635,
+      "grad_norm": 0.32195979356765747,
+      "learning_rate": 0.0002,
+      "loss": 1.1544,
+      "step": 820
+    },
+    {
+      "epoch": 0.6025408348457351,
+      "grad_norm": 0.2848738729953766,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 830
+    },
+    {
+      "epoch": 0.6098003629764065,
+      "grad_norm": 0.28015264868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.0937,
+      "step": 840
+    },
+    {
+      "epoch": 0.617059891107078,
+      "grad_norm": 0.37796008586883545,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 850
+    },
+    {
+      "epoch": 0.6243194192377496,
+      "grad_norm": 0.39311841130256653,
+      "learning_rate": 0.0002,
+      "loss": 1.1455,
+      "step": 860
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.2761685252189636,
+      "learning_rate": 0.0002,
+      "loss": 1.1066,
+      "step": 870
+    },
+    {
+      "epoch": 0.6388384754990926,
+      "grad_norm": 0.3826720118522644,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 880
+    },
+    {
+      "epoch": 0.6460980036297641,
+      "grad_norm": 0.30076679587364197,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 890
+    },
+    {
+      "epoch": 0.6533575317604355,
+      "grad_norm": 0.21997687220573425,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 900
+    },
+    {
+      "epoch": 0.6606170598911071,
+      "grad_norm": 0.32593777775764465,
+      "learning_rate": 0.0002,
+      "loss": 1.1603,
+      "step": 910
+    },
+    {
+      "epoch": 0.6678765880217786,
+      "grad_norm": 0.30347898602485657,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 920
+    },
+    {
+      "epoch": 0.6751361161524501,
+      "grad_norm": 0.44173774123191833,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 930
+    },
+    {
+      "epoch": 0.6823956442831216,
+      "grad_norm": 0.2507467269897461,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 940
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.29463833570480347,
+      "learning_rate": 0.0002,
+      "loss": 1.3001,
+      "step": 950
+    },
+    {
+      "epoch": 0.6969147005444646,
+      "grad_norm": 0.9363154172897339,
+      "learning_rate": 0.0002,
+      "loss": 1.1651,
+      "step": 960
+    },
+    {
+      "epoch": 0.7041742286751361,
+      "grad_norm": 0.3236212134361267,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 970
+    },
+    {
+      "epoch": 0.7114337568058077,
+      "grad_norm": 0.3123254179954529,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 980
+    },
+    {
+      "epoch": 0.7186932849364791,
+      "grad_norm": 0.3395805060863495,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 990
+    },
+    {
+      "epoch": 0.7259528130671506,
+      "grad_norm": 0.3240964412689209,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7332123411978222,
+      "grad_norm": 0.31902948021888733,
+      "learning_rate": 0.0002,
+      "loss": 1.0313,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7404718693284936,
+      "grad_norm": 0.4848408102989197,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7477313974591652,
+      "grad_norm": 0.33006033301353455,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7549909255898367,
+      "grad_norm": 0.2928730547428131,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7622504537205081,
+      "grad_norm": 0.3529164791107178,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7695099818511797,
+      "grad_norm": 0.2736213803291321,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7767695099818511,
+      "grad_norm": 0.7200686931610107,
+      "learning_rate": 0.0002,
+      "loss": 1.1169,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7840290381125227,
+      "grad_norm": 0.33396708965301514,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7912885662431942,
+      "grad_norm": 1.5760449171066284,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7985480943738656,
+      "grad_norm": 0.28138381242752075,
+      "learning_rate": 0.0002,
+      "loss": 1.1126,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8058076225045372,
+      "grad_norm": 0.2597472369670868,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8130671506352087,
+      "grad_norm": 0.3305445611476898,
+      "learning_rate": 0.0002,
+      "loss": 1.1177,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8203266787658802,
+      "grad_norm": 0.3934599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.0849,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8275862068965517,
+      "grad_norm": 0.3472191393375397,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8348457350272233,
+      "grad_norm": 0.2857365906238556,
+      "learning_rate": 0.0002,
+      "loss": 1.265,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.3207702934741974,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8493647912885662,
+      "grad_norm": 0.3176484704017639,
+      "learning_rate": 0.0002,
+      "loss": 1.1669,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8566243194192378,
+      "grad_norm": 0.40685558319091797,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8638838475499092,
+      "grad_norm": 0.31125199794769287,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8711433756805808,
+      "grad_norm": 0.7361181378364563,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8784029038112523,
+      "grad_norm": 0.33699527382850647,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8856624319419237,
+      "grad_norm": 0.3315220773220062,
+      "learning_rate": 0.0002,
+      "loss": 1.11,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8929219600725953,
+      "grad_norm": 0.6256054043769836,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9001814882032668,
+      "grad_norm": 0.3692137897014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1738,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9074410163339383,
+      "grad_norm": 0.3538484573364258,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9147005444646098,
+      "grad_norm": 0.27863040566444397,
+      "learning_rate": 0.0002,
+      "loss": 1.0878,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9219600725952813,
+      "grad_norm": 0.3322528302669525,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9292196007259528,
+      "grad_norm": 0.7553173303604126,
+      "learning_rate": 0.0002,
+      "loss": 1.0158,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9364791288566243,
+      "grad_norm": 0.4856191575527191,
+      "learning_rate": 0.0002,
+      "loss": 1.198,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9437386569872959,
+      "grad_norm": 0.3668074905872345,
+      "learning_rate": 0.0002,
+      "loss": 1.1678,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9509981851179673,
+      "grad_norm": 0.29851067066192627,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9582577132486388,
+      "grad_norm": 0.276664674282074,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9655172413793104,
+      "grad_norm": 0.2941018342971802,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9727767695099818,
+      "grad_norm": 0.3505859076976776,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9800362976406534,
+      "grad_norm": 0.3067687451839447,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9872958257713249,
+      "grad_norm": 0.27151066064834595,
+      "learning_rate": 0.0002,
+      "loss": 1.1511,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9945553539019963,
+      "grad_norm": 0.36370083689689636,
+      "learning_rate": 0.0002,
+      "loss": 1.1919,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9996370235934664,
+      "eval_loss": 1.1381088495254517,
+      "eval_runtime": 96.0848,
+      "eval_samples_per_second": 4.538,
+      "eval_steps_per_second": 0.572,
+      "step": 1377
+    },
+    {
+      "epoch": 1.0018148820326678,
+      "grad_norm": 0.27980583906173706,
+      "learning_rate": 0.0002,
+      "loss": 1.0082,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0090744101633393,
+      "grad_norm": 0.26713913679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0163339382940109,
+      "grad_norm": 0.3089541494846344,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0235934664246824,
+      "grad_norm": 0.4188242256641388,
+      "learning_rate": 0.0002,
+      "loss": 0.9649,
+      "step": 1410
+    },
+    {
+      "epoch": 1.030852994555354,
+      "grad_norm": 0.5246463418006897,
+      "learning_rate": 0.0002,
+      "loss": 0.9821,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0381125226860255,
+      "grad_norm": 0.2728777825832367,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0453720508166968,
+      "grad_norm": 0.38167616724967957,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.4439380168914795,
+      "learning_rate": 0.0002,
+      "loss": 1.0229,
+      "step": 1450
+    },
+    {
+      "epoch": 1.05989110707804,
+      "grad_norm": 0.30954182147979736,
+      "learning_rate": 0.0002,
+      "loss": 1.0451,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0671506352087115,
+      "grad_norm": 0.4022280275821686,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 1470
+    },
+    {
+      "epoch": 1.074410163339383,
+      "grad_norm": 0.7390811443328857,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0816696914700545,
+      "grad_norm": 0.3885486423969269,
+      "learning_rate": 0.0002,
+      "loss": 1.1378,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0889292196007259,
+      "grad_norm": 0.5275560617446899,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0961887477313974,
+      "grad_norm": 0.35112282633781433,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1510
+    },
+    {
+      "epoch": 1.103448275862069,
+      "grad_norm": 0.33714351058006287,
+      "learning_rate": 0.0002,
+      "loss": 1.0331,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1107078039927405,
+      "grad_norm": 0.31221693754196167,
+      "learning_rate": 0.0002,
+      "loss": 0.8723,
+      "step": 1530
+    },
+    {
+      "epoch": 1.117967332123412,
+      "grad_norm": 0.27549654245376587,
+      "learning_rate": 0.0002,
+      "loss": 1.0701,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1252268602540836,
+      "grad_norm": 0.8465521335601807,
+      "learning_rate": 0.0002,
+      "loss": 1.0114,
+      "step": 1550
+    },
+    {
+      "epoch": 1.132486388384755,
+      "grad_norm": 0.36125949025154114,
+      "learning_rate": 0.0002,
+      "loss": 0.9643,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1397459165154264,
+      "grad_norm": 0.37420371174812317,
+      "learning_rate": 0.0002,
+      "loss": 1.0744,
+      "step": 1570
+    },
+    {
+      "epoch": 1.147005444646098,
+      "grad_norm": 0.3294760584831238,
+      "learning_rate": 0.0002,
+      "loss": 1.165,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1542649727767695,
+      "grad_norm": 0.3881238102912903,
+      "learning_rate": 0.0002,
+      "loss": 0.9207,
+      "step": 1590
+    },
+    {
+      "epoch": 1.161524500907441,
+      "grad_norm": 0.4766491651535034,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1687840290381124,
+      "grad_norm": 0.395530104637146,
+      "learning_rate": 0.0002,
+      "loss": 1.0092,
+      "step": 1610
+    },
+    {
+      "epoch": 1.176043557168784,
+      "grad_norm": 0.3297106623649597,
+      "learning_rate": 0.0002,
+      "loss": 0.9895,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1833030852994555,
+      "grad_norm": 0.39528271555900574,
+      "learning_rate": 0.0002,
+      "loss": 1.1121,
+      "step": 1630
+    },
+    {
+      "epoch": 1.190562613430127,
+      "grad_norm": 0.3370221257209778,
+      "learning_rate": 0.0002,
+      "loss": 0.9202,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1978221415607986,
+      "grad_norm": 0.31922030448913574,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2050816696914701,
+      "grad_norm": 0.6142027378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2123411978221417,
+      "grad_norm": 0.44769710302352905,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 1670
+    },
+    {
+      "epoch": 1.219600725952813,
+      "grad_norm": 0.41233646869659424,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2268602540834845,
+      "grad_norm": 0.2928866147994995,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1690
+    },
+    {
+      "epoch": 1.234119782214156,
+      "grad_norm": 0.36913734674453735,
+      "learning_rate": 0.0002,
+      "loss": 0.9074,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2413793103448276,
+      "grad_norm": 0.5281891226768494,
+      "learning_rate": 0.0002,
+      "loss": 1.0847,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2486388384754992,
+      "grad_norm": 0.3374697268009186,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2558983666061705,
+      "grad_norm": 0.3802020847797394,
+      "learning_rate": 0.0002,
+      "loss": 0.9826,
+      "step": 1730
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.38048651814460754,
+      "learning_rate": 0.0002,
+      "loss": 1.0191,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2704174228675136,
+      "grad_norm": 0.6676169633865356,
+      "learning_rate": 0.0002,
+      "loss": 1.0142,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2776769509981851,
+      "grad_norm": 0.4075961410999298,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2849364791288567,
+      "grad_norm": 0.4374721348285675,
+      "learning_rate": 0.0002,
+      "loss": 1.0301,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2921960072595282,
+      "grad_norm": 0.4638824164867401,
+      "learning_rate": 0.0002,
+      "loss": 1.0601,
+      "step": 1780
+    },
+    {
+      "epoch": 1.2994555353901998,
+      "grad_norm": 0.38631564378738403,
+      "learning_rate": 0.0002,
+      "loss": 0.9593,
+      "step": 1790
+    },
+    {
+      "epoch": 1.306715063520871,
+      "grad_norm": 0.35873809456825256,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3139745916515426,
+      "grad_norm": 0.33602237701416016,
+      "learning_rate": 0.0002,
+      "loss": 0.9835,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3212341197822142,
+      "grad_norm": 0.46696463227272034,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3284936479128857,
+      "grad_norm": 0.368958979845047,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 1830
+    },
+    {
+      "epoch": 1.335753176043557,
+      "grad_norm": 0.3160957396030426,
+      "learning_rate": 0.0002,
+      "loss": 1.0456,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3430127041742286,
+      "grad_norm": 0.4511511027812958,
+      "learning_rate": 0.0002,
+      "loss": 1.0264,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3502722323049001,
+      "grad_norm": 0.6769845485687256,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3575317604355717,
+      "grad_norm": 0.3749309480190277,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3647912885662432,
+      "grad_norm": 0.553798496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.0223,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3720508166969148,
+      "grad_norm": 0.3538985550403595,
+      "learning_rate": 0.0002,
+      "loss": 0.9634,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.5501534938812256,
+      "learning_rate": 0.0002,
+      "loss": 1.0275,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3865698729582578,
+      "grad_norm": 0.4432051181793213,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3938294010889292,
+      "grad_norm": 0.41755786538124084,
+      "learning_rate": 0.0002,
+      "loss": 0.9082,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4010889292196007,
+      "grad_norm": 0.5626114010810852,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4083484573502723,
+      "grad_norm": 0.44038185477256775,
+      "learning_rate": 0.0002,
+      "loss": 1.0687,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4156079854809438,
+      "grad_norm": 0.3397001624107361,
+      "learning_rate": 0.0002,
+      "loss": 0.9454,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4228675136116151,
+      "grad_norm": 0.4325368106365204,
+      "learning_rate": 0.0002,
+      "loss": 0.943,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4301270417422867,
+      "grad_norm": 0.3900907039642334,
+      "learning_rate": 0.0002,
+      "loss": 1.0488,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4373865698729582,
+      "grad_norm": 0.369612455368042,
+      "learning_rate": 0.0002,
+      "loss": 0.9699,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4446460980036298,
+      "grad_norm": 0.4389338791370392,
+      "learning_rate": 0.0002,
+      "loss": 1.0609,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4519056261343013,
+      "grad_norm": 1.694450855255127,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4591651542649728,
+      "grad_norm": 0.516957700252533,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4664246823956444,
+      "grad_norm": 0.45515501499176025,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.4153139591217041,
+      "learning_rate": 0.0002,
+      "loss": 1.0188,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4809437386569873,
+      "grad_norm": 0.44353052973747253,
+      "learning_rate": 0.0002,
+      "loss": 1.1521,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4882032667876588,
+      "grad_norm": 0.570554256439209,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4954627949183303,
+      "grad_norm": 0.5742740035057068,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5027223230490017,
+      "grad_norm": 0.5890517830848694,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5099818511796732,
+      "grad_norm": 0.4162650406360626,
+      "learning_rate": 0.0002,
+      "loss": 1.0639,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5172413793103448,
+      "grad_norm": 0.7334717512130737,
+      "learning_rate": 0.0002,
+      "loss": 1.071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5245009074410163,
+      "grad_norm": 0.2963249683380127,
+      "learning_rate": 0.0002,
+      "loss": 0.8957,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5317604355716878,
+      "grad_norm": 0.30676454305648804,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5390199637023594,
+      "grad_norm": 0.35984641313552856,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 2120
+    },
+    {
+      "epoch": 1.546279491833031,
+      "grad_norm": 0.3384549617767334,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5535390199637025,
+      "grad_norm": 0.4725518226623535,
+      "learning_rate": 0.0002,
+      "loss": 0.9874,
+      "step": 2140
+    },
+    {
+      "epoch": 1.560798548094374,
+      "grad_norm": 0.3252685070037842,
+      "learning_rate": 0.0002,
+      "loss": 1.1166,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5680580762250453,
+      "grad_norm": 0.45043081045150757,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5753176043557169,
+      "grad_norm": 0.374208003282547,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5825771324863884,
+      "grad_norm": 0.5118404030799866,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5898366606170597,
+      "grad_norm": 0.482170969247818,
+      "learning_rate": 0.0002,
+      "loss": 1.0062,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5970961887477313,
+      "grad_norm": 0.5337533950805664,
+      "learning_rate": 0.0002,
+      "loss": 1.0872,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6043557168784028,
+      "grad_norm": 0.5195064544677734,
+      "learning_rate": 0.0002,
+      "loss": 1.0405,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6116152450090744,
+      "grad_norm": 0.30807098746299744,
+      "learning_rate": 0.0002,
+      "loss": 1.0454,
+      "step": 2220
+    },
+    {
+      "epoch": 1.618874773139746,
+      "grad_norm": 0.3962925672531128,
+      "learning_rate": 0.0002,
+      "loss": 1.0293,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6261343012704175,
+      "grad_norm": 0.7636962532997131,
+      "learning_rate": 0.0002,
+      "loss": 1.0137,
+      "step": 2240
+    },
+    {
+      "epoch": 1.633393829401089,
+      "grad_norm": 0.32380592823028564,
+      "learning_rate": 0.0002,
+      "loss": 0.991,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6406533575317606,
+      "grad_norm": 0.5767741799354553,
+      "learning_rate": 0.0002,
+      "loss": 0.9471,
+      "step": 2260
+    },
+    {
+      "epoch": 1.647912885662432,
+      "grad_norm": 0.39964812994003296,
+      "learning_rate": 0.0002,
+      "loss": 0.9524,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6551724137931034,
+      "grad_norm": 0.622629702091217,
+      "learning_rate": 0.0002,
+      "loss": 0.9866,
+      "step": 2280
+    },
+    {
+      "epoch": 1.662431941923775,
+      "grad_norm": 0.40202152729034424,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6696914700544465,
+      "grad_norm": 0.4467349052429199,
+      "learning_rate": 0.0002,
+      "loss": 0.9859,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6769509981851178,
+      "grad_norm": 0.5026949048042297,
+      "learning_rate": 0.0002,
+      "loss": 1.0312,
+      "step": 2310
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.43754541873931885,
+      "learning_rate": 0.0002,
+      "loss": 1.0776,
+      "step": 2320
+    },
+    {
+      "epoch": 1.691470054446461,
+      "grad_norm": 0.42869430780410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6987295825771325,
+      "grad_norm": 0.4192679524421692,
+      "learning_rate": 0.0002,
+      "loss": 1.0747,
+      "step": 2340
+    },
+    {
+      "epoch": 1.705989110707804,
+      "grad_norm": 0.3243155777454376,
+      "learning_rate": 0.0002,
+      "loss": 1.0054,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7132486388384756,
+      "grad_norm": 1.0514075756072998,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 2360
+    },
+    {
+      "epoch": 1.720508166969147,
+      "grad_norm": 0.4719122648239136,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7277676950998186,
+      "grad_norm": 0.3846144676208496,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 2380
+    },
+    {
+      "epoch": 1.73502722323049,
+      "grad_norm": 0.7266581058502197,
+      "learning_rate": 0.0002,
+      "loss": 0.9935,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7422867513611615,
+      "grad_norm": 0.6183241605758667,
+      "learning_rate": 0.0002,
+      "loss": 0.9509,
+      "step": 2400
+    },
+    {
+      "epoch": 1.749546279491833,
+      "grad_norm": 0.3658260405063629,
+      "learning_rate": 0.0002,
+      "loss": 0.9713,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7568058076225044,
+      "grad_norm": 0.6036322712898254,
+      "learning_rate": 0.0002,
+      "loss": 1.0033,
+      "step": 2420
+    },
+    {
+      "epoch": 1.764065335753176,
+      "grad_norm": 0.7872936129570007,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7713248638838475,
+      "grad_norm": 0.35946124792099,
+      "learning_rate": 0.0002,
+      "loss": 0.99,
+      "step": 2440
+    },
+    {
+      "epoch": 1.778584392014519,
+      "grad_norm": 0.3740338981151581,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2450
+    },
+    {
+      "epoch": 1.7858439201451906,
+      "grad_norm": 0.6150230169296265,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 2460
+    },
+    {
+      "epoch": 1.793103448275862,
+      "grad_norm": 0.4726075530052185,
+      "learning_rate": 0.0002,
+      "loss": 1.0373,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8003629764065336,
+      "grad_norm": 0.31292253732681274,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8076225045372052,
+      "grad_norm": 0.4463104009628296,
+      "learning_rate": 0.0002,
+      "loss": 1.0205,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8148820326678767,
+      "grad_norm": 0.7848200798034668,
+      "learning_rate": 0.0002,
+      "loss": 1.0771,
+      "step": 2500
+    },
+    {
+      "epoch": 1.822141560798548,
+      "grad_norm": 0.5562082529067993,
+      "learning_rate": 0.0002,
+      "loss": 0.9508,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8294010889292196,
+      "grad_norm": 0.39892756938934326,
+      "learning_rate": 0.0002,
+      "loss": 0.9343,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8366606170598911,
+      "grad_norm": 1.2923320531845093,
+      "learning_rate": 0.0002,
+      "loss": 1.0489,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8439201451905625,
+      "grad_norm": 0.6316490769386292,
+      "learning_rate": 0.0002,
+      "loss": 1.1203,
+      "step": 2540
+    },
+    {
+      "epoch": 1.851179673321234,
+      "grad_norm": 0.46100497245788574,
+      "learning_rate": 0.0002,
+      "loss": 0.9131,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8584392014519056,
+      "grad_norm": 0.7902987003326416,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 2560
+    },
+    {
+      "epoch": 1.865698729582577,
+      "grad_norm": 0.4596365690231323,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8729582577132486,
+      "grad_norm": 0.6592172384262085,
+      "learning_rate": 0.0002,
+      "loss": 0.9322,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8802177858439202,
+      "grad_norm": 0.5764662027359009,
+      "learning_rate": 0.0002,
+      "loss": 0.9549,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8874773139745917,
+      "grad_norm": 0.8421637415885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9485,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.5635305047035217,
+      "learning_rate": 0.0002,
+      "loss": 1.012,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9019963702359348,
+      "grad_norm": 0.46231237053871155,
+      "learning_rate": 0.0002,
+      "loss": 0.8907,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9092558983666061,
+      "grad_norm": 0.3944607079029083,
+      "learning_rate": 0.0002,
+      "loss": 0.9543,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9165154264972777,
+      "grad_norm": 0.4753907322883606,
+      "learning_rate": 0.0002,
+      "loss": 0.9964,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9237749546279492,
+      "grad_norm": 0.4151090979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.0217,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9310344827586206,
+      "grad_norm": 0.6793725490570068,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 2660
+    },
+    {
+      "epoch": 1.938294010889292,
+      "grad_norm": 0.339755117893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0401,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9455535390199636,
+      "grad_norm": 0.40789374709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.082,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9528130671506352,
+      "grad_norm": 0.3750005066394806,
+      "learning_rate": 0.0002,
+      "loss": 0.9973,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9600725952813067,
+      "grad_norm": 0.39684441685676575,
+      "learning_rate": 0.0002,
+      "loss": 0.9837,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9673321234119783,
+      "grad_norm": 0.378287672996521,
+      "learning_rate": 0.0002,
+      "loss": 1.1204,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9745916515426498,
+      "grad_norm": 0.3668482005596161,
+      "learning_rate": 0.0002,
+      "loss": 0.9284,
+      "step": 2720
+    },
+    {
+      "epoch": 1.9818511796733214,
+      "grad_norm": 0.49997565150260925,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 2730
+    },
+    {
+      "epoch": 1.989110707803993,
+      "grad_norm": 0.36852124333381653,
+      "learning_rate": 0.0002,
+      "loss": 0.9311,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9963702359346642,
+      "grad_norm": 0.5203380584716797,
+      "learning_rate": 0.0002,
+      "loss": 1.0898,
+      "step": 2750
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1086540222167969,
+      "eval_runtime": 95.6866,
+      "eval_samples_per_second": 4.557,
+      "eval_steps_per_second": 0.575,
+      "step": 2755
+    },
+    {
+      "epoch": 2.0036297640653356,
+      "grad_norm": 0.35921573638916016,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2760
+    },
+    {
+      "epoch": 2.010889292196007,
+      "grad_norm": 1.013861894607544,
+      "learning_rate": 0.0002,
+      "loss": 0.8968,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0181488203266786,
+      "grad_norm": 0.4425240159034729,
+      "learning_rate": 0.0002,
+      "loss": 0.7873,
+      "step": 2780
+    },
+    {
+      "epoch": 2.02540834845735,
+      "grad_norm": 0.60174161195755,
+      "learning_rate": 0.0002,
+      "loss": 0.8758,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0326678765880217,
+      "grad_norm": 0.47582098841667175,
+      "learning_rate": 0.0002,
+      "loss": 0.7989,
+      "step": 2800
+    },
+    {
+      "epoch": 2.0399274047186933,
+      "grad_norm": 0.6012811660766602,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 2810
+    },
+    {
+      "epoch": 2.047186932849365,
+      "grad_norm": 0.4444408118724823,
+      "learning_rate": 0.0002,
+      "loss": 0.8109,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0544464609800364,
+      "grad_norm": 0.4864003360271454,
+      "learning_rate": 0.0002,
+      "loss": 0.8621,
+      "step": 2830
+    },
+    {
+      "epoch": 2.061705989110708,
+      "grad_norm": 0.5104215741157532,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.6218489408493042,
+      "learning_rate": 0.0002,
+      "loss": 0.7789,
+      "step": 2850
+    },
+    {
+      "epoch": 2.076225045372051,
+      "grad_norm": 0.705784261226654,
+      "learning_rate": 0.0002,
+      "loss": 0.8778,
+      "step": 2860
+    },
+    {
+      "epoch": 2.0834845735027225,
+      "grad_norm": 0.48091503977775574,
+      "learning_rate": 0.0002,
+      "loss": 0.7851,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0907441016333936,
+      "grad_norm": 0.5062456727027893,
+      "learning_rate": 0.0002,
+      "loss": 0.8444,
+      "step": 2880
+    },
+    {
+      "epoch": 2.098003629764065,
+      "grad_norm": 0.4862022399902344,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 0.48264044523239136,
+      "learning_rate": 0.0002,
+      "loss": 0.8593,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1125226860254083,
+      "grad_norm": 0.43744346499443054,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2910
+    },
+    {
+      "epoch": 2.11978221415608,
+      "grad_norm": 0.5480492115020752,
+      "learning_rate": 0.0002,
+      "loss": 0.9099,
+      "step": 2920
+    },
+    {
+      "epoch": 2.1270417422867514,
+      "grad_norm": 0.5068560838699341,
+      "learning_rate": 0.0002,
+      "loss": 0.7727,
+      "step": 2930
+    },
+    {
+      "epoch": 2.134301270417423,
+      "grad_norm": 0.4650852680206299,
+      "learning_rate": 0.0002,
+      "loss": 0.8403,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1415607985480944,
+      "grad_norm": 0.4929981231689453,
+      "learning_rate": 0.0002,
+      "loss": 0.8467,
+      "step": 2950
+    },
+    {
+      "epoch": 2.148820326678766,
+      "grad_norm": 0.6537389755249023,
+      "learning_rate": 0.0002,
+      "loss": 0.8747,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1560798548094375,
+      "grad_norm": 0.8032940626144409,
+      "learning_rate": 0.0002,
+      "loss": 0.7663,
+      "step": 2970
+    },
+    {
+      "epoch": 2.163339382940109,
+      "grad_norm": 0.7131643891334534,
+      "learning_rate": 0.0002,
+      "loss": 0.7604,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1705989110707806,
+      "grad_norm": 0.6034275889396667,
+      "learning_rate": 0.0002,
+      "loss": 0.8424,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1778584392014517,
+      "grad_norm": 0.6081095933914185,
+      "learning_rate": 0.0002,
+      "loss": 0.8838,
+      "step": 3000
+    },
+    {
+      "epoch": 2.1851179673321233,
+      "grad_norm": 0.5706912875175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9013,
+      "step": 3010
+    },
+    {
+      "epoch": 2.192377495462795,
+      "grad_norm": 0.6742380261421204,
+      "learning_rate": 0.0002,
+      "loss": 0.8473,
+      "step": 3020
+    },
+    {
+      "epoch": 2.1996370235934664,
+      "grad_norm": 0.4847496449947357,
+      "learning_rate": 0.0002,
+      "loss": 0.8372,
+      "step": 3030
+    },
+    {
+      "epoch": 2.206896551724138,
+      "grad_norm": 0.5529342889785767,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2141560798548094,
+      "grad_norm": 0.6108783483505249,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3050
+    },
+    {
+      "epoch": 2.221415607985481,
+      "grad_norm": 0.8841571807861328,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2286751361161525,
+      "grad_norm": 0.4227530360221863,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 3070
+    },
+    {
+      "epoch": 2.235934664246824,
+      "grad_norm": 0.461935818195343,
+      "learning_rate": 0.0002,
+      "loss": 0.7925,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2431941923774956,
+      "grad_norm": 0.5407412648200989,
+      "learning_rate": 0.0002,
+      "loss": 0.8542,
+      "step": 3090
+    },
+    {
+      "epoch": 2.250453720508167,
+      "grad_norm": 0.5057998895645142,
+      "learning_rate": 0.0002,
+      "loss": 0.8392,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2577132486388383,
+      "grad_norm": 0.530057966709137,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 3110
+    },
+    {
+      "epoch": 2.26497277676951,
+      "grad_norm": 0.5066842436790466,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2722323049001814,
+      "grad_norm": 0.5069178342819214,
+      "learning_rate": 0.0002,
+      "loss": 0.7903,
+      "step": 3130
+    },
+    {
+      "epoch": 2.279491833030853,
+      "grad_norm": 0.6095499396324158,
+      "learning_rate": 0.0002,
+      "loss": 0.8458,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2867513611615244,
+      "grad_norm": 0.49524766206741333,
+      "learning_rate": 0.0002,
+      "loss": 0.8171,
+      "step": 3150
+    },
+    {
+      "epoch": 2.294010889292196,
+      "grad_norm": 0.5334409475326538,
+      "learning_rate": 0.0002,
+      "loss": 0.7568,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3012704174228675,
+      "grad_norm": 1.681748867034912,
+      "learning_rate": 0.0002,
+      "loss": 0.7533,
+      "step": 3170
+    },
+    {
+      "epoch": 2.308529945553539,
+      "grad_norm": 0.7225565314292908,
+      "learning_rate": 0.0002,
+      "loss": 0.8155,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3157894736842106,
+      "grad_norm": 0.5379388928413391,
+      "learning_rate": 0.0002,
+      "loss": 0.8034,
+      "step": 3190
+    },
+    {
+      "epoch": 2.323049001814882,
+      "grad_norm": 0.45770326256752014,
+      "learning_rate": 0.0002,
+      "loss": 0.8833,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3303085299455537,
+      "grad_norm": 0.6984533071517944,
+      "learning_rate": 0.0002,
+      "loss": 0.8419,
+      "step": 3210
+    },
+    {
+      "epoch": 2.337568058076225,
+      "grad_norm": 0.6725744605064392,
+      "learning_rate": 0.0002,
+      "loss": 0.7414,
+      "step": 3220
+    },
+    {
+      "epoch": 2.344827586206897,
+      "grad_norm": 1.1247471570968628,
+      "learning_rate": 0.0002,
+      "loss": 0.8657,
+      "step": 3230
+    },
+    {
+      "epoch": 2.352087114337568,
+      "grad_norm": 1.0240263938903809,
+      "learning_rate": 0.0002,
+      "loss": 0.7782,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3593466424682394,
+      "grad_norm": 0.5608096122741699,
+      "learning_rate": 0.0002,
+      "loss": 0.826,
+      "step": 3250
+    },
+    {
+      "epoch": 2.366606170598911,
+      "grad_norm": 0.8294990062713623,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3738656987295825,
+      "grad_norm": 0.6734224557876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9845,
+      "step": 3270
+    },
+    {
+      "epoch": 2.381125226860254,
+      "grad_norm": 0.6862800717353821,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3883847549909256,
+      "grad_norm": 0.5442930459976196,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 3290
+    },
+    {
+      "epoch": 2.395644283121597,
+      "grad_norm": 0.5745394229888916,
+      "learning_rate": 0.0002,
+      "loss": 0.8543,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4029038112522687,
+      "grad_norm": 0.6257799863815308,
+      "learning_rate": 0.0002,
+      "loss": 0.833,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4101633393829403,
+      "grad_norm": 0.5608420968055725,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3320
+    },
+    {
+      "epoch": 2.417422867513612,
+      "grad_norm": 0.5512017011642456,
+      "learning_rate": 0.0002,
+      "loss": 0.8745,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4246823956442833,
+      "grad_norm": 0.7452999353408813,
+      "learning_rate": 0.0002,
+      "loss": 0.8642,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4319419237749544,
+      "grad_norm": 0.4604301452636719,
+      "learning_rate": 0.0002,
+      "loss": 0.8417,
+      "step": 3350
+    },
+    {
+      "epoch": 2.439201451905626,
+      "grad_norm": 0.8225823640823364,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4464609800362975,
+      "grad_norm": 0.8604981303215027,
+      "learning_rate": 0.0002,
+      "loss": 0.8144,
+      "step": 3370
+    },
+    {
+      "epoch": 2.453720508166969,
+      "grad_norm": 0.6620925664901733,
+      "learning_rate": 0.0002,
+      "loss": 0.9276,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4609800362976406,
+      "grad_norm": 0.4750158488750458,
+      "learning_rate": 0.0002,
+      "loss": 0.8381,
+      "step": 3390
+    },
+    {
+      "epoch": 2.468239564428312,
+      "grad_norm": 0.6061418056488037,
+      "learning_rate": 0.0002,
+      "loss": 0.8249,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4754990925589837,
+      "grad_norm": 0.5977247357368469,
+      "learning_rate": 0.0002,
+      "loss": 0.8944,
+      "step": 3410
+    },
+    {
+      "epoch": 2.4827586206896552,
+      "grad_norm": 0.5004227757453918,
+      "learning_rate": 0.0002,
+      "loss": 0.8366,
+      "step": 3420
+    },
+    {
+      "epoch": 2.490018148820327,
+      "grad_norm": 0.46428972482681274,
+      "learning_rate": 0.0002,
+      "loss": 0.9133,
+      "step": 3430
+    },
+    {
+      "epoch": 2.4972776769509983,
+      "grad_norm": 0.7658395171165466,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 3440
+    },
+    {
+      "epoch": 2.50453720508167,
+      "grad_norm": 0.621819794178009,
+      "learning_rate": 0.0002,
+      "loss": 0.9066,
+      "step": 3450
+    },
+    {
+      "epoch": 2.511796733212341,
+      "grad_norm": 0.4556088149547577,
+      "learning_rate": 0.0002,
+      "loss": 0.8481,
+      "step": 3460
+    },
+    {
+      "epoch": 2.519056261343013,
+      "grad_norm": 0.6124140024185181,
+      "learning_rate": 0.0002,
+      "loss": 0.8522,
+      "step": 3470
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 0.6256231665611267,
+      "learning_rate": 0.0002,
+      "loss": 0.8158,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5335753176043556,
+      "grad_norm": 0.5464141964912415,
+      "learning_rate": 0.0002,
+      "loss": 0.9097,
+      "step": 3490
+    },
+    {
+      "epoch": 2.540834845735027,
+      "grad_norm": 0.51471346616745,
+      "learning_rate": 0.0002,
+      "loss": 0.8708,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5480943738656987,
+      "grad_norm": 0.5326165556907654,
+      "learning_rate": 0.0002,
+      "loss": 0.9042,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5553539019963702,
+      "grad_norm": 0.4750378429889679,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 3520
+    },
+    {
+      "epoch": 2.562613430127042,
+      "grad_norm": 0.5292279124259949,
+      "learning_rate": 0.0002,
+      "loss": 0.8248,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5698729582577133,
+      "grad_norm": 0.6145227551460266,
+      "learning_rate": 0.0002,
+      "loss": 0.8808,
+      "step": 3540
+    },
+    {
+      "epoch": 2.577132486388385,
+      "grad_norm": 0.8275189399719238,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5843920145190564,
+      "grad_norm": 0.5037438273429871,
+      "learning_rate": 0.0002,
+      "loss": 0.875,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5916515426497275,
+      "grad_norm": 0.5838707685470581,
+      "learning_rate": 0.0002,
+      "loss": 0.8324,
+      "step": 3570
+    },
+    {
+      "epoch": 2.5989110707803995,
+      "grad_norm": 0.5398710370063782,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6061705989110706,
+      "grad_norm": 0.6115376949310303,
+      "learning_rate": 0.0002,
+      "loss": 0.7843,
+      "step": 3590
+    },
+    {
+      "epoch": 2.613430127041742,
+      "grad_norm": 1.118809461593628,
+      "learning_rate": 0.0002,
+      "loss": 0.8142,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6206896551724137,
+      "grad_norm": 0.7811771631240845,
+      "learning_rate": 0.0002,
+      "loss": 0.8282,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6279491833030852,
+      "grad_norm": 0.5753175020217896,
+      "learning_rate": 0.0002,
+      "loss": 0.7997,
+      "step": 3620
+    },
+    {
+      "epoch": 2.635208711433757,
+      "grad_norm": 0.550829291343689,
+      "learning_rate": 0.0002,
+      "loss": 0.7705,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6424682395644283,
+      "grad_norm": 0.5360019207000732,
+      "learning_rate": 0.0002,
+      "loss": 0.8564,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6497277676951,
+      "grad_norm": 0.63050377368927,
+      "learning_rate": 0.0002,
+      "loss": 0.7648,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6569872958257714,
+      "grad_norm": 0.5833110213279724,
+      "learning_rate": 0.0002,
+      "loss": 0.9156,
+      "step": 3660
+    },
+    {
+      "epoch": 2.664246823956443,
+      "grad_norm": 0.5543047189712524,
+      "learning_rate": 0.0002,
+      "loss": 0.8033,
+      "step": 3670
+    },
+    {
+      "epoch": 2.671506352087114,
+      "grad_norm": 0.6842212080955505,
+      "learning_rate": 0.0002,
+      "loss": 0.8867,
+      "step": 3680
+    },
+    {
+      "epoch": 2.678765880217786,
+      "grad_norm": 0.6859333515167236,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 3690
+    },
+    {
+      "epoch": 2.686025408348457,
+      "grad_norm": 0.7038410902023315,
+      "learning_rate": 0.0002,
+      "loss": 0.8081,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6932849364791287,
+      "grad_norm": 0.447233647108078,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7005444646098002,
+      "grad_norm": 0.6453872919082642,
+      "learning_rate": 0.0002,
+      "loss": 0.8951,
+      "step": 3720
+    },
+    {
+      "epoch": 2.707803992740472,
+      "grad_norm": 0.8025672435760498,
+      "learning_rate": 0.0002,
+      "loss": 0.8292,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7150635208711433,
+      "grad_norm": 0.5997087955474854,
+      "learning_rate": 0.0002,
+      "loss": 0.8488,
+      "step": 3740
+    },
+    {
+      "epoch": 2.722323049001815,
+      "grad_norm": 0.6901142001152039,
+      "learning_rate": 0.0002,
+      "loss": 0.8308,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7295825771324864,
+      "grad_norm": 1.036145567893982,
+      "learning_rate": 0.0002,
+      "loss": 0.8517,
+      "step": 3760
+    },
+    {
+      "epoch": 2.736842105263158,
+      "grad_norm": 0.7207072377204895,
+      "learning_rate": 0.0002,
+      "loss": 0.8455,
+      "step": 3770
+    },
+    {
+      "epoch": 2.7441016333938295,
+      "grad_norm": 1.0452989339828491,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 3780
+    },
+    {
+      "epoch": 2.751361161524501,
+      "grad_norm": 0.5615278482437134,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.46439215540885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7658802177858437,
+      "grad_norm": 0.7134917974472046,
+      "learning_rate": 0.0002,
+      "loss": 0.7847,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7731397459165157,
+      "grad_norm": 0.5139115452766418,
+      "learning_rate": 0.0002,
+      "loss": 0.8315,
+      "step": 3820
+    },
+    {
+      "epoch": 2.780399274047187,
+      "grad_norm": 0.8595781326293945,
+      "learning_rate": 0.0002,
+      "loss": 0.8995,
+      "step": 3830
+    },
+    {
+      "epoch": 2.7876588021778583,
+      "grad_norm": 0.544614851474762,
+      "learning_rate": 0.0002,
+      "loss": 0.9079,
+      "step": 3840
+    },
+    {
+      "epoch": 2.79491833030853,
+      "grad_norm": 0.6073850393295288,
+      "learning_rate": 0.0002,
+      "loss": 0.8322,
+      "step": 3850
+    },
+    {
+      "epoch": 2.8021778584392014,
+      "grad_norm": 0.8224069476127625,
+      "learning_rate": 0.0002,
+      "loss": 0.8277,
+      "step": 3860
+    },
+    {
+      "epoch": 2.809437386569873,
+      "grad_norm": 0.5347970128059387,
+      "learning_rate": 0.0002,
+      "loss": 0.815,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8166969147005445,
+      "grad_norm": 0.6799601912498474,
+      "learning_rate": 0.0002,
+      "loss": 0.8219,
+      "step": 3880
+    },
+    {
+      "epoch": 2.823956442831216,
+      "grad_norm": 0.5219197869300842,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8312159709618876,
+      "grad_norm": 0.5710130333900452,
+      "learning_rate": 0.0002,
+      "loss": 0.8649,
+      "step": 3900
+    },
+    {
+      "epoch": 2.838475499092559,
+      "grad_norm": 0.5857213139533997,
+      "learning_rate": 0.0002,
+      "loss": 0.8741,
+      "step": 3910
+    },
+    {
+      "epoch": 2.8457350272232302,
+      "grad_norm": 0.5206644535064697,
+      "learning_rate": 0.0002,
+      "loss": 0.8154,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8529945553539022,
+      "grad_norm": 1.2902015447616577,
+      "learning_rate": 0.0002,
+      "loss": 0.9328,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8602540834845733,
+      "grad_norm": 0.5252797603607178,
+      "learning_rate": 0.0002,
+      "loss": 0.8192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.867513611615245,
+      "grad_norm": 0.5925108790397644,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8747731397459164,
+      "grad_norm": 0.9719610810279846,
+      "learning_rate": 0.0002,
+      "loss": 0.8611,
+      "step": 3960
+    },
+    {
+      "epoch": 2.882032667876588,
+      "grad_norm": 0.6834747195243835,
+      "learning_rate": 0.0002,
+      "loss": 0.8531,
+      "step": 3970
+    },
+    {
+      "epoch": 2.8892921960072595,
+      "grad_norm": 0.6952353119850159,
+      "learning_rate": 0.0002,
+      "loss": 0.8124,
+      "step": 3980
+    },
+    {
+      "epoch": 2.896551724137931,
+      "grad_norm": 0.49889910221099854,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9038112522686026,
+      "grad_norm": 0.5007026791572571,
+      "learning_rate": 0.0002,
+      "loss": 0.8197,
+      "step": 4000
+    },
+    {
+      "epoch": 2.911070780399274,
+      "grad_norm": 0.5474239587783813,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9183303085299457,
+      "grad_norm": 0.6617428660392761,
+      "learning_rate": 0.0002,
+      "loss": 0.8579,
+      "step": 4020
+    },
+    {
+      "epoch": 2.925589836660617,
+      "grad_norm": 0.6097776293754578,
+      "learning_rate": 0.0002,
+      "loss": 0.8583,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9328493647912888,
+      "grad_norm": 0.5985828638076782,
+      "learning_rate": 0.0002,
+      "loss": 0.8296,
+      "step": 4040
+    },
+    {
+      "epoch": 2.94010889292196,
+      "grad_norm": 0.769488513469696,
+      "learning_rate": 0.0002,
+      "loss": 0.9134,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 0.5167055130004883,
+      "learning_rate": 0.0002,
+      "loss": 0.8148,
+      "step": 4060
+    },
+    {
+      "epoch": 2.954627949183303,
+      "grad_norm": 0.6265496015548706,
+      "learning_rate": 0.0002,
+      "loss": 0.8665,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9618874773139745,
+      "grad_norm": 1.2644082307815552,
+      "learning_rate": 0.0002,
+      "loss": 0.9218,
+      "step": 4080
+    },
+    {
+      "epoch": 2.969147005444646,
+      "grad_norm": 0.6007561087608337,
+      "learning_rate": 0.0002,
+      "loss": 0.8026,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9764065335753176,
+      "grad_norm": 0.47984927892684937,
+      "learning_rate": 0.0002,
+      "loss": 0.8262,
+      "step": 4100
+    },
+    {
+      "epoch": 2.983666061705989,
+      "grad_norm": 1.128198504447937,
+      "learning_rate": 0.0002,
+      "loss": 0.7897,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9909255898366607,
+      "grad_norm": 0.526292085647583,
+      "learning_rate": 0.0002,
+      "loss": 0.7883,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9981851179673322,
+      "grad_norm": 0.5462674498558044,
+      "learning_rate": 0.0002,
+      "loss": 0.7801,
+      "step": 4130
+    },
+    {
+      "epoch": 2.9996370235934666,
+      "eval_loss": 1.1093357801437378,
+      "eval_runtime": 46.2498,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 4132
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11016,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.122606788083712e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..140bceabdec44fc473842cb9ace4a62a4dc9fe89
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-4132/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4abec5480aff960b5a7b03a48be1cdb1ae29ba8792469c451f5c2f269e4be93b
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7ea497b2c459907a15aa8a30d556a9e94b6b1d59
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ace6f198b7b601c431a919517fd6ec1f8a057a534c1ecefd21c2c303d649b96f
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..bb80c05d5b79d05ada17b8744555e2d8cd23b153
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05edafa7d9883453eabeb3309388918a362ca33bf36981f47d6f6feb0faeb3ef
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..5969ad87c7b111906b62d5bcd54e63d4f91b81de
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2961f1b3aaeaa95bebc77ef91a7b5d990dfe95bcaa7e2ad9a7fb22fa5ae8d2cf
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..c580d4796cbe45bdcb2a39acc06afba66279aab5
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35b86eb7ec49369219fd6c7b5185fd429cf5e2b9681284ddd6335300e9f880a5
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..446926ed076157d6bfdacd2e3ff5459f6c67760f
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/trainer_state.json
@@ -0,0 +1,3922 @@
+{
+  "best_metric": 1.1086540222167969,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 5510,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007259528130671506,
+      "grad_norm": 0.46010470390319824,
+      "learning_rate": 0.0002,
+      "loss": 2.4936,
+      "step": 10
+    },
+    {
+      "epoch": 0.014519056261343012,
+      "grad_norm": 0.5103179216384888,
+      "learning_rate": 0.0002,
+      "loss": 1.8442,
+      "step": 20
+    },
+    {
+      "epoch": 0.021778584392014518,
+      "grad_norm": 0.6282716989517212,
+      "learning_rate": 0.0002,
+      "loss": 1.559,
+      "step": 30
+    },
+    {
+      "epoch": 0.029038112522686024,
+      "grad_norm": 1.2480497360229492,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 40
+    },
+    {
+      "epoch": 0.036297640653357534,
+      "grad_norm": 0.4114855229854584,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 50
+    },
+    {
+      "epoch": 0.043557168784029036,
+      "grad_norm": 0.49482840299606323,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 60
+    },
+    {
+      "epoch": 0.050816696914700546,
+      "grad_norm": 0.4536272883415222,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 70
+    },
+    {
+      "epoch": 0.05807622504537205,
+      "grad_norm": 0.32328274846076965,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 80
+    },
+    {
+      "epoch": 0.06533575317604355,
+      "grad_norm": 0.40990468859672546,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 90
+    },
+    {
+      "epoch": 0.07259528130671507,
+      "grad_norm": 0.37273502349853516,
+      "learning_rate": 0.0002,
+      "loss": 1.2216,
+      "step": 100
+    },
+    {
+      "epoch": 0.07985480943738657,
+      "grad_norm": 0.3903871476650238,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 110
+    },
+    {
+      "epoch": 0.08711433756805807,
+      "grad_norm": 0.3272787928581238,
+      "learning_rate": 0.0002,
+      "loss": 1.3718,
+      "step": 120
+    },
+    {
+      "epoch": 0.09437386569872959,
+      "grad_norm": 0.3622824251651764,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 130
+    },
+    {
+      "epoch": 0.10163339382940109,
+      "grad_norm": 0.3503916561603546,
+      "learning_rate": 0.0002,
+      "loss": 1.2966,
+      "step": 140
+    },
+    {
+      "epoch": 0.1088929219600726,
+      "grad_norm": 0.32787832617759705,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 150
+    },
+    {
+      "epoch": 0.1161524500907441,
+      "grad_norm": 0.5822657942771912,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 160
+    },
+    {
+      "epoch": 0.12341197822141561,
+      "grad_norm": 0.28028249740600586,
+      "learning_rate": 0.0002,
+      "loss": 1.2091,
+      "step": 170
+    },
+    {
+      "epoch": 0.1306715063520871,
+      "grad_norm": 0.33602750301361084,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 180
+    },
+    {
+      "epoch": 0.13793103448275862,
+      "grad_norm": 0.29106274247169495,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 190
+    },
+    {
+      "epoch": 0.14519056261343014,
+      "grad_norm": 0.38753363490104675,
+      "learning_rate": 0.0002,
+      "loss": 1.3261,
+      "step": 200
+    },
+    {
+      "epoch": 0.15245009074410162,
+      "grad_norm": 0.361009418964386,
+      "learning_rate": 0.0002,
+      "loss": 1.1536,
+      "step": 210
+    },
+    {
+      "epoch": 0.15970961887477314,
+      "grad_norm": 0.6743836402893066,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 220
+    },
+    {
+      "epoch": 0.16696914700544466,
+      "grad_norm": 0.3923613727092743,
+      "learning_rate": 0.0002,
+      "loss": 1.264,
+      "step": 230
+    },
+    {
+      "epoch": 0.17422867513611615,
+      "grad_norm": 0.2809699773788452,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 240
+    },
+    {
+      "epoch": 0.18148820326678766,
+      "grad_norm": 0.3631494641304016,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 250
+    },
+    {
+      "epoch": 0.18874773139745918,
+      "grad_norm": 0.24658171832561493,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 260
+    },
+    {
+      "epoch": 0.19600725952813067,
+      "grad_norm": 0.5780664682388306,
+      "learning_rate": 0.0002,
+      "loss": 1.0065,
+      "step": 270
+    },
+    {
+      "epoch": 0.20326678765880218,
+      "grad_norm": 0.3056720495223999,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 280
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 0.2857084572315216,
+      "learning_rate": 0.0002,
+      "loss": 1.1503,
+      "step": 290
+    },
+    {
+      "epoch": 0.2177858439201452,
+      "grad_norm": 0.6645345687866211,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 300
+    },
+    {
+      "epoch": 0.2250453720508167,
+      "grad_norm": 0.5966078639030457,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 310
+    },
+    {
+      "epoch": 0.2323049001814882,
+      "grad_norm": 0.40937140583992004,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 320
+    },
+    {
+      "epoch": 0.2395644283121597,
+      "grad_norm": 0.5642806887626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 330
+    },
+    {
+      "epoch": 0.24682395644283123,
+      "grad_norm": 0.2750748097896576,
+      "learning_rate": 0.0002,
+      "loss": 1.1703,
+      "step": 340
+    },
+    {
+      "epoch": 0.2540834845735027,
+      "grad_norm": 0.34350234270095825,
+      "learning_rate": 0.0002,
+      "loss": 1.2598,
+      "step": 350
+    },
+    {
+      "epoch": 0.2613430127041742,
+      "grad_norm": 0.6767239570617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1942,
+      "step": 360
+    },
+    {
+      "epoch": 0.26860254083484575,
+      "grad_norm": 0.31006959080696106,
+      "learning_rate": 0.0002,
+      "loss": 1.1436,
+      "step": 370
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "grad_norm": 0.3825474679470062,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 380
+    },
+    {
+      "epoch": 0.2831215970961887,
+      "grad_norm": 0.30241551995277405,
+      "learning_rate": 0.0002,
+      "loss": 1.1523,
+      "step": 390
+    },
+    {
+      "epoch": 0.29038112522686027,
+      "grad_norm": 0.2962397336959839,
+      "learning_rate": 0.0002,
+      "loss": 1.1298,
+      "step": 400
+    },
+    {
+      "epoch": 0.29764065335753176,
+      "grad_norm": 0.2600369155406952,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 410
+    },
+    {
+      "epoch": 0.30490018148820325,
+      "grad_norm": 0.3675060272216797,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 420
+    },
+    {
+      "epoch": 0.3121597096188748,
+      "grad_norm": 0.3429498076438904,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 430
+    },
+    {
+      "epoch": 0.3194192377495463,
+      "grad_norm": 0.34311825037002563,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 440
+    },
+    {
+      "epoch": 0.32667876588021777,
+      "grad_norm": 0.37872210144996643,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 450
+    },
+    {
+      "epoch": 0.3339382940108893,
+      "grad_norm": 0.33271121978759766,
+      "learning_rate": 0.0002,
+      "loss": 1.1365,
+      "step": 460
+    },
+    {
+      "epoch": 0.3411978221415608,
+      "grad_norm": 0.34605276584625244,
+      "learning_rate": 0.0002,
+      "loss": 1.1572,
+      "step": 470
+    },
+    {
+      "epoch": 0.3484573502722323,
+      "grad_norm": 0.41050562262535095,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 480
+    },
+    {
+      "epoch": 0.35571687840290384,
+      "grad_norm": 0.2066836953163147,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 490
+    },
+    {
+      "epoch": 0.3629764065335753,
+      "grad_norm": 0.2859014868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.2072,
+      "step": 500
+    },
+    {
+      "epoch": 0.3702359346642468,
+      "grad_norm": 0.28763777017593384,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 510
+    },
+    {
+      "epoch": 0.37749546279491836,
+      "grad_norm": 0.2730471193790436,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 520
+    },
+    {
+      "epoch": 0.38475499092558985,
+      "grad_norm": 0.3968936800956726,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 530
+    },
+    {
+      "epoch": 0.39201451905626133,
+      "grad_norm": 0.3624701201915741,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 540
+    },
+    {
+      "epoch": 0.3992740471869328,
+      "grad_norm": 0.3303608298301697,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 550
+    },
+    {
+      "epoch": 0.40653357531760437,
+      "grad_norm": 0.33507466316223145,
+      "learning_rate": 0.0002,
+      "loss": 1.203,
+      "step": 560
+    },
+    {
+      "epoch": 0.41379310344827586,
+      "grad_norm": 0.3297670781612396,
+      "learning_rate": 0.0002,
+      "loss": 1.2294,
+      "step": 570
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.32334890961647034,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 580
+    },
+    {
+      "epoch": 0.4283121597096189,
+      "grad_norm": 0.30281195044517517,
+      "learning_rate": 0.0002,
+      "loss": 1.112,
+      "step": 590
+    },
+    {
+      "epoch": 0.4355716878402904,
+      "grad_norm": 0.5900027751922607,
+      "learning_rate": 0.0002,
+      "loss": 1.1352,
+      "step": 600
+    },
+    {
+      "epoch": 0.44283121597096187,
+      "grad_norm": 0.28437477350234985,
+      "learning_rate": 0.0002,
+      "loss": 1.1575,
+      "step": 610
+    },
+    {
+      "epoch": 0.4500907441016334,
+      "grad_norm": 0.39601704478263855,
+      "learning_rate": 0.0002,
+      "loss": 1.2009,
+      "step": 620
+    },
+    {
+      "epoch": 0.4573502722323049,
+      "grad_norm": 0.41971510648727417,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 630
+    },
+    {
+      "epoch": 0.4646098003629764,
+      "grad_norm": 0.33814409375190735,
+      "learning_rate": 0.0002,
+      "loss": 1.236,
+      "step": 640
+    },
+    {
+      "epoch": 0.47186932849364793,
+      "grad_norm": 0.575718104839325,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 650
+    },
+    {
+      "epoch": 0.4791288566243194,
+      "grad_norm": 0.37927401065826416,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 660
+    },
+    {
+      "epoch": 0.4863883847549909,
+      "grad_norm": 0.3224332630634308,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 670
+    },
+    {
+      "epoch": 0.49364791288566245,
+      "grad_norm": 0.32683515548706055,
+      "learning_rate": 0.0002,
+      "loss": 1.1354,
+      "step": 680
+    },
+    {
+      "epoch": 0.5009074410163339,
+      "grad_norm": 0.4316163659095764,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 690
+    },
+    {
+      "epoch": 0.5081669691470054,
+      "grad_norm": 0.342602401971817,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 700
+    },
+    {
+      "epoch": 0.515426497277677,
+      "grad_norm": 0.2794898748397827,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 710
+    },
+    {
+      "epoch": 0.5226860254083484,
+      "grad_norm": 0.3322339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.2395,
+      "step": 720
+    },
+    {
+      "epoch": 0.52994555353902,
+      "grad_norm": 0.3088509142398834,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 730
+    },
+    {
+      "epoch": 0.5372050816696915,
+      "grad_norm": 0.24444378912448883,
+      "learning_rate": 0.0002,
+      "loss": 1.1495,
+      "step": 740
+    },
+    {
+      "epoch": 0.5444646098003629,
+      "grad_norm": 0.3483171761035919,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 750
+    },
+    {
+      "epoch": 0.5517241379310345,
+      "grad_norm": 0.3425690531730652,
+      "learning_rate": 0.0002,
+      "loss": 1.2248,
+      "step": 760
+    },
+    {
+      "epoch": 0.558983666061706,
+      "grad_norm": 0.31841927766799927,
+      "learning_rate": 0.0002,
+      "loss": 1.1127,
+      "step": 770
+    },
+    {
+      "epoch": 0.5662431941923775,
+      "grad_norm": 0.39423868060112,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 780
+    },
+    {
+      "epoch": 0.573502722323049,
+      "grad_norm": 0.30328479409217834,
+      "learning_rate": 0.0002,
+      "loss": 1.0583,
+      "step": 790
+    },
+    {
+      "epoch": 0.5807622504537205,
+      "grad_norm": 0.24475938081741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0718,
+      "step": 800
+    },
+    {
+      "epoch": 0.588021778584392,
+      "grad_norm": 0.37132805585861206,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 810
+    },
+    {
+      "epoch": 0.5952813067150635,
+      "grad_norm": 0.32195979356765747,
+      "learning_rate": 0.0002,
+      "loss": 1.1544,
+      "step": 820
+    },
+    {
+      "epoch": 0.6025408348457351,
+      "grad_norm": 0.2848738729953766,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 830
+    },
+    {
+      "epoch": 0.6098003629764065,
+      "grad_norm": 0.28015264868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.0937,
+      "step": 840
+    },
+    {
+      "epoch": 0.617059891107078,
+      "grad_norm": 0.37796008586883545,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 850
+    },
+    {
+      "epoch": 0.6243194192377496,
+      "grad_norm": 0.39311841130256653,
+      "learning_rate": 0.0002,
+      "loss": 1.1455,
+      "step": 860
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.2761685252189636,
+      "learning_rate": 0.0002,
+      "loss": 1.1066,
+      "step": 870
+    },
+    {
+      "epoch": 0.6388384754990926,
+      "grad_norm": 0.3826720118522644,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 880
+    },
+    {
+      "epoch": 0.6460980036297641,
+      "grad_norm": 0.30076679587364197,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 890
+    },
+    {
+      "epoch": 0.6533575317604355,
+      "grad_norm": 0.21997687220573425,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 900
+    },
+    {
+      "epoch": 0.6606170598911071,
+      "grad_norm": 0.32593777775764465,
+      "learning_rate": 0.0002,
+      "loss": 1.1603,
+      "step": 910
+    },
+    {
+      "epoch": 0.6678765880217786,
+      "grad_norm": 0.30347898602485657,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 920
+    },
+    {
+      "epoch": 0.6751361161524501,
+      "grad_norm": 0.44173774123191833,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 930
+    },
+    {
+      "epoch": 0.6823956442831216,
+      "grad_norm": 0.2507467269897461,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 940
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.29463833570480347,
+      "learning_rate": 0.0002,
+      "loss": 1.3001,
+      "step": 950
+    },
+    {
+      "epoch": 0.6969147005444646,
+      "grad_norm": 0.9363154172897339,
+      "learning_rate": 0.0002,
+      "loss": 1.1651,
+      "step": 960
+    },
+    {
+      "epoch": 0.7041742286751361,
+      "grad_norm": 0.3236212134361267,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 970
+    },
+    {
+      "epoch": 0.7114337568058077,
+      "grad_norm": 0.3123254179954529,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 980
+    },
+    {
+      "epoch": 0.7186932849364791,
+      "grad_norm": 0.3395805060863495,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 990
+    },
+    {
+      "epoch": 0.7259528130671506,
+      "grad_norm": 0.3240964412689209,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7332123411978222,
+      "grad_norm": 0.31902948021888733,
+      "learning_rate": 0.0002,
+      "loss": 1.0313,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7404718693284936,
+      "grad_norm": 0.4848408102989197,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7477313974591652,
+      "grad_norm": 0.33006033301353455,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7549909255898367,
+      "grad_norm": 0.2928730547428131,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7622504537205081,
+      "grad_norm": 0.3529164791107178,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7695099818511797,
+      "grad_norm": 0.2736213803291321,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7767695099818511,
+      "grad_norm": 0.7200686931610107,
+      "learning_rate": 0.0002,
+      "loss": 1.1169,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7840290381125227,
+      "grad_norm": 0.33396708965301514,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7912885662431942,
+      "grad_norm": 1.5760449171066284,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7985480943738656,
+      "grad_norm": 0.28138381242752075,
+      "learning_rate": 0.0002,
+      "loss": 1.1126,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8058076225045372,
+      "grad_norm": 0.2597472369670868,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8130671506352087,
+      "grad_norm": 0.3305445611476898,
+      "learning_rate": 0.0002,
+      "loss": 1.1177,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8203266787658802,
+      "grad_norm": 0.3934599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.0849,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8275862068965517,
+      "grad_norm": 0.3472191393375397,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8348457350272233,
+      "grad_norm": 0.2857365906238556,
+      "learning_rate": 0.0002,
+      "loss": 1.265,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.3207702934741974,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8493647912885662,
+      "grad_norm": 0.3176484704017639,
+      "learning_rate": 0.0002,
+      "loss": 1.1669,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8566243194192378,
+      "grad_norm": 0.40685558319091797,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8638838475499092,
+      "grad_norm": 0.31125199794769287,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8711433756805808,
+      "grad_norm": 0.7361181378364563,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8784029038112523,
+      "grad_norm": 0.33699527382850647,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8856624319419237,
+      "grad_norm": 0.3315220773220062,
+      "learning_rate": 0.0002,
+      "loss": 1.11,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8929219600725953,
+      "grad_norm": 0.6256054043769836,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9001814882032668,
+      "grad_norm": 0.3692137897014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1738,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9074410163339383,
+      "grad_norm": 0.3538484573364258,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9147005444646098,
+      "grad_norm": 0.27863040566444397,
+      "learning_rate": 0.0002,
+      "loss": 1.0878,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9219600725952813,
+      "grad_norm": 0.3322528302669525,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9292196007259528,
+      "grad_norm": 0.7553173303604126,
+      "learning_rate": 0.0002,
+      "loss": 1.0158,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9364791288566243,
+      "grad_norm": 0.4856191575527191,
+      "learning_rate": 0.0002,
+      "loss": 1.198,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9437386569872959,
+      "grad_norm": 0.3668074905872345,
+      "learning_rate": 0.0002,
+      "loss": 1.1678,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9509981851179673,
+      "grad_norm": 0.29851067066192627,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9582577132486388,
+      "grad_norm": 0.276664674282074,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9655172413793104,
+      "grad_norm": 0.2941018342971802,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9727767695099818,
+      "grad_norm": 0.3505859076976776,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9800362976406534,
+      "grad_norm": 0.3067687451839447,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9872958257713249,
+      "grad_norm": 0.27151066064834595,
+      "learning_rate": 0.0002,
+      "loss": 1.1511,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9945553539019963,
+      "grad_norm": 0.36370083689689636,
+      "learning_rate": 0.0002,
+      "loss": 1.1919,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9996370235934664,
+      "eval_loss": 1.1381088495254517,
+      "eval_runtime": 96.0848,
+      "eval_samples_per_second": 4.538,
+      "eval_steps_per_second": 0.572,
+      "step": 1377
+    },
+    {
+      "epoch": 1.0018148820326678,
+      "grad_norm": 0.27980583906173706,
+      "learning_rate": 0.0002,
+      "loss": 1.0082,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0090744101633393,
+      "grad_norm": 0.26713913679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0163339382940109,
+      "grad_norm": 0.3089541494846344,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0235934664246824,
+      "grad_norm": 0.4188242256641388,
+      "learning_rate": 0.0002,
+      "loss": 0.9649,
+      "step": 1410
+    },
+    {
+      "epoch": 1.030852994555354,
+      "grad_norm": 0.5246463418006897,
+      "learning_rate": 0.0002,
+      "loss": 0.9821,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0381125226860255,
+      "grad_norm": 0.2728777825832367,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0453720508166968,
+      "grad_norm": 0.38167616724967957,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.4439380168914795,
+      "learning_rate": 0.0002,
+      "loss": 1.0229,
+      "step": 1450
+    },
+    {
+      "epoch": 1.05989110707804,
+      "grad_norm": 0.30954182147979736,
+      "learning_rate": 0.0002,
+      "loss": 1.0451,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0671506352087115,
+      "grad_norm": 0.4022280275821686,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 1470
+    },
+    {
+      "epoch": 1.074410163339383,
+      "grad_norm": 0.7390811443328857,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0816696914700545,
+      "grad_norm": 0.3885486423969269,
+      "learning_rate": 0.0002,
+      "loss": 1.1378,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0889292196007259,
+      "grad_norm": 0.5275560617446899,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0961887477313974,
+      "grad_norm": 0.35112282633781433,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1510
+    },
+    {
+      "epoch": 1.103448275862069,
+      "grad_norm": 0.33714351058006287,
+      "learning_rate": 0.0002,
+      "loss": 1.0331,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1107078039927405,
+      "grad_norm": 0.31221693754196167,
+      "learning_rate": 0.0002,
+      "loss": 0.8723,
+      "step": 1530
+    },
+    {
+      "epoch": 1.117967332123412,
+      "grad_norm": 0.27549654245376587,
+      "learning_rate": 0.0002,
+      "loss": 1.0701,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1252268602540836,
+      "grad_norm": 0.8465521335601807,
+      "learning_rate": 0.0002,
+      "loss": 1.0114,
+      "step": 1550
+    },
+    {
+      "epoch": 1.132486388384755,
+      "grad_norm": 0.36125949025154114,
+      "learning_rate": 0.0002,
+      "loss": 0.9643,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1397459165154264,
+      "grad_norm": 0.37420371174812317,
+      "learning_rate": 0.0002,
+      "loss": 1.0744,
+      "step": 1570
+    },
+    {
+      "epoch": 1.147005444646098,
+      "grad_norm": 0.3294760584831238,
+      "learning_rate": 0.0002,
+      "loss": 1.165,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1542649727767695,
+      "grad_norm": 0.3881238102912903,
+      "learning_rate": 0.0002,
+      "loss": 0.9207,
+      "step": 1590
+    },
+    {
+      "epoch": 1.161524500907441,
+      "grad_norm": 0.4766491651535034,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1687840290381124,
+      "grad_norm": 0.395530104637146,
+      "learning_rate": 0.0002,
+      "loss": 1.0092,
+      "step": 1610
+    },
+    {
+      "epoch": 1.176043557168784,
+      "grad_norm": 0.3297106623649597,
+      "learning_rate": 0.0002,
+      "loss": 0.9895,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1833030852994555,
+      "grad_norm": 0.39528271555900574,
+      "learning_rate": 0.0002,
+      "loss": 1.1121,
+      "step": 1630
+    },
+    {
+      "epoch": 1.190562613430127,
+      "grad_norm": 0.3370221257209778,
+      "learning_rate": 0.0002,
+      "loss": 0.9202,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1978221415607986,
+      "grad_norm": 0.31922030448913574,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2050816696914701,
+      "grad_norm": 0.6142027378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2123411978221417,
+      "grad_norm": 0.44769710302352905,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 1670
+    },
+    {
+      "epoch": 1.219600725952813,
+      "grad_norm": 0.41233646869659424,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2268602540834845,
+      "grad_norm": 0.2928866147994995,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1690
+    },
+    {
+      "epoch": 1.234119782214156,
+      "grad_norm": 0.36913734674453735,
+      "learning_rate": 0.0002,
+      "loss": 0.9074,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2413793103448276,
+      "grad_norm": 0.5281891226768494,
+      "learning_rate": 0.0002,
+      "loss": 1.0847,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2486388384754992,
+      "grad_norm": 0.3374697268009186,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2558983666061705,
+      "grad_norm": 0.3802020847797394,
+      "learning_rate": 0.0002,
+      "loss": 0.9826,
+      "step": 1730
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.38048651814460754,
+      "learning_rate": 0.0002,
+      "loss": 1.0191,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2704174228675136,
+      "grad_norm": 0.6676169633865356,
+      "learning_rate": 0.0002,
+      "loss": 1.0142,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2776769509981851,
+      "grad_norm": 0.4075961410999298,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2849364791288567,
+      "grad_norm": 0.4374721348285675,
+      "learning_rate": 0.0002,
+      "loss": 1.0301,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2921960072595282,
+      "grad_norm": 0.4638824164867401,
+      "learning_rate": 0.0002,
+      "loss": 1.0601,
+      "step": 1780
+    },
+    {
+      "epoch": 1.2994555353901998,
+      "grad_norm": 0.38631564378738403,
+      "learning_rate": 0.0002,
+      "loss": 0.9593,
+      "step": 1790
+    },
+    {
+      "epoch": 1.306715063520871,
+      "grad_norm": 0.35873809456825256,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3139745916515426,
+      "grad_norm": 0.33602237701416016,
+      "learning_rate": 0.0002,
+      "loss": 0.9835,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3212341197822142,
+      "grad_norm": 0.46696463227272034,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3284936479128857,
+      "grad_norm": 0.368958979845047,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 1830
+    },
+    {
+      "epoch": 1.335753176043557,
+      "grad_norm": 0.3160957396030426,
+      "learning_rate": 0.0002,
+      "loss": 1.0456,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3430127041742286,
+      "grad_norm": 0.4511511027812958,
+      "learning_rate": 0.0002,
+      "loss": 1.0264,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3502722323049001,
+      "grad_norm": 0.6769845485687256,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3575317604355717,
+      "grad_norm": 0.3749309480190277,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3647912885662432,
+      "grad_norm": 0.553798496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.0223,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3720508166969148,
+      "grad_norm": 0.3538985550403595,
+      "learning_rate": 0.0002,
+      "loss": 0.9634,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.5501534938812256,
+      "learning_rate": 0.0002,
+      "loss": 1.0275,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3865698729582578,
+      "grad_norm": 0.4432051181793213,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3938294010889292,
+      "grad_norm": 0.41755786538124084,
+      "learning_rate": 0.0002,
+      "loss": 0.9082,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4010889292196007,
+      "grad_norm": 0.5626114010810852,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4083484573502723,
+      "grad_norm": 0.44038185477256775,
+      "learning_rate": 0.0002,
+      "loss": 1.0687,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4156079854809438,
+      "grad_norm": 0.3397001624107361,
+      "learning_rate": 0.0002,
+      "loss": 0.9454,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4228675136116151,
+      "grad_norm": 0.4325368106365204,
+      "learning_rate": 0.0002,
+      "loss": 0.943,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4301270417422867,
+      "grad_norm": 0.3900907039642334,
+      "learning_rate": 0.0002,
+      "loss": 1.0488,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4373865698729582,
+      "grad_norm": 0.369612455368042,
+      "learning_rate": 0.0002,
+      "loss": 0.9699,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4446460980036298,
+      "grad_norm": 0.4389338791370392,
+      "learning_rate": 0.0002,
+      "loss": 1.0609,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4519056261343013,
+      "grad_norm": 1.694450855255127,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4591651542649728,
+      "grad_norm": 0.516957700252533,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4664246823956444,
+      "grad_norm": 0.45515501499176025,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.4153139591217041,
+      "learning_rate": 0.0002,
+      "loss": 1.0188,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4809437386569873,
+      "grad_norm": 0.44353052973747253,
+      "learning_rate": 0.0002,
+      "loss": 1.1521,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4882032667876588,
+      "grad_norm": 0.570554256439209,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4954627949183303,
+      "grad_norm": 0.5742740035057068,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5027223230490017,
+      "grad_norm": 0.5890517830848694,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5099818511796732,
+      "grad_norm": 0.4162650406360626,
+      "learning_rate": 0.0002,
+      "loss": 1.0639,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5172413793103448,
+      "grad_norm": 0.7334717512130737,
+      "learning_rate": 0.0002,
+      "loss": 1.071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5245009074410163,
+      "grad_norm": 0.2963249683380127,
+      "learning_rate": 0.0002,
+      "loss": 0.8957,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5317604355716878,
+      "grad_norm": 0.30676454305648804,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5390199637023594,
+      "grad_norm": 0.35984641313552856,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 2120
+    },
+    {
+      "epoch": 1.546279491833031,
+      "grad_norm": 0.3384549617767334,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5535390199637025,
+      "grad_norm": 0.4725518226623535,
+      "learning_rate": 0.0002,
+      "loss": 0.9874,
+      "step": 2140
+    },
+    {
+      "epoch": 1.560798548094374,
+      "grad_norm": 0.3252685070037842,
+      "learning_rate": 0.0002,
+      "loss": 1.1166,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5680580762250453,
+      "grad_norm": 0.45043081045150757,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5753176043557169,
+      "grad_norm": 0.374208003282547,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5825771324863884,
+      "grad_norm": 0.5118404030799866,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5898366606170597,
+      "grad_norm": 0.482170969247818,
+      "learning_rate": 0.0002,
+      "loss": 1.0062,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5970961887477313,
+      "grad_norm": 0.5337533950805664,
+      "learning_rate": 0.0002,
+      "loss": 1.0872,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6043557168784028,
+      "grad_norm": 0.5195064544677734,
+      "learning_rate": 0.0002,
+      "loss": 1.0405,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6116152450090744,
+      "grad_norm": 0.30807098746299744,
+      "learning_rate": 0.0002,
+      "loss": 1.0454,
+      "step": 2220
+    },
+    {
+      "epoch": 1.618874773139746,
+      "grad_norm": 0.3962925672531128,
+      "learning_rate": 0.0002,
+      "loss": 1.0293,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6261343012704175,
+      "grad_norm": 0.7636962532997131,
+      "learning_rate": 0.0002,
+      "loss": 1.0137,
+      "step": 2240
+    },
+    {
+      "epoch": 1.633393829401089,
+      "grad_norm": 0.32380592823028564,
+      "learning_rate": 0.0002,
+      "loss": 0.991,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6406533575317606,
+      "grad_norm": 0.5767741799354553,
+      "learning_rate": 0.0002,
+      "loss": 0.9471,
+      "step": 2260
+    },
+    {
+      "epoch": 1.647912885662432,
+      "grad_norm": 0.39964812994003296,
+      "learning_rate": 0.0002,
+      "loss": 0.9524,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6551724137931034,
+      "grad_norm": 0.622629702091217,
+      "learning_rate": 0.0002,
+      "loss": 0.9866,
+      "step": 2280
+    },
+    {
+      "epoch": 1.662431941923775,
+      "grad_norm": 0.40202152729034424,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6696914700544465,
+      "grad_norm": 0.4467349052429199,
+      "learning_rate": 0.0002,
+      "loss": 0.9859,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6769509981851178,
+      "grad_norm": 0.5026949048042297,
+      "learning_rate": 0.0002,
+      "loss": 1.0312,
+      "step": 2310
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.43754541873931885,
+      "learning_rate": 0.0002,
+      "loss": 1.0776,
+      "step": 2320
+    },
+    {
+      "epoch": 1.691470054446461,
+      "grad_norm": 0.42869430780410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6987295825771325,
+      "grad_norm": 0.4192679524421692,
+      "learning_rate": 0.0002,
+      "loss": 1.0747,
+      "step": 2340
+    },
+    {
+      "epoch": 1.705989110707804,
+      "grad_norm": 0.3243155777454376,
+      "learning_rate": 0.0002,
+      "loss": 1.0054,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7132486388384756,
+      "grad_norm": 1.0514075756072998,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 2360
+    },
+    {
+      "epoch": 1.720508166969147,
+      "grad_norm": 0.4719122648239136,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7277676950998186,
+      "grad_norm": 0.3846144676208496,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 2380
+    },
+    {
+      "epoch": 1.73502722323049,
+      "grad_norm": 0.7266581058502197,
+      "learning_rate": 0.0002,
+      "loss": 0.9935,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7422867513611615,
+      "grad_norm": 0.6183241605758667,
+      "learning_rate": 0.0002,
+      "loss": 0.9509,
+      "step": 2400
+    },
+    {
+      "epoch": 1.749546279491833,
+      "grad_norm": 0.3658260405063629,
+      "learning_rate": 0.0002,
+      "loss": 0.9713,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7568058076225044,
+      "grad_norm": 0.6036322712898254,
+      "learning_rate": 0.0002,
+      "loss": 1.0033,
+      "step": 2420
+    },
+    {
+      "epoch": 1.764065335753176,
+      "grad_norm": 0.7872936129570007,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7713248638838475,
+      "grad_norm": 0.35946124792099,
+      "learning_rate": 0.0002,
+      "loss": 0.99,
+      "step": 2440
+    },
+    {
+      "epoch": 1.778584392014519,
+      "grad_norm": 0.3740338981151581,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2450
+    },
+    {
+      "epoch": 1.7858439201451906,
+      "grad_norm": 0.6150230169296265,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 2460
+    },
+    {
+      "epoch": 1.793103448275862,
+      "grad_norm": 0.4726075530052185,
+      "learning_rate": 0.0002,
+      "loss": 1.0373,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8003629764065336,
+      "grad_norm": 0.31292253732681274,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8076225045372052,
+      "grad_norm": 0.4463104009628296,
+      "learning_rate": 0.0002,
+      "loss": 1.0205,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8148820326678767,
+      "grad_norm": 0.7848200798034668,
+      "learning_rate": 0.0002,
+      "loss": 1.0771,
+      "step": 2500
+    },
+    {
+      "epoch": 1.822141560798548,
+      "grad_norm": 0.5562082529067993,
+      "learning_rate": 0.0002,
+      "loss": 0.9508,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8294010889292196,
+      "grad_norm": 0.39892756938934326,
+      "learning_rate": 0.0002,
+      "loss": 0.9343,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8366606170598911,
+      "grad_norm": 1.2923320531845093,
+      "learning_rate": 0.0002,
+      "loss": 1.0489,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8439201451905625,
+      "grad_norm": 0.6316490769386292,
+      "learning_rate": 0.0002,
+      "loss": 1.1203,
+      "step": 2540
+    },
+    {
+      "epoch": 1.851179673321234,
+      "grad_norm": 0.46100497245788574,
+      "learning_rate": 0.0002,
+      "loss": 0.9131,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8584392014519056,
+      "grad_norm": 0.7902987003326416,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 2560
+    },
+    {
+      "epoch": 1.865698729582577,
+      "grad_norm": 0.4596365690231323,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8729582577132486,
+      "grad_norm": 0.6592172384262085,
+      "learning_rate": 0.0002,
+      "loss": 0.9322,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8802177858439202,
+      "grad_norm": 0.5764662027359009,
+      "learning_rate": 0.0002,
+      "loss": 0.9549,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8874773139745917,
+      "grad_norm": 0.8421637415885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9485,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.5635305047035217,
+      "learning_rate": 0.0002,
+      "loss": 1.012,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9019963702359348,
+      "grad_norm": 0.46231237053871155,
+      "learning_rate": 0.0002,
+      "loss": 0.8907,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9092558983666061,
+      "grad_norm": 0.3944607079029083,
+      "learning_rate": 0.0002,
+      "loss": 0.9543,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9165154264972777,
+      "grad_norm": 0.4753907322883606,
+      "learning_rate": 0.0002,
+      "loss": 0.9964,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9237749546279492,
+      "grad_norm": 0.4151090979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.0217,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9310344827586206,
+      "grad_norm": 0.6793725490570068,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 2660
+    },
+    {
+      "epoch": 1.938294010889292,
+      "grad_norm": 0.339755117893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0401,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9455535390199636,
+      "grad_norm": 0.40789374709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.082,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9528130671506352,
+      "grad_norm": 0.3750005066394806,
+      "learning_rate": 0.0002,
+      "loss": 0.9973,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9600725952813067,
+      "grad_norm": 0.39684441685676575,
+      "learning_rate": 0.0002,
+      "loss": 0.9837,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9673321234119783,
+      "grad_norm": 0.378287672996521,
+      "learning_rate": 0.0002,
+      "loss": 1.1204,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9745916515426498,
+      "grad_norm": 0.3668482005596161,
+      "learning_rate": 0.0002,
+      "loss": 0.9284,
+      "step": 2720
+    },
+    {
+      "epoch": 1.9818511796733214,
+      "grad_norm": 0.49997565150260925,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 2730
+    },
+    {
+      "epoch": 1.989110707803993,
+      "grad_norm": 0.36852124333381653,
+      "learning_rate": 0.0002,
+      "loss": 0.9311,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9963702359346642,
+      "grad_norm": 0.5203380584716797,
+      "learning_rate": 0.0002,
+      "loss": 1.0898,
+      "step": 2750
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1086540222167969,
+      "eval_runtime": 95.6866,
+      "eval_samples_per_second": 4.557,
+      "eval_steps_per_second": 0.575,
+      "step": 2755
+    },
+    {
+      "epoch": 2.0036297640653356,
+      "grad_norm": 0.35921573638916016,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2760
+    },
+    {
+      "epoch": 2.010889292196007,
+      "grad_norm": 1.013861894607544,
+      "learning_rate": 0.0002,
+      "loss": 0.8968,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0181488203266786,
+      "grad_norm": 0.4425240159034729,
+      "learning_rate": 0.0002,
+      "loss": 0.7873,
+      "step": 2780
+    },
+    {
+      "epoch": 2.02540834845735,
+      "grad_norm": 0.60174161195755,
+      "learning_rate": 0.0002,
+      "loss": 0.8758,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0326678765880217,
+      "grad_norm": 0.47582098841667175,
+      "learning_rate": 0.0002,
+      "loss": 0.7989,
+      "step": 2800
+    },
+    {
+      "epoch": 2.0399274047186933,
+      "grad_norm": 0.6012811660766602,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 2810
+    },
+    {
+      "epoch": 2.047186932849365,
+      "grad_norm": 0.4444408118724823,
+      "learning_rate": 0.0002,
+      "loss": 0.8109,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0544464609800364,
+      "grad_norm": 0.4864003360271454,
+      "learning_rate": 0.0002,
+      "loss": 0.8621,
+      "step": 2830
+    },
+    {
+      "epoch": 2.061705989110708,
+      "grad_norm": 0.5104215741157532,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.6218489408493042,
+      "learning_rate": 0.0002,
+      "loss": 0.7789,
+      "step": 2850
+    },
+    {
+      "epoch": 2.076225045372051,
+      "grad_norm": 0.705784261226654,
+      "learning_rate": 0.0002,
+      "loss": 0.8778,
+      "step": 2860
+    },
+    {
+      "epoch": 2.0834845735027225,
+      "grad_norm": 0.48091503977775574,
+      "learning_rate": 0.0002,
+      "loss": 0.7851,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0907441016333936,
+      "grad_norm": 0.5062456727027893,
+      "learning_rate": 0.0002,
+      "loss": 0.8444,
+      "step": 2880
+    },
+    {
+      "epoch": 2.098003629764065,
+      "grad_norm": 0.4862022399902344,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 0.48264044523239136,
+      "learning_rate": 0.0002,
+      "loss": 0.8593,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1125226860254083,
+      "grad_norm": 0.43744346499443054,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2910
+    },
+    {
+      "epoch": 2.11978221415608,
+      "grad_norm": 0.5480492115020752,
+      "learning_rate": 0.0002,
+      "loss": 0.9099,
+      "step": 2920
+    },
+    {
+      "epoch": 2.1270417422867514,
+      "grad_norm": 0.5068560838699341,
+      "learning_rate": 0.0002,
+      "loss": 0.7727,
+      "step": 2930
+    },
+    {
+      "epoch": 2.134301270417423,
+      "grad_norm": 0.4650852680206299,
+      "learning_rate": 0.0002,
+      "loss": 0.8403,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1415607985480944,
+      "grad_norm": 0.4929981231689453,
+      "learning_rate": 0.0002,
+      "loss": 0.8467,
+      "step": 2950
+    },
+    {
+      "epoch": 2.148820326678766,
+      "grad_norm": 0.6537389755249023,
+      "learning_rate": 0.0002,
+      "loss": 0.8747,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1560798548094375,
+      "grad_norm": 0.8032940626144409,
+      "learning_rate": 0.0002,
+      "loss": 0.7663,
+      "step": 2970
+    },
+    {
+      "epoch": 2.163339382940109,
+      "grad_norm": 0.7131643891334534,
+      "learning_rate": 0.0002,
+      "loss": 0.7604,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1705989110707806,
+      "grad_norm": 0.6034275889396667,
+      "learning_rate": 0.0002,
+      "loss": 0.8424,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1778584392014517,
+      "grad_norm": 0.6081095933914185,
+      "learning_rate": 0.0002,
+      "loss": 0.8838,
+      "step": 3000
+    },
+    {
+      "epoch": 2.1851179673321233,
+      "grad_norm": 0.5706912875175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9013,
+      "step": 3010
+    },
+    {
+      "epoch": 2.192377495462795,
+      "grad_norm": 0.6742380261421204,
+      "learning_rate": 0.0002,
+      "loss": 0.8473,
+      "step": 3020
+    },
+    {
+      "epoch": 2.1996370235934664,
+      "grad_norm": 0.4847496449947357,
+      "learning_rate": 0.0002,
+      "loss": 0.8372,
+      "step": 3030
+    },
+    {
+      "epoch": 2.206896551724138,
+      "grad_norm": 0.5529342889785767,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2141560798548094,
+      "grad_norm": 0.6108783483505249,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3050
+    },
+    {
+      "epoch": 2.221415607985481,
+      "grad_norm": 0.8841571807861328,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2286751361161525,
+      "grad_norm": 0.4227530360221863,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 3070
+    },
+    {
+      "epoch": 2.235934664246824,
+      "grad_norm": 0.461935818195343,
+      "learning_rate": 0.0002,
+      "loss": 0.7925,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2431941923774956,
+      "grad_norm": 0.5407412648200989,
+      "learning_rate": 0.0002,
+      "loss": 0.8542,
+      "step": 3090
+    },
+    {
+      "epoch": 2.250453720508167,
+      "grad_norm": 0.5057998895645142,
+      "learning_rate": 0.0002,
+      "loss": 0.8392,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2577132486388383,
+      "grad_norm": 0.530057966709137,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 3110
+    },
+    {
+      "epoch": 2.26497277676951,
+      "grad_norm": 0.5066842436790466,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2722323049001814,
+      "grad_norm": 0.5069178342819214,
+      "learning_rate": 0.0002,
+      "loss": 0.7903,
+      "step": 3130
+    },
+    {
+      "epoch": 2.279491833030853,
+      "grad_norm": 0.6095499396324158,
+      "learning_rate": 0.0002,
+      "loss": 0.8458,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2867513611615244,
+      "grad_norm": 0.49524766206741333,
+      "learning_rate": 0.0002,
+      "loss": 0.8171,
+      "step": 3150
+    },
+    {
+      "epoch": 2.294010889292196,
+      "grad_norm": 0.5334409475326538,
+      "learning_rate": 0.0002,
+      "loss": 0.7568,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3012704174228675,
+      "grad_norm": 1.681748867034912,
+      "learning_rate": 0.0002,
+      "loss": 0.7533,
+      "step": 3170
+    },
+    {
+      "epoch": 2.308529945553539,
+      "grad_norm": 0.7225565314292908,
+      "learning_rate": 0.0002,
+      "loss": 0.8155,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3157894736842106,
+      "grad_norm": 0.5379388928413391,
+      "learning_rate": 0.0002,
+      "loss": 0.8034,
+      "step": 3190
+    },
+    {
+      "epoch": 2.323049001814882,
+      "grad_norm": 0.45770326256752014,
+      "learning_rate": 0.0002,
+      "loss": 0.8833,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3303085299455537,
+      "grad_norm": 0.6984533071517944,
+      "learning_rate": 0.0002,
+      "loss": 0.8419,
+      "step": 3210
+    },
+    {
+      "epoch": 2.337568058076225,
+      "grad_norm": 0.6725744605064392,
+      "learning_rate": 0.0002,
+      "loss": 0.7414,
+      "step": 3220
+    },
+    {
+      "epoch": 2.344827586206897,
+      "grad_norm": 1.1247471570968628,
+      "learning_rate": 0.0002,
+      "loss": 0.8657,
+      "step": 3230
+    },
+    {
+      "epoch": 2.352087114337568,
+      "grad_norm": 1.0240263938903809,
+      "learning_rate": 0.0002,
+      "loss": 0.7782,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3593466424682394,
+      "grad_norm": 0.5608096122741699,
+      "learning_rate": 0.0002,
+      "loss": 0.826,
+      "step": 3250
+    },
+    {
+      "epoch": 2.366606170598911,
+      "grad_norm": 0.8294990062713623,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3738656987295825,
+      "grad_norm": 0.6734224557876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9845,
+      "step": 3270
+    },
+    {
+      "epoch": 2.381125226860254,
+      "grad_norm": 0.6862800717353821,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3883847549909256,
+      "grad_norm": 0.5442930459976196,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 3290
+    },
+    {
+      "epoch": 2.395644283121597,
+      "grad_norm": 0.5745394229888916,
+      "learning_rate": 0.0002,
+      "loss": 0.8543,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4029038112522687,
+      "grad_norm": 0.6257799863815308,
+      "learning_rate": 0.0002,
+      "loss": 0.833,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4101633393829403,
+      "grad_norm": 0.5608420968055725,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3320
+    },
+    {
+      "epoch": 2.417422867513612,
+      "grad_norm": 0.5512017011642456,
+      "learning_rate": 0.0002,
+      "loss": 0.8745,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4246823956442833,
+      "grad_norm": 0.7452999353408813,
+      "learning_rate": 0.0002,
+      "loss": 0.8642,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4319419237749544,
+      "grad_norm": 0.4604301452636719,
+      "learning_rate": 0.0002,
+      "loss": 0.8417,
+      "step": 3350
+    },
+    {
+      "epoch": 2.439201451905626,
+      "grad_norm": 0.8225823640823364,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4464609800362975,
+      "grad_norm": 0.8604981303215027,
+      "learning_rate": 0.0002,
+      "loss": 0.8144,
+      "step": 3370
+    },
+    {
+      "epoch": 2.453720508166969,
+      "grad_norm": 0.6620925664901733,
+      "learning_rate": 0.0002,
+      "loss": 0.9276,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4609800362976406,
+      "grad_norm": 0.4750158488750458,
+      "learning_rate": 0.0002,
+      "loss": 0.8381,
+      "step": 3390
+    },
+    {
+      "epoch": 2.468239564428312,
+      "grad_norm": 0.6061418056488037,
+      "learning_rate": 0.0002,
+      "loss": 0.8249,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4754990925589837,
+      "grad_norm": 0.5977247357368469,
+      "learning_rate": 0.0002,
+      "loss": 0.8944,
+      "step": 3410
+    },
+    {
+      "epoch": 2.4827586206896552,
+      "grad_norm": 0.5004227757453918,
+      "learning_rate": 0.0002,
+      "loss": 0.8366,
+      "step": 3420
+    },
+    {
+      "epoch": 2.490018148820327,
+      "grad_norm": 0.46428972482681274,
+      "learning_rate": 0.0002,
+      "loss": 0.9133,
+      "step": 3430
+    },
+    {
+      "epoch": 2.4972776769509983,
+      "grad_norm": 0.7658395171165466,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 3440
+    },
+    {
+      "epoch": 2.50453720508167,
+      "grad_norm": 0.621819794178009,
+      "learning_rate": 0.0002,
+      "loss": 0.9066,
+      "step": 3450
+    },
+    {
+      "epoch": 2.511796733212341,
+      "grad_norm": 0.4556088149547577,
+      "learning_rate": 0.0002,
+      "loss": 0.8481,
+      "step": 3460
+    },
+    {
+      "epoch": 2.519056261343013,
+      "grad_norm": 0.6124140024185181,
+      "learning_rate": 0.0002,
+      "loss": 0.8522,
+      "step": 3470
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 0.6256231665611267,
+      "learning_rate": 0.0002,
+      "loss": 0.8158,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5335753176043556,
+      "grad_norm": 0.5464141964912415,
+      "learning_rate": 0.0002,
+      "loss": 0.9097,
+      "step": 3490
+    },
+    {
+      "epoch": 2.540834845735027,
+      "grad_norm": 0.51471346616745,
+      "learning_rate": 0.0002,
+      "loss": 0.8708,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5480943738656987,
+      "grad_norm": 0.5326165556907654,
+      "learning_rate": 0.0002,
+      "loss": 0.9042,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5553539019963702,
+      "grad_norm": 0.4750378429889679,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 3520
+    },
+    {
+      "epoch": 2.562613430127042,
+      "grad_norm": 0.5292279124259949,
+      "learning_rate": 0.0002,
+      "loss": 0.8248,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5698729582577133,
+      "grad_norm": 0.6145227551460266,
+      "learning_rate": 0.0002,
+      "loss": 0.8808,
+      "step": 3540
+    },
+    {
+      "epoch": 2.577132486388385,
+      "grad_norm": 0.8275189399719238,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5843920145190564,
+      "grad_norm": 0.5037438273429871,
+      "learning_rate": 0.0002,
+      "loss": 0.875,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5916515426497275,
+      "grad_norm": 0.5838707685470581,
+      "learning_rate": 0.0002,
+      "loss": 0.8324,
+      "step": 3570
+    },
+    {
+      "epoch": 2.5989110707803995,
+      "grad_norm": 0.5398710370063782,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6061705989110706,
+      "grad_norm": 0.6115376949310303,
+      "learning_rate": 0.0002,
+      "loss": 0.7843,
+      "step": 3590
+    },
+    {
+      "epoch": 2.613430127041742,
+      "grad_norm": 1.118809461593628,
+      "learning_rate": 0.0002,
+      "loss": 0.8142,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6206896551724137,
+      "grad_norm": 0.7811771631240845,
+      "learning_rate": 0.0002,
+      "loss": 0.8282,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6279491833030852,
+      "grad_norm": 0.5753175020217896,
+      "learning_rate": 0.0002,
+      "loss": 0.7997,
+      "step": 3620
+    },
+    {
+      "epoch": 2.635208711433757,
+      "grad_norm": 0.550829291343689,
+      "learning_rate": 0.0002,
+      "loss": 0.7705,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6424682395644283,
+      "grad_norm": 0.5360019207000732,
+      "learning_rate": 0.0002,
+      "loss": 0.8564,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6497277676951,
+      "grad_norm": 0.63050377368927,
+      "learning_rate": 0.0002,
+      "loss": 0.7648,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6569872958257714,
+      "grad_norm": 0.5833110213279724,
+      "learning_rate": 0.0002,
+      "loss": 0.9156,
+      "step": 3660
+    },
+    {
+      "epoch": 2.664246823956443,
+      "grad_norm": 0.5543047189712524,
+      "learning_rate": 0.0002,
+      "loss": 0.8033,
+      "step": 3670
+    },
+    {
+      "epoch": 2.671506352087114,
+      "grad_norm": 0.6842212080955505,
+      "learning_rate": 0.0002,
+      "loss": 0.8867,
+      "step": 3680
+    },
+    {
+      "epoch": 2.678765880217786,
+      "grad_norm": 0.6859333515167236,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 3690
+    },
+    {
+      "epoch": 2.686025408348457,
+      "grad_norm": 0.7038410902023315,
+      "learning_rate": 0.0002,
+      "loss": 0.8081,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6932849364791287,
+      "grad_norm": 0.447233647108078,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7005444646098002,
+      "grad_norm": 0.6453872919082642,
+      "learning_rate": 0.0002,
+      "loss": 0.8951,
+      "step": 3720
+    },
+    {
+      "epoch": 2.707803992740472,
+      "grad_norm": 0.8025672435760498,
+      "learning_rate": 0.0002,
+      "loss": 0.8292,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7150635208711433,
+      "grad_norm": 0.5997087955474854,
+      "learning_rate": 0.0002,
+      "loss": 0.8488,
+      "step": 3740
+    },
+    {
+      "epoch": 2.722323049001815,
+      "grad_norm": 0.6901142001152039,
+      "learning_rate": 0.0002,
+      "loss": 0.8308,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7295825771324864,
+      "grad_norm": 1.036145567893982,
+      "learning_rate": 0.0002,
+      "loss": 0.8517,
+      "step": 3760
+    },
+    {
+      "epoch": 2.736842105263158,
+      "grad_norm": 0.7207072377204895,
+      "learning_rate": 0.0002,
+      "loss": 0.8455,
+      "step": 3770
+    },
+    {
+      "epoch": 2.7441016333938295,
+      "grad_norm": 1.0452989339828491,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 3780
+    },
+    {
+      "epoch": 2.751361161524501,
+      "grad_norm": 0.5615278482437134,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.46439215540885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7658802177858437,
+      "grad_norm": 0.7134917974472046,
+      "learning_rate": 0.0002,
+      "loss": 0.7847,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7731397459165157,
+      "grad_norm": 0.5139115452766418,
+      "learning_rate": 0.0002,
+      "loss": 0.8315,
+      "step": 3820
+    },
+    {
+      "epoch": 2.780399274047187,
+      "grad_norm": 0.8595781326293945,
+      "learning_rate": 0.0002,
+      "loss": 0.8995,
+      "step": 3830
+    },
+    {
+      "epoch": 2.7876588021778583,
+      "grad_norm": 0.544614851474762,
+      "learning_rate": 0.0002,
+      "loss": 0.9079,
+      "step": 3840
+    },
+    {
+      "epoch": 2.79491833030853,
+      "grad_norm": 0.6073850393295288,
+      "learning_rate": 0.0002,
+      "loss": 0.8322,
+      "step": 3850
+    },
+    {
+      "epoch": 2.8021778584392014,
+      "grad_norm": 0.8224069476127625,
+      "learning_rate": 0.0002,
+      "loss": 0.8277,
+      "step": 3860
+    },
+    {
+      "epoch": 2.809437386569873,
+      "grad_norm": 0.5347970128059387,
+      "learning_rate": 0.0002,
+      "loss": 0.815,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8166969147005445,
+      "grad_norm": 0.6799601912498474,
+      "learning_rate": 0.0002,
+      "loss": 0.8219,
+      "step": 3880
+    },
+    {
+      "epoch": 2.823956442831216,
+      "grad_norm": 0.5219197869300842,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8312159709618876,
+      "grad_norm": 0.5710130333900452,
+      "learning_rate": 0.0002,
+      "loss": 0.8649,
+      "step": 3900
+    },
+    {
+      "epoch": 2.838475499092559,
+      "grad_norm": 0.5857213139533997,
+      "learning_rate": 0.0002,
+      "loss": 0.8741,
+      "step": 3910
+    },
+    {
+      "epoch": 2.8457350272232302,
+      "grad_norm": 0.5206644535064697,
+      "learning_rate": 0.0002,
+      "loss": 0.8154,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8529945553539022,
+      "grad_norm": 1.2902015447616577,
+      "learning_rate": 0.0002,
+      "loss": 0.9328,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8602540834845733,
+      "grad_norm": 0.5252797603607178,
+      "learning_rate": 0.0002,
+      "loss": 0.8192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.867513611615245,
+      "grad_norm": 0.5925108790397644,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8747731397459164,
+      "grad_norm": 0.9719610810279846,
+      "learning_rate": 0.0002,
+      "loss": 0.8611,
+      "step": 3960
+    },
+    {
+      "epoch": 2.882032667876588,
+      "grad_norm": 0.6834747195243835,
+      "learning_rate": 0.0002,
+      "loss": 0.8531,
+      "step": 3970
+    },
+    {
+      "epoch": 2.8892921960072595,
+      "grad_norm": 0.6952353119850159,
+      "learning_rate": 0.0002,
+      "loss": 0.8124,
+      "step": 3980
+    },
+    {
+      "epoch": 2.896551724137931,
+      "grad_norm": 0.49889910221099854,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9038112522686026,
+      "grad_norm": 0.5007026791572571,
+      "learning_rate": 0.0002,
+      "loss": 0.8197,
+      "step": 4000
+    },
+    {
+      "epoch": 2.911070780399274,
+      "grad_norm": 0.5474239587783813,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9183303085299457,
+      "grad_norm": 0.6617428660392761,
+      "learning_rate": 0.0002,
+      "loss": 0.8579,
+      "step": 4020
+    },
+    {
+      "epoch": 2.925589836660617,
+      "grad_norm": 0.6097776293754578,
+      "learning_rate": 0.0002,
+      "loss": 0.8583,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9328493647912888,
+      "grad_norm": 0.5985828638076782,
+      "learning_rate": 0.0002,
+      "loss": 0.8296,
+      "step": 4040
+    },
+    {
+      "epoch": 2.94010889292196,
+      "grad_norm": 0.769488513469696,
+      "learning_rate": 0.0002,
+      "loss": 0.9134,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 0.5167055130004883,
+      "learning_rate": 0.0002,
+      "loss": 0.8148,
+      "step": 4060
+    },
+    {
+      "epoch": 2.954627949183303,
+      "grad_norm": 0.6265496015548706,
+      "learning_rate": 0.0002,
+      "loss": 0.8665,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9618874773139745,
+      "grad_norm": 1.2644082307815552,
+      "learning_rate": 0.0002,
+      "loss": 0.9218,
+      "step": 4080
+    },
+    {
+      "epoch": 2.969147005444646,
+      "grad_norm": 0.6007561087608337,
+      "learning_rate": 0.0002,
+      "loss": 0.8026,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9764065335753176,
+      "grad_norm": 0.47984927892684937,
+      "learning_rate": 0.0002,
+      "loss": 0.8262,
+      "step": 4100
+    },
+    {
+      "epoch": 2.983666061705989,
+      "grad_norm": 1.128198504447937,
+      "learning_rate": 0.0002,
+      "loss": 0.7897,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9909255898366607,
+      "grad_norm": 0.526292085647583,
+      "learning_rate": 0.0002,
+      "loss": 0.7883,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9981851179673322,
+      "grad_norm": 0.5462674498558044,
+      "learning_rate": 0.0002,
+      "loss": 0.7801,
+      "step": 4130
+    },
+    {
+      "epoch": 2.9996370235934666,
+      "eval_loss": 1.1093357801437378,
+      "eval_runtime": 46.2498,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 4132
+    },
+    {
+      "epoch": 3.0054446460980038,
+      "grad_norm": 0.48288026452064514,
+      "learning_rate": 0.0002,
+      "loss": 0.721,
+      "step": 4140
+    },
+    {
+      "epoch": 3.0127041742286753,
+      "grad_norm": 1.0181782245635986,
+      "learning_rate": 0.0002,
+      "loss": 0.6769,
+      "step": 4150
+    },
+    {
+      "epoch": 3.019963702359347,
+      "grad_norm": 0.7718019485473633,
+      "learning_rate": 0.0002,
+      "loss": 0.7185,
+      "step": 4160
+    },
+    {
+      "epoch": 3.027223230490018,
+      "grad_norm": 0.7492219805717468,
+      "learning_rate": 0.0002,
+      "loss": 0.6552,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0344827586206895,
+      "grad_norm": 0.9363632798194885,
+      "learning_rate": 0.0002,
+      "loss": 0.6678,
+      "step": 4180
+    },
+    {
+      "epoch": 3.041742286751361,
+      "grad_norm": 0.6888533234596252,
+      "learning_rate": 0.0002,
+      "loss": 0.7187,
+      "step": 4190
+    },
+    {
+      "epoch": 3.0490018148820326,
+      "grad_norm": 0.7072834968566895,
+      "learning_rate": 0.0002,
+      "loss": 0.6469,
+      "step": 4200
+    },
+    {
+      "epoch": 3.056261343012704,
+      "grad_norm": 0.7182047963142395,
+      "learning_rate": 0.0002,
+      "loss": 0.6387,
+      "step": 4210
+    },
+    {
+      "epoch": 3.0635208711433757,
+      "grad_norm": 0.7194355130195618,
+      "learning_rate": 0.0002,
+      "loss": 0.6385,
+      "step": 4220
+    },
+    {
+      "epoch": 3.0707803992740472,
+      "grad_norm": 0.9454023838043213,
+      "learning_rate": 0.0002,
+      "loss": 0.5812,
+      "step": 4230
+    },
+    {
+      "epoch": 3.0780399274047188,
+      "grad_norm": 0.838657557964325,
+      "learning_rate": 0.0002,
+      "loss": 0.6036,
+      "step": 4240
+    },
+    {
+      "epoch": 3.0852994555353903,
+      "grad_norm": 0.740113377571106,
+      "learning_rate": 0.0002,
+      "loss": 0.646,
+      "step": 4250
+    },
+    {
+      "epoch": 3.092558983666062,
+      "grad_norm": 0.6616561412811279,
+      "learning_rate": 0.0002,
+      "loss": 0.604,
+      "step": 4260
+    },
+    {
+      "epoch": 3.0998185117967334,
+      "grad_norm": 0.8846506476402283,
+      "learning_rate": 0.0002,
+      "loss": 0.6462,
+      "step": 4270
+    },
+    {
+      "epoch": 3.107078039927405,
+      "grad_norm": 0.6322125792503357,
+      "learning_rate": 0.0002,
+      "loss": 0.6037,
+      "step": 4280
+    },
+    {
+      "epoch": 3.114337568058076,
+      "grad_norm": 0.7461467385292053,
+      "learning_rate": 0.0002,
+      "loss": 0.5953,
+      "step": 4290
+    },
+    {
+      "epoch": 3.1215970961887476,
+      "grad_norm": 0.8251287341117859,
+      "learning_rate": 0.0002,
+      "loss": 0.6099,
+      "step": 4300
+    },
+    {
+      "epoch": 3.128856624319419,
+      "grad_norm": 0.8767673373222351,
+      "learning_rate": 0.0002,
+      "loss": 0.6284,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1361161524500907,
+      "grad_norm": 0.7758759260177612,
+      "learning_rate": 0.0002,
+      "loss": 0.7535,
+      "step": 4320
+    },
+    {
+      "epoch": 3.143375680580762,
+      "grad_norm": 1.1056879758834839,
+      "learning_rate": 0.0002,
+      "loss": 0.6624,
+      "step": 4330
+    },
+    {
+      "epoch": 3.1506352087114338,
+      "grad_norm": 0.8259835243225098,
+      "learning_rate": 0.0002,
+      "loss": 0.691,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 0.6607027053833008,
+      "learning_rate": 0.0002,
+      "loss": 0.6635,
+      "step": 4350
+    },
+    {
+      "epoch": 3.165154264972777,
+      "grad_norm": 0.7983301281929016,
+      "learning_rate": 0.0002,
+      "loss": 0.5911,
+      "step": 4360
+    },
+    {
+      "epoch": 3.1724137931034484,
+      "grad_norm": 0.6725239157676697,
+      "learning_rate": 0.0002,
+      "loss": 0.6496,
+      "step": 4370
+    },
+    {
+      "epoch": 3.17967332123412,
+      "grad_norm": 0.9052095413208008,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 4380
+    },
+    {
+      "epoch": 3.1869328493647915,
+      "grad_norm": 0.8131307363510132,
+      "learning_rate": 0.0002,
+      "loss": 0.6877,
+      "step": 4390
+    },
+    {
+      "epoch": 3.1941923774954626,
+      "grad_norm": 0.6435626149177551,
+      "learning_rate": 0.0002,
+      "loss": 0.6384,
+      "step": 4400
+    },
+    {
+      "epoch": 3.201451905626134,
+      "grad_norm": 0.84367436170578,
+      "learning_rate": 0.0002,
+      "loss": 0.5819,
+      "step": 4410
+    },
+    {
+      "epoch": 3.2087114337568057,
+      "grad_norm": 1.5018867254257202,
+      "learning_rate": 0.0002,
+      "loss": 0.6104,
+      "step": 4420
+    },
+    {
+      "epoch": 3.215970961887477,
+      "grad_norm": 0.7019091844558716,
+      "learning_rate": 0.0002,
+      "loss": 0.6838,
+      "step": 4430
+    },
+    {
+      "epoch": 3.2232304900181488,
+      "grad_norm": 0.9164197444915771,
+      "learning_rate": 0.0002,
+      "loss": 0.6153,
+      "step": 4440
+    },
+    {
+      "epoch": 3.2304900181488203,
+      "grad_norm": 0.7890861630439758,
+      "learning_rate": 0.0002,
+      "loss": 0.6618,
+      "step": 4450
+    },
+    {
+      "epoch": 3.237749546279492,
+      "grad_norm": 0.6517660617828369,
+      "learning_rate": 0.0002,
+      "loss": 0.6401,
+      "step": 4460
+    },
+    {
+      "epoch": 3.2450090744101634,
+      "grad_norm": 1.10188889503479,
+      "learning_rate": 0.0002,
+      "loss": 0.6699,
+      "step": 4470
+    },
+    {
+      "epoch": 3.252268602540835,
+      "grad_norm": 0.8158330917358398,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 4480
+    },
+    {
+      "epoch": 3.2595281306715065,
+      "grad_norm": 0.7663109302520752,
+      "learning_rate": 0.0002,
+      "loss": 0.7757,
+      "step": 4490
+    },
+    {
+      "epoch": 3.266787658802178,
+      "grad_norm": 0.8473444581031799,
+      "learning_rate": 0.0002,
+      "loss": 0.6539,
+      "step": 4500
+    },
+    {
+      "epoch": 3.274047186932849,
+      "grad_norm": 0.9724768996238708,
+      "learning_rate": 0.0002,
+      "loss": 0.6511,
+      "step": 4510
+    },
+    {
+      "epoch": 3.281306715063521,
+      "grad_norm": 0.8516759276390076,
+      "learning_rate": 0.0002,
+      "loss": 0.5464,
+      "step": 4520
+    },
+    {
+      "epoch": 3.288566243194192,
+      "grad_norm": 0.7543437480926514,
+      "learning_rate": 0.0002,
+      "loss": 0.6534,
+      "step": 4530
+    },
+    {
+      "epoch": 3.2958257713248638,
+      "grad_norm": 1.0472029447555542,
+      "learning_rate": 0.0002,
+      "loss": 0.6095,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3030852994555353,
+      "grad_norm": 0.6240826845169067,
+      "learning_rate": 0.0002,
+      "loss": 0.6216,
+      "step": 4550
+    },
+    {
+      "epoch": 3.310344827586207,
+      "grad_norm": 0.9957774877548218,
+      "learning_rate": 0.0002,
+      "loss": 0.6223,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3176043557168784,
+      "grad_norm": 0.6448912620544434,
+      "learning_rate": 0.0002,
+      "loss": 0.618,
+      "step": 4570
+    },
+    {
+      "epoch": 3.32486388384755,
+      "grad_norm": 0.7519692778587341,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 4580
+    },
+    {
+      "epoch": 3.3321234119782215,
+      "grad_norm": 0.7367453575134277,
+      "learning_rate": 0.0002,
+      "loss": 0.6672,
+      "step": 4590
+    },
+    {
+      "epoch": 3.339382940108893,
+      "grad_norm": 0.8064960837364197,
+      "learning_rate": 0.0002,
+      "loss": 0.6517,
+      "step": 4600
+    },
+    {
+      "epoch": 3.3466424682395646,
+      "grad_norm": 0.7664631009101868,
+      "learning_rate": 0.0002,
+      "loss": 0.6062,
+      "step": 4610
+    },
+    {
+      "epoch": 3.353901996370236,
+      "grad_norm": 0.7803396582603455,
+      "learning_rate": 0.0002,
+      "loss": 0.6834,
+      "step": 4620
+    },
+    {
+      "epoch": 3.3611615245009077,
+      "grad_norm": 0.9141599535942078,
+      "learning_rate": 0.0002,
+      "loss": 0.6961,
+      "step": 4630
+    },
+    {
+      "epoch": 3.3684210526315788,
+      "grad_norm": 0.9719856381416321,
+      "learning_rate": 0.0002,
+      "loss": 0.6889,
+      "step": 4640
+    },
+    {
+      "epoch": 3.3756805807622503,
+      "grad_norm": 0.9223218560218811,
+      "learning_rate": 0.0002,
+      "loss": 0.6914,
+      "step": 4650
+    },
+    {
+      "epoch": 3.382940108892922,
+      "grad_norm": 0.7289277911186218,
+      "learning_rate": 0.0002,
+      "loss": 0.5981,
+      "step": 4660
+    },
+    {
+      "epoch": 3.3901996370235934,
+      "grad_norm": 1.039724349975586,
+      "learning_rate": 0.0002,
+      "loss": 0.595,
+      "step": 4670
+    },
+    {
+      "epoch": 3.397459165154265,
+      "grad_norm": 1.397438883781433,
+      "learning_rate": 0.0002,
+      "loss": 0.8121,
+      "step": 4680
+    },
+    {
+      "epoch": 3.4047186932849365,
+      "grad_norm": 1.0069999694824219,
+      "learning_rate": 0.0002,
+      "loss": 0.6334,
+      "step": 4690
+    },
+    {
+      "epoch": 3.411978221415608,
+      "grad_norm": 0.816291332244873,
+      "learning_rate": 0.0002,
+      "loss": 0.6598,
+      "step": 4700
+    },
+    {
+      "epoch": 3.4192377495462796,
+      "grad_norm": 1.2831530570983887,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 4710
+    },
+    {
+      "epoch": 3.426497277676951,
+      "grad_norm": 0.9573889970779419,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 4720
+    },
+    {
+      "epoch": 3.4337568058076227,
+      "grad_norm": 0.7685632705688477,
+      "learning_rate": 0.0002,
+      "loss": 0.7279,
+      "step": 4730
+    },
+    {
+      "epoch": 3.441016333938294,
+      "grad_norm": 0.7019195556640625,
+      "learning_rate": 0.0002,
+      "loss": 0.6104,
+      "step": 4740
+    },
+    {
+      "epoch": 3.4482758620689653,
+      "grad_norm": 0.7244833707809448,
+      "learning_rate": 0.0002,
+      "loss": 0.7606,
+      "step": 4750
+    },
+    {
+      "epoch": 3.455535390199637,
+      "grad_norm": 1.3468551635742188,
+      "learning_rate": 0.0002,
+      "loss": 0.6951,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4627949183303084,
+      "grad_norm": 0.822846531867981,
+      "learning_rate": 0.0002,
+      "loss": 0.6945,
+      "step": 4770
+    },
+    {
+      "epoch": 3.47005444646098,
+      "grad_norm": 0.7311608195304871,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 4780
+    },
+    {
+      "epoch": 3.4773139745916515,
+      "grad_norm": 0.9466770887374878,
+      "learning_rate": 0.0002,
+      "loss": 0.7019,
+      "step": 4790
+    },
+    {
+      "epoch": 3.484573502722323,
+      "grad_norm": 1.1527155637741089,
+      "learning_rate": 0.0002,
+      "loss": 0.7767,
+      "step": 4800
+    },
+    {
+      "epoch": 3.4918330308529946,
+      "grad_norm": 1.1288906335830688,
+      "learning_rate": 0.0002,
+      "loss": 0.6882,
+      "step": 4810
+    },
+    {
+      "epoch": 3.499092558983666,
+      "grad_norm": 0.9096164107322693,
+      "learning_rate": 0.0002,
+      "loss": 0.6564,
+      "step": 4820
+    },
+    {
+      "epoch": 3.5063520871143377,
+      "grad_norm": 0.7988565564155579,
+      "learning_rate": 0.0002,
+      "loss": 0.6127,
+      "step": 4830
+    },
+    {
+      "epoch": 3.513611615245009,
+      "grad_norm": 0.7183415293693542,
+      "learning_rate": 0.0002,
+      "loss": 0.7004,
+      "step": 4840
+    },
+    {
+      "epoch": 3.5208711433756807,
+      "grad_norm": 0.6614915132522583,
+      "learning_rate": 0.0002,
+      "loss": 0.74,
+      "step": 4850
+    },
+    {
+      "epoch": 3.528130671506352,
+      "grad_norm": 0.8609521985054016,
+      "learning_rate": 0.0002,
+      "loss": 0.7271,
+      "step": 4860
+    },
+    {
+      "epoch": 3.535390199637024,
+      "grad_norm": 0.86552894115448,
+      "learning_rate": 0.0002,
+      "loss": 0.6664,
+      "step": 4870
+    },
+    {
+      "epoch": 3.542649727767695,
+      "grad_norm": 0.6926496028900146,
+      "learning_rate": 0.0002,
+      "loss": 0.6432,
+      "step": 4880
+    },
+    {
+      "epoch": 3.5499092558983665,
+      "grad_norm": 0.8157467246055603,
+      "learning_rate": 0.0002,
+      "loss": 0.7117,
+      "step": 4890
+    },
+    {
+      "epoch": 3.557168784029038,
+      "grad_norm": 0.9085357189178467,
+      "learning_rate": 0.0002,
+      "loss": 0.6201,
+      "step": 4900
+    },
+    {
+      "epoch": 3.5644283121597096,
+      "grad_norm": 0.6322644948959351,
+      "learning_rate": 0.0002,
+      "loss": 0.6521,
+      "step": 4910
+    },
+    {
+      "epoch": 3.571687840290381,
+      "grad_norm": 1.263205885887146,
+      "learning_rate": 0.0002,
+      "loss": 0.6607,
+      "step": 4920
+    },
+    {
+      "epoch": 3.5789473684210527,
+      "grad_norm": 0.8901070356369019,
+      "learning_rate": 0.0002,
+      "loss": 0.6657,
+      "step": 4930
+    },
+    {
+      "epoch": 3.586206896551724,
+      "grad_norm": 0.7983952164649963,
+      "learning_rate": 0.0002,
+      "loss": 0.6434,
+      "step": 4940
+    },
+    {
+      "epoch": 3.5934664246823957,
+      "grad_norm": 0.9887813925743103,
+      "learning_rate": 0.0002,
+      "loss": 0.6861,
+      "step": 4950
+    },
+    {
+      "epoch": 3.6007259528130673,
+      "grad_norm": 0.7895187735557556,
+      "learning_rate": 0.0002,
+      "loss": 0.6502,
+      "step": 4960
+    },
+    {
+      "epoch": 3.6079854809437384,
+      "grad_norm": 0.9685819745063782,
+      "learning_rate": 0.0002,
+      "loss": 0.7111,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6152450090744104,
+      "grad_norm": 0.6576591730117798,
+      "learning_rate": 0.0002,
+      "loss": 0.6915,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6225045372050815,
+      "grad_norm": 0.856985330581665,
+      "learning_rate": 0.0002,
+      "loss": 0.6195,
+      "step": 4990
+    },
+    {
+      "epoch": 3.629764065335753,
+      "grad_norm": 0.7230252623558044,
+      "learning_rate": 0.0002,
+      "loss": 0.6318,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6370235934664246,
+      "grad_norm": 0.8260893821716309,
+      "learning_rate": 0.0002,
+      "loss": 0.742,
+      "step": 5010
+    },
+    {
+      "epoch": 3.644283121597096,
+      "grad_norm": 0.7635950446128845,
+      "learning_rate": 0.0002,
+      "loss": 0.7223,
+      "step": 5020
+    },
+    {
+      "epoch": 3.6515426497277677,
+      "grad_norm": 0.7060768604278564,
+      "learning_rate": 0.0002,
+      "loss": 0.6837,
+      "step": 5030
+    },
+    {
+      "epoch": 3.658802177858439,
+      "grad_norm": 0.8020303249359131,
+      "learning_rate": 0.0002,
+      "loss": 0.6921,
+      "step": 5040
+    },
+    {
+      "epoch": 3.6660617059891107,
+      "grad_norm": 0.8530341386795044,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 5050
+    },
+    {
+      "epoch": 3.6733212341197823,
+      "grad_norm": 0.6667101979255676,
+      "learning_rate": 0.0002,
+      "loss": 0.7222,
+      "step": 5060
+    },
+    {
+      "epoch": 3.680580762250454,
+      "grad_norm": 0.7385406494140625,
+      "learning_rate": 0.0002,
+      "loss": 0.7081,
+      "step": 5070
+    },
+    {
+      "epoch": 3.6878402903811254,
+      "grad_norm": 0.7753380537033081,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 5080
+    },
+    {
+      "epoch": 3.695099818511797,
+      "grad_norm": 0.7516207098960876,
+      "learning_rate": 0.0002,
+      "loss": 0.6491,
+      "step": 5090
+    },
+    {
+      "epoch": 3.702359346642468,
+      "grad_norm": 0.8171586394309998,
+      "learning_rate": 0.0002,
+      "loss": 0.672,
+      "step": 5100
+    },
+    {
+      "epoch": 3.70961887477314,
+      "grad_norm": 1.0796279907226562,
+      "learning_rate": 0.0002,
+      "loss": 0.7459,
+      "step": 5110
+    },
+    {
+      "epoch": 3.716878402903811,
+      "grad_norm": 0.6957688927650452,
+      "learning_rate": 0.0002,
+      "loss": 0.5948,
+      "step": 5120
+    },
+    {
+      "epoch": 3.7241379310344827,
+      "grad_norm": 0.8550161719322205,
+      "learning_rate": 0.0002,
+      "loss": 0.7515,
+      "step": 5130
+    },
+    {
+      "epoch": 3.731397459165154,
+      "grad_norm": 0.9396728277206421,
+      "learning_rate": 0.0002,
+      "loss": 0.7286,
+      "step": 5140
+    },
+    {
+      "epoch": 3.7386569872958257,
+      "grad_norm": 1.4264805316925049,
+      "learning_rate": 0.0002,
+      "loss": 0.7594,
+      "step": 5150
+    },
+    {
+      "epoch": 3.7459165154264973,
+      "grad_norm": 0.8725108504295349,
+      "learning_rate": 0.0002,
+      "loss": 0.6575,
+      "step": 5160
+    },
+    {
+      "epoch": 3.753176043557169,
+      "grad_norm": 1.0346195697784424,
+      "learning_rate": 0.0002,
+      "loss": 0.6778,
+      "step": 5170
+    },
+    {
+      "epoch": 3.7604355716878404,
+      "grad_norm": 0.5395554304122925,
+      "learning_rate": 0.0002,
+      "loss": 0.6371,
+      "step": 5180
+    },
+    {
+      "epoch": 3.767695099818512,
+      "grad_norm": 1.3153616189956665,
+      "learning_rate": 0.0002,
+      "loss": 0.7308,
+      "step": 5190
+    },
+    {
+      "epoch": 3.7749546279491835,
+      "grad_norm": 0.9879828691482544,
+      "learning_rate": 0.0002,
+      "loss": 0.78,
+      "step": 5200
+    },
+    {
+      "epoch": 3.7822141560798546,
+      "grad_norm": 0.8876672983169556,
+      "learning_rate": 0.0002,
+      "loss": 0.7068,
+      "step": 5210
+    },
+    {
+      "epoch": 3.7894736842105265,
+      "grad_norm": 0.8363267779350281,
+      "learning_rate": 0.0002,
+      "loss": 0.6283,
+      "step": 5220
+    },
+    {
+      "epoch": 3.7967332123411976,
+      "grad_norm": 0.637294590473175,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 5230
+    },
+    {
+      "epoch": 3.803992740471869,
+      "grad_norm": 1.1408970355987549,
+      "learning_rate": 0.0002,
+      "loss": 0.6685,
+      "step": 5240
+    },
+    {
+      "epoch": 3.8112522686025407,
+      "grad_norm": 1.0128360986709595,
+      "learning_rate": 0.0002,
+      "loss": 0.6761,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8185117967332123,
+      "grad_norm": 0.8061144351959229,
+      "learning_rate": 0.0002,
+      "loss": 0.6764,
+      "step": 5260
+    },
+    {
+      "epoch": 3.825771324863884,
+      "grad_norm": 0.9626626968383789,
+      "learning_rate": 0.0002,
+      "loss": 0.7254,
+      "step": 5270
+    },
+    {
+      "epoch": 3.8330308529945554,
+      "grad_norm": 0.9013627171516418,
+      "learning_rate": 0.0002,
+      "loss": 0.7367,
+      "step": 5280
+    },
+    {
+      "epoch": 3.840290381125227,
+      "grad_norm": 0.8411344289779663,
+      "learning_rate": 0.0002,
+      "loss": 0.6806,
+      "step": 5290
+    },
+    {
+      "epoch": 3.8475499092558985,
+      "grad_norm": 0.7426059246063232,
+      "learning_rate": 0.0002,
+      "loss": 0.6818,
+      "step": 5300
+    },
+    {
+      "epoch": 3.85480943738657,
+      "grad_norm": 1.003413438796997,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 5310
+    },
+    {
+      "epoch": 3.862068965517241,
+      "grad_norm": 0.7527840733528137,
+      "learning_rate": 0.0002,
+      "loss": 0.8554,
+      "step": 5320
+    },
+    {
+      "epoch": 3.869328493647913,
+      "grad_norm": 0.738610565662384,
+      "learning_rate": 0.0002,
+      "loss": 0.7521,
+      "step": 5330
+    },
+    {
+      "epoch": 3.876588021778584,
+      "grad_norm": 0.7277999520301819,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 5340
+    },
+    {
+      "epoch": 3.8838475499092557,
+      "grad_norm": 0.5951359272003174,
+      "learning_rate": 0.0002,
+      "loss": 0.7503,
+      "step": 5350
+    },
+    {
+      "epoch": 3.8911070780399273,
+      "grad_norm": 1.043884038925171,
+      "learning_rate": 0.0002,
+      "loss": 0.7447,
+      "step": 5360
+    },
+    {
+      "epoch": 3.898366606170599,
+      "grad_norm": 0.8436498045921326,
+      "learning_rate": 0.0002,
+      "loss": 0.6862,
+      "step": 5370
+    },
+    {
+      "epoch": 3.9056261343012704,
+      "grad_norm": 0.5603365302085876,
+      "learning_rate": 0.0002,
+      "loss": 0.665,
+      "step": 5380
+    },
+    {
+      "epoch": 3.912885662431942,
+      "grad_norm": 1.0128886699676514,
+      "learning_rate": 0.0002,
+      "loss": 0.7098,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9201451905626135,
+      "grad_norm": 0.7970930337905884,
+      "learning_rate": 0.0002,
+      "loss": 0.6707,
+      "step": 5400
+    },
+    {
+      "epoch": 3.927404718693285,
+      "grad_norm": 0.7699369192123413,
+      "learning_rate": 0.0002,
+      "loss": 0.637,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9346642468239565,
+      "grad_norm": 0.800561249256134,
+      "learning_rate": 0.0002,
+      "loss": 0.6742,
+      "step": 5420
+    },
+    {
+      "epoch": 3.941923774954628,
+      "grad_norm": 0.8020331859588623,
+      "learning_rate": 0.0002,
+      "loss": 0.7208,
+      "step": 5430
+    },
+    {
+      "epoch": 3.9491833030852996,
+      "grad_norm": 0.7461140155792236,
+      "learning_rate": 0.0002,
+      "loss": 0.7294,
+      "step": 5440
+    },
+    {
+      "epoch": 3.9564428312159707,
+      "grad_norm": 0.8346918821334839,
+      "learning_rate": 0.0002,
+      "loss": 0.7013,
+      "step": 5450
+    },
+    {
+      "epoch": 3.9637023593466427,
+      "grad_norm": 0.9723302125930786,
+      "learning_rate": 0.0002,
+      "loss": 0.6289,
+      "step": 5460
+    },
+    {
+      "epoch": 3.970961887477314,
+      "grad_norm": 0.6809740662574768,
+      "learning_rate": 0.0002,
+      "loss": 0.8029,
+      "step": 5470
+    },
+    {
+      "epoch": 3.9782214156079854,
+      "grad_norm": 0.7353498339653015,
+      "learning_rate": 0.0002,
+      "loss": 0.6896,
+      "step": 5480
+    },
+    {
+      "epoch": 3.985480943738657,
+      "grad_norm": 0.748009443283081,
+      "learning_rate": 0.0002,
+      "loss": 0.6722,
+      "step": 5490
+    },
+    {
+      "epoch": 3.9927404718693285,
+      "grad_norm": 1.3656195402145386,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 5500
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.8402108550071716,
+      "learning_rate": 0.0002,
+      "loss": 0.7368,
+      "step": 5510
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.17229425907135,
+      "eval_runtime": 46.2554,
+      "eval_samples_per_second": 9.426,
+      "eval_steps_per_second": 1.189,
+      "step": 5510
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11016,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.830142384111616e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..140bceabdec44fc473842cb9ace4a62a4dc9fe89
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-5510/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4abec5480aff960b5a7b03a48be1cdb1ae29ba8792469c451f5c2f269e4be93b
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..7da450895fa1608a2cc58faccd6e26831db22107
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68eaeba0cc63008c5bd7c3bc5aa7bdd8e28a511348c857e109ff24809f57d6ee
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6ec6fee050830f7c9ff4a382547b40ff418b5b0e
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0a833460c3dd7d23d953c8af07fb3bf2b05669c63cabaf6fa51acb9f4050be7
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..ef2f7c31a7db238c7327dc554e587b7ae913704d
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84484997bb1fd50f9a979febb3f3a5e33d8908d9d7d1ad2009936d364854484c
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..966c1a8c7cc28c297f81d9de5accd07e8d81a61d
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c998afeb2f35200deae216475374398ad29855a63f049e215b018de796764f10
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..1596f14ed1b03d8480a2e4640c4f7cd958261a5b
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/trainer_state.json
@@ -0,0 +1,4889 @@
+{
+  "best_metric": 1.1086540222167969,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755",
+  "epoch": 4.999637023593467,
+  "eval_steps": 10,
+  "global_step": 6887,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007259528130671506,
+      "grad_norm": 0.46010470390319824,
+      "learning_rate": 0.0002,
+      "loss": 2.4936,
+      "step": 10
+    },
+    {
+      "epoch": 0.014519056261343012,
+      "grad_norm": 0.5103179216384888,
+      "learning_rate": 0.0002,
+      "loss": 1.8442,
+      "step": 20
+    },
+    {
+      "epoch": 0.021778584392014518,
+      "grad_norm": 0.6282716989517212,
+      "learning_rate": 0.0002,
+      "loss": 1.559,
+      "step": 30
+    },
+    {
+      "epoch": 0.029038112522686024,
+      "grad_norm": 1.2480497360229492,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 40
+    },
+    {
+      "epoch": 0.036297640653357534,
+      "grad_norm": 0.4114855229854584,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 50
+    },
+    {
+      "epoch": 0.043557168784029036,
+      "grad_norm": 0.49482840299606323,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 60
+    },
+    {
+      "epoch": 0.050816696914700546,
+      "grad_norm": 0.4536272883415222,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 70
+    },
+    {
+      "epoch": 0.05807622504537205,
+      "grad_norm": 0.32328274846076965,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 80
+    },
+    {
+      "epoch": 0.06533575317604355,
+      "grad_norm": 0.40990468859672546,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 90
+    },
+    {
+      "epoch": 0.07259528130671507,
+      "grad_norm": 0.37273502349853516,
+      "learning_rate": 0.0002,
+      "loss": 1.2216,
+      "step": 100
+    },
+    {
+      "epoch": 0.07985480943738657,
+      "grad_norm": 0.3903871476650238,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 110
+    },
+    {
+      "epoch": 0.08711433756805807,
+      "grad_norm": 0.3272787928581238,
+      "learning_rate": 0.0002,
+      "loss": 1.3718,
+      "step": 120
+    },
+    {
+      "epoch": 0.09437386569872959,
+      "grad_norm": 0.3622824251651764,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 130
+    },
+    {
+      "epoch": 0.10163339382940109,
+      "grad_norm": 0.3503916561603546,
+      "learning_rate": 0.0002,
+      "loss": 1.2966,
+      "step": 140
+    },
+    {
+      "epoch": 0.1088929219600726,
+      "grad_norm": 0.32787832617759705,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 150
+    },
+    {
+      "epoch": 0.1161524500907441,
+      "grad_norm": 0.5822657942771912,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 160
+    },
+    {
+      "epoch": 0.12341197822141561,
+      "grad_norm": 0.28028249740600586,
+      "learning_rate": 0.0002,
+      "loss": 1.2091,
+      "step": 170
+    },
+    {
+      "epoch": 0.1306715063520871,
+      "grad_norm": 0.33602750301361084,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 180
+    },
+    {
+      "epoch": 0.13793103448275862,
+      "grad_norm": 0.29106274247169495,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 190
+    },
+    {
+      "epoch": 0.14519056261343014,
+      "grad_norm": 0.38753363490104675,
+      "learning_rate": 0.0002,
+      "loss": 1.3261,
+      "step": 200
+    },
+    {
+      "epoch": 0.15245009074410162,
+      "grad_norm": 0.361009418964386,
+      "learning_rate": 0.0002,
+      "loss": 1.1536,
+      "step": 210
+    },
+    {
+      "epoch": 0.15970961887477314,
+      "grad_norm": 0.6743836402893066,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 220
+    },
+    {
+      "epoch": 0.16696914700544466,
+      "grad_norm": 0.3923613727092743,
+      "learning_rate": 0.0002,
+      "loss": 1.264,
+      "step": 230
+    },
+    {
+      "epoch": 0.17422867513611615,
+      "grad_norm": 0.2809699773788452,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 240
+    },
+    {
+      "epoch": 0.18148820326678766,
+      "grad_norm": 0.3631494641304016,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 250
+    },
+    {
+      "epoch": 0.18874773139745918,
+      "grad_norm": 0.24658171832561493,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 260
+    },
+    {
+      "epoch": 0.19600725952813067,
+      "grad_norm": 0.5780664682388306,
+      "learning_rate": 0.0002,
+      "loss": 1.0065,
+      "step": 270
+    },
+    {
+      "epoch": 0.20326678765880218,
+      "grad_norm": 0.3056720495223999,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 280
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 0.2857084572315216,
+      "learning_rate": 0.0002,
+      "loss": 1.1503,
+      "step": 290
+    },
+    {
+      "epoch": 0.2177858439201452,
+      "grad_norm": 0.6645345687866211,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 300
+    },
+    {
+      "epoch": 0.2250453720508167,
+      "grad_norm": 0.5966078639030457,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 310
+    },
+    {
+      "epoch": 0.2323049001814882,
+      "grad_norm": 0.40937140583992004,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 320
+    },
+    {
+      "epoch": 0.2395644283121597,
+      "grad_norm": 0.5642806887626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 330
+    },
+    {
+      "epoch": 0.24682395644283123,
+      "grad_norm": 0.2750748097896576,
+      "learning_rate": 0.0002,
+      "loss": 1.1703,
+      "step": 340
+    },
+    {
+      "epoch": 0.2540834845735027,
+      "grad_norm": 0.34350234270095825,
+      "learning_rate": 0.0002,
+      "loss": 1.2598,
+      "step": 350
+    },
+    {
+      "epoch": 0.2613430127041742,
+      "grad_norm": 0.6767239570617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1942,
+      "step": 360
+    },
+    {
+      "epoch": 0.26860254083484575,
+      "grad_norm": 0.31006959080696106,
+      "learning_rate": 0.0002,
+      "loss": 1.1436,
+      "step": 370
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "grad_norm": 0.3825474679470062,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 380
+    },
+    {
+      "epoch": 0.2831215970961887,
+      "grad_norm": 0.30241551995277405,
+      "learning_rate": 0.0002,
+      "loss": 1.1523,
+      "step": 390
+    },
+    {
+      "epoch": 0.29038112522686027,
+      "grad_norm": 0.2962397336959839,
+      "learning_rate": 0.0002,
+      "loss": 1.1298,
+      "step": 400
+    },
+    {
+      "epoch": 0.29764065335753176,
+      "grad_norm": 0.2600369155406952,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 410
+    },
+    {
+      "epoch": 0.30490018148820325,
+      "grad_norm": 0.3675060272216797,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 420
+    },
+    {
+      "epoch": 0.3121597096188748,
+      "grad_norm": 0.3429498076438904,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 430
+    },
+    {
+      "epoch": 0.3194192377495463,
+      "grad_norm": 0.34311825037002563,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 440
+    },
+    {
+      "epoch": 0.32667876588021777,
+      "grad_norm": 0.37872210144996643,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 450
+    },
+    {
+      "epoch": 0.3339382940108893,
+      "grad_norm": 0.33271121978759766,
+      "learning_rate": 0.0002,
+      "loss": 1.1365,
+      "step": 460
+    },
+    {
+      "epoch": 0.3411978221415608,
+      "grad_norm": 0.34605276584625244,
+      "learning_rate": 0.0002,
+      "loss": 1.1572,
+      "step": 470
+    },
+    {
+      "epoch": 0.3484573502722323,
+      "grad_norm": 0.41050562262535095,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 480
+    },
+    {
+      "epoch": 0.35571687840290384,
+      "grad_norm": 0.2066836953163147,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 490
+    },
+    {
+      "epoch": 0.3629764065335753,
+      "grad_norm": 0.2859014868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.2072,
+      "step": 500
+    },
+    {
+      "epoch": 0.3702359346642468,
+      "grad_norm": 0.28763777017593384,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 510
+    },
+    {
+      "epoch": 0.37749546279491836,
+      "grad_norm": 0.2730471193790436,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 520
+    },
+    {
+      "epoch": 0.38475499092558985,
+      "grad_norm": 0.3968936800956726,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 530
+    },
+    {
+      "epoch": 0.39201451905626133,
+      "grad_norm": 0.3624701201915741,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 540
+    },
+    {
+      "epoch": 0.3992740471869328,
+      "grad_norm": 0.3303608298301697,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 550
+    },
+    {
+      "epoch": 0.40653357531760437,
+      "grad_norm": 0.33507466316223145,
+      "learning_rate": 0.0002,
+      "loss": 1.203,
+      "step": 560
+    },
+    {
+      "epoch": 0.41379310344827586,
+      "grad_norm": 0.3297670781612396,
+      "learning_rate": 0.0002,
+      "loss": 1.2294,
+      "step": 570
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.32334890961647034,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 580
+    },
+    {
+      "epoch": 0.4283121597096189,
+      "grad_norm": 0.30281195044517517,
+      "learning_rate": 0.0002,
+      "loss": 1.112,
+      "step": 590
+    },
+    {
+      "epoch": 0.4355716878402904,
+      "grad_norm": 0.5900027751922607,
+      "learning_rate": 0.0002,
+      "loss": 1.1352,
+      "step": 600
+    },
+    {
+      "epoch": 0.44283121597096187,
+      "grad_norm": 0.28437477350234985,
+      "learning_rate": 0.0002,
+      "loss": 1.1575,
+      "step": 610
+    },
+    {
+      "epoch": 0.4500907441016334,
+      "grad_norm": 0.39601704478263855,
+      "learning_rate": 0.0002,
+      "loss": 1.2009,
+      "step": 620
+    },
+    {
+      "epoch": 0.4573502722323049,
+      "grad_norm": 0.41971510648727417,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 630
+    },
+    {
+      "epoch": 0.4646098003629764,
+      "grad_norm": 0.33814409375190735,
+      "learning_rate": 0.0002,
+      "loss": 1.236,
+      "step": 640
+    },
+    {
+      "epoch": 0.47186932849364793,
+      "grad_norm": 0.575718104839325,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 650
+    },
+    {
+      "epoch": 0.4791288566243194,
+      "grad_norm": 0.37927401065826416,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 660
+    },
+    {
+      "epoch": 0.4863883847549909,
+      "grad_norm": 0.3224332630634308,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 670
+    },
+    {
+      "epoch": 0.49364791288566245,
+      "grad_norm": 0.32683515548706055,
+      "learning_rate": 0.0002,
+      "loss": 1.1354,
+      "step": 680
+    },
+    {
+      "epoch": 0.5009074410163339,
+      "grad_norm": 0.4316163659095764,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 690
+    },
+    {
+      "epoch": 0.5081669691470054,
+      "grad_norm": 0.342602401971817,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 700
+    },
+    {
+      "epoch": 0.515426497277677,
+      "grad_norm": 0.2794898748397827,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 710
+    },
+    {
+      "epoch": 0.5226860254083484,
+      "grad_norm": 0.3322339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.2395,
+      "step": 720
+    },
+    {
+      "epoch": 0.52994555353902,
+      "grad_norm": 0.3088509142398834,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 730
+    },
+    {
+      "epoch": 0.5372050816696915,
+      "grad_norm": 0.24444378912448883,
+      "learning_rate": 0.0002,
+      "loss": 1.1495,
+      "step": 740
+    },
+    {
+      "epoch": 0.5444646098003629,
+      "grad_norm": 0.3483171761035919,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 750
+    },
+    {
+      "epoch": 0.5517241379310345,
+      "grad_norm": 0.3425690531730652,
+      "learning_rate": 0.0002,
+      "loss": 1.2248,
+      "step": 760
+    },
+    {
+      "epoch": 0.558983666061706,
+      "grad_norm": 0.31841927766799927,
+      "learning_rate": 0.0002,
+      "loss": 1.1127,
+      "step": 770
+    },
+    {
+      "epoch": 0.5662431941923775,
+      "grad_norm": 0.39423868060112,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 780
+    },
+    {
+      "epoch": 0.573502722323049,
+      "grad_norm": 0.30328479409217834,
+      "learning_rate": 0.0002,
+      "loss": 1.0583,
+      "step": 790
+    },
+    {
+      "epoch": 0.5807622504537205,
+      "grad_norm": 0.24475938081741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0718,
+      "step": 800
+    },
+    {
+      "epoch": 0.588021778584392,
+      "grad_norm": 0.37132805585861206,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 810
+    },
+    {
+      "epoch": 0.5952813067150635,
+      "grad_norm": 0.32195979356765747,
+      "learning_rate": 0.0002,
+      "loss": 1.1544,
+      "step": 820
+    },
+    {
+      "epoch": 0.6025408348457351,
+      "grad_norm": 0.2848738729953766,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 830
+    },
+    {
+      "epoch": 0.6098003629764065,
+      "grad_norm": 0.28015264868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.0937,
+      "step": 840
+    },
+    {
+      "epoch": 0.617059891107078,
+      "grad_norm": 0.37796008586883545,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 850
+    },
+    {
+      "epoch": 0.6243194192377496,
+      "grad_norm": 0.39311841130256653,
+      "learning_rate": 0.0002,
+      "loss": 1.1455,
+      "step": 860
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.2761685252189636,
+      "learning_rate": 0.0002,
+      "loss": 1.1066,
+      "step": 870
+    },
+    {
+      "epoch": 0.6388384754990926,
+      "grad_norm": 0.3826720118522644,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 880
+    },
+    {
+      "epoch": 0.6460980036297641,
+      "grad_norm": 0.30076679587364197,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 890
+    },
+    {
+      "epoch": 0.6533575317604355,
+      "grad_norm": 0.21997687220573425,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 900
+    },
+    {
+      "epoch": 0.6606170598911071,
+      "grad_norm": 0.32593777775764465,
+      "learning_rate": 0.0002,
+      "loss": 1.1603,
+      "step": 910
+    },
+    {
+      "epoch": 0.6678765880217786,
+      "grad_norm": 0.30347898602485657,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 920
+    },
+    {
+      "epoch": 0.6751361161524501,
+      "grad_norm": 0.44173774123191833,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 930
+    },
+    {
+      "epoch": 0.6823956442831216,
+      "grad_norm": 0.2507467269897461,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 940
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.29463833570480347,
+      "learning_rate": 0.0002,
+      "loss": 1.3001,
+      "step": 950
+    },
+    {
+      "epoch": 0.6969147005444646,
+      "grad_norm": 0.9363154172897339,
+      "learning_rate": 0.0002,
+      "loss": 1.1651,
+      "step": 960
+    },
+    {
+      "epoch": 0.7041742286751361,
+      "grad_norm": 0.3236212134361267,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 970
+    },
+    {
+      "epoch": 0.7114337568058077,
+      "grad_norm": 0.3123254179954529,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 980
+    },
+    {
+      "epoch": 0.7186932849364791,
+      "grad_norm": 0.3395805060863495,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 990
+    },
+    {
+      "epoch": 0.7259528130671506,
+      "grad_norm": 0.3240964412689209,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7332123411978222,
+      "grad_norm": 0.31902948021888733,
+      "learning_rate": 0.0002,
+      "loss": 1.0313,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7404718693284936,
+      "grad_norm": 0.4848408102989197,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7477313974591652,
+      "grad_norm": 0.33006033301353455,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7549909255898367,
+      "grad_norm": 0.2928730547428131,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7622504537205081,
+      "grad_norm": 0.3529164791107178,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7695099818511797,
+      "grad_norm": 0.2736213803291321,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7767695099818511,
+      "grad_norm": 0.7200686931610107,
+      "learning_rate": 0.0002,
+      "loss": 1.1169,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7840290381125227,
+      "grad_norm": 0.33396708965301514,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7912885662431942,
+      "grad_norm": 1.5760449171066284,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7985480943738656,
+      "grad_norm": 0.28138381242752075,
+      "learning_rate": 0.0002,
+      "loss": 1.1126,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8058076225045372,
+      "grad_norm": 0.2597472369670868,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8130671506352087,
+      "grad_norm": 0.3305445611476898,
+      "learning_rate": 0.0002,
+      "loss": 1.1177,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8203266787658802,
+      "grad_norm": 0.3934599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.0849,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8275862068965517,
+      "grad_norm": 0.3472191393375397,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8348457350272233,
+      "grad_norm": 0.2857365906238556,
+      "learning_rate": 0.0002,
+      "loss": 1.265,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.3207702934741974,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8493647912885662,
+      "grad_norm": 0.3176484704017639,
+      "learning_rate": 0.0002,
+      "loss": 1.1669,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8566243194192378,
+      "grad_norm": 0.40685558319091797,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8638838475499092,
+      "grad_norm": 0.31125199794769287,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8711433756805808,
+      "grad_norm": 0.7361181378364563,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8784029038112523,
+      "grad_norm": 0.33699527382850647,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8856624319419237,
+      "grad_norm": 0.3315220773220062,
+      "learning_rate": 0.0002,
+      "loss": 1.11,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8929219600725953,
+      "grad_norm": 0.6256054043769836,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9001814882032668,
+      "grad_norm": 0.3692137897014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1738,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9074410163339383,
+      "grad_norm": 0.3538484573364258,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9147005444646098,
+      "grad_norm": 0.27863040566444397,
+      "learning_rate": 0.0002,
+      "loss": 1.0878,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9219600725952813,
+      "grad_norm": 0.3322528302669525,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9292196007259528,
+      "grad_norm": 0.7553173303604126,
+      "learning_rate": 0.0002,
+      "loss": 1.0158,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9364791288566243,
+      "grad_norm": 0.4856191575527191,
+      "learning_rate": 0.0002,
+      "loss": 1.198,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9437386569872959,
+      "grad_norm": 0.3668074905872345,
+      "learning_rate": 0.0002,
+      "loss": 1.1678,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9509981851179673,
+      "grad_norm": 0.29851067066192627,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9582577132486388,
+      "grad_norm": 0.276664674282074,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9655172413793104,
+      "grad_norm": 0.2941018342971802,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9727767695099818,
+      "grad_norm": 0.3505859076976776,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9800362976406534,
+      "grad_norm": 0.3067687451839447,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9872958257713249,
+      "grad_norm": 0.27151066064834595,
+      "learning_rate": 0.0002,
+      "loss": 1.1511,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9945553539019963,
+      "grad_norm": 0.36370083689689636,
+      "learning_rate": 0.0002,
+      "loss": 1.1919,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9996370235934664,
+      "eval_loss": 1.1381088495254517,
+      "eval_runtime": 96.0848,
+      "eval_samples_per_second": 4.538,
+      "eval_steps_per_second": 0.572,
+      "step": 1377
+    },
+    {
+      "epoch": 1.0018148820326678,
+      "grad_norm": 0.27980583906173706,
+      "learning_rate": 0.0002,
+      "loss": 1.0082,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0090744101633393,
+      "grad_norm": 0.26713913679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0163339382940109,
+      "grad_norm": 0.3089541494846344,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0235934664246824,
+      "grad_norm": 0.4188242256641388,
+      "learning_rate": 0.0002,
+      "loss": 0.9649,
+      "step": 1410
+    },
+    {
+      "epoch": 1.030852994555354,
+      "grad_norm": 0.5246463418006897,
+      "learning_rate": 0.0002,
+      "loss": 0.9821,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0381125226860255,
+      "grad_norm": 0.2728777825832367,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0453720508166968,
+      "grad_norm": 0.38167616724967957,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.4439380168914795,
+      "learning_rate": 0.0002,
+      "loss": 1.0229,
+      "step": 1450
+    },
+    {
+      "epoch": 1.05989110707804,
+      "grad_norm": 0.30954182147979736,
+      "learning_rate": 0.0002,
+      "loss": 1.0451,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0671506352087115,
+      "grad_norm": 0.4022280275821686,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 1470
+    },
+    {
+      "epoch": 1.074410163339383,
+      "grad_norm": 0.7390811443328857,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0816696914700545,
+      "grad_norm": 0.3885486423969269,
+      "learning_rate": 0.0002,
+      "loss": 1.1378,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0889292196007259,
+      "grad_norm": 0.5275560617446899,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0961887477313974,
+      "grad_norm": 0.35112282633781433,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1510
+    },
+    {
+      "epoch": 1.103448275862069,
+      "grad_norm": 0.33714351058006287,
+      "learning_rate": 0.0002,
+      "loss": 1.0331,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1107078039927405,
+      "grad_norm": 0.31221693754196167,
+      "learning_rate": 0.0002,
+      "loss": 0.8723,
+      "step": 1530
+    },
+    {
+      "epoch": 1.117967332123412,
+      "grad_norm": 0.27549654245376587,
+      "learning_rate": 0.0002,
+      "loss": 1.0701,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1252268602540836,
+      "grad_norm": 0.8465521335601807,
+      "learning_rate": 0.0002,
+      "loss": 1.0114,
+      "step": 1550
+    },
+    {
+      "epoch": 1.132486388384755,
+      "grad_norm": 0.36125949025154114,
+      "learning_rate": 0.0002,
+      "loss": 0.9643,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1397459165154264,
+      "grad_norm": 0.37420371174812317,
+      "learning_rate": 0.0002,
+      "loss": 1.0744,
+      "step": 1570
+    },
+    {
+      "epoch": 1.147005444646098,
+      "grad_norm": 0.3294760584831238,
+      "learning_rate": 0.0002,
+      "loss": 1.165,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1542649727767695,
+      "grad_norm": 0.3881238102912903,
+      "learning_rate": 0.0002,
+      "loss": 0.9207,
+      "step": 1590
+    },
+    {
+      "epoch": 1.161524500907441,
+      "grad_norm": 0.4766491651535034,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1687840290381124,
+      "grad_norm": 0.395530104637146,
+      "learning_rate": 0.0002,
+      "loss": 1.0092,
+      "step": 1610
+    },
+    {
+      "epoch": 1.176043557168784,
+      "grad_norm": 0.3297106623649597,
+      "learning_rate": 0.0002,
+      "loss": 0.9895,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1833030852994555,
+      "grad_norm": 0.39528271555900574,
+      "learning_rate": 0.0002,
+      "loss": 1.1121,
+      "step": 1630
+    },
+    {
+      "epoch": 1.190562613430127,
+      "grad_norm": 0.3370221257209778,
+      "learning_rate": 0.0002,
+      "loss": 0.9202,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1978221415607986,
+      "grad_norm": 0.31922030448913574,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2050816696914701,
+      "grad_norm": 0.6142027378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2123411978221417,
+      "grad_norm": 0.44769710302352905,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 1670
+    },
+    {
+      "epoch": 1.219600725952813,
+      "grad_norm": 0.41233646869659424,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2268602540834845,
+      "grad_norm": 0.2928866147994995,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1690
+    },
+    {
+      "epoch": 1.234119782214156,
+      "grad_norm": 0.36913734674453735,
+      "learning_rate": 0.0002,
+      "loss": 0.9074,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2413793103448276,
+      "grad_norm": 0.5281891226768494,
+      "learning_rate": 0.0002,
+      "loss": 1.0847,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2486388384754992,
+      "grad_norm": 0.3374697268009186,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2558983666061705,
+      "grad_norm": 0.3802020847797394,
+      "learning_rate": 0.0002,
+      "loss": 0.9826,
+      "step": 1730
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.38048651814460754,
+      "learning_rate": 0.0002,
+      "loss": 1.0191,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2704174228675136,
+      "grad_norm": 0.6676169633865356,
+      "learning_rate": 0.0002,
+      "loss": 1.0142,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2776769509981851,
+      "grad_norm": 0.4075961410999298,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2849364791288567,
+      "grad_norm": 0.4374721348285675,
+      "learning_rate": 0.0002,
+      "loss": 1.0301,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2921960072595282,
+      "grad_norm": 0.4638824164867401,
+      "learning_rate": 0.0002,
+      "loss": 1.0601,
+      "step": 1780
+    },
+    {
+      "epoch": 1.2994555353901998,
+      "grad_norm": 0.38631564378738403,
+      "learning_rate": 0.0002,
+      "loss": 0.9593,
+      "step": 1790
+    },
+    {
+      "epoch": 1.306715063520871,
+      "grad_norm": 0.35873809456825256,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3139745916515426,
+      "grad_norm": 0.33602237701416016,
+      "learning_rate": 0.0002,
+      "loss": 0.9835,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3212341197822142,
+      "grad_norm": 0.46696463227272034,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3284936479128857,
+      "grad_norm": 0.368958979845047,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 1830
+    },
+    {
+      "epoch": 1.335753176043557,
+      "grad_norm": 0.3160957396030426,
+      "learning_rate": 0.0002,
+      "loss": 1.0456,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3430127041742286,
+      "grad_norm": 0.4511511027812958,
+      "learning_rate": 0.0002,
+      "loss": 1.0264,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3502722323049001,
+      "grad_norm": 0.6769845485687256,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3575317604355717,
+      "grad_norm": 0.3749309480190277,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3647912885662432,
+      "grad_norm": 0.553798496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.0223,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3720508166969148,
+      "grad_norm": 0.3538985550403595,
+      "learning_rate": 0.0002,
+      "loss": 0.9634,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.5501534938812256,
+      "learning_rate": 0.0002,
+      "loss": 1.0275,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3865698729582578,
+      "grad_norm": 0.4432051181793213,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3938294010889292,
+      "grad_norm": 0.41755786538124084,
+      "learning_rate": 0.0002,
+      "loss": 0.9082,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4010889292196007,
+      "grad_norm": 0.5626114010810852,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4083484573502723,
+      "grad_norm": 0.44038185477256775,
+      "learning_rate": 0.0002,
+      "loss": 1.0687,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4156079854809438,
+      "grad_norm": 0.3397001624107361,
+      "learning_rate": 0.0002,
+      "loss": 0.9454,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4228675136116151,
+      "grad_norm": 0.4325368106365204,
+      "learning_rate": 0.0002,
+      "loss": 0.943,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4301270417422867,
+      "grad_norm": 0.3900907039642334,
+      "learning_rate": 0.0002,
+      "loss": 1.0488,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4373865698729582,
+      "grad_norm": 0.369612455368042,
+      "learning_rate": 0.0002,
+      "loss": 0.9699,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4446460980036298,
+      "grad_norm": 0.4389338791370392,
+      "learning_rate": 0.0002,
+      "loss": 1.0609,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4519056261343013,
+      "grad_norm": 1.694450855255127,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4591651542649728,
+      "grad_norm": 0.516957700252533,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4664246823956444,
+      "grad_norm": 0.45515501499176025,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.4153139591217041,
+      "learning_rate": 0.0002,
+      "loss": 1.0188,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4809437386569873,
+      "grad_norm": 0.44353052973747253,
+      "learning_rate": 0.0002,
+      "loss": 1.1521,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4882032667876588,
+      "grad_norm": 0.570554256439209,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4954627949183303,
+      "grad_norm": 0.5742740035057068,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5027223230490017,
+      "grad_norm": 0.5890517830848694,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5099818511796732,
+      "grad_norm": 0.4162650406360626,
+      "learning_rate": 0.0002,
+      "loss": 1.0639,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5172413793103448,
+      "grad_norm": 0.7334717512130737,
+      "learning_rate": 0.0002,
+      "loss": 1.071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5245009074410163,
+      "grad_norm": 0.2963249683380127,
+      "learning_rate": 0.0002,
+      "loss": 0.8957,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5317604355716878,
+      "grad_norm": 0.30676454305648804,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5390199637023594,
+      "grad_norm": 0.35984641313552856,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 2120
+    },
+    {
+      "epoch": 1.546279491833031,
+      "grad_norm": 0.3384549617767334,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5535390199637025,
+      "grad_norm": 0.4725518226623535,
+      "learning_rate": 0.0002,
+      "loss": 0.9874,
+      "step": 2140
+    },
+    {
+      "epoch": 1.560798548094374,
+      "grad_norm": 0.3252685070037842,
+      "learning_rate": 0.0002,
+      "loss": 1.1166,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5680580762250453,
+      "grad_norm": 0.45043081045150757,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5753176043557169,
+      "grad_norm": 0.374208003282547,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5825771324863884,
+      "grad_norm": 0.5118404030799866,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5898366606170597,
+      "grad_norm": 0.482170969247818,
+      "learning_rate": 0.0002,
+      "loss": 1.0062,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5970961887477313,
+      "grad_norm": 0.5337533950805664,
+      "learning_rate": 0.0002,
+      "loss": 1.0872,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6043557168784028,
+      "grad_norm": 0.5195064544677734,
+      "learning_rate": 0.0002,
+      "loss": 1.0405,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6116152450090744,
+      "grad_norm": 0.30807098746299744,
+      "learning_rate": 0.0002,
+      "loss": 1.0454,
+      "step": 2220
+    },
+    {
+      "epoch": 1.618874773139746,
+      "grad_norm": 0.3962925672531128,
+      "learning_rate": 0.0002,
+      "loss": 1.0293,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6261343012704175,
+      "grad_norm": 0.7636962532997131,
+      "learning_rate": 0.0002,
+      "loss": 1.0137,
+      "step": 2240
+    },
+    {
+      "epoch": 1.633393829401089,
+      "grad_norm": 0.32380592823028564,
+      "learning_rate": 0.0002,
+      "loss": 0.991,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6406533575317606,
+      "grad_norm": 0.5767741799354553,
+      "learning_rate": 0.0002,
+      "loss": 0.9471,
+      "step": 2260
+    },
+    {
+      "epoch": 1.647912885662432,
+      "grad_norm": 0.39964812994003296,
+      "learning_rate": 0.0002,
+      "loss": 0.9524,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6551724137931034,
+      "grad_norm": 0.622629702091217,
+      "learning_rate": 0.0002,
+      "loss": 0.9866,
+      "step": 2280
+    },
+    {
+      "epoch": 1.662431941923775,
+      "grad_norm": 0.40202152729034424,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6696914700544465,
+      "grad_norm": 0.4467349052429199,
+      "learning_rate": 0.0002,
+      "loss": 0.9859,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6769509981851178,
+      "grad_norm": 0.5026949048042297,
+      "learning_rate": 0.0002,
+      "loss": 1.0312,
+      "step": 2310
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.43754541873931885,
+      "learning_rate": 0.0002,
+      "loss": 1.0776,
+      "step": 2320
+    },
+    {
+      "epoch": 1.691470054446461,
+      "grad_norm": 0.42869430780410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6987295825771325,
+      "grad_norm": 0.4192679524421692,
+      "learning_rate": 0.0002,
+      "loss": 1.0747,
+      "step": 2340
+    },
+    {
+      "epoch": 1.705989110707804,
+      "grad_norm": 0.3243155777454376,
+      "learning_rate": 0.0002,
+      "loss": 1.0054,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7132486388384756,
+      "grad_norm": 1.0514075756072998,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 2360
+    },
+    {
+      "epoch": 1.720508166969147,
+      "grad_norm": 0.4719122648239136,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7277676950998186,
+      "grad_norm": 0.3846144676208496,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 2380
+    },
+    {
+      "epoch": 1.73502722323049,
+      "grad_norm": 0.7266581058502197,
+      "learning_rate": 0.0002,
+      "loss": 0.9935,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7422867513611615,
+      "grad_norm": 0.6183241605758667,
+      "learning_rate": 0.0002,
+      "loss": 0.9509,
+      "step": 2400
+    },
+    {
+      "epoch": 1.749546279491833,
+      "grad_norm": 0.3658260405063629,
+      "learning_rate": 0.0002,
+      "loss": 0.9713,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7568058076225044,
+      "grad_norm": 0.6036322712898254,
+      "learning_rate": 0.0002,
+      "loss": 1.0033,
+      "step": 2420
+    },
+    {
+      "epoch": 1.764065335753176,
+      "grad_norm": 0.7872936129570007,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7713248638838475,
+      "grad_norm": 0.35946124792099,
+      "learning_rate": 0.0002,
+      "loss": 0.99,
+      "step": 2440
+    },
+    {
+      "epoch": 1.778584392014519,
+      "grad_norm": 0.3740338981151581,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2450
+    },
+    {
+      "epoch": 1.7858439201451906,
+      "grad_norm": 0.6150230169296265,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 2460
+    },
+    {
+      "epoch": 1.793103448275862,
+      "grad_norm": 0.4726075530052185,
+      "learning_rate": 0.0002,
+      "loss": 1.0373,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8003629764065336,
+      "grad_norm": 0.31292253732681274,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8076225045372052,
+      "grad_norm": 0.4463104009628296,
+      "learning_rate": 0.0002,
+      "loss": 1.0205,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8148820326678767,
+      "grad_norm": 0.7848200798034668,
+      "learning_rate": 0.0002,
+      "loss": 1.0771,
+      "step": 2500
+    },
+    {
+      "epoch": 1.822141560798548,
+      "grad_norm": 0.5562082529067993,
+      "learning_rate": 0.0002,
+      "loss": 0.9508,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8294010889292196,
+      "grad_norm": 0.39892756938934326,
+      "learning_rate": 0.0002,
+      "loss": 0.9343,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8366606170598911,
+      "grad_norm": 1.2923320531845093,
+      "learning_rate": 0.0002,
+      "loss": 1.0489,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8439201451905625,
+      "grad_norm": 0.6316490769386292,
+      "learning_rate": 0.0002,
+      "loss": 1.1203,
+      "step": 2540
+    },
+    {
+      "epoch": 1.851179673321234,
+      "grad_norm": 0.46100497245788574,
+      "learning_rate": 0.0002,
+      "loss": 0.9131,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8584392014519056,
+      "grad_norm": 0.7902987003326416,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 2560
+    },
+    {
+      "epoch": 1.865698729582577,
+      "grad_norm": 0.4596365690231323,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8729582577132486,
+      "grad_norm": 0.6592172384262085,
+      "learning_rate": 0.0002,
+      "loss": 0.9322,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8802177858439202,
+      "grad_norm": 0.5764662027359009,
+      "learning_rate": 0.0002,
+      "loss": 0.9549,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8874773139745917,
+      "grad_norm": 0.8421637415885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9485,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.5635305047035217,
+      "learning_rate": 0.0002,
+      "loss": 1.012,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9019963702359348,
+      "grad_norm": 0.46231237053871155,
+      "learning_rate": 0.0002,
+      "loss": 0.8907,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9092558983666061,
+      "grad_norm": 0.3944607079029083,
+      "learning_rate": 0.0002,
+      "loss": 0.9543,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9165154264972777,
+      "grad_norm": 0.4753907322883606,
+      "learning_rate": 0.0002,
+      "loss": 0.9964,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9237749546279492,
+      "grad_norm": 0.4151090979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.0217,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9310344827586206,
+      "grad_norm": 0.6793725490570068,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 2660
+    },
+    {
+      "epoch": 1.938294010889292,
+      "grad_norm": 0.339755117893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0401,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9455535390199636,
+      "grad_norm": 0.40789374709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.082,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9528130671506352,
+      "grad_norm": 0.3750005066394806,
+      "learning_rate": 0.0002,
+      "loss": 0.9973,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9600725952813067,
+      "grad_norm": 0.39684441685676575,
+      "learning_rate": 0.0002,
+      "loss": 0.9837,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9673321234119783,
+      "grad_norm": 0.378287672996521,
+      "learning_rate": 0.0002,
+      "loss": 1.1204,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9745916515426498,
+      "grad_norm": 0.3668482005596161,
+      "learning_rate": 0.0002,
+      "loss": 0.9284,
+      "step": 2720
+    },
+    {
+      "epoch": 1.9818511796733214,
+      "grad_norm": 0.49997565150260925,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 2730
+    },
+    {
+      "epoch": 1.989110707803993,
+      "grad_norm": 0.36852124333381653,
+      "learning_rate": 0.0002,
+      "loss": 0.9311,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9963702359346642,
+      "grad_norm": 0.5203380584716797,
+      "learning_rate": 0.0002,
+      "loss": 1.0898,
+      "step": 2750
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1086540222167969,
+      "eval_runtime": 95.6866,
+      "eval_samples_per_second": 4.557,
+      "eval_steps_per_second": 0.575,
+      "step": 2755
+    },
+    {
+      "epoch": 2.0036297640653356,
+      "grad_norm": 0.35921573638916016,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2760
+    },
+    {
+      "epoch": 2.010889292196007,
+      "grad_norm": 1.013861894607544,
+      "learning_rate": 0.0002,
+      "loss": 0.8968,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0181488203266786,
+      "grad_norm": 0.4425240159034729,
+      "learning_rate": 0.0002,
+      "loss": 0.7873,
+      "step": 2780
+    },
+    {
+      "epoch": 2.02540834845735,
+      "grad_norm": 0.60174161195755,
+      "learning_rate": 0.0002,
+      "loss": 0.8758,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0326678765880217,
+      "grad_norm": 0.47582098841667175,
+      "learning_rate": 0.0002,
+      "loss": 0.7989,
+      "step": 2800
+    },
+    {
+      "epoch": 2.0399274047186933,
+      "grad_norm": 0.6012811660766602,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 2810
+    },
+    {
+      "epoch": 2.047186932849365,
+      "grad_norm": 0.4444408118724823,
+      "learning_rate": 0.0002,
+      "loss": 0.8109,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0544464609800364,
+      "grad_norm": 0.4864003360271454,
+      "learning_rate": 0.0002,
+      "loss": 0.8621,
+      "step": 2830
+    },
+    {
+      "epoch": 2.061705989110708,
+      "grad_norm": 0.5104215741157532,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.6218489408493042,
+      "learning_rate": 0.0002,
+      "loss": 0.7789,
+      "step": 2850
+    },
+    {
+      "epoch": 2.076225045372051,
+      "grad_norm": 0.705784261226654,
+      "learning_rate": 0.0002,
+      "loss": 0.8778,
+      "step": 2860
+    },
+    {
+      "epoch": 2.0834845735027225,
+      "grad_norm": 0.48091503977775574,
+      "learning_rate": 0.0002,
+      "loss": 0.7851,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0907441016333936,
+      "grad_norm": 0.5062456727027893,
+      "learning_rate": 0.0002,
+      "loss": 0.8444,
+      "step": 2880
+    },
+    {
+      "epoch": 2.098003629764065,
+      "grad_norm": 0.4862022399902344,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 0.48264044523239136,
+      "learning_rate": 0.0002,
+      "loss": 0.8593,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1125226860254083,
+      "grad_norm": 0.43744346499443054,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2910
+    },
+    {
+      "epoch": 2.11978221415608,
+      "grad_norm": 0.5480492115020752,
+      "learning_rate": 0.0002,
+      "loss": 0.9099,
+      "step": 2920
+    },
+    {
+      "epoch": 2.1270417422867514,
+      "grad_norm": 0.5068560838699341,
+      "learning_rate": 0.0002,
+      "loss": 0.7727,
+      "step": 2930
+    },
+    {
+      "epoch": 2.134301270417423,
+      "grad_norm": 0.4650852680206299,
+      "learning_rate": 0.0002,
+      "loss": 0.8403,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1415607985480944,
+      "grad_norm": 0.4929981231689453,
+      "learning_rate": 0.0002,
+      "loss": 0.8467,
+      "step": 2950
+    },
+    {
+      "epoch": 2.148820326678766,
+      "grad_norm": 0.6537389755249023,
+      "learning_rate": 0.0002,
+      "loss": 0.8747,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1560798548094375,
+      "grad_norm": 0.8032940626144409,
+      "learning_rate": 0.0002,
+      "loss": 0.7663,
+      "step": 2970
+    },
+    {
+      "epoch": 2.163339382940109,
+      "grad_norm": 0.7131643891334534,
+      "learning_rate": 0.0002,
+      "loss": 0.7604,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1705989110707806,
+      "grad_norm": 0.6034275889396667,
+      "learning_rate": 0.0002,
+      "loss": 0.8424,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1778584392014517,
+      "grad_norm": 0.6081095933914185,
+      "learning_rate": 0.0002,
+      "loss": 0.8838,
+      "step": 3000
+    },
+    {
+      "epoch": 2.1851179673321233,
+      "grad_norm": 0.5706912875175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9013,
+      "step": 3010
+    },
+    {
+      "epoch": 2.192377495462795,
+      "grad_norm": 0.6742380261421204,
+      "learning_rate": 0.0002,
+      "loss": 0.8473,
+      "step": 3020
+    },
+    {
+      "epoch": 2.1996370235934664,
+      "grad_norm": 0.4847496449947357,
+      "learning_rate": 0.0002,
+      "loss": 0.8372,
+      "step": 3030
+    },
+    {
+      "epoch": 2.206896551724138,
+      "grad_norm": 0.5529342889785767,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2141560798548094,
+      "grad_norm": 0.6108783483505249,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3050
+    },
+    {
+      "epoch": 2.221415607985481,
+      "grad_norm": 0.8841571807861328,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2286751361161525,
+      "grad_norm": 0.4227530360221863,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 3070
+    },
+    {
+      "epoch": 2.235934664246824,
+      "grad_norm": 0.461935818195343,
+      "learning_rate": 0.0002,
+      "loss": 0.7925,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2431941923774956,
+      "grad_norm": 0.5407412648200989,
+      "learning_rate": 0.0002,
+      "loss": 0.8542,
+      "step": 3090
+    },
+    {
+      "epoch": 2.250453720508167,
+      "grad_norm": 0.5057998895645142,
+      "learning_rate": 0.0002,
+      "loss": 0.8392,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2577132486388383,
+      "grad_norm": 0.530057966709137,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 3110
+    },
+    {
+      "epoch": 2.26497277676951,
+      "grad_norm": 0.5066842436790466,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2722323049001814,
+      "grad_norm": 0.5069178342819214,
+      "learning_rate": 0.0002,
+      "loss": 0.7903,
+      "step": 3130
+    },
+    {
+      "epoch": 2.279491833030853,
+      "grad_norm": 0.6095499396324158,
+      "learning_rate": 0.0002,
+      "loss": 0.8458,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2867513611615244,
+      "grad_norm": 0.49524766206741333,
+      "learning_rate": 0.0002,
+      "loss": 0.8171,
+      "step": 3150
+    },
+    {
+      "epoch": 2.294010889292196,
+      "grad_norm": 0.5334409475326538,
+      "learning_rate": 0.0002,
+      "loss": 0.7568,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3012704174228675,
+      "grad_norm": 1.681748867034912,
+      "learning_rate": 0.0002,
+      "loss": 0.7533,
+      "step": 3170
+    },
+    {
+      "epoch": 2.308529945553539,
+      "grad_norm": 0.7225565314292908,
+      "learning_rate": 0.0002,
+      "loss": 0.8155,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3157894736842106,
+      "grad_norm": 0.5379388928413391,
+      "learning_rate": 0.0002,
+      "loss": 0.8034,
+      "step": 3190
+    },
+    {
+      "epoch": 2.323049001814882,
+      "grad_norm": 0.45770326256752014,
+      "learning_rate": 0.0002,
+      "loss": 0.8833,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3303085299455537,
+      "grad_norm": 0.6984533071517944,
+      "learning_rate": 0.0002,
+      "loss": 0.8419,
+      "step": 3210
+    },
+    {
+      "epoch": 2.337568058076225,
+      "grad_norm": 0.6725744605064392,
+      "learning_rate": 0.0002,
+      "loss": 0.7414,
+      "step": 3220
+    },
+    {
+      "epoch": 2.344827586206897,
+      "grad_norm": 1.1247471570968628,
+      "learning_rate": 0.0002,
+      "loss": 0.8657,
+      "step": 3230
+    },
+    {
+      "epoch": 2.352087114337568,
+      "grad_norm": 1.0240263938903809,
+      "learning_rate": 0.0002,
+      "loss": 0.7782,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3593466424682394,
+      "grad_norm": 0.5608096122741699,
+      "learning_rate": 0.0002,
+      "loss": 0.826,
+      "step": 3250
+    },
+    {
+      "epoch": 2.366606170598911,
+      "grad_norm": 0.8294990062713623,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3738656987295825,
+      "grad_norm": 0.6734224557876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9845,
+      "step": 3270
+    },
+    {
+      "epoch": 2.381125226860254,
+      "grad_norm": 0.6862800717353821,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3883847549909256,
+      "grad_norm": 0.5442930459976196,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 3290
+    },
+    {
+      "epoch": 2.395644283121597,
+      "grad_norm": 0.5745394229888916,
+      "learning_rate": 0.0002,
+      "loss": 0.8543,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4029038112522687,
+      "grad_norm": 0.6257799863815308,
+      "learning_rate": 0.0002,
+      "loss": 0.833,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4101633393829403,
+      "grad_norm": 0.5608420968055725,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3320
+    },
+    {
+      "epoch": 2.417422867513612,
+      "grad_norm": 0.5512017011642456,
+      "learning_rate": 0.0002,
+      "loss": 0.8745,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4246823956442833,
+      "grad_norm": 0.7452999353408813,
+      "learning_rate": 0.0002,
+      "loss": 0.8642,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4319419237749544,
+      "grad_norm": 0.4604301452636719,
+      "learning_rate": 0.0002,
+      "loss": 0.8417,
+      "step": 3350
+    },
+    {
+      "epoch": 2.439201451905626,
+      "grad_norm": 0.8225823640823364,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4464609800362975,
+      "grad_norm": 0.8604981303215027,
+      "learning_rate": 0.0002,
+      "loss": 0.8144,
+      "step": 3370
+    },
+    {
+      "epoch": 2.453720508166969,
+      "grad_norm": 0.6620925664901733,
+      "learning_rate": 0.0002,
+      "loss": 0.9276,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4609800362976406,
+      "grad_norm": 0.4750158488750458,
+      "learning_rate": 0.0002,
+      "loss": 0.8381,
+      "step": 3390
+    },
+    {
+      "epoch": 2.468239564428312,
+      "grad_norm": 0.6061418056488037,
+      "learning_rate": 0.0002,
+      "loss": 0.8249,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4754990925589837,
+      "grad_norm": 0.5977247357368469,
+      "learning_rate": 0.0002,
+      "loss": 0.8944,
+      "step": 3410
+    },
+    {
+      "epoch": 2.4827586206896552,
+      "grad_norm": 0.5004227757453918,
+      "learning_rate": 0.0002,
+      "loss": 0.8366,
+      "step": 3420
+    },
+    {
+      "epoch": 2.490018148820327,
+      "grad_norm": 0.46428972482681274,
+      "learning_rate": 0.0002,
+      "loss": 0.9133,
+      "step": 3430
+    },
+    {
+      "epoch": 2.4972776769509983,
+      "grad_norm": 0.7658395171165466,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 3440
+    },
+    {
+      "epoch": 2.50453720508167,
+      "grad_norm": 0.621819794178009,
+      "learning_rate": 0.0002,
+      "loss": 0.9066,
+      "step": 3450
+    },
+    {
+      "epoch": 2.511796733212341,
+      "grad_norm": 0.4556088149547577,
+      "learning_rate": 0.0002,
+      "loss": 0.8481,
+      "step": 3460
+    },
+    {
+      "epoch": 2.519056261343013,
+      "grad_norm": 0.6124140024185181,
+      "learning_rate": 0.0002,
+      "loss": 0.8522,
+      "step": 3470
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 0.6256231665611267,
+      "learning_rate": 0.0002,
+      "loss": 0.8158,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5335753176043556,
+      "grad_norm": 0.5464141964912415,
+      "learning_rate": 0.0002,
+      "loss": 0.9097,
+      "step": 3490
+    },
+    {
+      "epoch": 2.540834845735027,
+      "grad_norm": 0.51471346616745,
+      "learning_rate": 0.0002,
+      "loss": 0.8708,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5480943738656987,
+      "grad_norm": 0.5326165556907654,
+      "learning_rate": 0.0002,
+      "loss": 0.9042,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5553539019963702,
+      "grad_norm": 0.4750378429889679,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 3520
+    },
+    {
+      "epoch": 2.562613430127042,
+      "grad_norm": 0.5292279124259949,
+      "learning_rate": 0.0002,
+      "loss": 0.8248,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5698729582577133,
+      "grad_norm": 0.6145227551460266,
+      "learning_rate": 0.0002,
+      "loss": 0.8808,
+      "step": 3540
+    },
+    {
+      "epoch": 2.577132486388385,
+      "grad_norm": 0.8275189399719238,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5843920145190564,
+      "grad_norm": 0.5037438273429871,
+      "learning_rate": 0.0002,
+      "loss": 0.875,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5916515426497275,
+      "grad_norm": 0.5838707685470581,
+      "learning_rate": 0.0002,
+      "loss": 0.8324,
+      "step": 3570
+    },
+    {
+      "epoch": 2.5989110707803995,
+      "grad_norm": 0.5398710370063782,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6061705989110706,
+      "grad_norm": 0.6115376949310303,
+      "learning_rate": 0.0002,
+      "loss": 0.7843,
+      "step": 3590
+    },
+    {
+      "epoch": 2.613430127041742,
+      "grad_norm": 1.118809461593628,
+      "learning_rate": 0.0002,
+      "loss": 0.8142,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6206896551724137,
+      "grad_norm": 0.7811771631240845,
+      "learning_rate": 0.0002,
+      "loss": 0.8282,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6279491833030852,
+      "grad_norm": 0.5753175020217896,
+      "learning_rate": 0.0002,
+      "loss": 0.7997,
+      "step": 3620
+    },
+    {
+      "epoch": 2.635208711433757,
+      "grad_norm": 0.550829291343689,
+      "learning_rate": 0.0002,
+      "loss": 0.7705,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6424682395644283,
+      "grad_norm": 0.5360019207000732,
+      "learning_rate": 0.0002,
+      "loss": 0.8564,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6497277676951,
+      "grad_norm": 0.63050377368927,
+      "learning_rate": 0.0002,
+      "loss": 0.7648,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6569872958257714,
+      "grad_norm": 0.5833110213279724,
+      "learning_rate": 0.0002,
+      "loss": 0.9156,
+      "step": 3660
+    },
+    {
+      "epoch": 2.664246823956443,
+      "grad_norm": 0.5543047189712524,
+      "learning_rate": 0.0002,
+      "loss": 0.8033,
+      "step": 3670
+    },
+    {
+      "epoch": 2.671506352087114,
+      "grad_norm": 0.6842212080955505,
+      "learning_rate": 0.0002,
+      "loss": 0.8867,
+      "step": 3680
+    },
+    {
+      "epoch": 2.678765880217786,
+      "grad_norm": 0.6859333515167236,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 3690
+    },
+    {
+      "epoch": 2.686025408348457,
+      "grad_norm": 0.7038410902023315,
+      "learning_rate": 0.0002,
+      "loss": 0.8081,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6932849364791287,
+      "grad_norm": 0.447233647108078,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7005444646098002,
+      "grad_norm": 0.6453872919082642,
+      "learning_rate": 0.0002,
+      "loss": 0.8951,
+      "step": 3720
+    },
+    {
+      "epoch": 2.707803992740472,
+      "grad_norm": 0.8025672435760498,
+      "learning_rate": 0.0002,
+      "loss": 0.8292,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7150635208711433,
+      "grad_norm": 0.5997087955474854,
+      "learning_rate": 0.0002,
+      "loss": 0.8488,
+      "step": 3740
+    },
+    {
+      "epoch": 2.722323049001815,
+      "grad_norm": 0.6901142001152039,
+      "learning_rate": 0.0002,
+      "loss": 0.8308,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7295825771324864,
+      "grad_norm": 1.036145567893982,
+      "learning_rate": 0.0002,
+      "loss": 0.8517,
+      "step": 3760
+    },
+    {
+      "epoch": 2.736842105263158,
+      "grad_norm": 0.7207072377204895,
+      "learning_rate": 0.0002,
+      "loss": 0.8455,
+      "step": 3770
+    },
+    {
+      "epoch": 2.7441016333938295,
+      "grad_norm": 1.0452989339828491,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 3780
+    },
+    {
+      "epoch": 2.751361161524501,
+      "grad_norm": 0.5615278482437134,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.46439215540885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7658802177858437,
+      "grad_norm": 0.7134917974472046,
+      "learning_rate": 0.0002,
+      "loss": 0.7847,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7731397459165157,
+      "grad_norm": 0.5139115452766418,
+      "learning_rate": 0.0002,
+      "loss": 0.8315,
+      "step": 3820
+    },
+    {
+      "epoch": 2.780399274047187,
+      "grad_norm": 0.8595781326293945,
+      "learning_rate": 0.0002,
+      "loss": 0.8995,
+      "step": 3830
+    },
+    {
+      "epoch": 2.7876588021778583,
+      "grad_norm": 0.544614851474762,
+      "learning_rate": 0.0002,
+      "loss": 0.9079,
+      "step": 3840
+    },
+    {
+      "epoch": 2.79491833030853,
+      "grad_norm": 0.6073850393295288,
+      "learning_rate": 0.0002,
+      "loss": 0.8322,
+      "step": 3850
+    },
+    {
+      "epoch": 2.8021778584392014,
+      "grad_norm": 0.8224069476127625,
+      "learning_rate": 0.0002,
+      "loss": 0.8277,
+      "step": 3860
+    },
+    {
+      "epoch": 2.809437386569873,
+      "grad_norm": 0.5347970128059387,
+      "learning_rate": 0.0002,
+      "loss": 0.815,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8166969147005445,
+      "grad_norm": 0.6799601912498474,
+      "learning_rate": 0.0002,
+      "loss": 0.8219,
+      "step": 3880
+    },
+    {
+      "epoch": 2.823956442831216,
+      "grad_norm": 0.5219197869300842,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8312159709618876,
+      "grad_norm": 0.5710130333900452,
+      "learning_rate": 0.0002,
+      "loss": 0.8649,
+      "step": 3900
+    },
+    {
+      "epoch": 2.838475499092559,
+      "grad_norm": 0.5857213139533997,
+      "learning_rate": 0.0002,
+      "loss": 0.8741,
+      "step": 3910
+    },
+    {
+      "epoch": 2.8457350272232302,
+      "grad_norm": 0.5206644535064697,
+      "learning_rate": 0.0002,
+      "loss": 0.8154,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8529945553539022,
+      "grad_norm": 1.2902015447616577,
+      "learning_rate": 0.0002,
+      "loss": 0.9328,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8602540834845733,
+      "grad_norm": 0.5252797603607178,
+      "learning_rate": 0.0002,
+      "loss": 0.8192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.867513611615245,
+      "grad_norm": 0.5925108790397644,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8747731397459164,
+      "grad_norm": 0.9719610810279846,
+      "learning_rate": 0.0002,
+      "loss": 0.8611,
+      "step": 3960
+    },
+    {
+      "epoch": 2.882032667876588,
+      "grad_norm": 0.6834747195243835,
+      "learning_rate": 0.0002,
+      "loss": 0.8531,
+      "step": 3970
+    },
+    {
+      "epoch": 2.8892921960072595,
+      "grad_norm": 0.6952353119850159,
+      "learning_rate": 0.0002,
+      "loss": 0.8124,
+      "step": 3980
+    },
+    {
+      "epoch": 2.896551724137931,
+      "grad_norm": 0.49889910221099854,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9038112522686026,
+      "grad_norm": 0.5007026791572571,
+      "learning_rate": 0.0002,
+      "loss": 0.8197,
+      "step": 4000
+    },
+    {
+      "epoch": 2.911070780399274,
+      "grad_norm": 0.5474239587783813,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9183303085299457,
+      "grad_norm": 0.6617428660392761,
+      "learning_rate": 0.0002,
+      "loss": 0.8579,
+      "step": 4020
+    },
+    {
+      "epoch": 2.925589836660617,
+      "grad_norm": 0.6097776293754578,
+      "learning_rate": 0.0002,
+      "loss": 0.8583,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9328493647912888,
+      "grad_norm": 0.5985828638076782,
+      "learning_rate": 0.0002,
+      "loss": 0.8296,
+      "step": 4040
+    },
+    {
+      "epoch": 2.94010889292196,
+      "grad_norm": 0.769488513469696,
+      "learning_rate": 0.0002,
+      "loss": 0.9134,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 0.5167055130004883,
+      "learning_rate": 0.0002,
+      "loss": 0.8148,
+      "step": 4060
+    },
+    {
+      "epoch": 2.954627949183303,
+      "grad_norm": 0.6265496015548706,
+      "learning_rate": 0.0002,
+      "loss": 0.8665,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9618874773139745,
+      "grad_norm": 1.2644082307815552,
+      "learning_rate": 0.0002,
+      "loss": 0.9218,
+      "step": 4080
+    },
+    {
+      "epoch": 2.969147005444646,
+      "grad_norm": 0.6007561087608337,
+      "learning_rate": 0.0002,
+      "loss": 0.8026,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9764065335753176,
+      "grad_norm": 0.47984927892684937,
+      "learning_rate": 0.0002,
+      "loss": 0.8262,
+      "step": 4100
+    },
+    {
+      "epoch": 2.983666061705989,
+      "grad_norm": 1.128198504447937,
+      "learning_rate": 0.0002,
+      "loss": 0.7897,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9909255898366607,
+      "grad_norm": 0.526292085647583,
+      "learning_rate": 0.0002,
+      "loss": 0.7883,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9981851179673322,
+      "grad_norm": 0.5462674498558044,
+      "learning_rate": 0.0002,
+      "loss": 0.7801,
+      "step": 4130
+    },
+    {
+      "epoch": 2.9996370235934666,
+      "eval_loss": 1.1093357801437378,
+      "eval_runtime": 46.2498,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 4132
+    },
+    {
+      "epoch": 3.0054446460980038,
+      "grad_norm": 0.48288026452064514,
+      "learning_rate": 0.0002,
+      "loss": 0.721,
+      "step": 4140
+    },
+    {
+      "epoch": 3.0127041742286753,
+      "grad_norm": 1.0181782245635986,
+      "learning_rate": 0.0002,
+      "loss": 0.6769,
+      "step": 4150
+    },
+    {
+      "epoch": 3.019963702359347,
+      "grad_norm": 0.7718019485473633,
+      "learning_rate": 0.0002,
+      "loss": 0.7185,
+      "step": 4160
+    },
+    {
+      "epoch": 3.027223230490018,
+      "grad_norm": 0.7492219805717468,
+      "learning_rate": 0.0002,
+      "loss": 0.6552,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0344827586206895,
+      "grad_norm": 0.9363632798194885,
+      "learning_rate": 0.0002,
+      "loss": 0.6678,
+      "step": 4180
+    },
+    {
+      "epoch": 3.041742286751361,
+      "grad_norm": 0.6888533234596252,
+      "learning_rate": 0.0002,
+      "loss": 0.7187,
+      "step": 4190
+    },
+    {
+      "epoch": 3.0490018148820326,
+      "grad_norm": 0.7072834968566895,
+      "learning_rate": 0.0002,
+      "loss": 0.6469,
+      "step": 4200
+    },
+    {
+      "epoch": 3.056261343012704,
+      "grad_norm": 0.7182047963142395,
+      "learning_rate": 0.0002,
+      "loss": 0.6387,
+      "step": 4210
+    },
+    {
+      "epoch": 3.0635208711433757,
+      "grad_norm": 0.7194355130195618,
+      "learning_rate": 0.0002,
+      "loss": 0.6385,
+      "step": 4220
+    },
+    {
+      "epoch": 3.0707803992740472,
+      "grad_norm": 0.9454023838043213,
+      "learning_rate": 0.0002,
+      "loss": 0.5812,
+      "step": 4230
+    },
+    {
+      "epoch": 3.0780399274047188,
+      "grad_norm": 0.838657557964325,
+      "learning_rate": 0.0002,
+      "loss": 0.6036,
+      "step": 4240
+    },
+    {
+      "epoch": 3.0852994555353903,
+      "grad_norm": 0.740113377571106,
+      "learning_rate": 0.0002,
+      "loss": 0.646,
+      "step": 4250
+    },
+    {
+      "epoch": 3.092558983666062,
+      "grad_norm": 0.6616561412811279,
+      "learning_rate": 0.0002,
+      "loss": 0.604,
+      "step": 4260
+    },
+    {
+      "epoch": 3.0998185117967334,
+      "grad_norm": 0.8846506476402283,
+      "learning_rate": 0.0002,
+      "loss": 0.6462,
+      "step": 4270
+    },
+    {
+      "epoch": 3.107078039927405,
+      "grad_norm": 0.6322125792503357,
+      "learning_rate": 0.0002,
+      "loss": 0.6037,
+      "step": 4280
+    },
+    {
+      "epoch": 3.114337568058076,
+      "grad_norm": 0.7461467385292053,
+      "learning_rate": 0.0002,
+      "loss": 0.5953,
+      "step": 4290
+    },
+    {
+      "epoch": 3.1215970961887476,
+      "grad_norm": 0.8251287341117859,
+      "learning_rate": 0.0002,
+      "loss": 0.6099,
+      "step": 4300
+    },
+    {
+      "epoch": 3.128856624319419,
+      "grad_norm": 0.8767673373222351,
+      "learning_rate": 0.0002,
+      "loss": 0.6284,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1361161524500907,
+      "grad_norm": 0.7758759260177612,
+      "learning_rate": 0.0002,
+      "loss": 0.7535,
+      "step": 4320
+    },
+    {
+      "epoch": 3.143375680580762,
+      "grad_norm": 1.1056879758834839,
+      "learning_rate": 0.0002,
+      "loss": 0.6624,
+      "step": 4330
+    },
+    {
+      "epoch": 3.1506352087114338,
+      "grad_norm": 0.8259835243225098,
+      "learning_rate": 0.0002,
+      "loss": 0.691,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 0.6607027053833008,
+      "learning_rate": 0.0002,
+      "loss": 0.6635,
+      "step": 4350
+    },
+    {
+      "epoch": 3.165154264972777,
+      "grad_norm": 0.7983301281929016,
+      "learning_rate": 0.0002,
+      "loss": 0.5911,
+      "step": 4360
+    },
+    {
+      "epoch": 3.1724137931034484,
+      "grad_norm": 0.6725239157676697,
+      "learning_rate": 0.0002,
+      "loss": 0.6496,
+      "step": 4370
+    },
+    {
+      "epoch": 3.17967332123412,
+      "grad_norm": 0.9052095413208008,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 4380
+    },
+    {
+      "epoch": 3.1869328493647915,
+      "grad_norm": 0.8131307363510132,
+      "learning_rate": 0.0002,
+      "loss": 0.6877,
+      "step": 4390
+    },
+    {
+      "epoch": 3.1941923774954626,
+      "grad_norm": 0.6435626149177551,
+      "learning_rate": 0.0002,
+      "loss": 0.6384,
+      "step": 4400
+    },
+    {
+      "epoch": 3.201451905626134,
+      "grad_norm": 0.84367436170578,
+      "learning_rate": 0.0002,
+      "loss": 0.5819,
+      "step": 4410
+    },
+    {
+      "epoch": 3.2087114337568057,
+      "grad_norm": 1.5018867254257202,
+      "learning_rate": 0.0002,
+      "loss": 0.6104,
+      "step": 4420
+    },
+    {
+      "epoch": 3.215970961887477,
+      "grad_norm": 0.7019091844558716,
+      "learning_rate": 0.0002,
+      "loss": 0.6838,
+      "step": 4430
+    },
+    {
+      "epoch": 3.2232304900181488,
+      "grad_norm": 0.9164197444915771,
+      "learning_rate": 0.0002,
+      "loss": 0.6153,
+      "step": 4440
+    },
+    {
+      "epoch": 3.2304900181488203,
+      "grad_norm": 0.7890861630439758,
+      "learning_rate": 0.0002,
+      "loss": 0.6618,
+      "step": 4450
+    },
+    {
+      "epoch": 3.237749546279492,
+      "grad_norm": 0.6517660617828369,
+      "learning_rate": 0.0002,
+      "loss": 0.6401,
+      "step": 4460
+    },
+    {
+      "epoch": 3.2450090744101634,
+      "grad_norm": 1.10188889503479,
+      "learning_rate": 0.0002,
+      "loss": 0.6699,
+      "step": 4470
+    },
+    {
+      "epoch": 3.252268602540835,
+      "grad_norm": 0.8158330917358398,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 4480
+    },
+    {
+      "epoch": 3.2595281306715065,
+      "grad_norm": 0.7663109302520752,
+      "learning_rate": 0.0002,
+      "loss": 0.7757,
+      "step": 4490
+    },
+    {
+      "epoch": 3.266787658802178,
+      "grad_norm": 0.8473444581031799,
+      "learning_rate": 0.0002,
+      "loss": 0.6539,
+      "step": 4500
+    },
+    {
+      "epoch": 3.274047186932849,
+      "grad_norm": 0.9724768996238708,
+      "learning_rate": 0.0002,
+      "loss": 0.6511,
+      "step": 4510
+    },
+    {
+      "epoch": 3.281306715063521,
+      "grad_norm": 0.8516759276390076,
+      "learning_rate": 0.0002,
+      "loss": 0.5464,
+      "step": 4520
+    },
+    {
+      "epoch": 3.288566243194192,
+      "grad_norm": 0.7543437480926514,
+      "learning_rate": 0.0002,
+      "loss": 0.6534,
+      "step": 4530
+    },
+    {
+      "epoch": 3.2958257713248638,
+      "grad_norm": 1.0472029447555542,
+      "learning_rate": 0.0002,
+      "loss": 0.6095,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3030852994555353,
+      "grad_norm": 0.6240826845169067,
+      "learning_rate": 0.0002,
+      "loss": 0.6216,
+      "step": 4550
+    },
+    {
+      "epoch": 3.310344827586207,
+      "grad_norm": 0.9957774877548218,
+      "learning_rate": 0.0002,
+      "loss": 0.6223,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3176043557168784,
+      "grad_norm": 0.6448912620544434,
+      "learning_rate": 0.0002,
+      "loss": 0.618,
+      "step": 4570
+    },
+    {
+      "epoch": 3.32486388384755,
+      "grad_norm": 0.7519692778587341,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 4580
+    },
+    {
+      "epoch": 3.3321234119782215,
+      "grad_norm": 0.7367453575134277,
+      "learning_rate": 0.0002,
+      "loss": 0.6672,
+      "step": 4590
+    },
+    {
+      "epoch": 3.339382940108893,
+      "grad_norm": 0.8064960837364197,
+      "learning_rate": 0.0002,
+      "loss": 0.6517,
+      "step": 4600
+    },
+    {
+      "epoch": 3.3466424682395646,
+      "grad_norm": 0.7664631009101868,
+      "learning_rate": 0.0002,
+      "loss": 0.6062,
+      "step": 4610
+    },
+    {
+      "epoch": 3.353901996370236,
+      "grad_norm": 0.7803396582603455,
+      "learning_rate": 0.0002,
+      "loss": 0.6834,
+      "step": 4620
+    },
+    {
+      "epoch": 3.3611615245009077,
+      "grad_norm": 0.9141599535942078,
+      "learning_rate": 0.0002,
+      "loss": 0.6961,
+      "step": 4630
+    },
+    {
+      "epoch": 3.3684210526315788,
+      "grad_norm": 0.9719856381416321,
+      "learning_rate": 0.0002,
+      "loss": 0.6889,
+      "step": 4640
+    },
+    {
+      "epoch": 3.3756805807622503,
+      "grad_norm": 0.9223218560218811,
+      "learning_rate": 0.0002,
+      "loss": 0.6914,
+      "step": 4650
+    },
+    {
+      "epoch": 3.382940108892922,
+      "grad_norm": 0.7289277911186218,
+      "learning_rate": 0.0002,
+      "loss": 0.5981,
+      "step": 4660
+    },
+    {
+      "epoch": 3.3901996370235934,
+      "grad_norm": 1.039724349975586,
+      "learning_rate": 0.0002,
+      "loss": 0.595,
+      "step": 4670
+    },
+    {
+      "epoch": 3.397459165154265,
+      "grad_norm": 1.397438883781433,
+      "learning_rate": 0.0002,
+      "loss": 0.8121,
+      "step": 4680
+    },
+    {
+      "epoch": 3.4047186932849365,
+      "grad_norm": 1.0069999694824219,
+      "learning_rate": 0.0002,
+      "loss": 0.6334,
+      "step": 4690
+    },
+    {
+      "epoch": 3.411978221415608,
+      "grad_norm": 0.816291332244873,
+      "learning_rate": 0.0002,
+      "loss": 0.6598,
+      "step": 4700
+    },
+    {
+      "epoch": 3.4192377495462796,
+      "grad_norm": 1.2831530570983887,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 4710
+    },
+    {
+      "epoch": 3.426497277676951,
+      "grad_norm": 0.9573889970779419,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 4720
+    },
+    {
+      "epoch": 3.4337568058076227,
+      "grad_norm": 0.7685632705688477,
+      "learning_rate": 0.0002,
+      "loss": 0.7279,
+      "step": 4730
+    },
+    {
+      "epoch": 3.441016333938294,
+      "grad_norm": 0.7019195556640625,
+      "learning_rate": 0.0002,
+      "loss": 0.6104,
+      "step": 4740
+    },
+    {
+      "epoch": 3.4482758620689653,
+      "grad_norm": 0.7244833707809448,
+      "learning_rate": 0.0002,
+      "loss": 0.7606,
+      "step": 4750
+    },
+    {
+      "epoch": 3.455535390199637,
+      "grad_norm": 1.3468551635742188,
+      "learning_rate": 0.0002,
+      "loss": 0.6951,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4627949183303084,
+      "grad_norm": 0.822846531867981,
+      "learning_rate": 0.0002,
+      "loss": 0.6945,
+      "step": 4770
+    },
+    {
+      "epoch": 3.47005444646098,
+      "grad_norm": 0.7311608195304871,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 4780
+    },
+    {
+      "epoch": 3.4773139745916515,
+      "grad_norm": 0.9466770887374878,
+      "learning_rate": 0.0002,
+      "loss": 0.7019,
+      "step": 4790
+    },
+    {
+      "epoch": 3.484573502722323,
+      "grad_norm": 1.1527155637741089,
+      "learning_rate": 0.0002,
+      "loss": 0.7767,
+      "step": 4800
+    },
+    {
+      "epoch": 3.4918330308529946,
+      "grad_norm": 1.1288906335830688,
+      "learning_rate": 0.0002,
+      "loss": 0.6882,
+      "step": 4810
+    },
+    {
+      "epoch": 3.499092558983666,
+      "grad_norm": 0.9096164107322693,
+      "learning_rate": 0.0002,
+      "loss": 0.6564,
+      "step": 4820
+    },
+    {
+      "epoch": 3.5063520871143377,
+      "grad_norm": 0.7988565564155579,
+      "learning_rate": 0.0002,
+      "loss": 0.6127,
+      "step": 4830
+    },
+    {
+      "epoch": 3.513611615245009,
+      "grad_norm": 0.7183415293693542,
+      "learning_rate": 0.0002,
+      "loss": 0.7004,
+      "step": 4840
+    },
+    {
+      "epoch": 3.5208711433756807,
+      "grad_norm": 0.6614915132522583,
+      "learning_rate": 0.0002,
+      "loss": 0.74,
+      "step": 4850
+    },
+    {
+      "epoch": 3.528130671506352,
+      "grad_norm": 0.8609521985054016,
+      "learning_rate": 0.0002,
+      "loss": 0.7271,
+      "step": 4860
+    },
+    {
+      "epoch": 3.535390199637024,
+      "grad_norm": 0.86552894115448,
+      "learning_rate": 0.0002,
+      "loss": 0.6664,
+      "step": 4870
+    },
+    {
+      "epoch": 3.542649727767695,
+      "grad_norm": 0.6926496028900146,
+      "learning_rate": 0.0002,
+      "loss": 0.6432,
+      "step": 4880
+    },
+    {
+      "epoch": 3.5499092558983665,
+      "grad_norm": 0.8157467246055603,
+      "learning_rate": 0.0002,
+      "loss": 0.7117,
+      "step": 4890
+    },
+    {
+      "epoch": 3.557168784029038,
+      "grad_norm": 0.9085357189178467,
+      "learning_rate": 0.0002,
+      "loss": 0.6201,
+      "step": 4900
+    },
+    {
+      "epoch": 3.5644283121597096,
+      "grad_norm": 0.6322644948959351,
+      "learning_rate": 0.0002,
+      "loss": 0.6521,
+      "step": 4910
+    },
+    {
+      "epoch": 3.571687840290381,
+      "grad_norm": 1.263205885887146,
+      "learning_rate": 0.0002,
+      "loss": 0.6607,
+      "step": 4920
+    },
+    {
+      "epoch": 3.5789473684210527,
+      "grad_norm": 0.8901070356369019,
+      "learning_rate": 0.0002,
+      "loss": 0.6657,
+      "step": 4930
+    },
+    {
+      "epoch": 3.586206896551724,
+      "grad_norm": 0.7983952164649963,
+      "learning_rate": 0.0002,
+      "loss": 0.6434,
+      "step": 4940
+    },
+    {
+      "epoch": 3.5934664246823957,
+      "grad_norm": 0.9887813925743103,
+      "learning_rate": 0.0002,
+      "loss": 0.6861,
+      "step": 4950
+    },
+    {
+      "epoch": 3.6007259528130673,
+      "grad_norm": 0.7895187735557556,
+      "learning_rate": 0.0002,
+      "loss": 0.6502,
+      "step": 4960
+    },
+    {
+      "epoch": 3.6079854809437384,
+      "grad_norm": 0.9685819745063782,
+      "learning_rate": 0.0002,
+      "loss": 0.7111,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6152450090744104,
+      "grad_norm": 0.6576591730117798,
+      "learning_rate": 0.0002,
+      "loss": 0.6915,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6225045372050815,
+      "grad_norm": 0.856985330581665,
+      "learning_rate": 0.0002,
+      "loss": 0.6195,
+      "step": 4990
+    },
+    {
+      "epoch": 3.629764065335753,
+      "grad_norm": 0.7230252623558044,
+      "learning_rate": 0.0002,
+      "loss": 0.6318,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6370235934664246,
+      "grad_norm": 0.8260893821716309,
+      "learning_rate": 0.0002,
+      "loss": 0.742,
+      "step": 5010
+    },
+    {
+      "epoch": 3.644283121597096,
+      "grad_norm": 0.7635950446128845,
+      "learning_rate": 0.0002,
+      "loss": 0.7223,
+      "step": 5020
+    },
+    {
+      "epoch": 3.6515426497277677,
+      "grad_norm": 0.7060768604278564,
+      "learning_rate": 0.0002,
+      "loss": 0.6837,
+      "step": 5030
+    },
+    {
+      "epoch": 3.658802177858439,
+      "grad_norm": 0.8020303249359131,
+      "learning_rate": 0.0002,
+      "loss": 0.6921,
+      "step": 5040
+    },
+    {
+      "epoch": 3.6660617059891107,
+      "grad_norm": 0.8530341386795044,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 5050
+    },
+    {
+      "epoch": 3.6733212341197823,
+      "grad_norm": 0.6667101979255676,
+      "learning_rate": 0.0002,
+      "loss": 0.7222,
+      "step": 5060
+    },
+    {
+      "epoch": 3.680580762250454,
+      "grad_norm": 0.7385406494140625,
+      "learning_rate": 0.0002,
+      "loss": 0.7081,
+      "step": 5070
+    },
+    {
+      "epoch": 3.6878402903811254,
+      "grad_norm": 0.7753380537033081,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 5080
+    },
+    {
+      "epoch": 3.695099818511797,
+      "grad_norm": 0.7516207098960876,
+      "learning_rate": 0.0002,
+      "loss": 0.6491,
+      "step": 5090
+    },
+    {
+      "epoch": 3.702359346642468,
+      "grad_norm": 0.8171586394309998,
+      "learning_rate": 0.0002,
+      "loss": 0.672,
+      "step": 5100
+    },
+    {
+      "epoch": 3.70961887477314,
+      "grad_norm": 1.0796279907226562,
+      "learning_rate": 0.0002,
+      "loss": 0.7459,
+      "step": 5110
+    },
+    {
+      "epoch": 3.716878402903811,
+      "grad_norm": 0.6957688927650452,
+      "learning_rate": 0.0002,
+      "loss": 0.5948,
+      "step": 5120
+    },
+    {
+      "epoch": 3.7241379310344827,
+      "grad_norm": 0.8550161719322205,
+      "learning_rate": 0.0002,
+      "loss": 0.7515,
+      "step": 5130
+    },
+    {
+      "epoch": 3.731397459165154,
+      "grad_norm": 0.9396728277206421,
+      "learning_rate": 0.0002,
+      "loss": 0.7286,
+      "step": 5140
+    },
+    {
+      "epoch": 3.7386569872958257,
+      "grad_norm": 1.4264805316925049,
+      "learning_rate": 0.0002,
+      "loss": 0.7594,
+      "step": 5150
+    },
+    {
+      "epoch": 3.7459165154264973,
+      "grad_norm": 0.8725108504295349,
+      "learning_rate": 0.0002,
+      "loss": 0.6575,
+      "step": 5160
+    },
+    {
+      "epoch": 3.753176043557169,
+      "grad_norm": 1.0346195697784424,
+      "learning_rate": 0.0002,
+      "loss": 0.6778,
+      "step": 5170
+    },
+    {
+      "epoch": 3.7604355716878404,
+      "grad_norm": 0.5395554304122925,
+      "learning_rate": 0.0002,
+      "loss": 0.6371,
+      "step": 5180
+    },
+    {
+      "epoch": 3.767695099818512,
+      "grad_norm": 1.3153616189956665,
+      "learning_rate": 0.0002,
+      "loss": 0.7308,
+      "step": 5190
+    },
+    {
+      "epoch": 3.7749546279491835,
+      "grad_norm": 0.9879828691482544,
+      "learning_rate": 0.0002,
+      "loss": 0.78,
+      "step": 5200
+    },
+    {
+      "epoch": 3.7822141560798546,
+      "grad_norm": 0.8876672983169556,
+      "learning_rate": 0.0002,
+      "loss": 0.7068,
+      "step": 5210
+    },
+    {
+      "epoch": 3.7894736842105265,
+      "grad_norm": 0.8363267779350281,
+      "learning_rate": 0.0002,
+      "loss": 0.6283,
+      "step": 5220
+    },
+    {
+      "epoch": 3.7967332123411976,
+      "grad_norm": 0.637294590473175,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 5230
+    },
+    {
+      "epoch": 3.803992740471869,
+      "grad_norm": 1.1408970355987549,
+      "learning_rate": 0.0002,
+      "loss": 0.6685,
+      "step": 5240
+    },
+    {
+      "epoch": 3.8112522686025407,
+      "grad_norm": 1.0128360986709595,
+      "learning_rate": 0.0002,
+      "loss": 0.6761,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8185117967332123,
+      "grad_norm": 0.8061144351959229,
+      "learning_rate": 0.0002,
+      "loss": 0.6764,
+      "step": 5260
+    },
+    {
+      "epoch": 3.825771324863884,
+      "grad_norm": 0.9626626968383789,
+      "learning_rate": 0.0002,
+      "loss": 0.7254,
+      "step": 5270
+    },
+    {
+      "epoch": 3.8330308529945554,
+      "grad_norm": 0.9013627171516418,
+      "learning_rate": 0.0002,
+      "loss": 0.7367,
+      "step": 5280
+    },
+    {
+      "epoch": 3.840290381125227,
+      "grad_norm": 0.8411344289779663,
+      "learning_rate": 0.0002,
+      "loss": 0.6806,
+      "step": 5290
+    },
+    {
+      "epoch": 3.8475499092558985,
+      "grad_norm": 0.7426059246063232,
+      "learning_rate": 0.0002,
+      "loss": 0.6818,
+      "step": 5300
+    },
+    {
+      "epoch": 3.85480943738657,
+      "grad_norm": 1.003413438796997,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 5310
+    },
+    {
+      "epoch": 3.862068965517241,
+      "grad_norm": 0.7527840733528137,
+      "learning_rate": 0.0002,
+      "loss": 0.8554,
+      "step": 5320
+    },
+    {
+      "epoch": 3.869328493647913,
+      "grad_norm": 0.738610565662384,
+      "learning_rate": 0.0002,
+      "loss": 0.7521,
+      "step": 5330
+    },
+    {
+      "epoch": 3.876588021778584,
+      "grad_norm": 0.7277999520301819,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 5340
+    },
+    {
+      "epoch": 3.8838475499092557,
+      "grad_norm": 0.5951359272003174,
+      "learning_rate": 0.0002,
+      "loss": 0.7503,
+      "step": 5350
+    },
+    {
+      "epoch": 3.8911070780399273,
+      "grad_norm": 1.043884038925171,
+      "learning_rate": 0.0002,
+      "loss": 0.7447,
+      "step": 5360
+    },
+    {
+      "epoch": 3.898366606170599,
+      "grad_norm": 0.8436498045921326,
+      "learning_rate": 0.0002,
+      "loss": 0.6862,
+      "step": 5370
+    },
+    {
+      "epoch": 3.9056261343012704,
+      "grad_norm": 0.5603365302085876,
+      "learning_rate": 0.0002,
+      "loss": 0.665,
+      "step": 5380
+    },
+    {
+      "epoch": 3.912885662431942,
+      "grad_norm": 1.0128886699676514,
+      "learning_rate": 0.0002,
+      "loss": 0.7098,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9201451905626135,
+      "grad_norm": 0.7970930337905884,
+      "learning_rate": 0.0002,
+      "loss": 0.6707,
+      "step": 5400
+    },
+    {
+      "epoch": 3.927404718693285,
+      "grad_norm": 0.7699369192123413,
+      "learning_rate": 0.0002,
+      "loss": 0.637,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9346642468239565,
+      "grad_norm": 0.800561249256134,
+      "learning_rate": 0.0002,
+      "loss": 0.6742,
+      "step": 5420
+    },
+    {
+      "epoch": 3.941923774954628,
+      "grad_norm": 0.8020331859588623,
+      "learning_rate": 0.0002,
+      "loss": 0.7208,
+      "step": 5430
+    },
+    {
+      "epoch": 3.9491833030852996,
+      "grad_norm": 0.7461140155792236,
+      "learning_rate": 0.0002,
+      "loss": 0.7294,
+      "step": 5440
+    },
+    {
+      "epoch": 3.9564428312159707,
+      "grad_norm": 0.8346918821334839,
+      "learning_rate": 0.0002,
+      "loss": 0.7013,
+      "step": 5450
+    },
+    {
+      "epoch": 3.9637023593466427,
+      "grad_norm": 0.9723302125930786,
+      "learning_rate": 0.0002,
+      "loss": 0.6289,
+      "step": 5460
+    },
+    {
+      "epoch": 3.970961887477314,
+      "grad_norm": 0.6809740662574768,
+      "learning_rate": 0.0002,
+      "loss": 0.8029,
+      "step": 5470
+    },
+    {
+      "epoch": 3.9782214156079854,
+      "grad_norm": 0.7353498339653015,
+      "learning_rate": 0.0002,
+      "loss": 0.6896,
+      "step": 5480
+    },
+    {
+      "epoch": 3.985480943738657,
+      "grad_norm": 0.748009443283081,
+      "learning_rate": 0.0002,
+      "loss": 0.6722,
+      "step": 5490
+    },
+    {
+      "epoch": 3.9927404718693285,
+      "grad_norm": 1.3656195402145386,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 5500
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.8402108550071716,
+      "learning_rate": 0.0002,
+      "loss": 0.7368,
+      "step": 5510
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.17229425907135,
+      "eval_runtime": 46.2554,
+      "eval_samples_per_second": 9.426,
+      "eval_steps_per_second": 1.189,
+      "step": 5510
+    },
+    {
+      "epoch": 4.007259528130671,
+      "grad_norm": 0.8601235747337341,
+      "learning_rate": 0.0002,
+      "loss": 0.4637,
+      "step": 5520
+    },
+    {
+      "epoch": 4.014519056261343,
+      "grad_norm": 1.2635200023651123,
+      "learning_rate": 0.0002,
+      "loss": 0.4717,
+      "step": 5530
+    },
+    {
+      "epoch": 4.021778584392014,
+      "grad_norm": 1.0257477760314941,
+      "learning_rate": 0.0002,
+      "loss": 0.503,
+      "step": 5540
+    },
+    {
+      "epoch": 4.029038112522686,
+      "grad_norm": 0.9436745047569275,
+      "learning_rate": 0.0002,
+      "loss": 0.4547,
+      "step": 5550
+    },
+    {
+      "epoch": 4.036297640653357,
+      "grad_norm": 0.9443606734275818,
+      "learning_rate": 0.0002,
+      "loss": 0.459,
+      "step": 5560
+    },
+    {
+      "epoch": 4.043557168784029,
+      "grad_norm": 1.3965742588043213,
+      "learning_rate": 0.0002,
+      "loss": 0.5386,
+      "step": 5570
+    },
+    {
+      "epoch": 4.0508166969147,
+      "grad_norm": 0.8973520398139954,
+      "learning_rate": 0.0002,
+      "loss": 0.4248,
+      "step": 5580
+    },
+    {
+      "epoch": 4.058076225045372,
+      "grad_norm": 0.9998409748077393,
+      "learning_rate": 0.0002,
+      "loss": 0.4111,
+      "step": 5590
+    },
+    {
+      "epoch": 4.0653357531760435,
+      "grad_norm": 1.1213387250900269,
+      "learning_rate": 0.0002,
+      "loss": 0.4828,
+      "step": 5600
+    },
+    {
+      "epoch": 4.072595281306715,
+      "grad_norm": 0.7064558863639832,
+      "learning_rate": 0.0002,
+      "loss": 0.439,
+      "step": 5610
+    },
+    {
+      "epoch": 4.0798548094373865,
+      "grad_norm": 1.2390803098678589,
+      "learning_rate": 0.0002,
+      "loss": 0.4607,
+      "step": 5620
+    },
+    {
+      "epoch": 4.087114337568058,
+      "grad_norm": 1.123469591140747,
+      "learning_rate": 0.0002,
+      "loss": 0.5014,
+      "step": 5630
+    },
+    {
+      "epoch": 4.09437386569873,
+      "grad_norm": 1.229573369026184,
+      "learning_rate": 0.0002,
+      "loss": 0.513,
+      "step": 5640
+    },
+    {
+      "epoch": 4.101633393829401,
+      "grad_norm": 1.7182831764221191,
+      "learning_rate": 0.0002,
+      "loss": 0.5258,
+      "step": 5650
+    },
+    {
+      "epoch": 4.108892921960073,
+      "grad_norm": 0.894903302192688,
+      "learning_rate": 0.0002,
+      "loss": 0.5371,
+      "step": 5660
+    },
+    {
+      "epoch": 4.116152450090744,
+      "grad_norm": 0.8754552006721497,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 5670
+    },
+    {
+      "epoch": 4.123411978221416,
+      "grad_norm": 1.2401553392410278,
+      "learning_rate": 0.0002,
+      "loss": 0.491,
+      "step": 5680
+    },
+    {
+      "epoch": 4.130671506352087,
+      "grad_norm": 0.8631148934364319,
+      "learning_rate": 0.0002,
+      "loss": 0.4549,
+      "step": 5690
+    },
+    {
+      "epoch": 4.137931034482759,
+      "grad_norm": 1.1798022985458374,
+      "learning_rate": 0.0002,
+      "loss": 0.487,
+      "step": 5700
+    },
+    {
+      "epoch": 4.14519056261343,
+      "grad_norm": 0.8344549536705017,
+      "learning_rate": 0.0002,
+      "loss": 0.4522,
+      "step": 5710
+    },
+    {
+      "epoch": 4.152450090744102,
+      "grad_norm": 1.2342697381973267,
+      "learning_rate": 0.0002,
+      "loss": 0.4559,
+      "step": 5720
+    },
+    {
+      "epoch": 4.159709618874773,
+      "grad_norm": 1.1601094007492065,
+      "learning_rate": 0.0002,
+      "loss": 0.53,
+      "step": 5730
+    },
+    {
+      "epoch": 4.166969147005445,
+      "grad_norm": 1.2925703525543213,
+      "learning_rate": 0.0002,
+      "loss": 0.4755,
+      "step": 5740
+    },
+    {
+      "epoch": 4.174228675136116,
+      "grad_norm": 1.0870997905731201,
+      "learning_rate": 0.0002,
+      "loss": 0.4973,
+      "step": 5750
+    },
+    {
+      "epoch": 4.181488203266787,
+      "grad_norm": 0.9077792763710022,
+      "learning_rate": 0.0002,
+      "loss": 0.5184,
+      "step": 5760
+    },
+    {
+      "epoch": 4.188747731397459,
+      "grad_norm": 1.009273886680603,
+      "learning_rate": 0.0002,
+      "loss": 0.4606,
+      "step": 5770
+    },
+    {
+      "epoch": 4.19600725952813,
+      "grad_norm": 1.2465530633926392,
+      "learning_rate": 0.0002,
+      "loss": 0.5383,
+      "step": 5780
+    },
+    {
+      "epoch": 4.203266787658802,
+      "grad_norm": 1.2261253595352173,
+      "learning_rate": 0.0002,
+      "loss": 0.4938,
+      "step": 5790
+    },
+    {
+      "epoch": 4.2105263157894735,
+      "grad_norm": 1.1498041152954102,
+      "learning_rate": 0.0002,
+      "loss": 0.5255,
+      "step": 5800
+    },
+    {
+      "epoch": 4.217785843920145,
+      "grad_norm": 1.1966725587844849,
+      "learning_rate": 0.0002,
+      "loss": 0.5269,
+      "step": 5810
+    },
+    {
+      "epoch": 4.2250453720508165,
+      "grad_norm": 1.2651296854019165,
+      "learning_rate": 0.0002,
+      "loss": 0.5626,
+      "step": 5820
+    },
+    {
+      "epoch": 4.2323049001814885,
+      "grad_norm": 1.0388574600219727,
+      "learning_rate": 0.0002,
+      "loss": 0.5213,
+      "step": 5830
+    },
+    {
+      "epoch": 4.23956442831216,
+      "grad_norm": 1.3042771816253662,
+      "learning_rate": 0.0002,
+      "loss": 0.4965,
+      "step": 5840
+    },
+    {
+      "epoch": 4.246823956442832,
+      "grad_norm": 1.1127727031707764,
+      "learning_rate": 0.0002,
+      "loss": 0.5116,
+      "step": 5850
+    },
+    {
+      "epoch": 4.254083484573503,
+      "grad_norm": 0.9653958082199097,
+      "learning_rate": 0.0002,
+      "loss": 0.5197,
+      "step": 5860
+    },
+    {
+      "epoch": 4.261343012704174,
+      "grad_norm": 1.0500504970550537,
+      "learning_rate": 0.0002,
+      "loss": 0.4414,
+      "step": 5870
+    },
+    {
+      "epoch": 4.268602540834846,
+      "grad_norm": 1.1476165056228638,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 5880
+    },
+    {
+      "epoch": 4.275862068965517,
+      "grad_norm": 0.9424414038658142,
+      "learning_rate": 0.0002,
+      "loss": 0.4667,
+      "step": 5890
+    },
+    {
+      "epoch": 4.283121597096189,
+      "grad_norm": 1.3309166431427002,
+      "learning_rate": 0.0002,
+      "loss": 0.5039,
+      "step": 5900
+    },
+    {
+      "epoch": 4.29038112522686,
+      "grad_norm": 1.3025873899459839,
+      "learning_rate": 0.0002,
+      "loss": 0.5472,
+      "step": 5910
+    },
+    {
+      "epoch": 4.297640653357532,
+      "grad_norm": 1.1442325115203857,
+      "learning_rate": 0.0002,
+      "loss": 0.4644,
+      "step": 5920
+    },
+    {
+      "epoch": 4.304900181488203,
+      "grad_norm": 0.9820859432220459,
+      "learning_rate": 0.0002,
+      "loss": 0.5066,
+      "step": 5930
+    },
+    {
+      "epoch": 4.312159709618875,
+      "grad_norm": 0.9615740180015564,
+      "learning_rate": 0.0002,
+      "loss": 0.5224,
+      "step": 5940
+    },
+    {
+      "epoch": 4.319419237749546,
+      "grad_norm": 1.1627109050750732,
+      "learning_rate": 0.0002,
+      "loss": 0.5665,
+      "step": 5950
+    },
+    {
+      "epoch": 4.326678765880218,
+      "grad_norm": 0.9381322860717773,
+      "learning_rate": 0.0002,
+      "loss": 0.4852,
+      "step": 5960
+    },
+    {
+      "epoch": 4.333938294010889,
+      "grad_norm": 0.8154335618019104,
+      "learning_rate": 0.0002,
+      "loss": 0.4532,
+      "step": 5970
+    },
+    {
+      "epoch": 4.341197822141561,
+      "grad_norm": 0.877671480178833,
+      "learning_rate": 0.0002,
+      "loss": 0.5237,
+      "step": 5980
+    },
+    {
+      "epoch": 4.348457350272232,
+      "grad_norm": 1.1742031574249268,
+      "learning_rate": 0.0002,
+      "loss": 0.6113,
+      "step": 5990
+    },
+    {
+      "epoch": 4.3557168784029034,
+      "grad_norm": 1.0352917909622192,
+      "learning_rate": 0.0002,
+      "loss": 0.5704,
+      "step": 6000
+    },
+    {
+      "epoch": 4.362976406533575,
+      "grad_norm": 0.9963878989219666,
+      "learning_rate": 0.0002,
+      "loss": 0.4996,
+      "step": 6010
+    },
+    {
+      "epoch": 4.3702359346642465,
+      "grad_norm": 1.1892237663269043,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 6020
+    },
+    {
+      "epoch": 4.3774954627949185,
+      "grad_norm": 1.2516111135482788,
+      "learning_rate": 0.0002,
+      "loss": 0.5224,
+      "step": 6030
+    },
+    {
+      "epoch": 4.38475499092559,
+      "grad_norm": 1.2111951112747192,
+      "learning_rate": 0.0002,
+      "loss": 0.5952,
+      "step": 6040
+    },
+    {
+      "epoch": 4.392014519056262,
+      "grad_norm": 1.0820083618164062,
+      "learning_rate": 0.0002,
+      "loss": 0.4275,
+      "step": 6050
+    },
+    {
+      "epoch": 4.399274047186933,
+      "grad_norm": 1.033915638923645,
+      "learning_rate": 0.0002,
+      "loss": 0.5117,
+      "step": 6060
+    },
+    {
+      "epoch": 4.406533575317605,
+      "grad_norm": 1.0635870695114136,
+      "learning_rate": 0.0002,
+      "loss": 0.5431,
+      "step": 6070
+    },
+    {
+      "epoch": 4.413793103448276,
+      "grad_norm": 1.0520414113998413,
+      "learning_rate": 0.0002,
+      "loss": 0.5341,
+      "step": 6080
+    },
+    {
+      "epoch": 4.421052631578947,
+      "grad_norm": 1.0821926593780518,
+      "learning_rate": 0.0002,
+      "loss": 0.512,
+      "step": 6090
+    },
+    {
+      "epoch": 4.428312159709619,
+      "grad_norm": 1.0533246994018555,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 6100
+    },
+    {
+      "epoch": 4.43557168784029,
+      "grad_norm": 0.9231932759284973,
+      "learning_rate": 0.0002,
+      "loss": 0.4577,
+      "step": 6110
+    },
+    {
+      "epoch": 4.442831215970962,
+      "grad_norm": 0.9910260438919067,
+      "learning_rate": 0.0002,
+      "loss": 0.583,
+      "step": 6120
+    },
+    {
+      "epoch": 4.450090744101633,
+      "grad_norm": 1.061949372291565,
+      "learning_rate": 0.0002,
+      "loss": 0.4717,
+      "step": 6130
+    },
+    {
+      "epoch": 4.457350272232305,
+      "grad_norm": 1.2927039861679077,
+      "learning_rate": 0.0002,
+      "loss": 0.5893,
+      "step": 6140
+    },
+    {
+      "epoch": 4.464609800362976,
+      "grad_norm": 1.3966081142425537,
+      "learning_rate": 0.0002,
+      "loss": 0.4684,
+      "step": 6150
+    },
+    {
+      "epoch": 4.471869328493648,
+      "grad_norm": 1.3835992813110352,
+      "learning_rate": 0.0002,
+      "loss": 0.5507,
+      "step": 6160
+    },
+    {
+      "epoch": 4.479128856624319,
+      "grad_norm": 1.0892692804336548,
+      "learning_rate": 0.0002,
+      "loss": 0.5911,
+      "step": 6170
+    },
+    {
+      "epoch": 4.486388384754991,
+      "grad_norm": 1.0318800210952759,
+      "learning_rate": 0.0002,
+      "loss": 0.478,
+      "step": 6180
+    },
+    {
+      "epoch": 4.493647912885662,
+      "grad_norm": 0.8174677491188049,
+      "learning_rate": 0.0002,
+      "loss": 0.5198,
+      "step": 6190
+    },
+    {
+      "epoch": 4.500907441016334,
+      "grad_norm": 1.4157509803771973,
+      "learning_rate": 0.0002,
+      "loss": 0.5387,
+      "step": 6200
+    },
+    {
+      "epoch": 4.508166969147005,
+      "grad_norm": 1.5244114398956299,
+      "learning_rate": 0.0002,
+      "loss": 0.5868,
+      "step": 6210
+    },
+    {
+      "epoch": 4.5154264972776765,
+      "grad_norm": 0.8164850473403931,
+      "learning_rate": 0.0002,
+      "loss": 0.4642,
+      "step": 6220
+    },
+    {
+      "epoch": 4.5226860254083485,
+      "grad_norm": 1.2904746532440186,
+      "learning_rate": 0.0002,
+      "loss": 0.522,
+      "step": 6230
+    },
+    {
+      "epoch": 4.52994555353902,
+      "grad_norm": 0.7987732887268066,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 6240
+    },
+    {
+      "epoch": 4.537205081669692,
+      "grad_norm": 0.831040620803833,
+      "learning_rate": 0.0002,
+      "loss": 0.4615,
+      "step": 6250
+    },
+    {
+      "epoch": 4.544464609800363,
+      "grad_norm": 0.9545485973358154,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 6260
+    },
+    {
+      "epoch": 4.551724137931035,
+      "grad_norm": 0.9291793704032898,
+      "learning_rate": 0.0002,
+      "loss": 0.5515,
+      "step": 6270
+    },
+    {
+      "epoch": 4.558983666061706,
+      "grad_norm": 0.8977208733558655,
+      "learning_rate": 0.0002,
+      "loss": 0.4535,
+      "step": 6280
+    },
+    {
+      "epoch": 4.566243194192378,
+      "grad_norm": 1.1768537759780884,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 6290
+    },
+    {
+      "epoch": 4.573502722323049,
+      "grad_norm": 1.0688952207565308,
+      "learning_rate": 0.0002,
+      "loss": 0.5925,
+      "step": 6300
+    },
+    {
+      "epoch": 4.580762250453721,
+      "grad_norm": 0.8800966739654541,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 6310
+    },
+    {
+      "epoch": 4.588021778584392,
+      "grad_norm": 1.0911834239959717,
+      "learning_rate": 0.0002,
+      "loss": 0.6106,
+      "step": 6320
+    },
+    {
+      "epoch": 4.595281306715064,
+      "grad_norm": 1.1420872211456299,
+      "learning_rate": 0.0002,
+      "loss": 0.5109,
+      "step": 6330
+    },
+    {
+      "epoch": 4.602540834845735,
+      "grad_norm": 1.0215224027633667,
+      "learning_rate": 0.0002,
+      "loss": 0.5147,
+      "step": 6340
+    },
+    {
+      "epoch": 4.609800362976406,
+      "grad_norm": 0.9685489535331726,
+      "learning_rate": 0.0002,
+      "loss": 0.592,
+      "step": 6350
+    },
+    {
+      "epoch": 4.617059891107078,
+      "grad_norm": 1.12773597240448,
+      "learning_rate": 0.0002,
+      "loss": 0.5775,
+      "step": 6360
+    },
+    {
+      "epoch": 4.624319419237749,
+      "grad_norm": 1.0663973093032837,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 6370
+    },
+    {
+      "epoch": 4.631578947368421,
+      "grad_norm": 1.1707262992858887,
+      "learning_rate": 0.0002,
+      "loss": 0.512,
+      "step": 6380
+    },
+    {
+      "epoch": 4.638838475499092,
+      "grad_norm": 1.0672980546951294,
+      "learning_rate": 0.0002,
+      "loss": 0.5497,
+      "step": 6390
+    },
+    {
+      "epoch": 4.646098003629764,
+      "grad_norm": 1.1464333534240723,
+      "learning_rate": 0.0002,
+      "loss": 0.5699,
+      "step": 6400
+    },
+    {
+      "epoch": 4.653357531760435,
+      "grad_norm": 1.070230484008789,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 6410
+    },
+    {
+      "epoch": 4.660617059891107,
+      "grad_norm": 0.9673764109611511,
+      "learning_rate": 0.0002,
+      "loss": 0.5013,
+      "step": 6420
+    },
+    {
+      "epoch": 4.6678765880217785,
+      "grad_norm": 1.0189043283462524,
+      "learning_rate": 0.0002,
+      "loss": 0.5901,
+      "step": 6430
+    },
+    {
+      "epoch": 4.67513611615245,
+      "grad_norm": 1.185896396636963,
+      "learning_rate": 0.0002,
+      "loss": 0.5193,
+      "step": 6440
+    },
+    {
+      "epoch": 4.682395644283122,
+      "grad_norm": 1.0682812929153442,
+      "learning_rate": 0.0002,
+      "loss": 0.5318,
+      "step": 6450
+    },
+    {
+      "epoch": 4.689655172413794,
+      "grad_norm": 1.3586071729660034,
+      "learning_rate": 0.0002,
+      "loss": 0.5773,
+      "step": 6460
+    },
+    {
+      "epoch": 4.696914700544465,
+      "grad_norm": 0.6561792492866516,
+      "learning_rate": 0.0002,
+      "loss": 0.5482,
+      "step": 6470
+    },
+    {
+      "epoch": 4.704174228675136,
+      "grad_norm": 1.1394113302230835,
+      "learning_rate": 0.0002,
+      "loss": 0.5711,
+      "step": 6480
+    },
+    {
+      "epoch": 4.711433756805808,
+      "grad_norm": 0.9683151245117188,
+      "learning_rate": 0.0002,
+      "loss": 0.5325,
+      "step": 6490
+    },
+    {
+      "epoch": 4.718693284936479,
+      "grad_norm": 1.0247553586959839,
+      "learning_rate": 0.0002,
+      "loss": 0.5139,
+      "step": 6500
+    },
+    {
+      "epoch": 4.725952813067151,
+      "grad_norm": 0.8046169281005859,
+      "learning_rate": 0.0002,
+      "loss": 0.5794,
+      "step": 6510
+    },
+    {
+      "epoch": 4.733212341197822,
+      "grad_norm": 1.0710240602493286,
+      "learning_rate": 0.0002,
+      "loss": 0.5471,
+      "step": 6520
+    },
+    {
+      "epoch": 4.740471869328494,
+      "grad_norm": 0.9438924193382263,
+      "learning_rate": 0.0002,
+      "loss": 0.4805,
+      "step": 6530
+    },
+    {
+      "epoch": 4.747731397459165,
+      "grad_norm": 0.869162380695343,
+      "learning_rate": 0.0002,
+      "loss": 0.5404,
+      "step": 6540
+    },
+    {
+      "epoch": 4.754990925589837,
+      "grad_norm": 0.9776787161827087,
+      "learning_rate": 0.0002,
+      "loss": 0.6379,
+      "step": 6550
+    },
+    {
+      "epoch": 4.762250453720508,
+      "grad_norm": 1.1990505456924438,
+      "learning_rate": 0.0002,
+      "loss": 0.5288,
+      "step": 6560
+    },
+    {
+      "epoch": 4.769509981851179,
+      "grad_norm": 1.0582209825515747,
+      "learning_rate": 0.0002,
+      "loss": 0.5539,
+      "step": 6570
+    },
+    {
+      "epoch": 4.776769509981851,
+      "grad_norm": 0.9966367483139038,
+      "learning_rate": 0.0002,
+      "loss": 0.489,
+      "step": 6580
+    },
+    {
+      "epoch": 4.784029038112522,
+      "grad_norm": 0.9130612015724182,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 6590
+    },
+    {
+      "epoch": 4.791288566243194,
+      "grad_norm": 1.0950500965118408,
+      "learning_rate": 0.0002,
+      "loss": 0.5864,
+      "step": 6600
+    },
+    {
+      "epoch": 4.798548094373865,
+      "grad_norm": 1.108681321144104,
+      "learning_rate": 0.0002,
+      "loss": 0.5266,
+      "step": 6610
+    },
+    {
+      "epoch": 4.805807622504537,
+      "grad_norm": 1.1873763799667358,
+      "learning_rate": 0.0002,
+      "loss": 0.5875,
+      "step": 6620
+    },
+    {
+      "epoch": 4.8130671506352085,
+      "grad_norm": 1.305367112159729,
+      "learning_rate": 0.0002,
+      "loss": 0.5736,
+      "step": 6630
+    },
+    {
+      "epoch": 4.8203266787658805,
+      "grad_norm": 1.2801482677459717,
+      "learning_rate": 0.0002,
+      "loss": 0.5636,
+      "step": 6640
+    },
+    {
+      "epoch": 4.827586206896552,
+      "grad_norm": 1.26764976978302,
+      "learning_rate": 0.0002,
+      "loss": 0.582,
+      "step": 6650
+    },
+    {
+      "epoch": 4.834845735027224,
+      "grad_norm": 1.0018208026885986,
+      "learning_rate": 0.0002,
+      "loss": 0.5259,
+      "step": 6660
+    },
+    {
+      "epoch": 4.842105263157895,
+      "grad_norm": 1.2326326370239258,
+      "learning_rate": 0.0002,
+      "loss": 0.548,
+      "step": 6670
+    },
+    {
+      "epoch": 4.849364791288567,
+      "grad_norm": 0.9707282781600952,
+      "learning_rate": 0.0002,
+      "loss": 0.5933,
+      "step": 6680
+    },
+    {
+      "epoch": 4.856624319419238,
+      "grad_norm": 1.2772048711776733,
+      "learning_rate": 0.0002,
+      "loss": 0.5612,
+      "step": 6690
+    },
+    {
+      "epoch": 4.863883847549909,
+      "grad_norm": 2.6652262210845947,
+      "learning_rate": 0.0002,
+      "loss": 0.5346,
+      "step": 6700
+    },
+    {
+      "epoch": 4.871143375680581,
+      "grad_norm": 1.215828537940979,
+      "learning_rate": 0.0002,
+      "loss": 0.5428,
+      "step": 6710
+    },
+    {
+      "epoch": 4.878402903811252,
+      "grad_norm": 1.3704510927200317,
+      "learning_rate": 0.0002,
+      "loss": 0.6571,
+      "step": 6720
+    },
+    {
+      "epoch": 4.885662431941924,
+      "grad_norm": 0.7781757116317749,
+      "learning_rate": 0.0002,
+      "loss": 0.4963,
+      "step": 6730
+    },
+    {
+      "epoch": 4.892921960072595,
+      "grad_norm": 1.1883646249771118,
+      "learning_rate": 0.0002,
+      "loss": 0.5989,
+      "step": 6740
+    },
+    {
+      "epoch": 4.900181488203267,
+      "grad_norm": 0.9216066002845764,
+      "learning_rate": 0.0002,
+      "loss": 0.6067,
+      "step": 6750
+    },
+    {
+      "epoch": 4.907441016333938,
+      "grad_norm": 1.0558464527130127,
+      "learning_rate": 0.0002,
+      "loss": 0.5085,
+      "step": 6760
+    },
+    {
+      "epoch": 4.91470054446461,
+      "grad_norm": 1.032656192779541,
+      "learning_rate": 0.0002,
+      "loss": 0.5216,
+      "step": 6770
+    },
+    {
+      "epoch": 4.921960072595281,
+      "grad_norm": 1.1261441707611084,
+      "learning_rate": 0.0002,
+      "loss": 0.5426,
+      "step": 6780
+    },
+    {
+      "epoch": 4.929219600725952,
+      "grad_norm": 1.2178640365600586,
+      "learning_rate": 0.0002,
+      "loss": 0.5295,
+      "step": 6790
+    },
+    {
+      "epoch": 4.936479128856624,
+      "grad_norm": 1.5369361639022827,
+      "learning_rate": 0.0002,
+      "loss": 0.5476,
+      "step": 6800
+    },
+    {
+      "epoch": 4.943738656987296,
+      "grad_norm": 1.1188377141952515,
+      "learning_rate": 0.0002,
+      "loss": 0.5358,
+      "step": 6810
+    },
+    {
+      "epoch": 4.950998185117967,
+      "grad_norm": 1.2506113052368164,
+      "learning_rate": 0.0002,
+      "loss": 0.5483,
+      "step": 6820
+    },
+    {
+      "epoch": 4.9582577132486385,
+      "grad_norm": 0.8776047825813293,
+      "learning_rate": 0.0002,
+      "loss": 0.567,
+      "step": 6830
+    },
+    {
+      "epoch": 4.9655172413793105,
+      "grad_norm": 0.9700555205345154,
+      "learning_rate": 0.0002,
+      "loss": 0.5764,
+      "step": 6840
+    },
+    {
+      "epoch": 4.972776769509982,
+      "grad_norm": 1.2713534832000732,
+      "learning_rate": 0.0002,
+      "loss": 0.5396,
+      "step": 6850
+    },
+    {
+      "epoch": 4.980036297640654,
+      "grad_norm": 0.9855955243110657,
+      "learning_rate": 0.0002,
+      "loss": 0.5451,
+      "step": 6860
+    },
+    {
+      "epoch": 4.987295825771325,
+      "grad_norm": 0.8734853863716125,
+      "learning_rate": 0.0002,
+      "loss": 0.5884,
+      "step": 6870
+    },
+    {
+      "epoch": 4.994555353901997,
+      "grad_norm": 0.8065403699874878,
+      "learning_rate": 0.0002,
+      "loss": 0.5189,
+      "step": 6880
+    },
+    {
+      "epoch": 4.999637023593467,
+      "eval_loss": 1.3302682638168335,
+      "eval_runtime": 46.2496,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 6887
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11016,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.53767798013952e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..140bceabdec44fc473842cb9ace4a62a4dc9fe89
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-6887/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4abec5480aff960b5a7b03a48be1cdb1ae29ba8792469c451f5c2f269e4be93b
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..81a0479b82ca0455cfa8893b910606301b2664b2
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f824b3a2fc6297fba39203f80f82c2375a85fa203c4cb7c7bd8aa052ebaf1169
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..eb7e77c552c41d69148dfb3f1cc9598c98a50c9a
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:742dd61e226f0f53ee486c61d21a0b86f73e61db5c329460f2e81b9d4a3e9949
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..78f4ce556e5485ee856bd0d1b9988dbbed1a951c
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f815e186720e9ed960c1663c5ffe551ecf68125067fa5c89c0feb06e6c7a679
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..0cd3ff5a3b7962b8758e5d2affaa1b77d9549997
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eacf0861f52de0c0eef50849d3c288c99c87afb2d949a9bcc3e67628c7cd429f
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..8ed0ed427b6385b871dfa427a1ca09547c14f461
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/trainer_state.json
@@ -0,0 +1,5863 @@
+{
+  "best_metric": 1.1086540222167969,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 8265,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007259528130671506,
+      "grad_norm": 0.46010470390319824,
+      "learning_rate": 0.0002,
+      "loss": 2.4936,
+      "step": 10
+    },
+    {
+      "epoch": 0.014519056261343012,
+      "grad_norm": 0.5103179216384888,
+      "learning_rate": 0.0002,
+      "loss": 1.8442,
+      "step": 20
+    },
+    {
+      "epoch": 0.021778584392014518,
+      "grad_norm": 0.6282716989517212,
+      "learning_rate": 0.0002,
+      "loss": 1.559,
+      "step": 30
+    },
+    {
+      "epoch": 0.029038112522686024,
+      "grad_norm": 1.2480497360229492,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 40
+    },
+    {
+      "epoch": 0.036297640653357534,
+      "grad_norm": 0.4114855229854584,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 50
+    },
+    {
+      "epoch": 0.043557168784029036,
+      "grad_norm": 0.49482840299606323,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 60
+    },
+    {
+      "epoch": 0.050816696914700546,
+      "grad_norm": 0.4536272883415222,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 70
+    },
+    {
+      "epoch": 0.05807622504537205,
+      "grad_norm": 0.32328274846076965,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 80
+    },
+    {
+      "epoch": 0.06533575317604355,
+      "grad_norm": 0.40990468859672546,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 90
+    },
+    {
+      "epoch": 0.07259528130671507,
+      "grad_norm": 0.37273502349853516,
+      "learning_rate": 0.0002,
+      "loss": 1.2216,
+      "step": 100
+    },
+    {
+      "epoch": 0.07985480943738657,
+      "grad_norm": 0.3903871476650238,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 110
+    },
+    {
+      "epoch": 0.08711433756805807,
+      "grad_norm": 0.3272787928581238,
+      "learning_rate": 0.0002,
+      "loss": 1.3718,
+      "step": 120
+    },
+    {
+      "epoch": 0.09437386569872959,
+      "grad_norm": 0.3622824251651764,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 130
+    },
+    {
+      "epoch": 0.10163339382940109,
+      "grad_norm": 0.3503916561603546,
+      "learning_rate": 0.0002,
+      "loss": 1.2966,
+      "step": 140
+    },
+    {
+      "epoch": 0.1088929219600726,
+      "grad_norm": 0.32787832617759705,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 150
+    },
+    {
+      "epoch": 0.1161524500907441,
+      "grad_norm": 0.5822657942771912,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 160
+    },
+    {
+      "epoch": 0.12341197822141561,
+      "grad_norm": 0.28028249740600586,
+      "learning_rate": 0.0002,
+      "loss": 1.2091,
+      "step": 170
+    },
+    {
+      "epoch": 0.1306715063520871,
+      "grad_norm": 0.33602750301361084,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 180
+    },
+    {
+      "epoch": 0.13793103448275862,
+      "grad_norm": 0.29106274247169495,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 190
+    },
+    {
+      "epoch": 0.14519056261343014,
+      "grad_norm": 0.38753363490104675,
+      "learning_rate": 0.0002,
+      "loss": 1.3261,
+      "step": 200
+    },
+    {
+      "epoch": 0.15245009074410162,
+      "grad_norm": 0.361009418964386,
+      "learning_rate": 0.0002,
+      "loss": 1.1536,
+      "step": 210
+    },
+    {
+      "epoch": 0.15970961887477314,
+      "grad_norm": 0.6743836402893066,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 220
+    },
+    {
+      "epoch": 0.16696914700544466,
+      "grad_norm": 0.3923613727092743,
+      "learning_rate": 0.0002,
+      "loss": 1.264,
+      "step": 230
+    },
+    {
+      "epoch": 0.17422867513611615,
+      "grad_norm": 0.2809699773788452,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 240
+    },
+    {
+      "epoch": 0.18148820326678766,
+      "grad_norm": 0.3631494641304016,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 250
+    },
+    {
+      "epoch": 0.18874773139745918,
+      "grad_norm": 0.24658171832561493,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 260
+    },
+    {
+      "epoch": 0.19600725952813067,
+      "grad_norm": 0.5780664682388306,
+      "learning_rate": 0.0002,
+      "loss": 1.0065,
+      "step": 270
+    },
+    {
+      "epoch": 0.20326678765880218,
+      "grad_norm": 0.3056720495223999,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 280
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 0.2857084572315216,
+      "learning_rate": 0.0002,
+      "loss": 1.1503,
+      "step": 290
+    },
+    {
+      "epoch": 0.2177858439201452,
+      "grad_norm": 0.6645345687866211,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 300
+    },
+    {
+      "epoch": 0.2250453720508167,
+      "grad_norm": 0.5966078639030457,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 310
+    },
+    {
+      "epoch": 0.2323049001814882,
+      "grad_norm": 0.40937140583992004,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 320
+    },
+    {
+      "epoch": 0.2395644283121597,
+      "grad_norm": 0.5642806887626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 330
+    },
+    {
+      "epoch": 0.24682395644283123,
+      "grad_norm": 0.2750748097896576,
+      "learning_rate": 0.0002,
+      "loss": 1.1703,
+      "step": 340
+    },
+    {
+      "epoch": 0.2540834845735027,
+      "grad_norm": 0.34350234270095825,
+      "learning_rate": 0.0002,
+      "loss": 1.2598,
+      "step": 350
+    },
+    {
+      "epoch": 0.2613430127041742,
+      "grad_norm": 0.6767239570617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1942,
+      "step": 360
+    },
+    {
+      "epoch": 0.26860254083484575,
+      "grad_norm": 0.31006959080696106,
+      "learning_rate": 0.0002,
+      "loss": 1.1436,
+      "step": 370
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "grad_norm": 0.3825474679470062,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 380
+    },
+    {
+      "epoch": 0.2831215970961887,
+      "grad_norm": 0.30241551995277405,
+      "learning_rate": 0.0002,
+      "loss": 1.1523,
+      "step": 390
+    },
+    {
+      "epoch": 0.29038112522686027,
+      "grad_norm": 0.2962397336959839,
+      "learning_rate": 0.0002,
+      "loss": 1.1298,
+      "step": 400
+    },
+    {
+      "epoch": 0.29764065335753176,
+      "grad_norm": 0.2600369155406952,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 410
+    },
+    {
+      "epoch": 0.30490018148820325,
+      "grad_norm": 0.3675060272216797,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 420
+    },
+    {
+      "epoch": 0.3121597096188748,
+      "grad_norm": 0.3429498076438904,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 430
+    },
+    {
+      "epoch": 0.3194192377495463,
+      "grad_norm": 0.34311825037002563,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 440
+    },
+    {
+      "epoch": 0.32667876588021777,
+      "grad_norm": 0.37872210144996643,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 450
+    },
+    {
+      "epoch": 0.3339382940108893,
+      "grad_norm": 0.33271121978759766,
+      "learning_rate": 0.0002,
+      "loss": 1.1365,
+      "step": 460
+    },
+    {
+      "epoch": 0.3411978221415608,
+      "grad_norm": 0.34605276584625244,
+      "learning_rate": 0.0002,
+      "loss": 1.1572,
+      "step": 470
+    },
+    {
+      "epoch": 0.3484573502722323,
+      "grad_norm": 0.41050562262535095,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 480
+    },
+    {
+      "epoch": 0.35571687840290384,
+      "grad_norm": 0.2066836953163147,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 490
+    },
+    {
+      "epoch": 0.3629764065335753,
+      "grad_norm": 0.2859014868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.2072,
+      "step": 500
+    },
+    {
+      "epoch": 0.3702359346642468,
+      "grad_norm": 0.28763777017593384,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 510
+    },
+    {
+      "epoch": 0.37749546279491836,
+      "grad_norm": 0.2730471193790436,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 520
+    },
+    {
+      "epoch": 0.38475499092558985,
+      "grad_norm": 0.3968936800956726,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 530
+    },
+    {
+      "epoch": 0.39201451905626133,
+      "grad_norm": 0.3624701201915741,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 540
+    },
+    {
+      "epoch": 0.3992740471869328,
+      "grad_norm": 0.3303608298301697,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 550
+    },
+    {
+      "epoch": 0.40653357531760437,
+      "grad_norm": 0.33507466316223145,
+      "learning_rate": 0.0002,
+      "loss": 1.203,
+      "step": 560
+    },
+    {
+      "epoch": 0.41379310344827586,
+      "grad_norm": 0.3297670781612396,
+      "learning_rate": 0.0002,
+      "loss": 1.2294,
+      "step": 570
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.32334890961647034,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 580
+    },
+    {
+      "epoch": 0.4283121597096189,
+      "grad_norm": 0.30281195044517517,
+      "learning_rate": 0.0002,
+      "loss": 1.112,
+      "step": 590
+    },
+    {
+      "epoch": 0.4355716878402904,
+      "grad_norm": 0.5900027751922607,
+      "learning_rate": 0.0002,
+      "loss": 1.1352,
+      "step": 600
+    },
+    {
+      "epoch": 0.44283121597096187,
+      "grad_norm": 0.28437477350234985,
+      "learning_rate": 0.0002,
+      "loss": 1.1575,
+      "step": 610
+    },
+    {
+      "epoch": 0.4500907441016334,
+      "grad_norm": 0.39601704478263855,
+      "learning_rate": 0.0002,
+      "loss": 1.2009,
+      "step": 620
+    },
+    {
+      "epoch": 0.4573502722323049,
+      "grad_norm": 0.41971510648727417,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 630
+    },
+    {
+      "epoch": 0.4646098003629764,
+      "grad_norm": 0.33814409375190735,
+      "learning_rate": 0.0002,
+      "loss": 1.236,
+      "step": 640
+    },
+    {
+      "epoch": 0.47186932849364793,
+      "grad_norm": 0.575718104839325,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 650
+    },
+    {
+      "epoch": 0.4791288566243194,
+      "grad_norm": 0.37927401065826416,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 660
+    },
+    {
+      "epoch": 0.4863883847549909,
+      "grad_norm": 0.3224332630634308,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 670
+    },
+    {
+      "epoch": 0.49364791288566245,
+      "grad_norm": 0.32683515548706055,
+      "learning_rate": 0.0002,
+      "loss": 1.1354,
+      "step": 680
+    },
+    {
+      "epoch": 0.5009074410163339,
+      "grad_norm": 0.4316163659095764,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 690
+    },
+    {
+      "epoch": 0.5081669691470054,
+      "grad_norm": 0.342602401971817,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 700
+    },
+    {
+      "epoch": 0.515426497277677,
+      "grad_norm": 0.2794898748397827,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 710
+    },
+    {
+      "epoch": 0.5226860254083484,
+      "grad_norm": 0.3322339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.2395,
+      "step": 720
+    },
+    {
+      "epoch": 0.52994555353902,
+      "grad_norm": 0.3088509142398834,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 730
+    },
+    {
+      "epoch": 0.5372050816696915,
+      "grad_norm": 0.24444378912448883,
+      "learning_rate": 0.0002,
+      "loss": 1.1495,
+      "step": 740
+    },
+    {
+      "epoch": 0.5444646098003629,
+      "grad_norm": 0.3483171761035919,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 750
+    },
+    {
+      "epoch": 0.5517241379310345,
+      "grad_norm": 0.3425690531730652,
+      "learning_rate": 0.0002,
+      "loss": 1.2248,
+      "step": 760
+    },
+    {
+      "epoch": 0.558983666061706,
+      "grad_norm": 0.31841927766799927,
+      "learning_rate": 0.0002,
+      "loss": 1.1127,
+      "step": 770
+    },
+    {
+      "epoch": 0.5662431941923775,
+      "grad_norm": 0.39423868060112,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 780
+    },
+    {
+      "epoch": 0.573502722323049,
+      "grad_norm": 0.30328479409217834,
+      "learning_rate": 0.0002,
+      "loss": 1.0583,
+      "step": 790
+    },
+    {
+      "epoch": 0.5807622504537205,
+      "grad_norm": 0.24475938081741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0718,
+      "step": 800
+    },
+    {
+      "epoch": 0.588021778584392,
+      "grad_norm": 0.37132805585861206,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 810
+    },
+    {
+      "epoch": 0.5952813067150635,
+      "grad_norm": 0.32195979356765747,
+      "learning_rate": 0.0002,
+      "loss": 1.1544,
+      "step": 820
+    },
+    {
+      "epoch": 0.6025408348457351,
+      "grad_norm": 0.2848738729953766,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 830
+    },
+    {
+      "epoch": 0.6098003629764065,
+      "grad_norm": 0.28015264868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.0937,
+      "step": 840
+    },
+    {
+      "epoch": 0.617059891107078,
+      "grad_norm": 0.37796008586883545,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 850
+    },
+    {
+      "epoch": 0.6243194192377496,
+      "grad_norm": 0.39311841130256653,
+      "learning_rate": 0.0002,
+      "loss": 1.1455,
+      "step": 860
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.2761685252189636,
+      "learning_rate": 0.0002,
+      "loss": 1.1066,
+      "step": 870
+    },
+    {
+      "epoch": 0.6388384754990926,
+      "grad_norm": 0.3826720118522644,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 880
+    },
+    {
+      "epoch": 0.6460980036297641,
+      "grad_norm": 0.30076679587364197,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 890
+    },
+    {
+      "epoch": 0.6533575317604355,
+      "grad_norm": 0.21997687220573425,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 900
+    },
+    {
+      "epoch": 0.6606170598911071,
+      "grad_norm": 0.32593777775764465,
+      "learning_rate": 0.0002,
+      "loss": 1.1603,
+      "step": 910
+    },
+    {
+      "epoch": 0.6678765880217786,
+      "grad_norm": 0.30347898602485657,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 920
+    },
+    {
+      "epoch": 0.6751361161524501,
+      "grad_norm": 0.44173774123191833,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 930
+    },
+    {
+      "epoch": 0.6823956442831216,
+      "grad_norm": 0.2507467269897461,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 940
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.29463833570480347,
+      "learning_rate": 0.0002,
+      "loss": 1.3001,
+      "step": 950
+    },
+    {
+      "epoch": 0.6969147005444646,
+      "grad_norm": 0.9363154172897339,
+      "learning_rate": 0.0002,
+      "loss": 1.1651,
+      "step": 960
+    },
+    {
+      "epoch": 0.7041742286751361,
+      "grad_norm": 0.3236212134361267,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 970
+    },
+    {
+      "epoch": 0.7114337568058077,
+      "grad_norm": 0.3123254179954529,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 980
+    },
+    {
+      "epoch": 0.7186932849364791,
+      "grad_norm": 0.3395805060863495,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 990
+    },
+    {
+      "epoch": 0.7259528130671506,
+      "grad_norm": 0.3240964412689209,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7332123411978222,
+      "grad_norm": 0.31902948021888733,
+      "learning_rate": 0.0002,
+      "loss": 1.0313,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7404718693284936,
+      "grad_norm": 0.4848408102989197,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7477313974591652,
+      "grad_norm": 0.33006033301353455,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7549909255898367,
+      "grad_norm": 0.2928730547428131,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7622504537205081,
+      "grad_norm": 0.3529164791107178,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7695099818511797,
+      "grad_norm": 0.2736213803291321,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7767695099818511,
+      "grad_norm": 0.7200686931610107,
+      "learning_rate": 0.0002,
+      "loss": 1.1169,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7840290381125227,
+      "grad_norm": 0.33396708965301514,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7912885662431942,
+      "grad_norm": 1.5760449171066284,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7985480943738656,
+      "grad_norm": 0.28138381242752075,
+      "learning_rate": 0.0002,
+      "loss": 1.1126,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8058076225045372,
+      "grad_norm": 0.2597472369670868,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8130671506352087,
+      "grad_norm": 0.3305445611476898,
+      "learning_rate": 0.0002,
+      "loss": 1.1177,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8203266787658802,
+      "grad_norm": 0.3934599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.0849,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8275862068965517,
+      "grad_norm": 0.3472191393375397,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8348457350272233,
+      "grad_norm": 0.2857365906238556,
+      "learning_rate": 0.0002,
+      "loss": 1.265,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.3207702934741974,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8493647912885662,
+      "grad_norm": 0.3176484704017639,
+      "learning_rate": 0.0002,
+      "loss": 1.1669,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8566243194192378,
+      "grad_norm": 0.40685558319091797,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8638838475499092,
+      "grad_norm": 0.31125199794769287,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8711433756805808,
+      "grad_norm": 0.7361181378364563,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8784029038112523,
+      "grad_norm": 0.33699527382850647,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8856624319419237,
+      "grad_norm": 0.3315220773220062,
+      "learning_rate": 0.0002,
+      "loss": 1.11,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8929219600725953,
+      "grad_norm": 0.6256054043769836,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9001814882032668,
+      "grad_norm": 0.3692137897014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1738,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9074410163339383,
+      "grad_norm": 0.3538484573364258,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9147005444646098,
+      "grad_norm": 0.27863040566444397,
+      "learning_rate": 0.0002,
+      "loss": 1.0878,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9219600725952813,
+      "grad_norm": 0.3322528302669525,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9292196007259528,
+      "grad_norm": 0.7553173303604126,
+      "learning_rate": 0.0002,
+      "loss": 1.0158,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9364791288566243,
+      "grad_norm": 0.4856191575527191,
+      "learning_rate": 0.0002,
+      "loss": 1.198,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9437386569872959,
+      "grad_norm": 0.3668074905872345,
+      "learning_rate": 0.0002,
+      "loss": 1.1678,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9509981851179673,
+      "grad_norm": 0.29851067066192627,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9582577132486388,
+      "grad_norm": 0.276664674282074,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9655172413793104,
+      "grad_norm": 0.2941018342971802,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9727767695099818,
+      "grad_norm": 0.3505859076976776,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9800362976406534,
+      "grad_norm": 0.3067687451839447,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9872958257713249,
+      "grad_norm": 0.27151066064834595,
+      "learning_rate": 0.0002,
+      "loss": 1.1511,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9945553539019963,
+      "grad_norm": 0.36370083689689636,
+      "learning_rate": 0.0002,
+      "loss": 1.1919,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9996370235934664,
+      "eval_loss": 1.1381088495254517,
+      "eval_runtime": 96.0848,
+      "eval_samples_per_second": 4.538,
+      "eval_steps_per_second": 0.572,
+      "step": 1377
+    },
+    {
+      "epoch": 1.0018148820326678,
+      "grad_norm": 0.27980583906173706,
+      "learning_rate": 0.0002,
+      "loss": 1.0082,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0090744101633393,
+      "grad_norm": 0.26713913679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0163339382940109,
+      "grad_norm": 0.3089541494846344,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0235934664246824,
+      "grad_norm": 0.4188242256641388,
+      "learning_rate": 0.0002,
+      "loss": 0.9649,
+      "step": 1410
+    },
+    {
+      "epoch": 1.030852994555354,
+      "grad_norm": 0.5246463418006897,
+      "learning_rate": 0.0002,
+      "loss": 0.9821,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0381125226860255,
+      "grad_norm": 0.2728777825832367,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0453720508166968,
+      "grad_norm": 0.38167616724967957,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.4439380168914795,
+      "learning_rate": 0.0002,
+      "loss": 1.0229,
+      "step": 1450
+    },
+    {
+      "epoch": 1.05989110707804,
+      "grad_norm": 0.30954182147979736,
+      "learning_rate": 0.0002,
+      "loss": 1.0451,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0671506352087115,
+      "grad_norm": 0.4022280275821686,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 1470
+    },
+    {
+      "epoch": 1.074410163339383,
+      "grad_norm": 0.7390811443328857,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0816696914700545,
+      "grad_norm": 0.3885486423969269,
+      "learning_rate": 0.0002,
+      "loss": 1.1378,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0889292196007259,
+      "grad_norm": 0.5275560617446899,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0961887477313974,
+      "grad_norm": 0.35112282633781433,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1510
+    },
+    {
+      "epoch": 1.103448275862069,
+      "grad_norm": 0.33714351058006287,
+      "learning_rate": 0.0002,
+      "loss": 1.0331,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1107078039927405,
+      "grad_norm": 0.31221693754196167,
+      "learning_rate": 0.0002,
+      "loss": 0.8723,
+      "step": 1530
+    },
+    {
+      "epoch": 1.117967332123412,
+      "grad_norm": 0.27549654245376587,
+      "learning_rate": 0.0002,
+      "loss": 1.0701,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1252268602540836,
+      "grad_norm": 0.8465521335601807,
+      "learning_rate": 0.0002,
+      "loss": 1.0114,
+      "step": 1550
+    },
+    {
+      "epoch": 1.132486388384755,
+      "grad_norm": 0.36125949025154114,
+      "learning_rate": 0.0002,
+      "loss": 0.9643,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1397459165154264,
+      "grad_norm": 0.37420371174812317,
+      "learning_rate": 0.0002,
+      "loss": 1.0744,
+      "step": 1570
+    },
+    {
+      "epoch": 1.147005444646098,
+      "grad_norm": 0.3294760584831238,
+      "learning_rate": 0.0002,
+      "loss": 1.165,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1542649727767695,
+      "grad_norm": 0.3881238102912903,
+      "learning_rate": 0.0002,
+      "loss": 0.9207,
+      "step": 1590
+    },
+    {
+      "epoch": 1.161524500907441,
+      "grad_norm": 0.4766491651535034,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1687840290381124,
+      "grad_norm": 0.395530104637146,
+      "learning_rate": 0.0002,
+      "loss": 1.0092,
+      "step": 1610
+    },
+    {
+      "epoch": 1.176043557168784,
+      "grad_norm": 0.3297106623649597,
+      "learning_rate": 0.0002,
+      "loss": 0.9895,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1833030852994555,
+      "grad_norm": 0.39528271555900574,
+      "learning_rate": 0.0002,
+      "loss": 1.1121,
+      "step": 1630
+    },
+    {
+      "epoch": 1.190562613430127,
+      "grad_norm": 0.3370221257209778,
+      "learning_rate": 0.0002,
+      "loss": 0.9202,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1978221415607986,
+      "grad_norm": 0.31922030448913574,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2050816696914701,
+      "grad_norm": 0.6142027378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2123411978221417,
+      "grad_norm": 0.44769710302352905,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 1670
+    },
+    {
+      "epoch": 1.219600725952813,
+      "grad_norm": 0.41233646869659424,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2268602540834845,
+      "grad_norm": 0.2928866147994995,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1690
+    },
+    {
+      "epoch": 1.234119782214156,
+      "grad_norm": 0.36913734674453735,
+      "learning_rate": 0.0002,
+      "loss": 0.9074,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2413793103448276,
+      "grad_norm": 0.5281891226768494,
+      "learning_rate": 0.0002,
+      "loss": 1.0847,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2486388384754992,
+      "grad_norm": 0.3374697268009186,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2558983666061705,
+      "grad_norm": 0.3802020847797394,
+      "learning_rate": 0.0002,
+      "loss": 0.9826,
+      "step": 1730
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.38048651814460754,
+      "learning_rate": 0.0002,
+      "loss": 1.0191,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2704174228675136,
+      "grad_norm": 0.6676169633865356,
+      "learning_rate": 0.0002,
+      "loss": 1.0142,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2776769509981851,
+      "grad_norm": 0.4075961410999298,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2849364791288567,
+      "grad_norm": 0.4374721348285675,
+      "learning_rate": 0.0002,
+      "loss": 1.0301,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2921960072595282,
+      "grad_norm": 0.4638824164867401,
+      "learning_rate": 0.0002,
+      "loss": 1.0601,
+      "step": 1780
+    },
+    {
+      "epoch": 1.2994555353901998,
+      "grad_norm": 0.38631564378738403,
+      "learning_rate": 0.0002,
+      "loss": 0.9593,
+      "step": 1790
+    },
+    {
+      "epoch": 1.306715063520871,
+      "grad_norm": 0.35873809456825256,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3139745916515426,
+      "grad_norm": 0.33602237701416016,
+      "learning_rate": 0.0002,
+      "loss": 0.9835,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3212341197822142,
+      "grad_norm": 0.46696463227272034,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3284936479128857,
+      "grad_norm": 0.368958979845047,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 1830
+    },
+    {
+      "epoch": 1.335753176043557,
+      "grad_norm": 0.3160957396030426,
+      "learning_rate": 0.0002,
+      "loss": 1.0456,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3430127041742286,
+      "grad_norm": 0.4511511027812958,
+      "learning_rate": 0.0002,
+      "loss": 1.0264,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3502722323049001,
+      "grad_norm": 0.6769845485687256,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3575317604355717,
+      "grad_norm": 0.3749309480190277,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3647912885662432,
+      "grad_norm": 0.553798496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.0223,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3720508166969148,
+      "grad_norm": 0.3538985550403595,
+      "learning_rate": 0.0002,
+      "loss": 0.9634,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.5501534938812256,
+      "learning_rate": 0.0002,
+      "loss": 1.0275,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3865698729582578,
+      "grad_norm": 0.4432051181793213,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3938294010889292,
+      "grad_norm": 0.41755786538124084,
+      "learning_rate": 0.0002,
+      "loss": 0.9082,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4010889292196007,
+      "grad_norm": 0.5626114010810852,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4083484573502723,
+      "grad_norm": 0.44038185477256775,
+      "learning_rate": 0.0002,
+      "loss": 1.0687,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4156079854809438,
+      "grad_norm": 0.3397001624107361,
+      "learning_rate": 0.0002,
+      "loss": 0.9454,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4228675136116151,
+      "grad_norm": 0.4325368106365204,
+      "learning_rate": 0.0002,
+      "loss": 0.943,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4301270417422867,
+      "grad_norm": 0.3900907039642334,
+      "learning_rate": 0.0002,
+      "loss": 1.0488,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4373865698729582,
+      "grad_norm": 0.369612455368042,
+      "learning_rate": 0.0002,
+      "loss": 0.9699,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4446460980036298,
+      "grad_norm": 0.4389338791370392,
+      "learning_rate": 0.0002,
+      "loss": 1.0609,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4519056261343013,
+      "grad_norm": 1.694450855255127,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4591651542649728,
+      "grad_norm": 0.516957700252533,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4664246823956444,
+      "grad_norm": 0.45515501499176025,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.4153139591217041,
+      "learning_rate": 0.0002,
+      "loss": 1.0188,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4809437386569873,
+      "grad_norm": 0.44353052973747253,
+      "learning_rate": 0.0002,
+      "loss": 1.1521,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4882032667876588,
+      "grad_norm": 0.570554256439209,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4954627949183303,
+      "grad_norm": 0.5742740035057068,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5027223230490017,
+      "grad_norm": 0.5890517830848694,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5099818511796732,
+      "grad_norm": 0.4162650406360626,
+      "learning_rate": 0.0002,
+      "loss": 1.0639,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5172413793103448,
+      "grad_norm": 0.7334717512130737,
+      "learning_rate": 0.0002,
+      "loss": 1.071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5245009074410163,
+      "grad_norm": 0.2963249683380127,
+      "learning_rate": 0.0002,
+      "loss": 0.8957,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5317604355716878,
+      "grad_norm": 0.30676454305648804,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5390199637023594,
+      "grad_norm": 0.35984641313552856,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 2120
+    },
+    {
+      "epoch": 1.546279491833031,
+      "grad_norm": 0.3384549617767334,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5535390199637025,
+      "grad_norm": 0.4725518226623535,
+      "learning_rate": 0.0002,
+      "loss": 0.9874,
+      "step": 2140
+    },
+    {
+      "epoch": 1.560798548094374,
+      "grad_norm": 0.3252685070037842,
+      "learning_rate": 0.0002,
+      "loss": 1.1166,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5680580762250453,
+      "grad_norm": 0.45043081045150757,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5753176043557169,
+      "grad_norm": 0.374208003282547,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5825771324863884,
+      "grad_norm": 0.5118404030799866,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5898366606170597,
+      "grad_norm": 0.482170969247818,
+      "learning_rate": 0.0002,
+      "loss": 1.0062,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5970961887477313,
+      "grad_norm": 0.5337533950805664,
+      "learning_rate": 0.0002,
+      "loss": 1.0872,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6043557168784028,
+      "grad_norm": 0.5195064544677734,
+      "learning_rate": 0.0002,
+      "loss": 1.0405,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6116152450090744,
+      "grad_norm": 0.30807098746299744,
+      "learning_rate": 0.0002,
+      "loss": 1.0454,
+      "step": 2220
+    },
+    {
+      "epoch": 1.618874773139746,
+      "grad_norm": 0.3962925672531128,
+      "learning_rate": 0.0002,
+      "loss": 1.0293,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6261343012704175,
+      "grad_norm": 0.7636962532997131,
+      "learning_rate": 0.0002,
+      "loss": 1.0137,
+      "step": 2240
+    },
+    {
+      "epoch": 1.633393829401089,
+      "grad_norm": 0.32380592823028564,
+      "learning_rate": 0.0002,
+      "loss": 0.991,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6406533575317606,
+      "grad_norm": 0.5767741799354553,
+      "learning_rate": 0.0002,
+      "loss": 0.9471,
+      "step": 2260
+    },
+    {
+      "epoch": 1.647912885662432,
+      "grad_norm": 0.39964812994003296,
+      "learning_rate": 0.0002,
+      "loss": 0.9524,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6551724137931034,
+      "grad_norm": 0.622629702091217,
+      "learning_rate": 0.0002,
+      "loss": 0.9866,
+      "step": 2280
+    },
+    {
+      "epoch": 1.662431941923775,
+      "grad_norm": 0.40202152729034424,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6696914700544465,
+      "grad_norm": 0.4467349052429199,
+      "learning_rate": 0.0002,
+      "loss": 0.9859,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6769509981851178,
+      "grad_norm": 0.5026949048042297,
+      "learning_rate": 0.0002,
+      "loss": 1.0312,
+      "step": 2310
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.43754541873931885,
+      "learning_rate": 0.0002,
+      "loss": 1.0776,
+      "step": 2320
+    },
+    {
+      "epoch": 1.691470054446461,
+      "grad_norm": 0.42869430780410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6987295825771325,
+      "grad_norm": 0.4192679524421692,
+      "learning_rate": 0.0002,
+      "loss": 1.0747,
+      "step": 2340
+    },
+    {
+      "epoch": 1.705989110707804,
+      "grad_norm": 0.3243155777454376,
+      "learning_rate": 0.0002,
+      "loss": 1.0054,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7132486388384756,
+      "grad_norm": 1.0514075756072998,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 2360
+    },
+    {
+      "epoch": 1.720508166969147,
+      "grad_norm": 0.4719122648239136,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7277676950998186,
+      "grad_norm": 0.3846144676208496,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 2380
+    },
+    {
+      "epoch": 1.73502722323049,
+      "grad_norm": 0.7266581058502197,
+      "learning_rate": 0.0002,
+      "loss": 0.9935,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7422867513611615,
+      "grad_norm": 0.6183241605758667,
+      "learning_rate": 0.0002,
+      "loss": 0.9509,
+      "step": 2400
+    },
+    {
+      "epoch": 1.749546279491833,
+      "grad_norm": 0.3658260405063629,
+      "learning_rate": 0.0002,
+      "loss": 0.9713,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7568058076225044,
+      "grad_norm": 0.6036322712898254,
+      "learning_rate": 0.0002,
+      "loss": 1.0033,
+      "step": 2420
+    },
+    {
+      "epoch": 1.764065335753176,
+      "grad_norm": 0.7872936129570007,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7713248638838475,
+      "grad_norm": 0.35946124792099,
+      "learning_rate": 0.0002,
+      "loss": 0.99,
+      "step": 2440
+    },
+    {
+      "epoch": 1.778584392014519,
+      "grad_norm": 0.3740338981151581,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2450
+    },
+    {
+      "epoch": 1.7858439201451906,
+      "grad_norm": 0.6150230169296265,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 2460
+    },
+    {
+      "epoch": 1.793103448275862,
+      "grad_norm": 0.4726075530052185,
+      "learning_rate": 0.0002,
+      "loss": 1.0373,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8003629764065336,
+      "grad_norm": 0.31292253732681274,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8076225045372052,
+      "grad_norm": 0.4463104009628296,
+      "learning_rate": 0.0002,
+      "loss": 1.0205,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8148820326678767,
+      "grad_norm": 0.7848200798034668,
+      "learning_rate": 0.0002,
+      "loss": 1.0771,
+      "step": 2500
+    },
+    {
+      "epoch": 1.822141560798548,
+      "grad_norm": 0.5562082529067993,
+      "learning_rate": 0.0002,
+      "loss": 0.9508,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8294010889292196,
+      "grad_norm": 0.39892756938934326,
+      "learning_rate": 0.0002,
+      "loss": 0.9343,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8366606170598911,
+      "grad_norm": 1.2923320531845093,
+      "learning_rate": 0.0002,
+      "loss": 1.0489,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8439201451905625,
+      "grad_norm": 0.6316490769386292,
+      "learning_rate": 0.0002,
+      "loss": 1.1203,
+      "step": 2540
+    },
+    {
+      "epoch": 1.851179673321234,
+      "grad_norm": 0.46100497245788574,
+      "learning_rate": 0.0002,
+      "loss": 0.9131,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8584392014519056,
+      "grad_norm": 0.7902987003326416,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 2560
+    },
+    {
+      "epoch": 1.865698729582577,
+      "grad_norm": 0.4596365690231323,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8729582577132486,
+      "grad_norm": 0.6592172384262085,
+      "learning_rate": 0.0002,
+      "loss": 0.9322,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8802177858439202,
+      "grad_norm": 0.5764662027359009,
+      "learning_rate": 0.0002,
+      "loss": 0.9549,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8874773139745917,
+      "grad_norm": 0.8421637415885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9485,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.5635305047035217,
+      "learning_rate": 0.0002,
+      "loss": 1.012,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9019963702359348,
+      "grad_norm": 0.46231237053871155,
+      "learning_rate": 0.0002,
+      "loss": 0.8907,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9092558983666061,
+      "grad_norm": 0.3944607079029083,
+      "learning_rate": 0.0002,
+      "loss": 0.9543,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9165154264972777,
+      "grad_norm": 0.4753907322883606,
+      "learning_rate": 0.0002,
+      "loss": 0.9964,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9237749546279492,
+      "grad_norm": 0.4151090979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.0217,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9310344827586206,
+      "grad_norm": 0.6793725490570068,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 2660
+    },
+    {
+      "epoch": 1.938294010889292,
+      "grad_norm": 0.339755117893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0401,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9455535390199636,
+      "grad_norm": 0.40789374709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.082,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9528130671506352,
+      "grad_norm": 0.3750005066394806,
+      "learning_rate": 0.0002,
+      "loss": 0.9973,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9600725952813067,
+      "grad_norm": 0.39684441685676575,
+      "learning_rate": 0.0002,
+      "loss": 0.9837,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9673321234119783,
+      "grad_norm": 0.378287672996521,
+      "learning_rate": 0.0002,
+      "loss": 1.1204,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9745916515426498,
+      "grad_norm": 0.3668482005596161,
+      "learning_rate": 0.0002,
+      "loss": 0.9284,
+      "step": 2720
+    },
+    {
+      "epoch": 1.9818511796733214,
+      "grad_norm": 0.49997565150260925,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 2730
+    },
+    {
+      "epoch": 1.989110707803993,
+      "grad_norm": 0.36852124333381653,
+      "learning_rate": 0.0002,
+      "loss": 0.9311,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9963702359346642,
+      "grad_norm": 0.5203380584716797,
+      "learning_rate": 0.0002,
+      "loss": 1.0898,
+      "step": 2750
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1086540222167969,
+      "eval_runtime": 95.6866,
+      "eval_samples_per_second": 4.557,
+      "eval_steps_per_second": 0.575,
+      "step": 2755
+    },
+    {
+      "epoch": 2.0036297640653356,
+      "grad_norm": 0.35921573638916016,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2760
+    },
+    {
+      "epoch": 2.010889292196007,
+      "grad_norm": 1.013861894607544,
+      "learning_rate": 0.0002,
+      "loss": 0.8968,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0181488203266786,
+      "grad_norm": 0.4425240159034729,
+      "learning_rate": 0.0002,
+      "loss": 0.7873,
+      "step": 2780
+    },
+    {
+      "epoch": 2.02540834845735,
+      "grad_norm": 0.60174161195755,
+      "learning_rate": 0.0002,
+      "loss": 0.8758,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0326678765880217,
+      "grad_norm": 0.47582098841667175,
+      "learning_rate": 0.0002,
+      "loss": 0.7989,
+      "step": 2800
+    },
+    {
+      "epoch": 2.0399274047186933,
+      "grad_norm": 0.6012811660766602,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 2810
+    },
+    {
+      "epoch": 2.047186932849365,
+      "grad_norm": 0.4444408118724823,
+      "learning_rate": 0.0002,
+      "loss": 0.8109,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0544464609800364,
+      "grad_norm": 0.4864003360271454,
+      "learning_rate": 0.0002,
+      "loss": 0.8621,
+      "step": 2830
+    },
+    {
+      "epoch": 2.061705989110708,
+      "grad_norm": 0.5104215741157532,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.6218489408493042,
+      "learning_rate": 0.0002,
+      "loss": 0.7789,
+      "step": 2850
+    },
+    {
+      "epoch": 2.076225045372051,
+      "grad_norm": 0.705784261226654,
+      "learning_rate": 0.0002,
+      "loss": 0.8778,
+      "step": 2860
+    },
+    {
+      "epoch": 2.0834845735027225,
+      "grad_norm": 0.48091503977775574,
+      "learning_rate": 0.0002,
+      "loss": 0.7851,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0907441016333936,
+      "grad_norm": 0.5062456727027893,
+      "learning_rate": 0.0002,
+      "loss": 0.8444,
+      "step": 2880
+    },
+    {
+      "epoch": 2.098003629764065,
+      "grad_norm": 0.4862022399902344,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 0.48264044523239136,
+      "learning_rate": 0.0002,
+      "loss": 0.8593,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1125226860254083,
+      "grad_norm": 0.43744346499443054,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2910
+    },
+    {
+      "epoch": 2.11978221415608,
+      "grad_norm": 0.5480492115020752,
+      "learning_rate": 0.0002,
+      "loss": 0.9099,
+      "step": 2920
+    },
+    {
+      "epoch": 2.1270417422867514,
+      "grad_norm": 0.5068560838699341,
+      "learning_rate": 0.0002,
+      "loss": 0.7727,
+      "step": 2930
+    },
+    {
+      "epoch": 2.134301270417423,
+      "grad_norm": 0.4650852680206299,
+      "learning_rate": 0.0002,
+      "loss": 0.8403,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1415607985480944,
+      "grad_norm": 0.4929981231689453,
+      "learning_rate": 0.0002,
+      "loss": 0.8467,
+      "step": 2950
+    },
+    {
+      "epoch": 2.148820326678766,
+      "grad_norm": 0.6537389755249023,
+      "learning_rate": 0.0002,
+      "loss": 0.8747,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1560798548094375,
+      "grad_norm": 0.8032940626144409,
+      "learning_rate": 0.0002,
+      "loss": 0.7663,
+      "step": 2970
+    },
+    {
+      "epoch": 2.163339382940109,
+      "grad_norm": 0.7131643891334534,
+      "learning_rate": 0.0002,
+      "loss": 0.7604,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1705989110707806,
+      "grad_norm": 0.6034275889396667,
+      "learning_rate": 0.0002,
+      "loss": 0.8424,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1778584392014517,
+      "grad_norm": 0.6081095933914185,
+      "learning_rate": 0.0002,
+      "loss": 0.8838,
+      "step": 3000
+    },
+    {
+      "epoch": 2.1851179673321233,
+      "grad_norm": 0.5706912875175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9013,
+      "step": 3010
+    },
+    {
+      "epoch": 2.192377495462795,
+      "grad_norm": 0.6742380261421204,
+      "learning_rate": 0.0002,
+      "loss": 0.8473,
+      "step": 3020
+    },
+    {
+      "epoch": 2.1996370235934664,
+      "grad_norm": 0.4847496449947357,
+      "learning_rate": 0.0002,
+      "loss": 0.8372,
+      "step": 3030
+    },
+    {
+      "epoch": 2.206896551724138,
+      "grad_norm": 0.5529342889785767,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2141560798548094,
+      "grad_norm": 0.6108783483505249,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3050
+    },
+    {
+      "epoch": 2.221415607985481,
+      "grad_norm": 0.8841571807861328,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2286751361161525,
+      "grad_norm": 0.4227530360221863,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 3070
+    },
+    {
+      "epoch": 2.235934664246824,
+      "grad_norm": 0.461935818195343,
+      "learning_rate": 0.0002,
+      "loss": 0.7925,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2431941923774956,
+      "grad_norm": 0.5407412648200989,
+      "learning_rate": 0.0002,
+      "loss": 0.8542,
+      "step": 3090
+    },
+    {
+      "epoch": 2.250453720508167,
+      "grad_norm": 0.5057998895645142,
+      "learning_rate": 0.0002,
+      "loss": 0.8392,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2577132486388383,
+      "grad_norm": 0.530057966709137,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 3110
+    },
+    {
+      "epoch": 2.26497277676951,
+      "grad_norm": 0.5066842436790466,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2722323049001814,
+      "grad_norm": 0.5069178342819214,
+      "learning_rate": 0.0002,
+      "loss": 0.7903,
+      "step": 3130
+    },
+    {
+      "epoch": 2.279491833030853,
+      "grad_norm": 0.6095499396324158,
+      "learning_rate": 0.0002,
+      "loss": 0.8458,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2867513611615244,
+      "grad_norm": 0.49524766206741333,
+      "learning_rate": 0.0002,
+      "loss": 0.8171,
+      "step": 3150
+    },
+    {
+      "epoch": 2.294010889292196,
+      "grad_norm": 0.5334409475326538,
+      "learning_rate": 0.0002,
+      "loss": 0.7568,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3012704174228675,
+      "grad_norm": 1.681748867034912,
+      "learning_rate": 0.0002,
+      "loss": 0.7533,
+      "step": 3170
+    },
+    {
+      "epoch": 2.308529945553539,
+      "grad_norm": 0.7225565314292908,
+      "learning_rate": 0.0002,
+      "loss": 0.8155,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3157894736842106,
+      "grad_norm": 0.5379388928413391,
+      "learning_rate": 0.0002,
+      "loss": 0.8034,
+      "step": 3190
+    },
+    {
+      "epoch": 2.323049001814882,
+      "grad_norm": 0.45770326256752014,
+      "learning_rate": 0.0002,
+      "loss": 0.8833,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3303085299455537,
+      "grad_norm": 0.6984533071517944,
+      "learning_rate": 0.0002,
+      "loss": 0.8419,
+      "step": 3210
+    },
+    {
+      "epoch": 2.337568058076225,
+      "grad_norm": 0.6725744605064392,
+      "learning_rate": 0.0002,
+      "loss": 0.7414,
+      "step": 3220
+    },
+    {
+      "epoch": 2.344827586206897,
+      "grad_norm": 1.1247471570968628,
+      "learning_rate": 0.0002,
+      "loss": 0.8657,
+      "step": 3230
+    },
+    {
+      "epoch": 2.352087114337568,
+      "grad_norm": 1.0240263938903809,
+      "learning_rate": 0.0002,
+      "loss": 0.7782,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3593466424682394,
+      "grad_norm": 0.5608096122741699,
+      "learning_rate": 0.0002,
+      "loss": 0.826,
+      "step": 3250
+    },
+    {
+      "epoch": 2.366606170598911,
+      "grad_norm": 0.8294990062713623,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3738656987295825,
+      "grad_norm": 0.6734224557876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9845,
+      "step": 3270
+    },
+    {
+      "epoch": 2.381125226860254,
+      "grad_norm": 0.6862800717353821,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3883847549909256,
+      "grad_norm": 0.5442930459976196,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 3290
+    },
+    {
+      "epoch": 2.395644283121597,
+      "grad_norm": 0.5745394229888916,
+      "learning_rate": 0.0002,
+      "loss": 0.8543,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4029038112522687,
+      "grad_norm": 0.6257799863815308,
+      "learning_rate": 0.0002,
+      "loss": 0.833,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4101633393829403,
+      "grad_norm": 0.5608420968055725,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3320
+    },
+    {
+      "epoch": 2.417422867513612,
+      "grad_norm": 0.5512017011642456,
+      "learning_rate": 0.0002,
+      "loss": 0.8745,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4246823956442833,
+      "grad_norm": 0.7452999353408813,
+      "learning_rate": 0.0002,
+      "loss": 0.8642,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4319419237749544,
+      "grad_norm": 0.4604301452636719,
+      "learning_rate": 0.0002,
+      "loss": 0.8417,
+      "step": 3350
+    },
+    {
+      "epoch": 2.439201451905626,
+      "grad_norm": 0.8225823640823364,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4464609800362975,
+      "grad_norm": 0.8604981303215027,
+      "learning_rate": 0.0002,
+      "loss": 0.8144,
+      "step": 3370
+    },
+    {
+      "epoch": 2.453720508166969,
+      "grad_norm": 0.6620925664901733,
+      "learning_rate": 0.0002,
+      "loss": 0.9276,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4609800362976406,
+      "grad_norm": 0.4750158488750458,
+      "learning_rate": 0.0002,
+      "loss": 0.8381,
+      "step": 3390
+    },
+    {
+      "epoch": 2.468239564428312,
+      "grad_norm": 0.6061418056488037,
+      "learning_rate": 0.0002,
+      "loss": 0.8249,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4754990925589837,
+      "grad_norm": 0.5977247357368469,
+      "learning_rate": 0.0002,
+      "loss": 0.8944,
+      "step": 3410
+    },
+    {
+      "epoch": 2.4827586206896552,
+      "grad_norm": 0.5004227757453918,
+      "learning_rate": 0.0002,
+      "loss": 0.8366,
+      "step": 3420
+    },
+    {
+      "epoch": 2.490018148820327,
+      "grad_norm": 0.46428972482681274,
+      "learning_rate": 0.0002,
+      "loss": 0.9133,
+      "step": 3430
+    },
+    {
+      "epoch": 2.4972776769509983,
+      "grad_norm": 0.7658395171165466,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 3440
+    },
+    {
+      "epoch": 2.50453720508167,
+      "grad_norm": 0.621819794178009,
+      "learning_rate": 0.0002,
+      "loss": 0.9066,
+      "step": 3450
+    },
+    {
+      "epoch": 2.511796733212341,
+      "grad_norm": 0.4556088149547577,
+      "learning_rate": 0.0002,
+      "loss": 0.8481,
+      "step": 3460
+    },
+    {
+      "epoch": 2.519056261343013,
+      "grad_norm": 0.6124140024185181,
+      "learning_rate": 0.0002,
+      "loss": 0.8522,
+      "step": 3470
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 0.6256231665611267,
+      "learning_rate": 0.0002,
+      "loss": 0.8158,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5335753176043556,
+      "grad_norm": 0.5464141964912415,
+      "learning_rate": 0.0002,
+      "loss": 0.9097,
+      "step": 3490
+    },
+    {
+      "epoch": 2.540834845735027,
+      "grad_norm": 0.51471346616745,
+      "learning_rate": 0.0002,
+      "loss": 0.8708,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5480943738656987,
+      "grad_norm": 0.5326165556907654,
+      "learning_rate": 0.0002,
+      "loss": 0.9042,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5553539019963702,
+      "grad_norm": 0.4750378429889679,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 3520
+    },
+    {
+      "epoch": 2.562613430127042,
+      "grad_norm": 0.5292279124259949,
+      "learning_rate": 0.0002,
+      "loss": 0.8248,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5698729582577133,
+      "grad_norm": 0.6145227551460266,
+      "learning_rate": 0.0002,
+      "loss": 0.8808,
+      "step": 3540
+    },
+    {
+      "epoch": 2.577132486388385,
+      "grad_norm": 0.8275189399719238,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5843920145190564,
+      "grad_norm": 0.5037438273429871,
+      "learning_rate": 0.0002,
+      "loss": 0.875,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5916515426497275,
+      "grad_norm": 0.5838707685470581,
+      "learning_rate": 0.0002,
+      "loss": 0.8324,
+      "step": 3570
+    },
+    {
+      "epoch": 2.5989110707803995,
+      "grad_norm": 0.5398710370063782,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6061705989110706,
+      "grad_norm": 0.6115376949310303,
+      "learning_rate": 0.0002,
+      "loss": 0.7843,
+      "step": 3590
+    },
+    {
+      "epoch": 2.613430127041742,
+      "grad_norm": 1.118809461593628,
+      "learning_rate": 0.0002,
+      "loss": 0.8142,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6206896551724137,
+      "grad_norm": 0.7811771631240845,
+      "learning_rate": 0.0002,
+      "loss": 0.8282,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6279491833030852,
+      "grad_norm": 0.5753175020217896,
+      "learning_rate": 0.0002,
+      "loss": 0.7997,
+      "step": 3620
+    },
+    {
+      "epoch": 2.635208711433757,
+      "grad_norm": 0.550829291343689,
+      "learning_rate": 0.0002,
+      "loss": 0.7705,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6424682395644283,
+      "grad_norm": 0.5360019207000732,
+      "learning_rate": 0.0002,
+      "loss": 0.8564,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6497277676951,
+      "grad_norm": 0.63050377368927,
+      "learning_rate": 0.0002,
+      "loss": 0.7648,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6569872958257714,
+      "grad_norm": 0.5833110213279724,
+      "learning_rate": 0.0002,
+      "loss": 0.9156,
+      "step": 3660
+    },
+    {
+      "epoch": 2.664246823956443,
+      "grad_norm": 0.5543047189712524,
+      "learning_rate": 0.0002,
+      "loss": 0.8033,
+      "step": 3670
+    },
+    {
+      "epoch": 2.671506352087114,
+      "grad_norm": 0.6842212080955505,
+      "learning_rate": 0.0002,
+      "loss": 0.8867,
+      "step": 3680
+    },
+    {
+      "epoch": 2.678765880217786,
+      "grad_norm": 0.6859333515167236,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 3690
+    },
+    {
+      "epoch": 2.686025408348457,
+      "grad_norm": 0.7038410902023315,
+      "learning_rate": 0.0002,
+      "loss": 0.8081,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6932849364791287,
+      "grad_norm": 0.447233647108078,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7005444646098002,
+      "grad_norm": 0.6453872919082642,
+      "learning_rate": 0.0002,
+      "loss": 0.8951,
+      "step": 3720
+    },
+    {
+      "epoch": 2.707803992740472,
+      "grad_norm": 0.8025672435760498,
+      "learning_rate": 0.0002,
+      "loss": 0.8292,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7150635208711433,
+      "grad_norm": 0.5997087955474854,
+      "learning_rate": 0.0002,
+      "loss": 0.8488,
+      "step": 3740
+    },
+    {
+      "epoch": 2.722323049001815,
+      "grad_norm": 0.6901142001152039,
+      "learning_rate": 0.0002,
+      "loss": 0.8308,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7295825771324864,
+      "grad_norm": 1.036145567893982,
+      "learning_rate": 0.0002,
+      "loss": 0.8517,
+      "step": 3760
+    },
+    {
+      "epoch": 2.736842105263158,
+      "grad_norm": 0.7207072377204895,
+      "learning_rate": 0.0002,
+      "loss": 0.8455,
+      "step": 3770
+    },
+    {
+      "epoch": 2.7441016333938295,
+      "grad_norm": 1.0452989339828491,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 3780
+    },
+    {
+      "epoch": 2.751361161524501,
+      "grad_norm": 0.5615278482437134,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.46439215540885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7658802177858437,
+      "grad_norm": 0.7134917974472046,
+      "learning_rate": 0.0002,
+      "loss": 0.7847,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7731397459165157,
+      "grad_norm": 0.5139115452766418,
+      "learning_rate": 0.0002,
+      "loss": 0.8315,
+      "step": 3820
+    },
+    {
+      "epoch": 2.780399274047187,
+      "grad_norm": 0.8595781326293945,
+      "learning_rate": 0.0002,
+      "loss": 0.8995,
+      "step": 3830
+    },
+    {
+      "epoch": 2.7876588021778583,
+      "grad_norm": 0.544614851474762,
+      "learning_rate": 0.0002,
+      "loss": 0.9079,
+      "step": 3840
+    },
+    {
+      "epoch": 2.79491833030853,
+      "grad_norm": 0.6073850393295288,
+      "learning_rate": 0.0002,
+      "loss": 0.8322,
+      "step": 3850
+    },
+    {
+      "epoch": 2.8021778584392014,
+      "grad_norm": 0.8224069476127625,
+      "learning_rate": 0.0002,
+      "loss": 0.8277,
+      "step": 3860
+    },
+    {
+      "epoch": 2.809437386569873,
+      "grad_norm": 0.5347970128059387,
+      "learning_rate": 0.0002,
+      "loss": 0.815,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8166969147005445,
+      "grad_norm": 0.6799601912498474,
+      "learning_rate": 0.0002,
+      "loss": 0.8219,
+      "step": 3880
+    },
+    {
+      "epoch": 2.823956442831216,
+      "grad_norm": 0.5219197869300842,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8312159709618876,
+      "grad_norm": 0.5710130333900452,
+      "learning_rate": 0.0002,
+      "loss": 0.8649,
+      "step": 3900
+    },
+    {
+      "epoch": 2.838475499092559,
+      "grad_norm": 0.5857213139533997,
+      "learning_rate": 0.0002,
+      "loss": 0.8741,
+      "step": 3910
+    },
+    {
+      "epoch": 2.8457350272232302,
+      "grad_norm": 0.5206644535064697,
+      "learning_rate": 0.0002,
+      "loss": 0.8154,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8529945553539022,
+      "grad_norm": 1.2902015447616577,
+      "learning_rate": 0.0002,
+      "loss": 0.9328,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8602540834845733,
+      "grad_norm": 0.5252797603607178,
+      "learning_rate": 0.0002,
+      "loss": 0.8192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.867513611615245,
+      "grad_norm": 0.5925108790397644,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8747731397459164,
+      "grad_norm": 0.9719610810279846,
+      "learning_rate": 0.0002,
+      "loss": 0.8611,
+      "step": 3960
+    },
+    {
+      "epoch": 2.882032667876588,
+      "grad_norm": 0.6834747195243835,
+      "learning_rate": 0.0002,
+      "loss": 0.8531,
+      "step": 3970
+    },
+    {
+      "epoch": 2.8892921960072595,
+      "grad_norm": 0.6952353119850159,
+      "learning_rate": 0.0002,
+      "loss": 0.8124,
+      "step": 3980
+    },
+    {
+      "epoch": 2.896551724137931,
+      "grad_norm": 0.49889910221099854,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9038112522686026,
+      "grad_norm": 0.5007026791572571,
+      "learning_rate": 0.0002,
+      "loss": 0.8197,
+      "step": 4000
+    },
+    {
+      "epoch": 2.911070780399274,
+      "grad_norm": 0.5474239587783813,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9183303085299457,
+      "grad_norm": 0.6617428660392761,
+      "learning_rate": 0.0002,
+      "loss": 0.8579,
+      "step": 4020
+    },
+    {
+      "epoch": 2.925589836660617,
+      "grad_norm": 0.6097776293754578,
+      "learning_rate": 0.0002,
+      "loss": 0.8583,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9328493647912888,
+      "grad_norm": 0.5985828638076782,
+      "learning_rate": 0.0002,
+      "loss": 0.8296,
+      "step": 4040
+    },
+    {
+      "epoch": 2.94010889292196,
+      "grad_norm": 0.769488513469696,
+      "learning_rate": 0.0002,
+      "loss": 0.9134,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 0.5167055130004883,
+      "learning_rate": 0.0002,
+      "loss": 0.8148,
+      "step": 4060
+    },
+    {
+      "epoch": 2.954627949183303,
+      "grad_norm": 0.6265496015548706,
+      "learning_rate": 0.0002,
+      "loss": 0.8665,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9618874773139745,
+      "grad_norm": 1.2644082307815552,
+      "learning_rate": 0.0002,
+      "loss": 0.9218,
+      "step": 4080
+    },
+    {
+      "epoch": 2.969147005444646,
+      "grad_norm": 0.6007561087608337,
+      "learning_rate": 0.0002,
+      "loss": 0.8026,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9764065335753176,
+      "grad_norm": 0.47984927892684937,
+      "learning_rate": 0.0002,
+      "loss": 0.8262,
+      "step": 4100
+    },
+    {
+      "epoch": 2.983666061705989,
+      "grad_norm": 1.128198504447937,
+      "learning_rate": 0.0002,
+      "loss": 0.7897,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9909255898366607,
+      "grad_norm": 0.526292085647583,
+      "learning_rate": 0.0002,
+      "loss": 0.7883,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9981851179673322,
+      "grad_norm": 0.5462674498558044,
+      "learning_rate": 0.0002,
+      "loss": 0.7801,
+      "step": 4130
+    },
+    {
+      "epoch": 2.9996370235934666,
+      "eval_loss": 1.1093357801437378,
+      "eval_runtime": 46.2498,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 4132
+    },
+    {
+      "epoch": 3.0054446460980038,
+      "grad_norm": 0.48288026452064514,
+      "learning_rate": 0.0002,
+      "loss": 0.721,
+      "step": 4140
+    },
+    {
+      "epoch": 3.0127041742286753,
+      "grad_norm": 1.0181782245635986,
+      "learning_rate": 0.0002,
+      "loss": 0.6769,
+      "step": 4150
+    },
+    {
+      "epoch": 3.019963702359347,
+      "grad_norm": 0.7718019485473633,
+      "learning_rate": 0.0002,
+      "loss": 0.7185,
+      "step": 4160
+    },
+    {
+      "epoch": 3.027223230490018,
+      "grad_norm": 0.7492219805717468,
+      "learning_rate": 0.0002,
+      "loss": 0.6552,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0344827586206895,
+      "grad_norm": 0.9363632798194885,
+      "learning_rate": 0.0002,
+      "loss": 0.6678,
+      "step": 4180
+    },
+    {
+      "epoch": 3.041742286751361,
+      "grad_norm": 0.6888533234596252,
+      "learning_rate": 0.0002,
+      "loss": 0.7187,
+      "step": 4190
+    },
+    {
+      "epoch": 3.0490018148820326,
+      "grad_norm": 0.7072834968566895,
+      "learning_rate": 0.0002,
+      "loss": 0.6469,
+      "step": 4200
+    },
+    {
+      "epoch": 3.056261343012704,
+      "grad_norm": 0.7182047963142395,
+      "learning_rate": 0.0002,
+      "loss": 0.6387,
+      "step": 4210
+    },
+    {
+      "epoch": 3.0635208711433757,
+      "grad_norm": 0.7194355130195618,
+      "learning_rate": 0.0002,
+      "loss": 0.6385,
+      "step": 4220
+    },
+    {
+      "epoch": 3.0707803992740472,
+      "grad_norm": 0.9454023838043213,
+      "learning_rate": 0.0002,
+      "loss": 0.5812,
+      "step": 4230
+    },
+    {
+      "epoch": 3.0780399274047188,
+      "grad_norm": 0.838657557964325,
+      "learning_rate": 0.0002,
+      "loss": 0.6036,
+      "step": 4240
+    },
+    {
+      "epoch": 3.0852994555353903,
+      "grad_norm": 0.740113377571106,
+      "learning_rate": 0.0002,
+      "loss": 0.646,
+      "step": 4250
+    },
+    {
+      "epoch": 3.092558983666062,
+      "grad_norm": 0.6616561412811279,
+      "learning_rate": 0.0002,
+      "loss": 0.604,
+      "step": 4260
+    },
+    {
+      "epoch": 3.0998185117967334,
+      "grad_norm": 0.8846506476402283,
+      "learning_rate": 0.0002,
+      "loss": 0.6462,
+      "step": 4270
+    },
+    {
+      "epoch": 3.107078039927405,
+      "grad_norm": 0.6322125792503357,
+      "learning_rate": 0.0002,
+      "loss": 0.6037,
+      "step": 4280
+    },
+    {
+      "epoch": 3.114337568058076,
+      "grad_norm": 0.7461467385292053,
+      "learning_rate": 0.0002,
+      "loss": 0.5953,
+      "step": 4290
+    },
+    {
+      "epoch": 3.1215970961887476,
+      "grad_norm": 0.8251287341117859,
+      "learning_rate": 0.0002,
+      "loss": 0.6099,
+      "step": 4300
+    },
+    {
+      "epoch": 3.128856624319419,
+      "grad_norm": 0.8767673373222351,
+      "learning_rate": 0.0002,
+      "loss": 0.6284,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1361161524500907,
+      "grad_norm": 0.7758759260177612,
+      "learning_rate": 0.0002,
+      "loss": 0.7535,
+      "step": 4320
+    },
+    {
+      "epoch": 3.143375680580762,
+      "grad_norm": 1.1056879758834839,
+      "learning_rate": 0.0002,
+      "loss": 0.6624,
+      "step": 4330
+    },
+    {
+      "epoch": 3.1506352087114338,
+      "grad_norm": 0.8259835243225098,
+      "learning_rate": 0.0002,
+      "loss": 0.691,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 0.6607027053833008,
+      "learning_rate": 0.0002,
+      "loss": 0.6635,
+      "step": 4350
+    },
+    {
+      "epoch": 3.165154264972777,
+      "grad_norm": 0.7983301281929016,
+      "learning_rate": 0.0002,
+      "loss": 0.5911,
+      "step": 4360
+    },
+    {
+      "epoch": 3.1724137931034484,
+      "grad_norm": 0.6725239157676697,
+      "learning_rate": 0.0002,
+      "loss": 0.6496,
+      "step": 4370
+    },
+    {
+      "epoch": 3.17967332123412,
+      "grad_norm": 0.9052095413208008,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 4380
+    },
+    {
+      "epoch": 3.1869328493647915,
+      "grad_norm": 0.8131307363510132,
+      "learning_rate": 0.0002,
+      "loss": 0.6877,
+      "step": 4390
+    },
+    {
+      "epoch": 3.1941923774954626,
+      "grad_norm": 0.6435626149177551,
+      "learning_rate": 0.0002,
+      "loss": 0.6384,
+      "step": 4400
+    },
+    {
+      "epoch": 3.201451905626134,
+      "grad_norm": 0.84367436170578,
+      "learning_rate": 0.0002,
+      "loss": 0.5819,
+      "step": 4410
+    },
+    {
+      "epoch": 3.2087114337568057,
+      "grad_norm": 1.5018867254257202,
+      "learning_rate": 0.0002,
+      "loss": 0.6104,
+      "step": 4420
+    },
+    {
+      "epoch": 3.215970961887477,
+      "grad_norm": 0.7019091844558716,
+      "learning_rate": 0.0002,
+      "loss": 0.6838,
+      "step": 4430
+    },
+    {
+      "epoch": 3.2232304900181488,
+      "grad_norm": 0.9164197444915771,
+      "learning_rate": 0.0002,
+      "loss": 0.6153,
+      "step": 4440
+    },
+    {
+      "epoch": 3.2304900181488203,
+      "grad_norm": 0.7890861630439758,
+      "learning_rate": 0.0002,
+      "loss": 0.6618,
+      "step": 4450
+    },
+    {
+      "epoch": 3.237749546279492,
+      "grad_norm": 0.6517660617828369,
+      "learning_rate": 0.0002,
+      "loss": 0.6401,
+      "step": 4460
+    },
+    {
+      "epoch": 3.2450090744101634,
+      "grad_norm": 1.10188889503479,
+      "learning_rate": 0.0002,
+      "loss": 0.6699,
+      "step": 4470
+    },
+    {
+      "epoch": 3.252268602540835,
+      "grad_norm": 0.8158330917358398,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 4480
+    },
+    {
+      "epoch": 3.2595281306715065,
+      "grad_norm": 0.7663109302520752,
+      "learning_rate": 0.0002,
+      "loss": 0.7757,
+      "step": 4490
+    },
+    {
+      "epoch": 3.266787658802178,
+      "grad_norm": 0.8473444581031799,
+      "learning_rate": 0.0002,
+      "loss": 0.6539,
+      "step": 4500
+    },
+    {
+      "epoch": 3.274047186932849,
+      "grad_norm": 0.9724768996238708,
+      "learning_rate": 0.0002,
+      "loss": 0.6511,
+      "step": 4510
+    },
+    {
+      "epoch": 3.281306715063521,
+      "grad_norm": 0.8516759276390076,
+      "learning_rate": 0.0002,
+      "loss": 0.5464,
+      "step": 4520
+    },
+    {
+      "epoch": 3.288566243194192,
+      "grad_norm": 0.7543437480926514,
+      "learning_rate": 0.0002,
+      "loss": 0.6534,
+      "step": 4530
+    },
+    {
+      "epoch": 3.2958257713248638,
+      "grad_norm": 1.0472029447555542,
+      "learning_rate": 0.0002,
+      "loss": 0.6095,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3030852994555353,
+      "grad_norm": 0.6240826845169067,
+      "learning_rate": 0.0002,
+      "loss": 0.6216,
+      "step": 4550
+    },
+    {
+      "epoch": 3.310344827586207,
+      "grad_norm": 0.9957774877548218,
+      "learning_rate": 0.0002,
+      "loss": 0.6223,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3176043557168784,
+      "grad_norm": 0.6448912620544434,
+      "learning_rate": 0.0002,
+      "loss": 0.618,
+      "step": 4570
+    },
+    {
+      "epoch": 3.32486388384755,
+      "grad_norm": 0.7519692778587341,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 4580
+    },
+    {
+      "epoch": 3.3321234119782215,
+      "grad_norm": 0.7367453575134277,
+      "learning_rate": 0.0002,
+      "loss": 0.6672,
+      "step": 4590
+    },
+    {
+      "epoch": 3.339382940108893,
+      "grad_norm": 0.8064960837364197,
+      "learning_rate": 0.0002,
+      "loss": 0.6517,
+      "step": 4600
+    },
+    {
+      "epoch": 3.3466424682395646,
+      "grad_norm": 0.7664631009101868,
+      "learning_rate": 0.0002,
+      "loss": 0.6062,
+      "step": 4610
+    },
+    {
+      "epoch": 3.353901996370236,
+      "grad_norm": 0.7803396582603455,
+      "learning_rate": 0.0002,
+      "loss": 0.6834,
+      "step": 4620
+    },
+    {
+      "epoch": 3.3611615245009077,
+      "grad_norm": 0.9141599535942078,
+      "learning_rate": 0.0002,
+      "loss": 0.6961,
+      "step": 4630
+    },
+    {
+      "epoch": 3.3684210526315788,
+      "grad_norm": 0.9719856381416321,
+      "learning_rate": 0.0002,
+      "loss": 0.6889,
+      "step": 4640
+    },
+    {
+      "epoch": 3.3756805807622503,
+      "grad_norm": 0.9223218560218811,
+      "learning_rate": 0.0002,
+      "loss": 0.6914,
+      "step": 4650
+    },
+    {
+      "epoch": 3.382940108892922,
+      "grad_norm": 0.7289277911186218,
+      "learning_rate": 0.0002,
+      "loss": 0.5981,
+      "step": 4660
+    },
+    {
+      "epoch": 3.3901996370235934,
+      "grad_norm": 1.039724349975586,
+      "learning_rate": 0.0002,
+      "loss": 0.595,
+      "step": 4670
+    },
+    {
+      "epoch": 3.397459165154265,
+      "grad_norm": 1.397438883781433,
+      "learning_rate": 0.0002,
+      "loss": 0.8121,
+      "step": 4680
+    },
+    {
+      "epoch": 3.4047186932849365,
+      "grad_norm": 1.0069999694824219,
+      "learning_rate": 0.0002,
+      "loss": 0.6334,
+      "step": 4690
+    },
+    {
+      "epoch": 3.411978221415608,
+      "grad_norm": 0.816291332244873,
+      "learning_rate": 0.0002,
+      "loss": 0.6598,
+      "step": 4700
+    },
+    {
+      "epoch": 3.4192377495462796,
+      "grad_norm": 1.2831530570983887,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 4710
+    },
+    {
+      "epoch": 3.426497277676951,
+      "grad_norm": 0.9573889970779419,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 4720
+    },
+    {
+      "epoch": 3.4337568058076227,
+      "grad_norm": 0.7685632705688477,
+      "learning_rate": 0.0002,
+      "loss": 0.7279,
+      "step": 4730
+    },
+    {
+      "epoch": 3.441016333938294,
+      "grad_norm": 0.7019195556640625,
+      "learning_rate": 0.0002,
+      "loss": 0.6104,
+      "step": 4740
+    },
+    {
+      "epoch": 3.4482758620689653,
+      "grad_norm": 0.7244833707809448,
+      "learning_rate": 0.0002,
+      "loss": 0.7606,
+      "step": 4750
+    },
+    {
+      "epoch": 3.455535390199637,
+      "grad_norm": 1.3468551635742188,
+      "learning_rate": 0.0002,
+      "loss": 0.6951,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4627949183303084,
+      "grad_norm": 0.822846531867981,
+      "learning_rate": 0.0002,
+      "loss": 0.6945,
+      "step": 4770
+    },
+    {
+      "epoch": 3.47005444646098,
+      "grad_norm": 0.7311608195304871,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 4780
+    },
+    {
+      "epoch": 3.4773139745916515,
+      "grad_norm": 0.9466770887374878,
+      "learning_rate": 0.0002,
+      "loss": 0.7019,
+      "step": 4790
+    },
+    {
+      "epoch": 3.484573502722323,
+      "grad_norm": 1.1527155637741089,
+      "learning_rate": 0.0002,
+      "loss": 0.7767,
+      "step": 4800
+    },
+    {
+      "epoch": 3.4918330308529946,
+      "grad_norm": 1.1288906335830688,
+      "learning_rate": 0.0002,
+      "loss": 0.6882,
+      "step": 4810
+    },
+    {
+      "epoch": 3.499092558983666,
+      "grad_norm": 0.9096164107322693,
+      "learning_rate": 0.0002,
+      "loss": 0.6564,
+      "step": 4820
+    },
+    {
+      "epoch": 3.5063520871143377,
+      "grad_norm": 0.7988565564155579,
+      "learning_rate": 0.0002,
+      "loss": 0.6127,
+      "step": 4830
+    },
+    {
+      "epoch": 3.513611615245009,
+      "grad_norm": 0.7183415293693542,
+      "learning_rate": 0.0002,
+      "loss": 0.7004,
+      "step": 4840
+    },
+    {
+      "epoch": 3.5208711433756807,
+      "grad_norm": 0.6614915132522583,
+      "learning_rate": 0.0002,
+      "loss": 0.74,
+      "step": 4850
+    },
+    {
+      "epoch": 3.528130671506352,
+      "grad_norm": 0.8609521985054016,
+      "learning_rate": 0.0002,
+      "loss": 0.7271,
+      "step": 4860
+    },
+    {
+      "epoch": 3.535390199637024,
+      "grad_norm": 0.86552894115448,
+      "learning_rate": 0.0002,
+      "loss": 0.6664,
+      "step": 4870
+    },
+    {
+      "epoch": 3.542649727767695,
+      "grad_norm": 0.6926496028900146,
+      "learning_rate": 0.0002,
+      "loss": 0.6432,
+      "step": 4880
+    },
+    {
+      "epoch": 3.5499092558983665,
+      "grad_norm": 0.8157467246055603,
+      "learning_rate": 0.0002,
+      "loss": 0.7117,
+      "step": 4890
+    },
+    {
+      "epoch": 3.557168784029038,
+      "grad_norm": 0.9085357189178467,
+      "learning_rate": 0.0002,
+      "loss": 0.6201,
+      "step": 4900
+    },
+    {
+      "epoch": 3.5644283121597096,
+      "grad_norm": 0.6322644948959351,
+      "learning_rate": 0.0002,
+      "loss": 0.6521,
+      "step": 4910
+    },
+    {
+      "epoch": 3.571687840290381,
+      "grad_norm": 1.263205885887146,
+      "learning_rate": 0.0002,
+      "loss": 0.6607,
+      "step": 4920
+    },
+    {
+      "epoch": 3.5789473684210527,
+      "grad_norm": 0.8901070356369019,
+      "learning_rate": 0.0002,
+      "loss": 0.6657,
+      "step": 4930
+    },
+    {
+      "epoch": 3.586206896551724,
+      "grad_norm": 0.7983952164649963,
+      "learning_rate": 0.0002,
+      "loss": 0.6434,
+      "step": 4940
+    },
+    {
+      "epoch": 3.5934664246823957,
+      "grad_norm": 0.9887813925743103,
+      "learning_rate": 0.0002,
+      "loss": 0.6861,
+      "step": 4950
+    },
+    {
+      "epoch": 3.6007259528130673,
+      "grad_norm": 0.7895187735557556,
+      "learning_rate": 0.0002,
+      "loss": 0.6502,
+      "step": 4960
+    },
+    {
+      "epoch": 3.6079854809437384,
+      "grad_norm": 0.9685819745063782,
+      "learning_rate": 0.0002,
+      "loss": 0.7111,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6152450090744104,
+      "grad_norm": 0.6576591730117798,
+      "learning_rate": 0.0002,
+      "loss": 0.6915,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6225045372050815,
+      "grad_norm": 0.856985330581665,
+      "learning_rate": 0.0002,
+      "loss": 0.6195,
+      "step": 4990
+    },
+    {
+      "epoch": 3.629764065335753,
+      "grad_norm": 0.7230252623558044,
+      "learning_rate": 0.0002,
+      "loss": 0.6318,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6370235934664246,
+      "grad_norm": 0.8260893821716309,
+      "learning_rate": 0.0002,
+      "loss": 0.742,
+      "step": 5010
+    },
+    {
+      "epoch": 3.644283121597096,
+      "grad_norm": 0.7635950446128845,
+      "learning_rate": 0.0002,
+      "loss": 0.7223,
+      "step": 5020
+    },
+    {
+      "epoch": 3.6515426497277677,
+      "grad_norm": 0.7060768604278564,
+      "learning_rate": 0.0002,
+      "loss": 0.6837,
+      "step": 5030
+    },
+    {
+      "epoch": 3.658802177858439,
+      "grad_norm": 0.8020303249359131,
+      "learning_rate": 0.0002,
+      "loss": 0.6921,
+      "step": 5040
+    },
+    {
+      "epoch": 3.6660617059891107,
+      "grad_norm": 0.8530341386795044,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 5050
+    },
+    {
+      "epoch": 3.6733212341197823,
+      "grad_norm": 0.6667101979255676,
+      "learning_rate": 0.0002,
+      "loss": 0.7222,
+      "step": 5060
+    },
+    {
+      "epoch": 3.680580762250454,
+      "grad_norm": 0.7385406494140625,
+      "learning_rate": 0.0002,
+      "loss": 0.7081,
+      "step": 5070
+    },
+    {
+      "epoch": 3.6878402903811254,
+      "grad_norm": 0.7753380537033081,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 5080
+    },
+    {
+      "epoch": 3.695099818511797,
+      "grad_norm": 0.7516207098960876,
+      "learning_rate": 0.0002,
+      "loss": 0.6491,
+      "step": 5090
+    },
+    {
+      "epoch": 3.702359346642468,
+      "grad_norm": 0.8171586394309998,
+      "learning_rate": 0.0002,
+      "loss": 0.672,
+      "step": 5100
+    },
+    {
+      "epoch": 3.70961887477314,
+      "grad_norm": 1.0796279907226562,
+      "learning_rate": 0.0002,
+      "loss": 0.7459,
+      "step": 5110
+    },
+    {
+      "epoch": 3.716878402903811,
+      "grad_norm": 0.6957688927650452,
+      "learning_rate": 0.0002,
+      "loss": 0.5948,
+      "step": 5120
+    },
+    {
+      "epoch": 3.7241379310344827,
+      "grad_norm": 0.8550161719322205,
+      "learning_rate": 0.0002,
+      "loss": 0.7515,
+      "step": 5130
+    },
+    {
+      "epoch": 3.731397459165154,
+      "grad_norm": 0.9396728277206421,
+      "learning_rate": 0.0002,
+      "loss": 0.7286,
+      "step": 5140
+    },
+    {
+      "epoch": 3.7386569872958257,
+      "grad_norm": 1.4264805316925049,
+      "learning_rate": 0.0002,
+      "loss": 0.7594,
+      "step": 5150
+    },
+    {
+      "epoch": 3.7459165154264973,
+      "grad_norm": 0.8725108504295349,
+      "learning_rate": 0.0002,
+      "loss": 0.6575,
+      "step": 5160
+    },
+    {
+      "epoch": 3.753176043557169,
+      "grad_norm": 1.0346195697784424,
+      "learning_rate": 0.0002,
+      "loss": 0.6778,
+      "step": 5170
+    },
+    {
+      "epoch": 3.7604355716878404,
+      "grad_norm": 0.5395554304122925,
+      "learning_rate": 0.0002,
+      "loss": 0.6371,
+      "step": 5180
+    },
+    {
+      "epoch": 3.767695099818512,
+      "grad_norm": 1.3153616189956665,
+      "learning_rate": 0.0002,
+      "loss": 0.7308,
+      "step": 5190
+    },
+    {
+      "epoch": 3.7749546279491835,
+      "grad_norm": 0.9879828691482544,
+      "learning_rate": 0.0002,
+      "loss": 0.78,
+      "step": 5200
+    },
+    {
+      "epoch": 3.7822141560798546,
+      "grad_norm": 0.8876672983169556,
+      "learning_rate": 0.0002,
+      "loss": 0.7068,
+      "step": 5210
+    },
+    {
+      "epoch": 3.7894736842105265,
+      "grad_norm": 0.8363267779350281,
+      "learning_rate": 0.0002,
+      "loss": 0.6283,
+      "step": 5220
+    },
+    {
+      "epoch": 3.7967332123411976,
+      "grad_norm": 0.637294590473175,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 5230
+    },
+    {
+      "epoch": 3.803992740471869,
+      "grad_norm": 1.1408970355987549,
+      "learning_rate": 0.0002,
+      "loss": 0.6685,
+      "step": 5240
+    },
+    {
+      "epoch": 3.8112522686025407,
+      "grad_norm": 1.0128360986709595,
+      "learning_rate": 0.0002,
+      "loss": 0.6761,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8185117967332123,
+      "grad_norm": 0.8061144351959229,
+      "learning_rate": 0.0002,
+      "loss": 0.6764,
+      "step": 5260
+    },
+    {
+      "epoch": 3.825771324863884,
+      "grad_norm": 0.9626626968383789,
+      "learning_rate": 0.0002,
+      "loss": 0.7254,
+      "step": 5270
+    },
+    {
+      "epoch": 3.8330308529945554,
+      "grad_norm": 0.9013627171516418,
+      "learning_rate": 0.0002,
+      "loss": 0.7367,
+      "step": 5280
+    },
+    {
+      "epoch": 3.840290381125227,
+      "grad_norm": 0.8411344289779663,
+      "learning_rate": 0.0002,
+      "loss": 0.6806,
+      "step": 5290
+    },
+    {
+      "epoch": 3.8475499092558985,
+      "grad_norm": 0.7426059246063232,
+      "learning_rate": 0.0002,
+      "loss": 0.6818,
+      "step": 5300
+    },
+    {
+      "epoch": 3.85480943738657,
+      "grad_norm": 1.003413438796997,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 5310
+    },
+    {
+      "epoch": 3.862068965517241,
+      "grad_norm": 0.7527840733528137,
+      "learning_rate": 0.0002,
+      "loss": 0.8554,
+      "step": 5320
+    },
+    {
+      "epoch": 3.869328493647913,
+      "grad_norm": 0.738610565662384,
+      "learning_rate": 0.0002,
+      "loss": 0.7521,
+      "step": 5330
+    },
+    {
+      "epoch": 3.876588021778584,
+      "grad_norm": 0.7277999520301819,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 5340
+    },
+    {
+      "epoch": 3.8838475499092557,
+      "grad_norm": 0.5951359272003174,
+      "learning_rate": 0.0002,
+      "loss": 0.7503,
+      "step": 5350
+    },
+    {
+      "epoch": 3.8911070780399273,
+      "grad_norm": 1.043884038925171,
+      "learning_rate": 0.0002,
+      "loss": 0.7447,
+      "step": 5360
+    },
+    {
+      "epoch": 3.898366606170599,
+      "grad_norm": 0.8436498045921326,
+      "learning_rate": 0.0002,
+      "loss": 0.6862,
+      "step": 5370
+    },
+    {
+      "epoch": 3.9056261343012704,
+      "grad_norm": 0.5603365302085876,
+      "learning_rate": 0.0002,
+      "loss": 0.665,
+      "step": 5380
+    },
+    {
+      "epoch": 3.912885662431942,
+      "grad_norm": 1.0128886699676514,
+      "learning_rate": 0.0002,
+      "loss": 0.7098,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9201451905626135,
+      "grad_norm": 0.7970930337905884,
+      "learning_rate": 0.0002,
+      "loss": 0.6707,
+      "step": 5400
+    },
+    {
+      "epoch": 3.927404718693285,
+      "grad_norm": 0.7699369192123413,
+      "learning_rate": 0.0002,
+      "loss": 0.637,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9346642468239565,
+      "grad_norm": 0.800561249256134,
+      "learning_rate": 0.0002,
+      "loss": 0.6742,
+      "step": 5420
+    },
+    {
+      "epoch": 3.941923774954628,
+      "grad_norm": 0.8020331859588623,
+      "learning_rate": 0.0002,
+      "loss": 0.7208,
+      "step": 5430
+    },
+    {
+      "epoch": 3.9491833030852996,
+      "grad_norm": 0.7461140155792236,
+      "learning_rate": 0.0002,
+      "loss": 0.7294,
+      "step": 5440
+    },
+    {
+      "epoch": 3.9564428312159707,
+      "grad_norm": 0.8346918821334839,
+      "learning_rate": 0.0002,
+      "loss": 0.7013,
+      "step": 5450
+    },
+    {
+      "epoch": 3.9637023593466427,
+      "grad_norm": 0.9723302125930786,
+      "learning_rate": 0.0002,
+      "loss": 0.6289,
+      "step": 5460
+    },
+    {
+      "epoch": 3.970961887477314,
+      "grad_norm": 0.6809740662574768,
+      "learning_rate": 0.0002,
+      "loss": 0.8029,
+      "step": 5470
+    },
+    {
+      "epoch": 3.9782214156079854,
+      "grad_norm": 0.7353498339653015,
+      "learning_rate": 0.0002,
+      "loss": 0.6896,
+      "step": 5480
+    },
+    {
+      "epoch": 3.985480943738657,
+      "grad_norm": 0.748009443283081,
+      "learning_rate": 0.0002,
+      "loss": 0.6722,
+      "step": 5490
+    },
+    {
+      "epoch": 3.9927404718693285,
+      "grad_norm": 1.3656195402145386,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 5500
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.8402108550071716,
+      "learning_rate": 0.0002,
+      "loss": 0.7368,
+      "step": 5510
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.17229425907135,
+      "eval_runtime": 46.2554,
+      "eval_samples_per_second": 9.426,
+      "eval_steps_per_second": 1.189,
+      "step": 5510
+    },
+    {
+      "epoch": 4.007259528130671,
+      "grad_norm": 0.8601235747337341,
+      "learning_rate": 0.0002,
+      "loss": 0.4637,
+      "step": 5520
+    },
+    {
+      "epoch": 4.014519056261343,
+      "grad_norm": 1.2635200023651123,
+      "learning_rate": 0.0002,
+      "loss": 0.4717,
+      "step": 5530
+    },
+    {
+      "epoch": 4.021778584392014,
+      "grad_norm": 1.0257477760314941,
+      "learning_rate": 0.0002,
+      "loss": 0.503,
+      "step": 5540
+    },
+    {
+      "epoch": 4.029038112522686,
+      "grad_norm": 0.9436745047569275,
+      "learning_rate": 0.0002,
+      "loss": 0.4547,
+      "step": 5550
+    },
+    {
+      "epoch": 4.036297640653357,
+      "grad_norm": 0.9443606734275818,
+      "learning_rate": 0.0002,
+      "loss": 0.459,
+      "step": 5560
+    },
+    {
+      "epoch": 4.043557168784029,
+      "grad_norm": 1.3965742588043213,
+      "learning_rate": 0.0002,
+      "loss": 0.5386,
+      "step": 5570
+    },
+    {
+      "epoch": 4.0508166969147,
+      "grad_norm": 0.8973520398139954,
+      "learning_rate": 0.0002,
+      "loss": 0.4248,
+      "step": 5580
+    },
+    {
+      "epoch": 4.058076225045372,
+      "grad_norm": 0.9998409748077393,
+      "learning_rate": 0.0002,
+      "loss": 0.4111,
+      "step": 5590
+    },
+    {
+      "epoch": 4.0653357531760435,
+      "grad_norm": 1.1213387250900269,
+      "learning_rate": 0.0002,
+      "loss": 0.4828,
+      "step": 5600
+    },
+    {
+      "epoch": 4.072595281306715,
+      "grad_norm": 0.7064558863639832,
+      "learning_rate": 0.0002,
+      "loss": 0.439,
+      "step": 5610
+    },
+    {
+      "epoch": 4.0798548094373865,
+      "grad_norm": 1.2390803098678589,
+      "learning_rate": 0.0002,
+      "loss": 0.4607,
+      "step": 5620
+    },
+    {
+      "epoch": 4.087114337568058,
+      "grad_norm": 1.123469591140747,
+      "learning_rate": 0.0002,
+      "loss": 0.5014,
+      "step": 5630
+    },
+    {
+      "epoch": 4.09437386569873,
+      "grad_norm": 1.229573369026184,
+      "learning_rate": 0.0002,
+      "loss": 0.513,
+      "step": 5640
+    },
+    {
+      "epoch": 4.101633393829401,
+      "grad_norm": 1.7182831764221191,
+      "learning_rate": 0.0002,
+      "loss": 0.5258,
+      "step": 5650
+    },
+    {
+      "epoch": 4.108892921960073,
+      "grad_norm": 0.894903302192688,
+      "learning_rate": 0.0002,
+      "loss": 0.5371,
+      "step": 5660
+    },
+    {
+      "epoch": 4.116152450090744,
+      "grad_norm": 0.8754552006721497,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 5670
+    },
+    {
+      "epoch": 4.123411978221416,
+      "grad_norm": 1.2401553392410278,
+      "learning_rate": 0.0002,
+      "loss": 0.491,
+      "step": 5680
+    },
+    {
+      "epoch": 4.130671506352087,
+      "grad_norm": 0.8631148934364319,
+      "learning_rate": 0.0002,
+      "loss": 0.4549,
+      "step": 5690
+    },
+    {
+      "epoch": 4.137931034482759,
+      "grad_norm": 1.1798022985458374,
+      "learning_rate": 0.0002,
+      "loss": 0.487,
+      "step": 5700
+    },
+    {
+      "epoch": 4.14519056261343,
+      "grad_norm": 0.8344549536705017,
+      "learning_rate": 0.0002,
+      "loss": 0.4522,
+      "step": 5710
+    },
+    {
+      "epoch": 4.152450090744102,
+      "grad_norm": 1.2342697381973267,
+      "learning_rate": 0.0002,
+      "loss": 0.4559,
+      "step": 5720
+    },
+    {
+      "epoch": 4.159709618874773,
+      "grad_norm": 1.1601094007492065,
+      "learning_rate": 0.0002,
+      "loss": 0.53,
+      "step": 5730
+    },
+    {
+      "epoch": 4.166969147005445,
+      "grad_norm": 1.2925703525543213,
+      "learning_rate": 0.0002,
+      "loss": 0.4755,
+      "step": 5740
+    },
+    {
+      "epoch": 4.174228675136116,
+      "grad_norm": 1.0870997905731201,
+      "learning_rate": 0.0002,
+      "loss": 0.4973,
+      "step": 5750
+    },
+    {
+      "epoch": 4.181488203266787,
+      "grad_norm": 0.9077792763710022,
+      "learning_rate": 0.0002,
+      "loss": 0.5184,
+      "step": 5760
+    },
+    {
+      "epoch": 4.188747731397459,
+      "grad_norm": 1.009273886680603,
+      "learning_rate": 0.0002,
+      "loss": 0.4606,
+      "step": 5770
+    },
+    {
+      "epoch": 4.19600725952813,
+      "grad_norm": 1.2465530633926392,
+      "learning_rate": 0.0002,
+      "loss": 0.5383,
+      "step": 5780
+    },
+    {
+      "epoch": 4.203266787658802,
+      "grad_norm": 1.2261253595352173,
+      "learning_rate": 0.0002,
+      "loss": 0.4938,
+      "step": 5790
+    },
+    {
+      "epoch": 4.2105263157894735,
+      "grad_norm": 1.1498041152954102,
+      "learning_rate": 0.0002,
+      "loss": 0.5255,
+      "step": 5800
+    },
+    {
+      "epoch": 4.217785843920145,
+      "grad_norm": 1.1966725587844849,
+      "learning_rate": 0.0002,
+      "loss": 0.5269,
+      "step": 5810
+    },
+    {
+      "epoch": 4.2250453720508165,
+      "grad_norm": 1.2651296854019165,
+      "learning_rate": 0.0002,
+      "loss": 0.5626,
+      "step": 5820
+    },
+    {
+      "epoch": 4.2323049001814885,
+      "grad_norm": 1.0388574600219727,
+      "learning_rate": 0.0002,
+      "loss": 0.5213,
+      "step": 5830
+    },
+    {
+      "epoch": 4.23956442831216,
+      "grad_norm": 1.3042771816253662,
+      "learning_rate": 0.0002,
+      "loss": 0.4965,
+      "step": 5840
+    },
+    {
+      "epoch": 4.246823956442832,
+      "grad_norm": 1.1127727031707764,
+      "learning_rate": 0.0002,
+      "loss": 0.5116,
+      "step": 5850
+    },
+    {
+      "epoch": 4.254083484573503,
+      "grad_norm": 0.9653958082199097,
+      "learning_rate": 0.0002,
+      "loss": 0.5197,
+      "step": 5860
+    },
+    {
+      "epoch": 4.261343012704174,
+      "grad_norm": 1.0500504970550537,
+      "learning_rate": 0.0002,
+      "loss": 0.4414,
+      "step": 5870
+    },
+    {
+      "epoch": 4.268602540834846,
+      "grad_norm": 1.1476165056228638,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 5880
+    },
+    {
+      "epoch": 4.275862068965517,
+      "grad_norm": 0.9424414038658142,
+      "learning_rate": 0.0002,
+      "loss": 0.4667,
+      "step": 5890
+    },
+    {
+      "epoch": 4.283121597096189,
+      "grad_norm": 1.3309166431427002,
+      "learning_rate": 0.0002,
+      "loss": 0.5039,
+      "step": 5900
+    },
+    {
+      "epoch": 4.29038112522686,
+      "grad_norm": 1.3025873899459839,
+      "learning_rate": 0.0002,
+      "loss": 0.5472,
+      "step": 5910
+    },
+    {
+      "epoch": 4.297640653357532,
+      "grad_norm": 1.1442325115203857,
+      "learning_rate": 0.0002,
+      "loss": 0.4644,
+      "step": 5920
+    },
+    {
+      "epoch": 4.304900181488203,
+      "grad_norm": 0.9820859432220459,
+      "learning_rate": 0.0002,
+      "loss": 0.5066,
+      "step": 5930
+    },
+    {
+      "epoch": 4.312159709618875,
+      "grad_norm": 0.9615740180015564,
+      "learning_rate": 0.0002,
+      "loss": 0.5224,
+      "step": 5940
+    },
+    {
+      "epoch": 4.319419237749546,
+      "grad_norm": 1.1627109050750732,
+      "learning_rate": 0.0002,
+      "loss": 0.5665,
+      "step": 5950
+    },
+    {
+      "epoch": 4.326678765880218,
+      "grad_norm": 0.9381322860717773,
+      "learning_rate": 0.0002,
+      "loss": 0.4852,
+      "step": 5960
+    },
+    {
+      "epoch": 4.333938294010889,
+      "grad_norm": 0.8154335618019104,
+      "learning_rate": 0.0002,
+      "loss": 0.4532,
+      "step": 5970
+    },
+    {
+      "epoch": 4.341197822141561,
+      "grad_norm": 0.877671480178833,
+      "learning_rate": 0.0002,
+      "loss": 0.5237,
+      "step": 5980
+    },
+    {
+      "epoch": 4.348457350272232,
+      "grad_norm": 1.1742031574249268,
+      "learning_rate": 0.0002,
+      "loss": 0.6113,
+      "step": 5990
+    },
+    {
+      "epoch": 4.3557168784029034,
+      "grad_norm": 1.0352917909622192,
+      "learning_rate": 0.0002,
+      "loss": 0.5704,
+      "step": 6000
+    },
+    {
+      "epoch": 4.362976406533575,
+      "grad_norm": 0.9963878989219666,
+      "learning_rate": 0.0002,
+      "loss": 0.4996,
+      "step": 6010
+    },
+    {
+      "epoch": 4.3702359346642465,
+      "grad_norm": 1.1892237663269043,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 6020
+    },
+    {
+      "epoch": 4.3774954627949185,
+      "grad_norm": 1.2516111135482788,
+      "learning_rate": 0.0002,
+      "loss": 0.5224,
+      "step": 6030
+    },
+    {
+      "epoch": 4.38475499092559,
+      "grad_norm": 1.2111951112747192,
+      "learning_rate": 0.0002,
+      "loss": 0.5952,
+      "step": 6040
+    },
+    {
+      "epoch": 4.392014519056262,
+      "grad_norm": 1.0820083618164062,
+      "learning_rate": 0.0002,
+      "loss": 0.4275,
+      "step": 6050
+    },
+    {
+      "epoch": 4.399274047186933,
+      "grad_norm": 1.033915638923645,
+      "learning_rate": 0.0002,
+      "loss": 0.5117,
+      "step": 6060
+    },
+    {
+      "epoch": 4.406533575317605,
+      "grad_norm": 1.0635870695114136,
+      "learning_rate": 0.0002,
+      "loss": 0.5431,
+      "step": 6070
+    },
+    {
+      "epoch": 4.413793103448276,
+      "grad_norm": 1.0520414113998413,
+      "learning_rate": 0.0002,
+      "loss": 0.5341,
+      "step": 6080
+    },
+    {
+      "epoch": 4.421052631578947,
+      "grad_norm": 1.0821926593780518,
+      "learning_rate": 0.0002,
+      "loss": 0.512,
+      "step": 6090
+    },
+    {
+      "epoch": 4.428312159709619,
+      "grad_norm": 1.0533246994018555,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 6100
+    },
+    {
+      "epoch": 4.43557168784029,
+      "grad_norm": 0.9231932759284973,
+      "learning_rate": 0.0002,
+      "loss": 0.4577,
+      "step": 6110
+    },
+    {
+      "epoch": 4.442831215970962,
+      "grad_norm": 0.9910260438919067,
+      "learning_rate": 0.0002,
+      "loss": 0.583,
+      "step": 6120
+    },
+    {
+      "epoch": 4.450090744101633,
+      "grad_norm": 1.061949372291565,
+      "learning_rate": 0.0002,
+      "loss": 0.4717,
+      "step": 6130
+    },
+    {
+      "epoch": 4.457350272232305,
+      "grad_norm": 1.2927039861679077,
+      "learning_rate": 0.0002,
+      "loss": 0.5893,
+      "step": 6140
+    },
+    {
+      "epoch": 4.464609800362976,
+      "grad_norm": 1.3966081142425537,
+      "learning_rate": 0.0002,
+      "loss": 0.4684,
+      "step": 6150
+    },
+    {
+      "epoch": 4.471869328493648,
+      "grad_norm": 1.3835992813110352,
+      "learning_rate": 0.0002,
+      "loss": 0.5507,
+      "step": 6160
+    },
+    {
+      "epoch": 4.479128856624319,
+      "grad_norm": 1.0892692804336548,
+      "learning_rate": 0.0002,
+      "loss": 0.5911,
+      "step": 6170
+    },
+    {
+      "epoch": 4.486388384754991,
+      "grad_norm": 1.0318800210952759,
+      "learning_rate": 0.0002,
+      "loss": 0.478,
+      "step": 6180
+    },
+    {
+      "epoch": 4.493647912885662,
+      "grad_norm": 0.8174677491188049,
+      "learning_rate": 0.0002,
+      "loss": 0.5198,
+      "step": 6190
+    },
+    {
+      "epoch": 4.500907441016334,
+      "grad_norm": 1.4157509803771973,
+      "learning_rate": 0.0002,
+      "loss": 0.5387,
+      "step": 6200
+    },
+    {
+      "epoch": 4.508166969147005,
+      "grad_norm": 1.5244114398956299,
+      "learning_rate": 0.0002,
+      "loss": 0.5868,
+      "step": 6210
+    },
+    {
+      "epoch": 4.5154264972776765,
+      "grad_norm": 0.8164850473403931,
+      "learning_rate": 0.0002,
+      "loss": 0.4642,
+      "step": 6220
+    },
+    {
+      "epoch": 4.5226860254083485,
+      "grad_norm": 1.2904746532440186,
+      "learning_rate": 0.0002,
+      "loss": 0.522,
+      "step": 6230
+    },
+    {
+      "epoch": 4.52994555353902,
+      "grad_norm": 0.7987732887268066,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 6240
+    },
+    {
+      "epoch": 4.537205081669692,
+      "grad_norm": 0.831040620803833,
+      "learning_rate": 0.0002,
+      "loss": 0.4615,
+      "step": 6250
+    },
+    {
+      "epoch": 4.544464609800363,
+      "grad_norm": 0.9545485973358154,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 6260
+    },
+    {
+      "epoch": 4.551724137931035,
+      "grad_norm": 0.9291793704032898,
+      "learning_rate": 0.0002,
+      "loss": 0.5515,
+      "step": 6270
+    },
+    {
+      "epoch": 4.558983666061706,
+      "grad_norm": 0.8977208733558655,
+      "learning_rate": 0.0002,
+      "loss": 0.4535,
+      "step": 6280
+    },
+    {
+      "epoch": 4.566243194192378,
+      "grad_norm": 1.1768537759780884,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 6290
+    },
+    {
+      "epoch": 4.573502722323049,
+      "grad_norm": 1.0688952207565308,
+      "learning_rate": 0.0002,
+      "loss": 0.5925,
+      "step": 6300
+    },
+    {
+      "epoch": 4.580762250453721,
+      "grad_norm": 0.8800966739654541,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 6310
+    },
+    {
+      "epoch": 4.588021778584392,
+      "grad_norm": 1.0911834239959717,
+      "learning_rate": 0.0002,
+      "loss": 0.6106,
+      "step": 6320
+    },
+    {
+      "epoch": 4.595281306715064,
+      "grad_norm": 1.1420872211456299,
+      "learning_rate": 0.0002,
+      "loss": 0.5109,
+      "step": 6330
+    },
+    {
+      "epoch": 4.602540834845735,
+      "grad_norm": 1.0215224027633667,
+      "learning_rate": 0.0002,
+      "loss": 0.5147,
+      "step": 6340
+    },
+    {
+      "epoch": 4.609800362976406,
+      "grad_norm": 0.9685489535331726,
+      "learning_rate": 0.0002,
+      "loss": 0.592,
+      "step": 6350
+    },
+    {
+      "epoch": 4.617059891107078,
+      "grad_norm": 1.12773597240448,
+      "learning_rate": 0.0002,
+      "loss": 0.5775,
+      "step": 6360
+    },
+    {
+      "epoch": 4.624319419237749,
+      "grad_norm": 1.0663973093032837,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 6370
+    },
+    {
+      "epoch": 4.631578947368421,
+      "grad_norm": 1.1707262992858887,
+      "learning_rate": 0.0002,
+      "loss": 0.512,
+      "step": 6380
+    },
+    {
+      "epoch": 4.638838475499092,
+      "grad_norm": 1.0672980546951294,
+      "learning_rate": 0.0002,
+      "loss": 0.5497,
+      "step": 6390
+    },
+    {
+      "epoch": 4.646098003629764,
+      "grad_norm": 1.1464333534240723,
+      "learning_rate": 0.0002,
+      "loss": 0.5699,
+      "step": 6400
+    },
+    {
+      "epoch": 4.653357531760435,
+      "grad_norm": 1.070230484008789,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 6410
+    },
+    {
+      "epoch": 4.660617059891107,
+      "grad_norm": 0.9673764109611511,
+      "learning_rate": 0.0002,
+      "loss": 0.5013,
+      "step": 6420
+    },
+    {
+      "epoch": 4.6678765880217785,
+      "grad_norm": 1.0189043283462524,
+      "learning_rate": 0.0002,
+      "loss": 0.5901,
+      "step": 6430
+    },
+    {
+      "epoch": 4.67513611615245,
+      "grad_norm": 1.185896396636963,
+      "learning_rate": 0.0002,
+      "loss": 0.5193,
+      "step": 6440
+    },
+    {
+      "epoch": 4.682395644283122,
+      "grad_norm": 1.0682812929153442,
+      "learning_rate": 0.0002,
+      "loss": 0.5318,
+      "step": 6450
+    },
+    {
+      "epoch": 4.689655172413794,
+      "grad_norm": 1.3586071729660034,
+      "learning_rate": 0.0002,
+      "loss": 0.5773,
+      "step": 6460
+    },
+    {
+      "epoch": 4.696914700544465,
+      "grad_norm": 0.6561792492866516,
+      "learning_rate": 0.0002,
+      "loss": 0.5482,
+      "step": 6470
+    },
+    {
+      "epoch": 4.704174228675136,
+      "grad_norm": 1.1394113302230835,
+      "learning_rate": 0.0002,
+      "loss": 0.5711,
+      "step": 6480
+    },
+    {
+      "epoch": 4.711433756805808,
+      "grad_norm": 0.9683151245117188,
+      "learning_rate": 0.0002,
+      "loss": 0.5325,
+      "step": 6490
+    },
+    {
+      "epoch": 4.718693284936479,
+      "grad_norm": 1.0247553586959839,
+      "learning_rate": 0.0002,
+      "loss": 0.5139,
+      "step": 6500
+    },
+    {
+      "epoch": 4.725952813067151,
+      "grad_norm": 0.8046169281005859,
+      "learning_rate": 0.0002,
+      "loss": 0.5794,
+      "step": 6510
+    },
+    {
+      "epoch": 4.733212341197822,
+      "grad_norm": 1.0710240602493286,
+      "learning_rate": 0.0002,
+      "loss": 0.5471,
+      "step": 6520
+    },
+    {
+      "epoch": 4.740471869328494,
+      "grad_norm": 0.9438924193382263,
+      "learning_rate": 0.0002,
+      "loss": 0.4805,
+      "step": 6530
+    },
+    {
+      "epoch": 4.747731397459165,
+      "grad_norm": 0.869162380695343,
+      "learning_rate": 0.0002,
+      "loss": 0.5404,
+      "step": 6540
+    },
+    {
+      "epoch": 4.754990925589837,
+      "grad_norm": 0.9776787161827087,
+      "learning_rate": 0.0002,
+      "loss": 0.6379,
+      "step": 6550
+    },
+    {
+      "epoch": 4.762250453720508,
+      "grad_norm": 1.1990505456924438,
+      "learning_rate": 0.0002,
+      "loss": 0.5288,
+      "step": 6560
+    },
+    {
+      "epoch": 4.769509981851179,
+      "grad_norm": 1.0582209825515747,
+      "learning_rate": 0.0002,
+      "loss": 0.5539,
+      "step": 6570
+    },
+    {
+      "epoch": 4.776769509981851,
+      "grad_norm": 0.9966367483139038,
+      "learning_rate": 0.0002,
+      "loss": 0.489,
+      "step": 6580
+    },
+    {
+      "epoch": 4.784029038112522,
+      "grad_norm": 0.9130612015724182,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 6590
+    },
+    {
+      "epoch": 4.791288566243194,
+      "grad_norm": 1.0950500965118408,
+      "learning_rate": 0.0002,
+      "loss": 0.5864,
+      "step": 6600
+    },
+    {
+      "epoch": 4.798548094373865,
+      "grad_norm": 1.108681321144104,
+      "learning_rate": 0.0002,
+      "loss": 0.5266,
+      "step": 6610
+    },
+    {
+      "epoch": 4.805807622504537,
+      "grad_norm": 1.1873763799667358,
+      "learning_rate": 0.0002,
+      "loss": 0.5875,
+      "step": 6620
+    },
+    {
+      "epoch": 4.8130671506352085,
+      "grad_norm": 1.305367112159729,
+      "learning_rate": 0.0002,
+      "loss": 0.5736,
+      "step": 6630
+    },
+    {
+      "epoch": 4.8203266787658805,
+      "grad_norm": 1.2801482677459717,
+      "learning_rate": 0.0002,
+      "loss": 0.5636,
+      "step": 6640
+    },
+    {
+      "epoch": 4.827586206896552,
+      "grad_norm": 1.26764976978302,
+      "learning_rate": 0.0002,
+      "loss": 0.582,
+      "step": 6650
+    },
+    {
+      "epoch": 4.834845735027224,
+      "grad_norm": 1.0018208026885986,
+      "learning_rate": 0.0002,
+      "loss": 0.5259,
+      "step": 6660
+    },
+    {
+      "epoch": 4.842105263157895,
+      "grad_norm": 1.2326326370239258,
+      "learning_rate": 0.0002,
+      "loss": 0.548,
+      "step": 6670
+    },
+    {
+      "epoch": 4.849364791288567,
+      "grad_norm": 0.9707282781600952,
+      "learning_rate": 0.0002,
+      "loss": 0.5933,
+      "step": 6680
+    },
+    {
+      "epoch": 4.856624319419238,
+      "grad_norm": 1.2772048711776733,
+      "learning_rate": 0.0002,
+      "loss": 0.5612,
+      "step": 6690
+    },
+    {
+      "epoch": 4.863883847549909,
+      "grad_norm": 2.6652262210845947,
+      "learning_rate": 0.0002,
+      "loss": 0.5346,
+      "step": 6700
+    },
+    {
+      "epoch": 4.871143375680581,
+      "grad_norm": 1.215828537940979,
+      "learning_rate": 0.0002,
+      "loss": 0.5428,
+      "step": 6710
+    },
+    {
+      "epoch": 4.878402903811252,
+      "grad_norm": 1.3704510927200317,
+      "learning_rate": 0.0002,
+      "loss": 0.6571,
+      "step": 6720
+    },
+    {
+      "epoch": 4.885662431941924,
+      "grad_norm": 0.7781757116317749,
+      "learning_rate": 0.0002,
+      "loss": 0.4963,
+      "step": 6730
+    },
+    {
+      "epoch": 4.892921960072595,
+      "grad_norm": 1.1883646249771118,
+      "learning_rate": 0.0002,
+      "loss": 0.5989,
+      "step": 6740
+    },
+    {
+      "epoch": 4.900181488203267,
+      "grad_norm": 0.9216066002845764,
+      "learning_rate": 0.0002,
+      "loss": 0.6067,
+      "step": 6750
+    },
+    {
+      "epoch": 4.907441016333938,
+      "grad_norm": 1.0558464527130127,
+      "learning_rate": 0.0002,
+      "loss": 0.5085,
+      "step": 6760
+    },
+    {
+      "epoch": 4.91470054446461,
+      "grad_norm": 1.032656192779541,
+      "learning_rate": 0.0002,
+      "loss": 0.5216,
+      "step": 6770
+    },
+    {
+      "epoch": 4.921960072595281,
+      "grad_norm": 1.1261441707611084,
+      "learning_rate": 0.0002,
+      "loss": 0.5426,
+      "step": 6780
+    },
+    {
+      "epoch": 4.929219600725952,
+      "grad_norm": 1.2178640365600586,
+      "learning_rate": 0.0002,
+      "loss": 0.5295,
+      "step": 6790
+    },
+    {
+      "epoch": 4.936479128856624,
+      "grad_norm": 1.5369361639022827,
+      "learning_rate": 0.0002,
+      "loss": 0.5476,
+      "step": 6800
+    },
+    {
+      "epoch": 4.943738656987296,
+      "grad_norm": 1.1188377141952515,
+      "learning_rate": 0.0002,
+      "loss": 0.5358,
+      "step": 6810
+    },
+    {
+      "epoch": 4.950998185117967,
+      "grad_norm": 1.2506113052368164,
+      "learning_rate": 0.0002,
+      "loss": 0.5483,
+      "step": 6820
+    },
+    {
+      "epoch": 4.9582577132486385,
+      "grad_norm": 0.8776047825813293,
+      "learning_rate": 0.0002,
+      "loss": 0.567,
+      "step": 6830
+    },
+    {
+      "epoch": 4.9655172413793105,
+      "grad_norm": 0.9700555205345154,
+      "learning_rate": 0.0002,
+      "loss": 0.5764,
+      "step": 6840
+    },
+    {
+      "epoch": 4.972776769509982,
+      "grad_norm": 1.2713534832000732,
+      "learning_rate": 0.0002,
+      "loss": 0.5396,
+      "step": 6850
+    },
+    {
+      "epoch": 4.980036297640654,
+      "grad_norm": 0.9855955243110657,
+      "learning_rate": 0.0002,
+      "loss": 0.5451,
+      "step": 6860
+    },
+    {
+      "epoch": 4.987295825771325,
+      "grad_norm": 0.8734853863716125,
+      "learning_rate": 0.0002,
+      "loss": 0.5884,
+      "step": 6870
+    },
+    {
+      "epoch": 4.994555353901997,
+      "grad_norm": 0.8065403699874878,
+      "learning_rate": 0.0002,
+      "loss": 0.5189,
+      "step": 6880
+    },
+    {
+      "epoch": 4.999637023593467,
+      "eval_loss": 1.3302682638168335,
+      "eval_runtime": 46.2496,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 6887
+    },
+    {
+      "epoch": 5.001814882032668,
+      "grad_norm": 0.5163813829421997,
+      "learning_rate": 0.0002,
+      "loss": 0.4889,
+      "step": 6890
+    },
+    {
+      "epoch": 5.00907441016334,
+      "grad_norm": 1.1496137380599976,
+      "learning_rate": 0.0002,
+      "loss": 0.3545,
+      "step": 6900
+    },
+    {
+      "epoch": 5.016333938294011,
+      "grad_norm": 1.0133885145187378,
+      "learning_rate": 0.0002,
+      "loss": 0.39,
+      "step": 6910
+    },
+    {
+      "epoch": 5.023593466424682,
+      "grad_norm": 0.9479621052742004,
+      "learning_rate": 0.0002,
+      "loss": 0.3693,
+      "step": 6920
+    },
+    {
+      "epoch": 5.030852994555354,
+      "grad_norm": 0.8587583303451538,
+      "learning_rate": 0.0002,
+      "loss": 0.4012,
+      "step": 6930
+    },
+    {
+      "epoch": 5.038112522686025,
+      "grad_norm": 1.3314697742462158,
+      "learning_rate": 0.0002,
+      "loss": 0.3428,
+      "step": 6940
+    },
+    {
+      "epoch": 5.045372050816697,
+      "grad_norm": 1.195448875427246,
+      "learning_rate": 0.0002,
+      "loss": 0.3909,
+      "step": 6950
+    },
+    {
+      "epoch": 5.052631578947368,
+      "grad_norm": 1.2482256889343262,
+      "learning_rate": 0.0002,
+      "loss": 0.3322,
+      "step": 6960
+    },
+    {
+      "epoch": 5.05989110707804,
+      "grad_norm": 1.2011528015136719,
+      "learning_rate": 0.0002,
+      "loss": 0.3893,
+      "step": 6970
+    },
+    {
+      "epoch": 5.067150635208711,
+      "grad_norm": 1.3997188806533813,
+      "learning_rate": 0.0002,
+      "loss": 0.3265,
+      "step": 6980
+    },
+    {
+      "epoch": 5.074410163339383,
+      "grad_norm": 1.2147513628005981,
+      "learning_rate": 0.0002,
+      "loss": 0.3716,
+      "step": 6990
+    },
+    {
+      "epoch": 5.081669691470054,
+      "grad_norm": 1.6030137538909912,
+      "learning_rate": 0.0002,
+      "loss": 0.4053,
+      "step": 7000
+    },
+    {
+      "epoch": 5.088929219600726,
+      "grad_norm": 0.9466970562934875,
+      "learning_rate": 0.0002,
+      "loss": 0.3665,
+      "step": 7010
+    },
+    {
+      "epoch": 5.096188747731397,
+      "grad_norm": 1.4593411684036255,
+      "learning_rate": 0.0002,
+      "loss": 0.3451,
+      "step": 7020
+    },
+    {
+      "epoch": 5.103448275862069,
+      "grad_norm": 1.2196033000946045,
+      "learning_rate": 0.0002,
+      "loss": 0.3843,
+      "step": 7030
+    },
+    {
+      "epoch": 5.1107078039927405,
+      "grad_norm": 1.1341328620910645,
+      "learning_rate": 0.0002,
+      "loss": 0.3896,
+      "step": 7040
+    },
+    {
+      "epoch": 5.117967332123412,
+      "grad_norm": 1.2248976230621338,
+      "learning_rate": 0.0002,
+      "loss": 0.3627,
+      "step": 7050
+    },
+    {
+      "epoch": 5.125226860254084,
+      "grad_norm": 1.1620593070983887,
+      "learning_rate": 0.0002,
+      "loss": 0.3784,
+      "step": 7060
+    },
+    {
+      "epoch": 5.132486388384755,
+      "grad_norm": 0.9300723671913147,
+      "learning_rate": 0.0002,
+      "loss": 0.3678,
+      "step": 7070
+    },
+    {
+      "epoch": 5.139745916515427,
+      "grad_norm": 1.2265169620513916,
+      "learning_rate": 0.0002,
+      "loss": 0.3756,
+      "step": 7080
+    },
+    {
+      "epoch": 5.147005444646098,
+      "grad_norm": 1.4430373907089233,
+      "learning_rate": 0.0002,
+      "loss": 0.3595,
+      "step": 7090
+    },
+    {
+      "epoch": 5.15426497277677,
+      "grad_norm": 1.0821576118469238,
+      "learning_rate": 0.0002,
+      "loss": 0.3788,
+      "step": 7100
+    },
+    {
+      "epoch": 5.161524500907441,
+      "grad_norm": 1.2574739456176758,
+      "learning_rate": 0.0002,
+      "loss": 0.383,
+      "step": 7110
+    },
+    {
+      "epoch": 5.168784029038113,
+      "grad_norm": 1.1806069612503052,
+      "learning_rate": 0.0002,
+      "loss": 0.3692,
+      "step": 7120
+    },
+    {
+      "epoch": 5.176043557168784,
+      "grad_norm": 0.9900956153869629,
+      "learning_rate": 0.0002,
+      "loss": 0.3978,
+      "step": 7130
+    },
+    {
+      "epoch": 5.183303085299456,
+      "grad_norm": 1.2414425611495972,
+      "learning_rate": 0.0002,
+      "loss": 0.4358,
+      "step": 7140
+    },
+    {
+      "epoch": 5.190562613430127,
+      "grad_norm": 0.8220699429512024,
+      "learning_rate": 0.0002,
+      "loss": 0.3485,
+      "step": 7150
+    },
+    {
+      "epoch": 5.197822141560798,
+      "grad_norm": 1.29408860206604,
+      "learning_rate": 0.0002,
+      "loss": 0.3517,
+      "step": 7160
+    },
+    {
+      "epoch": 5.20508166969147,
+      "grad_norm": 0.8510639071464539,
+      "learning_rate": 0.0002,
+      "loss": 0.3405,
+      "step": 7170
+    },
+    {
+      "epoch": 5.212341197822141,
+      "grad_norm": 1.3448902368545532,
+      "learning_rate": 0.0002,
+      "loss": 0.4233,
+      "step": 7180
+    },
+    {
+      "epoch": 5.219600725952813,
+      "grad_norm": 1.054451584815979,
+      "learning_rate": 0.0002,
+      "loss": 0.3808,
+      "step": 7190
+    },
+    {
+      "epoch": 5.226860254083484,
+      "grad_norm": 1.3752713203430176,
+      "learning_rate": 0.0002,
+      "loss": 0.368,
+      "step": 7200
+    },
+    {
+      "epoch": 5.234119782214156,
+      "grad_norm": 1.4848095178604126,
+      "learning_rate": 0.0002,
+      "loss": 0.3844,
+      "step": 7210
+    },
+    {
+      "epoch": 5.241379310344827,
+      "grad_norm": 1.428842544555664,
+      "learning_rate": 0.0002,
+      "loss": 0.4187,
+      "step": 7220
+    },
+    {
+      "epoch": 5.248638838475499,
+      "grad_norm": 1.1703591346740723,
+      "learning_rate": 0.0002,
+      "loss": 0.3778,
+      "step": 7230
+    },
+    {
+      "epoch": 5.2558983666061705,
+      "grad_norm": 1.2180451154708862,
+      "learning_rate": 0.0002,
+      "loss": 0.417,
+      "step": 7240
+    },
+    {
+      "epoch": 5.2631578947368425,
+      "grad_norm": 1.094045877456665,
+      "learning_rate": 0.0002,
+      "loss": 0.3656,
+      "step": 7250
+    },
+    {
+      "epoch": 5.270417422867514,
+      "grad_norm": 0.9545766115188599,
+      "learning_rate": 0.0002,
+      "loss": 0.4331,
+      "step": 7260
+    },
+    {
+      "epoch": 5.277676950998185,
+      "grad_norm": 0.8356652855873108,
+      "learning_rate": 0.0002,
+      "loss": 0.3642,
+      "step": 7270
+    },
+    {
+      "epoch": 5.284936479128857,
+      "grad_norm": 1.148160457611084,
+      "learning_rate": 0.0002,
+      "loss": 0.3576,
+      "step": 7280
+    },
+    {
+      "epoch": 5.292196007259528,
+      "grad_norm": 1.2009977102279663,
+      "learning_rate": 0.0002,
+      "loss": 0.4178,
+      "step": 7290
+    },
+    {
+      "epoch": 5.2994555353902,
+      "grad_norm": 1.3283873796463013,
+      "learning_rate": 0.0002,
+      "loss": 0.3977,
+      "step": 7300
+    },
+    {
+      "epoch": 5.306715063520871,
+      "grad_norm": 0.9850481748580933,
+      "learning_rate": 0.0002,
+      "loss": 0.3853,
+      "step": 7310
+    },
+    {
+      "epoch": 5.313974591651543,
+      "grad_norm": 1.367550015449524,
+      "learning_rate": 0.0002,
+      "loss": 0.3645,
+      "step": 7320
+    },
+    {
+      "epoch": 5.321234119782214,
+      "grad_norm": 0.8602936863899231,
+      "learning_rate": 0.0002,
+      "loss": 0.3898,
+      "step": 7330
+    },
+    {
+      "epoch": 5.328493647912886,
+      "grad_norm": 1.1130679845809937,
+      "learning_rate": 0.0002,
+      "loss": 0.4173,
+      "step": 7340
+    },
+    {
+      "epoch": 5.335753176043557,
+      "grad_norm": 1.3002253770828247,
+      "learning_rate": 0.0002,
+      "loss": 0.3642,
+      "step": 7350
+    },
+    {
+      "epoch": 5.343012704174229,
+      "grad_norm": 1.6235289573669434,
+      "learning_rate": 0.0002,
+      "loss": 0.4138,
+      "step": 7360
+    },
+    {
+      "epoch": 5.3502722323049,
+      "grad_norm": 1.156379222869873,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 7370
+    },
+    {
+      "epoch": 5.357531760435572,
+      "grad_norm": 1.0569308996200562,
+      "learning_rate": 0.0002,
+      "loss": 0.3222,
+      "step": 7380
+    },
+    {
+      "epoch": 5.364791288566243,
+      "grad_norm": 1.6674021482467651,
+      "learning_rate": 0.0002,
+      "loss": 0.3573,
+      "step": 7390
+    },
+    {
+      "epoch": 5.372050816696914,
+      "grad_norm": 1.2962018251419067,
+      "learning_rate": 0.0002,
+      "loss": 0.4325,
+      "step": 7400
+    },
+    {
+      "epoch": 5.379310344827586,
+      "grad_norm": 1.1904195547103882,
+      "learning_rate": 0.0002,
+      "loss": 0.3809,
+      "step": 7410
+    },
+    {
+      "epoch": 5.386569872958257,
+      "grad_norm": 1.316245675086975,
+      "learning_rate": 0.0002,
+      "loss": 0.3728,
+      "step": 7420
+    },
+    {
+      "epoch": 5.393829401088929,
+      "grad_norm": 1.127570390701294,
+      "learning_rate": 0.0002,
+      "loss": 0.4096,
+      "step": 7430
+    },
+    {
+      "epoch": 5.4010889292196005,
+      "grad_norm": 1.3895777463912964,
+      "learning_rate": 0.0002,
+      "loss": 0.3933,
+      "step": 7440
+    },
+    {
+      "epoch": 5.4083484573502725,
+      "grad_norm": 1.626830816268921,
+      "learning_rate": 0.0002,
+      "loss": 0.4085,
+      "step": 7450
+    },
+    {
+      "epoch": 5.415607985480944,
+      "grad_norm": 1.3703926801681519,
+      "learning_rate": 0.0002,
+      "loss": 0.4186,
+      "step": 7460
+    },
+    {
+      "epoch": 5.422867513611616,
+      "grad_norm": 1.3854840993881226,
+      "learning_rate": 0.0002,
+      "loss": 0.3517,
+      "step": 7470
+    },
+    {
+      "epoch": 5.430127041742287,
+      "grad_norm": 1.107065200805664,
+      "learning_rate": 0.0002,
+      "loss": 0.3714,
+      "step": 7480
+    },
+    {
+      "epoch": 5.437386569872959,
+      "grad_norm": 0.7843456268310547,
+      "learning_rate": 0.0002,
+      "loss": 0.3855,
+      "step": 7490
+    },
+    {
+      "epoch": 5.44464609800363,
+      "grad_norm": 1.6692372560501099,
+      "learning_rate": 0.0002,
+      "loss": 0.4159,
+      "step": 7500
+    },
+    {
+      "epoch": 5.451905626134302,
+      "grad_norm": 1.2583858966827393,
+      "learning_rate": 0.0002,
+      "loss": 0.4185,
+      "step": 7510
+    },
+    {
+      "epoch": 5.459165154264973,
+      "grad_norm": 1.6827000379562378,
+      "learning_rate": 0.0002,
+      "loss": 0.4401,
+      "step": 7520
+    },
+    {
+      "epoch": 5.466424682395644,
+      "grad_norm": 1.6680560111999512,
+      "learning_rate": 0.0002,
+      "loss": 0.397,
+      "step": 7530
+    },
+    {
+      "epoch": 5.473684210526316,
+      "grad_norm": 1.3696072101593018,
+      "learning_rate": 0.0002,
+      "loss": 0.4193,
+      "step": 7540
+    },
+    {
+      "epoch": 5.480943738656987,
+      "grad_norm": 1.4523496627807617,
+      "learning_rate": 0.0002,
+      "loss": 0.4244,
+      "step": 7550
+    },
+    {
+      "epoch": 5.488203266787659,
+      "grad_norm": 1.3432692289352417,
+      "learning_rate": 0.0002,
+      "loss": 0.3609,
+      "step": 7560
+    },
+    {
+      "epoch": 5.49546279491833,
+      "grad_norm": 1.363818645477295,
+      "learning_rate": 0.0002,
+      "loss": 0.3675,
+      "step": 7570
+    },
+    {
+      "epoch": 5.502722323049002,
+      "grad_norm": 1.0176721811294556,
+      "learning_rate": 0.0002,
+      "loss": 0.3726,
+      "step": 7580
+    },
+    {
+      "epoch": 5.509981851179673,
+      "grad_norm": 1.1625547409057617,
+      "learning_rate": 0.0002,
+      "loss": 0.3751,
+      "step": 7590
+    },
+    {
+      "epoch": 5.517241379310345,
+      "grad_norm": 1.2480388879776,
+      "learning_rate": 0.0002,
+      "loss": 0.433,
+      "step": 7600
+    },
+    {
+      "epoch": 5.524500907441016,
+      "grad_norm": 1.341509222984314,
+      "learning_rate": 0.0002,
+      "loss": 0.4511,
+      "step": 7610
+    },
+    {
+      "epoch": 5.531760435571687,
+      "grad_norm": 1.7048436403274536,
+      "learning_rate": 0.0002,
+      "loss": 0.4642,
+      "step": 7620
+    },
+    {
+      "epoch": 5.539019963702359,
+      "grad_norm": 1.1435480117797852,
+      "learning_rate": 0.0002,
+      "loss": 0.4509,
+      "step": 7630
+    },
+    {
+      "epoch": 5.5462794918330305,
+      "grad_norm": 1.2381842136383057,
+      "learning_rate": 0.0002,
+      "loss": 0.4528,
+      "step": 7640
+    },
+    {
+      "epoch": 5.5535390199637025,
+      "grad_norm": 1.50786292552948,
+      "learning_rate": 0.0002,
+      "loss": 0.4496,
+      "step": 7650
+    },
+    {
+      "epoch": 5.560798548094374,
+      "grad_norm": 1.2263519763946533,
+      "learning_rate": 0.0002,
+      "loss": 0.4242,
+      "step": 7660
+    },
+    {
+      "epoch": 5.568058076225046,
+      "grad_norm": 1.2864696979522705,
+      "learning_rate": 0.0002,
+      "loss": 0.418,
+      "step": 7670
+    },
+    {
+      "epoch": 5.575317604355717,
+      "grad_norm": 1.4443191289901733,
+      "learning_rate": 0.0002,
+      "loss": 0.3832,
+      "step": 7680
+    },
+    {
+      "epoch": 5.582577132486389,
+      "grad_norm": 1.3360971212387085,
+      "learning_rate": 0.0002,
+      "loss": 0.3964,
+      "step": 7690
+    },
+    {
+      "epoch": 5.58983666061706,
+      "grad_norm": 1.391828179359436,
+      "learning_rate": 0.0002,
+      "loss": 0.4639,
+      "step": 7700
+    },
+    {
+      "epoch": 5.597096188747732,
+      "grad_norm": 1.3699384927749634,
+      "learning_rate": 0.0002,
+      "loss": 0.4722,
+      "step": 7710
+    },
+    {
+      "epoch": 5.604355716878403,
+      "grad_norm": 1.3778468370437622,
+      "learning_rate": 0.0002,
+      "loss": 0.4302,
+      "step": 7720
+    },
+    {
+      "epoch": 5.611615245009075,
+      "grad_norm": 1.1009501218795776,
+      "learning_rate": 0.0002,
+      "loss": 0.4179,
+      "step": 7730
+    },
+    {
+      "epoch": 5.618874773139746,
+      "grad_norm": 1.0410021543502808,
+      "learning_rate": 0.0002,
+      "loss": 0.4104,
+      "step": 7740
+    },
+    {
+      "epoch": 5.626134301270417,
+      "grad_norm": 1.1012226343154907,
+      "learning_rate": 0.0002,
+      "loss": 0.4489,
+      "step": 7750
+    },
+    {
+      "epoch": 5.633393829401089,
+      "grad_norm": 1.3246384859085083,
+      "learning_rate": 0.0002,
+      "loss": 0.4544,
+      "step": 7760
+    },
+    {
+      "epoch": 5.64065335753176,
+      "grad_norm": 1.4301716089248657,
+      "learning_rate": 0.0002,
+      "loss": 0.4381,
+      "step": 7770
+    },
+    {
+      "epoch": 5.647912885662432,
+      "grad_norm": 1.1368978023529053,
+      "learning_rate": 0.0002,
+      "loss": 0.4297,
+      "step": 7780
+    },
+    {
+      "epoch": 5.655172413793103,
+      "grad_norm": 1.3493064641952515,
+      "learning_rate": 0.0002,
+      "loss": 0.4063,
+      "step": 7790
+    },
+    {
+      "epoch": 5.662431941923775,
+      "grad_norm": 1.3328721523284912,
+      "learning_rate": 0.0002,
+      "loss": 0.4562,
+      "step": 7800
+    },
+    {
+      "epoch": 5.669691470054446,
+      "grad_norm": 1.3235671520233154,
+      "learning_rate": 0.0002,
+      "loss": 0.4075,
+      "step": 7810
+    },
+    {
+      "epoch": 5.676950998185118,
+      "grad_norm": 1.1961841583251953,
+      "learning_rate": 0.0002,
+      "loss": 0.4589,
+      "step": 7820
+    },
+    {
+      "epoch": 5.684210526315789,
+      "grad_norm": 1.4189636707305908,
+      "learning_rate": 0.0002,
+      "loss": 0.4503,
+      "step": 7830
+    },
+    {
+      "epoch": 5.691470054446461,
+      "grad_norm": 1.3551312685012817,
+      "learning_rate": 0.0002,
+      "loss": 0.4452,
+      "step": 7840
+    },
+    {
+      "epoch": 5.6987295825771325,
+      "grad_norm": 1.449987769126892,
+      "learning_rate": 0.0002,
+      "loss": 0.4268,
+      "step": 7850
+    },
+    {
+      "epoch": 5.7059891107078045,
+      "grad_norm": 1.1225156784057617,
+      "learning_rate": 0.0002,
+      "loss": 0.4141,
+      "step": 7860
+    },
+    {
+      "epoch": 5.713248638838476,
+      "grad_norm": 1.4734594821929932,
+      "learning_rate": 0.0002,
+      "loss": 0.41,
+      "step": 7870
+    },
+    {
+      "epoch": 5.720508166969147,
+      "grad_norm": 1.3793359994888306,
+      "learning_rate": 0.0002,
+      "loss": 0.4013,
+      "step": 7880
+    },
+    {
+      "epoch": 5.727767695099819,
+      "grad_norm": 1.2431834936141968,
+      "learning_rate": 0.0002,
+      "loss": 0.4065,
+      "step": 7890
+    },
+    {
+      "epoch": 5.73502722323049,
+      "grad_norm": 1.1158313751220703,
+      "learning_rate": 0.0002,
+      "loss": 0.4595,
+      "step": 7900
+    },
+    {
+      "epoch": 5.742286751361162,
+      "grad_norm": 1.212248682975769,
+      "learning_rate": 0.0002,
+      "loss": 0.4342,
+      "step": 7910
+    },
+    {
+      "epoch": 5.749546279491833,
+      "grad_norm": 1.5259995460510254,
+      "learning_rate": 0.0002,
+      "loss": 0.4611,
+      "step": 7920
+    },
+    {
+      "epoch": 5.756805807622505,
+      "grad_norm": 1.3909121751785278,
+      "learning_rate": 0.0002,
+      "loss": 0.4483,
+      "step": 7930
+    },
+    {
+      "epoch": 5.764065335753176,
+      "grad_norm": 1.2511249780654907,
+      "learning_rate": 0.0002,
+      "loss": 0.4325,
+      "step": 7940
+    },
+    {
+      "epoch": 5.771324863883848,
+      "grad_norm": 1.2511906623840332,
+      "learning_rate": 0.0002,
+      "loss": 0.4048,
+      "step": 7950
+    },
+    {
+      "epoch": 5.778584392014519,
+      "grad_norm": 1.1489921808242798,
+      "learning_rate": 0.0002,
+      "loss": 0.3715,
+      "step": 7960
+    },
+    {
+      "epoch": 5.78584392014519,
+      "grad_norm": 1.028943419456482,
+      "learning_rate": 0.0002,
+      "loss": 0.4196,
+      "step": 7970
+    },
+    {
+      "epoch": 5.793103448275862,
+      "grad_norm": 1.0820423364639282,
+      "learning_rate": 0.0002,
+      "loss": 0.4334,
+      "step": 7980
+    },
+    {
+      "epoch": 5.800362976406533,
+      "grad_norm": 1.296520471572876,
+      "learning_rate": 0.0002,
+      "loss": 0.3917,
+      "step": 7990
+    },
+    {
+      "epoch": 5.807622504537205,
+      "grad_norm": 1.3597749471664429,
+      "learning_rate": 0.0002,
+      "loss": 0.4509,
+      "step": 8000
+    },
+    {
+      "epoch": 5.814882032667876,
+      "grad_norm": 0.8741790652275085,
+      "learning_rate": 0.0002,
+      "loss": 0.4535,
+      "step": 8010
+    },
+    {
+      "epoch": 5.822141560798548,
+      "grad_norm": 1.1471822261810303,
+      "learning_rate": 0.0002,
+      "loss": 0.4239,
+      "step": 8020
+    },
+    {
+      "epoch": 5.829401088929219,
+      "grad_norm": 1.2997334003448486,
+      "learning_rate": 0.0002,
+      "loss": 0.5042,
+      "step": 8030
+    },
+    {
+      "epoch": 5.836660617059891,
+      "grad_norm": 1.1027175188064575,
+      "learning_rate": 0.0002,
+      "loss": 0.4758,
+      "step": 8040
+    },
+    {
+      "epoch": 5.8439201451905625,
+      "grad_norm": 1.2695307731628418,
+      "learning_rate": 0.0002,
+      "loss": 0.4192,
+      "step": 8050
+    },
+    {
+      "epoch": 5.8511796733212345,
+      "grad_norm": 1.5275461673736572,
+      "learning_rate": 0.0002,
+      "loss": 0.5173,
+      "step": 8060
+    },
+    {
+      "epoch": 5.8584392014519056,
+      "grad_norm": 1.3059501647949219,
+      "learning_rate": 0.0002,
+      "loss": 0.5012,
+      "step": 8070
+    },
+    {
+      "epoch": 5.8656987295825775,
+      "grad_norm": 1.57442045211792,
+      "learning_rate": 0.0002,
+      "loss": 0.4425,
+      "step": 8080
+    },
+    {
+      "epoch": 5.872958257713249,
+      "grad_norm": 1.119564414024353,
+      "learning_rate": 0.0002,
+      "loss": 0.4261,
+      "step": 8090
+    },
+    {
+      "epoch": 5.88021778584392,
+      "grad_norm": 1.6517373323440552,
+      "learning_rate": 0.0002,
+      "loss": 0.465,
+      "step": 8100
+    },
+    {
+      "epoch": 5.887477313974592,
+      "grad_norm": 1.4093554019927979,
+      "learning_rate": 0.0002,
+      "loss": 0.4406,
+      "step": 8110
+    },
+    {
+      "epoch": 5.894736842105263,
+      "grad_norm": 1.278843641281128,
+      "learning_rate": 0.0002,
+      "loss": 0.4433,
+      "step": 8120
+    },
+    {
+      "epoch": 5.901996370235935,
+      "grad_norm": 1.2042944431304932,
+      "learning_rate": 0.0002,
+      "loss": 0.4007,
+      "step": 8130
+    },
+    {
+      "epoch": 5.909255898366606,
+      "grad_norm": 1.1788326501846313,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 8140
+    },
+    {
+      "epoch": 5.916515426497278,
+      "grad_norm": 1.4364569187164307,
+      "learning_rate": 0.0002,
+      "loss": 0.4506,
+      "step": 8150
+    },
+    {
+      "epoch": 5.923774954627949,
+      "grad_norm": 1.1704283952713013,
+      "learning_rate": 0.0002,
+      "loss": 0.4651,
+      "step": 8160
+    },
+    {
+      "epoch": 5.931034482758621,
+      "grad_norm": 1.040814995765686,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 8170
+    },
+    {
+      "epoch": 5.938294010889292,
+      "grad_norm": 1.1367416381835938,
+      "learning_rate": 0.0002,
+      "loss": 0.4038,
+      "step": 8180
+    },
+    {
+      "epoch": 5.945553539019964,
+      "grad_norm": 1.3401511907577515,
+      "learning_rate": 0.0002,
+      "loss": 0.4387,
+      "step": 8190
+    },
+    {
+      "epoch": 5.952813067150635,
+      "grad_norm": 1.1154041290283203,
+      "learning_rate": 0.0002,
+      "loss": 0.4396,
+      "step": 8200
+    },
+    {
+      "epoch": 5.960072595281307,
+      "grad_norm": 1.426089882850647,
+      "learning_rate": 0.0002,
+      "loss": 0.4744,
+      "step": 8210
+    },
+    {
+      "epoch": 5.967332123411978,
+      "grad_norm": 1.3170222043991089,
+      "learning_rate": 0.0002,
+      "loss": 0.4105,
+      "step": 8220
+    },
+    {
+      "epoch": 5.974591651542649,
+      "grad_norm": 1.1960029602050781,
+      "learning_rate": 0.0002,
+      "loss": 0.4137,
+      "step": 8230
+    },
+    {
+      "epoch": 5.981851179673321,
+      "grad_norm": 1.0843931436538696,
+      "learning_rate": 0.0002,
+      "loss": 0.423,
+      "step": 8240
+    },
+    {
+      "epoch": 5.9891107078039925,
+      "grad_norm": 1.050421118736267,
+      "learning_rate": 0.0002,
+      "loss": 0.459,
+      "step": 8250
+    },
+    {
+      "epoch": 5.9963702359346644,
+      "grad_norm": 1.0183138847351074,
+      "learning_rate": 0.0002,
+      "loss": 0.3993,
+      "step": 8260
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.4677470922470093,
+      "eval_runtime": 46.2504,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 8265
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11016,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.245213576167424e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..140bceabdec44fc473842cb9ace4a62a4dc9fe89
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-8265/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4abec5480aff960b5a7b03a48be1cdb1ae29ba8792469c451f5c2f269e4be93b
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/README.md b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..503a34a03e25483aa99213835fd87bfc8289a3fe
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/README.md
@@ -0,0 +1,202 @@
+---
+base_model: google/gemma-2-9b-it
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/adapter_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..e04e4441e0c1b29f69c16b26c142944e440b8076
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-2-9b-it",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/adapter_model.safetensors b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a29550672985a02772104c9364a326af8516392f
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbfa96848843eee83f689b4aa8c81d3536f5fda32736d2f1bc88c7aaf8d8d581
+size 143153376
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/optimizer.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..9015ff4f0ebdd6af207e0f9bedb65376cb721252
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45bf5cb7fa854fc4154de94b62b2a2f6065395ab838e15396207bc7140a60ac0
+size 72886650
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/rng_state.pth b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..3ebd19de3bde8f3039b2d46160e1963c74508e88
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bfe69a522f129b0b868fa3f22242a0f3a02106525c851adceef61abf24ab6ab
+size 14244
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/scheduler.pt b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..f5a3b6fa20bce38a9ee756d031b747473d48d900
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fafd2ca8d1fb4578cad77d5f424303ecbe66d43ed15127356f9414a5f5e0997e
+size 1064
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/trainer_state.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..34cc6d54c103f47037a381a3332231f4248cb873
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/trainer_state.json
@@ -0,0 +1,6837 @@
+{
+  "best_metric": 1.1086540222167969,
+  "best_model_checkpoint": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755",
+  "epoch": 6.999637023593467,
+  "eval_steps": 10,
+  "global_step": 9642,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007259528130671506,
+      "grad_norm": 0.46010470390319824,
+      "learning_rate": 0.0002,
+      "loss": 2.4936,
+      "step": 10
+    },
+    {
+      "epoch": 0.014519056261343012,
+      "grad_norm": 0.5103179216384888,
+      "learning_rate": 0.0002,
+      "loss": 1.8442,
+      "step": 20
+    },
+    {
+      "epoch": 0.021778584392014518,
+      "grad_norm": 0.6282716989517212,
+      "learning_rate": 0.0002,
+      "loss": 1.559,
+      "step": 30
+    },
+    {
+      "epoch": 0.029038112522686024,
+      "grad_norm": 1.2480497360229492,
+      "learning_rate": 0.0002,
+      "loss": 1.3618,
+      "step": 40
+    },
+    {
+      "epoch": 0.036297640653357534,
+      "grad_norm": 0.4114855229854584,
+      "learning_rate": 0.0002,
+      "loss": 1.3874,
+      "step": 50
+    },
+    {
+      "epoch": 0.043557168784029036,
+      "grad_norm": 0.49482840299606323,
+      "learning_rate": 0.0002,
+      "loss": 1.2836,
+      "step": 60
+    },
+    {
+      "epoch": 0.050816696914700546,
+      "grad_norm": 0.4536272883415222,
+      "learning_rate": 0.0002,
+      "loss": 1.1728,
+      "step": 70
+    },
+    {
+      "epoch": 0.05807622504537205,
+      "grad_norm": 0.32328274846076965,
+      "learning_rate": 0.0002,
+      "loss": 1.1125,
+      "step": 80
+    },
+    {
+      "epoch": 0.06533575317604355,
+      "grad_norm": 0.40990468859672546,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 90
+    },
+    {
+      "epoch": 0.07259528130671507,
+      "grad_norm": 0.37273502349853516,
+      "learning_rate": 0.0002,
+      "loss": 1.2216,
+      "step": 100
+    },
+    {
+      "epoch": 0.07985480943738657,
+      "grad_norm": 0.3903871476650238,
+      "learning_rate": 0.0002,
+      "loss": 1.2259,
+      "step": 110
+    },
+    {
+      "epoch": 0.08711433756805807,
+      "grad_norm": 0.3272787928581238,
+      "learning_rate": 0.0002,
+      "loss": 1.3718,
+      "step": 120
+    },
+    {
+      "epoch": 0.09437386569872959,
+      "grad_norm": 0.3622824251651764,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 130
+    },
+    {
+      "epoch": 0.10163339382940109,
+      "grad_norm": 0.3503916561603546,
+      "learning_rate": 0.0002,
+      "loss": 1.2966,
+      "step": 140
+    },
+    {
+      "epoch": 0.1088929219600726,
+      "grad_norm": 0.32787832617759705,
+      "learning_rate": 0.0002,
+      "loss": 1.0913,
+      "step": 150
+    },
+    {
+      "epoch": 0.1161524500907441,
+      "grad_norm": 0.5822657942771912,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 160
+    },
+    {
+      "epoch": 0.12341197822141561,
+      "grad_norm": 0.28028249740600586,
+      "learning_rate": 0.0002,
+      "loss": 1.2091,
+      "step": 170
+    },
+    {
+      "epoch": 0.1306715063520871,
+      "grad_norm": 0.33602750301361084,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 180
+    },
+    {
+      "epoch": 0.13793103448275862,
+      "grad_norm": 0.29106274247169495,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 190
+    },
+    {
+      "epoch": 0.14519056261343014,
+      "grad_norm": 0.38753363490104675,
+      "learning_rate": 0.0002,
+      "loss": 1.3261,
+      "step": 200
+    },
+    {
+      "epoch": 0.15245009074410162,
+      "grad_norm": 0.361009418964386,
+      "learning_rate": 0.0002,
+      "loss": 1.1536,
+      "step": 210
+    },
+    {
+      "epoch": 0.15970961887477314,
+      "grad_norm": 0.6743836402893066,
+      "learning_rate": 0.0002,
+      "loss": 1.1483,
+      "step": 220
+    },
+    {
+      "epoch": 0.16696914700544466,
+      "grad_norm": 0.3923613727092743,
+      "learning_rate": 0.0002,
+      "loss": 1.264,
+      "step": 230
+    },
+    {
+      "epoch": 0.17422867513611615,
+      "grad_norm": 0.2809699773788452,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 240
+    },
+    {
+      "epoch": 0.18148820326678766,
+      "grad_norm": 0.3631494641304016,
+      "learning_rate": 0.0002,
+      "loss": 1.1826,
+      "step": 250
+    },
+    {
+      "epoch": 0.18874773139745918,
+      "grad_norm": 0.24658171832561493,
+      "learning_rate": 0.0002,
+      "loss": 1.1584,
+      "step": 260
+    },
+    {
+      "epoch": 0.19600725952813067,
+      "grad_norm": 0.5780664682388306,
+      "learning_rate": 0.0002,
+      "loss": 1.0065,
+      "step": 270
+    },
+    {
+      "epoch": 0.20326678765880218,
+      "grad_norm": 0.3056720495223999,
+      "learning_rate": 0.0002,
+      "loss": 1.1251,
+      "step": 280
+    },
+    {
+      "epoch": 0.21052631578947367,
+      "grad_norm": 0.2857084572315216,
+      "learning_rate": 0.0002,
+      "loss": 1.1503,
+      "step": 290
+    },
+    {
+      "epoch": 0.2177858439201452,
+      "grad_norm": 0.6645345687866211,
+      "learning_rate": 0.0002,
+      "loss": 1.1215,
+      "step": 300
+    },
+    {
+      "epoch": 0.2250453720508167,
+      "grad_norm": 0.5966078639030457,
+      "learning_rate": 0.0002,
+      "loss": 1.127,
+      "step": 310
+    },
+    {
+      "epoch": 0.2323049001814882,
+      "grad_norm": 0.40937140583992004,
+      "learning_rate": 0.0002,
+      "loss": 1.1974,
+      "step": 320
+    },
+    {
+      "epoch": 0.2395644283121597,
+      "grad_norm": 0.5642806887626648,
+      "learning_rate": 0.0002,
+      "loss": 1.2385,
+      "step": 330
+    },
+    {
+      "epoch": 0.24682395644283123,
+      "grad_norm": 0.2750748097896576,
+      "learning_rate": 0.0002,
+      "loss": 1.1703,
+      "step": 340
+    },
+    {
+      "epoch": 0.2540834845735027,
+      "grad_norm": 0.34350234270095825,
+      "learning_rate": 0.0002,
+      "loss": 1.2598,
+      "step": 350
+    },
+    {
+      "epoch": 0.2613430127041742,
+      "grad_norm": 0.6767239570617676,
+      "learning_rate": 0.0002,
+      "loss": 1.1942,
+      "step": 360
+    },
+    {
+      "epoch": 0.26860254083484575,
+      "grad_norm": 0.31006959080696106,
+      "learning_rate": 0.0002,
+      "loss": 1.1436,
+      "step": 370
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "grad_norm": 0.3825474679470062,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 380
+    },
+    {
+      "epoch": 0.2831215970961887,
+      "grad_norm": 0.30241551995277405,
+      "learning_rate": 0.0002,
+      "loss": 1.1523,
+      "step": 390
+    },
+    {
+      "epoch": 0.29038112522686027,
+      "grad_norm": 0.2962397336959839,
+      "learning_rate": 0.0002,
+      "loss": 1.1298,
+      "step": 400
+    },
+    {
+      "epoch": 0.29764065335753176,
+      "grad_norm": 0.2600369155406952,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 410
+    },
+    {
+      "epoch": 0.30490018148820325,
+      "grad_norm": 0.3675060272216797,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 420
+    },
+    {
+      "epoch": 0.3121597096188748,
+      "grad_norm": 0.3429498076438904,
+      "learning_rate": 0.0002,
+      "loss": 1.156,
+      "step": 430
+    },
+    {
+      "epoch": 0.3194192377495463,
+      "grad_norm": 0.34311825037002563,
+      "learning_rate": 0.0002,
+      "loss": 1.2741,
+      "step": 440
+    },
+    {
+      "epoch": 0.32667876588021777,
+      "grad_norm": 0.37872210144996643,
+      "learning_rate": 0.0002,
+      "loss": 1.3523,
+      "step": 450
+    },
+    {
+      "epoch": 0.3339382940108893,
+      "grad_norm": 0.33271121978759766,
+      "learning_rate": 0.0002,
+      "loss": 1.1365,
+      "step": 460
+    },
+    {
+      "epoch": 0.3411978221415608,
+      "grad_norm": 0.34605276584625244,
+      "learning_rate": 0.0002,
+      "loss": 1.1572,
+      "step": 470
+    },
+    {
+      "epoch": 0.3484573502722323,
+      "grad_norm": 0.41050562262535095,
+      "learning_rate": 0.0002,
+      "loss": 1.2755,
+      "step": 480
+    },
+    {
+      "epoch": 0.35571687840290384,
+      "grad_norm": 0.2066836953163147,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 490
+    },
+    {
+      "epoch": 0.3629764065335753,
+      "grad_norm": 0.2859014868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.2072,
+      "step": 500
+    },
+    {
+      "epoch": 0.3702359346642468,
+      "grad_norm": 0.28763777017593384,
+      "learning_rate": 0.0002,
+      "loss": 1.1435,
+      "step": 510
+    },
+    {
+      "epoch": 0.37749546279491836,
+      "grad_norm": 0.2730471193790436,
+      "learning_rate": 0.0002,
+      "loss": 1.1341,
+      "step": 520
+    },
+    {
+      "epoch": 0.38475499092558985,
+      "grad_norm": 0.3968936800956726,
+      "learning_rate": 0.0002,
+      "loss": 1.348,
+      "step": 530
+    },
+    {
+      "epoch": 0.39201451905626133,
+      "grad_norm": 0.3624701201915741,
+      "learning_rate": 0.0002,
+      "loss": 1.2363,
+      "step": 540
+    },
+    {
+      "epoch": 0.3992740471869328,
+      "grad_norm": 0.3303608298301697,
+      "learning_rate": 0.0002,
+      "loss": 1.1206,
+      "step": 550
+    },
+    {
+      "epoch": 0.40653357531760437,
+      "grad_norm": 0.33507466316223145,
+      "learning_rate": 0.0002,
+      "loss": 1.203,
+      "step": 560
+    },
+    {
+      "epoch": 0.41379310344827586,
+      "grad_norm": 0.3297670781612396,
+      "learning_rate": 0.0002,
+      "loss": 1.2294,
+      "step": 570
+    },
+    {
+      "epoch": 0.42105263157894735,
+      "grad_norm": 0.32334890961647034,
+      "learning_rate": 0.0002,
+      "loss": 1.2,
+      "step": 580
+    },
+    {
+      "epoch": 0.4283121597096189,
+      "grad_norm": 0.30281195044517517,
+      "learning_rate": 0.0002,
+      "loss": 1.112,
+      "step": 590
+    },
+    {
+      "epoch": 0.4355716878402904,
+      "grad_norm": 0.5900027751922607,
+      "learning_rate": 0.0002,
+      "loss": 1.1352,
+      "step": 600
+    },
+    {
+      "epoch": 0.44283121597096187,
+      "grad_norm": 0.28437477350234985,
+      "learning_rate": 0.0002,
+      "loss": 1.1575,
+      "step": 610
+    },
+    {
+      "epoch": 0.4500907441016334,
+      "grad_norm": 0.39601704478263855,
+      "learning_rate": 0.0002,
+      "loss": 1.2009,
+      "step": 620
+    },
+    {
+      "epoch": 0.4573502722323049,
+      "grad_norm": 0.41971510648727417,
+      "learning_rate": 0.0002,
+      "loss": 1.2302,
+      "step": 630
+    },
+    {
+      "epoch": 0.4646098003629764,
+      "grad_norm": 0.33814409375190735,
+      "learning_rate": 0.0002,
+      "loss": 1.236,
+      "step": 640
+    },
+    {
+      "epoch": 0.47186932849364793,
+      "grad_norm": 0.575718104839325,
+      "learning_rate": 0.0002,
+      "loss": 1.2471,
+      "step": 650
+    },
+    {
+      "epoch": 0.4791288566243194,
+      "grad_norm": 0.37927401065826416,
+      "learning_rate": 0.0002,
+      "loss": 1.1012,
+      "step": 660
+    },
+    {
+      "epoch": 0.4863883847549909,
+      "grad_norm": 0.3224332630634308,
+      "learning_rate": 0.0002,
+      "loss": 1.1552,
+      "step": 670
+    },
+    {
+      "epoch": 0.49364791288566245,
+      "grad_norm": 0.32683515548706055,
+      "learning_rate": 0.0002,
+      "loss": 1.1354,
+      "step": 680
+    },
+    {
+      "epoch": 0.5009074410163339,
+      "grad_norm": 0.4316163659095764,
+      "learning_rate": 0.0002,
+      "loss": 1.207,
+      "step": 690
+    },
+    {
+      "epoch": 0.5081669691470054,
+      "grad_norm": 0.342602401971817,
+      "learning_rate": 0.0002,
+      "loss": 1.2116,
+      "step": 700
+    },
+    {
+      "epoch": 0.515426497277677,
+      "grad_norm": 0.2794898748397827,
+      "learning_rate": 0.0002,
+      "loss": 1.0433,
+      "step": 710
+    },
+    {
+      "epoch": 0.5226860254083484,
+      "grad_norm": 0.3322339951992035,
+      "learning_rate": 0.0002,
+      "loss": 1.2395,
+      "step": 720
+    },
+    {
+      "epoch": 0.52994555353902,
+      "grad_norm": 0.3088509142398834,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 730
+    },
+    {
+      "epoch": 0.5372050816696915,
+      "grad_norm": 0.24444378912448883,
+      "learning_rate": 0.0002,
+      "loss": 1.1495,
+      "step": 740
+    },
+    {
+      "epoch": 0.5444646098003629,
+      "grad_norm": 0.3483171761035919,
+      "learning_rate": 0.0002,
+      "loss": 1.2176,
+      "step": 750
+    },
+    {
+      "epoch": 0.5517241379310345,
+      "grad_norm": 0.3425690531730652,
+      "learning_rate": 0.0002,
+      "loss": 1.2248,
+      "step": 760
+    },
+    {
+      "epoch": 0.558983666061706,
+      "grad_norm": 0.31841927766799927,
+      "learning_rate": 0.0002,
+      "loss": 1.1127,
+      "step": 770
+    },
+    {
+      "epoch": 0.5662431941923775,
+      "grad_norm": 0.39423868060112,
+      "learning_rate": 0.0002,
+      "loss": 1.1841,
+      "step": 780
+    },
+    {
+      "epoch": 0.573502722323049,
+      "grad_norm": 0.30328479409217834,
+      "learning_rate": 0.0002,
+      "loss": 1.0583,
+      "step": 790
+    },
+    {
+      "epoch": 0.5807622504537205,
+      "grad_norm": 0.24475938081741333,
+      "learning_rate": 0.0002,
+      "loss": 1.0718,
+      "step": 800
+    },
+    {
+      "epoch": 0.588021778584392,
+      "grad_norm": 0.37132805585861206,
+      "learning_rate": 0.0002,
+      "loss": 1.2006,
+      "step": 810
+    },
+    {
+      "epoch": 0.5952813067150635,
+      "grad_norm": 0.32195979356765747,
+      "learning_rate": 0.0002,
+      "loss": 1.1544,
+      "step": 820
+    },
+    {
+      "epoch": 0.6025408348457351,
+      "grad_norm": 0.2848738729953766,
+      "learning_rate": 0.0002,
+      "loss": 0.9937,
+      "step": 830
+    },
+    {
+      "epoch": 0.6098003629764065,
+      "grad_norm": 0.28015264868736267,
+      "learning_rate": 0.0002,
+      "loss": 1.0937,
+      "step": 840
+    },
+    {
+      "epoch": 0.617059891107078,
+      "grad_norm": 0.37796008586883545,
+      "learning_rate": 0.0002,
+      "loss": 1.1234,
+      "step": 850
+    },
+    {
+      "epoch": 0.6243194192377496,
+      "grad_norm": 0.39311841130256653,
+      "learning_rate": 0.0002,
+      "loss": 1.1455,
+      "step": 860
+    },
+    {
+      "epoch": 0.631578947368421,
+      "grad_norm": 0.2761685252189636,
+      "learning_rate": 0.0002,
+      "loss": 1.1066,
+      "step": 870
+    },
+    {
+      "epoch": 0.6388384754990926,
+      "grad_norm": 0.3826720118522644,
+      "learning_rate": 0.0002,
+      "loss": 1.0859,
+      "step": 880
+    },
+    {
+      "epoch": 0.6460980036297641,
+      "grad_norm": 0.30076679587364197,
+      "learning_rate": 0.0002,
+      "loss": 1.1182,
+      "step": 890
+    },
+    {
+      "epoch": 0.6533575317604355,
+      "grad_norm": 0.21997687220573425,
+      "learning_rate": 0.0002,
+      "loss": 1.0927,
+      "step": 900
+    },
+    {
+      "epoch": 0.6606170598911071,
+      "grad_norm": 0.32593777775764465,
+      "learning_rate": 0.0002,
+      "loss": 1.1603,
+      "step": 910
+    },
+    {
+      "epoch": 0.6678765880217786,
+      "grad_norm": 0.30347898602485657,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 920
+    },
+    {
+      "epoch": 0.6751361161524501,
+      "grad_norm": 0.44173774123191833,
+      "learning_rate": 0.0002,
+      "loss": 1.1642,
+      "step": 930
+    },
+    {
+      "epoch": 0.6823956442831216,
+      "grad_norm": 0.2507467269897461,
+      "learning_rate": 0.0002,
+      "loss": 1.1832,
+      "step": 940
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.29463833570480347,
+      "learning_rate": 0.0002,
+      "loss": 1.3001,
+      "step": 950
+    },
+    {
+      "epoch": 0.6969147005444646,
+      "grad_norm": 0.9363154172897339,
+      "learning_rate": 0.0002,
+      "loss": 1.1651,
+      "step": 960
+    },
+    {
+      "epoch": 0.7041742286751361,
+      "grad_norm": 0.3236212134361267,
+      "learning_rate": 0.0002,
+      "loss": 1.158,
+      "step": 970
+    },
+    {
+      "epoch": 0.7114337568058077,
+      "grad_norm": 0.3123254179954529,
+      "learning_rate": 0.0002,
+      "loss": 1.1078,
+      "step": 980
+    },
+    {
+      "epoch": 0.7186932849364791,
+      "grad_norm": 0.3395805060863495,
+      "learning_rate": 0.0002,
+      "loss": 1.1578,
+      "step": 990
+    },
+    {
+      "epoch": 0.7259528130671506,
+      "grad_norm": 0.3240964412689209,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7332123411978222,
+      "grad_norm": 0.31902948021888733,
+      "learning_rate": 0.0002,
+      "loss": 1.0313,
+      "step": 1010
+    },
+    {
+      "epoch": 0.7404718693284936,
+      "grad_norm": 0.4848408102989197,
+      "learning_rate": 0.0002,
+      "loss": 1.1664,
+      "step": 1020
+    },
+    {
+      "epoch": 0.7477313974591652,
+      "grad_norm": 0.33006033301353455,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 1030
+    },
+    {
+      "epoch": 0.7549909255898367,
+      "grad_norm": 0.2928730547428131,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7622504537205081,
+      "grad_norm": 0.3529164791107178,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7695099818511797,
+      "grad_norm": 0.2736213803291321,
+      "learning_rate": 0.0002,
+      "loss": 1.1274,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7767695099818511,
+      "grad_norm": 0.7200686931610107,
+      "learning_rate": 0.0002,
+      "loss": 1.1169,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7840290381125227,
+      "grad_norm": 0.33396708965301514,
+      "learning_rate": 0.0002,
+      "loss": 1.1171,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7912885662431942,
+      "grad_norm": 1.5760449171066284,
+      "learning_rate": 0.0002,
+      "loss": 1.2447,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7985480943738656,
+      "grad_norm": 0.28138381242752075,
+      "learning_rate": 0.0002,
+      "loss": 1.1126,
+      "step": 1100
+    },
+    {
+      "epoch": 0.8058076225045372,
+      "grad_norm": 0.2597472369670868,
+      "learning_rate": 0.0002,
+      "loss": 1.2012,
+      "step": 1110
+    },
+    {
+      "epoch": 0.8130671506352087,
+      "grad_norm": 0.3305445611476898,
+      "learning_rate": 0.0002,
+      "loss": 1.1177,
+      "step": 1120
+    },
+    {
+      "epoch": 0.8203266787658802,
+      "grad_norm": 0.3934599459171295,
+      "learning_rate": 0.0002,
+      "loss": 1.0849,
+      "step": 1130
+    },
+    {
+      "epoch": 0.8275862068965517,
+      "grad_norm": 0.3472191393375397,
+      "learning_rate": 0.0002,
+      "loss": 1.0889,
+      "step": 1140
+    },
+    {
+      "epoch": 0.8348457350272233,
+      "grad_norm": 0.2857365906238556,
+      "learning_rate": 0.0002,
+      "loss": 1.265,
+      "step": 1150
+    },
+    {
+      "epoch": 0.8421052631578947,
+      "grad_norm": 0.3207702934741974,
+      "learning_rate": 0.0002,
+      "loss": 1.03,
+      "step": 1160
+    },
+    {
+      "epoch": 0.8493647912885662,
+      "grad_norm": 0.3176484704017639,
+      "learning_rate": 0.0002,
+      "loss": 1.1669,
+      "step": 1170
+    },
+    {
+      "epoch": 0.8566243194192378,
+      "grad_norm": 0.40685558319091797,
+      "learning_rate": 0.0002,
+      "loss": 1.1386,
+      "step": 1180
+    },
+    {
+      "epoch": 0.8638838475499092,
+      "grad_norm": 0.31125199794769287,
+      "learning_rate": 0.0002,
+      "loss": 1.1383,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8711433756805808,
+      "grad_norm": 0.7361181378364563,
+      "learning_rate": 0.0002,
+      "loss": 1.1512,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8784029038112523,
+      "grad_norm": 0.33699527382850647,
+      "learning_rate": 0.0002,
+      "loss": 0.9983,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8856624319419237,
+      "grad_norm": 0.3315220773220062,
+      "learning_rate": 0.0002,
+      "loss": 1.11,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8929219600725953,
+      "grad_norm": 0.6256054043769836,
+      "learning_rate": 0.0002,
+      "loss": 1.0266,
+      "step": 1230
+    },
+    {
+      "epoch": 0.9001814882032668,
+      "grad_norm": 0.3692137897014618,
+      "learning_rate": 0.0002,
+      "loss": 1.1738,
+      "step": 1240
+    },
+    {
+      "epoch": 0.9074410163339383,
+      "grad_norm": 0.3538484573364258,
+      "learning_rate": 0.0002,
+      "loss": 1.1112,
+      "step": 1250
+    },
+    {
+      "epoch": 0.9147005444646098,
+      "grad_norm": 0.27863040566444397,
+      "learning_rate": 0.0002,
+      "loss": 1.0878,
+      "step": 1260
+    },
+    {
+      "epoch": 0.9219600725952813,
+      "grad_norm": 0.3322528302669525,
+      "learning_rate": 0.0002,
+      "loss": 1.0826,
+      "step": 1270
+    },
+    {
+      "epoch": 0.9292196007259528,
+      "grad_norm": 0.7553173303604126,
+      "learning_rate": 0.0002,
+      "loss": 1.0158,
+      "step": 1280
+    },
+    {
+      "epoch": 0.9364791288566243,
+      "grad_norm": 0.4856191575527191,
+      "learning_rate": 0.0002,
+      "loss": 1.198,
+      "step": 1290
+    },
+    {
+      "epoch": 0.9437386569872959,
+      "grad_norm": 0.3668074905872345,
+      "learning_rate": 0.0002,
+      "loss": 1.1678,
+      "step": 1300
+    },
+    {
+      "epoch": 0.9509981851179673,
+      "grad_norm": 0.29851067066192627,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1310
+    },
+    {
+      "epoch": 0.9582577132486388,
+      "grad_norm": 0.276664674282074,
+      "learning_rate": 0.0002,
+      "loss": 1.1299,
+      "step": 1320
+    },
+    {
+      "epoch": 0.9655172413793104,
+      "grad_norm": 0.2941018342971802,
+      "learning_rate": 0.0002,
+      "loss": 1.0597,
+      "step": 1330
+    },
+    {
+      "epoch": 0.9727767695099818,
+      "grad_norm": 0.3505859076976776,
+      "learning_rate": 0.0002,
+      "loss": 1.1731,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9800362976406534,
+      "grad_norm": 0.3067687451839447,
+      "learning_rate": 0.0002,
+      "loss": 1.1809,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9872958257713249,
+      "grad_norm": 0.27151066064834595,
+      "learning_rate": 0.0002,
+      "loss": 1.1511,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9945553539019963,
+      "grad_norm": 0.36370083689689636,
+      "learning_rate": 0.0002,
+      "loss": 1.1919,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9996370235934664,
+      "eval_loss": 1.1381088495254517,
+      "eval_runtime": 96.0848,
+      "eval_samples_per_second": 4.538,
+      "eval_steps_per_second": 0.572,
+      "step": 1377
+    },
+    {
+      "epoch": 1.0018148820326678,
+      "grad_norm": 0.27980583906173706,
+      "learning_rate": 0.0002,
+      "loss": 1.0082,
+      "step": 1380
+    },
+    {
+      "epoch": 1.0090744101633393,
+      "grad_norm": 0.26713913679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1390
+    },
+    {
+      "epoch": 1.0163339382940109,
+      "grad_norm": 0.3089541494846344,
+      "learning_rate": 0.0002,
+      "loss": 1.0241,
+      "step": 1400
+    },
+    {
+      "epoch": 1.0235934664246824,
+      "grad_norm": 0.4188242256641388,
+      "learning_rate": 0.0002,
+      "loss": 0.9649,
+      "step": 1410
+    },
+    {
+      "epoch": 1.030852994555354,
+      "grad_norm": 0.5246463418006897,
+      "learning_rate": 0.0002,
+      "loss": 0.9821,
+      "step": 1420
+    },
+    {
+      "epoch": 1.0381125226860255,
+      "grad_norm": 0.2728777825832367,
+      "learning_rate": 0.0002,
+      "loss": 1.0143,
+      "step": 1430
+    },
+    {
+      "epoch": 1.0453720508166968,
+      "grad_norm": 0.38167616724967957,
+      "learning_rate": 0.0002,
+      "loss": 1.075,
+      "step": 1440
+    },
+    {
+      "epoch": 1.0526315789473684,
+      "grad_norm": 0.4439380168914795,
+      "learning_rate": 0.0002,
+      "loss": 1.0229,
+      "step": 1450
+    },
+    {
+      "epoch": 1.05989110707804,
+      "grad_norm": 0.30954182147979736,
+      "learning_rate": 0.0002,
+      "loss": 1.0451,
+      "step": 1460
+    },
+    {
+      "epoch": 1.0671506352087115,
+      "grad_norm": 0.4022280275821686,
+      "learning_rate": 0.0002,
+      "loss": 1.0762,
+      "step": 1470
+    },
+    {
+      "epoch": 1.074410163339383,
+      "grad_norm": 0.7390811443328857,
+      "learning_rate": 0.0002,
+      "loss": 1.0754,
+      "step": 1480
+    },
+    {
+      "epoch": 1.0816696914700545,
+      "grad_norm": 0.3885486423969269,
+      "learning_rate": 0.0002,
+      "loss": 1.1378,
+      "step": 1490
+    },
+    {
+      "epoch": 1.0889292196007259,
+      "grad_norm": 0.5275560617446899,
+      "learning_rate": 0.0002,
+      "loss": 1.005,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0961887477313974,
+      "grad_norm": 0.35112282633781433,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1510
+    },
+    {
+      "epoch": 1.103448275862069,
+      "grad_norm": 0.33714351058006287,
+      "learning_rate": 0.0002,
+      "loss": 1.0331,
+      "step": 1520
+    },
+    {
+      "epoch": 1.1107078039927405,
+      "grad_norm": 0.31221693754196167,
+      "learning_rate": 0.0002,
+      "loss": 0.8723,
+      "step": 1530
+    },
+    {
+      "epoch": 1.117967332123412,
+      "grad_norm": 0.27549654245376587,
+      "learning_rate": 0.0002,
+      "loss": 1.0701,
+      "step": 1540
+    },
+    {
+      "epoch": 1.1252268602540836,
+      "grad_norm": 0.8465521335601807,
+      "learning_rate": 0.0002,
+      "loss": 1.0114,
+      "step": 1550
+    },
+    {
+      "epoch": 1.132486388384755,
+      "grad_norm": 0.36125949025154114,
+      "learning_rate": 0.0002,
+      "loss": 0.9643,
+      "step": 1560
+    },
+    {
+      "epoch": 1.1397459165154264,
+      "grad_norm": 0.37420371174812317,
+      "learning_rate": 0.0002,
+      "loss": 1.0744,
+      "step": 1570
+    },
+    {
+      "epoch": 1.147005444646098,
+      "grad_norm": 0.3294760584831238,
+      "learning_rate": 0.0002,
+      "loss": 1.165,
+      "step": 1580
+    },
+    {
+      "epoch": 1.1542649727767695,
+      "grad_norm": 0.3881238102912903,
+      "learning_rate": 0.0002,
+      "loss": 0.9207,
+      "step": 1590
+    },
+    {
+      "epoch": 1.161524500907441,
+      "grad_norm": 0.4766491651535034,
+      "learning_rate": 0.0002,
+      "loss": 1.0983,
+      "step": 1600
+    },
+    {
+      "epoch": 1.1687840290381124,
+      "grad_norm": 0.395530104637146,
+      "learning_rate": 0.0002,
+      "loss": 1.0092,
+      "step": 1610
+    },
+    {
+      "epoch": 1.176043557168784,
+      "grad_norm": 0.3297106623649597,
+      "learning_rate": 0.0002,
+      "loss": 0.9895,
+      "step": 1620
+    },
+    {
+      "epoch": 1.1833030852994555,
+      "grad_norm": 0.39528271555900574,
+      "learning_rate": 0.0002,
+      "loss": 1.1121,
+      "step": 1630
+    },
+    {
+      "epoch": 1.190562613430127,
+      "grad_norm": 0.3370221257209778,
+      "learning_rate": 0.0002,
+      "loss": 0.9202,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1978221415607986,
+      "grad_norm": 0.31922030448913574,
+      "learning_rate": 0.0002,
+      "loss": 1.102,
+      "step": 1650
+    },
+    {
+      "epoch": 1.2050816696914701,
+      "grad_norm": 0.6142027378082275,
+      "learning_rate": 0.0002,
+      "loss": 1.0565,
+      "step": 1660
+    },
+    {
+      "epoch": 1.2123411978221417,
+      "grad_norm": 0.44769710302352905,
+      "learning_rate": 0.0002,
+      "loss": 0.9541,
+      "step": 1670
+    },
+    {
+      "epoch": 1.219600725952813,
+      "grad_norm": 0.41233646869659424,
+      "learning_rate": 0.0002,
+      "loss": 1.2501,
+      "step": 1680
+    },
+    {
+      "epoch": 1.2268602540834845,
+      "grad_norm": 0.2928866147994995,
+      "learning_rate": 0.0002,
+      "loss": 1.0147,
+      "step": 1690
+    },
+    {
+      "epoch": 1.234119782214156,
+      "grad_norm": 0.36913734674453735,
+      "learning_rate": 0.0002,
+      "loss": 0.9074,
+      "step": 1700
+    },
+    {
+      "epoch": 1.2413793103448276,
+      "grad_norm": 0.5281891226768494,
+      "learning_rate": 0.0002,
+      "loss": 1.0847,
+      "step": 1710
+    },
+    {
+      "epoch": 1.2486388384754992,
+      "grad_norm": 0.3374697268009186,
+      "learning_rate": 0.0002,
+      "loss": 0.9482,
+      "step": 1720
+    },
+    {
+      "epoch": 1.2558983666061705,
+      "grad_norm": 0.3802020847797394,
+      "learning_rate": 0.0002,
+      "loss": 0.9826,
+      "step": 1730
+    },
+    {
+      "epoch": 1.263157894736842,
+      "grad_norm": 0.38048651814460754,
+      "learning_rate": 0.0002,
+      "loss": 1.0191,
+      "step": 1740
+    },
+    {
+      "epoch": 1.2704174228675136,
+      "grad_norm": 0.6676169633865356,
+      "learning_rate": 0.0002,
+      "loss": 1.0142,
+      "step": 1750
+    },
+    {
+      "epoch": 1.2776769509981851,
+      "grad_norm": 0.4075961410999298,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 1760
+    },
+    {
+      "epoch": 1.2849364791288567,
+      "grad_norm": 0.4374721348285675,
+      "learning_rate": 0.0002,
+      "loss": 1.0301,
+      "step": 1770
+    },
+    {
+      "epoch": 1.2921960072595282,
+      "grad_norm": 0.4638824164867401,
+      "learning_rate": 0.0002,
+      "loss": 1.0601,
+      "step": 1780
+    },
+    {
+      "epoch": 1.2994555353901998,
+      "grad_norm": 0.38631564378738403,
+      "learning_rate": 0.0002,
+      "loss": 0.9593,
+      "step": 1790
+    },
+    {
+      "epoch": 1.306715063520871,
+      "grad_norm": 0.35873809456825256,
+      "learning_rate": 0.0002,
+      "loss": 1.0042,
+      "step": 1800
+    },
+    {
+      "epoch": 1.3139745916515426,
+      "grad_norm": 0.33602237701416016,
+      "learning_rate": 0.0002,
+      "loss": 0.9835,
+      "step": 1810
+    },
+    {
+      "epoch": 1.3212341197822142,
+      "grad_norm": 0.46696463227272034,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 1820
+    },
+    {
+      "epoch": 1.3284936479128857,
+      "grad_norm": 0.368958979845047,
+      "learning_rate": 0.0002,
+      "loss": 1.0123,
+      "step": 1830
+    },
+    {
+      "epoch": 1.335753176043557,
+      "grad_norm": 0.3160957396030426,
+      "learning_rate": 0.0002,
+      "loss": 1.0456,
+      "step": 1840
+    },
+    {
+      "epoch": 1.3430127041742286,
+      "grad_norm": 0.4511511027812958,
+      "learning_rate": 0.0002,
+      "loss": 1.0264,
+      "step": 1850
+    },
+    {
+      "epoch": 1.3502722323049001,
+      "grad_norm": 0.6769845485687256,
+      "learning_rate": 0.0002,
+      "loss": 1.0179,
+      "step": 1860
+    },
+    {
+      "epoch": 1.3575317604355717,
+      "grad_norm": 0.3749309480190277,
+      "learning_rate": 0.0002,
+      "loss": 1.011,
+      "step": 1870
+    },
+    {
+      "epoch": 1.3647912885662432,
+      "grad_norm": 0.553798496723175,
+      "learning_rate": 0.0002,
+      "loss": 1.0223,
+      "step": 1880
+    },
+    {
+      "epoch": 1.3720508166969148,
+      "grad_norm": 0.3538985550403595,
+      "learning_rate": 0.0002,
+      "loss": 0.9634,
+      "step": 1890
+    },
+    {
+      "epoch": 1.3793103448275863,
+      "grad_norm": 0.5501534938812256,
+      "learning_rate": 0.0002,
+      "loss": 1.0275,
+      "step": 1900
+    },
+    {
+      "epoch": 1.3865698729582578,
+      "grad_norm": 0.4432051181793213,
+      "learning_rate": 0.0002,
+      "loss": 1.0461,
+      "step": 1910
+    },
+    {
+      "epoch": 1.3938294010889292,
+      "grad_norm": 0.41755786538124084,
+      "learning_rate": 0.0002,
+      "loss": 0.9082,
+      "step": 1920
+    },
+    {
+      "epoch": 1.4010889292196007,
+      "grad_norm": 0.5626114010810852,
+      "learning_rate": 0.0002,
+      "loss": 1.0858,
+      "step": 1930
+    },
+    {
+      "epoch": 1.4083484573502723,
+      "grad_norm": 0.44038185477256775,
+      "learning_rate": 0.0002,
+      "loss": 1.0687,
+      "step": 1940
+    },
+    {
+      "epoch": 1.4156079854809438,
+      "grad_norm": 0.3397001624107361,
+      "learning_rate": 0.0002,
+      "loss": 0.9454,
+      "step": 1950
+    },
+    {
+      "epoch": 1.4228675136116151,
+      "grad_norm": 0.4325368106365204,
+      "learning_rate": 0.0002,
+      "loss": 0.943,
+      "step": 1960
+    },
+    {
+      "epoch": 1.4301270417422867,
+      "grad_norm": 0.3900907039642334,
+      "learning_rate": 0.0002,
+      "loss": 1.0488,
+      "step": 1970
+    },
+    {
+      "epoch": 1.4373865698729582,
+      "grad_norm": 0.369612455368042,
+      "learning_rate": 0.0002,
+      "loss": 0.9699,
+      "step": 1980
+    },
+    {
+      "epoch": 1.4446460980036298,
+      "grad_norm": 0.4389338791370392,
+      "learning_rate": 0.0002,
+      "loss": 1.0609,
+      "step": 1990
+    },
+    {
+      "epoch": 1.4519056261343013,
+      "grad_norm": 1.694450855255127,
+      "learning_rate": 0.0002,
+      "loss": 1.042,
+      "step": 2000
+    },
+    {
+      "epoch": 1.4591651542649728,
+      "grad_norm": 0.516957700252533,
+      "learning_rate": 0.0002,
+      "loss": 0.997,
+      "step": 2010
+    },
+    {
+      "epoch": 1.4664246823956444,
+      "grad_norm": 0.45515501499176025,
+      "learning_rate": 0.0002,
+      "loss": 1.0979,
+      "step": 2020
+    },
+    {
+      "epoch": 1.4736842105263157,
+      "grad_norm": 0.4153139591217041,
+      "learning_rate": 0.0002,
+      "loss": 1.0188,
+      "step": 2030
+    },
+    {
+      "epoch": 1.4809437386569873,
+      "grad_norm": 0.44353052973747253,
+      "learning_rate": 0.0002,
+      "loss": 1.1521,
+      "step": 2040
+    },
+    {
+      "epoch": 1.4882032667876588,
+      "grad_norm": 0.570554256439209,
+      "learning_rate": 0.0002,
+      "loss": 0.9653,
+      "step": 2050
+    },
+    {
+      "epoch": 1.4954627949183303,
+      "grad_norm": 0.5742740035057068,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2060
+    },
+    {
+      "epoch": 1.5027223230490017,
+      "grad_norm": 0.5890517830848694,
+      "learning_rate": 0.0002,
+      "loss": 1.0963,
+      "step": 2070
+    },
+    {
+      "epoch": 1.5099818511796732,
+      "grad_norm": 0.4162650406360626,
+      "learning_rate": 0.0002,
+      "loss": 1.0639,
+      "step": 2080
+    },
+    {
+      "epoch": 1.5172413793103448,
+      "grad_norm": 0.7334717512130737,
+      "learning_rate": 0.0002,
+      "loss": 1.071,
+      "step": 2090
+    },
+    {
+      "epoch": 1.5245009074410163,
+      "grad_norm": 0.2963249683380127,
+      "learning_rate": 0.0002,
+      "loss": 0.8957,
+      "step": 2100
+    },
+    {
+      "epoch": 1.5317604355716878,
+      "grad_norm": 0.30676454305648804,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 2110
+    },
+    {
+      "epoch": 1.5390199637023594,
+      "grad_norm": 0.35984641313552856,
+      "learning_rate": 0.0002,
+      "loss": 0.9967,
+      "step": 2120
+    },
+    {
+      "epoch": 1.546279491833031,
+      "grad_norm": 0.3384549617767334,
+      "learning_rate": 0.0002,
+      "loss": 1.0352,
+      "step": 2130
+    },
+    {
+      "epoch": 1.5535390199637025,
+      "grad_norm": 0.4725518226623535,
+      "learning_rate": 0.0002,
+      "loss": 0.9874,
+      "step": 2140
+    },
+    {
+      "epoch": 1.560798548094374,
+      "grad_norm": 0.3252685070037842,
+      "learning_rate": 0.0002,
+      "loss": 1.1166,
+      "step": 2150
+    },
+    {
+      "epoch": 1.5680580762250453,
+      "grad_norm": 0.45043081045150757,
+      "learning_rate": 0.0002,
+      "loss": 0.9127,
+      "step": 2160
+    },
+    {
+      "epoch": 1.5753176043557169,
+      "grad_norm": 0.374208003282547,
+      "learning_rate": 0.0002,
+      "loss": 1.0767,
+      "step": 2170
+    },
+    {
+      "epoch": 1.5825771324863884,
+      "grad_norm": 0.5118404030799866,
+      "learning_rate": 0.0002,
+      "loss": 1.0255,
+      "step": 2180
+    },
+    {
+      "epoch": 1.5898366606170597,
+      "grad_norm": 0.482170969247818,
+      "learning_rate": 0.0002,
+      "loss": 1.0062,
+      "step": 2190
+    },
+    {
+      "epoch": 1.5970961887477313,
+      "grad_norm": 0.5337533950805664,
+      "learning_rate": 0.0002,
+      "loss": 1.0872,
+      "step": 2200
+    },
+    {
+      "epoch": 1.6043557168784028,
+      "grad_norm": 0.5195064544677734,
+      "learning_rate": 0.0002,
+      "loss": 1.0405,
+      "step": 2210
+    },
+    {
+      "epoch": 1.6116152450090744,
+      "grad_norm": 0.30807098746299744,
+      "learning_rate": 0.0002,
+      "loss": 1.0454,
+      "step": 2220
+    },
+    {
+      "epoch": 1.618874773139746,
+      "grad_norm": 0.3962925672531128,
+      "learning_rate": 0.0002,
+      "loss": 1.0293,
+      "step": 2230
+    },
+    {
+      "epoch": 1.6261343012704175,
+      "grad_norm": 0.7636962532997131,
+      "learning_rate": 0.0002,
+      "loss": 1.0137,
+      "step": 2240
+    },
+    {
+      "epoch": 1.633393829401089,
+      "grad_norm": 0.32380592823028564,
+      "learning_rate": 0.0002,
+      "loss": 0.991,
+      "step": 2250
+    },
+    {
+      "epoch": 1.6406533575317606,
+      "grad_norm": 0.5767741799354553,
+      "learning_rate": 0.0002,
+      "loss": 0.9471,
+      "step": 2260
+    },
+    {
+      "epoch": 1.647912885662432,
+      "grad_norm": 0.39964812994003296,
+      "learning_rate": 0.0002,
+      "loss": 0.9524,
+      "step": 2270
+    },
+    {
+      "epoch": 1.6551724137931034,
+      "grad_norm": 0.622629702091217,
+      "learning_rate": 0.0002,
+      "loss": 0.9866,
+      "step": 2280
+    },
+    {
+      "epoch": 1.662431941923775,
+      "grad_norm": 0.40202152729034424,
+      "learning_rate": 0.0002,
+      "loss": 0.9609,
+      "step": 2290
+    },
+    {
+      "epoch": 1.6696914700544465,
+      "grad_norm": 0.4467349052429199,
+      "learning_rate": 0.0002,
+      "loss": 0.9859,
+      "step": 2300
+    },
+    {
+      "epoch": 1.6769509981851178,
+      "grad_norm": 0.5026949048042297,
+      "learning_rate": 0.0002,
+      "loss": 1.0312,
+      "step": 2310
+    },
+    {
+      "epoch": 1.6842105263157894,
+      "grad_norm": 0.43754541873931885,
+      "learning_rate": 0.0002,
+      "loss": 1.0776,
+      "step": 2320
+    },
+    {
+      "epoch": 1.691470054446461,
+      "grad_norm": 0.42869430780410767,
+      "learning_rate": 0.0002,
+      "loss": 0.9786,
+      "step": 2330
+    },
+    {
+      "epoch": 1.6987295825771325,
+      "grad_norm": 0.4192679524421692,
+      "learning_rate": 0.0002,
+      "loss": 1.0747,
+      "step": 2340
+    },
+    {
+      "epoch": 1.705989110707804,
+      "grad_norm": 0.3243155777454376,
+      "learning_rate": 0.0002,
+      "loss": 1.0054,
+      "step": 2350
+    },
+    {
+      "epoch": 1.7132486388384756,
+      "grad_norm": 1.0514075756072998,
+      "learning_rate": 0.0002,
+      "loss": 0.9165,
+      "step": 2360
+    },
+    {
+      "epoch": 1.720508166969147,
+      "grad_norm": 0.4719122648239136,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 2370
+    },
+    {
+      "epoch": 1.7277676950998186,
+      "grad_norm": 0.3846144676208496,
+      "learning_rate": 0.0002,
+      "loss": 0.9913,
+      "step": 2380
+    },
+    {
+      "epoch": 1.73502722323049,
+      "grad_norm": 0.7266581058502197,
+      "learning_rate": 0.0002,
+      "loss": 0.9935,
+      "step": 2390
+    },
+    {
+      "epoch": 1.7422867513611615,
+      "grad_norm": 0.6183241605758667,
+      "learning_rate": 0.0002,
+      "loss": 0.9509,
+      "step": 2400
+    },
+    {
+      "epoch": 1.749546279491833,
+      "grad_norm": 0.3658260405063629,
+      "learning_rate": 0.0002,
+      "loss": 0.9713,
+      "step": 2410
+    },
+    {
+      "epoch": 1.7568058076225044,
+      "grad_norm": 0.6036322712898254,
+      "learning_rate": 0.0002,
+      "loss": 1.0033,
+      "step": 2420
+    },
+    {
+      "epoch": 1.764065335753176,
+      "grad_norm": 0.7872936129570007,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 2430
+    },
+    {
+      "epoch": 1.7713248638838475,
+      "grad_norm": 0.35946124792099,
+      "learning_rate": 0.0002,
+      "loss": 0.99,
+      "step": 2440
+    },
+    {
+      "epoch": 1.778584392014519,
+      "grad_norm": 0.3740338981151581,
+      "learning_rate": 0.0002,
+      "loss": 0.8845,
+      "step": 2450
+    },
+    {
+      "epoch": 1.7858439201451906,
+      "grad_norm": 0.6150230169296265,
+      "learning_rate": 0.0002,
+      "loss": 0.9333,
+      "step": 2460
+    },
+    {
+      "epoch": 1.793103448275862,
+      "grad_norm": 0.4726075530052185,
+      "learning_rate": 0.0002,
+      "loss": 1.0373,
+      "step": 2470
+    },
+    {
+      "epoch": 1.8003629764065336,
+      "grad_norm": 0.31292253732681274,
+      "learning_rate": 0.0002,
+      "loss": 0.974,
+      "step": 2480
+    },
+    {
+      "epoch": 1.8076225045372052,
+      "grad_norm": 0.4463104009628296,
+      "learning_rate": 0.0002,
+      "loss": 1.0205,
+      "step": 2490
+    },
+    {
+      "epoch": 1.8148820326678767,
+      "grad_norm": 0.7848200798034668,
+      "learning_rate": 0.0002,
+      "loss": 1.0771,
+      "step": 2500
+    },
+    {
+      "epoch": 1.822141560798548,
+      "grad_norm": 0.5562082529067993,
+      "learning_rate": 0.0002,
+      "loss": 0.9508,
+      "step": 2510
+    },
+    {
+      "epoch": 1.8294010889292196,
+      "grad_norm": 0.39892756938934326,
+      "learning_rate": 0.0002,
+      "loss": 0.9343,
+      "step": 2520
+    },
+    {
+      "epoch": 1.8366606170598911,
+      "grad_norm": 1.2923320531845093,
+      "learning_rate": 0.0002,
+      "loss": 1.0489,
+      "step": 2530
+    },
+    {
+      "epoch": 1.8439201451905625,
+      "grad_norm": 0.6316490769386292,
+      "learning_rate": 0.0002,
+      "loss": 1.1203,
+      "step": 2540
+    },
+    {
+      "epoch": 1.851179673321234,
+      "grad_norm": 0.46100497245788574,
+      "learning_rate": 0.0002,
+      "loss": 0.9131,
+      "step": 2550
+    },
+    {
+      "epoch": 1.8584392014519056,
+      "grad_norm": 0.7902987003326416,
+      "learning_rate": 0.0002,
+      "loss": 1.0977,
+      "step": 2560
+    },
+    {
+      "epoch": 1.865698729582577,
+      "grad_norm": 0.4596365690231323,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 2570
+    },
+    {
+      "epoch": 1.8729582577132486,
+      "grad_norm": 0.6592172384262085,
+      "learning_rate": 0.0002,
+      "loss": 0.9322,
+      "step": 2580
+    },
+    {
+      "epoch": 1.8802177858439202,
+      "grad_norm": 0.5764662027359009,
+      "learning_rate": 0.0002,
+      "loss": 0.9549,
+      "step": 2590
+    },
+    {
+      "epoch": 1.8874773139745917,
+      "grad_norm": 0.8421637415885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9485,
+      "step": 2600
+    },
+    {
+      "epoch": 1.8947368421052633,
+      "grad_norm": 0.5635305047035217,
+      "learning_rate": 0.0002,
+      "loss": 1.012,
+      "step": 2610
+    },
+    {
+      "epoch": 1.9019963702359348,
+      "grad_norm": 0.46231237053871155,
+      "learning_rate": 0.0002,
+      "loss": 0.8907,
+      "step": 2620
+    },
+    {
+      "epoch": 1.9092558983666061,
+      "grad_norm": 0.3944607079029083,
+      "learning_rate": 0.0002,
+      "loss": 0.9543,
+      "step": 2630
+    },
+    {
+      "epoch": 1.9165154264972777,
+      "grad_norm": 0.4753907322883606,
+      "learning_rate": 0.0002,
+      "loss": 0.9964,
+      "step": 2640
+    },
+    {
+      "epoch": 1.9237749546279492,
+      "grad_norm": 0.4151090979576111,
+      "learning_rate": 0.0002,
+      "loss": 1.0217,
+      "step": 2650
+    },
+    {
+      "epoch": 1.9310344827586206,
+      "grad_norm": 0.6793725490570068,
+      "learning_rate": 0.0002,
+      "loss": 1.0397,
+      "step": 2660
+    },
+    {
+      "epoch": 1.938294010889292,
+      "grad_norm": 0.339755117893219,
+      "learning_rate": 0.0002,
+      "loss": 1.0401,
+      "step": 2670
+    },
+    {
+      "epoch": 1.9455535390199636,
+      "grad_norm": 0.40789374709129333,
+      "learning_rate": 0.0002,
+      "loss": 1.082,
+      "step": 2680
+    },
+    {
+      "epoch": 1.9528130671506352,
+      "grad_norm": 0.3750005066394806,
+      "learning_rate": 0.0002,
+      "loss": 0.9973,
+      "step": 2690
+    },
+    {
+      "epoch": 1.9600725952813067,
+      "grad_norm": 0.39684441685676575,
+      "learning_rate": 0.0002,
+      "loss": 0.9837,
+      "step": 2700
+    },
+    {
+      "epoch": 1.9673321234119783,
+      "grad_norm": 0.378287672996521,
+      "learning_rate": 0.0002,
+      "loss": 1.1204,
+      "step": 2710
+    },
+    {
+      "epoch": 1.9745916515426498,
+      "grad_norm": 0.3668482005596161,
+      "learning_rate": 0.0002,
+      "loss": 0.9284,
+      "step": 2720
+    },
+    {
+      "epoch": 1.9818511796733214,
+      "grad_norm": 0.49997565150260925,
+      "learning_rate": 0.0002,
+      "loss": 1.0512,
+      "step": 2730
+    },
+    {
+      "epoch": 1.989110707803993,
+      "grad_norm": 0.36852124333381653,
+      "learning_rate": 0.0002,
+      "loss": 0.9311,
+      "step": 2740
+    },
+    {
+      "epoch": 1.9963702359346642,
+      "grad_norm": 0.5203380584716797,
+      "learning_rate": 0.0002,
+      "loss": 1.0898,
+      "step": 2750
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.1086540222167969,
+      "eval_runtime": 95.6866,
+      "eval_samples_per_second": 4.557,
+      "eval_steps_per_second": 0.575,
+      "step": 2755
+    },
+    {
+      "epoch": 2.0036297640653356,
+      "grad_norm": 0.35921573638916016,
+      "learning_rate": 0.0002,
+      "loss": 0.9791,
+      "step": 2760
+    },
+    {
+      "epoch": 2.010889292196007,
+      "grad_norm": 1.013861894607544,
+      "learning_rate": 0.0002,
+      "loss": 0.8968,
+      "step": 2770
+    },
+    {
+      "epoch": 2.0181488203266786,
+      "grad_norm": 0.4425240159034729,
+      "learning_rate": 0.0002,
+      "loss": 0.7873,
+      "step": 2780
+    },
+    {
+      "epoch": 2.02540834845735,
+      "grad_norm": 0.60174161195755,
+      "learning_rate": 0.0002,
+      "loss": 0.8758,
+      "step": 2790
+    },
+    {
+      "epoch": 2.0326678765880217,
+      "grad_norm": 0.47582098841667175,
+      "learning_rate": 0.0002,
+      "loss": 0.7989,
+      "step": 2800
+    },
+    {
+      "epoch": 2.0399274047186933,
+      "grad_norm": 0.6012811660766602,
+      "learning_rate": 0.0002,
+      "loss": 0.8898,
+      "step": 2810
+    },
+    {
+      "epoch": 2.047186932849365,
+      "grad_norm": 0.4444408118724823,
+      "learning_rate": 0.0002,
+      "loss": 0.8109,
+      "step": 2820
+    },
+    {
+      "epoch": 2.0544464609800364,
+      "grad_norm": 0.4864003360271454,
+      "learning_rate": 0.0002,
+      "loss": 0.8621,
+      "step": 2830
+    },
+    {
+      "epoch": 2.061705989110708,
+      "grad_norm": 0.5104215741157532,
+      "learning_rate": 0.0002,
+      "loss": 0.8817,
+      "step": 2840
+    },
+    {
+      "epoch": 2.0689655172413794,
+      "grad_norm": 0.6218489408493042,
+      "learning_rate": 0.0002,
+      "loss": 0.7789,
+      "step": 2850
+    },
+    {
+      "epoch": 2.076225045372051,
+      "grad_norm": 0.705784261226654,
+      "learning_rate": 0.0002,
+      "loss": 0.8778,
+      "step": 2860
+    },
+    {
+      "epoch": 2.0834845735027225,
+      "grad_norm": 0.48091503977775574,
+      "learning_rate": 0.0002,
+      "loss": 0.7851,
+      "step": 2870
+    },
+    {
+      "epoch": 2.0907441016333936,
+      "grad_norm": 0.5062456727027893,
+      "learning_rate": 0.0002,
+      "loss": 0.8444,
+      "step": 2880
+    },
+    {
+      "epoch": 2.098003629764065,
+      "grad_norm": 0.4862022399902344,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 2890
+    },
+    {
+      "epoch": 2.1052631578947367,
+      "grad_norm": 0.48264044523239136,
+      "learning_rate": 0.0002,
+      "loss": 0.8593,
+      "step": 2900
+    },
+    {
+      "epoch": 2.1125226860254083,
+      "grad_norm": 0.43744346499443054,
+      "learning_rate": 0.0002,
+      "loss": 0.8325,
+      "step": 2910
+    },
+    {
+      "epoch": 2.11978221415608,
+      "grad_norm": 0.5480492115020752,
+      "learning_rate": 0.0002,
+      "loss": 0.9099,
+      "step": 2920
+    },
+    {
+      "epoch": 2.1270417422867514,
+      "grad_norm": 0.5068560838699341,
+      "learning_rate": 0.0002,
+      "loss": 0.7727,
+      "step": 2930
+    },
+    {
+      "epoch": 2.134301270417423,
+      "grad_norm": 0.4650852680206299,
+      "learning_rate": 0.0002,
+      "loss": 0.8403,
+      "step": 2940
+    },
+    {
+      "epoch": 2.1415607985480944,
+      "grad_norm": 0.4929981231689453,
+      "learning_rate": 0.0002,
+      "loss": 0.8467,
+      "step": 2950
+    },
+    {
+      "epoch": 2.148820326678766,
+      "grad_norm": 0.6537389755249023,
+      "learning_rate": 0.0002,
+      "loss": 0.8747,
+      "step": 2960
+    },
+    {
+      "epoch": 2.1560798548094375,
+      "grad_norm": 0.8032940626144409,
+      "learning_rate": 0.0002,
+      "loss": 0.7663,
+      "step": 2970
+    },
+    {
+      "epoch": 2.163339382940109,
+      "grad_norm": 0.7131643891334534,
+      "learning_rate": 0.0002,
+      "loss": 0.7604,
+      "step": 2980
+    },
+    {
+      "epoch": 2.1705989110707806,
+      "grad_norm": 0.6034275889396667,
+      "learning_rate": 0.0002,
+      "loss": 0.8424,
+      "step": 2990
+    },
+    {
+      "epoch": 2.1778584392014517,
+      "grad_norm": 0.6081095933914185,
+      "learning_rate": 0.0002,
+      "loss": 0.8838,
+      "step": 3000
+    },
+    {
+      "epoch": 2.1851179673321233,
+      "grad_norm": 0.5706912875175476,
+      "learning_rate": 0.0002,
+      "loss": 0.9013,
+      "step": 3010
+    },
+    {
+      "epoch": 2.192377495462795,
+      "grad_norm": 0.6742380261421204,
+      "learning_rate": 0.0002,
+      "loss": 0.8473,
+      "step": 3020
+    },
+    {
+      "epoch": 2.1996370235934664,
+      "grad_norm": 0.4847496449947357,
+      "learning_rate": 0.0002,
+      "loss": 0.8372,
+      "step": 3030
+    },
+    {
+      "epoch": 2.206896551724138,
+      "grad_norm": 0.5529342889785767,
+      "learning_rate": 0.0002,
+      "loss": 0.8487,
+      "step": 3040
+    },
+    {
+      "epoch": 2.2141560798548094,
+      "grad_norm": 0.6108783483505249,
+      "learning_rate": 0.0002,
+      "loss": 0.9057,
+      "step": 3050
+    },
+    {
+      "epoch": 2.221415607985481,
+      "grad_norm": 0.8841571807861328,
+      "learning_rate": 0.0002,
+      "loss": 0.9659,
+      "step": 3060
+    },
+    {
+      "epoch": 2.2286751361161525,
+      "grad_norm": 0.4227530360221863,
+      "learning_rate": 0.0002,
+      "loss": 0.8538,
+      "step": 3070
+    },
+    {
+      "epoch": 2.235934664246824,
+      "grad_norm": 0.461935818195343,
+      "learning_rate": 0.0002,
+      "loss": 0.7925,
+      "step": 3080
+    },
+    {
+      "epoch": 2.2431941923774956,
+      "grad_norm": 0.5407412648200989,
+      "learning_rate": 0.0002,
+      "loss": 0.8542,
+      "step": 3090
+    },
+    {
+      "epoch": 2.250453720508167,
+      "grad_norm": 0.5057998895645142,
+      "learning_rate": 0.0002,
+      "loss": 0.8392,
+      "step": 3100
+    },
+    {
+      "epoch": 2.2577132486388383,
+      "grad_norm": 0.530057966709137,
+      "learning_rate": 0.0002,
+      "loss": 0.7898,
+      "step": 3110
+    },
+    {
+      "epoch": 2.26497277676951,
+      "grad_norm": 0.5066842436790466,
+      "learning_rate": 0.0002,
+      "loss": 0.8507,
+      "step": 3120
+    },
+    {
+      "epoch": 2.2722323049001814,
+      "grad_norm": 0.5069178342819214,
+      "learning_rate": 0.0002,
+      "loss": 0.7903,
+      "step": 3130
+    },
+    {
+      "epoch": 2.279491833030853,
+      "grad_norm": 0.6095499396324158,
+      "learning_rate": 0.0002,
+      "loss": 0.8458,
+      "step": 3140
+    },
+    {
+      "epoch": 2.2867513611615244,
+      "grad_norm": 0.49524766206741333,
+      "learning_rate": 0.0002,
+      "loss": 0.8171,
+      "step": 3150
+    },
+    {
+      "epoch": 2.294010889292196,
+      "grad_norm": 0.5334409475326538,
+      "learning_rate": 0.0002,
+      "loss": 0.7568,
+      "step": 3160
+    },
+    {
+      "epoch": 2.3012704174228675,
+      "grad_norm": 1.681748867034912,
+      "learning_rate": 0.0002,
+      "loss": 0.7533,
+      "step": 3170
+    },
+    {
+      "epoch": 2.308529945553539,
+      "grad_norm": 0.7225565314292908,
+      "learning_rate": 0.0002,
+      "loss": 0.8155,
+      "step": 3180
+    },
+    {
+      "epoch": 2.3157894736842106,
+      "grad_norm": 0.5379388928413391,
+      "learning_rate": 0.0002,
+      "loss": 0.8034,
+      "step": 3190
+    },
+    {
+      "epoch": 2.323049001814882,
+      "grad_norm": 0.45770326256752014,
+      "learning_rate": 0.0002,
+      "loss": 0.8833,
+      "step": 3200
+    },
+    {
+      "epoch": 2.3303085299455537,
+      "grad_norm": 0.6984533071517944,
+      "learning_rate": 0.0002,
+      "loss": 0.8419,
+      "step": 3210
+    },
+    {
+      "epoch": 2.337568058076225,
+      "grad_norm": 0.6725744605064392,
+      "learning_rate": 0.0002,
+      "loss": 0.7414,
+      "step": 3220
+    },
+    {
+      "epoch": 2.344827586206897,
+      "grad_norm": 1.1247471570968628,
+      "learning_rate": 0.0002,
+      "loss": 0.8657,
+      "step": 3230
+    },
+    {
+      "epoch": 2.352087114337568,
+      "grad_norm": 1.0240263938903809,
+      "learning_rate": 0.0002,
+      "loss": 0.7782,
+      "step": 3240
+    },
+    {
+      "epoch": 2.3593466424682394,
+      "grad_norm": 0.5608096122741699,
+      "learning_rate": 0.0002,
+      "loss": 0.826,
+      "step": 3250
+    },
+    {
+      "epoch": 2.366606170598911,
+      "grad_norm": 0.8294990062713623,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3260
+    },
+    {
+      "epoch": 2.3738656987295825,
+      "grad_norm": 0.6734224557876587,
+      "learning_rate": 0.0002,
+      "loss": 0.9845,
+      "step": 3270
+    },
+    {
+      "epoch": 2.381125226860254,
+      "grad_norm": 0.6862800717353821,
+      "learning_rate": 0.0002,
+      "loss": 0.7921,
+      "step": 3280
+    },
+    {
+      "epoch": 2.3883847549909256,
+      "grad_norm": 0.5442930459976196,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 3290
+    },
+    {
+      "epoch": 2.395644283121597,
+      "grad_norm": 0.5745394229888916,
+      "learning_rate": 0.0002,
+      "loss": 0.8543,
+      "step": 3300
+    },
+    {
+      "epoch": 2.4029038112522687,
+      "grad_norm": 0.6257799863815308,
+      "learning_rate": 0.0002,
+      "loss": 0.833,
+      "step": 3310
+    },
+    {
+      "epoch": 2.4101633393829403,
+      "grad_norm": 0.5608420968055725,
+      "learning_rate": 0.0002,
+      "loss": 0.8191,
+      "step": 3320
+    },
+    {
+      "epoch": 2.417422867513612,
+      "grad_norm": 0.5512017011642456,
+      "learning_rate": 0.0002,
+      "loss": 0.8745,
+      "step": 3330
+    },
+    {
+      "epoch": 2.4246823956442833,
+      "grad_norm": 0.7452999353408813,
+      "learning_rate": 0.0002,
+      "loss": 0.8642,
+      "step": 3340
+    },
+    {
+      "epoch": 2.4319419237749544,
+      "grad_norm": 0.4604301452636719,
+      "learning_rate": 0.0002,
+      "loss": 0.8417,
+      "step": 3350
+    },
+    {
+      "epoch": 2.439201451905626,
+      "grad_norm": 0.8225823640823364,
+      "learning_rate": 0.0002,
+      "loss": 0.9148,
+      "step": 3360
+    },
+    {
+      "epoch": 2.4464609800362975,
+      "grad_norm": 0.8604981303215027,
+      "learning_rate": 0.0002,
+      "loss": 0.8144,
+      "step": 3370
+    },
+    {
+      "epoch": 2.453720508166969,
+      "grad_norm": 0.6620925664901733,
+      "learning_rate": 0.0002,
+      "loss": 0.9276,
+      "step": 3380
+    },
+    {
+      "epoch": 2.4609800362976406,
+      "grad_norm": 0.4750158488750458,
+      "learning_rate": 0.0002,
+      "loss": 0.8381,
+      "step": 3390
+    },
+    {
+      "epoch": 2.468239564428312,
+      "grad_norm": 0.6061418056488037,
+      "learning_rate": 0.0002,
+      "loss": 0.8249,
+      "step": 3400
+    },
+    {
+      "epoch": 2.4754990925589837,
+      "grad_norm": 0.5977247357368469,
+      "learning_rate": 0.0002,
+      "loss": 0.8944,
+      "step": 3410
+    },
+    {
+      "epoch": 2.4827586206896552,
+      "grad_norm": 0.5004227757453918,
+      "learning_rate": 0.0002,
+      "loss": 0.8366,
+      "step": 3420
+    },
+    {
+      "epoch": 2.490018148820327,
+      "grad_norm": 0.46428972482681274,
+      "learning_rate": 0.0002,
+      "loss": 0.9133,
+      "step": 3430
+    },
+    {
+      "epoch": 2.4972776769509983,
+      "grad_norm": 0.7658395171165466,
+      "learning_rate": 0.0002,
+      "loss": 0.8534,
+      "step": 3440
+    },
+    {
+      "epoch": 2.50453720508167,
+      "grad_norm": 0.621819794178009,
+      "learning_rate": 0.0002,
+      "loss": 0.9066,
+      "step": 3450
+    },
+    {
+      "epoch": 2.511796733212341,
+      "grad_norm": 0.4556088149547577,
+      "learning_rate": 0.0002,
+      "loss": 0.8481,
+      "step": 3460
+    },
+    {
+      "epoch": 2.519056261343013,
+      "grad_norm": 0.6124140024185181,
+      "learning_rate": 0.0002,
+      "loss": 0.8522,
+      "step": 3470
+    },
+    {
+      "epoch": 2.526315789473684,
+      "grad_norm": 0.6256231665611267,
+      "learning_rate": 0.0002,
+      "loss": 0.8158,
+      "step": 3480
+    },
+    {
+      "epoch": 2.5335753176043556,
+      "grad_norm": 0.5464141964912415,
+      "learning_rate": 0.0002,
+      "loss": 0.9097,
+      "step": 3490
+    },
+    {
+      "epoch": 2.540834845735027,
+      "grad_norm": 0.51471346616745,
+      "learning_rate": 0.0002,
+      "loss": 0.8708,
+      "step": 3500
+    },
+    {
+      "epoch": 2.5480943738656987,
+      "grad_norm": 0.5326165556907654,
+      "learning_rate": 0.0002,
+      "loss": 0.9042,
+      "step": 3510
+    },
+    {
+      "epoch": 2.5553539019963702,
+      "grad_norm": 0.4750378429889679,
+      "learning_rate": 0.0002,
+      "loss": 0.8749,
+      "step": 3520
+    },
+    {
+      "epoch": 2.562613430127042,
+      "grad_norm": 0.5292279124259949,
+      "learning_rate": 0.0002,
+      "loss": 0.8248,
+      "step": 3530
+    },
+    {
+      "epoch": 2.5698729582577133,
+      "grad_norm": 0.6145227551460266,
+      "learning_rate": 0.0002,
+      "loss": 0.8808,
+      "step": 3540
+    },
+    {
+      "epoch": 2.577132486388385,
+      "grad_norm": 0.8275189399719238,
+      "learning_rate": 0.0002,
+      "loss": 0.846,
+      "step": 3550
+    },
+    {
+      "epoch": 2.5843920145190564,
+      "grad_norm": 0.5037438273429871,
+      "learning_rate": 0.0002,
+      "loss": 0.875,
+      "step": 3560
+    },
+    {
+      "epoch": 2.5916515426497275,
+      "grad_norm": 0.5838707685470581,
+      "learning_rate": 0.0002,
+      "loss": 0.8324,
+      "step": 3570
+    },
+    {
+      "epoch": 2.5989110707803995,
+      "grad_norm": 0.5398710370063782,
+      "learning_rate": 0.0002,
+      "loss": 0.8716,
+      "step": 3580
+    },
+    {
+      "epoch": 2.6061705989110706,
+      "grad_norm": 0.6115376949310303,
+      "learning_rate": 0.0002,
+      "loss": 0.7843,
+      "step": 3590
+    },
+    {
+      "epoch": 2.613430127041742,
+      "grad_norm": 1.118809461593628,
+      "learning_rate": 0.0002,
+      "loss": 0.8142,
+      "step": 3600
+    },
+    {
+      "epoch": 2.6206896551724137,
+      "grad_norm": 0.7811771631240845,
+      "learning_rate": 0.0002,
+      "loss": 0.8282,
+      "step": 3610
+    },
+    {
+      "epoch": 2.6279491833030852,
+      "grad_norm": 0.5753175020217896,
+      "learning_rate": 0.0002,
+      "loss": 0.7997,
+      "step": 3620
+    },
+    {
+      "epoch": 2.635208711433757,
+      "grad_norm": 0.550829291343689,
+      "learning_rate": 0.0002,
+      "loss": 0.7705,
+      "step": 3630
+    },
+    {
+      "epoch": 2.6424682395644283,
+      "grad_norm": 0.5360019207000732,
+      "learning_rate": 0.0002,
+      "loss": 0.8564,
+      "step": 3640
+    },
+    {
+      "epoch": 2.6497277676951,
+      "grad_norm": 0.63050377368927,
+      "learning_rate": 0.0002,
+      "loss": 0.7648,
+      "step": 3650
+    },
+    {
+      "epoch": 2.6569872958257714,
+      "grad_norm": 0.5833110213279724,
+      "learning_rate": 0.0002,
+      "loss": 0.9156,
+      "step": 3660
+    },
+    {
+      "epoch": 2.664246823956443,
+      "grad_norm": 0.5543047189712524,
+      "learning_rate": 0.0002,
+      "loss": 0.8033,
+      "step": 3670
+    },
+    {
+      "epoch": 2.671506352087114,
+      "grad_norm": 0.6842212080955505,
+      "learning_rate": 0.0002,
+      "loss": 0.8867,
+      "step": 3680
+    },
+    {
+      "epoch": 2.678765880217786,
+      "grad_norm": 0.6859333515167236,
+      "learning_rate": 0.0002,
+      "loss": 0.8569,
+      "step": 3690
+    },
+    {
+      "epoch": 2.686025408348457,
+      "grad_norm": 0.7038410902023315,
+      "learning_rate": 0.0002,
+      "loss": 0.8081,
+      "step": 3700
+    },
+    {
+      "epoch": 2.6932849364791287,
+      "grad_norm": 0.447233647108078,
+      "learning_rate": 0.0002,
+      "loss": 0.8686,
+      "step": 3710
+    },
+    {
+      "epoch": 2.7005444646098002,
+      "grad_norm": 0.6453872919082642,
+      "learning_rate": 0.0002,
+      "loss": 0.8951,
+      "step": 3720
+    },
+    {
+      "epoch": 2.707803992740472,
+      "grad_norm": 0.8025672435760498,
+      "learning_rate": 0.0002,
+      "loss": 0.8292,
+      "step": 3730
+    },
+    {
+      "epoch": 2.7150635208711433,
+      "grad_norm": 0.5997087955474854,
+      "learning_rate": 0.0002,
+      "loss": 0.8488,
+      "step": 3740
+    },
+    {
+      "epoch": 2.722323049001815,
+      "grad_norm": 0.6901142001152039,
+      "learning_rate": 0.0002,
+      "loss": 0.8308,
+      "step": 3750
+    },
+    {
+      "epoch": 2.7295825771324864,
+      "grad_norm": 1.036145567893982,
+      "learning_rate": 0.0002,
+      "loss": 0.8517,
+      "step": 3760
+    },
+    {
+      "epoch": 2.736842105263158,
+      "grad_norm": 0.7207072377204895,
+      "learning_rate": 0.0002,
+      "loss": 0.8455,
+      "step": 3770
+    },
+    {
+      "epoch": 2.7441016333938295,
+      "grad_norm": 1.0452989339828491,
+      "learning_rate": 0.0002,
+      "loss": 0.8009,
+      "step": 3780
+    },
+    {
+      "epoch": 2.751361161524501,
+      "grad_norm": 0.5615278482437134,
+      "learning_rate": 0.0002,
+      "loss": 0.8868,
+      "step": 3790
+    },
+    {
+      "epoch": 2.7586206896551726,
+      "grad_norm": 0.46439215540885925,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3800
+    },
+    {
+      "epoch": 2.7658802177858437,
+      "grad_norm": 0.7134917974472046,
+      "learning_rate": 0.0002,
+      "loss": 0.7847,
+      "step": 3810
+    },
+    {
+      "epoch": 2.7731397459165157,
+      "grad_norm": 0.5139115452766418,
+      "learning_rate": 0.0002,
+      "loss": 0.8315,
+      "step": 3820
+    },
+    {
+      "epoch": 2.780399274047187,
+      "grad_norm": 0.8595781326293945,
+      "learning_rate": 0.0002,
+      "loss": 0.8995,
+      "step": 3830
+    },
+    {
+      "epoch": 2.7876588021778583,
+      "grad_norm": 0.544614851474762,
+      "learning_rate": 0.0002,
+      "loss": 0.9079,
+      "step": 3840
+    },
+    {
+      "epoch": 2.79491833030853,
+      "grad_norm": 0.6073850393295288,
+      "learning_rate": 0.0002,
+      "loss": 0.8322,
+      "step": 3850
+    },
+    {
+      "epoch": 2.8021778584392014,
+      "grad_norm": 0.8224069476127625,
+      "learning_rate": 0.0002,
+      "loss": 0.8277,
+      "step": 3860
+    },
+    {
+      "epoch": 2.809437386569873,
+      "grad_norm": 0.5347970128059387,
+      "learning_rate": 0.0002,
+      "loss": 0.815,
+      "step": 3870
+    },
+    {
+      "epoch": 2.8166969147005445,
+      "grad_norm": 0.6799601912498474,
+      "learning_rate": 0.0002,
+      "loss": 0.8219,
+      "step": 3880
+    },
+    {
+      "epoch": 2.823956442831216,
+      "grad_norm": 0.5219197869300842,
+      "learning_rate": 0.0002,
+      "loss": 0.8796,
+      "step": 3890
+    },
+    {
+      "epoch": 2.8312159709618876,
+      "grad_norm": 0.5710130333900452,
+      "learning_rate": 0.0002,
+      "loss": 0.8649,
+      "step": 3900
+    },
+    {
+      "epoch": 2.838475499092559,
+      "grad_norm": 0.5857213139533997,
+      "learning_rate": 0.0002,
+      "loss": 0.8741,
+      "step": 3910
+    },
+    {
+      "epoch": 2.8457350272232302,
+      "grad_norm": 0.5206644535064697,
+      "learning_rate": 0.0002,
+      "loss": 0.8154,
+      "step": 3920
+    },
+    {
+      "epoch": 2.8529945553539022,
+      "grad_norm": 1.2902015447616577,
+      "learning_rate": 0.0002,
+      "loss": 0.9328,
+      "step": 3930
+    },
+    {
+      "epoch": 2.8602540834845733,
+      "grad_norm": 0.5252797603607178,
+      "learning_rate": 0.0002,
+      "loss": 0.8192,
+      "step": 3940
+    },
+    {
+      "epoch": 2.867513611615245,
+      "grad_norm": 0.5925108790397644,
+      "learning_rate": 0.0002,
+      "loss": 0.9047,
+      "step": 3950
+    },
+    {
+      "epoch": 2.8747731397459164,
+      "grad_norm": 0.9719610810279846,
+      "learning_rate": 0.0002,
+      "loss": 0.8611,
+      "step": 3960
+    },
+    {
+      "epoch": 2.882032667876588,
+      "grad_norm": 0.6834747195243835,
+      "learning_rate": 0.0002,
+      "loss": 0.8531,
+      "step": 3970
+    },
+    {
+      "epoch": 2.8892921960072595,
+      "grad_norm": 0.6952353119850159,
+      "learning_rate": 0.0002,
+      "loss": 0.8124,
+      "step": 3980
+    },
+    {
+      "epoch": 2.896551724137931,
+      "grad_norm": 0.49889910221099854,
+      "learning_rate": 0.0002,
+      "loss": 0.8365,
+      "step": 3990
+    },
+    {
+      "epoch": 2.9038112522686026,
+      "grad_norm": 0.5007026791572571,
+      "learning_rate": 0.0002,
+      "loss": 0.8197,
+      "step": 4000
+    },
+    {
+      "epoch": 2.911070780399274,
+      "grad_norm": 0.5474239587783813,
+      "learning_rate": 0.0002,
+      "loss": 0.7752,
+      "step": 4010
+    },
+    {
+      "epoch": 2.9183303085299457,
+      "grad_norm": 0.6617428660392761,
+      "learning_rate": 0.0002,
+      "loss": 0.8579,
+      "step": 4020
+    },
+    {
+      "epoch": 2.925589836660617,
+      "grad_norm": 0.6097776293754578,
+      "learning_rate": 0.0002,
+      "loss": 0.8583,
+      "step": 4030
+    },
+    {
+      "epoch": 2.9328493647912888,
+      "grad_norm": 0.5985828638076782,
+      "learning_rate": 0.0002,
+      "loss": 0.8296,
+      "step": 4040
+    },
+    {
+      "epoch": 2.94010889292196,
+      "grad_norm": 0.769488513469696,
+      "learning_rate": 0.0002,
+      "loss": 0.9134,
+      "step": 4050
+    },
+    {
+      "epoch": 2.9473684210526314,
+      "grad_norm": 0.5167055130004883,
+      "learning_rate": 0.0002,
+      "loss": 0.8148,
+      "step": 4060
+    },
+    {
+      "epoch": 2.954627949183303,
+      "grad_norm": 0.6265496015548706,
+      "learning_rate": 0.0002,
+      "loss": 0.8665,
+      "step": 4070
+    },
+    {
+      "epoch": 2.9618874773139745,
+      "grad_norm": 1.2644082307815552,
+      "learning_rate": 0.0002,
+      "loss": 0.9218,
+      "step": 4080
+    },
+    {
+      "epoch": 2.969147005444646,
+      "grad_norm": 0.6007561087608337,
+      "learning_rate": 0.0002,
+      "loss": 0.8026,
+      "step": 4090
+    },
+    {
+      "epoch": 2.9764065335753176,
+      "grad_norm": 0.47984927892684937,
+      "learning_rate": 0.0002,
+      "loss": 0.8262,
+      "step": 4100
+    },
+    {
+      "epoch": 2.983666061705989,
+      "grad_norm": 1.128198504447937,
+      "learning_rate": 0.0002,
+      "loss": 0.7897,
+      "step": 4110
+    },
+    {
+      "epoch": 2.9909255898366607,
+      "grad_norm": 0.526292085647583,
+      "learning_rate": 0.0002,
+      "loss": 0.7883,
+      "step": 4120
+    },
+    {
+      "epoch": 2.9981851179673322,
+      "grad_norm": 0.5462674498558044,
+      "learning_rate": 0.0002,
+      "loss": 0.7801,
+      "step": 4130
+    },
+    {
+      "epoch": 2.9996370235934666,
+      "eval_loss": 1.1093357801437378,
+      "eval_runtime": 46.2498,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 4132
+    },
+    {
+      "epoch": 3.0054446460980038,
+      "grad_norm": 0.48288026452064514,
+      "learning_rate": 0.0002,
+      "loss": 0.721,
+      "step": 4140
+    },
+    {
+      "epoch": 3.0127041742286753,
+      "grad_norm": 1.0181782245635986,
+      "learning_rate": 0.0002,
+      "loss": 0.6769,
+      "step": 4150
+    },
+    {
+      "epoch": 3.019963702359347,
+      "grad_norm": 0.7718019485473633,
+      "learning_rate": 0.0002,
+      "loss": 0.7185,
+      "step": 4160
+    },
+    {
+      "epoch": 3.027223230490018,
+      "grad_norm": 0.7492219805717468,
+      "learning_rate": 0.0002,
+      "loss": 0.6552,
+      "step": 4170
+    },
+    {
+      "epoch": 3.0344827586206895,
+      "grad_norm": 0.9363632798194885,
+      "learning_rate": 0.0002,
+      "loss": 0.6678,
+      "step": 4180
+    },
+    {
+      "epoch": 3.041742286751361,
+      "grad_norm": 0.6888533234596252,
+      "learning_rate": 0.0002,
+      "loss": 0.7187,
+      "step": 4190
+    },
+    {
+      "epoch": 3.0490018148820326,
+      "grad_norm": 0.7072834968566895,
+      "learning_rate": 0.0002,
+      "loss": 0.6469,
+      "step": 4200
+    },
+    {
+      "epoch": 3.056261343012704,
+      "grad_norm": 0.7182047963142395,
+      "learning_rate": 0.0002,
+      "loss": 0.6387,
+      "step": 4210
+    },
+    {
+      "epoch": 3.0635208711433757,
+      "grad_norm": 0.7194355130195618,
+      "learning_rate": 0.0002,
+      "loss": 0.6385,
+      "step": 4220
+    },
+    {
+      "epoch": 3.0707803992740472,
+      "grad_norm": 0.9454023838043213,
+      "learning_rate": 0.0002,
+      "loss": 0.5812,
+      "step": 4230
+    },
+    {
+      "epoch": 3.0780399274047188,
+      "grad_norm": 0.838657557964325,
+      "learning_rate": 0.0002,
+      "loss": 0.6036,
+      "step": 4240
+    },
+    {
+      "epoch": 3.0852994555353903,
+      "grad_norm": 0.740113377571106,
+      "learning_rate": 0.0002,
+      "loss": 0.646,
+      "step": 4250
+    },
+    {
+      "epoch": 3.092558983666062,
+      "grad_norm": 0.6616561412811279,
+      "learning_rate": 0.0002,
+      "loss": 0.604,
+      "step": 4260
+    },
+    {
+      "epoch": 3.0998185117967334,
+      "grad_norm": 0.8846506476402283,
+      "learning_rate": 0.0002,
+      "loss": 0.6462,
+      "step": 4270
+    },
+    {
+      "epoch": 3.107078039927405,
+      "grad_norm": 0.6322125792503357,
+      "learning_rate": 0.0002,
+      "loss": 0.6037,
+      "step": 4280
+    },
+    {
+      "epoch": 3.114337568058076,
+      "grad_norm": 0.7461467385292053,
+      "learning_rate": 0.0002,
+      "loss": 0.5953,
+      "step": 4290
+    },
+    {
+      "epoch": 3.1215970961887476,
+      "grad_norm": 0.8251287341117859,
+      "learning_rate": 0.0002,
+      "loss": 0.6099,
+      "step": 4300
+    },
+    {
+      "epoch": 3.128856624319419,
+      "grad_norm": 0.8767673373222351,
+      "learning_rate": 0.0002,
+      "loss": 0.6284,
+      "step": 4310
+    },
+    {
+      "epoch": 3.1361161524500907,
+      "grad_norm": 0.7758759260177612,
+      "learning_rate": 0.0002,
+      "loss": 0.7535,
+      "step": 4320
+    },
+    {
+      "epoch": 3.143375680580762,
+      "grad_norm": 1.1056879758834839,
+      "learning_rate": 0.0002,
+      "loss": 0.6624,
+      "step": 4330
+    },
+    {
+      "epoch": 3.1506352087114338,
+      "grad_norm": 0.8259835243225098,
+      "learning_rate": 0.0002,
+      "loss": 0.691,
+      "step": 4340
+    },
+    {
+      "epoch": 3.1578947368421053,
+      "grad_norm": 0.6607027053833008,
+      "learning_rate": 0.0002,
+      "loss": 0.6635,
+      "step": 4350
+    },
+    {
+      "epoch": 3.165154264972777,
+      "grad_norm": 0.7983301281929016,
+      "learning_rate": 0.0002,
+      "loss": 0.5911,
+      "step": 4360
+    },
+    {
+      "epoch": 3.1724137931034484,
+      "grad_norm": 0.6725239157676697,
+      "learning_rate": 0.0002,
+      "loss": 0.6496,
+      "step": 4370
+    },
+    {
+      "epoch": 3.17967332123412,
+      "grad_norm": 0.9052095413208008,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 4380
+    },
+    {
+      "epoch": 3.1869328493647915,
+      "grad_norm": 0.8131307363510132,
+      "learning_rate": 0.0002,
+      "loss": 0.6877,
+      "step": 4390
+    },
+    {
+      "epoch": 3.1941923774954626,
+      "grad_norm": 0.6435626149177551,
+      "learning_rate": 0.0002,
+      "loss": 0.6384,
+      "step": 4400
+    },
+    {
+      "epoch": 3.201451905626134,
+      "grad_norm": 0.84367436170578,
+      "learning_rate": 0.0002,
+      "loss": 0.5819,
+      "step": 4410
+    },
+    {
+      "epoch": 3.2087114337568057,
+      "grad_norm": 1.5018867254257202,
+      "learning_rate": 0.0002,
+      "loss": 0.6104,
+      "step": 4420
+    },
+    {
+      "epoch": 3.215970961887477,
+      "grad_norm": 0.7019091844558716,
+      "learning_rate": 0.0002,
+      "loss": 0.6838,
+      "step": 4430
+    },
+    {
+      "epoch": 3.2232304900181488,
+      "grad_norm": 0.9164197444915771,
+      "learning_rate": 0.0002,
+      "loss": 0.6153,
+      "step": 4440
+    },
+    {
+      "epoch": 3.2304900181488203,
+      "grad_norm": 0.7890861630439758,
+      "learning_rate": 0.0002,
+      "loss": 0.6618,
+      "step": 4450
+    },
+    {
+      "epoch": 3.237749546279492,
+      "grad_norm": 0.6517660617828369,
+      "learning_rate": 0.0002,
+      "loss": 0.6401,
+      "step": 4460
+    },
+    {
+      "epoch": 3.2450090744101634,
+      "grad_norm": 1.10188889503479,
+      "learning_rate": 0.0002,
+      "loss": 0.6699,
+      "step": 4470
+    },
+    {
+      "epoch": 3.252268602540835,
+      "grad_norm": 0.8158330917358398,
+      "learning_rate": 0.0002,
+      "loss": 0.6356,
+      "step": 4480
+    },
+    {
+      "epoch": 3.2595281306715065,
+      "grad_norm": 0.7663109302520752,
+      "learning_rate": 0.0002,
+      "loss": 0.7757,
+      "step": 4490
+    },
+    {
+      "epoch": 3.266787658802178,
+      "grad_norm": 0.8473444581031799,
+      "learning_rate": 0.0002,
+      "loss": 0.6539,
+      "step": 4500
+    },
+    {
+      "epoch": 3.274047186932849,
+      "grad_norm": 0.9724768996238708,
+      "learning_rate": 0.0002,
+      "loss": 0.6511,
+      "step": 4510
+    },
+    {
+      "epoch": 3.281306715063521,
+      "grad_norm": 0.8516759276390076,
+      "learning_rate": 0.0002,
+      "loss": 0.5464,
+      "step": 4520
+    },
+    {
+      "epoch": 3.288566243194192,
+      "grad_norm": 0.7543437480926514,
+      "learning_rate": 0.0002,
+      "loss": 0.6534,
+      "step": 4530
+    },
+    {
+      "epoch": 3.2958257713248638,
+      "grad_norm": 1.0472029447555542,
+      "learning_rate": 0.0002,
+      "loss": 0.6095,
+      "step": 4540
+    },
+    {
+      "epoch": 3.3030852994555353,
+      "grad_norm": 0.6240826845169067,
+      "learning_rate": 0.0002,
+      "loss": 0.6216,
+      "step": 4550
+    },
+    {
+      "epoch": 3.310344827586207,
+      "grad_norm": 0.9957774877548218,
+      "learning_rate": 0.0002,
+      "loss": 0.6223,
+      "step": 4560
+    },
+    {
+      "epoch": 3.3176043557168784,
+      "grad_norm": 0.6448912620544434,
+      "learning_rate": 0.0002,
+      "loss": 0.618,
+      "step": 4570
+    },
+    {
+      "epoch": 3.32486388384755,
+      "grad_norm": 0.7519692778587341,
+      "learning_rate": 0.0002,
+      "loss": 0.6188,
+      "step": 4580
+    },
+    {
+      "epoch": 3.3321234119782215,
+      "grad_norm": 0.7367453575134277,
+      "learning_rate": 0.0002,
+      "loss": 0.6672,
+      "step": 4590
+    },
+    {
+      "epoch": 3.339382940108893,
+      "grad_norm": 0.8064960837364197,
+      "learning_rate": 0.0002,
+      "loss": 0.6517,
+      "step": 4600
+    },
+    {
+      "epoch": 3.3466424682395646,
+      "grad_norm": 0.7664631009101868,
+      "learning_rate": 0.0002,
+      "loss": 0.6062,
+      "step": 4610
+    },
+    {
+      "epoch": 3.353901996370236,
+      "grad_norm": 0.7803396582603455,
+      "learning_rate": 0.0002,
+      "loss": 0.6834,
+      "step": 4620
+    },
+    {
+      "epoch": 3.3611615245009077,
+      "grad_norm": 0.9141599535942078,
+      "learning_rate": 0.0002,
+      "loss": 0.6961,
+      "step": 4630
+    },
+    {
+      "epoch": 3.3684210526315788,
+      "grad_norm": 0.9719856381416321,
+      "learning_rate": 0.0002,
+      "loss": 0.6889,
+      "step": 4640
+    },
+    {
+      "epoch": 3.3756805807622503,
+      "grad_norm": 0.9223218560218811,
+      "learning_rate": 0.0002,
+      "loss": 0.6914,
+      "step": 4650
+    },
+    {
+      "epoch": 3.382940108892922,
+      "grad_norm": 0.7289277911186218,
+      "learning_rate": 0.0002,
+      "loss": 0.5981,
+      "step": 4660
+    },
+    {
+      "epoch": 3.3901996370235934,
+      "grad_norm": 1.039724349975586,
+      "learning_rate": 0.0002,
+      "loss": 0.595,
+      "step": 4670
+    },
+    {
+      "epoch": 3.397459165154265,
+      "grad_norm": 1.397438883781433,
+      "learning_rate": 0.0002,
+      "loss": 0.8121,
+      "step": 4680
+    },
+    {
+      "epoch": 3.4047186932849365,
+      "grad_norm": 1.0069999694824219,
+      "learning_rate": 0.0002,
+      "loss": 0.6334,
+      "step": 4690
+    },
+    {
+      "epoch": 3.411978221415608,
+      "grad_norm": 0.816291332244873,
+      "learning_rate": 0.0002,
+      "loss": 0.6598,
+      "step": 4700
+    },
+    {
+      "epoch": 3.4192377495462796,
+      "grad_norm": 1.2831530570983887,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 4710
+    },
+    {
+      "epoch": 3.426497277676951,
+      "grad_norm": 0.9573889970779419,
+      "learning_rate": 0.0002,
+      "loss": 0.6625,
+      "step": 4720
+    },
+    {
+      "epoch": 3.4337568058076227,
+      "grad_norm": 0.7685632705688477,
+      "learning_rate": 0.0002,
+      "loss": 0.7279,
+      "step": 4730
+    },
+    {
+      "epoch": 3.441016333938294,
+      "grad_norm": 0.7019195556640625,
+      "learning_rate": 0.0002,
+      "loss": 0.6104,
+      "step": 4740
+    },
+    {
+      "epoch": 3.4482758620689653,
+      "grad_norm": 0.7244833707809448,
+      "learning_rate": 0.0002,
+      "loss": 0.7606,
+      "step": 4750
+    },
+    {
+      "epoch": 3.455535390199637,
+      "grad_norm": 1.3468551635742188,
+      "learning_rate": 0.0002,
+      "loss": 0.6951,
+      "step": 4760
+    },
+    {
+      "epoch": 3.4627949183303084,
+      "grad_norm": 0.822846531867981,
+      "learning_rate": 0.0002,
+      "loss": 0.6945,
+      "step": 4770
+    },
+    {
+      "epoch": 3.47005444646098,
+      "grad_norm": 0.7311608195304871,
+      "learning_rate": 0.0002,
+      "loss": 0.6431,
+      "step": 4780
+    },
+    {
+      "epoch": 3.4773139745916515,
+      "grad_norm": 0.9466770887374878,
+      "learning_rate": 0.0002,
+      "loss": 0.7019,
+      "step": 4790
+    },
+    {
+      "epoch": 3.484573502722323,
+      "grad_norm": 1.1527155637741089,
+      "learning_rate": 0.0002,
+      "loss": 0.7767,
+      "step": 4800
+    },
+    {
+      "epoch": 3.4918330308529946,
+      "grad_norm": 1.1288906335830688,
+      "learning_rate": 0.0002,
+      "loss": 0.6882,
+      "step": 4810
+    },
+    {
+      "epoch": 3.499092558983666,
+      "grad_norm": 0.9096164107322693,
+      "learning_rate": 0.0002,
+      "loss": 0.6564,
+      "step": 4820
+    },
+    {
+      "epoch": 3.5063520871143377,
+      "grad_norm": 0.7988565564155579,
+      "learning_rate": 0.0002,
+      "loss": 0.6127,
+      "step": 4830
+    },
+    {
+      "epoch": 3.513611615245009,
+      "grad_norm": 0.7183415293693542,
+      "learning_rate": 0.0002,
+      "loss": 0.7004,
+      "step": 4840
+    },
+    {
+      "epoch": 3.5208711433756807,
+      "grad_norm": 0.6614915132522583,
+      "learning_rate": 0.0002,
+      "loss": 0.74,
+      "step": 4850
+    },
+    {
+      "epoch": 3.528130671506352,
+      "grad_norm": 0.8609521985054016,
+      "learning_rate": 0.0002,
+      "loss": 0.7271,
+      "step": 4860
+    },
+    {
+      "epoch": 3.535390199637024,
+      "grad_norm": 0.86552894115448,
+      "learning_rate": 0.0002,
+      "loss": 0.6664,
+      "step": 4870
+    },
+    {
+      "epoch": 3.542649727767695,
+      "grad_norm": 0.6926496028900146,
+      "learning_rate": 0.0002,
+      "loss": 0.6432,
+      "step": 4880
+    },
+    {
+      "epoch": 3.5499092558983665,
+      "grad_norm": 0.8157467246055603,
+      "learning_rate": 0.0002,
+      "loss": 0.7117,
+      "step": 4890
+    },
+    {
+      "epoch": 3.557168784029038,
+      "grad_norm": 0.9085357189178467,
+      "learning_rate": 0.0002,
+      "loss": 0.6201,
+      "step": 4900
+    },
+    {
+      "epoch": 3.5644283121597096,
+      "grad_norm": 0.6322644948959351,
+      "learning_rate": 0.0002,
+      "loss": 0.6521,
+      "step": 4910
+    },
+    {
+      "epoch": 3.571687840290381,
+      "grad_norm": 1.263205885887146,
+      "learning_rate": 0.0002,
+      "loss": 0.6607,
+      "step": 4920
+    },
+    {
+      "epoch": 3.5789473684210527,
+      "grad_norm": 0.8901070356369019,
+      "learning_rate": 0.0002,
+      "loss": 0.6657,
+      "step": 4930
+    },
+    {
+      "epoch": 3.586206896551724,
+      "grad_norm": 0.7983952164649963,
+      "learning_rate": 0.0002,
+      "loss": 0.6434,
+      "step": 4940
+    },
+    {
+      "epoch": 3.5934664246823957,
+      "grad_norm": 0.9887813925743103,
+      "learning_rate": 0.0002,
+      "loss": 0.6861,
+      "step": 4950
+    },
+    {
+      "epoch": 3.6007259528130673,
+      "grad_norm": 0.7895187735557556,
+      "learning_rate": 0.0002,
+      "loss": 0.6502,
+      "step": 4960
+    },
+    {
+      "epoch": 3.6079854809437384,
+      "grad_norm": 0.9685819745063782,
+      "learning_rate": 0.0002,
+      "loss": 0.7111,
+      "step": 4970
+    },
+    {
+      "epoch": 3.6152450090744104,
+      "grad_norm": 0.6576591730117798,
+      "learning_rate": 0.0002,
+      "loss": 0.6915,
+      "step": 4980
+    },
+    {
+      "epoch": 3.6225045372050815,
+      "grad_norm": 0.856985330581665,
+      "learning_rate": 0.0002,
+      "loss": 0.6195,
+      "step": 4990
+    },
+    {
+      "epoch": 3.629764065335753,
+      "grad_norm": 0.7230252623558044,
+      "learning_rate": 0.0002,
+      "loss": 0.6318,
+      "step": 5000
+    },
+    {
+      "epoch": 3.6370235934664246,
+      "grad_norm": 0.8260893821716309,
+      "learning_rate": 0.0002,
+      "loss": 0.742,
+      "step": 5010
+    },
+    {
+      "epoch": 3.644283121597096,
+      "grad_norm": 0.7635950446128845,
+      "learning_rate": 0.0002,
+      "loss": 0.7223,
+      "step": 5020
+    },
+    {
+      "epoch": 3.6515426497277677,
+      "grad_norm": 0.7060768604278564,
+      "learning_rate": 0.0002,
+      "loss": 0.6837,
+      "step": 5030
+    },
+    {
+      "epoch": 3.658802177858439,
+      "grad_norm": 0.8020303249359131,
+      "learning_rate": 0.0002,
+      "loss": 0.6921,
+      "step": 5040
+    },
+    {
+      "epoch": 3.6660617059891107,
+      "grad_norm": 0.8530341386795044,
+      "learning_rate": 0.0002,
+      "loss": 0.6446,
+      "step": 5050
+    },
+    {
+      "epoch": 3.6733212341197823,
+      "grad_norm": 0.6667101979255676,
+      "learning_rate": 0.0002,
+      "loss": 0.7222,
+      "step": 5060
+    },
+    {
+      "epoch": 3.680580762250454,
+      "grad_norm": 0.7385406494140625,
+      "learning_rate": 0.0002,
+      "loss": 0.7081,
+      "step": 5070
+    },
+    {
+      "epoch": 3.6878402903811254,
+      "grad_norm": 0.7753380537033081,
+      "learning_rate": 0.0002,
+      "loss": 0.7035,
+      "step": 5080
+    },
+    {
+      "epoch": 3.695099818511797,
+      "grad_norm": 0.7516207098960876,
+      "learning_rate": 0.0002,
+      "loss": 0.6491,
+      "step": 5090
+    },
+    {
+      "epoch": 3.702359346642468,
+      "grad_norm": 0.8171586394309998,
+      "learning_rate": 0.0002,
+      "loss": 0.672,
+      "step": 5100
+    },
+    {
+      "epoch": 3.70961887477314,
+      "grad_norm": 1.0796279907226562,
+      "learning_rate": 0.0002,
+      "loss": 0.7459,
+      "step": 5110
+    },
+    {
+      "epoch": 3.716878402903811,
+      "grad_norm": 0.6957688927650452,
+      "learning_rate": 0.0002,
+      "loss": 0.5948,
+      "step": 5120
+    },
+    {
+      "epoch": 3.7241379310344827,
+      "grad_norm": 0.8550161719322205,
+      "learning_rate": 0.0002,
+      "loss": 0.7515,
+      "step": 5130
+    },
+    {
+      "epoch": 3.731397459165154,
+      "grad_norm": 0.9396728277206421,
+      "learning_rate": 0.0002,
+      "loss": 0.7286,
+      "step": 5140
+    },
+    {
+      "epoch": 3.7386569872958257,
+      "grad_norm": 1.4264805316925049,
+      "learning_rate": 0.0002,
+      "loss": 0.7594,
+      "step": 5150
+    },
+    {
+      "epoch": 3.7459165154264973,
+      "grad_norm": 0.8725108504295349,
+      "learning_rate": 0.0002,
+      "loss": 0.6575,
+      "step": 5160
+    },
+    {
+      "epoch": 3.753176043557169,
+      "grad_norm": 1.0346195697784424,
+      "learning_rate": 0.0002,
+      "loss": 0.6778,
+      "step": 5170
+    },
+    {
+      "epoch": 3.7604355716878404,
+      "grad_norm": 0.5395554304122925,
+      "learning_rate": 0.0002,
+      "loss": 0.6371,
+      "step": 5180
+    },
+    {
+      "epoch": 3.767695099818512,
+      "grad_norm": 1.3153616189956665,
+      "learning_rate": 0.0002,
+      "loss": 0.7308,
+      "step": 5190
+    },
+    {
+      "epoch": 3.7749546279491835,
+      "grad_norm": 0.9879828691482544,
+      "learning_rate": 0.0002,
+      "loss": 0.78,
+      "step": 5200
+    },
+    {
+      "epoch": 3.7822141560798546,
+      "grad_norm": 0.8876672983169556,
+      "learning_rate": 0.0002,
+      "loss": 0.7068,
+      "step": 5210
+    },
+    {
+      "epoch": 3.7894736842105265,
+      "grad_norm": 0.8363267779350281,
+      "learning_rate": 0.0002,
+      "loss": 0.6283,
+      "step": 5220
+    },
+    {
+      "epoch": 3.7967332123411976,
+      "grad_norm": 0.637294590473175,
+      "learning_rate": 0.0002,
+      "loss": 0.6255,
+      "step": 5230
+    },
+    {
+      "epoch": 3.803992740471869,
+      "grad_norm": 1.1408970355987549,
+      "learning_rate": 0.0002,
+      "loss": 0.6685,
+      "step": 5240
+    },
+    {
+      "epoch": 3.8112522686025407,
+      "grad_norm": 1.0128360986709595,
+      "learning_rate": 0.0002,
+      "loss": 0.6761,
+      "step": 5250
+    },
+    {
+      "epoch": 3.8185117967332123,
+      "grad_norm": 0.8061144351959229,
+      "learning_rate": 0.0002,
+      "loss": 0.6764,
+      "step": 5260
+    },
+    {
+      "epoch": 3.825771324863884,
+      "grad_norm": 0.9626626968383789,
+      "learning_rate": 0.0002,
+      "loss": 0.7254,
+      "step": 5270
+    },
+    {
+      "epoch": 3.8330308529945554,
+      "grad_norm": 0.9013627171516418,
+      "learning_rate": 0.0002,
+      "loss": 0.7367,
+      "step": 5280
+    },
+    {
+      "epoch": 3.840290381125227,
+      "grad_norm": 0.8411344289779663,
+      "learning_rate": 0.0002,
+      "loss": 0.6806,
+      "step": 5290
+    },
+    {
+      "epoch": 3.8475499092558985,
+      "grad_norm": 0.7426059246063232,
+      "learning_rate": 0.0002,
+      "loss": 0.6818,
+      "step": 5300
+    },
+    {
+      "epoch": 3.85480943738657,
+      "grad_norm": 1.003413438796997,
+      "learning_rate": 0.0002,
+      "loss": 0.6748,
+      "step": 5310
+    },
+    {
+      "epoch": 3.862068965517241,
+      "grad_norm": 0.7527840733528137,
+      "learning_rate": 0.0002,
+      "loss": 0.8554,
+      "step": 5320
+    },
+    {
+      "epoch": 3.869328493647913,
+      "grad_norm": 0.738610565662384,
+      "learning_rate": 0.0002,
+      "loss": 0.7521,
+      "step": 5330
+    },
+    {
+      "epoch": 3.876588021778584,
+      "grad_norm": 0.7277999520301819,
+      "learning_rate": 0.0002,
+      "loss": 0.7266,
+      "step": 5340
+    },
+    {
+      "epoch": 3.8838475499092557,
+      "grad_norm": 0.5951359272003174,
+      "learning_rate": 0.0002,
+      "loss": 0.7503,
+      "step": 5350
+    },
+    {
+      "epoch": 3.8911070780399273,
+      "grad_norm": 1.043884038925171,
+      "learning_rate": 0.0002,
+      "loss": 0.7447,
+      "step": 5360
+    },
+    {
+      "epoch": 3.898366606170599,
+      "grad_norm": 0.8436498045921326,
+      "learning_rate": 0.0002,
+      "loss": 0.6862,
+      "step": 5370
+    },
+    {
+      "epoch": 3.9056261343012704,
+      "grad_norm": 0.5603365302085876,
+      "learning_rate": 0.0002,
+      "loss": 0.665,
+      "step": 5380
+    },
+    {
+      "epoch": 3.912885662431942,
+      "grad_norm": 1.0128886699676514,
+      "learning_rate": 0.0002,
+      "loss": 0.7098,
+      "step": 5390
+    },
+    {
+      "epoch": 3.9201451905626135,
+      "grad_norm": 0.7970930337905884,
+      "learning_rate": 0.0002,
+      "loss": 0.6707,
+      "step": 5400
+    },
+    {
+      "epoch": 3.927404718693285,
+      "grad_norm": 0.7699369192123413,
+      "learning_rate": 0.0002,
+      "loss": 0.637,
+      "step": 5410
+    },
+    {
+      "epoch": 3.9346642468239565,
+      "grad_norm": 0.800561249256134,
+      "learning_rate": 0.0002,
+      "loss": 0.6742,
+      "step": 5420
+    },
+    {
+      "epoch": 3.941923774954628,
+      "grad_norm": 0.8020331859588623,
+      "learning_rate": 0.0002,
+      "loss": 0.7208,
+      "step": 5430
+    },
+    {
+      "epoch": 3.9491833030852996,
+      "grad_norm": 0.7461140155792236,
+      "learning_rate": 0.0002,
+      "loss": 0.7294,
+      "step": 5440
+    },
+    {
+      "epoch": 3.9564428312159707,
+      "grad_norm": 0.8346918821334839,
+      "learning_rate": 0.0002,
+      "loss": 0.7013,
+      "step": 5450
+    },
+    {
+      "epoch": 3.9637023593466427,
+      "grad_norm": 0.9723302125930786,
+      "learning_rate": 0.0002,
+      "loss": 0.6289,
+      "step": 5460
+    },
+    {
+      "epoch": 3.970961887477314,
+      "grad_norm": 0.6809740662574768,
+      "learning_rate": 0.0002,
+      "loss": 0.8029,
+      "step": 5470
+    },
+    {
+      "epoch": 3.9782214156079854,
+      "grad_norm": 0.7353498339653015,
+      "learning_rate": 0.0002,
+      "loss": 0.6896,
+      "step": 5480
+    },
+    {
+      "epoch": 3.985480943738657,
+      "grad_norm": 0.748009443283081,
+      "learning_rate": 0.0002,
+      "loss": 0.6722,
+      "step": 5490
+    },
+    {
+      "epoch": 3.9927404718693285,
+      "grad_norm": 1.3656195402145386,
+      "learning_rate": 0.0002,
+      "loss": 0.6866,
+      "step": 5500
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.8402108550071716,
+      "learning_rate": 0.0002,
+      "loss": 0.7368,
+      "step": 5510
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.17229425907135,
+      "eval_runtime": 46.2554,
+      "eval_samples_per_second": 9.426,
+      "eval_steps_per_second": 1.189,
+      "step": 5510
+    },
+    {
+      "epoch": 4.007259528130671,
+      "grad_norm": 0.8601235747337341,
+      "learning_rate": 0.0002,
+      "loss": 0.4637,
+      "step": 5520
+    },
+    {
+      "epoch": 4.014519056261343,
+      "grad_norm": 1.2635200023651123,
+      "learning_rate": 0.0002,
+      "loss": 0.4717,
+      "step": 5530
+    },
+    {
+      "epoch": 4.021778584392014,
+      "grad_norm": 1.0257477760314941,
+      "learning_rate": 0.0002,
+      "loss": 0.503,
+      "step": 5540
+    },
+    {
+      "epoch": 4.029038112522686,
+      "grad_norm": 0.9436745047569275,
+      "learning_rate": 0.0002,
+      "loss": 0.4547,
+      "step": 5550
+    },
+    {
+      "epoch": 4.036297640653357,
+      "grad_norm": 0.9443606734275818,
+      "learning_rate": 0.0002,
+      "loss": 0.459,
+      "step": 5560
+    },
+    {
+      "epoch": 4.043557168784029,
+      "grad_norm": 1.3965742588043213,
+      "learning_rate": 0.0002,
+      "loss": 0.5386,
+      "step": 5570
+    },
+    {
+      "epoch": 4.0508166969147,
+      "grad_norm": 0.8973520398139954,
+      "learning_rate": 0.0002,
+      "loss": 0.4248,
+      "step": 5580
+    },
+    {
+      "epoch": 4.058076225045372,
+      "grad_norm": 0.9998409748077393,
+      "learning_rate": 0.0002,
+      "loss": 0.4111,
+      "step": 5590
+    },
+    {
+      "epoch": 4.0653357531760435,
+      "grad_norm": 1.1213387250900269,
+      "learning_rate": 0.0002,
+      "loss": 0.4828,
+      "step": 5600
+    },
+    {
+      "epoch": 4.072595281306715,
+      "grad_norm": 0.7064558863639832,
+      "learning_rate": 0.0002,
+      "loss": 0.439,
+      "step": 5610
+    },
+    {
+      "epoch": 4.0798548094373865,
+      "grad_norm": 1.2390803098678589,
+      "learning_rate": 0.0002,
+      "loss": 0.4607,
+      "step": 5620
+    },
+    {
+      "epoch": 4.087114337568058,
+      "grad_norm": 1.123469591140747,
+      "learning_rate": 0.0002,
+      "loss": 0.5014,
+      "step": 5630
+    },
+    {
+      "epoch": 4.09437386569873,
+      "grad_norm": 1.229573369026184,
+      "learning_rate": 0.0002,
+      "loss": 0.513,
+      "step": 5640
+    },
+    {
+      "epoch": 4.101633393829401,
+      "grad_norm": 1.7182831764221191,
+      "learning_rate": 0.0002,
+      "loss": 0.5258,
+      "step": 5650
+    },
+    {
+      "epoch": 4.108892921960073,
+      "grad_norm": 0.894903302192688,
+      "learning_rate": 0.0002,
+      "loss": 0.5371,
+      "step": 5660
+    },
+    {
+      "epoch": 4.116152450090744,
+      "grad_norm": 0.8754552006721497,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 5670
+    },
+    {
+      "epoch": 4.123411978221416,
+      "grad_norm": 1.2401553392410278,
+      "learning_rate": 0.0002,
+      "loss": 0.491,
+      "step": 5680
+    },
+    {
+      "epoch": 4.130671506352087,
+      "grad_norm": 0.8631148934364319,
+      "learning_rate": 0.0002,
+      "loss": 0.4549,
+      "step": 5690
+    },
+    {
+      "epoch": 4.137931034482759,
+      "grad_norm": 1.1798022985458374,
+      "learning_rate": 0.0002,
+      "loss": 0.487,
+      "step": 5700
+    },
+    {
+      "epoch": 4.14519056261343,
+      "grad_norm": 0.8344549536705017,
+      "learning_rate": 0.0002,
+      "loss": 0.4522,
+      "step": 5710
+    },
+    {
+      "epoch": 4.152450090744102,
+      "grad_norm": 1.2342697381973267,
+      "learning_rate": 0.0002,
+      "loss": 0.4559,
+      "step": 5720
+    },
+    {
+      "epoch": 4.159709618874773,
+      "grad_norm": 1.1601094007492065,
+      "learning_rate": 0.0002,
+      "loss": 0.53,
+      "step": 5730
+    },
+    {
+      "epoch": 4.166969147005445,
+      "grad_norm": 1.2925703525543213,
+      "learning_rate": 0.0002,
+      "loss": 0.4755,
+      "step": 5740
+    },
+    {
+      "epoch": 4.174228675136116,
+      "grad_norm": 1.0870997905731201,
+      "learning_rate": 0.0002,
+      "loss": 0.4973,
+      "step": 5750
+    },
+    {
+      "epoch": 4.181488203266787,
+      "grad_norm": 0.9077792763710022,
+      "learning_rate": 0.0002,
+      "loss": 0.5184,
+      "step": 5760
+    },
+    {
+      "epoch": 4.188747731397459,
+      "grad_norm": 1.009273886680603,
+      "learning_rate": 0.0002,
+      "loss": 0.4606,
+      "step": 5770
+    },
+    {
+      "epoch": 4.19600725952813,
+      "grad_norm": 1.2465530633926392,
+      "learning_rate": 0.0002,
+      "loss": 0.5383,
+      "step": 5780
+    },
+    {
+      "epoch": 4.203266787658802,
+      "grad_norm": 1.2261253595352173,
+      "learning_rate": 0.0002,
+      "loss": 0.4938,
+      "step": 5790
+    },
+    {
+      "epoch": 4.2105263157894735,
+      "grad_norm": 1.1498041152954102,
+      "learning_rate": 0.0002,
+      "loss": 0.5255,
+      "step": 5800
+    },
+    {
+      "epoch": 4.217785843920145,
+      "grad_norm": 1.1966725587844849,
+      "learning_rate": 0.0002,
+      "loss": 0.5269,
+      "step": 5810
+    },
+    {
+      "epoch": 4.2250453720508165,
+      "grad_norm": 1.2651296854019165,
+      "learning_rate": 0.0002,
+      "loss": 0.5626,
+      "step": 5820
+    },
+    {
+      "epoch": 4.2323049001814885,
+      "grad_norm": 1.0388574600219727,
+      "learning_rate": 0.0002,
+      "loss": 0.5213,
+      "step": 5830
+    },
+    {
+      "epoch": 4.23956442831216,
+      "grad_norm": 1.3042771816253662,
+      "learning_rate": 0.0002,
+      "loss": 0.4965,
+      "step": 5840
+    },
+    {
+      "epoch": 4.246823956442832,
+      "grad_norm": 1.1127727031707764,
+      "learning_rate": 0.0002,
+      "loss": 0.5116,
+      "step": 5850
+    },
+    {
+      "epoch": 4.254083484573503,
+      "grad_norm": 0.9653958082199097,
+      "learning_rate": 0.0002,
+      "loss": 0.5197,
+      "step": 5860
+    },
+    {
+      "epoch": 4.261343012704174,
+      "grad_norm": 1.0500504970550537,
+      "learning_rate": 0.0002,
+      "loss": 0.4414,
+      "step": 5870
+    },
+    {
+      "epoch": 4.268602540834846,
+      "grad_norm": 1.1476165056228638,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 5880
+    },
+    {
+      "epoch": 4.275862068965517,
+      "grad_norm": 0.9424414038658142,
+      "learning_rate": 0.0002,
+      "loss": 0.4667,
+      "step": 5890
+    },
+    {
+      "epoch": 4.283121597096189,
+      "grad_norm": 1.3309166431427002,
+      "learning_rate": 0.0002,
+      "loss": 0.5039,
+      "step": 5900
+    },
+    {
+      "epoch": 4.29038112522686,
+      "grad_norm": 1.3025873899459839,
+      "learning_rate": 0.0002,
+      "loss": 0.5472,
+      "step": 5910
+    },
+    {
+      "epoch": 4.297640653357532,
+      "grad_norm": 1.1442325115203857,
+      "learning_rate": 0.0002,
+      "loss": 0.4644,
+      "step": 5920
+    },
+    {
+      "epoch": 4.304900181488203,
+      "grad_norm": 0.9820859432220459,
+      "learning_rate": 0.0002,
+      "loss": 0.5066,
+      "step": 5930
+    },
+    {
+      "epoch": 4.312159709618875,
+      "grad_norm": 0.9615740180015564,
+      "learning_rate": 0.0002,
+      "loss": 0.5224,
+      "step": 5940
+    },
+    {
+      "epoch": 4.319419237749546,
+      "grad_norm": 1.1627109050750732,
+      "learning_rate": 0.0002,
+      "loss": 0.5665,
+      "step": 5950
+    },
+    {
+      "epoch": 4.326678765880218,
+      "grad_norm": 0.9381322860717773,
+      "learning_rate": 0.0002,
+      "loss": 0.4852,
+      "step": 5960
+    },
+    {
+      "epoch": 4.333938294010889,
+      "grad_norm": 0.8154335618019104,
+      "learning_rate": 0.0002,
+      "loss": 0.4532,
+      "step": 5970
+    },
+    {
+      "epoch": 4.341197822141561,
+      "grad_norm": 0.877671480178833,
+      "learning_rate": 0.0002,
+      "loss": 0.5237,
+      "step": 5980
+    },
+    {
+      "epoch": 4.348457350272232,
+      "grad_norm": 1.1742031574249268,
+      "learning_rate": 0.0002,
+      "loss": 0.6113,
+      "step": 5990
+    },
+    {
+      "epoch": 4.3557168784029034,
+      "grad_norm": 1.0352917909622192,
+      "learning_rate": 0.0002,
+      "loss": 0.5704,
+      "step": 6000
+    },
+    {
+      "epoch": 4.362976406533575,
+      "grad_norm": 0.9963878989219666,
+      "learning_rate": 0.0002,
+      "loss": 0.4996,
+      "step": 6010
+    },
+    {
+      "epoch": 4.3702359346642465,
+      "grad_norm": 1.1892237663269043,
+      "learning_rate": 0.0002,
+      "loss": 0.4813,
+      "step": 6020
+    },
+    {
+      "epoch": 4.3774954627949185,
+      "grad_norm": 1.2516111135482788,
+      "learning_rate": 0.0002,
+      "loss": 0.5224,
+      "step": 6030
+    },
+    {
+      "epoch": 4.38475499092559,
+      "grad_norm": 1.2111951112747192,
+      "learning_rate": 0.0002,
+      "loss": 0.5952,
+      "step": 6040
+    },
+    {
+      "epoch": 4.392014519056262,
+      "grad_norm": 1.0820083618164062,
+      "learning_rate": 0.0002,
+      "loss": 0.4275,
+      "step": 6050
+    },
+    {
+      "epoch": 4.399274047186933,
+      "grad_norm": 1.033915638923645,
+      "learning_rate": 0.0002,
+      "loss": 0.5117,
+      "step": 6060
+    },
+    {
+      "epoch": 4.406533575317605,
+      "grad_norm": 1.0635870695114136,
+      "learning_rate": 0.0002,
+      "loss": 0.5431,
+      "step": 6070
+    },
+    {
+      "epoch": 4.413793103448276,
+      "grad_norm": 1.0520414113998413,
+      "learning_rate": 0.0002,
+      "loss": 0.5341,
+      "step": 6080
+    },
+    {
+      "epoch": 4.421052631578947,
+      "grad_norm": 1.0821926593780518,
+      "learning_rate": 0.0002,
+      "loss": 0.512,
+      "step": 6090
+    },
+    {
+      "epoch": 4.428312159709619,
+      "grad_norm": 1.0533246994018555,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 6100
+    },
+    {
+      "epoch": 4.43557168784029,
+      "grad_norm": 0.9231932759284973,
+      "learning_rate": 0.0002,
+      "loss": 0.4577,
+      "step": 6110
+    },
+    {
+      "epoch": 4.442831215970962,
+      "grad_norm": 0.9910260438919067,
+      "learning_rate": 0.0002,
+      "loss": 0.583,
+      "step": 6120
+    },
+    {
+      "epoch": 4.450090744101633,
+      "grad_norm": 1.061949372291565,
+      "learning_rate": 0.0002,
+      "loss": 0.4717,
+      "step": 6130
+    },
+    {
+      "epoch": 4.457350272232305,
+      "grad_norm": 1.2927039861679077,
+      "learning_rate": 0.0002,
+      "loss": 0.5893,
+      "step": 6140
+    },
+    {
+      "epoch": 4.464609800362976,
+      "grad_norm": 1.3966081142425537,
+      "learning_rate": 0.0002,
+      "loss": 0.4684,
+      "step": 6150
+    },
+    {
+      "epoch": 4.471869328493648,
+      "grad_norm": 1.3835992813110352,
+      "learning_rate": 0.0002,
+      "loss": 0.5507,
+      "step": 6160
+    },
+    {
+      "epoch": 4.479128856624319,
+      "grad_norm": 1.0892692804336548,
+      "learning_rate": 0.0002,
+      "loss": 0.5911,
+      "step": 6170
+    },
+    {
+      "epoch": 4.486388384754991,
+      "grad_norm": 1.0318800210952759,
+      "learning_rate": 0.0002,
+      "loss": 0.478,
+      "step": 6180
+    },
+    {
+      "epoch": 4.493647912885662,
+      "grad_norm": 0.8174677491188049,
+      "learning_rate": 0.0002,
+      "loss": 0.5198,
+      "step": 6190
+    },
+    {
+      "epoch": 4.500907441016334,
+      "grad_norm": 1.4157509803771973,
+      "learning_rate": 0.0002,
+      "loss": 0.5387,
+      "step": 6200
+    },
+    {
+      "epoch": 4.508166969147005,
+      "grad_norm": 1.5244114398956299,
+      "learning_rate": 0.0002,
+      "loss": 0.5868,
+      "step": 6210
+    },
+    {
+      "epoch": 4.5154264972776765,
+      "grad_norm": 0.8164850473403931,
+      "learning_rate": 0.0002,
+      "loss": 0.4642,
+      "step": 6220
+    },
+    {
+      "epoch": 4.5226860254083485,
+      "grad_norm": 1.2904746532440186,
+      "learning_rate": 0.0002,
+      "loss": 0.522,
+      "step": 6230
+    },
+    {
+      "epoch": 4.52994555353902,
+      "grad_norm": 0.7987732887268066,
+      "learning_rate": 0.0002,
+      "loss": 0.5103,
+      "step": 6240
+    },
+    {
+      "epoch": 4.537205081669692,
+      "grad_norm": 0.831040620803833,
+      "learning_rate": 0.0002,
+      "loss": 0.4615,
+      "step": 6250
+    },
+    {
+      "epoch": 4.544464609800363,
+      "grad_norm": 0.9545485973358154,
+      "learning_rate": 0.0002,
+      "loss": 0.5065,
+      "step": 6260
+    },
+    {
+      "epoch": 4.551724137931035,
+      "grad_norm": 0.9291793704032898,
+      "learning_rate": 0.0002,
+      "loss": 0.5515,
+      "step": 6270
+    },
+    {
+      "epoch": 4.558983666061706,
+      "grad_norm": 0.8977208733558655,
+      "learning_rate": 0.0002,
+      "loss": 0.4535,
+      "step": 6280
+    },
+    {
+      "epoch": 4.566243194192378,
+      "grad_norm": 1.1768537759780884,
+      "learning_rate": 0.0002,
+      "loss": 0.544,
+      "step": 6290
+    },
+    {
+      "epoch": 4.573502722323049,
+      "grad_norm": 1.0688952207565308,
+      "learning_rate": 0.0002,
+      "loss": 0.5925,
+      "step": 6300
+    },
+    {
+      "epoch": 4.580762250453721,
+      "grad_norm": 0.8800966739654541,
+      "learning_rate": 0.0002,
+      "loss": 0.5207,
+      "step": 6310
+    },
+    {
+      "epoch": 4.588021778584392,
+      "grad_norm": 1.0911834239959717,
+      "learning_rate": 0.0002,
+      "loss": 0.6106,
+      "step": 6320
+    },
+    {
+      "epoch": 4.595281306715064,
+      "grad_norm": 1.1420872211456299,
+      "learning_rate": 0.0002,
+      "loss": 0.5109,
+      "step": 6330
+    },
+    {
+      "epoch": 4.602540834845735,
+      "grad_norm": 1.0215224027633667,
+      "learning_rate": 0.0002,
+      "loss": 0.5147,
+      "step": 6340
+    },
+    {
+      "epoch": 4.609800362976406,
+      "grad_norm": 0.9685489535331726,
+      "learning_rate": 0.0002,
+      "loss": 0.592,
+      "step": 6350
+    },
+    {
+      "epoch": 4.617059891107078,
+      "grad_norm": 1.12773597240448,
+      "learning_rate": 0.0002,
+      "loss": 0.5775,
+      "step": 6360
+    },
+    {
+      "epoch": 4.624319419237749,
+      "grad_norm": 1.0663973093032837,
+      "learning_rate": 0.0002,
+      "loss": 0.5966,
+      "step": 6370
+    },
+    {
+      "epoch": 4.631578947368421,
+      "grad_norm": 1.1707262992858887,
+      "learning_rate": 0.0002,
+      "loss": 0.512,
+      "step": 6380
+    },
+    {
+      "epoch": 4.638838475499092,
+      "grad_norm": 1.0672980546951294,
+      "learning_rate": 0.0002,
+      "loss": 0.5497,
+      "step": 6390
+    },
+    {
+      "epoch": 4.646098003629764,
+      "grad_norm": 1.1464333534240723,
+      "learning_rate": 0.0002,
+      "loss": 0.5699,
+      "step": 6400
+    },
+    {
+      "epoch": 4.653357531760435,
+      "grad_norm": 1.070230484008789,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 6410
+    },
+    {
+      "epoch": 4.660617059891107,
+      "grad_norm": 0.9673764109611511,
+      "learning_rate": 0.0002,
+      "loss": 0.5013,
+      "step": 6420
+    },
+    {
+      "epoch": 4.6678765880217785,
+      "grad_norm": 1.0189043283462524,
+      "learning_rate": 0.0002,
+      "loss": 0.5901,
+      "step": 6430
+    },
+    {
+      "epoch": 4.67513611615245,
+      "grad_norm": 1.185896396636963,
+      "learning_rate": 0.0002,
+      "loss": 0.5193,
+      "step": 6440
+    },
+    {
+      "epoch": 4.682395644283122,
+      "grad_norm": 1.0682812929153442,
+      "learning_rate": 0.0002,
+      "loss": 0.5318,
+      "step": 6450
+    },
+    {
+      "epoch": 4.689655172413794,
+      "grad_norm": 1.3586071729660034,
+      "learning_rate": 0.0002,
+      "loss": 0.5773,
+      "step": 6460
+    },
+    {
+      "epoch": 4.696914700544465,
+      "grad_norm": 0.6561792492866516,
+      "learning_rate": 0.0002,
+      "loss": 0.5482,
+      "step": 6470
+    },
+    {
+      "epoch": 4.704174228675136,
+      "grad_norm": 1.1394113302230835,
+      "learning_rate": 0.0002,
+      "loss": 0.5711,
+      "step": 6480
+    },
+    {
+      "epoch": 4.711433756805808,
+      "grad_norm": 0.9683151245117188,
+      "learning_rate": 0.0002,
+      "loss": 0.5325,
+      "step": 6490
+    },
+    {
+      "epoch": 4.718693284936479,
+      "grad_norm": 1.0247553586959839,
+      "learning_rate": 0.0002,
+      "loss": 0.5139,
+      "step": 6500
+    },
+    {
+      "epoch": 4.725952813067151,
+      "grad_norm": 0.8046169281005859,
+      "learning_rate": 0.0002,
+      "loss": 0.5794,
+      "step": 6510
+    },
+    {
+      "epoch": 4.733212341197822,
+      "grad_norm": 1.0710240602493286,
+      "learning_rate": 0.0002,
+      "loss": 0.5471,
+      "step": 6520
+    },
+    {
+      "epoch": 4.740471869328494,
+      "grad_norm": 0.9438924193382263,
+      "learning_rate": 0.0002,
+      "loss": 0.4805,
+      "step": 6530
+    },
+    {
+      "epoch": 4.747731397459165,
+      "grad_norm": 0.869162380695343,
+      "learning_rate": 0.0002,
+      "loss": 0.5404,
+      "step": 6540
+    },
+    {
+      "epoch": 4.754990925589837,
+      "grad_norm": 0.9776787161827087,
+      "learning_rate": 0.0002,
+      "loss": 0.6379,
+      "step": 6550
+    },
+    {
+      "epoch": 4.762250453720508,
+      "grad_norm": 1.1990505456924438,
+      "learning_rate": 0.0002,
+      "loss": 0.5288,
+      "step": 6560
+    },
+    {
+      "epoch": 4.769509981851179,
+      "grad_norm": 1.0582209825515747,
+      "learning_rate": 0.0002,
+      "loss": 0.5539,
+      "step": 6570
+    },
+    {
+      "epoch": 4.776769509981851,
+      "grad_norm": 0.9966367483139038,
+      "learning_rate": 0.0002,
+      "loss": 0.489,
+      "step": 6580
+    },
+    {
+      "epoch": 4.784029038112522,
+      "grad_norm": 0.9130612015724182,
+      "learning_rate": 0.0002,
+      "loss": 0.5514,
+      "step": 6590
+    },
+    {
+      "epoch": 4.791288566243194,
+      "grad_norm": 1.0950500965118408,
+      "learning_rate": 0.0002,
+      "loss": 0.5864,
+      "step": 6600
+    },
+    {
+      "epoch": 4.798548094373865,
+      "grad_norm": 1.108681321144104,
+      "learning_rate": 0.0002,
+      "loss": 0.5266,
+      "step": 6610
+    },
+    {
+      "epoch": 4.805807622504537,
+      "grad_norm": 1.1873763799667358,
+      "learning_rate": 0.0002,
+      "loss": 0.5875,
+      "step": 6620
+    },
+    {
+      "epoch": 4.8130671506352085,
+      "grad_norm": 1.305367112159729,
+      "learning_rate": 0.0002,
+      "loss": 0.5736,
+      "step": 6630
+    },
+    {
+      "epoch": 4.8203266787658805,
+      "grad_norm": 1.2801482677459717,
+      "learning_rate": 0.0002,
+      "loss": 0.5636,
+      "step": 6640
+    },
+    {
+      "epoch": 4.827586206896552,
+      "grad_norm": 1.26764976978302,
+      "learning_rate": 0.0002,
+      "loss": 0.582,
+      "step": 6650
+    },
+    {
+      "epoch": 4.834845735027224,
+      "grad_norm": 1.0018208026885986,
+      "learning_rate": 0.0002,
+      "loss": 0.5259,
+      "step": 6660
+    },
+    {
+      "epoch": 4.842105263157895,
+      "grad_norm": 1.2326326370239258,
+      "learning_rate": 0.0002,
+      "loss": 0.548,
+      "step": 6670
+    },
+    {
+      "epoch": 4.849364791288567,
+      "grad_norm": 0.9707282781600952,
+      "learning_rate": 0.0002,
+      "loss": 0.5933,
+      "step": 6680
+    },
+    {
+      "epoch": 4.856624319419238,
+      "grad_norm": 1.2772048711776733,
+      "learning_rate": 0.0002,
+      "loss": 0.5612,
+      "step": 6690
+    },
+    {
+      "epoch": 4.863883847549909,
+      "grad_norm": 2.6652262210845947,
+      "learning_rate": 0.0002,
+      "loss": 0.5346,
+      "step": 6700
+    },
+    {
+      "epoch": 4.871143375680581,
+      "grad_norm": 1.215828537940979,
+      "learning_rate": 0.0002,
+      "loss": 0.5428,
+      "step": 6710
+    },
+    {
+      "epoch": 4.878402903811252,
+      "grad_norm": 1.3704510927200317,
+      "learning_rate": 0.0002,
+      "loss": 0.6571,
+      "step": 6720
+    },
+    {
+      "epoch": 4.885662431941924,
+      "grad_norm": 0.7781757116317749,
+      "learning_rate": 0.0002,
+      "loss": 0.4963,
+      "step": 6730
+    },
+    {
+      "epoch": 4.892921960072595,
+      "grad_norm": 1.1883646249771118,
+      "learning_rate": 0.0002,
+      "loss": 0.5989,
+      "step": 6740
+    },
+    {
+      "epoch": 4.900181488203267,
+      "grad_norm": 0.9216066002845764,
+      "learning_rate": 0.0002,
+      "loss": 0.6067,
+      "step": 6750
+    },
+    {
+      "epoch": 4.907441016333938,
+      "grad_norm": 1.0558464527130127,
+      "learning_rate": 0.0002,
+      "loss": 0.5085,
+      "step": 6760
+    },
+    {
+      "epoch": 4.91470054446461,
+      "grad_norm": 1.032656192779541,
+      "learning_rate": 0.0002,
+      "loss": 0.5216,
+      "step": 6770
+    },
+    {
+      "epoch": 4.921960072595281,
+      "grad_norm": 1.1261441707611084,
+      "learning_rate": 0.0002,
+      "loss": 0.5426,
+      "step": 6780
+    },
+    {
+      "epoch": 4.929219600725952,
+      "grad_norm": 1.2178640365600586,
+      "learning_rate": 0.0002,
+      "loss": 0.5295,
+      "step": 6790
+    },
+    {
+      "epoch": 4.936479128856624,
+      "grad_norm": 1.5369361639022827,
+      "learning_rate": 0.0002,
+      "loss": 0.5476,
+      "step": 6800
+    },
+    {
+      "epoch": 4.943738656987296,
+      "grad_norm": 1.1188377141952515,
+      "learning_rate": 0.0002,
+      "loss": 0.5358,
+      "step": 6810
+    },
+    {
+      "epoch": 4.950998185117967,
+      "grad_norm": 1.2506113052368164,
+      "learning_rate": 0.0002,
+      "loss": 0.5483,
+      "step": 6820
+    },
+    {
+      "epoch": 4.9582577132486385,
+      "grad_norm": 0.8776047825813293,
+      "learning_rate": 0.0002,
+      "loss": 0.567,
+      "step": 6830
+    },
+    {
+      "epoch": 4.9655172413793105,
+      "grad_norm": 0.9700555205345154,
+      "learning_rate": 0.0002,
+      "loss": 0.5764,
+      "step": 6840
+    },
+    {
+      "epoch": 4.972776769509982,
+      "grad_norm": 1.2713534832000732,
+      "learning_rate": 0.0002,
+      "loss": 0.5396,
+      "step": 6850
+    },
+    {
+      "epoch": 4.980036297640654,
+      "grad_norm": 0.9855955243110657,
+      "learning_rate": 0.0002,
+      "loss": 0.5451,
+      "step": 6860
+    },
+    {
+      "epoch": 4.987295825771325,
+      "grad_norm": 0.8734853863716125,
+      "learning_rate": 0.0002,
+      "loss": 0.5884,
+      "step": 6870
+    },
+    {
+      "epoch": 4.994555353901997,
+      "grad_norm": 0.8065403699874878,
+      "learning_rate": 0.0002,
+      "loss": 0.5189,
+      "step": 6880
+    },
+    {
+      "epoch": 4.999637023593467,
+      "eval_loss": 1.3302682638168335,
+      "eval_runtime": 46.2496,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 6887
+    },
+    {
+      "epoch": 5.001814882032668,
+      "grad_norm": 0.5163813829421997,
+      "learning_rate": 0.0002,
+      "loss": 0.4889,
+      "step": 6890
+    },
+    {
+      "epoch": 5.00907441016334,
+      "grad_norm": 1.1496137380599976,
+      "learning_rate": 0.0002,
+      "loss": 0.3545,
+      "step": 6900
+    },
+    {
+      "epoch": 5.016333938294011,
+      "grad_norm": 1.0133885145187378,
+      "learning_rate": 0.0002,
+      "loss": 0.39,
+      "step": 6910
+    },
+    {
+      "epoch": 5.023593466424682,
+      "grad_norm": 0.9479621052742004,
+      "learning_rate": 0.0002,
+      "loss": 0.3693,
+      "step": 6920
+    },
+    {
+      "epoch": 5.030852994555354,
+      "grad_norm": 0.8587583303451538,
+      "learning_rate": 0.0002,
+      "loss": 0.4012,
+      "step": 6930
+    },
+    {
+      "epoch": 5.038112522686025,
+      "grad_norm": 1.3314697742462158,
+      "learning_rate": 0.0002,
+      "loss": 0.3428,
+      "step": 6940
+    },
+    {
+      "epoch": 5.045372050816697,
+      "grad_norm": 1.195448875427246,
+      "learning_rate": 0.0002,
+      "loss": 0.3909,
+      "step": 6950
+    },
+    {
+      "epoch": 5.052631578947368,
+      "grad_norm": 1.2482256889343262,
+      "learning_rate": 0.0002,
+      "loss": 0.3322,
+      "step": 6960
+    },
+    {
+      "epoch": 5.05989110707804,
+      "grad_norm": 1.2011528015136719,
+      "learning_rate": 0.0002,
+      "loss": 0.3893,
+      "step": 6970
+    },
+    {
+      "epoch": 5.067150635208711,
+      "grad_norm": 1.3997188806533813,
+      "learning_rate": 0.0002,
+      "loss": 0.3265,
+      "step": 6980
+    },
+    {
+      "epoch": 5.074410163339383,
+      "grad_norm": 1.2147513628005981,
+      "learning_rate": 0.0002,
+      "loss": 0.3716,
+      "step": 6990
+    },
+    {
+      "epoch": 5.081669691470054,
+      "grad_norm": 1.6030137538909912,
+      "learning_rate": 0.0002,
+      "loss": 0.4053,
+      "step": 7000
+    },
+    {
+      "epoch": 5.088929219600726,
+      "grad_norm": 0.9466970562934875,
+      "learning_rate": 0.0002,
+      "loss": 0.3665,
+      "step": 7010
+    },
+    {
+      "epoch": 5.096188747731397,
+      "grad_norm": 1.4593411684036255,
+      "learning_rate": 0.0002,
+      "loss": 0.3451,
+      "step": 7020
+    },
+    {
+      "epoch": 5.103448275862069,
+      "grad_norm": 1.2196033000946045,
+      "learning_rate": 0.0002,
+      "loss": 0.3843,
+      "step": 7030
+    },
+    {
+      "epoch": 5.1107078039927405,
+      "grad_norm": 1.1341328620910645,
+      "learning_rate": 0.0002,
+      "loss": 0.3896,
+      "step": 7040
+    },
+    {
+      "epoch": 5.117967332123412,
+      "grad_norm": 1.2248976230621338,
+      "learning_rate": 0.0002,
+      "loss": 0.3627,
+      "step": 7050
+    },
+    {
+      "epoch": 5.125226860254084,
+      "grad_norm": 1.1620593070983887,
+      "learning_rate": 0.0002,
+      "loss": 0.3784,
+      "step": 7060
+    },
+    {
+      "epoch": 5.132486388384755,
+      "grad_norm": 0.9300723671913147,
+      "learning_rate": 0.0002,
+      "loss": 0.3678,
+      "step": 7070
+    },
+    {
+      "epoch": 5.139745916515427,
+      "grad_norm": 1.2265169620513916,
+      "learning_rate": 0.0002,
+      "loss": 0.3756,
+      "step": 7080
+    },
+    {
+      "epoch": 5.147005444646098,
+      "grad_norm": 1.4430373907089233,
+      "learning_rate": 0.0002,
+      "loss": 0.3595,
+      "step": 7090
+    },
+    {
+      "epoch": 5.15426497277677,
+      "grad_norm": 1.0821576118469238,
+      "learning_rate": 0.0002,
+      "loss": 0.3788,
+      "step": 7100
+    },
+    {
+      "epoch": 5.161524500907441,
+      "grad_norm": 1.2574739456176758,
+      "learning_rate": 0.0002,
+      "loss": 0.383,
+      "step": 7110
+    },
+    {
+      "epoch": 5.168784029038113,
+      "grad_norm": 1.1806069612503052,
+      "learning_rate": 0.0002,
+      "loss": 0.3692,
+      "step": 7120
+    },
+    {
+      "epoch": 5.176043557168784,
+      "grad_norm": 0.9900956153869629,
+      "learning_rate": 0.0002,
+      "loss": 0.3978,
+      "step": 7130
+    },
+    {
+      "epoch": 5.183303085299456,
+      "grad_norm": 1.2414425611495972,
+      "learning_rate": 0.0002,
+      "loss": 0.4358,
+      "step": 7140
+    },
+    {
+      "epoch": 5.190562613430127,
+      "grad_norm": 0.8220699429512024,
+      "learning_rate": 0.0002,
+      "loss": 0.3485,
+      "step": 7150
+    },
+    {
+      "epoch": 5.197822141560798,
+      "grad_norm": 1.29408860206604,
+      "learning_rate": 0.0002,
+      "loss": 0.3517,
+      "step": 7160
+    },
+    {
+      "epoch": 5.20508166969147,
+      "grad_norm": 0.8510639071464539,
+      "learning_rate": 0.0002,
+      "loss": 0.3405,
+      "step": 7170
+    },
+    {
+      "epoch": 5.212341197822141,
+      "grad_norm": 1.3448902368545532,
+      "learning_rate": 0.0002,
+      "loss": 0.4233,
+      "step": 7180
+    },
+    {
+      "epoch": 5.219600725952813,
+      "grad_norm": 1.054451584815979,
+      "learning_rate": 0.0002,
+      "loss": 0.3808,
+      "step": 7190
+    },
+    {
+      "epoch": 5.226860254083484,
+      "grad_norm": 1.3752713203430176,
+      "learning_rate": 0.0002,
+      "loss": 0.368,
+      "step": 7200
+    },
+    {
+      "epoch": 5.234119782214156,
+      "grad_norm": 1.4848095178604126,
+      "learning_rate": 0.0002,
+      "loss": 0.3844,
+      "step": 7210
+    },
+    {
+      "epoch": 5.241379310344827,
+      "grad_norm": 1.428842544555664,
+      "learning_rate": 0.0002,
+      "loss": 0.4187,
+      "step": 7220
+    },
+    {
+      "epoch": 5.248638838475499,
+      "grad_norm": 1.1703591346740723,
+      "learning_rate": 0.0002,
+      "loss": 0.3778,
+      "step": 7230
+    },
+    {
+      "epoch": 5.2558983666061705,
+      "grad_norm": 1.2180451154708862,
+      "learning_rate": 0.0002,
+      "loss": 0.417,
+      "step": 7240
+    },
+    {
+      "epoch": 5.2631578947368425,
+      "grad_norm": 1.094045877456665,
+      "learning_rate": 0.0002,
+      "loss": 0.3656,
+      "step": 7250
+    },
+    {
+      "epoch": 5.270417422867514,
+      "grad_norm": 0.9545766115188599,
+      "learning_rate": 0.0002,
+      "loss": 0.4331,
+      "step": 7260
+    },
+    {
+      "epoch": 5.277676950998185,
+      "grad_norm": 0.8356652855873108,
+      "learning_rate": 0.0002,
+      "loss": 0.3642,
+      "step": 7270
+    },
+    {
+      "epoch": 5.284936479128857,
+      "grad_norm": 1.148160457611084,
+      "learning_rate": 0.0002,
+      "loss": 0.3576,
+      "step": 7280
+    },
+    {
+      "epoch": 5.292196007259528,
+      "grad_norm": 1.2009977102279663,
+      "learning_rate": 0.0002,
+      "loss": 0.4178,
+      "step": 7290
+    },
+    {
+      "epoch": 5.2994555353902,
+      "grad_norm": 1.3283873796463013,
+      "learning_rate": 0.0002,
+      "loss": 0.3977,
+      "step": 7300
+    },
+    {
+      "epoch": 5.306715063520871,
+      "grad_norm": 0.9850481748580933,
+      "learning_rate": 0.0002,
+      "loss": 0.3853,
+      "step": 7310
+    },
+    {
+      "epoch": 5.313974591651543,
+      "grad_norm": 1.367550015449524,
+      "learning_rate": 0.0002,
+      "loss": 0.3645,
+      "step": 7320
+    },
+    {
+      "epoch": 5.321234119782214,
+      "grad_norm": 0.8602936863899231,
+      "learning_rate": 0.0002,
+      "loss": 0.3898,
+      "step": 7330
+    },
+    {
+      "epoch": 5.328493647912886,
+      "grad_norm": 1.1130679845809937,
+      "learning_rate": 0.0002,
+      "loss": 0.4173,
+      "step": 7340
+    },
+    {
+      "epoch": 5.335753176043557,
+      "grad_norm": 1.3002253770828247,
+      "learning_rate": 0.0002,
+      "loss": 0.3642,
+      "step": 7350
+    },
+    {
+      "epoch": 5.343012704174229,
+      "grad_norm": 1.6235289573669434,
+      "learning_rate": 0.0002,
+      "loss": 0.4138,
+      "step": 7360
+    },
+    {
+      "epoch": 5.3502722323049,
+      "grad_norm": 1.156379222869873,
+      "learning_rate": 0.0002,
+      "loss": 0.4779,
+      "step": 7370
+    },
+    {
+      "epoch": 5.357531760435572,
+      "grad_norm": 1.0569308996200562,
+      "learning_rate": 0.0002,
+      "loss": 0.3222,
+      "step": 7380
+    },
+    {
+      "epoch": 5.364791288566243,
+      "grad_norm": 1.6674021482467651,
+      "learning_rate": 0.0002,
+      "loss": 0.3573,
+      "step": 7390
+    },
+    {
+      "epoch": 5.372050816696914,
+      "grad_norm": 1.2962018251419067,
+      "learning_rate": 0.0002,
+      "loss": 0.4325,
+      "step": 7400
+    },
+    {
+      "epoch": 5.379310344827586,
+      "grad_norm": 1.1904195547103882,
+      "learning_rate": 0.0002,
+      "loss": 0.3809,
+      "step": 7410
+    },
+    {
+      "epoch": 5.386569872958257,
+      "grad_norm": 1.316245675086975,
+      "learning_rate": 0.0002,
+      "loss": 0.3728,
+      "step": 7420
+    },
+    {
+      "epoch": 5.393829401088929,
+      "grad_norm": 1.127570390701294,
+      "learning_rate": 0.0002,
+      "loss": 0.4096,
+      "step": 7430
+    },
+    {
+      "epoch": 5.4010889292196005,
+      "grad_norm": 1.3895777463912964,
+      "learning_rate": 0.0002,
+      "loss": 0.3933,
+      "step": 7440
+    },
+    {
+      "epoch": 5.4083484573502725,
+      "grad_norm": 1.626830816268921,
+      "learning_rate": 0.0002,
+      "loss": 0.4085,
+      "step": 7450
+    },
+    {
+      "epoch": 5.415607985480944,
+      "grad_norm": 1.3703926801681519,
+      "learning_rate": 0.0002,
+      "loss": 0.4186,
+      "step": 7460
+    },
+    {
+      "epoch": 5.422867513611616,
+      "grad_norm": 1.3854840993881226,
+      "learning_rate": 0.0002,
+      "loss": 0.3517,
+      "step": 7470
+    },
+    {
+      "epoch": 5.430127041742287,
+      "grad_norm": 1.107065200805664,
+      "learning_rate": 0.0002,
+      "loss": 0.3714,
+      "step": 7480
+    },
+    {
+      "epoch": 5.437386569872959,
+      "grad_norm": 0.7843456268310547,
+      "learning_rate": 0.0002,
+      "loss": 0.3855,
+      "step": 7490
+    },
+    {
+      "epoch": 5.44464609800363,
+      "grad_norm": 1.6692372560501099,
+      "learning_rate": 0.0002,
+      "loss": 0.4159,
+      "step": 7500
+    },
+    {
+      "epoch": 5.451905626134302,
+      "grad_norm": 1.2583858966827393,
+      "learning_rate": 0.0002,
+      "loss": 0.4185,
+      "step": 7510
+    },
+    {
+      "epoch": 5.459165154264973,
+      "grad_norm": 1.6827000379562378,
+      "learning_rate": 0.0002,
+      "loss": 0.4401,
+      "step": 7520
+    },
+    {
+      "epoch": 5.466424682395644,
+      "grad_norm": 1.6680560111999512,
+      "learning_rate": 0.0002,
+      "loss": 0.397,
+      "step": 7530
+    },
+    {
+      "epoch": 5.473684210526316,
+      "grad_norm": 1.3696072101593018,
+      "learning_rate": 0.0002,
+      "loss": 0.4193,
+      "step": 7540
+    },
+    {
+      "epoch": 5.480943738656987,
+      "grad_norm": 1.4523496627807617,
+      "learning_rate": 0.0002,
+      "loss": 0.4244,
+      "step": 7550
+    },
+    {
+      "epoch": 5.488203266787659,
+      "grad_norm": 1.3432692289352417,
+      "learning_rate": 0.0002,
+      "loss": 0.3609,
+      "step": 7560
+    },
+    {
+      "epoch": 5.49546279491833,
+      "grad_norm": 1.363818645477295,
+      "learning_rate": 0.0002,
+      "loss": 0.3675,
+      "step": 7570
+    },
+    {
+      "epoch": 5.502722323049002,
+      "grad_norm": 1.0176721811294556,
+      "learning_rate": 0.0002,
+      "loss": 0.3726,
+      "step": 7580
+    },
+    {
+      "epoch": 5.509981851179673,
+      "grad_norm": 1.1625547409057617,
+      "learning_rate": 0.0002,
+      "loss": 0.3751,
+      "step": 7590
+    },
+    {
+      "epoch": 5.517241379310345,
+      "grad_norm": 1.2480388879776,
+      "learning_rate": 0.0002,
+      "loss": 0.433,
+      "step": 7600
+    },
+    {
+      "epoch": 5.524500907441016,
+      "grad_norm": 1.341509222984314,
+      "learning_rate": 0.0002,
+      "loss": 0.4511,
+      "step": 7610
+    },
+    {
+      "epoch": 5.531760435571687,
+      "grad_norm": 1.7048436403274536,
+      "learning_rate": 0.0002,
+      "loss": 0.4642,
+      "step": 7620
+    },
+    {
+      "epoch": 5.539019963702359,
+      "grad_norm": 1.1435480117797852,
+      "learning_rate": 0.0002,
+      "loss": 0.4509,
+      "step": 7630
+    },
+    {
+      "epoch": 5.5462794918330305,
+      "grad_norm": 1.2381842136383057,
+      "learning_rate": 0.0002,
+      "loss": 0.4528,
+      "step": 7640
+    },
+    {
+      "epoch": 5.5535390199637025,
+      "grad_norm": 1.50786292552948,
+      "learning_rate": 0.0002,
+      "loss": 0.4496,
+      "step": 7650
+    },
+    {
+      "epoch": 5.560798548094374,
+      "grad_norm": 1.2263519763946533,
+      "learning_rate": 0.0002,
+      "loss": 0.4242,
+      "step": 7660
+    },
+    {
+      "epoch": 5.568058076225046,
+      "grad_norm": 1.2864696979522705,
+      "learning_rate": 0.0002,
+      "loss": 0.418,
+      "step": 7670
+    },
+    {
+      "epoch": 5.575317604355717,
+      "grad_norm": 1.4443191289901733,
+      "learning_rate": 0.0002,
+      "loss": 0.3832,
+      "step": 7680
+    },
+    {
+      "epoch": 5.582577132486389,
+      "grad_norm": 1.3360971212387085,
+      "learning_rate": 0.0002,
+      "loss": 0.3964,
+      "step": 7690
+    },
+    {
+      "epoch": 5.58983666061706,
+      "grad_norm": 1.391828179359436,
+      "learning_rate": 0.0002,
+      "loss": 0.4639,
+      "step": 7700
+    },
+    {
+      "epoch": 5.597096188747732,
+      "grad_norm": 1.3699384927749634,
+      "learning_rate": 0.0002,
+      "loss": 0.4722,
+      "step": 7710
+    },
+    {
+      "epoch": 5.604355716878403,
+      "grad_norm": 1.3778468370437622,
+      "learning_rate": 0.0002,
+      "loss": 0.4302,
+      "step": 7720
+    },
+    {
+      "epoch": 5.611615245009075,
+      "grad_norm": 1.1009501218795776,
+      "learning_rate": 0.0002,
+      "loss": 0.4179,
+      "step": 7730
+    },
+    {
+      "epoch": 5.618874773139746,
+      "grad_norm": 1.0410021543502808,
+      "learning_rate": 0.0002,
+      "loss": 0.4104,
+      "step": 7740
+    },
+    {
+      "epoch": 5.626134301270417,
+      "grad_norm": 1.1012226343154907,
+      "learning_rate": 0.0002,
+      "loss": 0.4489,
+      "step": 7750
+    },
+    {
+      "epoch": 5.633393829401089,
+      "grad_norm": 1.3246384859085083,
+      "learning_rate": 0.0002,
+      "loss": 0.4544,
+      "step": 7760
+    },
+    {
+      "epoch": 5.64065335753176,
+      "grad_norm": 1.4301716089248657,
+      "learning_rate": 0.0002,
+      "loss": 0.4381,
+      "step": 7770
+    },
+    {
+      "epoch": 5.647912885662432,
+      "grad_norm": 1.1368978023529053,
+      "learning_rate": 0.0002,
+      "loss": 0.4297,
+      "step": 7780
+    },
+    {
+      "epoch": 5.655172413793103,
+      "grad_norm": 1.3493064641952515,
+      "learning_rate": 0.0002,
+      "loss": 0.4063,
+      "step": 7790
+    },
+    {
+      "epoch": 5.662431941923775,
+      "grad_norm": 1.3328721523284912,
+      "learning_rate": 0.0002,
+      "loss": 0.4562,
+      "step": 7800
+    },
+    {
+      "epoch": 5.669691470054446,
+      "grad_norm": 1.3235671520233154,
+      "learning_rate": 0.0002,
+      "loss": 0.4075,
+      "step": 7810
+    },
+    {
+      "epoch": 5.676950998185118,
+      "grad_norm": 1.1961841583251953,
+      "learning_rate": 0.0002,
+      "loss": 0.4589,
+      "step": 7820
+    },
+    {
+      "epoch": 5.684210526315789,
+      "grad_norm": 1.4189636707305908,
+      "learning_rate": 0.0002,
+      "loss": 0.4503,
+      "step": 7830
+    },
+    {
+      "epoch": 5.691470054446461,
+      "grad_norm": 1.3551312685012817,
+      "learning_rate": 0.0002,
+      "loss": 0.4452,
+      "step": 7840
+    },
+    {
+      "epoch": 5.6987295825771325,
+      "grad_norm": 1.449987769126892,
+      "learning_rate": 0.0002,
+      "loss": 0.4268,
+      "step": 7850
+    },
+    {
+      "epoch": 5.7059891107078045,
+      "grad_norm": 1.1225156784057617,
+      "learning_rate": 0.0002,
+      "loss": 0.4141,
+      "step": 7860
+    },
+    {
+      "epoch": 5.713248638838476,
+      "grad_norm": 1.4734594821929932,
+      "learning_rate": 0.0002,
+      "loss": 0.41,
+      "step": 7870
+    },
+    {
+      "epoch": 5.720508166969147,
+      "grad_norm": 1.3793359994888306,
+      "learning_rate": 0.0002,
+      "loss": 0.4013,
+      "step": 7880
+    },
+    {
+      "epoch": 5.727767695099819,
+      "grad_norm": 1.2431834936141968,
+      "learning_rate": 0.0002,
+      "loss": 0.4065,
+      "step": 7890
+    },
+    {
+      "epoch": 5.73502722323049,
+      "grad_norm": 1.1158313751220703,
+      "learning_rate": 0.0002,
+      "loss": 0.4595,
+      "step": 7900
+    },
+    {
+      "epoch": 5.742286751361162,
+      "grad_norm": 1.212248682975769,
+      "learning_rate": 0.0002,
+      "loss": 0.4342,
+      "step": 7910
+    },
+    {
+      "epoch": 5.749546279491833,
+      "grad_norm": 1.5259995460510254,
+      "learning_rate": 0.0002,
+      "loss": 0.4611,
+      "step": 7920
+    },
+    {
+      "epoch": 5.756805807622505,
+      "grad_norm": 1.3909121751785278,
+      "learning_rate": 0.0002,
+      "loss": 0.4483,
+      "step": 7930
+    },
+    {
+      "epoch": 5.764065335753176,
+      "grad_norm": 1.2511249780654907,
+      "learning_rate": 0.0002,
+      "loss": 0.4325,
+      "step": 7940
+    },
+    {
+      "epoch": 5.771324863883848,
+      "grad_norm": 1.2511906623840332,
+      "learning_rate": 0.0002,
+      "loss": 0.4048,
+      "step": 7950
+    },
+    {
+      "epoch": 5.778584392014519,
+      "grad_norm": 1.1489921808242798,
+      "learning_rate": 0.0002,
+      "loss": 0.3715,
+      "step": 7960
+    },
+    {
+      "epoch": 5.78584392014519,
+      "grad_norm": 1.028943419456482,
+      "learning_rate": 0.0002,
+      "loss": 0.4196,
+      "step": 7970
+    },
+    {
+      "epoch": 5.793103448275862,
+      "grad_norm": 1.0820423364639282,
+      "learning_rate": 0.0002,
+      "loss": 0.4334,
+      "step": 7980
+    },
+    {
+      "epoch": 5.800362976406533,
+      "grad_norm": 1.296520471572876,
+      "learning_rate": 0.0002,
+      "loss": 0.3917,
+      "step": 7990
+    },
+    {
+      "epoch": 5.807622504537205,
+      "grad_norm": 1.3597749471664429,
+      "learning_rate": 0.0002,
+      "loss": 0.4509,
+      "step": 8000
+    },
+    {
+      "epoch": 5.814882032667876,
+      "grad_norm": 0.8741790652275085,
+      "learning_rate": 0.0002,
+      "loss": 0.4535,
+      "step": 8010
+    },
+    {
+      "epoch": 5.822141560798548,
+      "grad_norm": 1.1471822261810303,
+      "learning_rate": 0.0002,
+      "loss": 0.4239,
+      "step": 8020
+    },
+    {
+      "epoch": 5.829401088929219,
+      "grad_norm": 1.2997334003448486,
+      "learning_rate": 0.0002,
+      "loss": 0.5042,
+      "step": 8030
+    },
+    {
+      "epoch": 5.836660617059891,
+      "grad_norm": 1.1027175188064575,
+      "learning_rate": 0.0002,
+      "loss": 0.4758,
+      "step": 8040
+    },
+    {
+      "epoch": 5.8439201451905625,
+      "grad_norm": 1.2695307731628418,
+      "learning_rate": 0.0002,
+      "loss": 0.4192,
+      "step": 8050
+    },
+    {
+      "epoch": 5.8511796733212345,
+      "grad_norm": 1.5275461673736572,
+      "learning_rate": 0.0002,
+      "loss": 0.5173,
+      "step": 8060
+    },
+    {
+      "epoch": 5.8584392014519056,
+      "grad_norm": 1.3059501647949219,
+      "learning_rate": 0.0002,
+      "loss": 0.5012,
+      "step": 8070
+    },
+    {
+      "epoch": 5.8656987295825775,
+      "grad_norm": 1.57442045211792,
+      "learning_rate": 0.0002,
+      "loss": 0.4425,
+      "step": 8080
+    },
+    {
+      "epoch": 5.872958257713249,
+      "grad_norm": 1.119564414024353,
+      "learning_rate": 0.0002,
+      "loss": 0.4261,
+      "step": 8090
+    },
+    {
+      "epoch": 5.88021778584392,
+      "grad_norm": 1.6517373323440552,
+      "learning_rate": 0.0002,
+      "loss": 0.465,
+      "step": 8100
+    },
+    {
+      "epoch": 5.887477313974592,
+      "grad_norm": 1.4093554019927979,
+      "learning_rate": 0.0002,
+      "loss": 0.4406,
+      "step": 8110
+    },
+    {
+      "epoch": 5.894736842105263,
+      "grad_norm": 1.278843641281128,
+      "learning_rate": 0.0002,
+      "loss": 0.4433,
+      "step": 8120
+    },
+    {
+      "epoch": 5.901996370235935,
+      "grad_norm": 1.2042944431304932,
+      "learning_rate": 0.0002,
+      "loss": 0.4007,
+      "step": 8130
+    },
+    {
+      "epoch": 5.909255898366606,
+      "grad_norm": 1.1788326501846313,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 8140
+    },
+    {
+      "epoch": 5.916515426497278,
+      "grad_norm": 1.4364569187164307,
+      "learning_rate": 0.0002,
+      "loss": 0.4506,
+      "step": 8150
+    },
+    {
+      "epoch": 5.923774954627949,
+      "grad_norm": 1.1704283952713013,
+      "learning_rate": 0.0002,
+      "loss": 0.4651,
+      "step": 8160
+    },
+    {
+      "epoch": 5.931034482758621,
+      "grad_norm": 1.040814995765686,
+      "learning_rate": 0.0002,
+      "loss": 0.3972,
+      "step": 8170
+    },
+    {
+      "epoch": 5.938294010889292,
+      "grad_norm": 1.1367416381835938,
+      "learning_rate": 0.0002,
+      "loss": 0.4038,
+      "step": 8180
+    },
+    {
+      "epoch": 5.945553539019964,
+      "grad_norm": 1.3401511907577515,
+      "learning_rate": 0.0002,
+      "loss": 0.4387,
+      "step": 8190
+    },
+    {
+      "epoch": 5.952813067150635,
+      "grad_norm": 1.1154041290283203,
+      "learning_rate": 0.0002,
+      "loss": 0.4396,
+      "step": 8200
+    },
+    {
+      "epoch": 5.960072595281307,
+      "grad_norm": 1.426089882850647,
+      "learning_rate": 0.0002,
+      "loss": 0.4744,
+      "step": 8210
+    },
+    {
+      "epoch": 5.967332123411978,
+      "grad_norm": 1.3170222043991089,
+      "learning_rate": 0.0002,
+      "loss": 0.4105,
+      "step": 8220
+    },
+    {
+      "epoch": 5.974591651542649,
+      "grad_norm": 1.1960029602050781,
+      "learning_rate": 0.0002,
+      "loss": 0.4137,
+      "step": 8230
+    },
+    {
+      "epoch": 5.981851179673321,
+      "grad_norm": 1.0843931436538696,
+      "learning_rate": 0.0002,
+      "loss": 0.423,
+      "step": 8240
+    },
+    {
+      "epoch": 5.9891107078039925,
+      "grad_norm": 1.050421118736267,
+      "learning_rate": 0.0002,
+      "loss": 0.459,
+      "step": 8250
+    },
+    {
+      "epoch": 5.9963702359346644,
+      "grad_norm": 1.0183138847351074,
+      "learning_rate": 0.0002,
+      "loss": 0.3993,
+      "step": 8260
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.4677470922470093,
+      "eval_runtime": 46.2504,
+      "eval_samples_per_second": 9.427,
+      "eval_steps_per_second": 1.189,
+      "step": 8265
+    },
+    {
+      "epoch": 6.0036297640653356,
+      "grad_norm": 1.1702998876571655,
+      "learning_rate": 0.0002,
+      "loss": 0.3947,
+      "step": 8270
+    },
+    {
+      "epoch": 6.0108892921960075,
+      "grad_norm": 1.5389727354049683,
+      "learning_rate": 0.0002,
+      "loss": 0.2854,
+      "step": 8280
+    },
+    {
+      "epoch": 6.018148820326679,
+      "grad_norm": 1.502568244934082,
+      "learning_rate": 0.0002,
+      "loss": 0.2603,
+      "step": 8290
+    },
+    {
+      "epoch": 6.025408348457351,
+      "grad_norm": 1.3846043348312378,
+      "learning_rate": 0.0002,
+      "loss": 0.3329,
+      "step": 8300
+    },
+    {
+      "epoch": 6.032667876588022,
+      "grad_norm": 1.173553228378296,
+      "learning_rate": 0.0002,
+      "loss": 0.2651,
+      "step": 8310
+    },
+    {
+      "epoch": 6.039927404718694,
+      "grad_norm": 1.5325932502746582,
+      "learning_rate": 0.0002,
+      "loss": 0.3142,
+      "step": 8320
+    },
+    {
+      "epoch": 6.047186932849365,
+      "grad_norm": 1.303783655166626,
+      "learning_rate": 0.0002,
+      "loss": 0.2511,
+      "step": 8330
+    },
+    {
+      "epoch": 6.054446460980036,
+      "grad_norm": 0.9408994913101196,
+      "learning_rate": 0.0002,
+      "loss": 0.2352,
+      "step": 8340
+    },
+    {
+      "epoch": 6.061705989110708,
+      "grad_norm": 1.5430388450622559,
+      "learning_rate": 0.0002,
+      "loss": 0.2548,
+      "step": 8350
+    },
+    {
+      "epoch": 6.068965517241379,
+      "grad_norm": 0.8765342235565186,
+      "learning_rate": 0.0002,
+      "loss": 0.2682,
+      "step": 8360
+    },
+    {
+      "epoch": 6.076225045372051,
+      "grad_norm": 1.2363157272338867,
+      "learning_rate": 0.0002,
+      "loss": 0.2614,
+      "step": 8370
+    },
+    {
+      "epoch": 6.083484573502722,
+      "grad_norm": 1.21284818649292,
+      "learning_rate": 0.0002,
+      "loss": 0.294,
+      "step": 8380
+    },
+    {
+      "epoch": 6.090744101633394,
+      "grad_norm": 1.3261712789535522,
+      "learning_rate": 0.0002,
+      "loss": 0.2498,
+      "step": 8390
+    },
+    {
+      "epoch": 6.098003629764065,
+      "grad_norm": 1.077317714691162,
+      "learning_rate": 0.0002,
+      "loss": 0.2649,
+      "step": 8400
+    },
+    {
+      "epoch": 6.105263157894737,
+      "grad_norm": 0.9873808026313782,
+      "learning_rate": 0.0002,
+      "loss": 0.269,
+      "step": 8410
+    },
+    {
+      "epoch": 6.112522686025408,
+      "grad_norm": 1.032258152961731,
+      "learning_rate": 0.0002,
+      "loss": 0.2736,
+      "step": 8420
+    },
+    {
+      "epoch": 6.11978221415608,
+      "grad_norm": 1.1014811992645264,
+      "learning_rate": 0.0002,
+      "loss": 0.2854,
+      "step": 8430
+    },
+    {
+      "epoch": 6.127041742286751,
+      "grad_norm": 1.4264203310012817,
+      "learning_rate": 0.0002,
+      "loss": 0.2924,
+      "step": 8440
+    },
+    {
+      "epoch": 6.1343012704174225,
+      "grad_norm": 1.4086531400680542,
+      "learning_rate": 0.0002,
+      "loss": 0.3388,
+      "step": 8450
+    },
+    {
+      "epoch": 6.1415607985480944,
+      "grad_norm": 1.3842453956604004,
+      "learning_rate": 0.0002,
+      "loss": 0.2786,
+      "step": 8460
+    },
+    {
+      "epoch": 6.1488203266787655,
+      "grad_norm": 1.4356757402420044,
+      "learning_rate": 0.0002,
+      "loss": 0.3201,
+      "step": 8470
+    },
+    {
+      "epoch": 6.1560798548094375,
+      "grad_norm": 1.193315029144287,
+      "learning_rate": 0.0002,
+      "loss": 0.2908,
+      "step": 8480
+    },
+    {
+      "epoch": 6.163339382940109,
+      "grad_norm": 1.0623924732208252,
+      "learning_rate": 0.0002,
+      "loss": 0.342,
+      "step": 8490
+    },
+    {
+      "epoch": 6.170598911070781,
+      "grad_norm": 1.5484434366226196,
+      "learning_rate": 0.0002,
+      "loss": 0.3257,
+      "step": 8500
+    },
+    {
+      "epoch": 6.177858439201452,
+      "grad_norm": 1.3520029783248901,
+      "learning_rate": 0.0002,
+      "loss": 0.2861,
+      "step": 8510
+    },
+    {
+      "epoch": 6.185117967332124,
+      "grad_norm": 1.2773103713989258,
+      "learning_rate": 0.0002,
+      "loss": 0.3242,
+      "step": 8520
+    },
+    {
+      "epoch": 6.192377495462795,
+      "grad_norm": 1.4675105810165405,
+      "learning_rate": 0.0002,
+      "loss": 0.3108,
+      "step": 8530
+    },
+    {
+      "epoch": 6.199637023593467,
+      "grad_norm": 1.2118732929229736,
+      "learning_rate": 0.0002,
+      "loss": 0.3044,
+      "step": 8540
+    },
+    {
+      "epoch": 6.206896551724138,
+      "grad_norm": 1.264024257659912,
+      "learning_rate": 0.0002,
+      "loss": 0.2726,
+      "step": 8550
+    },
+    {
+      "epoch": 6.21415607985481,
+      "grad_norm": 1.406931757926941,
+      "learning_rate": 0.0002,
+      "loss": 0.306,
+      "step": 8560
+    },
+    {
+      "epoch": 6.221415607985481,
+      "grad_norm": 1.385459542274475,
+      "learning_rate": 0.0002,
+      "loss": 0.2904,
+      "step": 8570
+    },
+    {
+      "epoch": 6.228675136116152,
+      "grad_norm": 1.9336168766021729,
+      "learning_rate": 0.0002,
+      "loss": 0.3413,
+      "step": 8580
+    },
+    {
+      "epoch": 6.235934664246824,
+      "grad_norm": 0.9880136847496033,
+      "learning_rate": 0.0002,
+      "loss": 0.2769,
+      "step": 8590
+    },
+    {
+      "epoch": 6.243194192377495,
+      "grad_norm": 1.3870339393615723,
+      "learning_rate": 0.0002,
+      "loss": 0.3035,
+      "step": 8600
+    },
+    {
+      "epoch": 6.250453720508167,
+      "grad_norm": 1.2303647994995117,
+      "learning_rate": 0.0002,
+      "loss": 0.286,
+      "step": 8610
+    },
+    {
+      "epoch": 6.257713248638838,
+      "grad_norm": 1.5406211614608765,
+      "learning_rate": 0.0002,
+      "loss": 0.3113,
+      "step": 8620
+    },
+    {
+      "epoch": 6.26497277676951,
+      "grad_norm": 1.2436790466308594,
+      "learning_rate": 0.0002,
+      "loss": 0.292,
+      "step": 8630
+    },
+    {
+      "epoch": 6.272232304900181,
+      "grad_norm": 0.8844212293624878,
+      "learning_rate": 0.0002,
+      "loss": 0.3102,
+      "step": 8640
+    },
+    {
+      "epoch": 6.279491833030853,
+      "grad_norm": 1.2846336364746094,
+      "learning_rate": 0.0002,
+      "loss": 0.3373,
+      "step": 8650
+    },
+    {
+      "epoch": 6.286751361161524,
+      "grad_norm": 1.593814730644226,
+      "learning_rate": 0.0002,
+      "loss": 0.3535,
+      "step": 8660
+    },
+    {
+      "epoch": 6.2940108892921955,
+      "grad_norm": 1.2277469635009766,
+      "learning_rate": 0.0002,
+      "loss": 0.3413,
+      "step": 8670
+    },
+    {
+      "epoch": 6.3012704174228675,
+      "grad_norm": 1.2574384212493896,
+      "learning_rate": 0.0002,
+      "loss": 0.2958,
+      "step": 8680
+    },
+    {
+      "epoch": 6.308529945553539,
+      "grad_norm": 1.335150957107544,
+      "learning_rate": 0.0002,
+      "loss": 0.3251,
+      "step": 8690
+    },
+    {
+      "epoch": 6.315789473684211,
+      "grad_norm": 1.3140437602996826,
+      "learning_rate": 0.0002,
+      "loss": 0.3009,
+      "step": 8700
+    },
+    {
+      "epoch": 6.323049001814882,
+      "grad_norm": 1.1689209938049316,
+      "learning_rate": 0.0002,
+      "loss": 0.2783,
+      "step": 8710
+    },
+    {
+      "epoch": 6.330308529945554,
+      "grad_norm": 1.6448503732681274,
+      "learning_rate": 0.0002,
+      "loss": 0.3476,
+      "step": 8720
+    },
+    {
+      "epoch": 6.337568058076225,
+      "grad_norm": 0.9944145679473877,
+      "learning_rate": 0.0002,
+      "loss": 0.2934,
+      "step": 8730
+    },
+    {
+      "epoch": 6.344827586206897,
+      "grad_norm": 1.1775634288787842,
+      "learning_rate": 0.0002,
+      "loss": 0.3315,
+      "step": 8740
+    },
+    {
+      "epoch": 6.352087114337568,
+      "grad_norm": 1.8438639640808105,
+      "learning_rate": 0.0002,
+      "loss": 0.3514,
+      "step": 8750
+    },
+    {
+      "epoch": 6.35934664246824,
+      "grad_norm": 1.062495470046997,
+      "learning_rate": 0.0002,
+      "loss": 0.303,
+      "step": 8760
+    },
+    {
+      "epoch": 6.366606170598911,
+      "grad_norm": 1.3224315643310547,
+      "learning_rate": 0.0002,
+      "loss": 0.2737,
+      "step": 8770
+    },
+    {
+      "epoch": 6.373865698729583,
+      "grad_norm": 1.399844765663147,
+      "learning_rate": 0.0002,
+      "loss": 0.3445,
+      "step": 8780
+    },
+    {
+      "epoch": 6.381125226860254,
+      "grad_norm": 1.0409915447235107,
+      "learning_rate": 0.0002,
+      "loss": 0.3277,
+      "step": 8790
+    },
+    {
+      "epoch": 6.388384754990925,
+      "grad_norm": 1.5657726526260376,
+      "learning_rate": 0.0002,
+      "loss": 0.3218,
+      "step": 8800
+    },
+    {
+      "epoch": 6.395644283121597,
+      "grad_norm": 1.4098644256591797,
+      "learning_rate": 0.0002,
+      "loss": 0.3031,
+      "step": 8810
+    },
+    {
+      "epoch": 6.402903811252268,
+      "grad_norm": 1.5154732465744019,
+      "learning_rate": 0.0002,
+      "loss": 0.3133,
+      "step": 8820
+    },
+    {
+      "epoch": 6.41016333938294,
+      "grad_norm": 1.1139698028564453,
+      "learning_rate": 0.0002,
+      "loss": 0.3111,
+      "step": 8830
+    },
+    {
+      "epoch": 6.417422867513611,
+      "grad_norm": 1.4149729013442993,
+      "learning_rate": 0.0002,
+      "loss": 0.3553,
+      "step": 8840
+    },
+    {
+      "epoch": 6.424682395644283,
+      "grad_norm": 1.2632299661636353,
+      "learning_rate": 0.0002,
+      "loss": 0.287,
+      "step": 8850
+    },
+    {
+      "epoch": 6.431941923774954,
+      "grad_norm": 1.6636109352111816,
+      "learning_rate": 0.0002,
+      "loss": 0.3198,
+      "step": 8860
+    },
+    {
+      "epoch": 6.439201451905626,
+      "grad_norm": 1.4149386882781982,
+      "learning_rate": 0.0002,
+      "loss": 0.3749,
+      "step": 8870
+    },
+    {
+      "epoch": 6.4464609800362975,
+      "grad_norm": 1.1396206617355347,
+      "learning_rate": 0.0002,
+      "loss": 0.3504,
+      "step": 8880
+    },
+    {
+      "epoch": 6.4537205081669695,
+      "grad_norm": 1.2188775539398193,
+      "learning_rate": 0.0002,
+      "loss": 0.3328,
+      "step": 8890
+    },
+    {
+      "epoch": 6.460980036297641,
+      "grad_norm": 0.9740369319915771,
+      "learning_rate": 0.0002,
+      "loss": 0.3427,
+      "step": 8900
+    },
+    {
+      "epoch": 6.468239564428313,
+      "grad_norm": 1.228569746017456,
+      "learning_rate": 0.0002,
+      "loss": 0.3223,
+      "step": 8910
+    },
+    {
+      "epoch": 6.475499092558984,
+      "grad_norm": 1.5019789934158325,
+      "learning_rate": 0.0002,
+      "loss": 0.3151,
+      "step": 8920
+    },
+    {
+      "epoch": 6.482758620689655,
+      "grad_norm": 1.3320101499557495,
+      "learning_rate": 0.0002,
+      "loss": 0.2916,
+      "step": 8930
+    },
+    {
+      "epoch": 6.490018148820327,
+      "grad_norm": 1.5551502704620361,
+      "learning_rate": 0.0002,
+      "loss": 0.298,
+      "step": 8940
+    },
+    {
+      "epoch": 6.497277676950998,
+      "grad_norm": 1.470131754875183,
+      "learning_rate": 0.0002,
+      "loss": 0.3238,
+      "step": 8950
+    },
+    {
+      "epoch": 6.50453720508167,
+      "grad_norm": 1.1803025007247925,
+      "learning_rate": 0.0002,
+      "loss": 0.2808,
+      "step": 8960
+    },
+    {
+      "epoch": 6.511796733212341,
+      "grad_norm": 1.3505640029907227,
+      "learning_rate": 0.0002,
+      "loss": 0.3025,
+      "step": 8970
+    },
+    {
+      "epoch": 6.519056261343013,
+      "grad_norm": 1.13093900680542,
+      "learning_rate": 0.0002,
+      "loss": 0.3124,
+      "step": 8980
+    },
+    {
+      "epoch": 6.526315789473684,
+      "grad_norm": 1.347386121749878,
+      "learning_rate": 0.0002,
+      "loss": 0.3454,
+      "step": 8990
+    },
+    {
+      "epoch": 6.533575317604356,
+      "grad_norm": 1.7879165410995483,
+      "learning_rate": 0.0002,
+      "loss": 0.3532,
+      "step": 9000
+    },
+    {
+      "epoch": 6.540834845735027,
+      "grad_norm": 1.2168169021606445,
+      "learning_rate": 0.0002,
+      "loss": 0.3382,
+      "step": 9010
+    },
+    {
+      "epoch": 6.548094373865698,
+      "grad_norm": 1.1758877038955688,
+      "learning_rate": 0.0002,
+      "loss": 0.3413,
+      "step": 9020
+    },
+    {
+      "epoch": 6.55535390199637,
+      "grad_norm": 1.7366445064544678,
+      "learning_rate": 0.0002,
+      "loss": 0.2806,
+      "step": 9030
+    },
+    {
+      "epoch": 6.562613430127042,
+      "grad_norm": 1.5919222831726074,
+      "learning_rate": 0.0002,
+      "loss": 0.3437,
+      "step": 9040
+    },
+    {
+      "epoch": 6.569872958257713,
+      "grad_norm": 1.336863398551941,
+      "learning_rate": 0.0002,
+      "loss": 0.3261,
+      "step": 9050
+    },
+    {
+      "epoch": 6.577132486388384,
+      "grad_norm": 1.1769421100616455,
+      "learning_rate": 0.0002,
+      "loss": 0.3103,
+      "step": 9060
+    },
+    {
+      "epoch": 6.584392014519056,
+      "grad_norm": 1.0048751831054688,
+      "learning_rate": 0.0002,
+      "loss": 0.3295,
+      "step": 9070
+    },
+    {
+      "epoch": 6.5916515426497275,
+      "grad_norm": 1.5268515348434448,
+      "learning_rate": 0.0002,
+      "loss": 0.3156,
+      "step": 9080
+    },
+    {
+      "epoch": 6.5989110707803995,
+      "grad_norm": 1.434610366821289,
+      "learning_rate": 0.0002,
+      "loss": 0.3752,
+      "step": 9090
+    },
+    {
+      "epoch": 6.606170598911071,
+      "grad_norm": 1.1151410341262817,
+      "learning_rate": 0.0002,
+      "loss": 0.3375,
+      "step": 9100
+    },
+    {
+      "epoch": 6.613430127041743,
+      "grad_norm": 1.6690642833709717,
+      "learning_rate": 0.0002,
+      "loss": 0.363,
+      "step": 9110
+    },
+    {
+      "epoch": 6.620689655172414,
+      "grad_norm": 1.4495552778244019,
+      "learning_rate": 0.0002,
+      "loss": 0.3703,
+      "step": 9120
+    },
+    {
+      "epoch": 6.627949183303086,
+      "grad_norm": 1.377621054649353,
+      "learning_rate": 0.0002,
+      "loss": 0.3648,
+      "step": 9130
+    },
+    {
+      "epoch": 6.635208711433757,
+      "grad_norm": 1.5459434986114502,
+      "learning_rate": 0.0002,
+      "loss": 0.3766,
+      "step": 9140
+    },
+    {
+      "epoch": 6.642468239564428,
+      "grad_norm": 1.0920850038528442,
+      "learning_rate": 0.0002,
+      "loss": 0.3196,
+      "step": 9150
+    },
+    {
+      "epoch": 6.6497277676951,
+      "grad_norm": 1.6708381175994873,
+      "learning_rate": 0.0002,
+      "loss": 0.3505,
+      "step": 9160
+    },
+    {
+      "epoch": 6.656987295825771,
+      "grad_norm": 1.747514009475708,
+      "learning_rate": 0.0002,
+      "loss": 0.368,
+      "step": 9170
+    },
+    {
+      "epoch": 6.664246823956443,
+      "grad_norm": 1.133466362953186,
+      "learning_rate": 0.0002,
+      "loss": 0.3099,
+      "step": 9180
+    },
+    {
+      "epoch": 6.671506352087114,
+      "grad_norm": 1.394358515739441,
+      "learning_rate": 0.0002,
+      "loss": 0.3175,
+      "step": 9190
+    },
+    {
+      "epoch": 6.678765880217786,
+      "grad_norm": 0.9258374571800232,
+      "learning_rate": 0.0002,
+      "loss": 0.2981,
+      "step": 9200
+    },
+    {
+      "epoch": 6.686025408348457,
+      "grad_norm": 1.3750739097595215,
+      "learning_rate": 0.0002,
+      "loss": 0.3723,
+      "step": 9210
+    },
+    {
+      "epoch": 6.693284936479129,
+      "grad_norm": 0.8604967594146729,
+      "learning_rate": 0.0002,
+      "loss": 0.3441,
+      "step": 9220
+    },
+    {
+      "epoch": 6.7005444646098,
+      "grad_norm": 1.6074559688568115,
+      "learning_rate": 0.0002,
+      "loss": 0.3775,
+      "step": 9230
+    },
+    {
+      "epoch": 6.707803992740472,
+      "grad_norm": 0.9576877355575562,
+      "learning_rate": 0.0002,
+      "loss": 0.3139,
+      "step": 9240
+    },
+    {
+      "epoch": 6.715063520871143,
+      "grad_norm": 1.7193048000335693,
+      "learning_rate": 0.0002,
+      "loss": 0.3633,
+      "step": 9250
+    },
+    {
+      "epoch": 6.722323049001815,
+      "grad_norm": 1.3131844997406006,
+      "learning_rate": 0.0002,
+      "loss": 0.3139,
+      "step": 9260
+    },
+    {
+      "epoch": 6.729582577132486,
+      "grad_norm": 1.2978184223175049,
+      "learning_rate": 0.0002,
+      "loss": 0.3121,
+      "step": 9270
+    },
+    {
+      "epoch": 6.7368421052631575,
+      "grad_norm": 1.4792617559432983,
+      "learning_rate": 0.0002,
+      "loss": 0.3534,
+      "step": 9280
+    },
+    {
+      "epoch": 6.7441016333938295,
+      "grad_norm": 1.1265567541122437,
+      "learning_rate": 0.0002,
+      "loss": 0.3429,
+      "step": 9290
+    },
+    {
+      "epoch": 6.751361161524501,
+      "grad_norm": 1.8553377389907837,
+      "learning_rate": 0.0002,
+      "loss": 0.3526,
+      "step": 9300
+    },
+    {
+      "epoch": 6.758620689655173,
+      "grad_norm": 1.3602519035339355,
+      "learning_rate": 0.0002,
+      "loss": 0.3666,
+      "step": 9310
+    },
+    {
+      "epoch": 6.765880217785844,
+      "grad_norm": 1.2874794006347656,
+      "learning_rate": 0.0002,
+      "loss": 0.2922,
+      "step": 9320
+    },
+    {
+      "epoch": 6.773139745916516,
+      "grad_norm": 1.4834712743759155,
+      "learning_rate": 0.0002,
+      "loss": 0.3816,
+      "step": 9330
+    },
+    {
+      "epoch": 6.780399274047187,
+      "grad_norm": 2.0824034214019775,
+      "learning_rate": 0.0002,
+      "loss": 0.3557,
+      "step": 9340
+    },
+    {
+      "epoch": 6.787658802177859,
+      "grad_norm": 1.2267698049545288,
+      "learning_rate": 0.0002,
+      "loss": 0.3174,
+      "step": 9350
+    },
+    {
+      "epoch": 6.79491833030853,
+      "grad_norm": 1.4485498666763306,
+      "learning_rate": 0.0002,
+      "loss": 0.3665,
+      "step": 9360
+    },
+    {
+      "epoch": 6.802177858439201,
+      "grad_norm": 1.3199396133422852,
+      "learning_rate": 0.0002,
+      "loss": 0.3676,
+      "step": 9370
+    },
+    {
+      "epoch": 6.809437386569873,
+      "grad_norm": 1.2552456855773926,
+      "learning_rate": 0.0002,
+      "loss": 0.298,
+      "step": 9380
+    },
+    {
+      "epoch": 6.816696914700545,
+      "grad_norm": 1.3895127773284912,
+      "learning_rate": 0.0002,
+      "loss": 0.3152,
+      "step": 9390
+    },
+    {
+      "epoch": 6.823956442831216,
+      "grad_norm": 1.7637823820114136,
+      "learning_rate": 0.0002,
+      "loss": 0.3375,
+      "step": 9400
+    },
+    {
+      "epoch": 6.831215970961887,
+      "grad_norm": 1.6004475355148315,
+      "learning_rate": 0.0002,
+      "loss": 0.3234,
+      "step": 9410
+    },
+    {
+      "epoch": 6.838475499092559,
+      "grad_norm": 1.4133695363998413,
+      "learning_rate": 0.0002,
+      "loss": 0.3364,
+      "step": 9420
+    },
+    {
+      "epoch": 6.84573502722323,
+      "grad_norm": 1.1583502292633057,
+      "learning_rate": 0.0002,
+      "loss": 0.3656,
+      "step": 9430
+    },
+    {
+      "epoch": 6.852994555353902,
+      "grad_norm": 1.3769075870513916,
+      "learning_rate": 0.0002,
+      "loss": 0.3499,
+      "step": 9440
+    },
+    {
+      "epoch": 6.860254083484573,
+      "grad_norm": 1.1831218004226685,
+      "learning_rate": 0.0002,
+      "loss": 0.3333,
+      "step": 9450
+    },
+    {
+      "epoch": 6.867513611615245,
+      "grad_norm": 1.6092621088027954,
+      "learning_rate": 0.0002,
+      "loss": 0.3501,
+      "step": 9460
+    },
+    {
+      "epoch": 6.874773139745916,
+      "grad_norm": 1.3850210905075073,
+      "learning_rate": 0.0002,
+      "loss": 0.3933,
+      "step": 9470
+    },
+    {
+      "epoch": 6.882032667876588,
+      "grad_norm": 1.4119619131088257,
+      "learning_rate": 0.0002,
+      "loss": 0.3868,
+      "step": 9480
+    },
+    {
+      "epoch": 6.8892921960072595,
+      "grad_norm": 1.3494242429733276,
+      "learning_rate": 0.0002,
+      "loss": 0.3939,
+      "step": 9490
+    },
+    {
+      "epoch": 6.896551724137931,
+      "grad_norm": 1.3130041360855103,
+      "learning_rate": 0.0002,
+      "loss": 0.3217,
+      "step": 9500
+    },
+    {
+      "epoch": 6.903811252268603,
+      "grad_norm": 1.169256329536438,
+      "learning_rate": 0.0002,
+      "loss": 0.3738,
+      "step": 9510
+    },
+    {
+      "epoch": 6.911070780399274,
+      "grad_norm": 1.7475035190582275,
+      "learning_rate": 0.0002,
+      "loss": 0.408,
+      "step": 9520
+    },
+    {
+      "epoch": 6.918330308529946,
+      "grad_norm": 1.440434217453003,
+      "learning_rate": 0.0002,
+      "loss": 0.3407,
+      "step": 9530
+    },
+    {
+      "epoch": 6.925589836660617,
+      "grad_norm": 1.6768704652786255,
+      "learning_rate": 0.0002,
+      "loss": 0.3707,
+      "step": 9540
+    },
+    {
+      "epoch": 6.932849364791289,
+      "grad_norm": 1.3720577955245972,
+      "learning_rate": 0.0002,
+      "loss": 0.3283,
+      "step": 9550
+    },
+    {
+      "epoch": 6.94010889292196,
+      "grad_norm": 1.8140523433685303,
+      "learning_rate": 0.0002,
+      "loss": 0.3257,
+      "step": 9560
+    },
+    {
+      "epoch": 6.947368421052632,
+      "grad_norm": 1.1828241348266602,
+      "learning_rate": 0.0002,
+      "loss": 0.3308,
+      "step": 9570
+    },
+    {
+      "epoch": 6.954627949183303,
+      "grad_norm": 1.2755135297775269,
+      "learning_rate": 0.0002,
+      "loss": 0.3536,
+      "step": 9580
+    },
+    {
+      "epoch": 6.961887477313975,
+      "grad_norm": 1.622009038925171,
+      "learning_rate": 0.0002,
+      "loss": 0.3711,
+      "step": 9590
+    },
+    {
+      "epoch": 6.969147005444646,
+      "grad_norm": 1.1543664932250977,
+      "learning_rate": 0.0002,
+      "loss": 0.3529,
+      "step": 9600
+    },
+    {
+      "epoch": 6.976406533575318,
+      "grad_norm": 1.6755319833755493,
+      "learning_rate": 0.0002,
+      "loss": 0.416,
+      "step": 9610
+    },
+    {
+      "epoch": 6.983666061705989,
+      "grad_norm": 1.3726437091827393,
+      "learning_rate": 0.0002,
+      "loss": 0.339,
+      "step": 9620
+    },
+    {
+      "epoch": 6.99092558983666,
+      "grad_norm": 1.1605958938598633,
+      "learning_rate": 0.0002,
+      "loss": 0.3709,
+      "step": 9630
+    },
+    {
+      "epoch": 6.998185117967332,
+      "grad_norm": 1.5371781587600708,
+      "learning_rate": 0.0002,
+      "loss": 0.371,
+      "step": 9640
+    },
+    {
+      "epoch": 6.999637023593467,
+      "eval_loss": 1.6280181407928467,
+      "eval_runtime": 46.1964,
+      "eval_samples_per_second": 9.438,
+      "eval_steps_per_second": 1.191,
+      "step": 9642
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11016,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.952749172195328e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..140bceabdec44fc473842cb9ace4a62a4dc9fe89
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-9642/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4abec5480aff960b5a7b03a48be1cdb1ae29ba8792469c451f5c2f269e4be93b
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/special_tokens_map.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..0acb52c84d6ea33178bee426ec6706bfba8ba637
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/special_tokens_map.json
@@ -0,0 +1,28 @@
+{
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<eos>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/tokenizer.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..af0eac5c0056f83b8f3fcdb79165f8847111c305
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f289bc05132635a8bc7aca7aa21255efd5e18f3710f43e3cdb96bcd41be4922
+size 17525357
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/tokenizer.model b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..796efe9ab515c15e146ce7588e6d7b9b8134dbf8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/tokenizer_config.json b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..1aa249f4dc9f84e87ad8983458e7800ae5bf5454
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/tokenizer_config.json
@@ -0,0 +1,2013 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "5": {
+      "content": "<2mass>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "6": {
+      "content": "[@BOS@]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "7": {
+      "content": "<unused0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "8": {
+      "content": "<unused1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "9": {
+      "content": "<unused2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "10": {
+      "content": "<unused3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "11": {
+      "content": "<unused4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "12": {
+      "content": "<unused5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "13": {
+      "content": "<unused6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "14": {
+      "content": "<unused7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "15": {
+      "content": "<unused8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "16": {
+      "content": "<unused9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "17": {
+      "content": "<unused10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "18": {
+      "content": "<unused11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "19": {
+      "content": "<unused12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "20": {
+      "content": "<unused13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "21": {
+      "content": "<unused14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "22": {
+      "content": "<unused15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "23": {
+      "content": "<unused16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "24": {
+      "content": "<unused17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "25": {
+      "content": "<unused18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "26": {
+      "content": "<unused19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "27": {
+      "content": "<unused20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "28": {
+      "content": "<unused21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "29": {
+      "content": "<unused22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "30": {
+      "content": "<unused23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "31": {
+      "content": "<unused24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "32": {
+      "content": "<unused25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "33": {
+      "content": "<unused26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "34": {
+      "content": "<unused27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "35": {
+      "content": "<unused28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "36": {
+      "content": "<unused29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "37": {
+      "content": "<unused30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "38": {
+      "content": "<unused31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "39": {
+      "content": "<unused32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "40": {
+      "content": "<unused33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "41": {
+      "content": "<unused34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "42": {
+      "content": "<unused35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "43": {
+      "content": "<unused36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "44": {
+      "content": "<unused37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "45": {
+      "content": "<unused38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "46": {
+      "content": "<unused39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "47": {
+      "content": "<unused40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "48": {
+      "content": "<unused41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "49": {
+      "content": "<unused42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50": {
+      "content": "<unused43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "51": {
+      "content": "<unused44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "52": {
+      "content": "<unused45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "53": {
+      "content": "<unused46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "54": {
+      "content": "<unused47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "55": {
+      "content": "<unused48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "56": {
+      "content": "<unused49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "57": {
+      "content": "<unused50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "58": {
+      "content": "<unused51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "59": {
+      "content": "<unused52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "60": {
+      "content": "<unused53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "61": {
+      "content": "<unused54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "62": {
+      "content": "<unused55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "63": {
+      "content": "<unused56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "64": {
+      "content": "<unused57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "65": {
+      "content": "<unused58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "66": {
+      "content": "<unused59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "67": {
+      "content": "<unused60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "68": {
+      "content": "<unused61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "69": {
+      "content": "<unused62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "70": {
+      "content": "<unused63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "71": {
+      "content": "<unused64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "72": {
+      "content": "<unused65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "73": {
+      "content": "<unused66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "74": {
+      "content": "<unused67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "75": {
+      "content": "<unused68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "76": {
+      "content": "<unused69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "77": {
+      "content": "<unused70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "78": {
+      "content": "<unused71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "79": {
+      "content": "<unused72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "80": {
+      "content": "<unused73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "81": {
+      "content": "<unused74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "82": {
+      "content": "<unused75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "83": {
+      "content": "<unused76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "84": {
+      "content": "<unused77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "85": {
+      "content": "<unused78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "86": {
+      "content": "<unused79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "87": {
+      "content": "<unused80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "88": {
+      "content": "<unused81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "89": {
+      "content": "<unused82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "90": {
+      "content": "<unused83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "91": {
+      "content": "<unused84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "92": {
+      "content": "<unused85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "93": {
+      "content": "<unused86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "94": {
+      "content": "<unused87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "95": {
+      "content": "<unused88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "96": {
+      "content": "<unused89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "97": {
+      "content": "<unused90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "98": {
+      "content": "<unused91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "99": {
+      "content": "<unused92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "100": {
+      "content": "<unused93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "101": {
+      "content": "<unused94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "102": {
+      "content": "<unused95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "103": {
+      "content": "<unused96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "104": {
+      "content": "<unused97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "105": {
+      "content": "<unused98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "106": {
+      "content": "<start_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "107": {
+      "content": "<end_of_turn>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "108": {
+      "content": "\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "109": {
+      "content": "\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "110": {
+      "content": "\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "111": {
+      "content": "\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "112": {
+      "content": "\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "113": {
+      "content": "\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "114": {
+      "content": "\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "115": {
+      "content": "\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "116": {
+      "content": "\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "117": {
+      "content": "\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "118": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "119": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "120": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "121": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "122": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "123": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "124": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "125": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "126": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "127": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "128": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "129": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "130": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "131": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "132": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "133": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "134": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "135": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "136": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "137": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "138": {
+      "content": "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "139": {
+      "content": "▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "140": {
+      "content": "▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "141": {
+      "content": "▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "142": {
+      "content": "▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "143": {
+      "content": "▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "144": {
+      "content": "▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "145": {
+      "content": "▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "146": {
+      "content": "▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "147": {
+      "content": "▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "148": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "149": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "150": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "152": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "153": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "154": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "155": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "156": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "157": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "158": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "159": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "160": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "161": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "162": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "163": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "164": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "165": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "166": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "167": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "168": {
+      "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "169": {
+      "content": "<table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "170": {
+      "content": "<caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "171": {
+      "content": "<thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "172": {
+      "content": "<tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "173": {
+      "content": "<tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "174": {
+      "content": "<tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "175": {
+      "content": "<th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "176": {
+      "content": "<td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "177": {
+      "content": "</table>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "178": {
+      "content": "</caption>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "179": {
+      "content": "</thead>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "180": {
+      "content": "</tbody>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "181": {
+      "content": "</tfoot>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "182": {
+      "content": "</tr>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "183": {
+      "content": "</th>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "184": {
+      "content": "</td>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "185": {
+      "content": "<h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "186": {
+      "content": "<h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "187": {
+      "content": "<h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "188": {
+      "content": "<h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "189": {
+      "content": "<h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "190": {
+      "content": "<h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "191": {
+      "content": "<blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "192": {
+      "content": "</h1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "193": {
+      "content": "</h2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "194": {
+      "content": "</h3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "195": {
+      "content": "</h4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "196": {
+      "content": "</h5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "197": {
+      "content": "</h6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "198": {
+      "content": "</blockquote>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "199": {
+      "content": "<strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "200": {
+      "content": "<em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "201": {
+      "content": "<b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "202": {
+      "content": "<i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "203": {
+      "content": "<u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "204": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "205": {
+      "content": "<sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "206": {
+      "content": "<sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "207": {
+      "content": "<code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "208": {
+      "content": "</strong>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "209": {
+      "content": "</em>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "210": {
+      "content": "</b>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "211": {
+      "content": "</i>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "212": {
+      "content": "</u>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "213": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "214": {
+      "content": "</sub>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "215": {
+      "content": "</sup>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "216": {
+      "content": "</code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255968": {
+      "content": "[toxicity=0]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255969": {
+      "content": "\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255970": {
+      "content": "\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255971": {
+      "content": "\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255972": {
+      "content": "\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255973": {
+      "content": "\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255974": {
+      "content": "\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255975": {
+      "content": "\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255976": {
+      "content": "\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255977": {
+      "content": "\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255978": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255979": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255980": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255981": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255982": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255983": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255984": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255985": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255986": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255987": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255988": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255989": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255990": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255991": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255992": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255993": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255994": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255995": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255996": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255997": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255998": {
+      "content": "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "255999": {
+      "content": "<unused99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<start_of_turn>",
+    "<end_of_turn>"
+  ],
+  "bos_token": "<bos>",
+  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<eos>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<eos>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/training_args.bin b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..140bceabdec44fc473842cb9ace4a62a4dc9fe89
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4abec5480aff960b5a7b03a48be1cdb1ae29ba8792469c451f5c2f269e4be93b
+size 5560
diff --git a/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/training_log.jsonl b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..13bc1bb55cfbc6e4f997e42f0cd1a650ac7068f8
--- /dev/null
+++ b/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/training_log.jsonl
@@ -0,0 +1,8 @@
+{"epoch": 0.9996370235934664, "step": 1377, "epoch_duration": 4561.8241751194, "total_accumulated_duration": 4561.8241751194, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 11696.9921875}, "avg_memory_reserved": {"GPU_0": 12758.0}, "peak_memory_reserved": {"GPU_0": 12758.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4936, "grad_norm": 0.46010470390319824, "learning_rate": 0.0002, "epoch": 0.007259528130671506, "step": 10}, {"loss": 1.8442, "grad_norm": 0.5103179216384888, "learning_rate": 0.0002, "epoch": 0.014519056261343012, "step": 20}, {"loss": 1.559, "grad_norm": 0.6282716989517212, "learning_rate": 0.0002, "epoch": 0.021778584392014518, "step": 30}, {"loss": 1.3618, "grad_norm": 1.2480497360229492, "learning_rate": 0.0002, "epoch": 0.029038112522686024, "step": 40}, {"loss": 1.3874, "grad_norm": 0.4114855229854584, "learning_rate": 0.0002, "epoch": 0.036297640653357534, "step": 50}, {"loss": 1.2836, "grad_norm": 0.49482840299606323, "learning_rate": 0.0002, "epoch": 0.043557168784029036, "step": 60}, {"loss": 1.1728, "grad_norm": 0.4536272883415222, "learning_rate": 0.0002, "epoch": 0.050816696914700546, "step": 70}, {"loss": 1.1125, "grad_norm": 0.32328274846076965, "learning_rate": 0.0002, "epoch": 0.05807622504537205, "step": 80}, {"loss": 1.2447, "grad_norm": 0.40990468859672546, "learning_rate": 0.0002, "epoch": 0.06533575317604355, "step": 90}, {"loss": 1.2216, "grad_norm": 0.37273502349853516, "learning_rate": 0.0002, "epoch": 0.07259528130671507, "step": 100}, {"loss": 1.2259, "grad_norm": 0.3903871476650238, "learning_rate": 0.0002, "epoch": 0.07985480943738657, "step": 110}, {"loss": 1.3718, "grad_norm": 0.3272787928581238, "learning_rate": 0.0002, "epoch": 0.08711433756805807, "step": 120}, {"loss": 1.1832, "grad_norm": 0.3622824251651764, "learning_rate": 0.0002, "epoch": 0.09437386569872959, "step": 130}, {"loss": 1.2966, "grad_norm": 0.3503916561603546, "learning_rate": 0.0002, "epoch": 0.10163339382940109, "step": 140}, {"loss": 1.0913, "grad_norm": 0.32787832617759705, "learning_rate": 0.0002, "epoch": 0.1088929219600726, "step": 150}, {"loss": 1.2143, "grad_norm": 0.5822657942771912, "learning_rate": 0.0002, "epoch": 0.1161524500907441, "step": 160}, {"loss": 1.2091, "grad_norm": 0.28028249740600586, "learning_rate": 0.0002, "epoch": 0.12341197822141561, "step": 170}, {"loss": 1.1731, "grad_norm": 0.33602750301361084, "learning_rate": 0.0002, "epoch": 0.1306715063520871, "step": 180}, {"loss": 1.127, "grad_norm": 0.29106274247169495, "learning_rate": 0.0002, "epoch": 0.13793103448275862, "step": 190}, {"loss": 1.3261, "grad_norm": 0.38753363490104675, "learning_rate": 0.0002, "epoch": 0.14519056261343014, "step": 200}, {"loss": 1.1536, "grad_norm": 0.361009418964386, "learning_rate": 0.0002, "epoch": 0.15245009074410162, "step": 210}, {"loss": 1.1483, "grad_norm": 0.6743836402893066, "learning_rate": 0.0002, "epoch": 0.15970961887477314, "step": 220}, {"loss": 1.264, "grad_norm": 0.3923613727092743, "learning_rate": 0.0002, "epoch": 0.16696914700544466, "step": 230}, {"loss": 1.0437, "grad_norm": 0.2809699773788452, "learning_rate": 0.0002, "epoch": 0.17422867513611615, "step": 240}, {"loss": 1.1826, "grad_norm": 0.3631494641304016, "learning_rate": 0.0002, "epoch": 0.18148820326678766, "step": 250}, {"loss": 1.1584, "grad_norm": 0.24658171832561493, "learning_rate": 0.0002, "epoch": 0.18874773139745918, "step": 260}, {"loss": 1.0065, "grad_norm": 0.5780664682388306, "learning_rate": 0.0002, "epoch": 0.19600725952813067, "step": 270}, {"loss": 1.1251, "grad_norm": 0.3056720495223999, "learning_rate": 0.0002, "epoch": 0.20326678765880218, "step": 280}, {"loss": 1.1503, "grad_norm": 0.2857084572315216, "learning_rate": 0.0002, "epoch": 0.21052631578947367, "step": 290}, {"loss": 1.1215, "grad_norm": 0.6645345687866211, "learning_rate": 0.0002, "epoch": 0.2177858439201452, "step": 300}, {"loss": 1.127, "grad_norm": 0.5966078639030457, "learning_rate": 0.0002, "epoch": 0.2250453720508167, "step": 310}, {"loss": 1.1974, "grad_norm": 0.40937140583992004, "learning_rate": 0.0002, "epoch": 0.2323049001814882, "step": 320}, {"loss": 1.2385, "grad_norm": 0.5642806887626648, "learning_rate": 0.0002, "epoch": 0.2395644283121597, "step": 330}, {"loss": 1.1703, "grad_norm": 0.2750748097896576, "learning_rate": 0.0002, "epoch": 0.24682395644283123, "step": 340}, {"loss": 1.2598, "grad_norm": 0.34350234270095825, "learning_rate": 0.0002, "epoch": 0.2540834845735027, "step": 350}, {"loss": 1.1942, "grad_norm": 0.6767239570617676, "learning_rate": 0.0002, "epoch": 0.2613430127041742, "step": 360}, {"loss": 1.1436, "grad_norm": 0.31006959080696106, "learning_rate": 0.0002, "epoch": 0.26860254083484575, "step": 370}, {"loss": 1.2084, "grad_norm": 0.3825474679470062, "learning_rate": 0.0002, "epoch": 0.27586206896551724, "step": 380}, {"loss": 1.1523, "grad_norm": 0.30241551995277405, "learning_rate": 0.0002, "epoch": 0.2831215970961887, "step": 390}, {"loss": 1.1298, "grad_norm": 0.2962397336959839, "learning_rate": 0.0002, "epoch": 0.29038112522686027, "step": 400}, {"loss": 1.1299, "grad_norm": 0.2600369155406952, "learning_rate": 0.0002, "epoch": 0.29764065335753176, "step": 410}, {"loss": 1.1366, "grad_norm": 0.3675060272216797, "learning_rate": 0.0002, "epoch": 0.30490018148820325, "step": 420}, {"loss": 1.156, "grad_norm": 0.3429498076438904, "learning_rate": 0.0002, "epoch": 0.3121597096188748, "step": 430}, {"loss": 1.2741, "grad_norm": 0.34311825037002563, "learning_rate": 0.0002, "epoch": 0.3194192377495463, "step": 440}, {"loss": 1.3523, "grad_norm": 0.37872210144996643, "learning_rate": 0.0002, "epoch": 0.32667876588021777, "step": 450}, {"loss": 1.1365, "grad_norm": 0.33271121978759766, "learning_rate": 0.0002, "epoch": 0.3339382940108893, "step": 460}, {"loss": 1.1572, "grad_norm": 0.34605276584625244, "learning_rate": 0.0002, "epoch": 0.3411978221415608, "step": 470}, {"loss": 1.2755, "grad_norm": 0.41050562262535095, "learning_rate": 0.0002, "epoch": 0.3484573502722323, "step": 480}, {"loss": 1.075, "grad_norm": 0.2066836953163147, "learning_rate": 0.0002, "epoch": 0.35571687840290384, "step": 490}, {"loss": 1.2072, "grad_norm": 0.2859014868736267, "learning_rate": 0.0002, "epoch": 0.3629764065335753, "step": 500}, {"loss": 1.1435, "grad_norm": 0.28763777017593384, "learning_rate": 0.0002, "epoch": 0.3702359346642468, "step": 510}, {"loss": 1.1341, "grad_norm": 0.2730471193790436, "learning_rate": 0.0002, "epoch": 0.37749546279491836, "step": 520}, {"loss": 1.348, "grad_norm": 0.3968936800956726, "learning_rate": 0.0002, "epoch": 0.38475499092558985, "step": 530}, {"loss": 1.2363, "grad_norm": 0.3624701201915741, "learning_rate": 0.0002, "epoch": 0.39201451905626133, "step": 540}, {"loss": 1.1206, "grad_norm": 0.3303608298301697, "learning_rate": 0.0002, "epoch": 0.3992740471869328, "step": 550}, {"loss": 1.203, "grad_norm": 0.33507466316223145, "learning_rate": 0.0002, "epoch": 0.40653357531760437, "step": 560}, {"loss": 1.2294, "grad_norm": 0.3297670781612396, "learning_rate": 0.0002, "epoch": 0.41379310344827586, "step": 570}, {"loss": 1.2, "grad_norm": 0.32334890961647034, "learning_rate": 0.0002, "epoch": 0.42105263157894735, "step": 580}, {"loss": 1.112, "grad_norm": 0.30281195044517517, "learning_rate": 0.0002, "epoch": 0.4283121597096189, "step": 590}, {"loss": 1.1352, "grad_norm": 0.5900027751922607, "learning_rate": 0.0002, "epoch": 0.4355716878402904, "step": 600}, {"loss": 1.1575, "grad_norm": 0.28437477350234985, "learning_rate": 0.0002, "epoch": 0.44283121597096187, "step": 610}, {"loss": 1.2009, "grad_norm": 0.39601704478263855, "learning_rate": 0.0002, "epoch": 0.4500907441016334, "step": 620}, {"loss": 1.2302, "grad_norm": 0.41971510648727417, "learning_rate": 0.0002, "epoch": 0.4573502722323049, "step": 630}, {"loss": 1.236, "grad_norm": 0.33814409375190735, "learning_rate": 0.0002, "epoch": 0.4646098003629764, "step": 640}, {"loss": 1.2471, "grad_norm": 0.575718104839325, "learning_rate": 0.0002, "epoch": 0.47186932849364793, "step": 650}, {"loss": 1.1012, "grad_norm": 0.37927401065826416, "learning_rate": 0.0002, "epoch": 0.4791288566243194, "step": 660}, {"loss": 1.1552, "grad_norm": 0.3224332630634308, "learning_rate": 0.0002, "epoch": 0.4863883847549909, "step": 670}, {"loss": 1.1354, "grad_norm": 0.32683515548706055, "learning_rate": 0.0002, "epoch": 0.49364791288566245, "step": 680}, {"loss": 1.207, "grad_norm": 0.4316163659095764, "learning_rate": 0.0002, "epoch": 0.5009074410163339, "step": 690}, {"loss": 1.2116, "grad_norm": 0.342602401971817, "learning_rate": 0.0002, "epoch": 0.5081669691470054, "step": 700}, {"loss": 1.0433, "grad_norm": 0.2794898748397827, "learning_rate": 0.0002, "epoch": 0.515426497277677, "step": 710}, {"loss": 1.2395, "grad_norm": 0.3322339951992035, "learning_rate": 0.0002, "epoch": 0.5226860254083484, "step": 720}, {"loss": 1.0762, "grad_norm": 0.3088509142398834, "learning_rate": 0.0002, "epoch": 0.52994555353902, "step": 730}, {"loss": 1.1495, "grad_norm": 0.24444378912448883, "learning_rate": 0.0002, "epoch": 0.5372050816696915, "step": 740}, {"loss": 1.2176, "grad_norm": 0.3483171761035919, "learning_rate": 0.0002, "epoch": 0.5444646098003629, "step": 750}, {"loss": 1.2248, "grad_norm": 0.3425690531730652, "learning_rate": 0.0002, "epoch": 0.5517241379310345, "step": 760}, {"loss": 1.1127, "grad_norm": 0.31841927766799927, "learning_rate": 0.0002, "epoch": 0.558983666061706, "step": 770}, {"loss": 1.1841, "grad_norm": 0.39423868060112, "learning_rate": 0.0002, "epoch": 0.5662431941923775, "step": 780}, {"loss": 1.0583, "grad_norm": 0.30328479409217834, "learning_rate": 0.0002, "epoch": 0.573502722323049, "step": 790}, {"loss": 1.0718, "grad_norm": 0.24475938081741333, "learning_rate": 0.0002, "epoch": 0.5807622504537205, "step": 800}, {"loss": 1.2006, "grad_norm": 0.37132805585861206, "learning_rate": 0.0002, "epoch": 0.588021778584392, "step": 810}, {"loss": 1.1544, "grad_norm": 0.32195979356765747, "learning_rate": 0.0002, "epoch": 0.5952813067150635, "step": 820}, {"loss": 0.9937, "grad_norm": 0.2848738729953766, "learning_rate": 0.0002, "epoch": 0.6025408348457351, "step": 830}, {"loss": 1.0937, "grad_norm": 0.28015264868736267, "learning_rate": 0.0002, "epoch": 0.6098003629764065, "step": 840}, {"loss": 1.1234, "grad_norm": 0.37796008586883545, "learning_rate": 0.0002, "epoch": 0.617059891107078, "step": 850}, {"loss": 1.1455, "grad_norm": 0.39311841130256653, "learning_rate": 0.0002, "epoch": 0.6243194192377496, "step": 860}, {"loss": 1.1066, "grad_norm": 0.2761685252189636, "learning_rate": 0.0002, "epoch": 0.631578947368421, "step": 870}, {"loss": 1.0859, "grad_norm": 0.3826720118522644, "learning_rate": 0.0002, "epoch": 0.6388384754990926, "step": 880}, {"loss": 1.1182, "grad_norm": 0.30076679587364197, "learning_rate": 0.0002, "epoch": 0.6460980036297641, "step": 890}, {"loss": 1.0927, "grad_norm": 0.21997687220573425, "learning_rate": 0.0002, "epoch": 0.6533575317604355, "step": 900}, {"loss": 1.1603, "grad_norm": 0.32593777775764465, "learning_rate": 0.0002, "epoch": 0.6606170598911071, "step": 910}, {"loss": 1.1578, "grad_norm": 0.30347898602485657, "learning_rate": 0.0002, "epoch": 0.6678765880217786, "step": 920}, {"loss": 1.1642, "grad_norm": 0.44173774123191833, "learning_rate": 0.0002, "epoch": 0.6751361161524501, "step": 930}, {"loss": 1.1832, "grad_norm": 0.2507467269897461, "learning_rate": 0.0002, "epoch": 0.6823956442831216, "step": 940}, {"loss": 1.3001, "grad_norm": 0.29463833570480347, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 950}, {"loss": 1.1651, "grad_norm": 0.9363154172897339, "learning_rate": 0.0002, "epoch": 0.6969147005444646, "step": 960}, {"loss": 1.158, "grad_norm": 0.3236212134361267, "learning_rate": 0.0002, "epoch": 0.7041742286751361, "step": 970}, {"loss": 1.1078, "grad_norm": 0.3123254179954529, "learning_rate": 0.0002, "epoch": 0.7114337568058077, "step": 980}, {"loss": 1.1578, "grad_norm": 0.3395805060863495, "learning_rate": 0.0002, "epoch": 0.7186932849364791, "step": 990}, {"loss": 1.0754, "grad_norm": 0.3240964412689209, "learning_rate": 0.0002, "epoch": 0.7259528130671506, "step": 1000}, {"loss": 1.0313, "grad_norm": 0.31902948021888733, "learning_rate": 0.0002, "epoch": 0.7332123411978222, "step": 1010}, {"loss": 1.1664, "grad_norm": 0.4848408102989197, "learning_rate": 0.0002, "epoch": 0.7404718693284936, "step": 1020}, {"loss": 1.0838, "grad_norm": 0.33006033301353455, "learning_rate": 0.0002, "epoch": 0.7477313974591652, "step": 1030}, {"loss": 1.0979, "grad_norm": 0.2928730547428131, "learning_rate": 0.0002, "epoch": 0.7549909255898367, "step": 1040}, {"loss": 1.1351, "grad_norm": 0.3529164791107178, "learning_rate": 0.0002, "epoch": 0.7622504537205081, "step": 1050}, {"loss": 1.1274, "grad_norm": 0.2736213803291321, "learning_rate": 0.0002, "epoch": 0.7695099818511797, "step": 1060}, {"loss": 1.1169, "grad_norm": 0.7200686931610107, "learning_rate": 0.0002, "epoch": 0.7767695099818511, "step": 1070}, {"loss": 1.1171, "grad_norm": 0.33396708965301514, "learning_rate": 0.0002, "epoch": 0.7840290381125227, "step": 1080}, {"loss": 1.2447, "grad_norm": 1.5760449171066284, "learning_rate": 0.0002, "epoch": 0.7912885662431942, "step": 1090}, {"loss": 1.1126, "grad_norm": 0.28138381242752075, "learning_rate": 0.0002, "epoch": 0.7985480943738656, "step": 1100}, {"loss": 1.2012, "grad_norm": 0.2597472369670868, "learning_rate": 0.0002, "epoch": 0.8058076225045372, "step": 1110}, {"loss": 1.1177, "grad_norm": 0.3305445611476898, "learning_rate": 0.0002, "epoch": 0.8130671506352087, "step": 1120}, {"loss": 1.0849, "grad_norm": 0.3934599459171295, "learning_rate": 0.0002, "epoch": 0.8203266787658802, "step": 1130}, {"loss": 1.0889, "grad_norm": 0.3472191393375397, "learning_rate": 0.0002, "epoch": 0.8275862068965517, "step": 1140}, {"loss": 1.265, "grad_norm": 0.2857365906238556, "learning_rate": 0.0002, "epoch": 0.8348457350272233, "step": 1150}, {"loss": 1.03, "grad_norm": 0.3207702934741974, "learning_rate": 0.0002, "epoch": 0.8421052631578947, "step": 1160}, {"loss": 1.1669, "grad_norm": 0.3176484704017639, "learning_rate": 0.0002, "epoch": 0.8493647912885662, "step": 1170}, {"loss": 1.1386, "grad_norm": 0.40685558319091797, "learning_rate": 0.0002, "epoch": 0.8566243194192378, "step": 1180}, {"loss": 1.1383, "grad_norm": 0.31125199794769287, "learning_rate": 0.0002, "epoch": 0.8638838475499092, "step": 1190}, {"loss": 1.1512, "grad_norm": 0.7361181378364563, "learning_rate": 0.0002, "epoch": 0.8711433756805808, "step": 1200}, {"loss": 0.9983, "grad_norm": 0.33699527382850647, "learning_rate": 0.0002, "epoch": 0.8784029038112523, "step": 1210}, {"loss": 1.11, "grad_norm": 0.3315220773220062, "learning_rate": 0.0002, "epoch": 0.8856624319419237, "step": 1220}, {"loss": 1.0266, "grad_norm": 0.6256054043769836, "learning_rate": 0.0002, "epoch": 0.8929219600725953, "step": 1230}, {"loss": 1.1738, "grad_norm": 0.3692137897014618, "learning_rate": 0.0002, "epoch": 0.9001814882032668, "step": 1240}, {"loss": 1.1112, "grad_norm": 0.3538484573364258, "learning_rate": 0.0002, "epoch": 0.9074410163339383, "step": 1250}, {"loss": 1.0878, "grad_norm": 0.27863040566444397, "learning_rate": 0.0002, "epoch": 0.9147005444646098, "step": 1260}, {"loss": 1.0826, "grad_norm": 0.3322528302669525, "learning_rate": 0.0002, "epoch": 0.9219600725952813, "step": 1270}, {"loss": 1.0158, "grad_norm": 0.7553173303604126, "learning_rate": 0.0002, "epoch": 0.9292196007259528, "step": 1280}, {"loss": 1.198, "grad_norm": 0.4856191575527191, "learning_rate": 0.0002, "epoch": 0.9364791288566243, "step": 1290}, {"loss": 1.1678, "grad_norm": 0.3668074905872345, "learning_rate": 0.0002, "epoch": 0.9437386569872959, "step": 1300}, {"loss": 1.1809, "grad_norm": 0.29851067066192627, "learning_rate": 0.0002, "epoch": 0.9509981851179673, "step": 1310}, {"loss": 1.1299, "grad_norm": 0.276664674282074, "learning_rate": 0.0002, "epoch": 0.9582577132486388, "step": 1320}, {"loss": 1.0597, "grad_norm": 0.2941018342971802, "learning_rate": 0.0002, "epoch": 0.9655172413793104, "step": 1330}, {"loss": 1.1731, "grad_norm": 0.3505859076976776, "learning_rate": 0.0002, "epoch": 0.9727767695099818, "step": 1340}, {"loss": 1.1809, "grad_norm": 0.3067687451839447, "learning_rate": 0.0002, "epoch": 0.9800362976406534, "step": 1350}, {"loss": 1.1511, "grad_norm": 0.27151066064834595, "learning_rate": 0.0002, "epoch": 0.9872958257713249, "step": 1360}, {"loss": 1.1919, "grad_norm": 0.36370083689689636, "learning_rate": 0.0002, "epoch": 0.9945553539019963, "step": 1370}]}
+{"epoch": 2.0, "step": 2755, "epoch_duration": 4702.509291410446, "total_accumulated_duration": 9264.333466529846, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-1377", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4936, "grad_norm": 0.46010470390319824, "learning_rate": 0.0002, "epoch": 0.007259528130671506, "step": 10}, {"loss": 1.8442, "grad_norm": 0.5103179216384888, "learning_rate": 0.0002, "epoch": 0.014519056261343012, "step": 20}, {"loss": 1.559, "grad_norm": 0.6282716989517212, "learning_rate": 0.0002, "epoch": 0.021778584392014518, "step": 30}, {"loss": 1.3618, "grad_norm": 1.2480497360229492, "learning_rate": 0.0002, "epoch": 0.029038112522686024, "step": 40}, {"loss": 1.3874, "grad_norm": 0.4114855229854584, "learning_rate": 0.0002, "epoch": 0.036297640653357534, "step": 50}, {"loss": 1.2836, "grad_norm": 0.49482840299606323, "learning_rate": 0.0002, "epoch": 0.043557168784029036, "step": 60}, {"loss": 1.1728, "grad_norm": 0.4536272883415222, "learning_rate": 0.0002, "epoch": 0.050816696914700546, "step": 70}, {"loss": 1.1125, "grad_norm": 0.32328274846076965, "learning_rate": 0.0002, "epoch": 0.05807622504537205, "step": 80}, {"loss": 1.2447, "grad_norm": 0.40990468859672546, "learning_rate": 0.0002, "epoch": 0.06533575317604355, "step": 90}, {"loss": 1.2216, "grad_norm": 0.37273502349853516, "learning_rate": 0.0002, "epoch": 0.07259528130671507, "step": 100}, {"loss": 1.2259, "grad_norm": 0.3903871476650238, "learning_rate": 0.0002, "epoch": 0.07985480943738657, "step": 110}, {"loss": 1.3718, "grad_norm": 0.3272787928581238, "learning_rate": 0.0002, "epoch": 0.08711433756805807, "step": 120}, {"loss": 1.1832, "grad_norm": 0.3622824251651764, "learning_rate": 0.0002, "epoch": 0.09437386569872959, "step": 130}, {"loss": 1.2966, "grad_norm": 0.3503916561603546, "learning_rate": 0.0002, "epoch": 0.10163339382940109, "step": 140}, {"loss": 1.0913, "grad_norm": 0.32787832617759705, "learning_rate": 0.0002, "epoch": 0.1088929219600726, "step": 150}, {"loss": 1.2143, "grad_norm": 0.5822657942771912, "learning_rate": 0.0002, "epoch": 0.1161524500907441, "step": 160}, {"loss": 1.2091, "grad_norm": 0.28028249740600586, "learning_rate": 0.0002, "epoch": 0.12341197822141561, "step": 170}, {"loss": 1.1731, "grad_norm": 0.33602750301361084, "learning_rate": 0.0002, "epoch": 0.1306715063520871, "step": 180}, {"loss": 1.127, "grad_norm": 0.29106274247169495, "learning_rate": 0.0002, "epoch": 0.13793103448275862, "step": 190}, {"loss": 1.3261, "grad_norm": 0.38753363490104675, "learning_rate": 0.0002, "epoch": 0.14519056261343014, "step": 200}, {"loss": 1.1536, "grad_norm": 0.361009418964386, "learning_rate": 0.0002, "epoch": 0.15245009074410162, "step": 210}, {"loss": 1.1483, "grad_norm": 0.6743836402893066, "learning_rate": 0.0002, "epoch": 0.15970961887477314, "step": 220}, {"loss": 1.264, "grad_norm": 0.3923613727092743, "learning_rate": 0.0002, "epoch": 0.16696914700544466, "step": 230}, {"loss": 1.0437, "grad_norm": 0.2809699773788452, "learning_rate": 0.0002, "epoch": 0.17422867513611615, "step": 240}, {"loss": 1.1826, "grad_norm": 0.3631494641304016, "learning_rate": 0.0002, "epoch": 0.18148820326678766, "step": 250}, {"loss": 1.1584, "grad_norm": 0.24658171832561493, "learning_rate": 0.0002, "epoch": 0.18874773139745918, "step": 260}, {"loss": 1.0065, "grad_norm": 0.5780664682388306, "learning_rate": 0.0002, "epoch": 0.19600725952813067, "step": 270}, {"loss": 1.1251, "grad_norm": 0.3056720495223999, "learning_rate": 0.0002, "epoch": 0.20326678765880218, "step": 280}, {"loss": 1.1503, "grad_norm": 0.2857084572315216, "learning_rate": 0.0002, "epoch": 0.21052631578947367, "step": 290}, {"loss": 1.1215, "grad_norm": 0.6645345687866211, "learning_rate": 0.0002, "epoch": 0.2177858439201452, "step": 300}, {"loss": 1.127, "grad_norm": 0.5966078639030457, "learning_rate": 0.0002, "epoch": 0.2250453720508167, "step": 310}, {"loss": 1.1974, "grad_norm": 0.40937140583992004, "learning_rate": 0.0002, "epoch": 0.2323049001814882, "step": 320}, {"loss": 1.2385, "grad_norm": 0.5642806887626648, "learning_rate": 0.0002, "epoch": 0.2395644283121597, "step": 330}, {"loss": 1.1703, "grad_norm": 0.2750748097896576, "learning_rate": 0.0002, "epoch": 0.24682395644283123, "step": 340}, {"loss": 1.2598, "grad_norm": 0.34350234270095825, "learning_rate": 0.0002, "epoch": 0.2540834845735027, "step": 350}, {"loss": 1.1942, "grad_norm": 0.6767239570617676, "learning_rate": 0.0002, "epoch": 0.2613430127041742, "step": 360}, {"loss": 1.1436, "grad_norm": 0.31006959080696106, "learning_rate": 0.0002, "epoch": 0.26860254083484575, "step": 370}, {"loss": 1.2084, "grad_norm": 0.3825474679470062, "learning_rate": 0.0002, "epoch": 0.27586206896551724, "step": 380}, {"loss": 1.1523, "grad_norm": 0.30241551995277405, "learning_rate": 0.0002, "epoch": 0.2831215970961887, "step": 390}, {"loss": 1.1298, "grad_norm": 0.2962397336959839, "learning_rate": 0.0002, "epoch": 0.29038112522686027, "step": 400}, {"loss": 1.1299, "grad_norm": 0.2600369155406952, "learning_rate": 0.0002, "epoch": 0.29764065335753176, "step": 410}, {"loss": 1.1366, "grad_norm": 0.3675060272216797, "learning_rate": 0.0002, "epoch": 0.30490018148820325, "step": 420}, {"loss": 1.156, "grad_norm": 0.3429498076438904, "learning_rate": 0.0002, "epoch": 0.3121597096188748, "step": 430}, {"loss": 1.2741, "grad_norm": 0.34311825037002563, "learning_rate": 0.0002, "epoch": 0.3194192377495463, "step": 440}, {"loss": 1.3523, "grad_norm": 0.37872210144996643, "learning_rate": 0.0002, "epoch": 0.32667876588021777, "step": 450}, {"loss": 1.1365, "grad_norm": 0.33271121978759766, "learning_rate": 0.0002, "epoch": 0.3339382940108893, "step": 460}, {"loss": 1.1572, "grad_norm": 0.34605276584625244, "learning_rate": 0.0002, "epoch": 0.3411978221415608, "step": 470}, {"loss": 1.2755, "grad_norm": 0.41050562262535095, "learning_rate": 0.0002, "epoch": 0.3484573502722323, "step": 480}, {"loss": 1.075, "grad_norm": 0.2066836953163147, "learning_rate": 0.0002, "epoch": 0.35571687840290384, "step": 490}, {"loss": 1.2072, "grad_norm": 0.2859014868736267, "learning_rate": 0.0002, "epoch": 0.3629764065335753, "step": 500}, {"loss": 1.1435, "grad_norm": 0.28763777017593384, "learning_rate": 0.0002, "epoch": 0.3702359346642468, "step": 510}, {"loss": 1.1341, "grad_norm": 0.2730471193790436, "learning_rate": 0.0002, "epoch": 0.37749546279491836, "step": 520}, {"loss": 1.348, "grad_norm": 0.3968936800956726, "learning_rate": 0.0002, "epoch": 0.38475499092558985, "step": 530}, {"loss": 1.2363, "grad_norm": 0.3624701201915741, "learning_rate": 0.0002, "epoch": 0.39201451905626133, "step": 540}, {"loss": 1.1206, "grad_norm": 0.3303608298301697, "learning_rate": 0.0002, "epoch": 0.3992740471869328, "step": 550}, {"loss": 1.203, "grad_norm": 0.33507466316223145, "learning_rate": 0.0002, "epoch": 0.40653357531760437, "step": 560}, {"loss": 1.2294, "grad_norm": 0.3297670781612396, "learning_rate": 0.0002, "epoch": 0.41379310344827586, "step": 570}, {"loss": 1.2, "grad_norm": 0.32334890961647034, "learning_rate": 0.0002, "epoch": 0.42105263157894735, "step": 580}, {"loss": 1.112, "grad_norm": 0.30281195044517517, "learning_rate": 0.0002, "epoch": 0.4283121597096189, "step": 590}, {"loss": 1.1352, "grad_norm": 0.5900027751922607, "learning_rate": 0.0002, "epoch": 0.4355716878402904, "step": 600}, {"loss": 1.1575, "grad_norm": 0.28437477350234985, "learning_rate": 0.0002, "epoch": 0.44283121597096187, "step": 610}, {"loss": 1.2009, "grad_norm": 0.39601704478263855, "learning_rate": 0.0002, "epoch": 0.4500907441016334, "step": 620}, {"loss": 1.2302, "grad_norm": 0.41971510648727417, "learning_rate": 0.0002, "epoch": 0.4573502722323049, "step": 630}, {"loss": 1.236, "grad_norm": 0.33814409375190735, "learning_rate": 0.0002, "epoch": 0.4646098003629764, "step": 640}, {"loss": 1.2471, "grad_norm": 0.575718104839325, "learning_rate": 0.0002, "epoch": 0.47186932849364793, "step": 650}, {"loss": 1.1012, "grad_norm": 0.37927401065826416, "learning_rate": 0.0002, "epoch": 0.4791288566243194, "step": 660}, {"loss": 1.1552, "grad_norm": 0.3224332630634308, "learning_rate": 0.0002, "epoch": 0.4863883847549909, "step": 670}, {"loss": 1.1354, "grad_norm": 0.32683515548706055, "learning_rate": 0.0002, "epoch": 0.49364791288566245, "step": 680}, {"loss": 1.207, "grad_norm": 0.4316163659095764, "learning_rate": 0.0002, "epoch": 0.5009074410163339, "step": 690}, {"loss": 1.2116, "grad_norm": 0.342602401971817, "learning_rate": 0.0002, "epoch": 0.5081669691470054, "step": 700}, {"loss": 1.0433, "grad_norm": 0.2794898748397827, "learning_rate": 0.0002, "epoch": 0.515426497277677, "step": 710}, {"loss": 1.2395, "grad_norm": 0.3322339951992035, "learning_rate": 0.0002, "epoch": 0.5226860254083484, "step": 720}, {"loss": 1.0762, "grad_norm": 0.3088509142398834, "learning_rate": 0.0002, "epoch": 0.52994555353902, "step": 730}, {"loss": 1.1495, "grad_norm": 0.24444378912448883, "learning_rate": 0.0002, "epoch": 0.5372050816696915, "step": 740}, {"loss": 1.2176, "grad_norm": 0.3483171761035919, "learning_rate": 0.0002, "epoch": 0.5444646098003629, "step": 750}, {"loss": 1.2248, "grad_norm": 0.3425690531730652, "learning_rate": 0.0002, "epoch": 0.5517241379310345, "step": 760}, {"loss": 1.1127, "grad_norm": 0.31841927766799927, "learning_rate": 0.0002, "epoch": 0.558983666061706, "step": 770}, {"loss": 1.1841, "grad_norm": 0.39423868060112, "learning_rate": 0.0002, "epoch": 0.5662431941923775, "step": 780}, {"loss": 1.0583, "grad_norm": 0.30328479409217834, "learning_rate": 0.0002, "epoch": 0.573502722323049, "step": 790}, {"loss": 1.0718, "grad_norm": 0.24475938081741333, "learning_rate": 0.0002, "epoch": 0.5807622504537205, "step": 800}, {"loss": 1.2006, "grad_norm": 0.37132805585861206, "learning_rate": 0.0002, "epoch": 0.588021778584392, "step": 810}, {"loss": 1.1544, "grad_norm": 0.32195979356765747, "learning_rate": 0.0002, "epoch": 0.5952813067150635, "step": 820}, {"loss": 0.9937, "grad_norm": 0.2848738729953766, "learning_rate": 0.0002, "epoch": 0.6025408348457351, "step": 830}, {"loss": 1.0937, "grad_norm": 0.28015264868736267, "learning_rate": 0.0002, "epoch": 0.6098003629764065, "step": 840}, {"loss": 1.1234, "grad_norm": 0.37796008586883545, "learning_rate": 0.0002, "epoch": 0.617059891107078, "step": 850}, {"loss": 1.1455, "grad_norm": 0.39311841130256653, "learning_rate": 0.0002, "epoch": 0.6243194192377496, "step": 860}, {"loss": 1.1066, "grad_norm": 0.2761685252189636, "learning_rate": 0.0002, "epoch": 0.631578947368421, "step": 870}, {"loss": 1.0859, "grad_norm": 0.3826720118522644, "learning_rate": 0.0002, "epoch": 0.6388384754990926, "step": 880}, {"loss": 1.1182, "grad_norm": 0.30076679587364197, "learning_rate": 0.0002, "epoch": 0.6460980036297641, "step": 890}, {"loss": 1.0927, "grad_norm": 0.21997687220573425, "learning_rate": 0.0002, "epoch": 0.6533575317604355, "step": 900}, {"loss": 1.1603, "grad_norm": 0.32593777775764465, "learning_rate": 0.0002, "epoch": 0.6606170598911071, "step": 910}, {"loss": 1.1578, "grad_norm": 0.30347898602485657, "learning_rate": 0.0002, "epoch": 0.6678765880217786, "step": 920}, {"loss": 1.1642, "grad_norm": 0.44173774123191833, "learning_rate": 0.0002, "epoch": 0.6751361161524501, "step": 930}, {"loss": 1.1832, "grad_norm": 0.2507467269897461, "learning_rate": 0.0002, "epoch": 0.6823956442831216, "step": 940}, {"loss": 1.3001, "grad_norm": 0.29463833570480347, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 950}, {"loss": 1.1651, "grad_norm": 0.9363154172897339, "learning_rate": 0.0002, "epoch": 0.6969147005444646, "step": 960}, {"loss": 1.158, "grad_norm": 0.3236212134361267, "learning_rate": 0.0002, "epoch": 0.7041742286751361, "step": 970}, {"loss": 1.1078, "grad_norm": 0.3123254179954529, "learning_rate": 0.0002, "epoch": 0.7114337568058077, "step": 980}, {"loss": 1.1578, "grad_norm": 0.3395805060863495, "learning_rate": 0.0002, "epoch": 0.7186932849364791, "step": 990}, {"loss": 1.0754, "grad_norm": 0.3240964412689209, "learning_rate": 0.0002, "epoch": 0.7259528130671506, "step": 1000}, {"loss": 1.0313, "grad_norm": 0.31902948021888733, "learning_rate": 0.0002, "epoch": 0.7332123411978222, "step": 1010}, {"loss": 1.1664, "grad_norm": 0.4848408102989197, "learning_rate": 0.0002, "epoch": 0.7404718693284936, "step": 1020}, {"loss": 1.0838, "grad_norm": 0.33006033301353455, "learning_rate": 0.0002, "epoch": 0.7477313974591652, "step": 1030}, {"loss": 1.0979, "grad_norm": 0.2928730547428131, "learning_rate": 0.0002, "epoch": 0.7549909255898367, "step": 1040}, {"loss": 1.1351, "grad_norm": 0.3529164791107178, "learning_rate": 0.0002, "epoch": 0.7622504537205081, "step": 1050}, {"loss": 1.1274, "grad_norm": 0.2736213803291321, "learning_rate": 0.0002, "epoch": 0.7695099818511797, "step": 1060}, {"loss": 1.1169, "grad_norm": 0.7200686931610107, "learning_rate": 0.0002, "epoch": 0.7767695099818511, "step": 1070}, {"loss": 1.1171, "grad_norm": 0.33396708965301514, "learning_rate": 0.0002, "epoch": 0.7840290381125227, "step": 1080}, {"loss": 1.2447, "grad_norm": 1.5760449171066284, "learning_rate": 0.0002, "epoch": 0.7912885662431942, "step": 1090}, {"loss": 1.1126, "grad_norm": 0.28138381242752075, "learning_rate": 0.0002, "epoch": 0.7985480943738656, "step": 1100}, {"loss": 1.2012, "grad_norm": 0.2597472369670868, "learning_rate": 0.0002, "epoch": 0.8058076225045372, "step": 1110}, {"loss": 1.1177, "grad_norm": 0.3305445611476898, "learning_rate": 0.0002, "epoch": 0.8130671506352087, "step": 1120}, {"loss": 1.0849, "grad_norm": 0.3934599459171295, "learning_rate": 0.0002, "epoch": 0.8203266787658802, "step": 1130}, {"loss": 1.0889, "grad_norm": 0.3472191393375397, "learning_rate": 0.0002, "epoch": 0.8275862068965517, "step": 1140}, {"loss": 1.265, "grad_norm": 0.2857365906238556, "learning_rate": 0.0002, "epoch": 0.8348457350272233, "step": 1150}, {"loss": 1.03, "grad_norm": 0.3207702934741974, "learning_rate": 0.0002, "epoch": 0.8421052631578947, "step": 1160}, {"loss": 1.1669, "grad_norm": 0.3176484704017639, "learning_rate": 0.0002, "epoch": 0.8493647912885662, "step": 1170}, {"loss": 1.1386, "grad_norm": 0.40685558319091797, "learning_rate": 0.0002, "epoch": 0.8566243194192378, "step": 1180}, {"loss": 1.1383, "grad_norm": 0.31125199794769287, "learning_rate": 0.0002, "epoch": 0.8638838475499092, "step": 1190}, {"loss": 1.1512, "grad_norm": 0.7361181378364563, "learning_rate": 0.0002, "epoch": 0.8711433756805808, "step": 1200}, {"loss": 0.9983, "grad_norm": 0.33699527382850647, "learning_rate": 0.0002, "epoch": 0.8784029038112523, "step": 1210}, {"loss": 1.11, "grad_norm": 0.3315220773220062, "learning_rate": 0.0002, "epoch": 0.8856624319419237, "step": 1220}, {"loss": 1.0266, "grad_norm": 0.6256054043769836, "learning_rate": 0.0002, "epoch": 0.8929219600725953, "step": 1230}, {"loss": 1.1738, "grad_norm": 0.3692137897014618, "learning_rate": 0.0002, "epoch": 0.9001814882032668, "step": 1240}, {"loss": 1.1112, "grad_norm": 0.3538484573364258, "learning_rate": 0.0002, "epoch": 0.9074410163339383, "step": 1250}, {"loss": 1.0878, "grad_norm": 0.27863040566444397, "learning_rate": 0.0002, "epoch": 0.9147005444646098, "step": 1260}, {"loss": 1.0826, "grad_norm": 0.3322528302669525, "learning_rate": 0.0002, "epoch": 0.9219600725952813, "step": 1270}, {"loss": 1.0158, "grad_norm": 0.7553173303604126, "learning_rate": 0.0002, "epoch": 0.9292196007259528, "step": 1280}, {"loss": 1.198, "grad_norm": 0.4856191575527191, "learning_rate": 0.0002, "epoch": 0.9364791288566243, "step": 1290}, {"loss": 1.1678, "grad_norm": 0.3668074905872345, "learning_rate": 0.0002, "epoch": 0.9437386569872959, "step": 1300}, {"loss": 1.1809, "grad_norm": 0.29851067066192627, "learning_rate": 0.0002, "epoch": 0.9509981851179673, "step": 1310}, {"loss": 1.1299, "grad_norm": 0.276664674282074, "learning_rate": 0.0002, "epoch": 0.9582577132486388, "step": 1320}, {"loss": 1.0597, "grad_norm": 0.2941018342971802, "learning_rate": 0.0002, "epoch": 0.9655172413793104, "step": 1330}, {"loss": 1.1731, "grad_norm": 0.3505859076976776, "learning_rate": 0.0002, "epoch": 0.9727767695099818, "step": 1340}, {"loss": 1.1809, "grad_norm": 0.3067687451839447, "learning_rate": 0.0002, "epoch": 0.9800362976406534, "step": 1350}, {"loss": 1.1511, "grad_norm": 0.27151066064834595, "learning_rate": 0.0002, "epoch": 0.9872958257713249, "step": 1360}, {"loss": 1.1919, "grad_norm": 0.36370083689689636, "learning_rate": 0.0002, "epoch": 0.9945553539019963, "step": 1370}, {"eval_loss": 1.1381088495254517, "eval_runtime": 96.0848, "eval_samples_per_second": 4.538, "eval_steps_per_second": 0.572, "epoch": 0.9996370235934664, "step": 1377}, {"loss": 1.0082, "grad_norm": 0.27980583906173706, "learning_rate": 0.0002, "epoch": 1.0018148820326678, "step": 1380}, {"loss": 1.0565, "grad_norm": 0.26713913679122925, "learning_rate": 0.0002, "epoch": 1.0090744101633393, "step": 1390}, {"loss": 1.0241, "grad_norm": 0.3089541494846344, "learning_rate": 0.0002, "epoch": 1.0163339382940109, "step": 1400}, {"loss": 0.9649, "grad_norm": 0.4188242256641388, "learning_rate": 0.0002, "epoch": 1.0235934664246824, "step": 1410}, {"loss": 0.9821, "grad_norm": 0.5246463418006897, "learning_rate": 0.0002, "epoch": 1.030852994555354, "step": 1420}, {"loss": 1.0143, "grad_norm": 0.2728777825832367, "learning_rate": 0.0002, "epoch": 1.0381125226860255, "step": 1430}, {"loss": 1.075, "grad_norm": 0.38167616724967957, "learning_rate": 0.0002, "epoch": 1.0453720508166968, "step": 1440}, {"loss": 1.0229, "grad_norm": 0.4439380168914795, "learning_rate": 0.0002, "epoch": 1.0526315789473684, "step": 1450}, {"loss": 1.0451, "grad_norm": 0.30954182147979736, "learning_rate": 0.0002, "epoch": 1.05989110707804, "step": 1460}, {"loss": 1.0762, "grad_norm": 0.4022280275821686, "learning_rate": 0.0002, "epoch": 1.0671506352087115, "step": 1470}, {"loss": 1.0754, "grad_norm": 0.7390811443328857, "learning_rate": 0.0002, "epoch": 1.074410163339383, "step": 1480}, {"loss": 1.1378, "grad_norm": 0.3885486423969269, "learning_rate": 0.0002, "epoch": 1.0816696914700545, "step": 1490}, {"loss": 1.005, "grad_norm": 0.5275560617446899, "learning_rate": 0.0002, "epoch": 1.0889292196007259, "step": 1500}, {"loss": 1.0147, "grad_norm": 0.35112282633781433, "learning_rate": 0.0002, "epoch": 1.0961887477313974, "step": 1510}, {"loss": 1.0331, "grad_norm": 0.33714351058006287, "learning_rate": 0.0002, "epoch": 1.103448275862069, "step": 1520}, {"loss": 0.8723, "grad_norm": 0.31221693754196167, "learning_rate": 0.0002, "epoch": 1.1107078039927405, "step": 1530}, {"loss": 1.0701, "grad_norm": 0.27549654245376587, "learning_rate": 0.0002, "epoch": 1.117967332123412, "step": 1540}, {"loss": 1.0114, "grad_norm": 0.8465521335601807, "learning_rate": 0.0002, "epoch": 1.1252268602540836, "step": 1550}, {"loss": 0.9643, "grad_norm": 0.36125949025154114, "learning_rate": 0.0002, "epoch": 1.132486388384755, "step": 1560}, {"loss": 1.0744, "grad_norm": 0.37420371174812317, "learning_rate": 0.0002, "epoch": 1.1397459165154264, "step": 1570}, {"loss": 1.165, "grad_norm": 0.3294760584831238, "learning_rate": 0.0002, "epoch": 1.147005444646098, "step": 1580}, {"loss": 0.9207, "grad_norm": 0.3881238102912903, "learning_rate": 0.0002, "epoch": 1.1542649727767695, "step": 1590}, {"loss": 1.0983, "grad_norm": 0.4766491651535034, "learning_rate": 0.0002, "epoch": 1.161524500907441, "step": 1600}, {"loss": 1.0092, "grad_norm": 0.395530104637146, "learning_rate": 0.0002, "epoch": 1.1687840290381124, "step": 1610}, {"loss": 0.9895, "grad_norm": 0.3297106623649597, "learning_rate": 0.0002, "epoch": 1.176043557168784, "step": 1620}, {"loss": 1.1121, "grad_norm": 0.39528271555900574, "learning_rate": 0.0002, "epoch": 1.1833030852994555, "step": 1630}, {"loss": 0.9202, "grad_norm": 0.3370221257209778, "learning_rate": 0.0002, "epoch": 1.190562613430127, "step": 1640}, {"loss": 1.102, "grad_norm": 0.31922030448913574, "learning_rate": 0.0002, "epoch": 1.1978221415607986, "step": 1650}, {"loss": 1.0565, "grad_norm": 0.6142027378082275, "learning_rate": 0.0002, "epoch": 1.2050816696914701, "step": 1660}, {"loss": 0.9541, "grad_norm": 0.44769710302352905, "learning_rate": 0.0002, "epoch": 1.2123411978221417, "step": 1670}, {"loss": 1.2501, "grad_norm": 0.41233646869659424, "learning_rate": 0.0002, "epoch": 1.219600725952813, "step": 1680}, {"loss": 1.0147, "grad_norm": 0.2928866147994995, "learning_rate": 0.0002, "epoch": 1.2268602540834845, "step": 1690}, {"loss": 0.9074, "grad_norm": 0.36913734674453735, "learning_rate": 0.0002, "epoch": 1.234119782214156, "step": 1700}, {"loss": 1.0847, "grad_norm": 0.5281891226768494, "learning_rate": 0.0002, "epoch": 1.2413793103448276, "step": 1710}, {"loss": 0.9482, "grad_norm": 0.3374697268009186, "learning_rate": 0.0002, "epoch": 1.2486388384754992, "step": 1720}, {"loss": 0.9826, "grad_norm": 0.3802020847797394, "learning_rate": 0.0002, "epoch": 1.2558983666061705, "step": 1730}, {"loss": 1.0191, "grad_norm": 0.38048651814460754, "learning_rate": 0.0002, "epoch": 1.263157894736842, "step": 1740}, {"loss": 1.0142, "grad_norm": 0.6676169633865356, "learning_rate": 0.0002, "epoch": 1.2704174228675136, "step": 1750}, {"loss": 0.992, "grad_norm": 0.4075961410999298, "learning_rate": 0.0002, "epoch": 1.2776769509981851, "step": 1760}, {"loss": 1.0301, "grad_norm": 0.4374721348285675, "learning_rate": 0.0002, "epoch": 1.2849364791288567, "step": 1770}, {"loss": 1.0601, "grad_norm": 0.4638824164867401, "learning_rate": 0.0002, "epoch": 1.2921960072595282, "step": 1780}, {"loss": 0.9593, "grad_norm": 0.38631564378738403, "learning_rate": 0.0002, "epoch": 1.2994555353901998, "step": 1790}, {"loss": 1.0042, "grad_norm": 0.35873809456825256, "learning_rate": 0.0002, "epoch": 1.306715063520871, "step": 1800}, {"loss": 0.9835, "grad_norm": 0.33602237701416016, "learning_rate": 0.0002, "epoch": 1.3139745916515426, "step": 1810}, {"loss": 1.1032, "grad_norm": 0.46696463227272034, "learning_rate": 0.0002, "epoch": 1.3212341197822142, "step": 1820}, {"loss": 1.0123, "grad_norm": 0.368958979845047, "learning_rate": 0.0002, "epoch": 1.3284936479128857, "step": 1830}, {"loss": 1.0456, "grad_norm": 0.3160957396030426, "learning_rate": 0.0002, "epoch": 1.335753176043557, "step": 1840}, {"loss": 1.0264, "grad_norm": 0.4511511027812958, "learning_rate": 0.0002, "epoch": 1.3430127041742286, "step": 1850}, {"loss": 1.0179, "grad_norm": 0.6769845485687256, "learning_rate": 0.0002, "epoch": 1.3502722323049001, "step": 1860}, {"loss": 1.011, "grad_norm": 0.3749309480190277, "learning_rate": 0.0002, "epoch": 1.3575317604355717, "step": 1870}, {"loss": 1.0223, "grad_norm": 0.553798496723175, "learning_rate": 0.0002, "epoch": 1.3647912885662432, "step": 1880}, {"loss": 0.9634, "grad_norm": 0.3538985550403595, "learning_rate": 0.0002, "epoch": 1.3720508166969148, "step": 1890}, {"loss": 1.0275, "grad_norm": 0.5501534938812256, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1900}, {"loss": 1.0461, "grad_norm": 0.4432051181793213, "learning_rate": 0.0002, "epoch": 1.3865698729582578, "step": 1910}, {"loss": 0.9082, "grad_norm": 0.41755786538124084, "learning_rate": 0.0002, "epoch": 1.3938294010889292, "step": 1920}, {"loss": 1.0858, "grad_norm": 0.5626114010810852, "learning_rate": 0.0002, "epoch": 1.4010889292196007, "step": 1930}, {"loss": 1.0687, "grad_norm": 0.44038185477256775, "learning_rate": 0.0002, "epoch": 1.4083484573502723, "step": 1940}, {"loss": 0.9454, "grad_norm": 0.3397001624107361, "learning_rate": 0.0002, "epoch": 1.4156079854809438, "step": 1950}, {"loss": 0.943, "grad_norm": 0.4325368106365204, "learning_rate": 0.0002, "epoch": 1.4228675136116151, "step": 1960}, {"loss": 1.0488, "grad_norm": 0.3900907039642334, "learning_rate": 0.0002, "epoch": 1.4301270417422867, "step": 1970}, {"loss": 0.9699, "grad_norm": 0.369612455368042, "learning_rate": 0.0002, "epoch": 1.4373865698729582, "step": 1980}, {"loss": 1.0609, "grad_norm": 0.4389338791370392, "learning_rate": 0.0002, "epoch": 1.4446460980036298, "step": 1990}, {"loss": 1.042, "grad_norm": 1.694450855255127, "learning_rate": 0.0002, "epoch": 1.4519056261343013, "step": 2000}, {"loss": 0.997, "grad_norm": 0.516957700252533, "learning_rate": 0.0002, "epoch": 1.4591651542649728, "step": 2010}, {"loss": 1.0979, "grad_norm": 0.45515501499176025, "learning_rate": 0.0002, "epoch": 1.4664246823956444, "step": 2020}, {"loss": 1.0188, "grad_norm": 0.4153139591217041, "learning_rate": 0.0002, "epoch": 1.4736842105263157, "step": 2030}, {"loss": 1.1521, "grad_norm": 0.44353052973747253, "learning_rate": 0.0002, "epoch": 1.4809437386569873, "step": 2040}, {"loss": 0.9653, "grad_norm": 0.570554256439209, "learning_rate": 0.0002, "epoch": 1.4882032667876588, "step": 2050}, {"loss": 1.0352, "grad_norm": 0.5742740035057068, "learning_rate": 0.0002, "epoch": 1.4954627949183303, "step": 2060}, {"loss": 1.0963, "grad_norm": 0.5890517830848694, "learning_rate": 0.0002, "epoch": 1.5027223230490017, "step": 2070}, {"loss": 1.0639, "grad_norm": 0.4162650406360626, "learning_rate": 0.0002, "epoch": 1.5099818511796732, "step": 2080}, {"loss": 1.071, "grad_norm": 0.7334717512130737, "learning_rate": 0.0002, "epoch": 1.5172413793103448, "step": 2090}, {"loss": 0.8957, "grad_norm": 0.2963249683380127, "learning_rate": 0.0002, "epoch": 1.5245009074410163, "step": 2100}, {"loss": 1.0446, "grad_norm": 0.30676454305648804, "learning_rate": 0.0002, "epoch": 1.5317604355716878, "step": 2110}, {"loss": 0.9967, "grad_norm": 0.35984641313552856, "learning_rate": 0.0002, "epoch": 1.5390199637023594, "step": 2120}, {"loss": 1.0352, "grad_norm": 0.3384549617767334, "learning_rate": 0.0002, "epoch": 1.546279491833031, "step": 2130}, {"loss": 0.9874, "grad_norm": 0.4725518226623535, "learning_rate": 0.0002, "epoch": 1.5535390199637025, "step": 2140}, {"loss": 1.1166, "grad_norm": 0.3252685070037842, "learning_rate": 0.0002, "epoch": 1.560798548094374, "step": 2150}, {"loss": 0.9127, "grad_norm": 0.45043081045150757, "learning_rate": 0.0002, "epoch": 1.5680580762250453, "step": 2160}, {"loss": 1.0767, "grad_norm": 0.374208003282547, "learning_rate": 0.0002, "epoch": 1.5753176043557169, "step": 2170}, {"loss": 1.0255, "grad_norm": 0.5118404030799866, "learning_rate": 0.0002, "epoch": 1.5825771324863884, "step": 2180}, {"loss": 1.0062, "grad_norm": 0.482170969247818, "learning_rate": 0.0002, "epoch": 1.5898366606170597, "step": 2190}, {"loss": 1.0872, "grad_norm": 0.5337533950805664, "learning_rate": 0.0002, "epoch": 1.5970961887477313, "step": 2200}, {"loss": 1.0405, "grad_norm": 0.5195064544677734, "learning_rate": 0.0002, "epoch": 1.6043557168784028, "step": 2210}, {"loss": 1.0454, "grad_norm": 0.30807098746299744, "learning_rate": 0.0002, "epoch": 1.6116152450090744, "step": 2220}, {"loss": 1.0293, "grad_norm": 0.3962925672531128, "learning_rate": 0.0002, "epoch": 1.618874773139746, "step": 2230}, {"loss": 1.0137, "grad_norm": 0.7636962532997131, "learning_rate": 0.0002, "epoch": 1.6261343012704175, "step": 2240}, {"loss": 0.991, "grad_norm": 0.32380592823028564, "learning_rate": 0.0002, "epoch": 1.633393829401089, "step": 2250}, {"loss": 0.9471, "grad_norm": 0.5767741799354553, "learning_rate": 0.0002, "epoch": 1.6406533575317606, "step": 2260}, {"loss": 0.9524, "grad_norm": 0.39964812994003296, "learning_rate": 0.0002, "epoch": 1.647912885662432, "step": 2270}, {"loss": 0.9866, "grad_norm": 0.622629702091217, "learning_rate": 0.0002, "epoch": 1.6551724137931034, "step": 2280}, {"loss": 0.9609, "grad_norm": 0.40202152729034424, "learning_rate": 0.0002, "epoch": 1.662431941923775, "step": 2290}, {"loss": 0.9859, "grad_norm": 0.4467349052429199, "learning_rate": 0.0002, "epoch": 1.6696914700544465, "step": 2300}, {"loss": 1.0312, "grad_norm": 0.5026949048042297, "learning_rate": 0.0002, "epoch": 1.6769509981851178, "step": 2310}, {"loss": 1.0776, "grad_norm": 0.43754541873931885, "learning_rate": 0.0002, "epoch": 1.6842105263157894, "step": 2320}, {"loss": 0.9786, "grad_norm": 0.42869430780410767, "learning_rate": 0.0002, "epoch": 1.691470054446461, "step": 2330}, {"loss": 1.0747, "grad_norm": 0.4192679524421692, "learning_rate": 0.0002, "epoch": 1.6987295825771325, "step": 2340}, {"loss": 1.0054, "grad_norm": 0.3243155777454376, "learning_rate": 0.0002, "epoch": 1.705989110707804, "step": 2350}, {"loss": 0.9165, "grad_norm": 1.0514075756072998, "learning_rate": 0.0002, "epoch": 1.7132486388384756, "step": 2360}, {"loss": 1.1353, "grad_norm": 0.4719122648239136, "learning_rate": 0.0002, "epoch": 1.720508166969147, "step": 2370}, {"loss": 0.9913, "grad_norm": 0.3846144676208496, "learning_rate": 0.0002, "epoch": 1.7277676950998186, "step": 2380}, {"loss": 0.9935, "grad_norm": 0.7266581058502197, "learning_rate": 0.0002, "epoch": 1.73502722323049, "step": 2390}, {"loss": 0.9509, "grad_norm": 0.6183241605758667, "learning_rate": 0.0002, "epoch": 1.7422867513611615, "step": 2400}, {"loss": 0.9713, "grad_norm": 0.3658260405063629, "learning_rate": 0.0002, "epoch": 1.749546279491833, "step": 2410}, {"loss": 1.0033, "grad_norm": 0.6036322712898254, "learning_rate": 0.0002, "epoch": 1.7568058076225044, "step": 2420}, {"loss": 1.0526, "grad_norm": 0.7872936129570007, "learning_rate": 0.0002, "epoch": 1.764065335753176, "step": 2430}, {"loss": 0.99, "grad_norm": 0.35946124792099, "learning_rate": 0.0002, "epoch": 1.7713248638838475, "step": 2440}, {"loss": 0.8845, "grad_norm": 0.3740338981151581, "learning_rate": 0.0002, "epoch": 1.778584392014519, "step": 2450}, {"loss": 0.9333, "grad_norm": 0.6150230169296265, "learning_rate": 0.0002, "epoch": 1.7858439201451906, "step": 2460}, {"loss": 1.0373, "grad_norm": 0.4726075530052185, "learning_rate": 0.0002, "epoch": 1.793103448275862, "step": 2470}, {"loss": 0.974, "grad_norm": 0.31292253732681274, "learning_rate": 0.0002, "epoch": 1.8003629764065336, "step": 2480}, {"loss": 1.0205, "grad_norm": 0.4463104009628296, "learning_rate": 0.0002, "epoch": 1.8076225045372052, "step": 2490}, {"loss": 1.0771, "grad_norm": 0.7848200798034668, "learning_rate": 0.0002, "epoch": 1.8148820326678767, "step": 2500}, {"loss": 0.9508, "grad_norm": 0.5562082529067993, "learning_rate": 0.0002, "epoch": 1.822141560798548, "step": 2510}, {"loss": 0.9343, "grad_norm": 0.39892756938934326, "learning_rate": 0.0002, "epoch": 1.8294010889292196, "step": 2520}, {"loss": 1.0489, "grad_norm": 1.2923320531845093, "learning_rate": 0.0002, "epoch": 1.8366606170598911, "step": 2530}, {"loss": 1.1203, "grad_norm": 0.6316490769386292, "learning_rate": 0.0002, "epoch": 1.8439201451905625, "step": 2540}, {"loss": 0.9131, "grad_norm": 0.46100497245788574, "learning_rate": 0.0002, "epoch": 1.851179673321234, "step": 2550}, {"loss": 1.0977, "grad_norm": 0.7902987003326416, "learning_rate": 0.0002, "epoch": 1.8584392014519056, "step": 2560}, {"loss": 0.9702, "grad_norm": 0.4596365690231323, "learning_rate": 0.0002, "epoch": 1.865698729582577, "step": 2570}, {"loss": 0.9322, "grad_norm": 0.6592172384262085, "learning_rate": 0.0002, "epoch": 1.8729582577132486, "step": 2580}, {"loss": 0.9549, "grad_norm": 0.5764662027359009, "learning_rate": 0.0002, "epoch": 1.8802177858439202, "step": 2590}, {"loss": 0.9485, "grad_norm": 0.8421637415885925, "learning_rate": 0.0002, "epoch": 1.8874773139745917, "step": 2600}, {"loss": 1.012, "grad_norm": 0.5635305047035217, "learning_rate": 0.0002, "epoch": 1.8947368421052633, "step": 2610}, {"loss": 0.8907, "grad_norm": 0.46231237053871155, "learning_rate": 0.0002, "epoch": 1.9019963702359348, "step": 2620}, {"loss": 0.9543, "grad_norm": 0.3944607079029083, "learning_rate": 0.0002, "epoch": 1.9092558983666061, "step": 2630}, {"loss": 0.9964, "grad_norm": 0.4753907322883606, "learning_rate": 0.0002, "epoch": 1.9165154264972777, "step": 2640}, {"loss": 1.0217, "grad_norm": 0.4151090979576111, "learning_rate": 0.0002, "epoch": 1.9237749546279492, "step": 2650}, {"loss": 1.0397, "grad_norm": 0.6793725490570068, "learning_rate": 0.0002, "epoch": 1.9310344827586206, "step": 2660}, {"loss": 1.0401, "grad_norm": 0.339755117893219, "learning_rate": 0.0002, "epoch": 1.938294010889292, "step": 2670}, {"loss": 1.082, "grad_norm": 0.40789374709129333, "learning_rate": 0.0002, "epoch": 1.9455535390199636, "step": 2680}, {"loss": 0.9973, "grad_norm": 0.3750005066394806, "learning_rate": 0.0002, "epoch": 1.9528130671506352, "step": 2690}, {"loss": 0.9837, "grad_norm": 0.39684441685676575, "learning_rate": 0.0002, "epoch": 1.9600725952813067, "step": 2700}, {"loss": 1.1204, "grad_norm": 0.378287672996521, "learning_rate": 0.0002, "epoch": 1.9673321234119783, "step": 2710}, {"loss": 0.9284, "grad_norm": 0.3668482005596161, "learning_rate": 0.0002, "epoch": 1.9745916515426498, "step": 2720}, {"loss": 1.0512, "grad_norm": 0.49997565150260925, "learning_rate": 0.0002, "epoch": 1.9818511796733214, "step": 2730}, {"loss": 0.9311, "grad_norm": 0.36852124333381653, "learning_rate": 0.0002, "epoch": 1.989110707803993, "step": 2740}, {"loss": 1.0898, "grad_norm": 0.5203380584716797, "learning_rate": 0.0002, "epoch": 1.9963702359346642, "step": 2750}]}
+{"epoch": 2.9996370235934666, "step": 4132, "epoch_duration": 2965.006893157959, "total_accumulated_duration": 12229.340359687805, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4936, "grad_norm": 0.46010470390319824, "learning_rate": 0.0002, "epoch": 0.007259528130671506, "step": 10}, {"loss": 1.8442, "grad_norm": 0.5103179216384888, "learning_rate": 0.0002, "epoch": 0.014519056261343012, "step": 20}, {"loss": 1.559, "grad_norm": 0.6282716989517212, "learning_rate": 0.0002, "epoch": 0.021778584392014518, "step": 30}, {"loss": 1.3618, "grad_norm": 1.2480497360229492, "learning_rate": 0.0002, "epoch": 0.029038112522686024, "step": 40}, {"loss": 1.3874, "grad_norm": 0.4114855229854584, "learning_rate": 0.0002, "epoch": 0.036297640653357534, "step": 50}, {"loss": 1.2836, "grad_norm": 0.49482840299606323, "learning_rate": 0.0002, "epoch": 0.043557168784029036, "step": 60}, {"loss": 1.1728, "grad_norm": 0.4536272883415222, "learning_rate": 0.0002, "epoch": 0.050816696914700546, "step": 70}, {"loss": 1.1125, "grad_norm": 0.32328274846076965, "learning_rate": 0.0002, "epoch": 0.05807622504537205, "step": 80}, {"loss": 1.2447, "grad_norm": 0.40990468859672546, "learning_rate": 0.0002, "epoch": 0.06533575317604355, "step": 90}, {"loss": 1.2216, "grad_norm": 0.37273502349853516, "learning_rate": 0.0002, "epoch": 0.07259528130671507, "step": 100}, {"loss": 1.2259, "grad_norm": 0.3903871476650238, "learning_rate": 0.0002, "epoch": 0.07985480943738657, "step": 110}, {"loss": 1.3718, "grad_norm": 0.3272787928581238, "learning_rate": 0.0002, "epoch": 0.08711433756805807, "step": 120}, {"loss": 1.1832, "grad_norm": 0.3622824251651764, "learning_rate": 0.0002, "epoch": 0.09437386569872959, "step": 130}, {"loss": 1.2966, "grad_norm": 0.3503916561603546, "learning_rate": 0.0002, "epoch": 0.10163339382940109, "step": 140}, {"loss": 1.0913, "grad_norm": 0.32787832617759705, "learning_rate": 0.0002, "epoch": 0.1088929219600726, "step": 150}, {"loss": 1.2143, "grad_norm": 0.5822657942771912, "learning_rate": 0.0002, "epoch": 0.1161524500907441, "step": 160}, {"loss": 1.2091, "grad_norm": 0.28028249740600586, "learning_rate": 0.0002, "epoch": 0.12341197822141561, "step": 170}, {"loss": 1.1731, "grad_norm": 0.33602750301361084, "learning_rate": 0.0002, "epoch": 0.1306715063520871, "step": 180}, {"loss": 1.127, "grad_norm": 0.29106274247169495, "learning_rate": 0.0002, "epoch": 0.13793103448275862, "step": 190}, {"loss": 1.3261, "grad_norm": 0.38753363490104675, "learning_rate": 0.0002, "epoch": 0.14519056261343014, "step": 200}, {"loss": 1.1536, "grad_norm": 0.361009418964386, "learning_rate": 0.0002, "epoch": 0.15245009074410162, "step": 210}, {"loss": 1.1483, "grad_norm": 0.6743836402893066, "learning_rate": 0.0002, "epoch": 0.15970961887477314, "step": 220}, {"loss": 1.264, "grad_norm": 0.3923613727092743, "learning_rate": 0.0002, "epoch": 0.16696914700544466, "step": 230}, {"loss": 1.0437, "grad_norm": 0.2809699773788452, "learning_rate": 0.0002, "epoch": 0.17422867513611615, "step": 240}, {"loss": 1.1826, "grad_norm": 0.3631494641304016, "learning_rate": 0.0002, "epoch": 0.18148820326678766, "step": 250}, {"loss": 1.1584, "grad_norm": 0.24658171832561493, "learning_rate": 0.0002, "epoch": 0.18874773139745918, "step": 260}, {"loss": 1.0065, "grad_norm": 0.5780664682388306, "learning_rate": 0.0002, "epoch": 0.19600725952813067, "step": 270}, {"loss": 1.1251, "grad_norm": 0.3056720495223999, "learning_rate": 0.0002, "epoch": 0.20326678765880218, "step": 280}, {"loss": 1.1503, "grad_norm": 0.2857084572315216, "learning_rate": 0.0002, "epoch": 0.21052631578947367, "step": 290}, {"loss": 1.1215, "grad_norm": 0.6645345687866211, "learning_rate": 0.0002, "epoch": 0.2177858439201452, "step": 300}, {"loss": 1.127, "grad_norm": 0.5966078639030457, "learning_rate": 0.0002, "epoch": 0.2250453720508167, "step": 310}, {"loss": 1.1974, "grad_norm": 0.40937140583992004, "learning_rate": 0.0002, "epoch": 0.2323049001814882, "step": 320}, {"loss": 1.2385, "grad_norm": 0.5642806887626648, "learning_rate": 0.0002, "epoch": 0.2395644283121597, "step": 330}, {"loss": 1.1703, "grad_norm": 0.2750748097896576, "learning_rate": 0.0002, "epoch": 0.24682395644283123, "step": 340}, {"loss": 1.2598, "grad_norm": 0.34350234270095825, "learning_rate": 0.0002, "epoch": 0.2540834845735027, "step": 350}, {"loss": 1.1942, "grad_norm": 0.6767239570617676, "learning_rate": 0.0002, "epoch": 0.2613430127041742, "step": 360}, {"loss": 1.1436, "grad_norm": 0.31006959080696106, "learning_rate": 0.0002, "epoch": 0.26860254083484575, "step": 370}, {"loss": 1.2084, "grad_norm": 0.3825474679470062, "learning_rate": 0.0002, "epoch": 0.27586206896551724, "step": 380}, {"loss": 1.1523, "grad_norm": 0.30241551995277405, "learning_rate": 0.0002, "epoch": 0.2831215970961887, "step": 390}, {"loss": 1.1298, "grad_norm": 0.2962397336959839, "learning_rate": 0.0002, "epoch": 0.29038112522686027, "step": 400}, {"loss": 1.1299, "grad_norm": 0.2600369155406952, "learning_rate": 0.0002, "epoch": 0.29764065335753176, "step": 410}, {"loss": 1.1366, "grad_norm": 0.3675060272216797, "learning_rate": 0.0002, "epoch": 0.30490018148820325, "step": 420}, {"loss": 1.156, "grad_norm": 0.3429498076438904, "learning_rate": 0.0002, "epoch": 0.3121597096188748, "step": 430}, {"loss": 1.2741, "grad_norm": 0.34311825037002563, "learning_rate": 0.0002, "epoch": 0.3194192377495463, "step": 440}, {"loss": 1.3523, "grad_norm": 0.37872210144996643, "learning_rate": 0.0002, "epoch": 0.32667876588021777, "step": 450}, {"loss": 1.1365, "grad_norm": 0.33271121978759766, "learning_rate": 0.0002, "epoch": 0.3339382940108893, "step": 460}, {"loss": 1.1572, "grad_norm": 0.34605276584625244, "learning_rate": 0.0002, "epoch": 0.3411978221415608, "step": 470}, {"loss": 1.2755, "grad_norm": 0.41050562262535095, "learning_rate": 0.0002, "epoch": 0.3484573502722323, "step": 480}, {"loss": 1.075, "grad_norm": 0.2066836953163147, "learning_rate": 0.0002, "epoch": 0.35571687840290384, "step": 490}, {"loss": 1.2072, "grad_norm": 0.2859014868736267, "learning_rate": 0.0002, "epoch": 0.3629764065335753, "step": 500}, {"loss": 1.1435, "grad_norm": 0.28763777017593384, "learning_rate": 0.0002, "epoch": 0.3702359346642468, "step": 510}, {"loss": 1.1341, "grad_norm": 0.2730471193790436, "learning_rate": 0.0002, "epoch": 0.37749546279491836, "step": 520}, {"loss": 1.348, "grad_norm": 0.3968936800956726, "learning_rate": 0.0002, "epoch": 0.38475499092558985, "step": 530}, {"loss": 1.2363, "grad_norm": 0.3624701201915741, "learning_rate": 0.0002, "epoch": 0.39201451905626133, "step": 540}, {"loss": 1.1206, "grad_norm": 0.3303608298301697, "learning_rate": 0.0002, "epoch": 0.3992740471869328, "step": 550}, {"loss": 1.203, "grad_norm": 0.33507466316223145, "learning_rate": 0.0002, "epoch": 0.40653357531760437, "step": 560}, {"loss": 1.2294, "grad_norm": 0.3297670781612396, "learning_rate": 0.0002, "epoch": 0.41379310344827586, "step": 570}, {"loss": 1.2, "grad_norm": 0.32334890961647034, "learning_rate": 0.0002, "epoch": 0.42105263157894735, "step": 580}, {"loss": 1.112, "grad_norm": 0.30281195044517517, "learning_rate": 0.0002, "epoch": 0.4283121597096189, "step": 590}, {"loss": 1.1352, "grad_norm": 0.5900027751922607, "learning_rate": 0.0002, "epoch": 0.4355716878402904, "step": 600}, {"loss": 1.1575, "grad_norm": 0.28437477350234985, "learning_rate": 0.0002, "epoch": 0.44283121597096187, "step": 610}, {"loss": 1.2009, "grad_norm": 0.39601704478263855, "learning_rate": 0.0002, "epoch": 0.4500907441016334, "step": 620}, {"loss": 1.2302, "grad_norm": 0.41971510648727417, "learning_rate": 0.0002, "epoch": 0.4573502722323049, "step": 630}, {"loss": 1.236, "grad_norm": 0.33814409375190735, "learning_rate": 0.0002, "epoch": 0.4646098003629764, "step": 640}, {"loss": 1.2471, "grad_norm": 0.575718104839325, "learning_rate": 0.0002, "epoch": 0.47186932849364793, "step": 650}, {"loss": 1.1012, "grad_norm": 0.37927401065826416, "learning_rate": 0.0002, "epoch": 0.4791288566243194, "step": 660}, {"loss": 1.1552, "grad_norm": 0.3224332630634308, "learning_rate": 0.0002, "epoch": 0.4863883847549909, "step": 670}, {"loss": 1.1354, "grad_norm": 0.32683515548706055, "learning_rate": 0.0002, "epoch": 0.49364791288566245, "step": 680}, {"loss": 1.207, "grad_norm": 0.4316163659095764, "learning_rate": 0.0002, "epoch": 0.5009074410163339, "step": 690}, {"loss": 1.2116, "grad_norm": 0.342602401971817, "learning_rate": 0.0002, "epoch": 0.5081669691470054, "step": 700}, {"loss": 1.0433, "grad_norm": 0.2794898748397827, "learning_rate": 0.0002, "epoch": 0.515426497277677, "step": 710}, {"loss": 1.2395, "grad_norm": 0.3322339951992035, "learning_rate": 0.0002, "epoch": 0.5226860254083484, "step": 720}, {"loss": 1.0762, "grad_norm": 0.3088509142398834, "learning_rate": 0.0002, "epoch": 0.52994555353902, "step": 730}, {"loss": 1.1495, "grad_norm": 0.24444378912448883, "learning_rate": 0.0002, "epoch": 0.5372050816696915, "step": 740}, {"loss": 1.2176, "grad_norm": 0.3483171761035919, "learning_rate": 0.0002, "epoch": 0.5444646098003629, "step": 750}, {"loss": 1.2248, "grad_norm": 0.3425690531730652, "learning_rate": 0.0002, "epoch": 0.5517241379310345, "step": 760}, {"loss": 1.1127, "grad_norm": 0.31841927766799927, "learning_rate": 0.0002, "epoch": 0.558983666061706, "step": 770}, {"loss": 1.1841, "grad_norm": 0.39423868060112, "learning_rate": 0.0002, "epoch": 0.5662431941923775, "step": 780}, {"loss": 1.0583, "grad_norm": 0.30328479409217834, "learning_rate": 0.0002, "epoch": 0.573502722323049, "step": 790}, {"loss": 1.0718, "grad_norm": 0.24475938081741333, "learning_rate": 0.0002, "epoch": 0.5807622504537205, "step": 800}, {"loss": 1.2006, "grad_norm": 0.37132805585861206, "learning_rate": 0.0002, "epoch": 0.588021778584392, "step": 810}, {"loss": 1.1544, "grad_norm": 0.32195979356765747, "learning_rate": 0.0002, "epoch": 0.5952813067150635, "step": 820}, {"loss": 0.9937, "grad_norm": 0.2848738729953766, "learning_rate": 0.0002, "epoch": 0.6025408348457351, "step": 830}, {"loss": 1.0937, "grad_norm": 0.28015264868736267, "learning_rate": 0.0002, "epoch": 0.6098003629764065, "step": 840}, {"loss": 1.1234, "grad_norm": 0.37796008586883545, "learning_rate": 0.0002, "epoch": 0.617059891107078, "step": 850}, {"loss": 1.1455, "grad_norm": 0.39311841130256653, "learning_rate": 0.0002, "epoch": 0.6243194192377496, "step": 860}, {"loss": 1.1066, "grad_norm": 0.2761685252189636, "learning_rate": 0.0002, "epoch": 0.631578947368421, "step": 870}, {"loss": 1.0859, "grad_norm": 0.3826720118522644, "learning_rate": 0.0002, "epoch": 0.6388384754990926, "step": 880}, {"loss": 1.1182, "grad_norm": 0.30076679587364197, "learning_rate": 0.0002, "epoch": 0.6460980036297641, "step": 890}, {"loss": 1.0927, "grad_norm": 0.21997687220573425, "learning_rate": 0.0002, "epoch": 0.6533575317604355, "step": 900}, {"loss": 1.1603, "grad_norm": 0.32593777775764465, "learning_rate": 0.0002, "epoch": 0.6606170598911071, "step": 910}, {"loss": 1.1578, "grad_norm": 0.30347898602485657, "learning_rate": 0.0002, "epoch": 0.6678765880217786, "step": 920}, {"loss": 1.1642, "grad_norm": 0.44173774123191833, "learning_rate": 0.0002, "epoch": 0.6751361161524501, "step": 930}, {"loss": 1.1832, "grad_norm": 0.2507467269897461, "learning_rate": 0.0002, "epoch": 0.6823956442831216, "step": 940}, {"loss": 1.3001, "grad_norm": 0.29463833570480347, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 950}, {"loss": 1.1651, "grad_norm": 0.9363154172897339, "learning_rate": 0.0002, "epoch": 0.6969147005444646, "step": 960}, {"loss": 1.158, "grad_norm": 0.3236212134361267, "learning_rate": 0.0002, "epoch": 0.7041742286751361, "step": 970}, {"loss": 1.1078, "grad_norm": 0.3123254179954529, "learning_rate": 0.0002, "epoch": 0.7114337568058077, "step": 980}, {"loss": 1.1578, "grad_norm": 0.3395805060863495, "learning_rate": 0.0002, "epoch": 0.7186932849364791, "step": 990}, {"loss": 1.0754, "grad_norm": 0.3240964412689209, "learning_rate": 0.0002, "epoch": 0.7259528130671506, "step": 1000}, {"loss": 1.0313, "grad_norm": 0.31902948021888733, "learning_rate": 0.0002, "epoch": 0.7332123411978222, "step": 1010}, {"loss": 1.1664, "grad_norm": 0.4848408102989197, "learning_rate": 0.0002, "epoch": 0.7404718693284936, "step": 1020}, {"loss": 1.0838, "grad_norm": 0.33006033301353455, "learning_rate": 0.0002, "epoch": 0.7477313974591652, "step": 1030}, {"loss": 1.0979, "grad_norm": 0.2928730547428131, "learning_rate": 0.0002, "epoch": 0.7549909255898367, "step": 1040}, {"loss": 1.1351, "grad_norm": 0.3529164791107178, "learning_rate": 0.0002, "epoch": 0.7622504537205081, "step": 1050}, {"loss": 1.1274, "grad_norm": 0.2736213803291321, "learning_rate": 0.0002, "epoch": 0.7695099818511797, "step": 1060}, {"loss": 1.1169, "grad_norm": 0.7200686931610107, "learning_rate": 0.0002, "epoch": 0.7767695099818511, "step": 1070}, {"loss": 1.1171, "grad_norm": 0.33396708965301514, "learning_rate": 0.0002, "epoch": 0.7840290381125227, "step": 1080}, {"loss": 1.2447, "grad_norm": 1.5760449171066284, "learning_rate": 0.0002, "epoch": 0.7912885662431942, "step": 1090}, {"loss": 1.1126, "grad_norm": 0.28138381242752075, "learning_rate": 0.0002, "epoch": 0.7985480943738656, "step": 1100}, {"loss": 1.2012, "grad_norm": 0.2597472369670868, "learning_rate": 0.0002, "epoch": 0.8058076225045372, "step": 1110}, {"loss": 1.1177, "grad_norm": 0.3305445611476898, "learning_rate": 0.0002, "epoch": 0.8130671506352087, "step": 1120}, {"loss": 1.0849, "grad_norm": 0.3934599459171295, "learning_rate": 0.0002, "epoch": 0.8203266787658802, "step": 1130}, {"loss": 1.0889, "grad_norm": 0.3472191393375397, "learning_rate": 0.0002, "epoch": 0.8275862068965517, "step": 1140}, {"loss": 1.265, "grad_norm": 0.2857365906238556, "learning_rate": 0.0002, "epoch": 0.8348457350272233, "step": 1150}, {"loss": 1.03, "grad_norm": 0.3207702934741974, "learning_rate": 0.0002, "epoch": 0.8421052631578947, "step": 1160}, {"loss": 1.1669, "grad_norm": 0.3176484704017639, "learning_rate": 0.0002, "epoch": 0.8493647912885662, "step": 1170}, {"loss": 1.1386, "grad_norm": 0.40685558319091797, "learning_rate": 0.0002, "epoch": 0.8566243194192378, "step": 1180}, {"loss": 1.1383, "grad_norm": 0.31125199794769287, "learning_rate": 0.0002, "epoch": 0.8638838475499092, "step": 1190}, {"loss": 1.1512, "grad_norm": 0.7361181378364563, "learning_rate": 0.0002, "epoch": 0.8711433756805808, "step": 1200}, {"loss": 0.9983, "grad_norm": 0.33699527382850647, "learning_rate": 0.0002, "epoch": 0.8784029038112523, "step": 1210}, {"loss": 1.11, "grad_norm": 0.3315220773220062, "learning_rate": 0.0002, "epoch": 0.8856624319419237, "step": 1220}, {"loss": 1.0266, "grad_norm": 0.6256054043769836, "learning_rate": 0.0002, "epoch": 0.8929219600725953, "step": 1230}, {"loss": 1.1738, "grad_norm": 0.3692137897014618, "learning_rate": 0.0002, "epoch": 0.9001814882032668, "step": 1240}, {"loss": 1.1112, "grad_norm": 0.3538484573364258, "learning_rate": 0.0002, "epoch": 0.9074410163339383, "step": 1250}, {"loss": 1.0878, "grad_norm": 0.27863040566444397, "learning_rate": 0.0002, "epoch": 0.9147005444646098, "step": 1260}, {"loss": 1.0826, "grad_norm": 0.3322528302669525, "learning_rate": 0.0002, "epoch": 0.9219600725952813, "step": 1270}, {"loss": 1.0158, "grad_norm": 0.7553173303604126, "learning_rate": 0.0002, "epoch": 0.9292196007259528, "step": 1280}, {"loss": 1.198, "grad_norm": 0.4856191575527191, "learning_rate": 0.0002, "epoch": 0.9364791288566243, "step": 1290}, {"loss": 1.1678, "grad_norm": 0.3668074905872345, "learning_rate": 0.0002, "epoch": 0.9437386569872959, "step": 1300}, {"loss": 1.1809, "grad_norm": 0.29851067066192627, "learning_rate": 0.0002, "epoch": 0.9509981851179673, "step": 1310}, {"loss": 1.1299, "grad_norm": 0.276664674282074, "learning_rate": 0.0002, "epoch": 0.9582577132486388, "step": 1320}, {"loss": 1.0597, "grad_norm": 0.2941018342971802, "learning_rate": 0.0002, "epoch": 0.9655172413793104, "step": 1330}, {"loss": 1.1731, "grad_norm": 0.3505859076976776, "learning_rate": 0.0002, "epoch": 0.9727767695099818, "step": 1340}, {"loss": 1.1809, "grad_norm": 0.3067687451839447, "learning_rate": 0.0002, "epoch": 0.9800362976406534, "step": 1350}, {"loss": 1.1511, "grad_norm": 0.27151066064834595, "learning_rate": 0.0002, "epoch": 0.9872958257713249, "step": 1360}, {"loss": 1.1919, "grad_norm": 0.36370083689689636, "learning_rate": 0.0002, "epoch": 0.9945553539019963, "step": 1370}, {"eval_loss": 1.1381088495254517, "eval_runtime": 96.0848, "eval_samples_per_second": 4.538, "eval_steps_per_second": 0.572, "epoch": 0.9996370235934664, "step": 1377}, {"loss": 1.0082, "grad_norm": 0.27980583906173706, "learning_rate": 0.0002, "epoch": 1.0018148820326678, "step": 1380}, {"loss": 1.0565, "grad_norm": 0.26713913679122925, "learning_rate": 0.0002, "epoch": 1.0090744101633393, "step": 1390}, {"loss": 1.0241, "grad_norm": 0.3089541494846344, "learning_rate": 0.0002, "epoch": 1.0163339382940109, "step": 1400}, {"loss": 0.9649, "grad_norm": 0.4188242256641388, "learning_rate": 0.0002, "epoch": 1.0235934664246824, "step": 1410}, {"loss": 0.9821, "grad_norm": 0.5246463418006897, "learning_rate": 0.0002, "epoch": 1.030852994555354, "step": 1420}, {"loss": 1.0143, "grad_norm": 0.2728777825832367, "learning_rate": 0.0002, "epoch": 1.0381125226860255, "step": 1430}, {"loss": 1.075, "grad_norm": 0.38167616724967957, "learning_rate": 0.0002, "epoch": 1.0453720508166968, "step": 1440}, {"loss": 1.0229, "grad_norm": 0.4439380168914795, "learning_rate": 0.0002, "epoch": 1.0526315789473684, "step": 1450}, {"loss": 1.0451, "grad_norm": 0.30954182147979736, "learning_rate": 0.0002, "epoch": 1.05989110707804, "step": 1460}, {"loss": 1.0762, "grad_norm": 0.4022280275821686, "learning_rate": 0.0002, "epoch": 1.0671506352087115, "step": 1470}, {"loss": 1.0754, "grad_norm": 0.7390811443328857, "learning_rate": 0.0002, "epoch": 1.074410163339383, "step": 1480}, {"loss": 1.1378, "grad_norm": 0.3885486423969269, "learning_rate": 0.0002, "epoch": 1.0816696914700545, "step": 1490}, {"loss": 1.005, "grad_norm": 0.5275560617446899, "learning_rate": 0.0002, "epoch": 1.0889292196007259, "step": 1500}, {"loss": 1.0147, "grad_norm": 0.35112282633781433, "learning_rate": 0.0002, "epoch": 1.0961887477313974, "step": 1510}, {"loss": 1.0331, "grad_norm": 0.33714351058006287, "learning_rate": 0.0002, "epoch": 1.103448275862069, "step": 1520}, {"loss": 0.8723, "grad_norm": 0.31221693754196167, "learning_rate": 0.0002, "epoch": 1.1107078039927405, "step": 1530}, {"loss": 1.0701, "grad_norm": 0.27549654245376587, "learning_rate": 0.0002, "epoch": 1.117967332123412, "step": 1540}, {"loss": 1.0114, "grad_norm": 0.8465521335601807, "learning_rate": 0.0002, "epoch": 1.1252268602540836, "step": 1550}, {"loss": 0.9643, "grad_norm": 0.36125949025154114, "learning_rate": 0.0002, "epoch": 1.132486388384755, "step": 1560}, {"loss": 1.0744, "grad_norm": 0.37420371174812317, "learning_rate": 0.0002, "epoch": 1.1397459165154264, "step": 1570}, {"loss": 1.165, "grad_norm": 0.3294760584831238, "learning_rate": 0.0002, "epoch": 1.147005444646098, "step": 1580}, {"loss": 0.9207, "grad_norm": 0.3881238102912903, "learning_rate": 0.0002, "epoch": 1.1542649727767695, "step": 1590}, {"loss": 1.0983, "grad_norm": 0.4766491651535034, "learning_rate": 0.0002, "epoch": 1.161524500907441, "step": 1600}, {"loss": 1.0092, "grad_norm": 0.395530104637146, "learning_rate": 0.0002, "epoch": 1.1687840290381124, "step": 1610}, {"loss": 0.9895, "grad_norm": 0.3297106623649597, "learning_rate": 0.0002, "epoch": 1.176043557168784, "step": 1620}, {"loss": 1.1121, "grad_norm": 0.39528271555900574, "learning_rate": 0.0002, "epoch": 1.1833030852994555, "step": 1630}, {"loss": 0.9202, "grad_norm": 0.3370221257209778, "learning_rate": 0.0002, "epoch": 1.190562613430127, "step": 1640}, {"loss": 1.102, "grad_norm": 0.31922030448913574, "learning_rate": 0.0002, "epoch": 1.1978221415607986, "step": 1650}, {"loss": 1.0565, "grad_norm": 0.6142027378082275, "learning_rate": 0.0002, "epoch": 1.2050816696914701, "step": 1660}, {"loss": 0.9541, "grad_norm": 0.44769710302352905, "learning_rate": 0.0002, "epoch": 1.2123411978221417, "step": 1670}, {"loss": 1.2501, "grad_norm": 0.41233646869659424, "learning_rate": 0.0002, "epoch": 1.219600725952813, "step": 1680}, {"loss": 1.0147, "grad_norm": 0.2928866147994995, "learning_rate": 0.0002, "epoch": 1.2268602540834845, "step": 1690}, {"loss": 0.9074, "grad_norm": 0.36913734674453735, "learning_rate": 0.0002, "epoch": 1.234119782214156, "step": 1700}, {"loss": 1.0847, "grad_norm": 0.5281891226768494, "learning_rate": 0.0002, "epoch": 1.2413793103448276, "step": 1710}, {"loss": 0.9482, "grad_norm": 0.3374697268009186, "learning_rate": 0.0002, "epoch": 1.2486388384754992, "step": 1720}, {"loss": 0.9826, "grad_norm": 0.3802020847797394, "learning_rate": 0.0002, "epoch": 1.2558983666061705, "step": 1730}, {"loss": 1.0191, "grad_norm": 0.38048651814460754, "learning_rate": 0.0002, "epoch": 1.263157894736842, "step": 1740}, {"loss": 1.0142, "grad_norm": 0.6676169633865356, "learning_rate": 0.0002, "epoch": 1.2704174228675136, "step": 1750}, {"loss": 0.992, "grad_norm": 0.4075961410999298, "learning_rate": 0.0002, "epoch": 1.2776769509981851, "step": 1760}, {"loss": 1.0301, "grad_norm": 0.4374721348285675, "learning_rate": 0.0002, "epoch": 1.2849364791288567, "step": 1770}, {"loss": 1.0601, "grad_norm": 0.4638824164867401, "learning_rate": 0.0002, "epoch": 1.2921960072595282, "step": 1780}, {"loss": 0.9593, "grad_norm": 0.38631564378738403, "learning_rate": 0.0002, "epoch": 1.2994555353901998, "step": 1790}, {"loss": 1.0042, "grad_norm": 0.35873809456825256, "learning_rate": 0.0002, "epoch": 1.306715063520871, "step": 1800}, {"loss": 0.9835, "grad_norm": 0.33602237701416016, "learning_rate": 0.0002, "epoch": 1.3139745916515426, "step": 1810}, {"loss": 1.1032, "grad_norm": 0.46696463227272034, "learning_rate": 0.0002, "epoch": 1.3212341197822142, "step": 1820}, {"loss": 1.0123, "grad_norm": 0.368958979845047, "learning_rate": 0.0002, "epoch": 1.3284936479128857, "step": 1830}, {"loss": 1.0456, "grad_norm": 0.3160957396030426, "learning_rate": 0.0002, "epoch": 1.335753176043557, "step": 1840}, {"loss": 1.0264, "grad_norm": 0.4511511027812958, "learning_rate": 0.0002, "epoch": 1.3430127041742286, "step": 1850}, {"loss": 1.0179, "grad_norm": 0.6769845485687256, "learning_rate": 0.0002, "epoch": 1.3502722323049001, "step": 1860}, {"loss": 1.011, "grad_norm": 0.3749309480190277, "learning_rate": 0.0002, "epoch": 1.3575317604355717, "step": 1870}, {"loss": 1.0223, "grad_norm": 0.553798496723175, "learning_rate": 0.0002, "epoch": 1.3647912885662432, "step": 1880}, {"loss": 0.9634, "grad_norm": 0.3538985550403595, "learning_rate": 0.0002, "epoch": 1.3720508166969148, "step": 1890}, {"loss": 1.0275, "grad_norm": 0.5501534938812256, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1900}, {"loss": 1.0461, "grad_norm": 0.4432051181793213, "learning_rate": 0.0002, "epoch": 1.3865698729582578, "step": 1910}, {"loss": 0.9082, "grad_norm": 0.41755786538124084, "learning_rate": 0.0002, "epoch": 1.3938294010889292, "step": 1920}, {"loss": 1.0858, "grad_norm": 0.5626114010810852, "learning_rate": 0.0002, "epoch": 1.4010889292196007, "step": 1930}, {"loss": 1.0687, "grad_norm": 0.44038185477256775, "learning_rate": 0.0002, "epoch": 1.4083484573502723, "step": 1940}, {"loss": 0.9454, "grad_norm": 0.3397001624107361, "learning_rate": 0.0002, "epoch": 1.4156079854809438, "step": 1950}, {"loss": 0.943, "grad_norm": 0.4325368106365204, "learning_rate": 0.0002, "epoch": 1.4228675136116151, "step": 1960}, {"loss": 1.0488, "grad_norm": 0.3900907039642334, "learning_rate": 0.0002, "epoch": 1.4301270417422867, "step": 1970}, {"loss": 0.9699, "grad_norm": 0.369612455368042, "learning_rate": 0.0002, "epoch": 1.4373865698729582, "step": 1980}, {"loss": 1.0609, "grad_norm": 0.4389338791370392, "learning_rate": 0.0002, "epoch": 1.4446460980036298, "step": 1990}, {"loss": 1.042, "grad_norm": 1.694450855255127, "learning_rate": 0.0002, "epoch": 1.4519056261343013, "step": 2000}, {"loss": 0.997, "grad_norm": 0.516957700252533, "learning_rate": 0.0002, "epoch": 1.4591651542649728, "step": 2010}, {"loss": 1.0979, "grad_norm": 0.45515501499176025, "learning_rate": 0.0002, "epoch": 1.4664246823956444, "step": 2020}, {"loss": 1.0188, "grad_norm": 0.4153139591217041, "learning_rate": 0.0002, "epoch": 1.4736842105263157, "step": 2030}, {"loss": 1.1521, "grad_norm": 0.44353052973747253, "learning_rate": 0.0002, "epoch": 1.4809437386569873, "step": 2040}, {"loss": 0.9653, "grad_norm": 0.570554256439209, "learning_rate": 0.0002, "epoch": 1.4882032667876588, "step": 2050}, {"loss": 1.0352, "grad_norm": 0.5742740035057068, "learning_rate": 0.0002, "epoch": 1.4954627949183303, "step": 2060}, {"loss": 1.0963, "grad_norm": 0.5890517830848694, "learning_rate": 0.0002, "epoch": 1.5027223230490017, "step": 2070}, {"loss": 1.0639, "grad_norm": 0.4162650406360626, "learning_rate": 0.0002, "epoch": 1.5099818511796732, "step": 2080}, {"loss": 1.071, "grad_norm": 0.7334717512130737, "learning_rate": 0.0002, "epoch": 1.5172413793103448, "step": 2090}, {"loss": 0.8957, "grad_norm": 0.2963249683380127, "learning_rate": 0.0002, "epoch": 1.5245009074410163, "step": 2100}, {"loss": 1.0446, "grad_norm": 0.30676454305648804, "learning_rate": 0.0002, "epoch": 1.5317604355716878, "step": 2110}, {"loss": 0.9967, "grad_norm": 0.35984641313552856, "learning_rate": 0.0002, "epoch": 1.5390199637023594, "step": 2120}, {"loss": 1.0352, "grad_norm": 0.3384549617767334, "learning_rate": 0.0002, "epoch": 1.546279491833031, "step": 2130}, {"loss": 0.9874, "grad_norm": 0.4725518226623535, "learning_rate": 0.0002, "epoch": 1.5535390199637025, "step": 2140}, {"loss": 1.1166, "grad_norm": 0.3252685070037842, "learning_rate": 0.0002, "epoch": 1.560798548094374, "step": 2150}, {"loss": 0.9127, "grad_norm": 0.45043081045150757, "learning_rate": 0.0002, "epoch": 1.5680580762250453, "step": 2160}, {"loss": 1.0767, "grad_norm": 0.374208003282547, "learning_rate": 0.0002, "epoch": 1.5753176043557169, "step": 2170}, {"loss": 1.0255, "grad_norm": 0.5118404030799866, "learning_rate": 0.0002, "epoch": 1.5825771324863884, "step": 2180}, {"loss": 1.0062, "grad_norm": 0.482170969247818, "learning_rate": 0.0002, "epoch": 1.5898366606170597, "step": 2190}, {"loss": 1.0872, "grad_norm": 0.5337533950805664, "learning_rate": 0.0002, "epoch": 1.5970961887477313, "step": 2200}, {"loss": 1.0405, "grad_norm": 0.5195064544677734, "learning_rate": 0.0002, "epoch": 1.6043557168784028, "step": 2210}, {"loss": 1.0454, "grad_norm": 0.30807098746299744, "learning_rate": 0.0002, "epoch": 1.6116152450090744, "step": 2220}, {"loss": 1.0293, "grad_norm": 0.3962925672531128, "learning_rate": 0.0002, "epoch": 1.618874773139746, "step": 2230}, {"loss": 1.0137, "grad_norm": 0.7636962532997131, "learning_rate": 0.0002, "epoch": 1.6261343012704175, "step": 2240}, {"loss": 0.991, "grad_norm": 0.32380592823028564, "learning_rate": 0.0002, "epoch": 1.633393829401089, "step": 2250}, {"loss": 0.9471, "grad_norm": 0.5767741799354553, "learning_rate": 0.0002, "epoch": 1.6406533575317606, "step": 2260}, {"loss": 0.9524, "grad_norm": 0.39964812994003296, "learning_rate": 0.0002, "epoch": 1.647912885662432, "step": 2270}, {"loss": 0.9866, "grad_norm": 0.622629702091217, "learning_rate": 0.0002, "epoch": 1.6551724137931034, "step": 2280}, {"loss": 0.9609, "grad_norm": 0.40202152729034424, "learning_rate": 0.0002, "epoch": 1.662431941923775, "step": 2290}, {"loss": 0.9859, "grad_norm": 0.4467349052429199, "learning_rate": 0.0002, "epoch": 1.6696914700544465, "step": 2300}, {"loss": 1.0312, "grad_norm": 0.5026949048042297, "learning_rate": 0.0002, "epoch": 1.6769509981851178, "step": 2310}, {"loss": 1.0776, "grad_norm": 0.43754541873931885, "learning_rate": 0.0002, "epoch": 1.6842105263157894, "step": 2320}, {"loss": 0.9786, "grad_norm": 0.42869430780410767, "learning_rate": 0.0002, "epoch": 1.691470054446461, "step": 2330}, {"loss": 1.0747, "grad_norm": 0.4192679524421692, "learning_rate": 0.0002, "epoch": 1.6987295825771325, "step": 2340}, {"loss": 1.0054, "grad_norm": 0.3243155777454376, "learning_rate": 0.0002, "epoch": 1.705989110707804, "step": 2350}, {"loss": 0.9165, "grad_norm": 1.0514075756072998, "learning_rate": 0.0002, "epoch": 1.7132486388384756, "step": 2360}, {"loss": 1.1353, "grad_norm": 0.4719122648239136, "learning_rate": 0.0002, "epoch": 1.720508166969147, "step": 2370}, {"loss": 0.9913, "grad_norm": 0.3846144676208496, "learning_rate": 0.0002, "epoch": 1.7277676950998186, "step": 2380}, {"loss": 0.9935, "grad_norm": 0.7266581058502197, "learning_rate": 0.0002, "epoch": 1.73502722323049, "step": 2390}, {"loss": 0.9509, "grad_norm": 0.6183241605758667, "learning_rate": 0.0002, "epoch": 1.7422867513611615, "step": 2400}, {"loss": 0.9713, "grad_norm": 0.3658260405063629, "learning_rate": 0.0002, "epoch": 1.749546279491833, "step": 2410}, {"loss": 1.0033, "grad_norm": 0.6036322712898254, "learning_rate": 0.0002, "epoch": 1.7568058076225044, "step": 2420}, {"loss": 1.0526, "grad_norm": 0.7872936129570007, "learning_rate": 0.0002, "epoch": 1.764065335753176, "step": 2430}, {"loss": 0.99, "grad_norm": 0.35946124792099, "learning_rate": 0.0002, "epoch": 1.7713248638838475, "step": 2440}, {"loss": 0.8845, "grad_norm": 0.3740338981151581, "learning_rate": 0.0002, "epoch": 1.778584392014519, "step": 2450}, {"loss": 0.9333, "grad_norm": 0.6150230169296265, "learning_rate": 0.0002, "epoch": 1.7858439201451906, "step": 2460}, {"loss": 1.0373, "grad_norm": 0.4726075530052185, "learning_rate": 0.0002, "epoch": 1.793103448275862, "step": 2470}, {"loss": 0.974, "grad_norm": 0.31292253732681274, "learning_rate": 0.0002, "epoch": 1.8003629764065336, "step": 2480}, {"loss": 1.0205, "grad_norm": 0.4463104009628296, "learning_rate": 0.0002, "epoch": 1.8076225045372052, "step": 2490}, {"loss": 1.0771, "grad_norm": 0.7848200798034668, "learning_rate": 0.0002, "epoch": 1.8148820326678767, "step": 2500}, {"loss": 0.9508, "grad_norm": 0.5562082529067993, "learning_rate": 0.0002, "epoch": 1.822141560798548, "step": 2510}, {"loss": 0.9343, "grad_norm": 0.39892756938934326, "learning_rate": 0.0002, "epoch": 1.8294010889292196, "step": 2520}, {"loss": 1.0489, "grad_norm": 1.2923320531845093, "learning_rate": 0.0002, "epoch": 1.8366606170598911, "step": 2530}, {"loss": 1.1203, "grad_norm": 0.6316490769386292, "learning_rate": 0.0002, "epoch": 1.8439201451905625, "step": 2540}, {"loss": 0.9131, "grad_norm": 0.46100497245788574, "learning_rate": 0.0002, "epoch": 1.851179673321234, "step": 2550}, {"loss": 1.0977, "grad_norm": 0.7902987003326416, "learning_rate": 0.0002, "epoch": 1.8584392014519056, "step": 2560}, {"loss": 0.9702, "grad_norm": 0.4596365690231323, "learning_rate": 0.0002, "epoch": 1.865698729582577, "step": 2570}, {"loss": 0.9322, "grad_norm": 0.6592172384262085, "learning_rate": 0.0002, "epoch": 1.8729582577132486, "step": 2580}, {"loss": 0.9549, "grad_norm": 0.5764662027359009, "learning_rate": 0.0002, "epoch": 1.8802177858439202, "step": 2590}, {"loss": 0.9485, "grad_norm": 0.8421637415885925, "learning_rate": 0.0002, "epoch": 1.8874773139745917, "step": 2600}, {"loss": 1.012, "grad_norm": 0.5635305047035217, "learning_rate": 0.0002, "epoch": 1.8947368421052633, "step": 2610}, {"loss": 0.8907, "grad_norm": 0.46231237053871155, "learning_rate": 0.0002, "epoch": 1.9019963702359348, "step": 2620}, {"loss": 0.9543, "grad_norm": 0.3944607079029083, "learning_rate": 0.0002, "epoch": 1.9092558983666061, "step": 2630}, {"loss": 0.9964, "grad_norm": 0.4753907322883606, "learning_rate": 0.0002, "epoch": 1.9165154264972777, "step": 2640}, {"loss": 1.0217, "grad_norm": 0.4151090979576111, "learning_rate": 0.0002, "epoch": 1.9237749546279492, "step": 2650}, {"loss": 1.0397, "grad_norm": 0.6793725490570068, "learning_rate": 0.0002, "epoch": 1.9310344827586206, "step": 2660}, {"loss": 1.0401, "grad_norm": 0.339755117893219, "learning_rate": 0.0002, "epoch": 1.938294010889292, "step": 2670}, {"loss": 1.082, "grad_norm": 0.40789374709129333, "learning_rate": 0.0002, "epoch": 1.9455535390199636, "step": 2680}, {"loss": 0.9973, "grad_norm": 0.3750005066394806, "learning_rate": 0.0002, "epoch": 1.9528130671506352, "step": 2690}, {"loss": 0.9837, "grad_norm": 0.39684441685676575, "learning_rate": 0.0002, "epoch": 1.9600725952813067, "step": 2700}, {"loss": 1.1204, "grad_norm": 0.378287672996521, "learning_rate": 0.0002, "epoch": 1.9673321234119783, "step": 2710}, {"loss": 0.9284, "grad_norm": 0.3668482005596161, "learning_rate": 0.0002, "epoch": 1.9745916515426498, "step": 2720}, {"loss": 1.0512, "grad_norm": 0.49997565150260925, "learning_rate": 0.0002, "epoch": 1.9818511796733214, "step": 2730}, {"loss": 0.9311, "grad_norm": 0.36852124333381653, "learning_rate": 0.0002, "epoch": 1.989110707803993, "step": 2740}, {"loss": 1.0898, "grad_norm": 0.5203380584716797, "learning_rate": 0.0002, "epoch": 1.9963702359346642, "step": 2750}, {"eval_loss": 1.1086540222167969, "eval_runtime": 95.6866, "eval_samples_per_second": 4.557, "eval_steps_per_second": 0.575, "epoch": 2.0, "step": 2755}, {"loss": 0.9791, "grad_norm": 0.35921573638916016, "learning_rate": 0.0002, "epoch": 2.0036297640653356, "step": 2760}, {"loss": 0.8968, "grad_norm": 1.013861894607544, "learning_rate": 0.0002, "epoch": 2.010889292196007, "step": 2770}, {"loss": 0.7873, "grad_norm": 0.4425240159034729, "learning_rate": 0.0002, "epoch": 2.0181488203266786, "step": 2780}, {"loss": 0.8758, "grad_norm": 0.60174161195755, "learning_rate": 0.0002, "epoch": 2.02540834845735, "step": 2790}, {"loss": 0.7989, "grad_norm": 0.47582098841667175, "learning_rate": 0.0002, "epoch": 2.0326678765880217, "step": 2800}, {"loss": 0.8898, "grad_norm": 0.6012811660766602, "learning_rate": 0.0002, "epoch": 2.0399274047186933, "step": 2810}, {"loss": 0.8109, "grad_norm": 0.4444408118724823, "learning_rate": 0.0002, "epoch": 2.047186932849365, "step": 2820}, {"loss": 0.8621, "grad_norm": 0.4864003360271454, "learning_rate": 0.0002, "epoch": 2.0544464609800364, "step": 2830}, {"loss": 0.8817, "grad_norm": 0.5104215741157532, "learning_rate": 0.0002, "epoch": 2.061705989110708, "step": 2840}, {"loss": 0.7789, "grad_norm": 0.6218489408493042, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 2850}, {"loss": 0.8778, "grad_norm": 0.705784261226654, "learning_rate": 0.0002, "epoch": 2.076225045372051, "step": 2860}, {"loss": 0.7851, "grad_norm": 0.48091503977775574, "learning_rate": 0.0002, "epoch": 2.0834845735027225, "step": 2870}, {"loss": 0.8444, "grad_norm": 0.5062456727027893, "learning_rate": 0.0002, "epoch": 2.0907441016333936, "step": 2880}, {"loss": 0.7898, "grad_norm": 0.4862022399902344, "learning_rate": 0.0002, "epoch": 2.098003629764065, "step": 2890}, {"loss": 0.8593, "grad_norm": 0.48264044523239136, "learning_rate": 0.0002, "epoch": 2.1052631578947367, "step": 2900}, {"loss": 0.8325, "grad_norm": 0.43744346499443054, "learning_rate": 0.0002, "epoch": 2.1125226860254083, "step": 2910}, {"loss": 0.9099, "grad_norm": 0.5480492115020752, "learning_rate": 0.0002, "epoch": 2.11978221415608, "step": 2920}, {"loss": 0.7727, "grad_norm": 0.5068560838699341, "learning_rate": 0.0002, "epoch": 2.1270417422867514, "step": 2930}, {"loss": 0.8403, "grad_norm": 0.4650852680206299, "learning_rate": 0.0002, "epoch": 2.134301270417423, "step": 2940}, {"loss": 0.8467, "grad_norm": 0.4929981231689453, "learning_rate": 0.0002, "epoch": 2.1415607985480944, "step": 2950}, {"loss": 0.8747, "grad_norm": 0.6537389755249023, "learning_rate": 0.0002, "epoch": 2.148820326678766, "step": 2960}, {"loss": 0.7663, "grad_norm": 0.8032940626144409, "learning_rate": 0.0002, "epoch": 2.1560798548094375, "step": 2970}, {"loss": 0.7604, "grad_norm": 0.7131643891334534, "learning_rate": 0.0002, "epoch": 2.163339382940109, "step": 2980}, {"loss": 0.8424, "grad_norm": 0.6034275889396667, "learning_rate": 0.0002, "epoch": 2.1705989110707806, "step": 2990}, {"loss": 0.8838, "grad_norm": 0.6081095933914185, "learning_rate": 0.0002, "epoch": 2.1778584392014517, "step": 3000}, {"loss": 0.9013, "grad_norm": 0.5706912875175476, "learning_rate": 0.0002, "epoch": 2.1851179673321233, "step": 3010}, {"loss": 0.8473, "grad_norm": 0.6742380261421204, "learning_rate": 0.0002, "epoch": 2.192377495462795, "step": 3020}, {"loss": 0.8372, "grad_norm": 0.4847496449947357, "learning_rate": 0.0002, "epoch": 2.1996370235934664, "step": 3030}, {"loss": 0.8487, "grad_norm": 0.5529342889785767, "learning_rate": 0.0002, "epoch": 2.206896551724138, "step": 3040}, {"loss": 0.9057, "grad_norm": 0.6108783483505249, "learning_rate": 0.0002, "epoch": 2.2141560798548094, "step": 3050}, {"loss": 0.9659, "grad_norm": 0.8841571807861328, "learning_rate": 0.0002, "epoch": 2.221415607985481, "step": 3060}, {"loss": 0.8538, "grad_norm": 0.4227530360221863, "learning_rate": 0.0002, "epoch": 2.2286751361161525, "step": 3070}, {"loss": 0.7925, "grad_norm": 0.461935818195343, "learning_rate": 0.0002, "epoch": 2.235934664246824, "step": 3080}, {"loss": 0.8542, "grad_norm": 0.5407412648200989, "learning_rate": 0.0002, "epoch": 2.2431941923774956, "step": 3090}, {"loss": 0.8392, "grad_norm": 0.5057998895645142, "learning_rate": 0.0002, "epoch": 2.250453720508167, "step": 3100}, {"loss": 0.7898, "grad_norm": 0.530057966709137, "learning_rate": 0.0002, "epoch": 2.2577132486388383, "step": 3110}, {"loss": 0.8507, "grad_norm": 0.5066842436790466, "learning_rate": 0.0002, "epoch": 2.26497277676951, "step": 3120}, {"loss": 0.7903, "grad_norm": 0.5069178342819214, "learning_rate": 0.0002, "epoch": 2.2722323049001814, "step": 3130}, {"loss": 0.8458, "grad_norm": 0.6095499396324158, "learning_rate": 0.0002, "epoch": 2.279491833030853, "step": 3140}, {"loss": 0.8171, "grad_norm": 0.49524766206741333, "learning_rate": 0.0002, "epoch": 2.2867513611615244, "step": 3150}, {"loss": 0.7568, "grad_norm": 0.5334409475326538, "learning_rate": 0.0002, "epoch": 2.294010889292196, "step": 3160}, {"loss": 0.7533, "grad_norm": 1.681748867034912, "learning_rate": 0.0002, "epoch": 2.3012704174228675, "step": 3170}, {"loss": 0.8155, "grad_norm": 0.7225565314292908, "learning_rate": 0.0002, "epoch": 2.308529945553539, "step": 3180}, {"loss": 0.8034, "grad_norm": 0.5379388928413391, "learning_rate": 0.0002, "epoch": 2.3157894736842106, "step": 3190}, {"loss": 0.8833, "grad_norm": 0.45770326256752014, "learning_rate": 0.0002, "epoch": 2.323049001814882, "step": 3200}, {"loss": 0.8419, "grad_norm": 0.6984533071517944, "learning_rate": 0.0002, "epoch": 2.3303085299455537, "step": 3210}, {"loss": 0.7414, "grad_norm": 0.6725744605064392, "learning_rate": 0.0002, "epoch": 2.337568058076225, "step": 3220}, {"loss": 0.8657, "grad_norm": 1.1247471570968628, "learning_rate": 0.0002, "epoch": 2.344827586206897, "step": 3230}, {"loss": 0.7782, "grad_norm": 1.0240263938903809, "learning_rate": 0.0002, "epoch": 2.352087114337568, "step": 3240}, {"loss": 0.826, "grad_norm": 0.5608096122741699, "learning_rate": 0.0002, "epoch": 2.3593466424682394, "step": 3250}, {"loss": 0.8191, "grad_norm": 0.8294990062713623, "learning_rate": 0.0002, "epoch": 2.366606170598911, "step": 3260}, {"loss": 0.9845, "grad_norm": 0.6734224557876587, "learning_rate": 0.0002, "epoch": 2.3738656987295825, "step": 3270}, {"loss": 0.7921, "grad_norm": 0.6862800717353821, "learning_rate": 0.0002, "epoch": 2.381125226860254, "step": 3280}, {"loss": 0.7752, "grad_norm": 0.5442930459976196, "learning_rate": 0.0002, "epoch": 2.3883847549909256, "step": 3290}, {"loss": 0.8543, "grad_norm": 0.5745394229888916, "learning_rate": 0.0002, "epoch": 2.395644283121597, "step": 3300}, {"loss": 0.833, "grad_norm": 0.6257799863815308, "learning_rate": 0.0002, "epoch": 2.4029038112522687, "step": 3310}, {"loss": 0.8191, "grad_norm": 0.5608420968055725, "learning_rate": 0.0002, "epoch": 2.4101633393829403, "step": 3320}, {"loss": 0.8745, "grad_norm": 0.5512017011642456, "learning_rate": 0.0002, "epoch": 2.417422867513612, "step": 3330}, {"loss": 0.8642, "grad_norm": 0.7452999353408813, "learning_rate": 0.0002, "epoch": 2.4246823956442833, "step": 3340}, {"loss": 0.8417, "grad_norm": 0.4604301452636719, "learning_rate": 0.0002, "epoch": 2.4319419237749544, "step": 3350}, {"loss": 0.9148, "grad_norm": 0.8225823640823364, "learning_rate": 0.0002, "epoch": 2.439201451905626, "step": 3360}, {"loss": 0.8144, "grad_norm": 0.8604981303215027, "learning_rate": 0.0002, "epoch": 2.4464609800362975, "step": 3370}, {"loss": 0.9276, "grad_norm": 0.6620925664901733, "learning_rate": 0.0002, "epoch": 2.453720508166969, "step": 3380}, {"loss": 0.8381, "grad_norm": 0.4750158488750458, "learning_rate": 0.0002, "epoch": 2.4609800362976406, "step": 3390}, {"loss": 0.8249, "grad_norm": 0.6061418056488037, "learning_rate": 0.0002, "epoch": 2.468239564428312, "step": 3400}, {"loss": 0.8944, "grad_norm": 0.5977247357368469, "learning_rate": 0.0002, "epoch": 2.4754990925589837, "step": 3410}, {"loss": 0.8366, "grad_norm": 0.5004227757453918, "learning_rate": 0.0002, "epoch": 2.4827586206896552, "step": 3420}, {"loss": 0.9133, "grad_norm": 0.46428972482681274, "learning_rate": 0.0002, "epoch": 2.490018148820327, "step": 3430}, {"loss": 0.8534, "grad_norm": 0.7658395171165466, "learning_rate": 0.0002, "epoch": 2.4972776769509983, "step": 3440}, {"loss": 0.9066, "grad_norm": 0.621819794178009, "learning_rate": 0.0002, "epoch": 2.50453720508167, "step": 3450}, {"loss": 0.8481, "grad_norm": 0.4556088149547577, "learning_rate": 0.0002, "epoch": 2.511796733212341, "step": 3460}, {"loss": 0.8522, "grad_norm": 0.6124140024185181, "learning_rate": 0.0002, "epoch": 2.519056261343013, "step": 3470}, {"loss": 0.8158, "grad_norm": 0.6256231665611267, "learning_rate": 0.0002, "epoch": 2.526315789473684, "step": 3480}, {"loss": 0.9097, "grad_norm": 0.5464141964912415, "learning_rate": 0.0002, "epoch": 2.5335753176043556, "step": 3490}, {"loss": 0.8708, "grad_norm": 0.51471346616745, "learning_rate": 0.0002, "epoch": 2.540834845735027, "step": 3500}, {"loss": 0.9042, "grad_norm": 0.5326165556907654, "learning_rate": 0.0002, "epoch": 2.5480943738656987, "step": 3510}, {"loss": 0.8749, "grad_norm": 0.4750378429889679, "learning_rate": 0.0002, "epoch": 2.5553539019963702, "step": 3520}, {"loss": 0.8248, "grad_norm": 0.5292279124259949, "learning_rate": 0.0002, "epoch": 2.562613430127042, "step": 3530}, {"loss": 0.8808, "grad_norm": 0.6145227551460266, "learning_rate": 0.0002, "epoch": 2.5698729582577133, "step": 3540}, {"loss": 0.846, "grad_norm": 0.8275189399719238, "learning_rate": 0.0002, "epoch": 2.577132486388385, "step": 3550}, {"loss": 0.875, "grad_norm": 0.5037438273429871, "learning_rate": 0.0002, "epoch": 2.5843920145190564, "step": 3560}, {"loss": 0.8324, "grad_norm": 0.5838707685470581, "learning_rate": 0.0002, "epoch": 2.5916515426497275, "step": 3570}, {"loss": 0.8716, "grad_norm": 0.5398710370063782, "learning_rate": 0.0002, "epoch": 2.5989110707803995, "step": 3580}, {"loss": 0.7843, "grad_norm": 0.6115376949310303, "learning_rate": 0.0002, "epoch": 2.6061705989110706, "step": 3590}, {"loss": 0.8142, "grad_norm": 1.118809461593628, "learning_rate": 0.0002, "epoch": 2.613430127041742, "step": 3600}, {"loss": 0.8282, "grad_norm": 0.7811771631240845, "learning_rate": 0.0002, "epoch": 2.6206896551724137, "step": 3610}, {"loss": 0.7997, "grad_norm": 0.5753175020217896, "learning_rate": 0.0002, "epoch": 2.6279491833030852, "step": 3620}, {"loss": 0.7705, "grad_norm": 0.550829291343689, "learning_rate": 0.0002, "epoch": 2.635208711433757, "step": 3630}, {"loss": 0.8564, "grad_norm": 0.5360019207000732, "learning_rate": 0.0002, "epoch": 2.6424682395644283, "step": 3640}, {"loss": 0.7648, "grad_norm": 0.63050377368927, "learning_rate": 0.0002, "epoch": 2.6497277676951, "step": 3650}, {"loss": 0.9156, "grad_norm": 0.5833110213279724, "learning_rate": 0.0002, "epoch": 2.6569872958257714, "step": 3660}, {"loss": 0.8033, "grad_norm": 0.5543047189712524, "learning_rate": 0.0002, "epoch": 2.664246823956443, "step": 3670}, {"loss": 0.8867, "grad_norm": 0.6842212080955505, "learning_rate": 0.0002, "epoch": 2.671506352087114, "step": 3680}, {"loss": 0.8569, "grad_norm": 0.6859333515167236, "learning_rate": 0.0002, "epoch": 2.678765880217786, "step": 3690}, {"loss": 0.8081, "grad_norm": 0.7038410902023315, "learning_rate": 0.0002, "epoch": 2.686025408348457, "step": 3700}, {"loss": 0.8686, "grad_norm": 0.447233647108078, "learning_rate": 0.0002, "epoch": 2.6932849364791287, "step": 3710}, {"loss": 0.8951, "grad_norm": 0.6453872919082642, "learning_rate": 0.0002, "epoch": 2.7005444646098002, "step": 3720}, {"loss": 0.8292, "grad_norm": 0.8025672435760498, "learning_rate": 0.0002, "epoch": 2.707803992740472, "step": 3730}, {"loss": 0.8488, "grad_norm": 0.5997087955474854, "learning_rate": 0.0002, "epoch": 2.7150635208711433, "step": 3740}, {"loss": 0.8308, "grad_norm": 0.6901142001152039, "learning_rate": 0.0002, "epoch": 2.722323049001815, "step": 3750}, {"loss": 0.8517, "grad_norm": 1.036145567893982, "learning_rate": 0.0002, "epoch": 2.7295825771324864, "step": 3760}, {"loss": 0.8455, "grad_norm": 0.7207072377204895, "learning_rate": 0.0002, "epoch": 2.736842105263158, "step": 3770}, {"loss": 0.8009, "grad_norm": 1.0452989339828491, "learning_rate": 0.0002, "epoch": 2.7441016333938295, "step": 3780}, {"loss": 0.8868, "grad_norm": 0.5615278482437134, "learning_rate": 0.0002, "epoch": 2.751361161524501, "step": 3790}, {"loss": 0.9047, "grad_norm": 0.46439215540885925, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 3800}, {"loss": 0.7847, "grad_norm": 0.7134917974472046, "learning_rate": 0.0002, "epoch": 2.7658802177858437, "step": 3810}, {"loss": 0.8315, "grad_norm": 0.5139115452766418, "learning_rate": 0.0002, "epoch": 2.7731397459165157, "step": 3820}, {"loss": 0.8995, "grad_norm": 0.8595781326293945, "learning_rate": 0.0002, "epoch": 2.780399274047187, "step": 3830}, {"loss": 0.9079, "grad_norm": 0.544614851474762, "learning_rate": 0.0002, "epoch": 2.7876588021778583, "step": 3840}, {"loss": 0.8322, "grad_norm": 0.6073850393295288, "learning_rate": 0.0002, "epoch": 2.79491833030853, "step": 3850}, {"loss": 0.8277, "grad_norm": 0.8224069476127625, "learning_rate": 0.0002, "epoch": 2.8021778584392014, "step": 3860}, {"loss": 0.815, "grad_norm": 0.5347970128059387, "learning_rate": 0.0002, "epoch": 2.809437386569873, "step": 3870}, {"loss": 0.8219, "grad_norm": 0.6799601912498474, "learning_rate": 0.0002, "epoch": 2.8166969147005445, "step": 3880}, {"loss": 0.8796, "grad_norm": 0.5219197869300842, "learning_rate": 0.0002, "epoch": 2.823956442831216, "step": 3890}, {"loss": 0.8649, "grad_norm": 0.5710130333900452, "learning_rate": 0.0002, "epoch": 2.8312159709618876, "step": 3900}, {"loss": 0.8741, "grad_norm": 0.5857213139533997, "learning_rate": 0.0002, "epoch": 2.838475499092559, "step": 3910}, {"loss": 0.8154, "grad_norm": 0.5206644535064697, "learning_rate": 0.0002, "epoch": 2.8457350272232302, "step": 3920}, {"loss": 0.9328, "grad_norm": 1.2902015447616577, "learning_rate": 0.0002, "epoch": 2.8529945553539022, "step": 3930}, {"loss": 0.8192, "grad_norm": 0.5252797603607178, "learning_rate": 0.0002, "epoch": 2.8602540834845733, "step": 3940}, {"loss": 0.9047, "grad_norm": 0.5925108790397644, "learning_rate": 0.0002, "epoch": 2.867513611615245, "step": 3950}, {"loss": 0.8611, "grad_norm": 0.9719610810279846, "learning_rate": 0.0002, "epoch": 2.8747731397459164, "step": 3960}, {"loss": 0.8531, "grad_norm": 0.6834747195243835, "learning_rate": 0.0002, "epoch": 2.882032667876588, "step": 3970}, {"loss": 0.8124, "grad_norm": 0.6952353119850159, "learning_rate": 0.0002, "epoch": 2.8892921960072595, "step": 3980}, {"loss": 0.8365, "grad_norm": 0.49889910221099854, "learning_rate": 0.0002, "epoch": 2.896551724137931, "step": 3990}, {"loss": 0.8197, "grad_norm": 0.5007026791572571, "learning_rate": 0.0002, "epoch": 2.9038112522686026, "step": 4000}, {"loss": 0.7752, "grad_norm": 0.5474239587783813, "learning_rate": 0.0002, "epoch": 2.911070780399274, "step": 4010}, {"loss": 0.8579, "grad_norm": 0.6617428660392761, "learning_rate": 0.0002, "epoch": 2.9183303085299457, "step": 4020}, {"loss": 0.8583, "grad_norm": 0.6097776293754578, "learning_rate": 0.0002, "epoch": 2.925589836660617, "step": 4030}, {"loss": 0.8296, "grad_norm": 0.5985828638076782, "learning_rate": 0.0002, "epoch": 2.9328493647912888, "step": 4040}, {"loss": 0.9134, "grad_norm": 0.769488513469696, "learning_rate": 0.0002, "epoch": 2.94010889292196, "step": 4050}, {"loss": 0.8148, "grad_norm": 0.5167055130004883, "learning_rate": 0.0002, "epoch": 2.9473684210526314, "step": 4060}, {"loss": 0.8665, "grad_norm": 0.6265496015548706, "learning_rate": 0.0002, "epoch": 2.954627949183303, "step": 4070}, {"loss": 0.9218, "grad_norm": 1.2644082307815552, "learning_rate": 0.0002, "epoch": 2.9618874773139745, "step": 4080}, {"loss": 0.8026, "grad_norm": 0.6007561087608337, "learning_rate": 0.0002, "epoch": 2.969147005444646, "step": 4090}, {"loss": 0.8262, "grad_norm": 0.47984927892684937, "learning_rate": 0.0002, "epoch": 2.9764065335753176, "step": 4100}, {"loss": 0.7897, "grad_norm": 1.128198504447937, "learning_rate": 0.0002, "epoch": 2.983666061705989, "step": 4110}, {"loss": 0.7883, "grad_norm": 0.526292085647583, "learning_rate": 0.0002, "epoch": 2.9909255898366607, "step": 4120}, {"loss": 0.7801, "grad_norm": 0.5462674498558044, "learning_rate": 0.0002, "epoch": 2.9981851179673322, "step": 4130}]}
+{"epoch": 4.0, "step": 5510, "epoch_duration": 2056.9718704223633, "total_accumulated_duration": 14286.312230110168, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4936, "grad_norm": 0.46010470390319824, "learning_rate": 0.0002, "epoch": 0.007259528130671506, "step": 10}, {"loss": 1.8442, "grad_norm": 0.5103179216384888, "learning_rate": 0.0002, "epoch": 0.014519056261343012, "step": 20}, {"loss": 1.559, "grad_norm": 0.6282716989517212, "learning_rate": 0.0002, "epoch": 0.021778584392014518, "step": 30}, {"loss": 1.3618, "grad_norm": 1.2480497360229492, "learning_rate": 0.0002, "epoch": 0.029038112522686024, "step": 40}, {"loss": 1.3874, "grad_norm": 0.4114855229854584, "learning_rate": 0.0002, "epoch": 0.036297640653357534, "step": 50}, {"loss": 1.2836, "grad_norm": 0.49482840299606323, "learning_rate": 0.0002, "epoch": 0.043557168784029036, "step": 60}, {"loss": 1.1728, "grad_norm": 0.4536272883415222, "learning_rate": 0.0002, "epoch": 0.050816696914700546, "step": 70}, {"loss": 1.1125, "grad_norm": 0.32328274846076965, "learning_rate": 0.0002, "epoch": 0.05807622504537205, "step": 80}, {"loss": 1.2447, "grad_norm": 0.40990468859672546, "learning_rate": 0.0002, "epoch": 0.06533575317604355, "step": 90}, {"loss": 1.2216, "grad_norm": 0.37273502349853516, "learning_rate": 0.0002, "epoch": 0.07259528130671507, "step": 100}, {"loss": 1.2259, "grad_norm": 0.3903871476650238, "learning_rate": 0.0002, "epoch": 0.07985480943738657, "step": 110}, {"loss": 1.3718, "grad_norm": 0.3272787928581238, "learning_rate": 0.0002, "epoch": 0.08711433756805807, "step": 120}, {"loss": 1.1832, "grad_norm": 0.3622824251651764, "learning_rate": 0.0002, "epoch": 0.09437386569872959, "step": 130}, {"loss": 1.2966, "grad_norm": 0.3503916561603546, "learning_rate": 0.0002, "epoch": 0.10163339382940109, "step": 140}, {"loss": 1.0913, "grad_norm": 0.32787832617759705, "learning_rate": 0.0002, "epoch": 0.1088929219600726, "step": 150}, {"loss": 1.2143, "grad_norm": 0.5822657942771912, "learning_rate": 0.0002, "epoch": 0.1161524500907441, "step": 160}, {"loss": 1.2091, "grad_norm": 0.28028249740600586, "learning_rate": 0.0002, "epoch": 0.12341197822141561, "step": 170}, {"loss": 1.1731, "grad_norm": 0.33602750301361084, "learning_rate": 0.0002, "epoch": 0.1306715063520871, "step": 180}, {"loss": 1.127, "grad_norm": 0.29106274247169495, "learning_rate": 0.0002, "epoch": 0.13793103448275862, "step": 190}, {"loss": 1.3261, "grad_norm": 0.38753363490104675, "learning_rate": 0.0002, "epoch": 0.14519056261343014, "step": 200}, {"loss": 1.1536, "grad_norm": 0.361009418964386, "learning_rate": 0.0002, "epoch": 0.15245009074410162, "step": 210}, {"loss": 1.1483, "grad_norm": 0.6743836402893066, "learning_rate": 0.0002, "epoch": 0.15970961887477314, "step": 220}, {"loss": 1.264, "grad_norm": 0.3923613727092743, "learning_rate": 0.0002, "epoch": 0.16696914700544466, "step": 230}, {"loss": 1.0437, "grad_norm": 0.2809699773788452, "learning_rate": 0.0002, "epoch": 0.17422867513611615, "step": 240}, {"loss": 1.1826, "grad_norm": 0.3631494641304016, "learning_rate": 0.0002, "epoch": 0.18148820326678766, "step": 250}, {"loss": 1.1584, "grad_norm": 0.24658171832561493, "learning_rate": 0.0002, "epoch": 0.18874773139745918, "step": 260}, {"loss": 1.0065, "grad_norm": 0.5780664682388306, "learning_rate": 0.0002, "epoch": 0.19600725952813067, "step": 270}, {"loss": 1.1251, "grad_norm": 0.3056720495223999, "learning_rate": 0.0002, "epoch": 0.20326678765880218, "step": 280}, {"loss": 1.1503, "grad_norm": 0.2857084572315216, "learning_rate": 0.0002, "epoch": 0.21052631578947367, "step": 290}, {"loss": 1.1215, "grad_norm": 0.6645345687866211, "learning_rate": 0.0002, "epoch": 0.2177858439201452, "step": 300}, {"loss": 1.127, "grad_norm": 0.5966078639030457, "learning_rate": 0.0002, "epoch": 0.2250453720508167, "step": 310}, {"loss": 1.1974, "grad_norm": 0.40937140583992004, "learning_rate": 0.0002, "epoch": 0.2323049001814882, "step": 320}, {"loss": 1.2385, "grad_norm": 0.5642806887626648, "learning_rate": 0.0002, "epoch": 0.2395644283121597, "step": 330}, {"loss": 1.1703, "grad_norm": 0.2750748097896576, "learning_rate": 0.0002, "epoch": 0.24682395644283123, "step": 340}, {"loss": 1.2598, "grad_norm": 0.34350234270095825, "learning_rate": 0.0002, "epoch": 0.2540834845735027, "step": 350}, {"loss": 1.1942, "grad_norm": 0.6767239570617676, "learning_rate": 0.0002, "epoch": 0.2613430127041742, "step": 360}, {"loss": 1.1436, "grad_norm": 0.31006959080696106, "learning_rate": 0.0002, "epoch": 0.26860254083484575, "step": 370}, {"loss": 1.2084, "grad_norm": 0.3825474679470062, "learning_rate": 0.0002, "epoch": 0.27586206896551724, "step": 380}, {"loss": 1.1523, "grad_norm": 0.30241551995277405, "learning_rate": 0.0002, "epoch": 0.2831215970961887, "step": 390}, {"loss": 1.1298, "grad_norm": 0.2962397336959839, "learning_rate": 0.0002, "epoch": 0.29038112522686027, "step": 400}, {"loss": 1.1299, "grad_norm": 0.2600369155406952, "learning_rate": 0.0002, "epoch": 0.29764065335753176, "step": 410}, {"loss": 1.1366, "grad_norm": 0.3675060272216797, "learning_rate": 0.0002, "epoch": 0.30490018148820325, "step": 420}, {"loss": 1.156, "grad_norm": 0.3429498076438904, "learning_rate": 0.0002, "epoch": 0.3121597096188748, "step": 430}, {"loss": 1.2741, "grad_norm": 0.34311825037002563, "learning_rate": 0.0002, "epoch": 0.3194192377495463, "step": 440}, {"loss": 1.3523, "grad_norm": 0.37872210144996643, "learning_rate": 0.0002, "epoch": 0.32667876588021777, "step": 450}, {"loss": 1.1365, "grad_norm": 0.33271121978759766, "learning_rate": 0.0002, "epoch": 0.3339382940108893, "step": 460}, {"loss": 1.1572, "grad_norm": 0.34605276584625244, "learning_rate": 0.0002, "epoch": 0.3411978221415608, "step": 470}, {"loss": 1.2755, "grad_norm": 0.41050562262535095, "learning_rate": 0.0002, "epoch": 0.3484573502722323, "step": 480}, {"loss": 1.075, "grad_norm": 0.2066836953163147, "learning_rate": 0.0002, "epoch": 0.35571687840290384, "step": 490}, {"loss": 1.2072, "grad_norm": 0.2859014868736267, "learning_rate": 0.0002, "epoch": 0.3629764065335753, "step": 500}, {"loss": 1.1435, "grad_norm": 0.28763777017593384, "learning_rate": 0.0002, "epoch": 0.3702359346642468, "step": 510}, {"loss": 1.1341, "grad_norm": 0.2730471193790436, "learning_rate": 0.0002, "epoch": 0.37749546279491836, "step": 520}, {"loss": 1.348, "grad_norm": 0.3968936800956726, "learning_rate": 0.0002, "epoch": 0.38475499092558985, "step": 530}, {"loss": 1.2363, "grad_norm": 0.3624701201915741, "learning_rate": 0.0002, "epoch": 0.39201451905626133, "step": 540}, {"loss": 1.1206, "grad_norm": 0.3303608298301697, "learning_rate": 0.0002, "epoch": 0.3992740471869328, "step": 550}, {"loss": 1.203, "grad_norm": 0.33507466316223145, "learning_rate": 0.0002, "epoch": 0.40653357531760437, "step": 560}, {"loss": 1.2294, "grad_norm": 0.3297670781612396, "learning_rate": 0.0002, "epoch": 0.41379310344827586, "step": 570}, {"loss": 1.2, "grad_norm": 0.32334890961647034, "learning_rate": 0.0002, "epoch": 0.42105263157894735, "step": 580}, {"loss": 1.112, "grad_norm": 0.30281195044517517, "learning_rate": 0.0002, "epoch": 0.4283121597096189, "step": 590}, {"loss": 1.1352, "grad_norm": 0.5900027751922607, "learning_rate": 0.0002, "epoch": 0.4355716878402904, "step": 600}, {"loss": 1.1575, "grad_norm": 0.28437477350234985, "learning_rate": 0.0002, "epoch": 0.44283121597096187, "step": 610}, {"loss": 1.2009, "grad_norm": 0.39601704478263855, "learning_rate": 0.0002, "epoch": 0.4500907441016334, "step": 620}, {"loss": 1.2302, "grad_norm": 0.41971510648727417, "learning_rate": 0.0002, "epoch": 0.4573502722323049, "step": 630}, {"loss": 1.236, "grad_norm": 0.33814409375190735, "learning_rate": 0.0002, "epoch": 0.4646098003629764, "step": 640}, {"loss": 1.2471, "grad_norm": 0.575718104839325, "learning_rate": 0.0002, "epoch": 0.47186932849364793, "step": 650}, {"loss": 1.1012, "grad_norm": 0.37927401065826416, "learning_rate": 0.0002, "epoch": 0.4791288566243194, "step": 660}, {"loss": 1.1552, "grad_norm": 0.3224332630634308, "learning_rate": 0.0002, "epoch": 0.4863883847549909, "step": 670}, {"loss": 1.1354, "grad_norm": 0.32683515548706055, "learning_rate": 0.0002, "epoch": 0.49364791288566245, "step": 680}, {"loss": 1.207, "grad_norm": 0.4316163659095764, "learning_rate": 0.0002, "epoch": 0.5009074410163339, "step": 690}, {"loss": 1.2116, "grad_norm": 0.342602401971817, "learning_rate": 0.0002, "epoch": 0.5081669691470054, "step": 700}, {"loss": 1.0433, "grad_norm": 0.2794898748397827, "learning_rate": 0.0002, "epoch": 0.515426497277677, "step": 710}, {"loss": 1.2395, "grad_norm": 0.3322339951992035, "learning_rate": 0.0002, "epoch": 0.5226860254083484, "step": 720}, {"loss": 1.0762, "grad_norm": 0.3088509142398834, "learning_rate": 0.0002, "epoch": 0.52994555353902, "step": 730}, {"loss": 1.1495, "grad_norm": 0.24444378912448883, "learning_rate": 0.0002, "epoch": 0.5372050816696915, "step": 740}, {"loss": 1.2176, "grad_norm": 0.3483171761035919, "learning_rate": 0.0002, "epoch": 0.5444646098003629, "step": 750}, {"loss": 1.2248, "grad_norm": 0.3425690531730652, "learning_rate": 0.0002, "epoch": 0.5517241379310345, "step": 760}, {"loss": 1.1127, "grad_norm": 0.31841927766799927, "learning_rate": 0.0002, "epoch": 0.558983666061706, "step": 770}, {"loss": 1.1841, "grad_norm": 0.39423868060112, "learning_rate": 0.0002, "epoch": 0.5662431941923775, "step": 780}, {"loss": 1.0583, "grad_norm": 0.30328479409217834, "learning_rate": 0.0002, "epoch": 0.573502722323049, "step": 790}, {"loss": 1.0718, "grad_norm": 0.24475938081741333, "learning_rate": 0.0002, "epoch": 0.5807622504537205, "step": 800}, {"loss": 1.2006, "grad_norm": 0.37132805585861206, "learning_rate": 0.0002, "epoch": 0.588021778584392, "step": 810}, {"loss": 1.1544, "grad_norm": 0.32195979356765747, "learning_rate": 0.0002, "epoch": 0.5952813067150635, "step": 820}, {"loss": 0.9937, "grad_norm": 0.2848738729953766, "learning_rate": 0.0002, "epoch": 0.6025408348457351, "step": 830}, {"loss": 1.0937, "grad_norm": 0.28015264868736267, "learning_rate": 0.0002, "epoch": 0.6098003629764065, "step": 840}, {"loss": 1.1234, "grad_norm": 0.37796008586883545, "learning_rate": 0.0002, "epoch": 0.617059891107078, "step": 850}, {"loss": 1.1455, "grad_norm": 0.39311841130256653, "learning_rate": 0.0002, "epoch": 0.6243194192377496, "step": 860}, {"loss": 1.1066, "grad_norm": 0.2761685252189636, "learning_rate": 0.0002, "epoch": 0.631578947368421, "step": 870}, {"loss": 1.0859, "grad_norm": 0.3826720118522644, "learning_rate": 0.0002, "epoch": 0.6388384754990926, "step": 880}, {"loss": 1.1182, "grad_norm": 0.30076679587364197, "learning_rate": 0.0002, "epoch": 0.6460980036297641, "step": 890}, {"loss": 1.0927, "grad_norm": 0.21997687220573425, "learning_rate": 0.0002, "epoch": 0.6533575317604355, "step": 900}, {"loss": 1.1603, "grad_norm": 0.32593777775764465, "learning_rate": 0.0002, "epoch": 0.6606170598911071, "step": 910}, {"loss": 1.1578, "grad_norm": 0.30347898602485657, "learning_rate": 0.0002, "epoch": 0.6678765880217786, "step": 920}, {"loss": 1.1642, "grad_norm": 0.44173774123191833, "learning_rate": 0.0002, "epoch": 0.6751361161524501, "step": 930}, {"loss": 1.1832, "grad_norm": 0.2507467269897461, "learning_rate": 0.0002, "epoch": 0.6823956442831216, "step": 940}, {"loss": 1.3001, "grad_norm": 0.29463833570480347, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 950}, {"loss": 1.1651, "grad_norm": 0.9363154172897339, "learning_rate": 0.0002, "epoch": 0.6969147005444646, "step": 960}, {"loss": 1.158, "grad_norm": 0.3236212134361267, "learning_rate": 0.0002, "epoch": 0.7041742286751361, "step": 970}, {"loss": 1.1078, "grad_norm": 0.3123254179954529, "learning_rate": 0.0002, "epoch": 0.7114337568058077, "step": 980}, {"loss": 1.1578, "grad_norm": 0.3395805060863495, "learning_rate": 0.0002, "epoch": 0.7186932849364791, "step": 990}, {"loss": 1.0754, "grad_norm": 0.3240964412689209, "learning_rate": 0.0002, "epoch": 0.7259528130671506, "step": 1000}, {"loss": 1.0313, "grad_norm": 0.31902948021888733, "learning_rate": 0.0002, "epoch": 0.7332123411978222, "step": 1010}, {"loss": 1.1664, "grad_norm": 0.4848408102989197, "learning_rate": 0.0002, "epoch": 0.7404718693284936, "step": 1020}, {"loss": 1.0838, "grad_norm": 0.33006033301353455, "learning_rate": 0.0002, "epoch": 0.7477313974591652, "step": 1030}, {"loss": 1.0979, "grad_norm": 0.2928730547428131, "learning_rate": 0.0002, "epoch": 0.7549909255898367, "step": 1040}, {"loss": 1.1351, "grad_norm": 0.3529164791107178, "learning_rate": 0.0002, "epoch": 0.7622504537205081, "step": 1050}, {"loss": 1.1274, "grad_norm": 0.2736213803291321, "learning_rate": 0.0002, "epoch": 0.7695099818511797, "step": 1060}, {"loss": 1.1169, "grad_norm": 0.7200686931610107, "learning_rate": 0.0002, "epoch": 0.7767695099818511, "step": 1070}, {"loss": 1.1171, "grad_norm": 0.33396708965301514, "learning_rate": 0.0002, "epoch": 0.7840290381125227, "step": 1080}, {"loss": 1.2447, "grad_norm": 1.5760449171066284, "learning_rate": 0.0002, "epoch": 0.7912885662431942, "step": 1090}, {"loss": 1.1126, "grad_norm": 0.28138381242752075, "learning_rate": 0.0002, "epoch": 0.7985480943738656, "step": 1100}, {"loss": 1.2012, "grad_norm": 0.2597472369670868, "learning_rate": 0.0002, "epoch": 0.8058076225045372, "step": 1110}, {"loss": 1.1177, "grad_norm": 0.3305445611476898, "learning_rate": 0.0002, "epoch": 0.8130671506352087, "step": 1120}, {"loss": 1.0849, "grad_norm": 0.3934599459171295, "learning_rate": 0.0002, "epoch": 0.8203266787658802, "step": 1130}, {"loss": 1.0889, "grad_norm": 0.3472191393375397, "learning_rate": 0.0002, "epoch": 0.8275862068965517, "step": 1140}, {"loss": 1.265, "grad_norm": 0.2857365906238556, "learning_rate": 0.0002, "epoch": 0.8348457350272233, "step": 1150}, {"loss": 1.03, "grad_norm": 0.3207702934741974, "learning_rate": 0.0002, "epoch": 0.8421052631578947, "step": 1160}, {"loss": 1.1669, "grad_norm": 0.3176484704017639, "learning_rate": 0.0002, "epoch": 0.8493647912885662, "step": 1170}, {"loss": 1.1386, "grad_norm": 0.40685558319091797, "learning_rate": 0.0002, "epoch": 0.8566243194192378, "step": 1180}, {"loss": 1.1383, "grad_norm": 0.31125199794769287, "learning_rate": 0.0002, "epoch": 0.8638838475499092, "step": 1190}, {"loss": 1.1512, "grad_norm": 0.7361181378364563, "learning_rate": 0.0002, "epoch": 0.8711433756805808, "step": 1200}, {"loss": 0.9983, "grad_norm": 0.33699527382850647, "learning_rate": 0.0002, "epoch": 0.8784029038112523, "step": 1210}, {"loss": 1.11, "grad_norm": 0.3315220773220062, "learning_rate": 0.0002, "epoch": 0.8856624319419237, "step": 1220}, {"loss": 1.0266, "grad_norm": 0.6256054043769836, "learning_rate": 0.0002, "epoch": 0.8929219600725953, "step": 1230}, {"loss": 1.1738, "grad_norm": 0.3692137897014618, "learning_rate": 0.0002, "epoch": 0.9001814882032668, "step": 1240}, {"loss": 1.1112, "grad_norm": 0.3538484573364258, "learning_rate": 0.0002, "epoch": 0.9074410163339383, "step": 1250}, {"loss": 1.0878, "grad_norm": 0.27863040566444397, "learning_rate": 0.0002, "epoch": 0.9147005444646098, "step": 1260}, {"loss": 1.0826, "grad_norm": 0.3322528302669525, "learning_rate": 0.0002, "epoch": 0.9219600725952813, "step": 1270}, {"loss": 1.0158, "grad_norm": 0.7553173303604126, "learning_rate": 0.0002, "epoch": 0.9292196007259528, "step": 1280}, {"loss": 1.198, "grad_norm": 0.4856191575527191, "learning_rate": 0.0002, "epoch": 0.9364791288566243, "step": 1290}, {"loss": 1.1678, "grad_norm": 0.3668074905872345, "learning_rate": 0.0002, "epoch": 0.9437386569872959, "step": 1300}, {"loss": 1.1809, "grad_norm": 0.29851067066192627, "learning_rate": 0.0002, "epoch": 0.9509981851179673, "step": 1310}, {"loss": 1.1299, "grad_norm": 0.276664674282074, "learning_rate": 0.0002, "epoch": 0.9582577132486388, "step": 1320}, {"loss": 1.0597, "grad_norm": 0.2941018342971802, "learning_rate": 0.0002, "epoch": 0.9655172413793104, "step": 1330}, {"loss": 1.1731, "grad_norm": 0.3505859076976776, "learning_rate": 0.0002, "epoch": 0.9727767695099818, "step": 1340}, {"loss": 1.1809, "grad_norm": 0.3067687451839447, "learning_rate": 0.0002, "epoch": 0.9800362976406534, "step": 1350}, {"loss": 1.1511, "grad_norm": 0.27151066064834595, "learning_rate": 0.0002, "epoch": 0.9872958257713249, "step": 1360}, {"loss": 1.1919, "grad_norm": 0.36370083689689636, "learning_rate": 0.0002, "epoch": 0.9945553539019963, "step": 1370}, {"eval_loss": 1.1381088495254517, "eval_runtime": 96.0848, "eval_samples_per_second": 4.538, "eval_steps_per_second": 0.572, "epoch": 0.9996370235934664, "step": 1377}, {"loss": 1.0082, "grad_norm": 0.27980583906173706, "learning_rate": 0.0002, "epoch": 1.0018148820326678, "step": 1380}, {"loss": 1.0565, "grad_norm": 0.26713913679122925, "learning_rate": 0.0002, "epoch": 1.0090744101633393, "step": 1390}, {"loss": 1.0241, "grad_norm": 0.3089541494846344, "learning_rate": 0.0002, "epoch": 1.0163339382940109, "step": 1400}, {"loss": 0.9649, "grad_norm": 0.4188242256641388, "learning_rate": 0.0002, "epoch": 1.0235934664246824, "step": 1410}, {"loss": 0.9821, "grad_norm": 0.5246463418006897, "learning_rate": 0.0002, "epoch": 1.030852994555354, "step": 1420}, {"loss": 1.0143, "grad_norm": 0.2728777825832367, "learning_rate": 0.0002, "epoch": 1.0381125226860255, "step": 1430}, {"loss": 1.075, "grad_norm": 0.38167616724967957, "learning_rate": 0.0002, "epoch": 1.0453720508166968, "step": 1440}, {"loss": 1.0229, "grad_norm": 0.4439380168914795, "learning_rate": 0.0002, "epoch": 1.0526315789473684, "step": 1450}, {"loss": 1.0451, "grad_norm": 0.30954182147979736, "learning_rate": 0.0002, "epoch": 1.05989110707804, "step": 1460}, {"loss": 1.0762, "grad_norm": 0.4022280275821686, "learning_rate": 0.0002, "epoch": 1.0671506352087115, "step": 1470}, {"loss": 1.0754, "grad_norm": 0.7390811443328857, "learning_rate": 0.0002, "epoch": 1.074410163339383, "step": 1480}, {"loss": 1.1378, "grad_norm": 0.3885486423969269, "learning_rate": 0.0002, "epoch": 1.0816696914700545, "step": 1490}, {"loss": 1.005, "grad_norm": 0.5275560617446899, "learning_rate": 0.0002, "epoch": 1.0889292196007259, "step": 1500}, {"loss": 1.0147, "grad_norm": 0.35112282633781433, "learning_rate": 0.0002, "epoch": 1.0961887477313974, "step": 1510}, {"loss": 1.0331, "grad_norm": 0.33714351058006287, "learning_rate": 0.0002, "epoch": 1.103448275862069, "step": 1520}, {"loss": 0.8723, "grad_norm": 0.31221693754196167, "learning_rate": 0.0002, "epoch": 1.1107078039927405, "step": 1530}, {"loss": 1.0701, "grad_norm": 0.27549654245376587, "learning_rate": 0.0002, "epoch": 1.117967332123412, "step": 1540}, {"loss": 1.0114, "grad_norm": 0.8465521335601807, "learning_rate": 0.0002, "epoch": 1.1252268602540836, "step": 1550}, {"loss": 0.9643, "grad_norm": 0.36125949025154114, "learning_rate": 0.0002, "epoch": 1.132486388384755, "step": 1560}, {"loss": 1.0744, "grad_norm": 0.37420371174812317, "learning_rate": 0.0002, "epoch": 1.1397459165154264, "step": 1570}, {"loss": 1.165, "grad_norm": 0.3294760584831238, "learning_rate": 0.0002, "epoch": 1.147005444646098, "step": 1580}, {"loss": 0.9207, "grad_norm": 0.3881238102912903, "learning_rate": 0.0002, "epoch": 1.1542649727767695, "step": 1590}, {"loss": 1.0983, "grad_norm": 0.4766491651535034, "learning_rate": 0.0002, "epoch": 1.161524500907441, "step": 1600}, {"loss": 1.0092, "grad_norm": 0.395530104637146, "learning_rate": 0.0002, "epoch": 1.1687840290381124, "step": 1610}, {"loss": 0.9895, "grad_norm": 0.3297106623649597, "learning_rate": 0.0002, "epoch": 1.176043557168784, "step": 1620}, {"loss": 1.1121, "grad_norm": 0.39528271555900574, "learning_rate": 0.0002, "epoch": 1.1833030852994555, "step": 1630}, {"loss": 0.9202, "grad_norm": 0.3370221257209778, "learning_rate": 0.0002, "epoch": 1.190562613430127, "step": 1640}, {"loss": 1.102, "grad_norm": 0.31922030448913574, "learning_rate": 0.0002, "epoch": 1.1978221415607986, "step": 1650}, {"loss": 1.0565, "grad_norm": 0.6142027378082275, "learning_rate": 0.0002, "epoch": 1.2050816696914701, "step": 1660}, {"loss": 0.9541, "grad_norm": 0.44769710302352905, "learning_rate": 0.0002, "epoch": 1.2123411978221417, "step": 1670}, {"loss": 1.2501, "grad_norm": 0.41233646869659424, "learning_rate": 0.0002, "epoch": 1.219600725952813, "step": 1680}, {"loss": 1.0147, "grad_norm": 0.2928866147994995, "learning_rate": 0.0002, "epoch": 1.2268602540834845, "step": 1690}, {"loss": 0.9074, "grad_norm": 0.36913734674453735, "learning_rate": 0.0002, "epoch": 1.234119782214156, "step": 1700}, {"loss": 1.0847, "grad_norm": 0.5281891226768494, "learning_rate": 0.0002, "epoch": 1.2413793103448276, "step": 1710}, {"loss": 0.9482, "grad_norm": 0.3374697268009186, "learning_rate": 0.0002, "epoch": 1.2486388384754992, "step": 1720}, {"loss": 0.9826, "grad_norm": 0.3802020847797394, "learning_rate": 0.0002, "epoch": 1.2558983666061705, "step": 1730}, {"loss": 1.0191, "grad_norm": 0.38048651814460754, "learning_rate": 0.0002, "epoch": 1.263157894736842, "step": 1740}, {"loss": 1.0142, "grad_norm": 0.6676169633865356, "learning_rate": 0.0002, "epoch": 1.2704174228675136, "step": 1750}, {"loss": 0.992, "grad_norm": 0.4075961410999298, "learning_rate": 0.0002, "epoch": 1.2776769509981851, "step": 1760}, {"loss": 1.0301, "grad_norm": 0.4374721348285675, "learning_rate": 0.0002, "epoch": 1.2849364791288567, "step": 1770}, {"loss": 1.0601, "grad_norm": 0.4638824164867401, "learning_rate": 0.0002, "epoch": 1.2921960072595282, "step": 1780}, {"loss": 0.9593, "grad_norm": 0.38631564378738403, "learning_rate": 0.0002, "epoch": 1.2994555353901998, "step": 1790}, {"loss": 1.0042, "grad_norm": 0.35873809456825256, "learning_rate": 0.0002, "epoch": 1.306715063520871, "step": 1800}, {"loss": 0.9835, "grad_norm": 0.33602237701416016, "learning_rate": 0.0002, "epoch": 1.3139745916515426, "step": 1810}, {"loss": 1.1032, "grad_norm": 0.46696463227272034, "learning_rate": 0.0002, "epoch": 1.3212341197822142, "step": 1820}, {"loss": 1.0123, "grad_norm": 0.368958979845047, "learning_rate": 0.0002, "epoch": 1.3284936479128857, "step": 1830}, {"loss": 1.0456, "grad_norm": 0.3160957396030426, "learning_rate": 0.0002, "epoch": 1.335753176043557, "step": 1840}, {"loss": 1.0264, "grad_norm": 0.4511511027812958, "learning_rate": 0.0002, "epoch": 1.3430127041742286, "step": 1850}, {"loss": 1.0179, "grad_norm": 0.6769845485687256, "learning_rate": 0.0002, "epoch": 1.3502722323049001, "step": 1860}, {"loss": 1.011, "grad_norm": 0.3749309480190277, "learning_rate": 0.0002, "epoch": 1.3575317604355717, "step": 1870}, {"loss": 1.0223, "grad_norm": 0.553798496723175, "learning_rate": 0.0002, "epoch": 1.3647912885662432, "step": 1880}, {"loss": 0.9634, "grad_norm": 0.3538985550403595, "learning_rate": 0.0002, "epoch": 1.3720508166969148, "step": 1890}, {"loss": 1.0275, "grad_norm": 0.5501534938812256, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1900}, {"loss": 1.0461, "grad_norm": 0.4432051181793213, "learning_rate": 0.0002, "epoch": 1.3865698729582578, "step": 1910}, {"loss": 0.9082, "grad_norm": 0.41755786538124084, "learning_rate": 0.0002, "epoch": 1.3938294010889292, "step": 1920}, {"loss": 1.0858, "grad_norm": 0.5626114010810852, "learning_rate": 0.0002, "epoch": 1.4010889292196007, "step": 1930}, {"loss": 1.0687, "grad_norm": 0.44038185477256775, "learning_rate": 0.0002, "epoch": 1.4083484573502723, "step": 1940}, {"loss": 0.9454, "grad_norm": 0.3397001624107361, "learning_rate": 0.0002, "epoch": 1.4156079854809438, "step": 1950}, {"loss": 0.943, "grad_norm": 0.4325368106365204, "learning_rate": 0.0002, "epoch": 1.4228675136116151, "step": 1960}, {"loss": 1.0488, "grad_norm": 0.3900907039642334, "learning_rate": 0.0002, "epoch": 1.4301270417422867, "step": 1970}, {"loss": 0.9699, "grad_norm": 0.369612455368042, "learning_rate": 0.0002, "epoch": 1.4373865698729582, "step": 1980}, {"loss": 1.0609, "grad_norm": 0.4389338791370392, "learning_rate": 0.0002, "epoch": 1.4446460980036298, "step": 1990}, {"loss": 1.042, "grad_norm": 1.694450855255127, "learning_rate": 0.0002, "epoch": 1.4519056261343013, "step": 2000}, {"loss": 0.997, "grad_norm": 0.516957700252533, "learning_rate": 0.0002, "epoch": 1.4591651542649728, "step": 2010}, {"loss": 1.0979, "grad_norm": 0.45515501499176025, "learning_rate": 0.0002, "epoch": 1.4664246823956444, "step": 2020}, {"loss": 1.0188, "grad_norm": 0.4153139591217041, "learning_rate": 0.0002, "epoch": 1.4736842105263157, "step": 2030}, {"loss": 1.1521, "grad_norm": 0.44353052973747253, "learning_rate": 0.0002, "epoch": 1.4809437386569873, "step": 2040}, {"loss": 0.9653, "grad_norm": 0.570554256439209, "learning_rate": 0.0002, "epoch": 1.4882032667876588, "step": 2050}, {"loss": 1.0352, "grad_norm": 0.5742740035057068, "learning_rate": 0.0002, "epoch": 1.4954627949183303, "step": 2060}, {"loss": 1.0963, "grad_norm": 0.5890517830848694, "learning_rate": 0.0002, "epoch": 1.5027223230490017, "step": 2070}, {"loss": 1.0639, "grad_norm": 0.4162650406360626, "learning_rate": 0.0002, "epoch": 1.5099818511796732, "step": 2080}, {"loss": 1.071, "grad_norm": 0.7334717512130737, "learning_rate": 0.0002, "epoch": 1.5172413793103448, "step": 2090}, {"loss": 0.8957, "grad_norm": 0.2963249683380127, "learning_rate": 0.0002, "epoch": 1.5245009074410163, "step": 2100}, {"loss": 1.0446, "grad_norm": 0.30676454305648804, "learning_rate": 0.0002, "epoch": 1.5317604355716878, "step": 2110}, {"loss": 0.9967, "grad_norm": 0.35984641313552856, "learning_rate": 0.0002, "epoch": 1.5390199637023594, "step": 2120}, {"loss": 1.0352, "grad_norm": 0.3384549617767334, "learning_rate": 0.0002, "epoch": 1.546279491833031, "step": 2130}, {"loss": 0.9874, "grad_norm": 0.4725518226623535, "learning_rate": 0.0002, "epoch": 1.5535390199637025, "step": 2140}, {"loss": 1.1166, "grad_norm": 0.3252685070037842, "learning_rate": 0.0002, "epoch": 1.560798548094374, "step": 2150}, {"loss": 0.9127, "grad_norm": 0.45043081045150757, "learning_rate": 0.0002, "epoch": 1.5680580762250453, "step": 2160}, {"loss": 1.0767, "grad_norm": 0.374208003282547, "learning_rate": 0.0002, "epoch": 1.5753176043557169, "step": 2170}, {"loss": 1.0255, "grad_norm": 0.5118404030799866, "learning_rate": 0.0002, "epoch": 1.5825771324863884, "step": 2180}, {"loss": 1.0062, "grad_norm": 0.482170969247818, "learning_rate": 0.0002, "epoch": 1.5898366606170597, "step": 2190}, {"loss": 1.0872, "grad_norm": 0.5337533950805664, "learning_rate": 0.0002, "epoch": 1.5970961887477313, "step": 2200}, {"loss": 1.0405, "grad_norm": 0.5195064544677734, "learning_rate": 0.0002, "epoch": 1.6043557168784028, "step": 2210}, {"loss": 1.0454, "grad_norm": 0.30807098746299744, "learning_rate": 0.0002, "epoch": 1.6116152450090744, "step": 2220}, {"loss": 1.0293, "grad_norm": 0.3962925672531128, "learning_rate": 0.0002, "epoch": 1.618874773139746, "step": 2230}, {"loss": 1.0137, "grad_norm": 0.7636962532997131, "learning_rate": 0.0002, "epoch": 1.6261343012704175, "step": 2240}, {"loss": 0.991, "grad_norm": 0.32380592823028564, "learning_rate": 0.0002, "epoch": 1.633393829401089, "step": 2250}, {"loss": 0.9471, "grad_norm": 0.5767741799354553, "learning_rate": 0.0002, "epoch": 1.6406533575317606, "step": 2260}, {"loss": 0.9524, "grad_norm": 0.39964812994003296, "learning_rate": 0.0002, "epoch": 1.647912885662432, "step": 2270}, {"loss": 0.9866, "grad_norm": 0.622629702091217, "learning_rate": 0.0002, "epoch": 1.6551724137931034, "step": 2280}, {"loss": 0.9609, "grad_norm": 0.40202152729034424, "learning_rate": 0.0002, "epoch": 1.662431941923775, "step": 2290}, {"loss": 0.9859, "grad_norm": 0.4467349052429199, "learning_rate": 0.0002, "epoch": 1.6696914700544465, "step": 2300}, {"loss": 1.0312, "grad_norm": 0.5026949048042297, "learning_rate": 0.0002, "epoch": 1.6769509981851178, "step": 2310}, {"loss": 1.0776, "grad_norm": 0.43754541873931885, "learning_rate": 0.0002, "epoch": 1.6842105263157894, "step": 2320}, {"loss": 0.9786, "grad_norm": 0.42869430780410767, "learning_rate": 0.0002, "epoch": 1.691470054446461, "step": 2330}, {"loss": 1.0747, "grad_norm": 0.4192679524421692, "learning_rate": 0.0002, "epoch": 1.6987295825771325, "step": 2340}, {"loss": 1.0054, "grad_norm": 0.3243155777454376, "learning_rate": 0.0002, "epoch": 1.705989110707804, "step": 2350}, {"loss": 0.9165, "grad_norm": 1.0514075756072998, "learning_rate": 0.0002, "epoch": 1.7132486388384756, "step": 2360}, {"loss": 1.1353, "grad_norm": 0.4719122648239136, "learning_rate": 0.0002, "epoch": 1.720508166969147, "step": 2370}, {"loss": 0.9913, "grad_norm": 0.3846144676208496, "learning_rate": 0.0002, "epoch": 1.7277676950998186, "step": 2380}, {"loss": 0.9935, "grad_norm": 0.7266581058502197, "learning_rate": 0.0002, "epoch": 1.73502722323049, "step": 2390}, {"loss": 0.9509, "grad_norm": 0.6183241605758667, "learning_rate": 0.0002, "epoch": 1.7422867513611615, "step": 2400}, {"loss": 0.9713, "grad_norm": 0.3658260405063629, "learning_rate": 0.0002, "epoch": 1.749546279491833, "step": 2410}, {"loss": 1.0033, "grad_norm": 0.6036322712898254, "learning_rate": 0.0002, "epoch": 1.7568058076225044, "step": 2420}, {"loss": 1.0526, "grad_norm": 0.7872936129570007, "learning_rate": 0.0002, "epoch": 1.764065335753176, "step": 2430}, {"loss": 0.99, "grad_norm": 0.35946124792099, "learning_rate": 0.0002, "epoch": 1.7713248638838475, "step": 2440}, {"loss": 0.8845, "grad_norm": 0.3740338981151581, "learning_rate": 0.0002, "epoch": 1.778584392014519, "step": 2450}, {"loss": 0.9333, "grad_norm": 0.6150230169296265, "learning_rate": 0.0002, "epoch": 1.7858439201451906, "step": 2460}, {"loss": 1.0373, "grad_norm": 0.4726075530052185, "learning_rate": 0.0002, "epoch": 1.793103448275862, "step": 2470}, {"loss": 0.974, "grad_norm": 0.31292253732681274, "learning_rate": 0.0002, "epoch": 1.8003629764065336, "step": 2480}, {"loss": 1.0205, "grad_norm": 0.4463104009628296, "learning_rate": 0.0002, "epoch": 1.8076225045372052, "step": 2490}, {"loss": 1.0771, "grad_norm": 0.7848200798034668, "learning_rate": 0.0002, "epoch": 1.8148820326678767, "step": 2500}, {"loss": 0.9508, "grad_norm": 0.5562082529067993, "learning_rate": 0.0002, "epoch": 1.822141560798548, "step": 2510}, {"loss": 0.9343, "grad_norm": 0.39892756938934326, "learning_rate": 0.0002, "epoch": 1.8294010889292196, "step": 2520}, {"loss": 1.0489, "grad_norm": 1.2923320531845093, "learning_rate": 0.0002, "epoch": 1.8366606170598911, "step": 2530}, {"loss": 1.1203, "grad_norm": 0.6316490769386292, "learning_rate": 0.0002, "epoch": 1.8439201451905625, "step": 2540}, {"loss": 0.9131, "grad_norm": 0.46100497245788574, "learning_rate": 0.0002, "epoch": 1.851179673321234, "step": 2550}, {"loss": 1.0977, "grad_norm": 0.7902987003326416, "learning_rate": 0.0002, "epoch": 1.8584392014519056, "step": 2560}, {"loss": 0.9702, "grad_norm": 0.4596365690231323, "learning_rate": 0.0002, "epoch": 1.865698729582577, "step": 2570}, {"loss": 0.9322, "grad_norm": 0.6592172384262085, "learning_rate": 0.0002, "epoch": 1.8729582577132486, "step": 2580}, {"loss": 0.9549, "grad_norm": 0.5764662027359009, "learning_rate": 0.0002, "epoch": 1.8802177858439202, "step": 2590}, {"loss": 0.9485, "grad_norm": 0.8421637415885925, "learning_rate": 0.0002, "epoch": 1.8874773139745917, "step": 2600}, {"loss": 1.012, "grad_norm": 0.5635305047035217, "learning_rate": 0.0002, "epoch": 1.8947368421052633, "step": 2610}, {"loss": 0.8907, "grad_norm": 0.46231237053871155, "learning_rate": 0.0002, "epoch": 1.9019963702359348, "step": 2620}, {"loss": 0.9543, "grad_norm": 0.3944607079029083, "learning_rate": 0.0002, "epoch": 1.9092558983666061, "step": 2630}, {"loss": 0.9964, "grad_norm": 0.4753907322883606, "learning_rate": 0.0002, "epoch": 1.9165154264972777, "step": 2640}, {"loss": 1.0217, "grad_norm": 0.4151090979576111, "learning_rate": 0.0002, "epoch": 1.9237749546279492, "step": 2650}, {"loss": 1.0397, "grad_norm": 0.6793725490570068, "learning_rate": 0.0002, "epoch": 1.9310344827586206, "step": 2660}, {"loss": 1.0401, "grad_norm": 0.339755117893219, "learning_rate": 0.0002, "epoch": 1.938294010889292, "step": 2670}, {"loss": 1.082, "grad_norm": 0.40789374709129333, "learning_rate": 0.0002, "epoch": 1.9455535390199636, "step": 2680}, {"loss": 0.9973, "grad_norm": 0.3750005066394806, "learning_rate": 0.0002, "epoch": 1.9528130671506352, "step": 2690}, {"loss": 0.9837, "grad_norm": 0.39684441685676575, "learning_rate": 0.0002, "epoch": 1.9600725952813067, "step": 2700}, {"loss": 1.1204, "grad_norm": 0.378287672996521, "learning_rate": 0.0002, "epoch": 1.9673321234119783, "step": 2710}, {"loss": 0.9284, "grad_norm": 0.3668482005596161, "learning_rate": 0.0002, "epoch": 1.9745916515426498, "step": 2720}, {"loss": 1.0512, "grad_norm": 0.49997565150260925, "learning_rate": 0.0002, "epoch": 1.9818511796733214, "step": 2730}, {"loss": 0.9311, "grad_norm": 0.36852124333381653, "learning_rate": 0.0002, "epoch": 1.989110707803993, "step": 2740}, {"loss": 1.0898, "grad_norm": 0.5203380584716797, "learning_rate": 0.0002, "epoch": 1.9963702359346642, "step": 2750}, {"eval_loss": 1.1086540222167969, "eval_runtime": 95.6866, "eval_samples_per_second": 4.557, "eval_steps_per_second": 0.575, "epoch": 2.0, "step": 2755}, {"loss": 0.9791, "grad_norm": 0.35921573638916016, "learning_rate": 0.0002, "epoch": 2.0036297640653356, "step": 2760}, {"loss": 0.8968, "grad_norm": 1.013861894607544, "learning_rate": 0.0002, "epoch": 2.010889292196007, "step": 2770}, {"loss": 0.7873, "grad_norm": 0.4425240159034729, "learning_rate": 0.0002, "epoch": 2.0181488203266786, "step": 2780}, {"loss": 0.8758, "grad_norm": 0.60174161195755, "learning_rate": 0.0002, "epoch": 2.02540834845735, "step": 2790}, {"loss": 0.7989, "grad_norm": 0.47582098841667175, "learning_rate": 0.0002, "epoch": 2.0326678765880217, "step": 2800}, {"loss": 0.8898, "grad_norm": 0.6012811660766602, "learning_rate": 0.0002, "epoch": 2.0399274047186933, "step": 2810}, {"loss": 0.8109, "grad_norm": 0.4444408118724823, "learning_rate": 0.0002, "epoch": 2.047186932849365, "step": 2820}, {"loss": 0.8621, "grad_norm": 0.4864003360271454, "learning_rate": 0.0002, "epoch": 2.0544464609800364, "step": 2830}, {"loss": 0.8817, "grad_norm": 0.5104215741157532, "learning_rate": 0.0002, "epoch": 2.061705989110708, "step": 2840}, {"loss": 0.7789, "grad_norm": 0.6218489408493042, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 2850}, {"loss": 0.8778, "grad_norm": 0.705784261226654, "learning_rate": 0.0002, "epoch": 2.076225045372051, "step": 2860}, {"loss": 0.7851, "grad_norm": 0.48091503977775574, "learning_rate": 0.0002, "epoch": 2.0834845735027225, "step": 2870}, {"loss": 0.8444, "grad_norm": 0.5062456727027893, "learning_rate": 0.0002, "epoch": 2.0907441016333936, "step": 2880}, {"loss": 0.7898, "grad_norm": 0.4862022399902344, "learning_rate": 0.0002, "epoch": 2.098003629764065, "step": 2890}, {"loss": 0.8593, "grad_norm": 0.48264044523239136, "learning_rate": 0.0002, "epoch": 2.1052631578947367, "step": 2900}, {"loss": 0.8325, "grad_norm": 0.43744346499443054, "learning_rate": 0.0002, "epoch": 2.1125226860254083, "step": 2910}, {"loss": 0.9099, "grad_norm": 0.5480492115020752, "learning_rate": 0.0002, "epoch": 2.11978221415608, "step": 2920}, {"loss": 0.7727, "grad_norm": 0.5068560838699341, "learning_rate": 0.0002, "epoch": 2.1270417422867514, "step": 2930}, {"loss": 0.8403, "grad_norm": 0.4650852680206299, "learning_rate": 0.0002, "epoch": 2.134301270417423, "step": 2940}, {"loss": 0.8467, "grad_norm": 0.4929981231689453, "learning_rate": 0.0002, "epoch": 2.1415607985480944, "step": 2950}, {"loss": 0.8747, "grad_norm": 0.6537389755249023, "learning_rate": 0.0002, "epoch": 2.148820326678766, "step": 2960}, {"loss": 0.7663, "grad_norm": 0.8032940626144409, "learning_rate": 0.0002, "epoch": 2.1560798548094375, "step": 2970}, {"loss": 0.7604, "grad_norm": 0.7131643891334534, "learning_rate": 0.0002, "epoch": 2.163339382940109, "step": 2980}, {"loss": 0.8424, "grad_norm": 0.6034275889396667, "learning_rate": 0.0002, "epoch": 2.1705989110707806, "step": 2990}, {"loss": 0.8838, "grad_norm": 0.6081095933914185, "learning_rate": 0.0002, "epoch": 2.1778584392014517, "step": 3000}, {"loss": 0.9013, "grad_norm": 0.5706912875175476, "learning_rate": 0.0002, "epoch": 2.1851179673321233, "step": 3010}, {"loss": 0.8473, "grad_norm": 0.6742380261421204, "learning_rate": 0.0002, "epoch": 2.192377495462795, "step": 3020}, {"loss": 0.8372, "grad_norm": 0.4847496449947357, "learning_rate": 0.0002, "epoch": 2.1996370235934664, "step": 3030}, {"loss": 0.8487, "grad_norm": 0.5529342889785767, "learning_rate": 0.0002, "epoch": 2.206896551724138, "step": 3040}, {"loss": 0.9057, "grad_norm": 0.6108783483505249, "learning_rate": 0.0002, "epoch": 2.2141560798548094, "step": 3050}, {"loss": 0.9659, "grad_norm": 0.8841571807861328, "learning_rate": 0.0002, "epoch": 2.221415607985481, "step": 3060}, {"loss": 0.8538, "grad_norm": 0.4227530360221863, "learning_rate": 0.0002, "epoch": 2.2286751361161525, "step": 3070}, {"loss": 0.7925, "grad_norm": 0.461935818195343, "learning_rate": 0.0002, "epoch": 2.235934664246824, "step": 3080}, {"loss": 0.8542, "grad_norm": 0.5407412648200989, "learning_rate": 0.0002, "epoch": 2.2431941923774956, "step": 3090}, {"loss": 0.8392, "grad_norm": 0.5057998895645142, "learning_rate": 0.0002, "epoch": 2.250453720508167, "step": 3100}, {"loss": 0.7898, "grad_norm": 0.530057966709137, "learning_rate": 0.0002, "epoch": 2.2577132486388383, "step": 3110}, {"loss": 0.8507, "grad_norm": 0.5066842436790466, "learning_rate": 0.0002, "epoch": 2.26497277676951, "step": 3120}, {"loss": 0.7903, "grad_norm": 0.5069178342819214, "learning_rate": 0.0002, "epoch": 2.2722323049001814, "step": 3130}, {"loss": 0.8458, "grad_norm": 0.6095499396324158, "learning_rate": 0.0002, "epoch": 2.279491833030853, "step": 3140}, {"loss": 0.8171, "grad_norm": 0.49524766206741333, "learning_rate": 0.0002, "epoch": 2.2867513611615244, "step": 3150}, {"loss": 0.7568, "grad_norm": 0.5334409475326538, "learning_rate": 0.0002, "epoch": 2.294010889292196, "step": 3160}, {"loss": 0.7533, "grad_norm": 1.681748867034912, "learning_rate": 0.0002, "epoch": 2.3012704174228675, "step": 3170}, {"loss": 0.8155, "grad_norm": 0.7225565314292908, "learning_rate": 0.0002, "epoch": 2.308529945553539, "step": 3180}, {"loss": 0.8034, "grad_norm": 0.5379388928413391, "learning_rate": 0.0002, "epoch": 2.3157894736842106, "step": 3190}, {"loss": 0.8833, "grad_norm": 0.45770326256752014, "learning_rate": 0.0002, "epoch": 2.323049001814882, "step": 3200}, {"loss": 0.8419, "grad_norm": 0.6984533071517944, "learning_rate": 0.0002, "epoch": 2.3303085299455537, "step": 3210}, {"loss": 0.7414, "grad_norm": 0.6725744605064392, "learning_rate": 0.0002, "epoch": 2.337568058076225, "step": 3220}, {"loss": 0.8657, "grad_norm": 1.1247471570968628, "learning_rate": 0.0002, "epoch": 2.344827586206897, "step": 3230}, {"loss": 0.7782, "grad_norm": 1.0240263938903809, "learning_rate": 0.0002, "epoch": 2.352087114337568, "step": 3240}, {"loss": 0.826, "grad_norm": 0.5608096122741699, "learning_rate": 0.0002, "epoch": 2.3593466424682394, "step": 3250}, {"loss": 0.8191, "grad_norm": 0.8294990062713623, "learning_rate": 0.0002, "epoch": 2.366606170598911, "step": 3260}, {"loss": 0.9845, "grad_norm": 0.6734224557876587, "learning_rate": 0.0002, "epoch": 2.3738656987295825, "step": 3270}, {"loss": 0.7921, "grad_norm": 0.6862800717353821, "learning_rate": 0.0002, "epoch": 2.381125226860254, "step": 3280}, {"loss": 0.7752, "grad_norm": 0.5442930459976196, "learning_rate": 0.0002, "epoch": 2.3883847549909256, "step": 3290}, {"loss": 0.8543, "grad_norm": 0.5745394229888916, "learning_rate": 0.0002, "epoch": 2.395644283121597, "step": 3300}, {"loss": 0.833, "grad_norm": 0.6257799863815308, "learning_rate": 0.0002, "epoch": 2.4029038112522687, "step": 3310}, {"loss": 0.8191, "grad_norm": 0.5608420968055725, "learning_rate": 0.0002, "epoch": 2.4101633393829403, "step": 3320}, {"loss": 0.8745, "grad_norm": 0.5512017011642456, "learning_rate": 0.0002, "epoch": 2.417422867513612, "step": 3330}, {"loss": 0.8642, "grad_norm": 0.7452999353408813, "learning_rate": 0.0002, "epoch": 2.4246823956442833, "step": 3340}, {"loss": 0.8417, "grad_norm": 0.4604301452636719, "learning_rate": 0.0002, "epoch": 2.4319419237749544, "step": 3350}, {"loss": 0.9148, "grad_norm": 0.8225823640823364, "learning_rate": 0.0002, "epoch": 2.439201451905626, "step": 3360}, {"loss": 0.8144, "grad_norm": 0.8604981303215027, "learning_rate": 0.0002, "epoch": 2.4464609800362975, "step": 3370}, {"loss": 0.9276, "grad_norm": 0.6620925664901733, "learning_rate": 0.0002, "epoch": 2.453720508166969, "step": 3380}, {"loss": 0.8381, "grad_norm": 0.4750158488750458, "learning_rate": 0.0002, "epoch": 2.4609800362976406, "step": 3390}, {"loss": 0.8249, "grad_norm": 0.6061418056488037, "learning_rate": 0.0002, "epoch": 2.468239564428312, "step": 3400}, {"loss": 0.8944, "grad_norm": 0.5977247357368469, "learning_rate": 0.0002, "epoch": 2.4754990925589837, "step": 3410}, {"loss": 0.8366, "grad_norm": 0.5004227757453918, "learning_rate": 0.0002, "epoch": 2.4827586206896552, "step": 3420}, {"loss": 0.9133, "grad_norm": 0.46428972482681274, "learning_rate": 0.0002, "epoch": 2.490018148820327, "step": 3430}, {"loss": 0.8534, "grad_norm": 0.7658395171165466, "learning_rate": 0.0002, "epoch": 2.4972776769509983, "step": 3440}, {"loss": 0.9066, "grad_norm": 0.621819794178009, "learning_rate": 0.0002, "epoch": 2.50453720508167, "step": 3450}, {"loss": 0.8481, "grad_norm": 0.4556088149547577, "learning_rate": 0.0002, "epoch": 2.511796733212341, "step": 3460}, {"loss": 0.8522, "grad_norm": 0.6124140024185181, "learning_rate": 0.0002, "epoch": 2.519056261343013, "step": 3470}, {"loss": 0.8158, "grad_norm": 0.6256231665611267, "learning_rate": 0.0002, "epoch": 2.526315789473684, "step": 3480}, {"loss": 0.9097, "grad_norm": 0.5464141964912415, "learning_rate": 0.0002, "epoch": 2.5335753176043556, "step": 3490}, {"loss": 0.8708, "grad_norm": 0.51471346616745, "learning_rate": 0.0002, "epoch": 2.540834845735027, "step": 3500}, {"loss": 0.9042, "grad_norm": 0.5326165556907654, "learning_rate": 0.0002, "epoch": 2.5480943738656987, "step": 3510}, {"loss": 0.8749, "grad_norm": 0.4750378429889679, "learning_rate": 0.0002, "epoch": 2.5553539019963702, "step": 3520}, {"loss": 0.8248, "grad_norm": 0.5292279124259949, "learning_rate": 0.0002, "epoch": 2.562613430127042, "step": 3530}, {"loss": 0.8808, "grad_norm": 0.6145227551460266, "learning_rate": 0.0002, "epoch": 2.5698729582577133, "step": 3540}, {"loss": 0.846, "grad_norm": 0.8275189399719238, "learning_rate": 0.0002, "epoch": 2.577132486388385, "step": 3550}, {"loss": 0.875, "grad_norm": 0.5037438273429871, "learning_rate": 0.0002, "epoch": 2.5843920145190564, "step": 3560}, {"loss": 0.8324, "grad_norm": 0.5838707685470581, "learning_rate": 0.0002, "epoch": 2.5916515426497275, "step": 3570}, {"loss": 0.8716, "grad_norm": 0.5398710370063782, "learning_rate": 0.0002, "epoch": 2.5989110707803995, "step": 3580}, {"loss": 0.7843, "grad_norm": 0.6115376949310303, "learning_rate": 0.0002, "epoch": 2.6061705989110706, "step": 3590}, {"loss": 0.8142, "grad_norm": 1.118809461593628, "learning_rate": 0.0002, "epoch": 2.613430127041742, "step": 3600}, {"loss": 0.8282, "grad_norm": 0.7811771631240845, "learning_rate": 0.0002, "epoch": 2.6206896551724137, "step": 3610}, {"loss": 0.7997, "grad_norm": 0.5753175020217896, "learning_rate": 0.0002, "epoch": 2.6279491833030852, "step": 3620}, {"loss": 0.7705, "grad_norm": 0.550829291343689, "learning_rate": 0.0002, "epoch": 2.635208711433757, "step": 3630}, {"loss": 0.8564, "grad_norm": 0.5360019207000732, "learning_rate": 0.0002, "epoch": 2.6424682395644283, "step": 3640}, {"loss": 0.7648, "grad_norm": 0.63050377368927, "learning_rate": 0.0002, "epoch": 2.6497277676951, "step": 3650}, {"loss": 0.9156, "grad_norm": 0.5833110213279724, "learning_rate": 0.0002, "epoch": 2.6569872958257714, "step": 3660}, {"loss": 0.8033, "grad_norm": 0.5543047189712524, "learning_rate": 0.0002, "epoch": 2.664246823956443, "step": 3670}, {"loss": 0.8867, "grad_norm": 0.6842212080955505, "learning_rate": 0.0002, "epoch": 2.671506352087114, "step": 3680}, {"loss": 0.8569, "grad_norm": 0.6859333515167236, "learning_rate": 0.0002, "epoch": 2.678765880217786, "step": 3690}, {"loss": 0.8081, "grad_norm": 0.7038410902023315, "learning_rate": 0.0002, "epoch": 2.686025408348457, "step": 3700}, {"loss": 0.8686, "grad_norm": 0.447233647108078, "learning_rate": 0.0002, "epoch": 2.6932849364791287, "step": 3710}, {"loss": 0.8951, "grad_norm": 0.6453872919082642, "learning_rate": 0.0002, "epoch": 2.7005444646098002, "step": 3720}, {"loss": 0.8292, "grad_norm": 0.8025672435760498, "learning_rate": 0.0002, "epoch": 2.707803992740472, "step": 3730}, {"loss": 0.8488, "grad_norm": 0.5997087955474854, "learning_rate": 0.0002, "epoch": 2.7150635208711433, "step": 3740}, {"loss": 0.8308, "grad_norm": 0.6901142001152039, "learning_rate": 0.0002, "epoch": 2.722323049001815, "step": 3750}, {"loss": 0.8517, "grad_norm": 1.036145567893982, "learning_rate": 0.0002, "epoch": 2.7295825771324864, "step": 3760}, {"loss": 0.8455, "grad_norm": 0.7207072377204895, "learning_rate": 0.0002, "epoch": 2.736842105263158, "step": 3770}, {"loss": 0.8009, "grad_norm": 1.0452989339828491, "learning_rate": 0.0002, "epoch": 2.7441016333938295, "step": 3780}, {"loss": 0.8868, "grad_norm": 0.5615278482437134, "learning_rate": 0.0002, "epoch": 2.751361161524501, "step": 3790}, {"loss": 0.9047, "grad_norm": 0.46439215540885925, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 3800}, {"loss": 0.7847, "grad_norm": 0.7134917974472046, "learning_rate": 0.0002, "epoch": 2.7658802177858437, "step": 3810}, {"loss": 0.8315, "grad_norm": 0.5139115452766418, "learning_rate": 0.0002, "epoch": 2.7731397459165157, "step": 3820}, {"loss": 0.8995, "grad_norm": 0.8595781326293945, "learning_rate": 0.0002, "epoch": 2.780399274047187, "step": 3830}, {"loss": 0.9079, "grad_norm": 0.544614851474762, "learning_rate": 0.0002, "epoch": 2.7876588021778583, "step": 3840}, {"loss": 0.8322, "grad_norm": 0.6073850393295288, "learning_rate": 0.0002, "epoch": 2.79491833030853, "step": 3850}, {"loss": 0.8277, "grad_norm": 0.8224069476127625, "learning_rate": 0.0002, "epoch": 2.8021778584392014, "step": 3860}, {"loss": 0.815, "grad_norm": 0.5347970128059387, "learning_rate": 0.0002, "epoch": 2.809437386569873, "step": 3870}, {"loss": 0.8219, "grad_norm": 0.6799601912498474, "learning_rate": 0.0002, "epoch": 2.8166969147005445, "step": 3880}, {"loss": 0.8796, "grad_norm": 0.5219197869300842, "learning_rate": 0.0002, "epoch": 2.823956442831216, "step": 3890}, {"loss": 0.8649, "grad_norm": 0.5710130333900452, "learning_rate": 0.0002, "epoch": 2.8312159709618876, "step": 3900}, {"loss": 0.8741, "grad_norm": 0.5857213139533997, "learning_rate": 0.0002, "epoch": 2.838475499092559, "step": 3910}, {"loss": 0.8154, "grad_norm": 0.5206644535064697, "learning_rate": 0.0002, "epoch": 2.8457350272232302, "step": 3920}, {"loss": 0.9328, "grad_norm": 1.2902015447616577, "learning_rate": 0.0002, "epoch": 2.8529945553539022, "step": 3930}, {"loss": 0.8192, "grad_norm": 0.5252797603607178, "learning_rate": 0.0002, "epoch": 2.8602540834845733, "step": 3940}, {"loss": 0.9047, "grad_norm": 0.5925108790397644, "learning_rate": 0.0002, "epoch": 2.867513611615245, "step": 3950}, {"loss": 0.8611, "grad_norm": 0.9719610810279846, "learning_rate": 0.0002, "epoch": 2.8747731397459164, "step": 3960}, {"loss": 0.8531, "grad_norm": 0.6834747195243835, "learning_rate": 0.0002, "epoch": 2.882032667876588, "step": 3970}, {"loss": 0.8124, "grad_norm": 0.6952353119850159, "learning_rate": 0.0002, "epoch": 2.8892921960072595, "step": 3980}, {"loss": 0.8365, "grad_norm": 0.49889910221099854, "learning_rate": 0.0002, "epoch": 2.896551724137931, "step": 3990}, {"loss": 0.8197, "grad_norm": 0.5007026791572571, "learning_rate": 0.0002, "epoch": 2.9038112522686026, "step": 4000}, {"loss": 0.7752, "grad_norm": 0.5474239587783813, "learning_rate": 0.0002, "epoch": 2.911070780399274, "step": 4010}, {"loss": 0.8579, "grad_norm": 0.6617428660392761, "learning_rate": 0.0002, "epoch": 2.9183303085299457, "step": 4020}, {"loss": 0.8583, "grad_norm": 0.6097776293754578, "learning_rate": 0.0002, "epoch": 2.925589836660617, "step": 4030}, {"loss": 0.8296, "grad_norm": 0.5985828638076782, "learning_rate": 0.0002, "epoch": 2.9328493647912888, "step": 4040}, {"loss": 0.9134, "grad_norm": 0.769488513469696, "learning_rate": 0.0002, "epoch": 2.94010889292196, "step": 4050}, {"loss": 0.8148, "grad_norm": 0.5167055130004883, "learning_rate": 0.0002, "epoch": 2.9473684210526314, "step": 4060}, {"loss": 0.8665, "grad_norm": 0.6265496015548706, "learning_rate": 0.0002, "epoch": 2.954627949183303, "step": 4070}, {"loss": 0.9218, "grad_norm": 1.2644082307815552, "learning_rate": 0.0002, "epoch": 2.9618874773139745, "step": 4080}, {"loss": 0.8026, "grad_norm": 0.6007561087608337, "learning_rate": 0.0002, "epoch": 2.969147005444646, "step": 4090}, {"loss": 0.8262, "grad_norm": 0.47984927892684937, "learning_rate": 0.0002, "epoch": 2.9764065335753176, "step": 4100}, {"loss": 0.7897, "grad_norm": 1.128198504447937, "learning_rate": 0.0002, "epoch": 2.983666061705989, "step": 4110}, {"loss": 0.7883, "grad_norm": 0.526292085647583, "learning_rate": 0.0002, "epoch": 2.9909255898366607, "step": 4120}, {"loss": 0.7801, "grad_norm": 0.5462674498558044, "learning_rate": 0.0002, "epoch": 2.9981851179673322, "step": 4130}, {"eval_loss": 1.1093357801437378, "eval_runtime": 46.2498, "eval_samples_per_second": 9.427, "eval_steps_per_second": 1.189, "epoch": 2.9996370235934666, "step": 4132}, {"loss": 0.721, "grad_norm": 0.48288026452064514, "learning_rate": 0.0002, "epoch": 3.0054446460980038, "step": 4140}, {"loss": 0.6769, "grad_norm": 1.0181782245635986, "learning_rate": 0.0002, "epoch": 3.0127041742286753, "step": 4150}, {"loss": 0.7185, "grad_norm": 0.7718019485473633, "learning_rate": 0.0002, "epoch": 3.019963702359347, "step": 4160}, {"loss": 0.6552, "grad_norm": 0.7492219805717468, "learning_rate": 0.0002, "epoch": 3.027223230490018, "step": 4170}, {"loss": 0.6678, "grad_norm": 0.9363632798194885, "learning_rate": 0.0002, "epoch": 3.0344827586206895, "step": 4180}, {"loss": 0.7187, "grad_norm": 0.6888533234596252, "learning_rate": 0.0002, "epoch": 3.041742286751361, "step": 4190}, {"loss": 0.6469, "grad_norm": 0.7072834968566895, "learning_rate": 0.0002, "epoch": 3.0490018148820326, "step": 4200}, {"loss": 0.6387, "grad_norm": 0.7182047963142395, "learning_rate": 0.0002, "epoch": 3.056261343012704, "step": 4210}, {"loss": 0.6385, "grad_norm": 0.7194355130195618, "learning_rate": 0.0002, "epoch": 3.0635208711433757, "step": 4220}, {"loss": 0.5812, "grad_norm": 0.9454023838043213, "learning_rate": 0.0002, "epoch": 3.0707803992740472, "step": 4230}, {"loss": 0.6036, "grad_norm": 0.838657557964325, "learning_rate": 0.0002, "epoch": 3.0780399274047188, "step": 4240}, {"loss": 0.646, "grad_norm": 0.740113377571106, "learning_rate": 0.0002, "epoch": 3.0852994555353903, "step": 4250}, {"loss": 0.604, "grad_norm": 0.6616561412811279, "learning_rate": 0.0002, "epoch": 3.092558983666062, "step": 4260}, {"loss": 0.6462, "grad_norm": 0.8846506476402283, "learning_rate": 0.0002, "epoch": 3.0998185117967334, "step": 4270}, {"loss": 0.6037, "grad_norm": 0.6322125792503357, "learning_rate": 0.0002, "epoch": 3.107078039927405, "step": 4280}, {"loss": 0.5953, "grad_norm": 0.7461467385292053, "learning_rate": 0.0002, "epoch": 3.114337568058076, "step": 4290}, {"loss": 0.6099, "grad_norm": 0.8251287341117859, "learning_rate": 0.0002, "epoch": 3.1215970961887476, "step": 4300}, {"loss": 0.6284, "grad_norm": 0.8767673373222351, "learning_rate": 0.0002, "epoch": 3.128856624319419, "step": 4310}, {"loss": 0.7535, "grad_norm": 0.7758759260177612, "learning_rate": 0.0002, "epoch": 3.1361161524500907, "step": 4320}, {"loss": 0.6624, "grad_norm": 1.1056879758834839, "learning_rate": 0.0002, "epoch": 3.143375680580762, "step": 4330}, {"loss": 0.691, "grad_norm": 0.8259835243225098, "learning_rate": 0.0002, "epoch": 3.1506352087114338, "step": 4340}, {"loss": 0.6635, "grad_norm": 0.6607027053833008, "learning_rate": 0.0002, "epoch": 3.1578947368421053, "step": 4350}, {"loss": 0.5911, "grad_norm": 0.7983301281929016, "learning_rate": 0.0002, "epoch": 3.165154264972777, "step": 4360}, {"loss": 0.6496, "grad_norm": 0.6725239157676697, "learning_rate": 0.0002, "epoch": 3.1724137931034484, "step": 4370}, {"loss": 0.5966, "grad_norm": 0.9052095413208008, "learning_rate": 0.0002, "epoch": 3.17967332123412, "step": 4380}, {"loss": 0.6877, "grad_norm": 0.8131307363510132, "learning_rate": 0.0002, "epoch": 3.1869328493647915, "step": 4390}, {"loss": 0.6384, "grad_norm": 0.6435626149177551, "learning_rate": 0.0002, "epoch": 3.1941923774954626, "step": 4400}, {"loss": 0.5819, "grad_norm": 0.84367436170578, "learning_rate": 0.0002, "epoch": 3.201451905626134, "step": 4410}, {"loss": 0.6104, "grad_norm": 1.5018867254257202, "learning_rate": 0.0002, "epoch": 3.2087114337568057, "step": 4420}, {"loss": 0.6838, "grad_norm": 0.7019091844558716, "learning_rate": 0.0002, "epoch": 3.215970961887477, "step": 4430}, {"loss": 0.6153, "grad_norm": 0.9164197444915771, "learning_rate": 0.0002, "epoch": 3.2232304900181488, "step": 4440}, {"loss": 0.6618, "grad_norm": 0.7890861630439758, "learning_rate": 0.0002, "epoch": 3.2304900181488203, "step": 4450}, {"loss": 0.6401, "grad_norm": 0.6517660617828369, "learning_rate": 0.0002, "epoch": 3.237749546279492, "step": 4460}, {"loss": 0.6699, "grad_norm": 1.10188889503479, "learning_rate": 0.0002, "epoch": 3.2450090744101634, "step": 4470}, {"loss": 0.6356, "grad_norm": 0.8158330917358398, "learning_rate": 0.0002, "epoch": 3.252268602540835, "step": 4480}, {"loss": 0.7757, "grad_norm": 0.7663109302520752, "learning_rate": 0.0002, "epoch": 3.2595281306715065, "step": 4490}, {"loss": 0.6539, "grad_norm": 0.8473444581031799, "learning_rate": 0.0002, "epoch": 3.266787658802178, "step": 4500}, {"loss": 0.6511, "grad_norm": 0.9724768996238708, "learning_rate": 0.0002, "epoch": 3.274047186932849, "step": 4510}, {"loss": 0.5464, "grad_norm": 0.8516759276390076, "learning_rate": 0.0002, "epoch": 3.281306715063521, "step": 4520}, {"loss": 0.6534, "grad_norm": 0.7543437480926514, "learning_rate": 0.0002, "epoch": 3.288566243194192, "step": 4530}, {"loss": 0.6095, "grad_norm": 1.0472029447555542, "learning_rate": 0.0002, "epoch": 3.2958257713248638, "step": 4540}, {"loss": 0.6216, "grad_norm": 0.6240826845169067, "learning_rate": 0.0002, "epoch": 3.3030852994555353, "step": 4550}, {"loss": 0.6223, "grad_norm": 0.9957774877548218, "learning_rate": 0.0002, "epoch": 3.310344827586207, "step": 4560}, {"loss": 0.618, "grad_norm": 0.6448912620544434, "learning_rate": 0.0002, "epoch": 3.3176043557168784, "step": 4570}, {"loss": 0.6188, "grad_norm": 0.7519692778587341, "learning_rate": 0.0002, "epoch": 3.32486388384755, "step": 4580}, {"loss": 0.6672, "grad_norm": 0.7367453575134277, "learning_rate": 0.0002, "epoch": 3.3321234119782215, "step": 4590}, {"loss": 0.6517, "grad_norm": 0.8064960837364197, "learning_rate": 0.0002, "epoch": 3.339382940108893, "step": 4600}, {"loss": 0.6062, "grad_norm": 0.7664631009101868, "learning_rate": 0.0002, "epoch": 3.3466424682395646, "step": 4610}, {"loss": 0.6834, "grad_norm": 0.7803396582603455, "learning_rate": 0.0002, "epoch": 3.353901996370236, "step": 4620}, {"loss": 0.6961, "grad_norm": 0.9141599535942078, "learning_rate": 0.0002, "epoch": 3.3611615245009077, "step": 4630}, {"loss": 0.6889, "grad_norm": 0.9719856381416321, "learning_rate": 0.0002, "epoch": 3.3684210526315788, "step": 4640}, {"loss": 0.6914, "grad_norm": 0.9223218560218811, "learning_rate": 0.0002, "epoch": 3.3756805807622503, "step": 4650}, {"loss": 0.5981, "grad_norm": 0.7289277911186218, "learning_rate": 0.0002, "epoch": 3.382940108892922, "step": 4660}, {"loss": 0.595, "grad_norm": 1.039724349975586, "learning_rate": 0.0002, "epoch": 3.3901996370235934, "step": 4670}, {"loss": 0.8121, "grad_norm": 1.397438883781433, "learning_rate": 0.0002, "epoch": 3.397459165154265, "step": 4680}, {"loss": 0.6334, "grad_norm": 1.0069999694824219, "learning_rate": 0.0002, "epoch": 3.4047186932849365, "step": 4690}, {"loss": 0.6598, "grad_norm": 0.816291332244873, "learning_rate": 0.0002, "epoch": 3.411978221415608, "step": 4700}, {"loss": 0.6748, "grad_norm": 1.2831530570983887, "learning_rate": 0.0002, "epoch": 3.4192377495462796, "step": 4710}, {"loss": 0.6625, "grad_norm": 0.9573889970779419, "learning_rate": 0.0002, "epoch": 3.426497277676951, "step": 4720}, {"loss": 0.7279, "grad_norm": 0.7685632705688477, "learning_rate": 0.0002, "epoch": 3.4337568058076227, "step": 4730}, {"loss": 0.6104, "grad_norm": 0.7019195556640625, "learning_rate": 0.0002, "epoch": 3.441016333938294, "step": 4740}, {"loss": 0.7606, "grad_norm": 0.7244833707809448, "learning_rate": 0.0002, "epoch": 3.4482758620689653, "step": 4750}, {"loss": 0.6951, "grad_norm": 1.3468551635742188, "learning_rate": 0.0002, "epoch": 3.455535390199637, "step": 4760}, {"loss": 0.6945, "grad_norm": 0.822846531867981, "learning_rate": 0.0002, "epoch": 3.4627949183303084, "step": 4770}, {"loss": 0.6431, "grad_norm": 0.7311608195304871, "learning_rate": 0.0002, "epoch": 3.47005444646098, "step": 4780}, {"loss": 0.7019, "grad_norm": 0.9466770887374878, "learning_rate": 0.0002, "epoch": 3.4773139745916515, "step": 4790}, {"loss": 0.7767, "grad_norm": 1.1527155637741089, "learning_rate": 0.0002, "epoch": 3.484573502722323, "step": 4800}, {"loss": 0.6882, "grad_norm": 1.1288906335830688, "learning_rate": 0.0002, "epoch": 3.4918330308529946, "step": 4810}, {"loss": 0.6564, "grad_norm": 0.9096164107322693, "learning_rate": 0.0002, "epoch": 3.499092558983666, "step": 4820}, {"loss": 0.6127, "grad_norm": 0.7988565564155579, "learning_rate": 0.0002, "epoch": 3.5063520871143377, "step": 4830}, {"loss": 0.7004, "grad_norm": 0.7183415293693542, "learning_rate": 0.0002, "epoch": 3.513611615245009, "step": 4840}, {"loss": 0.74, "grad_norm": 0.6614915132522583, "learning_rate": 0.0002, "epoch": 3.5208711433756807, "step": 4850}, {"loss": 0.7271, "grad_norm": 0.8609521985054016, "learning_rate": 0.0002, "epoch": 3.528130671506352, "step": 4860}, {"loss": 0.6664, "grad_norm": 0.86552894115448, "learning_rate": 0.0002, "epoch": 3.535390199637024, "step": 4870}, {"loss": 0.6432, "grad_norm": 0.6926496028900146, "learning_rate": 0.0002, "epoch": 3.542649727767695, "step": 4880}, {"loss": 0.7117, "grad_norm": 0.8157467246055603, "learning_rate": 0.0002, "epoch": 3.5499092558983665, "step": 4890}, {"loss": 0.6201, "grad_norm": 0.9085357189178467, "learning_rate": 0.0002, "epoch": 3.557168784029038, "step": 4900}, {"loss": 0.6521, "grad_norm": 0.6322644948959351, "learning_rate": 0.0002, "epoch": 3.5644283121597096, "step": 4910}, {"loss": 0.6607, "grad_norm": 1.263205885887146, "learning_rate": 0.0002, "epoch": 3.571687840290381, "step": 4920}, {"loss": 0.6657, "grad_norm": 0.8901070356369019, "learning_rate": 0.0002, "epoch": 3.5789473684210527, "step": 4930}, {"loss": 0.6434, "grad_norm": 0.7983952164649963, "learning_rate": 0.0002, "epoch": 3.586206896551724, "step": 4940}, {"loss": 0.6861, "grad_norm": 0.9887813925743103, "learning_rate": 0.0002, "epoch": 3.5934664246823957, "step": 4950}, {"loss": 0.6502, "grad_norm": 0.7895187735557556, "learning_rate": 0.0002, "epoch": 3.6007259528130673, "step": 4960}, {"loss": 0.7111, "grad_norm": 0.9685819745063782, "learning_rate": 0.0002, "epoch": 3.6079854809437384, "step": 4970}, {"loss": 0.6915, "grad_norm": 0.6576591730117798, "learning_rate": 0.0002, "epoch": 3.6152450090744104, "step": 4980}, {"loss": 0.6195, "grad_norm": 0.856985330581665, "learning_rate": 0.0002, "epoch": 3.6225045372050815, "step": 4990}, {"loss": 0.6318, "grad_norm": 0.7230252623558044, "learning_rate": 0.0002, "epoch": 3.629764065335753, "step": 5000}, {"loss": 0.742, "grad_norm": 0.8260893821716309, "learning_rate": 0.0002, "epoch": 3.6370235934664246, "step": 5010}, {"loss": 0.7223, "grad_norm": 0.7635950446128845, "learning_rate": 0.0002, "epoch": 3.644283121597096, "step": 5020}, {"loss": 0.6837, "grad_norm": 0.7060768604278564, "learning_rate": 0.0002, "epoch": 3.6515426497277677, "step": 5030}, {"loss": 0.6921, "grad_norm": 0.8020303249359131, "learning_rate": 0.0002, "epoch": 3.658802177858439, "step": 5040}, {"loss": 0.6446, "grad_norm": 0.8530341386795044, "learning_rate": 0.0002, "epoch": 3.6660617059891107, "step": 5050}, {"loss": 0.7222, "grad_norm": 0.6667101979255676, "learning_rate": 0.0002, "epoch": 3.6733212341197823, "step": 5060}, {"loss": 0.7081, "grad_norm": 0.7385406494140625, "learning_rate": 0.0002, "epoch": 3.680580762250454, "step": 5070}, {"loss": 0.7035, "grad_norm": 0.7753380537033081, "learning_rate": 0.0002, "epoch": 3.6878402903811254, "step": 5080}, {"loss": 0.6491, "grad_norm": 0.7516207098960876, "learning_rate": 0.0002, "epoch": 3.695099818511797, "step": 5090}, {"loss": 0.672, "grad_norm": 0.8171586394309998, "learning_rate": 0.0002, "epoch": 3.702359346642468, "step": 5100}, {"loss": 0.7459, "grad_norm": 1.0796279907226562, "learning_rate": 0.0002, "epoch": 3.70961887477314, "step": 5110}, {"loss": 0.5948, "grad_norm": 0.6957688927650452, "learning_rate": 0.0002, "epoch": 3.716878402903811, "step": 5120}, {"loss": 0.7515, "grad_norm": 0.8550161719322205, "learning_rate": 0.0002, "epoch": 3.7241379310344827, "step": 5130}, {"loss": 0.7286, "grad_norm": 0.9396728277206421, "learning_rate": 0.0002, "epoch": 3.731397459165154, "step": 5140}, {"loss": 0.7594, "grad_norm": 1.4264805316925049, "learning_rate": 0.0002, "epoch": 3.7386569872958257, "step": 5150}, {"loss": 0.6575, "grad_norm": 0.8725108504295349, "learning_rate": 0.0002, "epoch": 3.7459165154264973, "step": 5160}, {"loss": 0.6778, "grad_norm": 1.0346195697784424, "learning_rate": 0.0002, "epoch": 3.753176043557169, "step": 5170}, {"loss": 0.6371, "grad_norm": 0.5395554304122925, "learning_rate": 0.0002, "epoch": 3.7604355716878404, "step": 5180}, {"loss": 0.7308, "grad_norm": 1.3153616189956665, "learning_rate": 0.0002, "epoch": 3.767695099818512, "step": 5190}, {"loss": 0.78, "grad_norm": 0.9879828691482544, "learning_rate": 0.0002, "epoch": 3.7749546279491835, "step": 5200}, {"loss": 0.7068, "grad_norm": 0.8876672983169556, "learning_rate": 0.0002, "epoch": 3.7822141560798546, "step": 5210}, {"loss": 0.6283, "grad_norm": 0.8363267779350281, "learning_rate": 0.0002, "epoch": 3.7894736842105265, "step": 5220}, {"loss": 0.6255, "grad_norm": 0.637294590473175, "learning_rate": 0.0002, "epoch": 3.7967332123411976, "step": 5230}, {"loss": 0.6685, "grad_norm": 1.1408970355987549, "learning_rate": 0.0002, "epoch": 3.803992740471869, "step": 5240}, {"loss": 0.6761, "grad_norm": 1.0128360986709595, "learning_rate": 0.0002, "epoch": 3.8112522686025407, "step": 5250}, {"loss": 0.6764, "grad_norm": 0.8061144351959229, "learning_rate": 0.0002, "epoch": 3.8185117967332123, "step": 5260}, {"loss": 0.7254, "grad_norm": 0.9626626968383789, "learning_rate": 0.0002, "epoch": 3.825771324863884, "step": 5270}, {"loss": 0.7367, "grad_norm": 0.9013627171516418, "learning_rate": 0.0002, "epoch": 3.8330308529945554, "step": 5280}, {"loss": 0.6806, "grad_norm": 0.8411344289779663, "learning_rate": 0.0002, "epoch": 3.840290381125227, "step": 5290}, {"loss": 0.6818, "grad_norm": 0.7426059246063232, "learning_rate": 0.0002, "epoch": 3.8475499092558985, "step": 5300}, {"loss": 0.6748, "grad_norm": 1.003413438796997, "learning_rate": 0.0002, "epoch": 3.85480943738657, "step": 5310}, {"loss": 0.8554, "grad_norm": 0.7527840733528137, "learning_rate": 0.0002, "epoch": 3.862068965517241, "step": 5320}, {"loss": 0.7521, "grad_norm": 0.738610565662384, "learning_rate": 0.0002, "epoch": 3.869328493647913, "step": 5330}, {"loss": 0.7266, "grad_norm": 0.7277999520301819, "learning_rate": 0.0002, "epoch": 3.876588021778584, "step": 5340}, {"loss": 0.7503, "grad_norm": 0.5951359272003174, "learning_rate": 0.0002, "epoch": 3.8838475499092557, "step": 5350}, {"loss": 0.7447, "grad_norm": 1.043884038925171, "learning_rate": 0.0002, "epoch": 3.8911070780399273, "step": 5360}, {"loss": 0.6862, "grad_norm": 0.8436498045921326, "learning_rate": 0.0002, "epoch": 3.898366606170599, "step": 5370}, {"loss": 0.665, "grad_norm": 0.5603365302085876, "learning_rate": 0.0002, "epoch": 3.9056261343012704, "step": 5380}, {"loss": 0.7098, "grad_norm": 1.0128886699676514, "learning_rate": 0.0002, "epoch": 3.912885662431942, "step": 5390}, {"loss": 0.6707, "grad_norm": 0.7970930337905884, "learning_rate": 0.0002, "epoch": 3.9201451905626135, "step": 5400}, {"loss": 0.637, "grad_norm": 0.7699369192123413, "learning_rate": 0.0002, "epoch": 3.927404718693285, "step": 5410}, {"loss": 0.6742, "grad_norm": 0.800561249256134, "learning_rate": 0.0002, "epoch": 3.9346642468239565, "step": 5420}, {"loss": 0.7208, "grad_norm": 0.8020331859588623, "learning_rate": 0.0002, "epoch": 3.941923774954628, "step": 5430}, {"loss": 0.7294, "grad_norm": 0.7461140155792236, "learning_rate": 0.0002, "epoch": 3.9491833030852996, "step": 5440}, {"loss": 0.7013, "grad_norm": 0.8346918821334839, "learning_rate": 0.0002, "epoch": 3.9564428312159707, "step": 5450}, {"loss": 0.6289, "grad_norm": 0.9723302125930786, "learning_rate": 0.0002, "epoch": 3.9637023593466427, "step": 5460}, {"loss": 0.8029, "grad_norm": 0.6809740662574768, "learning_rate": 0.0002, "epoch": 3.970961887477314, "step": 5470}, {"loss": 0.6896, "grad_norm": 0.7353498339653015, "learning_rate": 0.0002, "epoch": 3.9782214156079854, "step": 5480}, {"loss": 0.6722, "grad_norm": 0.748009443283081, "learning_rate": 0.0002, "epoch": 3.985480943738657, "step": 5490}, {"loss": 0.6866, "grad_norm": 1.3656195402145386, "learning_rate": 0.0002, "epoch": 3.9927404718693285, "step": 5500}, {"loss": 0.7368, "grad_norm": 0.8402108550071716, "learning_rate": 0.0002, "epoch": 4.0, "step": 5510}]}
+{"epoch": 4.999637023593467, "step": 6887, "epoch_duration": 2059.8379588127136, "total_accumulated_duration": 16346.150188922882, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4936, "grad_norm": 0.46010470390319824, "learning_rate": 0.0002, "epoch": 0.007259528130671506, "step": 10}, {"loss": 1.8442, "grad_norm": 0.5103179216384888, "learning_rate": 0.0002, "epoch": 0.014519056261343012, "step": 20}, {"loss": 1.559, "grad_norm": 0.6282716989517212, "learning_rate": 0.0002, "epoch": 0.021778584392014518, "step": 30}, {"loss": 1.3618, "grad_norm": 1.2480497360229492, "learning_rate": 0.0002, "epoch": 0.029038112522686024, "step": 40}, {"loss": 1.3874, "grad_norm": 0.4114855229854584, "learning_rate": 0.0002, "epoch": 0.036297640653357534, "step": 50}, {"loss": 1.2836, "grad_norm": 0.49482840299606323, "learning_rate": 0.0002, "epoch": 0.043557168784029036, "step": 60}, {"loss": 1.1728, "grad_norm": 0.4536272883415222, "learning_rate": 0.0002, "epoch": 0.050816696914700546, "step": 70}, {"loss": 1.1125, "grad_norm": 0.32328274846076965, "learning_rate": 0.0002, "epoch": 0.05807622504537205, "step": 80}, {"loss": 1.2447, "grad_norm": 0.40990468859672546, "learning_rate": 0.0002, "epoch": 0.06533575317604355, "step": 90}, {"loss": 1.2216, "grad_norm": 0.37273502349853516, "learning_rate": 0.0002, "epoch": 0.07259528130671507, "step": 100}, {"loss": 1.2259, "grad_norm": 0.3903871476650238, "learning_rate": 0.0002, "epoch": 0.07985480943738657, "step": 110}, {"loss": 1.3718, "grad_norm": 0.3272787928581238, "learning_rate": 0.0002, "epoch": 0.08711433756805807, "step": 120}, {"loss": 1.1832, "grad_norm": 0.3622824251651764, "learning_rate": 0.0002, "epoch": 0.09437386569872959, "step": 130}, {"loss": 1.2966, "grad_norm": 0.3503916561603546, "learning_rate": 0.0002, "epoch": 0.10163339382940109, "step": 140}, {"loss": 1.0913, "grad_norm": 0.32787832617759705, "learning_rate": 0.0002, "epoch": 0.1088929219600726, "step": 150}, {"loss": 1.2143, "grad_norm": 0.5822657942771912, "learning_rate": 0.0002, "epoch": 0.1161524500907441, "step": 160}, {"loss": 1.2091, "grad_norm": 0.28028249740600586, "learning_rate": 0.0002, "epoch": 0.12341197822141561, "step": 170}, {"loss": 1.1731, "grad_norm": 0.33602750301361084, "learning_rate": 0.0002, "epoch": 0.1306715063520871, "step": 180}, {"loss": 1.127, "grad_norm": 0.29106274247169495, "learning_rate": 0.0002, "epoch": 0.13793103448275862, "step": 190}, {"loss": 1.3261, "grad_norm": 0.38753363490104675, "learning_rate": 0.0002, "epoch": 0.14519056261343014, "step": 200}, {"loss": 1.1536, "grad_norm": 0.361009418964386, "learning_rate": 0.0002, "epoch": 0.15245009074410162, "step": 210}, {"loss": 1.1483, "grad_norm": 0.6743836402893066, "learning_rate": 0.0002, "epoch": 0.15970961887477314, "step": 220}, {"loss": 1.264, "grad_norm": 0.3923613727092743, "learning_rate": 0.0002, "epoch": 0.16696914700544466, "step": 230}, {"loss": 1.0437, "grad_norm": 0.2809699773788452, "learning_rate": 0.0002, "epoch": 0.17422867513611615, "step": 240}, {"loss": 1.1826, "grad_norm": 0.3631494641304016, "learning_rate": 0.0002, "epoch": 0.18148820326678766, "step": 250}, {"loss": 1.1584, "grad_norm": 0.24658171832561493, "learning_rate": 0.0002, "epoch": 0.18874773139745918, "step": 260}, {"loss": 1.0065, "grad_norm": 0.5780664682388306, "learning_rate": 0.0002, "epoch": 0.19600725952813067, "step": 270}, {"loss": 1.1251, "grad_norm": 0.3056720495223999, "learning_rate": 0.0002, "epoch": 0.20326678765880218, "step": 280}, {"loss": 1.1503, "grad_norm": 0.2857084572315216, "learning_rate": 0.0002, "epoch": 0.21052631578947367, "step": 290}, {"loss": 1.1215, "grad_norm": 0.6645345687866211, "learning_rate": 0.0002, "epoch": 0.2177858439201452, "step": 300}, {"loss": 1.127, "grad_norm": 0.5966078639030457, "learning_rate": 0.0002, "epoch": 0.2250453720508167, "step": 310}, {"loss": 1.1974, "grad_norm": 0.40937140583992004, "learning_rate": 0.0002, "epoch": 0.2323049001814882, "step": 320}, {"loss": 1.2385, "grad_norm": 0.5642806887626648, "learning_rate": 0.0002, "epoch": 0.2395644283121597, "step": 330}, {"loss": 1.1703, "grad_norm": 0.2750748097896576, "learning_rate": 0.0002, "epoch": 0.24682395644283123, "step": 340}, {"loss": 1.2598, "grad_norm": 0.34350234270095825, "learning_rate": 0.0002, "epoch": 0.2540834845735027, "step": 350}, {"loss": 1.1942, "grad_norm": 0.6767239570617676, "learning_rate": 0.0002, "epoch": 0.2613430127041742, "step": 360}, {"loss": 1.1436, "grad_norm": 0.31006959080696106, "learning_rate": 0.0002, "epoch": 0.26860254083484575, "step": 370}, {"loss": 1.2084, "grad_norm": 0.3825474679470062, "learning_rate": 0.0002, "epoch": 0.27586206896551724, "step": 380}, {"loss": 1.1523, "grad_norm": 0.30241551995277405, "learning_rate": 0.0002, "epoch": 0.2831215970961887, "step": 390}, {"loss": 1.1298, "grad_norm": 0.2962397336959839, "learning_rate": 0.0002, "epoch": 0.29038112522686027, "step": 400}, {"loss": 1.1299, "grad_norm": 0.2600369155406952, "learning_rate": 0.0002, "epoch": 0.29764065335753176, "step": 410}, {"loss": 1.1366, "grad_norm": 0.3675060272216797, "learning_rate": 0.0002, "epoch": 0.30490018148820325, "step": 420}, {"loss": 1.156, "grad_norm": 0.3429498076438904, "learning_rate": 0.0002, "epoch": 0.3121597096188748, "step": 430}, {"loss": 1.2741, "grad_norm": 0.34311825037002563, "learning_rate": 0.0002, "epoch": 0.3194192377495463, "step": 440}, {"loss": 1.3523, "grad_norm": 0.37872210144996643, "learning_rate": 0.0002, "epoch": 0.32667876588021777, "step": 450}, {"loss": 1.1365, "grad_norm": 0.33271121978759766, "learning_rate": 0.0002, "epoch": 0.3339382940108893, "step": 460}, {"loss": 1.1572, "grad_norm": 0.34605276584625244, "learning_rate": 0.0002, "epoch": 0.3411978221415608, "step": 470}, {"loss": 1.2755, "grad_norm": 0.41050562262535095, "learning_rate": 0.0002, "epoch": 0.3484573502722323, "step": 480}, {"loss": 1.075, "grad_norm": 0.2066836953163147, "learning_rate": 0.0002, "epoch": 0.35571687840290384, "step": 490}, {"loss": 1.2072, "grad_norm": 0.2859014868736267, "learning_rate": 0.0002, "epoch": 0.3629764065335753, "step": 500}, {"loss": 1.1435, "grad_norm": 0.28763777017593384, "learning_rate": 0.0002, "epoch": 0.3702359346642468, "step": 510}, {"loss": 1.1341, "grad_norm": 0.2730471193790436, "learning_rate": 0.0002, "epoch": 0.37749546279491836, "step": 520}, {"loss": 1.348, "grad_norm": 0.3968936800956726, "learning_rate": 0.0002, "epoch": 0.38475499092558985, "step": 530}, {"loss": 1.2363, "grad_norm": 0.3624701201915741, "learning_rate": 0.0002, "epoch": 0.39201451905626133, "step": 540}, {"loss": 1.1206, "grad_norm": 0.3303608298301697, "learning_rate": 0.0002, "epoch": 0.3992740471869328, "step": 550}, {"loss": 1.203, "grad_norm": 0.33507466316223145, "learning_rate": 0.0002, "epoch": 0.40653357531760437, "step": 560}, {"loss": 1.2294, "grad_norm": 0.3297670781612396, "learning_rate": 0.0002, "epoch": 0.41379310344827586, "step": 570}, {"loss": 1.2, "grad_norm": 0.32334890961647034, "learning_rate": 0.0002, "epoch": 0.42105263157894735, "step": 580}, {"loss": 1.112, "grad_norm": 0.30281195044517517, "learning_rate": 0.0002, "epoch": 0.4283121597096189, "step": 590}, {"loss": 1.1352, "grad_norm": 0.5900027751922607, "learning_rate": 0.0002, "epoch": 0.4355716878402904, "step": 600}, {"loss": 1.1575, "grad_norm": 0.28437477350234985, "learning_rate": 0.0002, "epoch": 0.44283121597096187, "step": 610}, {"loss": 1.2009, "grad_norm": 0.39601704478263855, "learning_rate": 0.0002, "epoch": 0.4500907441016334, "step": 620}, {"loss": 1.2302, "grad_norm": 0.41971510648727417, "learning_rate": 0.0002, "epoch": 0.4573502722323049, "step": 630}, {"loss": 1.236, "grad_norm": 0.33814409375190735, "learning_rate": 0.0002, "epoch": 0.4646098003629764, "step": 640}, {"loss": 1.2471, "grad_norm": 0.575718104839325, "learning_rate": 0.0002, "epoch": 0.47186932849364793, "step": 650}, {"loss": 1.1012, "grad_norm": 0.37927401065826416, "learning_rate": 0.0002, "epoch": 0.4791288566243194, "step": 660}, {"loss": 1.1552, "grad_norm": 0.3224332630634308, "learning_rate": 0.0002, "epoch": 0.4863883847549909, "step": 670}, {"loss": 1.1354, "grad_norm": 0.32683515548706055, "learning_rate": 0.0002, "epoch": 0.49364791288566245, "step": 680}, {"loss": 1.207, "grad_norm": 0.4316163659095764, "learning_rate": 0.0002, "epoch": 0.5009074410163339, "step": 690}, {"loss": 1.2116, "grad_norm": 0.342602401971817, "learning_rate": 0.0002, "epoch": 0.5081669691470054, "step": 700}, {"loss": 1.0433, "grad_norm": 0.2794898748397827, "learning_rate": 0.0002, "epoch": 0.515426497277677, "step": 710}, {"loss": 1.2395, "grad_norm": 0.3322339951992035, "learning_rate": 0.0002, "epoch": 0.5226860254083484, "step": 720}, {"loss": 1.0762, "grad_norm": 0.3088509142398834, "learning_rate": 0.0002, "epoch": 0.52994555353902, "step": 730}, {"loss": 1.1495, "grad_norm": 0.24444378912448883, "learning_rate": 0.0002, "epoch": 0.5372050816696915, "step": 740}, {"loss": 1.2176, "grad_norm": 0.3483171761035919, "learning_rate": 0.0002, "epoch": 0.5444646098003629, "step": 750}, {"loss": 1.2248, "grad_norm": 0.3425690531730652, "learning_rate": 0.0002, "epoch": 0.5517241379310345, "step": 760}, {"loss": 1.1127, "grad_norm": 0.31841927766799927, "learning_rate": 0.0002, "epoch": 0.558983666061706, "step": 770}, {"loss": 1.1841, "grad_norm": 0.39423868060112, "learning_rate": 0.0002, "epoch": 0.5662431941923775, "step": 780}, {"loss": 1.0583, "grad_norm": 0.30328479409217834, "learning_rate": 0.0002, "epoch": 0.573502722323049, "step": 790}, {"loss": 1.0718, "grad_norm": 0.24475938081741333, "learning_rate": 0.0002, "epoch": 0.5807622504537205, "step": 800}, {"loss": 1.2006, "grad_norm": 0.37132805585861206, "learning_rate": 0.0002, "epoch": 0.588021778584392, "step": 810}, {"loss": 1.1544, "grad_norm": 0.32195979356765747, "learning_rate": 0.0002, "epoch": 0.5952813067150635, "step": 820}, {"loss": 0.9937, "grad_norm": 0.2848738729953766, "learning_rate": 0.0002, "epoch": 0.6025408348457351, "step": 830}, {"loss": 1.0937, "grad_norm": 0.28015264868736267, "learning_rate": 0.0002, "epoch": 0.6098003629764065, "step": 840}, {"loss": 1.1234, "grad_norm": 0.37796008586883545, "learning_rate": 0.0002, "epoch": 0.617059891107078, "step": 850}, {"loss": 1.1455, "grad_norm": 0.39311841130256653, "learning_rate": 0.0002, "epoch": 0.6243194192377496, "step": 860}, {"loss": 1.1066, "grad_norm": 0.2761685252189636, "learning_rate": 0.0002, "epoch": 0.631578947368421, "step": 870}, {"loss": 1.0859, "grad_norm": 0.3826720118522644, "learning_rate": 0.0002, "epoch": 0.6388384754990926, "step": 880}, {"loss": 1.1182, "grad_norm": 0.30076679587364197, "learning_rate": 0.0002, "epoch": 0.6460980036297641, "step": 890}, {"loss": 1.0927, "grad_norm": 0.21997687220573425, "learning_rate": 0.0002, "epoch": 0.6533575317604355, "step": 900}, {"loss": 1.1603, "grad_norm": 0.32593777775764465, "learning_rate": 0.0002, "epoch": 0.6606170598911071, "step": 910}, {"loss": 1.1578, "grad_norm": 0.30347898602485657, "learning_rate": 0.0002, "epoch": 0.6678765880217786, "step": 920}, {"loss": 1.1642, "grad_norm": 0.44173774123191833, "learning_rate": 0.0002, "epoch": 0.6751361161524501, "step": 930}, {"loss": 1.1832, "grad_norm": 0.2507467269897461, "learning_rate": 0.0002, "epoch": 0.6823956442831216, "step": 940}, {"loss": 1.3001, "grad_norm": 0.29463833570480347, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 950}, {"loss": 1.1651, "grad_norm": 0.9363154172897339, "learning_rate": 0.0002, "epoch": 0.6969147005444646, "step": 960}, {"loss": 1.158, "grad_norm": 0.3236212134361267, "learning_rate": 0.0002, "epoch": 0.7041742286751361, "step": 970}, {"loss": 1.1078, "grad_norm": 0.3123254179954529, "learning_rate": 0.0002, "epoch": 0.7114337568058077, "step": 980}, {"loss": 1.1578, "grad_norm": 0.3395805060863495, "learning_rate": 0.0002, "epoch": 0.7186932849364791, "step": 990}, {"loss": 1.0754, "grad_norm": 0.3240964412689209, "learning_rate": 0.0002, "epoch": 0.7259528130671506, "step": 1000}, {"loss": 1.0313, "grad_norm": 0.31902948021888733, "learning_rate": 0.0002, "epoch": 0.7332123411978222, "step": 1010}, {"loss": 1.1664, "grad_norm": 0.4848408102989197, "learning_rate": 0.0002, "epoch": 0.7404718693284936, "step": 1020}, {"loss": 1.0838, "grad_norm": 0.33006033301353455, "learning_rate": 0.0002, "epoch": 0.7477313974591652, "step": 1030}, {"loss": 1.0979, "grad_norm": 0.2928730547428131, "learning_rate": 0.0002, "epoch": 0.7549909255898367, "step": 1040}, {"loss": 1.1351, "grad_norm": 0.3529164791107178, "learning_rate": 0.0002, "epoch": 0.7622504537205081, "step": 1050}, {"loss": 1.1274, "grad_norm": 0.2736213803291321, "learning_rate": 0.0002, "epoch": 0.7695099818511797, "step": 1060}, {"loss": 1.1169, "grad_norm": 0.7200686931610107, "learning_rate": 0.0002, "epoch": 0.7767695099818511, "step": 1070}, {"loss": 1.1171, "grad_norm": 0.33396708965301514, "learning_rate": 0.0002, "epoch": 0.7840290381125227, "step": 1080}, {"loss": 1.2447, "grad_norm": 1.5760449171066284, "learning_rate": 0.0002, "epoch": 0.7912885662431942, "step": 1090}, {"loss": 1.1126, "grad_norm": 0.28138381242752075, "learning_rate": 0.0002, "epoch": 0.7985480943738656, "step": 1100}, {"loss": 1.2012, "grad_norm": 0.2597472369670868, "learning_rate": 0.0002, "epoch": 0.8058076225045372, "step": 1110}, {"loss": 1.1177, "grad_norm": 0.3305445611476898, "learning_rate": 0.0002, "epoch": 0.8130671506352087, "step": 1120}, {"loss": 1.0849, "grad_norm": 0.3934599459171295, "learning_rate": 0.0002, "epoch": 0.8203266787658802, "step": 1130}, {"loss": 1.0889, "grad_norm": 0.3472191393375397, "learning_rate": 0.0002, "epoch": 0.8275862068965517, "step": 1140}, {"loss": 1.265, "grad_norm": 0.2857365906238556, "learning_rate": 0.0002, "epoch": 0.8348457350272233, "step": 1150}, {"loss": 1.03, "grad_norm": 0.3207702934741974, "learning_rate": 0.0002, "epoch": 0.8421052631578947, "step": 1160}, {"loss": 1.1669, "grad_norm": 0.3176484704017639, "learning_rate": 0.0002, "epoch": 0.8493647912885662, "step": 1170}, {"loss": 1.1386, "grad_norm": 0.40685558319091797, "learning_rate": 0.0002, "epoch": 0.8566243194192378, "step": 1180}, {"loss": 1.1383, "grad_norm": 0.31125199794769287, "learning_rate": 0.0002, "epoch": 0.8638838475499092, "step": 1190}, {"loss": 1.1512, "grad_norm": 0.7361181378364563, "learning_rate": 0.0002, "epoch": 0.8711433756805808, "step": 1200}, {"loss": 0.9983, "grad_norm": 0.33699527382850647, "learning_rate": 0.0002, "epoch": 0.8784029038112523, "step": 1210}, {"loss": 1.11, "grad_norm": 0.3315220773220062, "learning_rate": 0.0002, "epoch": 0.8856624319419237, "step": 1220}, {"loss": 1.0266, "grad_norm": 0.6256054043769836, "learning_rate": 0.0002, "epoch": 0.8929219600725953, "step": 1230}, {"loss": 1.1738, "grad_norm": 0.3692137897014618, "learning_rate": 0.0002, "epoch": 0.9001814882032668, "step": 1240}, {"loss": 1.1112, "grad_norm": 0.3538484573364258, "learning_rate": 0.0002, "epoch": 0.9074410163339383, "step": 1250}, {"loss": 1.0878, "grad_norm": 0.27863040566444397, "learning_rate": 0.0002, "epoch": 0.9147005444646098, "step": 1260}, {"loss": 1.0826, "grad_norm": 0.3322528302669525, "learning_rate": 0.0002, "epoch": 0.9219600725952813, "step": 1270}, {"loss": 1.0158, "grad_norm": 0.7553173303604126, "learning_rate": 0.0002, "epoch": 0.9292196007259528, "step": 1280}, {"loss": 1.198, "grad_norm": 0.4856191575527191, "learning_rate": 0.0002, "epoch": 0.9364791288566243, "step": 1290}, {"loss": 1.1678, "grad_norm": 0.3668074905872345, "learning_rate": 0.0002, "epoch": 0.9437386569872959, "step": 1300}, {"loss": 1.1809, "grad_norm": 0.29851067066192627, "learning_rate": 0.0002, "epoch": 0.9509981851179673, "step": 1310}, {"loss": 1.1299, "grad_norm": 0.276664674282074, "learning_rate": 0.0002, "epoch": 0.9582577132486388, "step": 1320}, {"loss": 1.0597, "grad_norm": 0.2941018342971802, "learning_rate": 0.0002, "epoch": 0.9655172413793104, "step": 1330}, {"loss": 1.1731, "grad_norm": 0.3505859076976776, "learning_rate": 0.0002, "epoch": 0.9727767695099818, "step": 1340}, {"loss": 1.1809, "grad_norm": 0.3067687451839447, "learning_rate": 0.0002, "epoch": 0.9800362976406534, "step": 1350}, {"loss": 1.1511, "grad_norm": 0.27151066064834595, "learning_rate": 0.0002, "epoch": 0.9872958257713249, "step": 1360}, {"loss": 1.1919, "grad_norm": 0.36370083689689636, "learning_rate": 0.0002, "epoch": 0.9945553539019963, "step": 1370}, {"eval_loss": 1.1381088495254517, "eval_runtime": 96.0848, "eval_samples_per_second": 4.538, "eval_steps_per_second": 0.572, "epoch": 0.9996370235934664, "step": 1377}, {"loss": 1.0082, "grad_norm": 0.27980583906173706, "learning_rate": 0.0002, "epoch": 1.0018148820326678, "step": 1380}, {"loss": 1.0565, "grad_norm": 0.26713913679122925, "learning_rate": 0.0002, "epoch": 1.0090744101633393, "step": 1390}, {"loss": 1.0241, "grad_norm": 0.3089541494846344, "learning_rate": 0.0002, "epoch": 1.0163339382940109, "step": 1400}, {"loss": 0.9649, "grad_norm": 0.4188242256641388, "learning_rate": 0.0002, "epoch": 1.0235934664246824, "step": 1410}, {"loss": 0.9821, "grad_norm": 0.5246463418006897, "learning_rate": 0.0002, "epoch": 1.030852994555354, "step": 1420}, {"loss": 1.0143, "grad_norm": 0.2728777825832367, "learning_rate": 0.0002, "epoch": 1.0381125226860255, "step": 1430}, {"loss": 1.075, "grad_norm": 0.38167616724967957, "learning_rate": 0.0002, "epoch": 1.0453720508166968, "step": 1440}, {"loss": 1.0229, "grad_norm": 0.4439380168914795, "learning_rate": 0.0002, "epoch": 1.0526315789473684, "step": 1450}, {"loss": 1.0451, "grad_norm": 0.30954182147979736, "learning_rate": 0.0002, "epoch": 1.05989110707804, "step": 1460}, {"loss": 1.0762, "grad_norm": 0.4022280275821686, "learning_rate": 0.0002, "epoch": 1.0671506352087115, "step": 1470}, {"loss": 1.0754, "grad_norm": 0.7390811443328857, "learning_rate": 0.0002, "epoch": 1.074410163339383, "step": 1480}, {"loss": 1.1378, "grad_norm": 0.3885486423969269, "learning_rate": 0.0002, "epoch": 1.0816696914700545, "step": 1490}, {"loss": 1.005, "grad_norm": 0.5275560617446899, "learning_rate": 0.0002, "epoch": 1.0889292196007259, "step": 1500}, {"loss": 1.0147, "grad_norm": 0.35112282633781433, "learning_rate": 0.0002, "epoch": 1.0961887477313974, "step": 1510}, {"loss": 1.0331, "grad_norm": 0.33714351058006287, "learning_rate": 0.0002, "epoch": 1.103448275862069, "step": 1520}, {"loss": 0.8723, "grad_norm": 0.31221693754196167, "learning_rate": 0.0002, "epoch": 1.1107078039927405, "step": 1530}, {"loss": 1.0701, "grad_norm": 0.27549654245376587, "learning_rate": 0.0002, "epoch": 1.117967332123412, "step": 1540}, {"loss": 1.0114, "grad_norm": 0.8465521335601807, "learning_rate": 0.0002, "epoch": 1.1252268602540836, "step": 1550}, {"loss": 0.9643, "grad_norm": 0.36125949025154114, "learning_rate": 0.0002, "epoch": 1.132486388384755, "step": 1560}, {"loss": 1.0744, "grad_norm": 0.37420371174812317, "learning_rate": 0.0002, "epoch": 1.1397459165154264, "step": 1570}, {"loss": 1.165, "grad_norm": 0.3294760584831238, "learning_rate": 0.0002, "epoch": 1.147005444646098, "step": 1580}, {"loss": 0.9207, "grad_norm": 0.3881238102912903, "learning_rate": 0.0002, "epoch": 1.1542649727767695, "step": 1590}, {"loss": 1.0983, "grad_norm": 0.4766491651535034, "learning_rate": 0.0002, "epoch": 1.161524500907441, "step": 1600}, {"loss": 1.0092, "grad_norm": 0.395530104637146, "learning_rate": 0.0002, "epoch": 1.1687840290381124, "step": 1610}, {"loss": 0.9895, "grad_norm": 0.3297106623649597, "learning_rate": 0.0002, "epoch": 1.176043557168784, "step": 1620}, {"loss": 1.1121, "grad_norm": 0.39528271555900574, "learning_rate": 0.0002, "epoch": 1.1833030852994555, "step": 1630}, {"loss": 0.9202, "grad_norm": 0.3370221257209778, "learning_rate": 0.0002, "epoch": 1.190562613430127, "step": 1640}, {"loss": 1.102, "grad_norm": 0.31922030448913574, "learning_rate": 0.0002, "epoch": 1.1978221415607986, "step": 1650}, {"loss": 1.0565, "grad_norm": 0.6142027378082275, "learning_rate": 0.0002, "epoch": 1.2050816696914701, "step": 1660}, {"loss": 0.9541, "grad_norm": 0.44769710302352905, "learning_rate": 0.0002, "epoch": 1.2123411978221417, "step": 1670}, {"loss": 1.2501, "grad_norm": 0.41233646869659424, "learning_rate": 0.0002, "epoch": 1.219600725952813, "step": 1680}, {"loss": 1.0147, "grad_norm": 0.2928866147994995, "learning_rate": 0.0002, "epoch": 1.2268602540834845, "step": 1690}, {"loss": 0.9074, "grad_norm": 0.36913734674453735, "learning_rate": 0.0002, "epoch": 1.234119782214156, "step": 1700}, {"loss": 1.0847, "grad_norm": 0.5281891226768494, "learning_rate": 0.0002, "epoch": 1.2413793103448276, "step": 1710}, {"loss": 0.9482, "grad_norm": 0.3374697268009186, "learning_rate": 0.0002, "epoch": 1.2486388384754992, "step": 1720}, {"loss": 0.9826, "grad_norm": 0.3802020847797394, "learning_rate": 0.0002, "epoch": 1.2558983666061705, "step": 1730}, {"loss": 1.0191, "grad_norm": 0.38048651814460754, "learning_rate": 0.0002, "epoch": 1.263157894736842, "step": 1740}, {"loss": 1.0142, "grad_norm": 0.6676169633865356, "learning_rate": 0.0002, "epoch": 1.2704174228675136, "step": 1750}, {"loss": 0.992, "grad_norm": 0.4075961410999298, "learning_rate": 0.0002, "epoch": 1.2776769509981851, "step": 1760}, {"loss": 1.0301, "grad_norm": 0.4374721348285675, "learning_rate": 0.0002, "epoch": 1.2849364791288567, "step": 1770}, {"loss": 1.0601, "grad_norm": 0.4638824164867401, "learning_rate": 0.0002, "epoch": 1.2921960072595282, "step": 1780}, {"loss": 0.9593, "grad_norm": 0.38631564378738403, "learning_rate": 0.0002, "epoch": 1.2994555353901998, "step": 1790}, {"loss": 1.0042, "grad_norm": 0.35873809456825256, "learning_rate": 0.0002, "epoch": 1.306715063520871, "step": 1800}, {"loss": 0.9835, "grad_norm": 0.33602237701416016, "learning_rate": 0.0002, "epoch": 1.3139745916515426, "step": 1810}, {"loss": 1.1032, "grad_norm": 0.46696463227272034, "learning_rate": 0.0002, "epoch": 1.3212341197822142, "step": 1820}, {"loss": 1.0123, "grad_norm": 0.368958979845047, "learning_rate": 0.0002, "epoch": 1.3284936479128857, "step": 1830}, {"loss": 1.0456, "grad_norm": 0.3160957396030426, "learning_rate": 0.0002, "epoch": 1.335753176043557, "step": 1840}, {"loss": 1.0264, "grad_norm": 0.4511511027812958, "learning_rate": 0.0002, "epoch": 1.3430127041742286, "step": 1850}, {"loss": 1.0179, "grad_norm": 0.6769845485687256, "learning_rate": 0.0002, "epoch": 1.3502722323049001, "step": 1860}, {"loss": 1.011, "grad_norm": 0.3749309480190277, "learning_rate": 0.0002, "epoch": 1.3575317604355717, "step": 1870}, {"loss": 1.0223, "grad_norm": 0.553798496723175, "learning_rate": 0.0002, "epoch": 1.3647912885662432, "step": 1880}, {"loss": 0.9634, "grad_norm": 0.3538985550403595, "learning_rate": 0.0002, "epoch": 1.3720508166969148, "step": 1890}, {"loss": 1.0275, "grad_norm": 0.5501534938812256, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1900}, {"loss": 1.0461, "grad_norm": 0.4432051181793213, "learning_rate": 0.0002, "epoch": 1.3865698729582578, "step": 1910}, {"loss": 0.9082, "grad_norm": 0.41755786538124084, "learning_rate": 0.0002, "epoch": 1.3938294010889292, "step": 1920}, {"loss": 1.0858, "grad_norm": 0.5626114010810852, "learning_rate": 0.0002, "epoch": 1.4010889292196007, "step": 1930}, {"loss": 1.0687, "grad_norm": 0.44038185477256775, "learning_rate": 0.0002, "epoch": 1.4083484573502723, "step": 1940}, {"loss": 0.9454, "grad_norm": 0.3397001624107361, "learning_rate": 0.0002, "epoch": 1.4156079854809438, "step": 1950}, {"loss": 0.943, "grad_norm": 0.4325368106365204, "learning_rate": 0.0002, "epoch": 1.4228675136116151, "step": 1960}, {"loss": 1.0488, "grad_norm": 0.3900907039642334, "learning_rate": 0.0002, "epoch": 1.4301270417422867, "step": 1970}, {"loss": 0.9699, "grad_norm": 0.369612455368042, "learning_rate": 0.0002, "epoch": 1.4373865698729582, "step": 1980}, {"loss": 1.0609, "grad_norm": 0.4389338791370392, "learning_rate": 0.0002, "epoch": 1.4446460980036298, "step": 1990}, {"loss": 1.042, "grad_norm": 1.694450855255127, "learning_rate": 0.0002, "epoch": 1.4519056261343013, "step": 2000}, {"loss": 0.997, "grad_norm": 0.516957700252533, "learning_rate": 0.0002, "epoch": 1.4591651542649728, "step": 2010}, {"loss": 1.0979, "grad_norm": 0.45515501499176025, "learning_rate": 0.0002, "epoch": 1.4664246823956444, "step": 2020}, {"loss": 1.0188, "grad_norm": 0.4153139591217041, "learning_rate": 0.0002, "epoch": 1.4736842105263157, "step": 2030}, {"loss": 1.1521, "grad_norm": 0.44353052973747253, "learning_rate": 0.0002, "epoch": 1.4809437386569873, "step": 2040}, {"loss": 0.9653, "grad_norm": 0.570554256439209, "learning_rate": 0.0002, "epoch": 1.4882032667876588, "step": 2050}, {"loss": 1.0352, "grad_norm": 0.5742740035057068, "learning_rate": 0.0002, "epoch": 1.4954627949183303, "step": 2060}, {"loss": 1.0963, "grad_norm": 0.5890517830848694, "learning_rate": 0.0002, "epoch": 1.5027223230490017, "step": 2070}, {"loss": 1.0639, "grad_norm": 0.4162650406360626, "learning_rate": 0.0002, "epoch": 1.5099818511796732, "step": 2080}, {"loss": 1.071, "grad_norm": 0.7334717512130737, "learning_rate": 0.0002, "epoch": 1.5172413793103448, "step": 2090}, {"loss": 0.8957, "grad_norm": 0.2963249683380127, "learning_rate": 0.0002, "epoch": 1.5245009074410163, "step": 2100}, {"loss": 1.0446, "grad_norm": 0.30676454305648804, "learning_rate": 0.0002, "epoch": 1.5317604355716878, "step": 2110}, {"loss": 0.9967, "grad_norm": 0.35984641313552856, "learning_rate": 0.0002, "epoch": 1.5390199637023594, "step": 2120}, {"loss": 1.0352, "grad_norm": 0.3384549617767334, "learning_rate": 0.0002, "epoch": 1.546279491833031, "step": 2130}, {"loss": 0.9874, "grad_norm": 0.4725518226623535, "learning_rate": 0.0002, "epoch": 1.5535390199637025, "step": 2140}, {"loss": 1.1166, "grad_norm": 0.3252685070037842, "learning_rate": 0.0002, "epoch": 1.560798548094374, "step": 2150}, {"loss": 0.9127, "grad_norm": 0.45043081045150757, "learning_rate": 0.0002, "epoch": 1.5680580762250453, "step": 2160}, {"loss": 1.0767, "grad_norm": 0.374208003282547, "learning_rate": 0.0002, "epoch": 1.5753176043557169, "step": 2170}, {"loss": 1.0255, "grad_norm": 0.5118404030799866, "learning_rate": 0.0002, "epoch": 1.5825771324863884, "step": 2180}, {"loss": 1.0062, "grad_norm": 0.482170969247818, "learning_rate": 0.0002, "epoch": 1.5898366606170597, "step": 2190}, {"loss": 1.0872, "grad_norm": 0.5337533950805664, "learning_rate": 0.0002, "epoch": 1.5970961887477313, "step": 2200}, {"loss": 1.0405, "grad_norm": 0.5195064544677734, "learning_rate": 0.0002, "epoch": 1.6043557168784028, "step": 2210}, {"loss": 1.0454, "grad_norm": 0.30807098746299744, "learning_rate": 0.0002, "epoch": 1.6116152450090744, "step": 2220}, {"loss": 1.0293, "grad_norm": 0.3962925672531128, "learning_rate": 0.0002, "epoch": 1.618874773139746, "step": 2230}, {"loss": 1.0137, "grad_norm": 0.7636962532997131, "learning_rate": 0.0002, "epoch": 1.6261343012704175, "step": 2240}, {"loss": 0.991, "grad_norm": 0.32380592823028564, "learning_rate": 0.0002, "epoch": 1.633393829401089, "step": 2250}, {"loss": 0.9471, "grad_norm": 0.5767741799354553, "learning_rate": 0.0002, "epoch": 1.6406533575317606, "step": 2260}, {"loss": 0.9524, "grad_norm": 0.39964812994003296, "learning_rate": 0.0002, "epoch": 1.647912885662432, "step": 2270}, {"loss": 0.9866, "grad_norm": 0.622629702091217, "learning_rate": 0.0002, "epoch": 1.6551724137931034, "step": 2280}, {"loss": 0.9609, "grad_norm": 0.40202152729034424, "learning_rate": 0.0002, "epoch": 1.662431941923775, "step": 2290}, {"loss": 0.9859, "grad_norm": 0.4467349052429199, "learning_rate": 0.0002, "epoch": 1.6696914700544465, "step": 2300}, {"loss": 1.0312, "grad_norm": 0.5026949048042297, "learning_rate": 0.0002, "epoch": 1.6769509981851178, "step": 2310}, {"loss": 1.0776, "grad_norm": 0.43754541873931885, "learning_rate": 0.0002, "epoch": 1.6842105263157894, "step": 2320}, {"loss": 0.9786, "grad_norm": 0.42869430780410767, "learning_rate": 0.0002, "epoch": 1.691470054446461, "step": 2330}, {"loss": 1.0747, "grad_norm": 0.4192679524421692, "learning_rate": 0.0002, "epoch": 1.6987295825771325, "step": 2340}, {"loss": 1.0054, "grad_norm": 0.3243155777454376, "learning_rate": 0.0002, "epoch": 1.705989110707804, "step": 2350}, {"loss": 0.9165, "grad_norm": 1.0514075756072998, "learning_rate": 0.0002, "epoch": 1.7132486388384756, "step": 2360}, {"loss": 1.1353, "grad_norm": 0.4719122648239136, "learning_rate": 0.0002, "epoch": 1.720508166969147, "step": 2370}, {"loss": 0.9913, "grad_norm": 0.3846144676208496, "learning_rate": 0.0002, "epoch": 1.7277676950998186, "step": 2380}, {"loss": 0.9935, "grad_norm": 0.7266581058502197, "learning_rate": 0.0002, "epoch": 1.73502722323049, "step": 2390}, {"loss": 0.9509, "grad_norm": 0.6183241605758667, "learning_rate": 0.0002, "epoch": 1.7422867513611615, "step": 2400}, {"loss": 0.9713, "grad_norm": 0.3658260405063629, "learning_rate": 0.0002, "epoch": 1.749546279491833, "step": 2410}, {"loss": 1.0033, "grad_norm": 0.6036322712898254, "learning_rate": 0.0002, "epoch": 1.7568058076225044, "step": 2420}, {"loss": 1.0526, "grad_norm": 0.7872936129570007, "learning_rate": 0.0002, "epoch": 1.764065335753176, "step": 2430}, {"loss": 0.99, "grad_norm": 0.35946124792099, "learning_rate": 0.0002, "epoch": 1.7713248638838475, "step": 2440}, {"loss": 0.8845, "grad_norm": 0.3740338981151581, "learning_rate": 0.0002, "epoch": 1.778584392014519, "step": 2450}, {"loss": 0.9333, "grad_norm": 0.6150230169296265, "learning_rate": 0.0002, "epoch": 1.7858439201451906, "step": 2460}, {"loss": 1.0373, "grad_norm": 0.4726075530052185, "learning_rate": 0.0002, "epoch": 1.793103448275862, "step": 2470}, {"loss": 0.974, "grad_norm": 0.31292253732681274, "learning_rate": 0.0002, "epoch": 1.8003629764065336, "step": 2480}, {"loss": 1.0205, "grad_norm": 0.4463104009628296, "learning_rate": 0.0002, "epoch": 1.8076225045372052, "step": 2490}, {"loss": 1.0771, "grad_norm": 0.7848200798034668, "learning_rate": 0.0002, "epoch": 1.8148820326678767, "step": 2500}, {"loss": 0.9508, "grad_norm": 0.5562082529067993, "learning_rate": 0.0002, "epoch": 1.822141560798548, "step": 2510}, {"loss": 0.9343, "grad_norm": 0.39892756938934326, "learning_rate": 0.0002, "epoch": 1.8294010889292196, "step": 2520}, {"loss": 1.0489, "grad_norm": 1.2923320531845093, "learning_rate": 0.0002, "epoch": 1.8366606170598911, "step": 2530}, {"loss": 1.1203, "grad_norm": 0.6316490769386292, "learning_rate": 0.0002, "epoch": 1.8439201451905625, "step": 2540}, {"loss": 0.9131, "grad_norm": 0.46100497245788574, "learning_rate": 0.0002, "epoch": 1.851179673321234, "step": 2550}, {"loss": 1.0977, "grad_norm": 0.7902987003326416, "learning_rate": 0.0002, "epoch": 1.8584392014519056, "step": 2560}, {"loss": 0.9702, "grad_norm": 0.4596365690231323, "learning_rate": 0.0002, "epoch": 1.865698729582577, "step": 2570}, {"loss": 0.9322, "grad_norm": 0.6592172384262085, "learning_rate": 0.0002, "epoch": 1.8729582577132486, "step": 2580}, {"loss": 0.9549, "grad_norm": 0.5764662027359009, "learning_rate": 0.0002, "epoch": 1.8802177858439202, "step": 2590}, {"loss": 0.9485, "grad_norm": 0.8421637415885925, "learning_rate": 0.0002, "epoch": 1.8874773139745917, "step": 2600}, {"loss": 1.012, "grad_norm": 0.5635305047035217, "learning_rate": 0.0002, "epoch": 1.8947368421052633, "step": 2610}, {"loss": 0.8907, "grad_norm": 0.46231237053871155, "learning_rate": 0.0002, "epoch": 1.9019963702359348, "step": 2620}, {"loss": 0.9543, "grad_norm": 0.3944607079029083, "learning_rate": 0.0002, "epoch": 1.9092558983666061, "step": 2630}, {"loss": 0.9964, "grad_norm": 0.4753907322883606, "learning_rate": 0.0002, "epoch": 1.9165154264972777, "step": 2640}, {"loss": 1.0217, "grad_norm": 0.4151090979576111, "learning_rate": 0.0002, "epoch": 1.9237749546279492, "step": 2650}, {"loss": 1.0397, "grad_norm": 0.6793725490570068, "learning_rate": 0.0002, "epoch": 1.9310344827586206, "step": 2660}, {"loss": 1.0401, "grad_norm": 0.339755117893219, "learning_rate": 0.0002, "epoch": 1.938294010889292, "step": 2670}, {"loss": 1.082, "grad_norm": 0.40789374709129333, "learning_rate": 0.0002, "epoch": 1.9455535390199636, "step": 2680}, {"loss": 0.9973, "grad_norm": 0.3750005066394806, "learning_rate": 0.0002, "epoch": 1.9528130671506352, "step": 2690}, {"loss": 0.9837, "grad_norm": 0.39684441685676575, "learning_rate": 0.0002, "epoch": 1.9600725952813067, "step": 2700}, {"loss": 1.1204, "grad_norm": 0.378287672996521, "learning_rate": 0.0002, "epoch": 1.9673321234119783, "step": 2710}, {"loss": 0.9284, "grad_norm": 0.3668482005596161, "learning_rate": 0.0002, "epoch": 1.9745916515426498, "step": 2720}, {"loss": 1.0512, "grad_norm": 0.49997565150260925, "learning_rate": 0.0002, "epoch": 1.9818511796733214, "step": 2730}, {"loss": 0.9311, "grad_norm": 0.36852124333381653, "learning_rate": 0.0002, "epoch": 1.989110707803993, "step": 2740}, {"loss": 1.0898, "grad_norm": 0.5203380584716797, "learning_rate": 0.0002, "epoch": 1.9963702359346642, "step": 2750}, {"eval_loss": 1.1086540222167969, "eval_runtime": 95.6866, "eval_samples_per_second": 4.557, "eval_steps_per_second": 0.575, "epoch": 2.0, "step": 2755}, {"loss": 0.9791, "grad_norm": 0.35921573638916016, "learning_rate": 0.0002, "epoch": 2.0036297640653356, "step": 2760}, {"loss": 0.8968, "grad_norm": 1.013861894607544, "learning_rate": 0.0002, "epoch": 2.010889292196007, "step": 2770}, {"loss": 0.7873, "grad_norm": 0.4425240159034729, "learning_rate": 0.0002, "epoch": 2.0181488203266786, "step": 2780}, {"loss": 0.8758, "grad_norm": 0.60174161195755, "learning_rate": 0.0002, "epoch": 2.02540834845735, "step": 2790}, {"loss": 0.7989, "grad_norm": 0.47582098841667175, "learning_rate": 0.0002, "epoch": 2.0326678765880217, "step": 2800}, {"loss": 0.8898, "grad_norm": 0.6012811660766602, "learning_rate": 0.0002, "epoch": 2.0399274047186933, "step": 2810}, {"loss": 0.8109, "grad_norm": 0.4444408118724823, "learning_rate": 0.0002, "epoch": 2.047186932849365, "step": 2820}, {"loss": 0.8621, "grad_norm": 0.4864003360271454, "learning_rate": 0.0002, "epoch": 2.0544464609800364, "step": 2830}, {"loss": 0.8817, "grad_norm": 0.5104215741157532, "learning_rate": 0.0002, "epoch": 2.061705989110708, "step": 2840}, {"loss": 0.7789, "grad_norm": 0.6218489408493042, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 2850}, {"loss": 0.8778, "grad_norm": 0.705784261226654, "learning_rate": 0.0002, "epoch": 2.076225045372051, "step": 2860}, {"loss": 0.7851, "grad_norm": 0.48091503977775574, "learning_rate": 0.0002, "epoch": 2.0834845735027225, "step": 2870}, {"loss": 0.8444, "grad_norm": 0.5062456727027893, "learning_rate": 0.0002, "epoch": 2.0907441016333936, "step": 2880}, {"loss": 0.7898, "grad_norm": 0.4862022399902344, "learning_rate": 0.0002, "epoch": 2.098003629764065, "step": 2890}, {"loss": 0.8593, "grad_norm": 0.48264044523239136, "learning_rate": 0.0002, "epoch": 2.1052631578947367, "step": 2900}, {"loss": 0.8325, "grad_norm": 0.43744346499443054, "learning_rate": 0.0002, "epoch": 2.1125226860254083, "step": 2910}, {"loss": 0.9099, "grad_norm": 0.5480492115020752, "learning_rate": 0.0002, "epoch": 2.11978221415608, "step": 2920}, {"loss": 0.7727, "grad_norm": 0.5068560838699341, "learning_rate": 0.0002, "epoch": 2.1270417422867514, "step": 2930}, {"loss": 0.8403, "grad_norm": 0.4650852680206299, "learning_rate": 0.0002, "epoch": 2.134301270417423, "step": 2940}, {"loss": 0.8467, "grad_norm": 0.4929981231689453, "learning_rate": 0.0002, "epoch": 2.1415607985480944, "step": 2950}, {"loss": 0.8747, "grad_norm": 0.6537389755249023, "learning_rate": 0.0002, "epoch": 2.148820326678766, "step": 2960}, {"loss": 0.7663, "grad_norm": 0.8032940626144409, "learning_rate": 0.0002, "epoch": 2.1560798548094375, "step": 2970}, {"loss": 0.7604, "grad_norm": 0.7131643891334534, "learning_rate": 0.0002, "epoch": 2.163339382940109, "step": 2980}, {"loss": 0.8424, "grad_norm": 0.6034275889396667, "learning_rate": 0.0002, "epoch": 2.1705989110707806, "step": 2990}, {"loss": 0.8838, "grad_norm": 0.6081095933914185, "learning_rate": 0.0002, "epoch": 2.1778584392014517, "step": 3000}, {"loss": 0.9013, "grad_norm": 0.5706912875175476, "learning_rate": 0.0002, "epoch": 2.1851179673321233, "step": 3010}, {"loss": 0.8473, "grad_norm": 0.6742380261421204, "learning_rate": 0.0002, "epoch": 2.192377495462795, "step": 3020}, {"loss": 0.8372, "grad_norm": 0.4847496449947357, "learning_rate": 0.0002, "epoch": 2.1996370235934664, "step": 3030}, {"loss": 0.8487, "grad_norm": 0.5529342889785767, "learning_rate": 0.0002, "epoch": 2.206896551724138, "step": 3040}, {"loss": 0.9057, "grad_norm": 0.6108783483505249, "learning_rate": 0.0002, "epoch": 2.2141560798548094, "step": 3050}, {"loss": 0.9659, "grad_norm": 0.8841571807861328, "learning_rate": 0.0002, "epoch": 2.221415607985481, "step": 3060}, {"loss": 0.8538, "grad_norm": 0.4227530360221863, "learning_rate": 0.0002, "epoch": 2.2286751361161525, "step": 3070}, {"loss": 0.7925, "grad_norm": 0.461935818195343, "learning_rate": 0.0002, "epoch": 2.235934664246824, "step": 3080}, {"loss": 0.8542, "grad_norm": 0.5407412648200989, "learning_rate": 0.0002, "epoch": 2.2431941923774956, "step": 3090}, {"loss": 0.8392, "grad_norm": 0.5057998895645142, "learning_rate": 0.0002, "epoch": 2.250453720508167, "step": 3100}, {"loss": 0.7898, "grad_norm": 0.530057966709137, "learning_rate": 0.0002, "epoch": 2.2577132486388383, "step": 3110}, {"loss": 0.8507, "grad_norm": 0.5066842436790466, "learning_rate": 0.0002, "epoch": 2.26497277676951, "step": 3120}, {"loss": 0.7903, "grad_norm": 0.5069178342819214, "learning_rate": 0.0002, "epoch": 2.2722323049001814, "step": 3130}, {"loss": 0.8458, "grad_norm": 0.6095499396324158, "learning_rate": 0.0002, "epoch": 2.279491833030853, "step": 3140}, {"loss": 0.8171, "grad_norm": 0.49524766206741333, "learning_rate": 0.0002, "epoch": 2.2867513611615244, "step": 3150}, {"loss": 0.7568, "grad_norm": 0.5334409475326538, "learning_rate": 0.0002, "epoch": 2.294010889292196, "step": 3160}, {"loss": 0.7533, "grad_norm": 1.681748867034912, "learning_rate": 0.0002, "epoch": 2.3012704174228675, "step": 3170}, {"loss": 0.8155, "grad_norm": 0.7225565314292908, "learning_rate": 0.0002, "epoch": 2.308529945553539, "step": 3180}, {"loss": 0.8034, "grad_norm": 0.5379388928413391, "learning_rate": 0.0002, "epoch": 2.3157894736842106, "step": 3190}, {"loss": 0.8833, "grad_norm": 0.45770326256752014, "learning_rate": 0.0002, "epoch": 2.323049001814882, "step": 3200}, {"loss": 0.8419, "grad_norm": 0.6984533071517944, "learning_rate": 0.0002, "epoch": 2.3303085299455537, "step": 3210}, {"loss": 0.7414, "grad_norm": 0.6725744605064392, "learning_rate": 0.0002, "epoch": 2.337568058076225, "step": 3220}, {"loss": 0.8657, "grad_norm": 1.1247471570968628, "learning_rate": 0.0002, "epoch": 2.344827586206897, "step": 3230}, {"loss": 0.7782, "grad_norm": 1.0240263938903809, "learning_rate": 0.0002, "epoch": 2.352087114337568, "step": 3240}, {"loss": 0.826, "grad_norm": 0.5608096122741699, "learning_rate": 0.0002, "epoch": 2.3593466424682394, "step": 3250}, {"loss": 0.8191, "grad_norm": 0.8294990062713623, "learning_rate": 0.0002, "epoch": 2.366606170598911, "step": 3260}, {"loss": 0.9845, "grad_norm": 0.6734224557876587, "learning_rate": 0.0002, "epoch": 2.3738656987295825, "step": 3270}, {"loss": 0.7921, "grad_norm": 0.6862800717353821, "learning_rate": 0.0002, "epoch": 2.381125226860254, "step": 3280}, {"loss": 0.7752, "grad_norm": 0.5442930459976196, "learning_rate": 0.0002, "epoch": 2.3883847549909256, "step": 3290}, {"loss": 0.8543, "grad_norm": 0.5745394229888916, "learning_rate": 0.0002, "epoch": 2.395644283121597, "step": 3300}, {"loss": 0.833, "grad_norm": 0.6257799863815308, "learning_rate": 0.0002, "epoch": 2.4029038112522687, "step": 3310}, {"loss": 0.8191, "grad_norm": 0.5608420968055725, "learning_rate": 0.0002, "epoch": 2.4101633393829403, "step": 3320}, {"loss": 0.8745, "grad_norm": 0.5512017011642456, "learning_rate": 0.0002, "epoch": 2.417422867513612, "step": 3330}, {"loss": 0.8642, "grad_norm": 0.7452999353408813, "learning_rate": 0.0002, "epoch": 2.4246823956442833, "step": 3340}, {"loss": 0.8417, "grad_norm": 0.4604301452636719, "learning_rate": 0.0002, "epoch": 2.4319419237749544, "step": 3350}, {"loss": 0.9148, "grad_norm": 0.8225823640823364, "learning_rate": 0.0002, "epoch": 2.439201451905626, "step": 3360}, {"loss": 0.8144, "grad_norm": 0.8604981303215027, "learning_rate": 0.0002, "epoch": 2.4464609800362975, "step": 3370}, {"loss": 0.9276, "grad_norm": 0.6620925664901733, "learning_rate": 0.0002, "epoch": 2.453720508166969, "step": 3380}, {"loss": 0.8381, "grad_norm": 0.4750158488750458, "learning_rate": 0.0002, "epoch": 2.4609800362976406, "step": 3390}, {"loss": 0.8249, "grad_norm": 0.6061418056488037, "learning_rate": 0.0002, "epoch": 2.468239564428312, "step": 3400}, {"loss": 0.8944, "grad_norm": 0.5977247357368469, "learning_rate": 0.0002, "epoch": 2.4754990925589837, "step": 3410}, {"loss": 0.8366, "grad_norm": 0.5004227757453918, "learning_rate": 0.0002, "epoch": 2.4827586206896552, "step": 3420}, {"loss": 0.9133, "grad_norm": 0.46428972482681274, "learning_rate": 0.0002, "epoch": 2.490018148820327, "step": 3430}, {"loss": 0.8534, "grad_norm": 0.7658395171165466, "learning_rate": 0.0002, "epoch": 2.4972776769509983, "step": 3440}, {"loss": 0.9066, "grad_norm": 0.621819794178009, "learning_rate": 0.0002, "epoch": 2.50453720508167, "step": 3450}, {"loss": 0.8481, "grad_norm": 0.4556088149547577, "learning_rate": 0.0002, "epoch": 2.511796733212341, "step": 3460}, {"loss": 0.8522, "grad_norm": 0.6124140024185181, "learning_rate": 0.0002, "epoch": 2.519056261343013, "step": 3470}, {"loss": 0.8158, "grad_norm": 0.6256231665611267, "learning_rate": 0.0002, "epoch": 2.526315789473684, "step": 3480}, {"loss": 0.9097, "grad_norm": 0.5464141964912415, "learning_rate": 0.0002, "epoch": 2.5335753176043556, "step": 3490}, {"loss": 0.8708, "grad_norm": 0.51471346616745, "learning_rate": 0.0002, "epoch": 2.540834845735027, "step": 3500}, {"loss": 0.9042, "grad_norm": 0.5326165556907654, "learning_rate": 0.0002, "epoch": 2.5480943738656987, "step": 3510}, {"loss": 0.8749, "grad_norm": 0.4750378429889679, "learning_rate": 0.0002, "epoch": 2.5553539019963702, "step": 3520}, {"loss": 0.8248, "grad_norm": 0.5292279124259949, "learning_rate": 0.0002, "epoch": 2.562613430127042, "step": 3530}, {"loss": 0.8808, "grad_norm": 0.6145227551460266, "learning_rate": 0.0002, "epoch": 2.5698729582577133, "step": 3540}, {"loss": 0.846, "grad_norm": 0.8275189399719238, "learning_rate": 0.0002, "epoch": 2.577132486388385, "step": 3550}, {"loss": 0.875, "grad_norm": 0.5037438273429871, "learning_rate": 0.0002, "epoch": 2.5843920145190564, "step": 3560}, {"loss": 0.8324, "grad_norm": 0.5838707685470581, "learning_rate": 0.0002, "epoch": 2.5916515426497275, "step": 3570}, {"loss": 0.8716, "grad_norm": 0.5398710370063782, "learning_rate": 0.0002, "epoch": 2.5989110707803995, "step": 3580}, {"loss": 0.7843, "grad_norm": 0.6115376949310303, "learning_rate": 0.0002, "epoch": 2.6061705989110706, "step": 3590}, {"loss": 0.8142, "grad_norm": 1.118809461593628, "learning_rate": 0.0002, "epoch": 2.613430127041742, "step": 3600}, {"loss": 0.8282, "grad_norm": 0.7811771631240845, "learning_rate": 0.0002, "epoch": 2.6206896551724137, "step": 3610}, {"loss": 0.7997, "grad_norm": 0.5753175020217896, "learning_rate": 0.0002, "epoch": 2.6279491833030852, "step": 3620}, {"loss": 0.7705, "grad_norm": 0.550829291343689, "learning_rate": 0.0002, "epoch": 2.635208711433757, "step": 3630}, {"loss": 0.8564, "grad_norm": 0.5360019207000732, "learning_rate": 0.0002, "epoch": 2.6424682395644283, "step": 3640}, {"loss": 0.7648, "grad_norm": 0.63050377368927, "learning_rate": 0.0002, "epoch": 2.6497277676951, "step": 3650}, {"loss": 0.9156, "grad_norm": 0.5833110213279724, "learning_rate": 0.0002, "epoch": 2.6569872958257714, "step": 3660}, {"loss": 0.8033, "grad_norm": 0.5543047189712524, "learning_rate": 0.0002, "epoch": 2.664246823956443, "step": 3670}, {"loss": 0.8867, "grad_norm": 0.6842212080955505, "learning_rate": 0.0002, "epoch": 2.671506352087114, "step": 3680}, {"loss": 0.8569, "grad_norm": 0.6859333515167236, "learning_rate": 0.0002, "epoch": 2.678765880217786, "step": 3690}, {"loss": 0.8081, "grad_norm": 0.7038410902023315, "learning_rate": 0.0002, "epoch": 2.686025408348457, "step": 3700}, {"loss": 0.8686, "grad_norm": 0.447233647108078, "learning_rate": 0.0002, "epoch": 2.6932849364791287, "step": 3710}, {"loss": 0.8951, "grad_norm": 0.6453872919082642, "learning_rate": 0.0002, "epoch": 2.7005444646098002, "step": 3720}, {"loss": 0.8292, "grad_norm": 0.8025672435760498, "learning_rate": 0.0002, "epoch": 2.707803992740472, "step": 3730}, {"loss": 0.8488, "grad_norm": 0.5997087955474854, "learning_rate": 0.0002, "epoch": 2.7150635208711433, "step": 3740}, {"loss": 0.8308, "grad_norm": 0.6901142001152039, "learning_rate": 0.0002, "epoch": 2.722323049001815, "step": 3750}, {"loss": 0.8517, "grad_norm": 1.036145567893982, "learning_rate": 0.0002, "epoch": 2.7295825771324864, "step": 3760}, {"loss": 0.8455, "grad_norm": 0.7207072377204895, "learning_rate": 0.0002, "epoch": 2.736842105263158, "step": 3770}, {"loss": 0.8009, "grad_norm": 1.0452989339828491, "learning_rate": 0.0002, "epoch": 2.7441016333938295, "step": 3780}, {"loss": 0.8868, "grad_norm": 0.5615278482437134, "learning_rate": 0.0002, "epoch": 2.751361161524501, "step": 3790}, {"loss": 0.9047, "grad_norm": 0.46439215540885925, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 3800}, {"loss": 0.7847, "grad_norm": 0.7134917974472046, "learning_rate": 0.0002, "epoch": 2.7658802177858437, "step": 3810}, {"loss": 0.8315, "grad_norm": 0.5139115452766418, "learning_rate": 0.0002, "epoch": 2.7731397459165157, "step": 3820}, {"loss": 0.8995, "grad_norm": 0.8595781326293945, "learning_rate": 0.0002, "epoch": 2.780399274047187, "step": 3830}, {"loss": 0.9079, "grad_norm": 0.544614851474762, "learning_rate": 0.0002, "epoch": 2.7876588021778583, "step": 3840}, {"loss": 0.8322, "grad_norm": 0.6073850393295288, "learning_rate": 0.0002, "epoch": 2.79491833030853, "step": 3850}, {"loss": 0.8277, "grad_norm": 0.8224069476127625, "learning_rate": 0.0002, "epoch": 2.8021778584392014, "step": 3860}, {"loss": 0.815, "grad_norm": 0.5347970128059387, "learning_rate": 0.0002, "epoch": 2.809437386569873, "step": 3870}, {"loss": 0.8219, "grad_norm": 0.6799601912498474, "learning_rate": 0.0002, "epoch": 2.8166969147005445, "step": 3880}, {"loss": 0.8796, "grad_norm": 0.5219197869300842, "learning_rate": 0.0002, "epoch": 2.823956442831216, "step": 3890}, {"loss": 0.8649, "grad_norm": 0.5710130333900452, "learning_rate": 0.0002, "epoch": 2.8312159709618876, "step": 3900}, {"loss": 0.8741, "grad_norm": 0.5857213139533997, "learning_rate": 0.0002, "epoch": 2.838475499092559, "step": 3910}, {"loss": 0.8154, "grad_norm": 0.5206644535064697, "learning_rate": 0.0002, "epoch": 2.8457350272232302, "step": 3920}, {"loss": 0.9328, "grad_norm": 1.2902015447616577, "learning_rate": 0.0002, "epoch": 2.8529945553539022, "step": 3930}, {"loss": 0.8192, "grad_norm": 0.5252797603607178, "learning_rate": 0.0002, "epoch": 2.8602540834845733, "step": 3940}, {"loss": 0.9047, "grad_norm": 0.5925108790397644, "learning_rate": 0.0002, "epoch": 2.867513611615245, "step": 3950}, {"loss": 0.8611, "grad_norm": 0.9719610810279846, "learning_rate": 0.0002, "epoch": 2.8747731397459164, "step": 3960}, {"loss": 0.8531, "grad_norm": 0.6834747195243835, "learning_rate": 0.0002, "epoch": 2.882032667876588, "step": 3970}, {"loss": 0.8124, "grad_norm": 0.6952353119850159, "learning_rate": 0.0002, "epoch": 2.8892921960072595, "step": 3980}, {"loss": 0.8365, "grad_norm": 0.49889910221099854, "learning_rate": 0.0002, "epoch": 2.896551724137931, "step": 3990}, {"loss": 0.8197, "grad_norm": 0.5007026791572571, "learning_rate": 0.0002, "epoch": 2.9038112522686026, "step": 4000}, {"loss": 0.7752, "grad_norm": 0.5474239587783813, "learning_rate": 0.0002, "epoch": 2.911070780399274, "step": 4010}, {"loss": 0.8579, "grad_norm": 0.6617428660392761, "learning_rate": 0.0002, "epoch": 2.9183303085299457, "step": 4020}, {"loss": 0.8583, "grad_norm": 0.6097776293754578, "learning_rate": 0.0002, "epoch": 2.925589836660617, "step": 4030}, {"loss": 0.8296, "grad_norm": 0.5985828638076782, "learning_rate": 0.0002, "epoch": 2.9328493647912888, "step": 4040}, {"loss": 0.9134, "grad_norm": 0.769488513469696, "learning_rate": 0.0002, "epoch": 2.94010889292196, "step": 4050}, {"loss": 0.8148, "grad_norm": 0.5167055130004883, "learning_rate": 0.0002, "epoch": 2.9473684210526314, "step": 4060}, {"loss": 0.8665, "grad_norm": 0.6265496015548706, "learning_rate": 0.0002, "epoch": 2.954627949183303, "step": 4070}, {"loss": 0.9218, "grad_norm": 1.2644082307815552, "learning_rate": 0.0002, "epoch": 2.9618874773139745, "step": 4080}, {"loss": 0.8026, "grad_norm": 0.6007561087608337, "learning_rate": 0.0002, "epoch": 2.969147005444646, "step": 4090}, {"loss": 0.8262, "grad_norm": 0.47984927892684937, "learning_rate": 0.0002, "epoch": 2.9764065335753176, "step": 4100}, {"loss": 0.7897, "grad_norm": 1.128198504447937, "learning_rate": 0.0002, "epoch": 2.983666061705989, "step": 4110}, {"loss": 0.7883, "grad_norm": 0.526292085647583, "learning_rate": 0.0002, "epoch": 2.9909255898366607, "step": 4120}, {"loss": 0.7801, "grad_norm": 0.5462674498558044, "learning_rate": 0.0002, "epoch": 2.9981851179673322, "step": 4130}, {"eval_loss": 1.1093357801437378, "eval_runtime": 46.2498, "eval_samples_per_second": 9.427, "eval_steps_per_second": 1.189, "epoch": 2.9996370235934666, "step": 4132}, {"loss": 0.721, "grad_norm": 0.48288026452064514, "learning_rate": 0.0002, "epoch": 3.0054446460980038, "step": 4140}, {"loss": 0.6769, "grad_norm": 1.0181782245635986, "learning_rate": 0.0002, "epoch": 3.0127041742286753, "step": 4150}, {"loss": 0.7185, "grad_norm": 0.7718019485473633, "learning_rate": 0.0002, "epoch": 3.019963702359347, "step": 4160}, {"loss": 0.6552, "grad_norm": 0.7492219805717468, "learning_rate": 0.0002, "epoch": 3.027223230490018, "step": 4170}, {"loss": 0.6678, "grad_norm": 0.9363632798194885, "learning_rate": 0.0002, "epoch": 3.0344827586206895, "step": 4180}, {"loss": 0.7187, "grad_norm": 0.6888533234596252, "learning_rate": 0.0002, "epoch": 3.041742286751361, "step": 4190}, {"loss": 0.6469, "grad_norm": 0.7072834968566895, "learning_rate": 0.0002, "epoch": 3.0490018148820326, "step": 4200}, {"loss": 0.6387, "grad_norm": 0.7182047963142395, "learning_rate": 0.0002, "epoch": 3.056261343012704, "step": 4210}, {"loss": 0.6385, "grad_norm": 0.7194355130195618, "learning_rate": 0.0002, "epoch": 3.0635208711433757, "step": 4220}, {"loss": 0.5812, "grad_norm": 0.9454023838043213, "learning_rate": 0.0002, "epoch": 3.0707803992740472, "step": 4230}, {"loss": 0.6036, "grad_norm": 0.838657557964325, "learning_rate": 0.0002, "epoch": 3.0780399274047188, "step": 4240}, {"loss": 0.646, "grad_norm": 0.740113377571106, "learning_rate": 0.0002, "epoch": 3.0852994555353903, "step": 4250}, {"loss": 0.604, "grad_norm": 0.6616561412811279, "learning_rate": 0.0002, "epoch": 3.092558983666062, "step": 4260}, {"loss": 0.6462, "grad_norm": 0.8846506476402283, "learning_rate": 0.0002, "epoch": 3.0998185117967334, "step": 4270}, {"loss": 0.6037, "grad_norm": 0.6322125792503357, "learning_rate": 0.0002, "epoch": 3.107078039927405, "step": 4280}, {"loss": 0.5953, "grad_norm": 0.7461467385292053, "learning_rate": 0.0002, "epoch": 3.114337568058076, "step": 4290}, {"loss": 0.6099, "grad_norm": 0.8251287341117859, "learning_rate": 0.0002, "epoch": 3.1215970961887476, "step": 4300}, {"loss": 0.6284, "grad_norm": 0.8767673373222351, "learning_rate": 0.0002, "epoch": 3.128856624319419, "step": 4310}, {"loss": 0.7535, "grad_norm": 0.7758759260177612, "learning_rate": 0.0002, "epoch": 3.1361161524500907, "step": 4320}, {"loss": 0.6624, "grad_norm": 1.1056879758834839, "learning_rate": 0.0002, "epoch": 3.143375680580762, "step": 4330}, {"loss": 0.691, "grad_norm": 0.8259835243225098, "learning_rate": 0.0002, "epoch": 3.1506352087114338, "step": 4340}, {"loss": 0.6635, "grad_norm": 0.6607027053833008, "learning_rate": 0.0002, "epoch": 3.1578947368421053, "step": 4350}, {"loss": 0.5911, "grad_norm": 0.7983301281929016, "learning_rate": 0.0002, "epoch": 3.165154264972777, "step": 4360}, {"loss": 0.6496, "grad_norm": 0.6725239157676697, "learning_rate": 0.0002, "epoch": 3.1724137931034484, "step": 4370}, {"loss": 0.5966, "grad_norm": 0.9052095413208008, "learning_rate": 0.0002, "epoch": 3.17967332123412, "step": 4380}, {"loss": 0.6877, "grad_norm": 0.8131307363510132, "learning_rate": 0.0002, "epoch": 3.1869328493647915, "step": 4390}, {"loss": 0.6384, "grad_norm": 0.6435626149177551, "learning_rate": 0.0002, "epoch": 3.1941923774954626, "step": 4400}, {"loss": 0.5819, "grad_norm": 0.84367436170578, "learning_rate": 0.0002, "epoch": 3.201451905626134, "step": 4410}, {"loss": 0.6104, "grad_norm": 1.5018867254257202, "learning_rate": 0.0002, "epoch": 3.2087114337568057, "step": 4420}, {"loss": 0.6838, "grad_norm": 0.7019091844558716, "learning_rate": 0.0002, "epoch": 3.215970961887477, "step": 4430}, {"loss": 0.6153, "grad_norm": 0.9164197444915771, "learning_rate": 0.0002, "epoch": 3.2232304900181488, "step": 4440}, {"loss": 0.6618, "grad_norm": 0.7890861630439758, "learning_rate": 0.0002, "epoch": 3.2304900181488203, "step": 4450}, {"loss": 0.6401, "grad_norm": 0.6517660617828369, "learning_rate": 0.0002, "epoch": 3.237749546279492, "step": 4460}, {"loss": 0.6699, "grad_norm": 1.10188889503479, "learning_rate": 0.0002, "epoch": 3.2450090744101634, "step": 4470}, {"loss": 0.6356, "grad_norm": 0.8158330917358398, "learning_rate": 0.0002, "epoch": 3.252268602540835, "step": 4480}, {"loss": 0.7757, "grad_norm": 0.7663109302520752, "learning_rate": 0.0002, "epoch": 3.2595281306715065, "step": 4490}, {"loss": 0.6539, "grad_norm": 0.8473444581031799, "learning_rate": 0.0002, "epoch": 3.266787658802178, "step": 4500}, {"loss": 0.6511, "grad_norm": 0.9724768996238708, "learning_rate": 0.0002, "epoch": 3.274047186932849, "step": 4510}, {"loss": 0.5464, "grad_norm": 0.8516759276390076, "learning_rate": 0.0002, "epoch": 3.281306715063521, "step": 4520}, {"loss": 0.6534, "grad_norm": 0.7543437480926514, "learning_rate": 0.0002, "epoch": 3.288566243194192, "step": 4530}, {"loss": 0.6095, "grad_norm": 1.0472029447555542, "learning_rate": 0.0002, "epoch": 3.2958257713248638, "step": 4540}, {"loss": 0.6216, "grad_norm": 0.6240826845169067, "learning_rate": 0.0002, "epoch": 3.3030852994555353, "step": 4550}, {"loss": 0.6223, "grad_norm": 0.9957774877548218, "learning_rate": 0.0002, "epoch": 3.310344827586207, "step": 4560}, {"loss": 0.618, "grad_norm": 0.6448912620544434, "learning_rate": 0.0002, "epoch": 3.3176043557168784, "step": 4570}, {"loss": 0.6188, "grad_norm": 0.7519692778587341, "learning_rate": 0.0002, "epoch": 3.32486388384755, "step": 4580}, {"loss": 0.6672, "grad_norm": 0.7367453575134277, "learning_rate": 0.0002, "epoch": 3.3321234119782215, "step": 4590}, {"loss": 0.6517, "grad_norm": 0.8064960837364197, "learning_rate": 0.0002, "epoch": 3.339382940108893, "step": 4600}, {"loss": 0.6062, "grad_norm": 0.7664631009101868, "learning_rate": 0.0002, "epoch": 3.3466424682395646, "step": 4610}, {"loss": 0.6834, "grad_norm": 0.7803396582603455, "learning_rate": 0.0002, "epoch": 3.353901996370236, "step": 4620}, {"loss": 0.6961, "grad_norm": 0.9141599535942078, "learning_rate": 0.0002, "epoch": 3.3611615245009077, "step": 4630}, {"loss": 0.6889, "grad_norm": 0.9719856381416321, "learning_rate": 0.0002, "epoch": 3.3684210526315788, "step": 4640}, {"loss": 0.6914, "grad_norm": 0.9223218560218811, "learning_rate": 0.0002, "epoch": 3.3756805807622503, "step": 4650}, {"loss": 0.5981, "grad_norm": 0.7289277911186218, "learning_rate": 0.0002, "epoch": 3.382940108892922, "step": 4660}, {"loss": 0.595, "grad_norm": 1.039724349975586, "learning_rate": 0.0002, "epoch": 3.3901996370235934, "step": 4670}, {"loss": 0.8121, "grad_norm": 1.397438883781433, "learning_rate": 0.0002, "epoch": 3.397459165154265, "step": 4680}, {"loss": 0.6334, "grad_norm": 1.0069999694824219, "learning_rate": 0.0002, "epoch": 3.4047186932849365, "step": 4690}, {"loss": 0.6598, "grad_norm": 0.816291332244873, "learning_rate": 0.0002, "epoch": 3.411978221415608, "step": 4700}, {"loss": 0.6748, "grad_norm": 1.2831530570983887, "learning_rate": 0.0002, "epoch": 3.4192377495462796, "step": 4710}, {"loss": 0.6625, "grad_norm": 0.9573889970779419, "learning_rate": 0.0002, "epoch": 3.426497277676951, "step": 4720}, {"loss": 0.7279, "grad_norm": 0.7685632705688477, "learning_rate": 0.0002, "epoch": 3.4337568058076227, "step": 4730}, {"loss": 0.6104, "grad_norm": 0.7019195556640625, "learning_rate": 0.0002, "epoch": 3.441016333938294, "step": 4740}, {"loss": 0.7606, "grad_norm": 0.7244833707809448, "learning_rate": 0.0002, "epoch": 3.4482758620689653, "step": 4750}, {"loss": 0.6951, "grad_norm": 1.3468551635742188, "learning_rate": 0.0002, "epoch": 3.455535390199637, "step": 4760}, {"loss": 0.6945, "grad_norm": 0.822846531867981, "learning_rate": 0.0002, "epoch": 3.4627949183303084, "step": 4770}, {"loss": 0.6431, "grad_norm": 0.7311608195304871, "learning_rate": 0.0002, "epoch": 3.47005444646098, "step": 4780}, {"loss": 0.7019, "grad_norm": 0.9466770887374878, "learning_rate": 0.0002, "epoch": 3.4773139745916515, "step": 4790}, {"loss": 0.7767, "grad_norm": 1.1527155637741089, "learning_rate": 0.0002, "epoch": 3.484573502722323, "step": 4800}, {"loss": 0.6882, "grad_norm": 1.1288906335830688, "learning_rate": 0.0002, "epoch": 3.4918330308529946, "step": 4810}, {"loss": 0.6564, "grad_norm": 0.9096164107322693, "learning_rate": 0.0002, "epoch": 3.499092558983666, "step": 4820}, {"loss": 0.6127, "grad_norm": 0.7988565564155579, "learning_rate": 0.0002, "epoch": 3.5063520871143377, "step": 4830}, {"loss": 0.7004, "grad_norm": 0.7183415293693542, "learning_rate": 0.0002, "epoch": 3.513611615245009, "step": 4840}, {"loss": 0.74, "grad_norm": 0.6614915132522583, "learning_rate": 0.0002, "epoch": 3.5208711433756807, "step": 4850}, {"loss": 0.7271, "grad_norm": 0.8609521985054016, "learning_rate": 0.0002, "epoch": 3.528130671506352, "step": 4860}, {"loss": 0.6664, "grad_norm": 0.86552894115448, "learning_rate": 0.0002, "epoch": 3.535390199637024, "step": 4870}, {"loss": 0.6432, "grad_norm": 0.6926496028900146, "learning_rate": 0.0002, "epoch": 3.542649727767695, "step": 4880}, {"loss": 0.7117, "grad_norm": 0.8157467246055603, "learning_rate": 0.0002, "epoch": 3.5499092558983665, "step": 4890}, {"loss": 0.6201, "grad_norm": 0.9085357189178467, "learning_rate": 0.0002, "epoch": 3.557168784029038, "step": 4900}, {"loss": 0.6521, "grad_norm": 0.6322644948959351, "learning_rate": 0.0002, "epoch": 3.5644283121597096, "step": 4910}, {"loss": 0.6607, "grad_norm": 1.263205885887146, "learning_rate": 0.0002, "epoch": 3.571687840290381, "step": 4920}, {"loss": 0.6657, "grad_norm": 0.8901070356369019, "learning_rate": 0.0002, "epoch": 3.5789473684210527, "step": 4930}, {"loss": 0.6434, "grad_norm": 0.7983952164649963, "learning_rate": 0.0002, "epoch": 3.586206896551724, "step": 4940}, {"loss": 0.6861, "grad_norm": 0.9887813925743103, "learning_rate": 0.0002, "epoch": 3.5934664246823957, "step": 4950}, {"loss": 0.6502, "grad_norm": 0.7895187735557556, "learning_rate": 0.0002, "epoch": 3.6007259528130673, "step": 4960}, {"loss": 0.7111, "grad_norm": 0.9685819745063782, "learning_rate": 0.0002, "epoch": 3.6079854809437384, "step": 4970}, {"loss": 0.6915, "grad_norm": 0.6576591730117798, "learning_rate": 0.0002, "epoch": 3.6152450090744104, "step": 4980}, {"loss": 0.6195, "grad_norm": 0.856985330581665, "learning_rate": 0.0002, "epoch": 3.6225045372050815, "step": 4990}, {"loss": 0.6318, "grad_norm": 0.7230252623558044, "learning_rate": 0.0002, "epoch": 3.629764065335753, "step": 5000}, {"loss": 0.742, "grad_norm": 0.8260893821716309, "learning_rate": 0.0002, "epoch": 3.6370235934664246, "step": 5010}, {"loss": 0.7223, "grad_norm": 0.7635950446128845, "learning_rate": 0.0002, "epoch": 3.644283121597096, "step": 5020}, {"loss": 0.6837, "grad_norm": 0.7060768604278564, "learning_rate": 0.0002, "epoch": 3.6515426497277677, "step": 5030}, {"loss": 0.6921, "grad_norm": 0.8020303249359131, "learning_rate": 0.0002, "epoch": 3.658802177858439, "step": 5040}, {"loss": 0.6446, "grad_norm": 0.8530341386795044, "learning_rate": 0.0002, "epoch": 3.6660617059891107, "step": 5050}, {"loss": 0.7222, "grad_norm": 0.6667101979255676, "learning_rate": 0.0002, "epoch": 3.6733212341197823, "step": 5060}, {"loss": 0.7081, "grad_norm": 0.7385406494140625, "learning_rate": 0.0002, "epoch": 3.680580762250454, "step": 5070}, {"loss": 0.7035, "grad_norm": 0.7753380537033081, "learning_rate": 0.0002, "epoch": 3.6878402903811254, "step": 5080}, {"loss": 0.6491, "grad_norm": 0.7516207098960876, "learning_rate": 0.0002, "epoch": 3.695099818511797, "step": 5090}, {"loss": 0.672, "grad_norm": 0.8171586394309998, "learning_rate": 0.0002, "epoch": 3.702359346642468, "step": 5100}, {"loss": 0.7459, "grad_norm": 1.0796279907226562, "learning_rate": 0.0002, "epoch": 3.70961887477314, "step": 5110}, {"loss": 0.5948, "grad_norm": 0.6957688927650452, "learning_rate": 0.0002, "epoch": 3.716878402903811, "step": 5120}, {"loss": 0.7515, "grad_norm": 0.8550161719322205, "learning_rate": 0.0002, "epoch": 3.7241379310344827, "step": 5130}, {"loss": 0.7286, "grad_norm": 0.9396728277206421, "learning_rate": 0.0002, "epoch": 3.731397459165154, "step": 5140}, {"loss": 0.7594, "grad_norm": 1.4264805316925049, "learning_rate": 0.0002, "epoch": 3.7386569872958257, "step": 5150}, {"loss": 0.6575, "grad_norm": 0.8725108504295349, "learning_rate": 0.0002, "epoch": 3.7459165154264973, "step": 5160}, {"loss": 0.6778, "grad_norm": 1.0346195697784424, "learning_rate": 0.0002, "epoch": 3.753176043557169, "step": 5170}, {"loss": 0.6371, "grad_norm": 0.5395554304122925, "learning_rate": 0.0002, "epoch": 3.7604355716878404, "step": 5180}, {"loss": 0.7308, "grad_norm": 1.3153616189956665, "learning_rate": 0.0002, "epoch": 3.767695099818512, "step": 5190}, {"loss": 0.78, "grad_norm": 0.9879828691482544, "learning_rate": 0.0002, "epoch": 3.7749546279491835, "step": 5200}, {"loss": 0.7068, "grad_norm": 0.8876672983169556, "learning_rate": 0.0002, "epoch": 3.7822141560798546, "step": 5210}, {"loss": 0.6283, "grad_norm": 0.8363267779350281, "learning_rate": 0.0002, "epoch": 3.7894736842105265, "step": 5220}, {"loss": 0.6255, "grad_norm": 0.637294590473175, "learning_rate": 0.0002, "epoch": 3.7967332123411976, "step": 5230}, {"loss": 0.6685, "grad_norm": 1.1408970355987549, "learning_rate": 0.0002, "epoch": 3.803992740471869, "step": 5240}, {"loss": 0.6761, "grad_norm": 1.0128360986709595, "learning_rate": 0.0002, "epoch": 3.8112522686025407, "step": 5250}, {"loss": 0.6764, "grad_norm": 0.8061144351959229, "learning_rate": 0.0002, "epoch": 3.8185117967332123, "step": 5260}, {"loss": 0.7254, "grad_norm": 0.9626626968383789, "learning_rate": 0.0002, "epoch": 3.825771324863884, "step": 5270}, {"loss": 0.7367, "grad_norm": 0.9013627171516418, "learning_rate": 0.0002, "epoch": 3.8330308529945554, "step": 5280}, {"loss": 0.6806, "grad_norm": 0.8411344289779663, "learning_rate": 0.0002, "epoch": 3.840290381125227, "step": 5290}, {"loss": 0.6818, "grad_norm": 0.7426059246063232, "learning_rate": 0.0002, "epoch": 3.8475499092558985, "step": 5300}, {"loss": 0.6748, "grad_norm": 1.003413438796997, "learning_rate": 0.0002, "epoch": 3.85480943738657, "step": 5310}, {"loss": 0.8554, "grad_norm": 0.7527840733528137, "learning_rate": 0.0002, "epoch": 3.862068965517241, "step": 5320}, {"loss": 0.7521, "grad_norm": 0.738610565662384, "learning_rate": 0.0002, "epoch": 3.869328493647913, "step": 5330}, {"loss": 0.7266, "grad_norm": 0.7277999520301819, "learning_rate": 0.0002, "epoch": 3.876588021778584, "step": 5340}, {"loss": 0.7503, "grad_norm": 0.5951359272003174, "learning_rate": 0.0002, "epoch": 3.8838475499092557, "step": 5350}, {"loss": 0.7447, "grad_norm": 1.043884038925171, "learning_rate": 0.0002, "epoch": 3.8911070780399273, "step": 5360}, {"loss": 0.6862, "grad_norm": 0.8436498045921326, "learning_rate": 0.0002, "epoch": 3.898366606170599, "step": 5370}, {"loss": 0.665, "grad_norm": 0.5603365302085876, "learning_rate": 0.0002, "epoch": 3.9056261343012704, "step": 5380}, {"loss": 0.7098, "grad_norm": 1.0128886699676514, "learning_rate": 0.0002, "epoch": 3.912885662431942, "step": 5390}, {"loss": 0.6707, "grad_norm": 0.7970930337905884, "learning_rate": 0.0002, "epoch": 3.9201451905626135, "step": 5400}, {"loss": 0.637, "grad_norm": 0.7699369192123413, "learning_rate": 0.0002, "epoch": 3.927404718693285, "step": 5410}, {"loss": 0.6742, "grad_norm": 0.800561249256134, "learning_rate": 0.0002, "epoch": 3.9346642468239565, "step": 5420}, {"loss": 0.7208, "grad_norm": 0.8020331859588623, "learning_rate": 0.0002, "epoch": 3.941923774954628, "step": 5430}, {"loss": 0.7294, "grad_norm": 0.7461140155792236, "learning_rate": 0.0002, "epoch": 3.9491833030852996, "step": 5440}, {"loss": 0.7013, "grad_norm": 0.8346918821334839, "learning_rate": 0.0002, "epoch": 3.9564428312159707, "step": 5450}, {"loss": 0.6289, "grad_norm": 0.9723302125930786, "learning_rate": 0.0002, "epoch": 3.9637023593466427, "step": 5460}, {"loss": 0.8029, "grad_norm": 0.6809740662574768, "learning_rate": 0.0002, "epoch": 3.970961887477314, "step": 5470}, {"loss": 0.6896, "grad_norm": 0.7353498339653015, "learning_rate": 0.0002, "epoch": 3.9782214156079854, "step": 5480}, {"loss": 0.6722, "grad_norm": 0.748009443283081, "learning_rate": 0.0002, "epoch": 3.985480943738657, "step": 5490}, {"loss": 0.6866, "grad_norm": 1.3656195402145386, "learning_rate": 0.0002, "epoch": 3.9927404718693285, "step": 5500}, {"loss": 0.7368, "grad_norm": 0.8402108550071716, "learning_rate": 0.0002, "epoch": 4.0, "step": 5510}, {"eval_loss": 1.17229425907135, "eval_runtime": 46.2554, "eval_samples_per_second": 9.426, "eval_steps_per_second": 1.189, "epoch": 4.0, "step": 5510}, {"loss": 0.4637, "grad_norm": 0.8601235747337341, "learning_rate": 0.0002, "epoch": 4.007259528130671, "step": 5520}, {"loss": 0.4717, "grad_norm": 1.2635200023651123, "learning_rate": 0.0002, "epoch": 4.014519056261343, "step": 5530}, {"loss": 0.503, "grad_norm": 1.0257477760314941, "learning_rate": 0.0002, "epoch": 4.021778584392014, "step": 5540}, {"loss": 0.4547, "grad_norm": 0.9436745047569275, "learning_rate": 0.0002, "epoch": 4.029038112522686, "step": 5550}, {"loss": 0.459, "grad_norm": 0.9443606734275818, "learning_rate": 0.0002, "epoch": 4.036297640653357, "step": 5560}, {"loss": 0.5386, "grad_norm": 1.3965742588043213, "learning_rate": 0.0002, "epoch": 4.043557168784029, "step": 5570}, {"loss": 0.4248, "grad_norm": 0.8973520398139954, "learning_rate": 0.0002, "epoch": 4.0508166969147, "step": 5580}, {"loss": 0.4111, "grad_norm": 0.9998409748077393, "learning_rate": 0.0002, "epoch": 4.058076225045372, "step": 5590}, {"loss": 0.4828, "grad_norm": 1.1213387250900269, "learning_rate": 0.0002, "epoch": 4.0653357531760435, "step": 5600}, {"loss": 0.439, "grad_norm": 0.7064558863639832, "learning_rate": 0.0002, "epoch": 4.072595281306715, "step": 5610}, {"loss": 0.4607, "grad_norm": 1.2390803098678589, "learning_rate": 0.0002, "epoch": 4.0798548094373865, "step": 5620}, {"loss": 0.5014, "grad_norm": 1.123469591140747, "learning_rate": 0.0002, "epoch": 4.087114337568058, "step": 5630}, {"loss": 0.513, "grad_norm": 1.229573369026184, "learning_rate": 0.0002, "epoch": 4.09437386569873, "step": 5640}, {"loss": 0.5258, "grad_norm": 1.7182831764221191, "learning_rate": 0.0002, "epoch": 4.101633393829401, "step": 5650}, {"loss": 0.5371, "grad_norm": 0.894903302192688, "learning_rate": 0.0002, "epoch": 4.108892921960073, "step": 5660}, {"loss": 0.4813, "grad_norm": 0.8754552006721497, "learning_rate": 0.0002, "epoch": 4.116152450090744, "step": 5670}, {"loss": 0.491, "grad_norm": 1.2401553392410278, "learning_rate": 0.0002, "epoch": 4.123411978221416, "step": 5680}, {"loss": 0.4549, "grad_norm": 0.8631148934364319, "learning_rate": 0.0002, "epoch": 4.130671506352087, "step": 5690}, {"loss": 0.487, "grad_norm": 1.1798022985458374, "learning_rate": 0.0002, "epoch": 4.137931034482759, "step": 5700}, {"loss": 0.4522, "grad_norm": 0.8344549536705017, "learning_rate": 0.0002, "epoch": 4.14519056261343, "step": 5710}, {"loss": 0.4559, "grad_norm": 1.2342697381973267, "learning_rate": 0.0002, "epoch": 4.152450090744102, "step": 5720}, {"loss": 0.53, "grad_norm": 1.1601094007492065, "learning_rate": 0.0002, "epoch": 4.159709618874773, "step": 5730}, {"loss": 0.4755, "grad_norm": 1.2925703525543213, "learning_rate": 0.0002, "epoch": 4.166969147005445, "step": 5740}, {"loss": 0.4973, "grad_norm": 1.0870997905731201, "learning_rate": 0.0002, "epoch": 4.174228675136116, "step": 5750}, {"loss": 0.5184, "grad_norm": 0.9077792763710022, "learning_rate": 0.0002, "epoch": 4.181488203266787, "step": 5760}, {"loss": 0.4606, "grad_norm": 1.009273886680603, "learning_rate": 0.0002, "epoch": 4.188747731397459, "step": 5770}, {"loss": 0.5383, "grad_norm": 1.2465530633926392, "learning_rate": 0.0002, "epoch": 4.19600725952813, "step": 5780}, {"loss": 0.4938, "grad_norm": 1.2261253595352173, "learning_rate": 0.0002, "epoch": 4.203266787658802, "step": 5790}, {"loss": 0.5255, "grad_norm": 1.1498041152954102, "learning_rate": 0.0002, "epoch": 4.2105263157894735, "step": 5800}, {"loss": 0.5269, "grad_norm": 1.1966725587844849, "learning_rate": 0.0002, "epoch": 4.217785843920145, "step": 5810}, {"loss": 0.5626, "grad_norm": 1.2651296854019165, "learning_rate": 0.0002, "epoch": 4.2250453720508165, "step": 5820}, {"loss": 0.5213, "grad_norm": 1.0388574600219727, "learning_rate": 0.0002, "epoch": 4.2323049001814885, "step": 5830}, {"loss": 0.4965, "grad_norm": 1.3042771816253662, "learning_rate": 0.0002, "epoch": 4.23956442831216, "step": 5840}, {"loss": 0.5116, "grad_norm": 1.1127727031707764, "learning_rate": 0.0002, "epoch": 4.246823956442832, "step": 5850}, {"loss": 0.5197, "grad_norm": 0.9653958082199097, "learning_rate": 0.0002, "epoch": 4.254083484573503, "step": 5860}, {"loss": 0.4414, "grad_norm": 1.0500504970550537, "learning_rate": 0.0002, "epoch": 4.261343012704174, "step": 5870}, {"loss": 0.544, "grad_norm": 1.1476165056228638, "learning_rate": 0.0002, "epoch": 4.268602540834846, "step": 5880}, {"loss": 0.4667, "grad_norm": 0.9424414038658142, "learning_rate": 0.0002, "epoch": 4.275862068965517, "step": 5890}, {"loss": 0.5039, "grad_norm": 1.3309166431427002, "learning_rate": 0.0002, "epoch": 4.283121597096189, "step": 5900}, {"loss": 0.5472, "grad_norm": 1.3025873899459839, "learning_rate": 0.0002, "epoch": 4.29038112522686, "step": 5910}, {"loss": 0.4644, "grad_norm": 1.1442325115203857, "learning_rate": 0.0002, "epoch": 4.297640653357532, "step": 5920}, {"loss": 0.5066, "grad_norm": 0.9820859432220459, "learning_rate": 0.0002, "epoch": 4.304900181488203, "step": 5930}, {"loss": 0.5224, "grad_norm": 0.9615740180015564, "learning_rate": 0.0002, "epoch": 4.312159709618875, "step": 5940}, {"loss": 0.5665, "grad_norm": 1.1627109050750732, "learning_rate": 0.0002, "epoch": 4.319419237749546, "step": 5950}, {"loss": 0.4852, "grad_norm": 0.9381322860717773, "learning_rate": 0.0002, "epoch": 4.326678765880218, "step": 5960}, {"loss": 0.4532, "grad_norm": 0.8154335618019104, "learning_rate": 0.0002, "epoch": 4.333938294010889, "step": 5970}, {"loss": 0.5237, "grad_norm": 0.877671480178833, "learning_rate": 0.0002, "epoch": 4.341197822141561, "step": 5980}, {"loss": 0.6113, "grad_norm": 1.1742031574249268, "learning_rate": 0.0002, "epoch": 4.348457350272232, "step": 5990}, {"loss": 0.5704, "grad_norm": 1.0352917909622192, "learning_rate": 0.0002, "epoch": 4.3557168784029034, "step": 6000}, {"loss": 0.4996, "grad_norm": 0.9963878989219666, "learning_rate": 0.0002, "epoch": 4.362976406533575, "step": 6010}, {"loss": 0.4813, "grad_norm": 1.1892237663269043, "learning_rate": 0.0002, "epoch": 4.3702359346642465, "step": 6020}, {"loss": 0.5224, "grad_norm": 1.2516111135482788, "learning_rate": 0.0002, "epoch": 4.3774954627949185, "step": 6030}, {"loss": 0.5952, "grad_norm": 1.2111951112747192, "learning_rate": 0.0002, "epoch": 4.38475499092559, "step": 6040}, {"loss": 0.4275, "grad_norm": 1.0820083618164062, "learning_rate": 0.0002, "epoch": 4.392014519056262, "step": 6050}, {"loss": 0.5117, "grad_norm": 1.033915638923645, "learning_rate": 0.0002, "epoch": 4.399274047186933, "step": 6060}, {"loss": 0.5431, "grad_norm": 1.0635870695114136, "learning_rate": 0.0002, "epoch": 4.406533575317605, "step": 6070}, {"loss": 0.5341, "grad_norm": 1.0520414113998413, "learning_rate": 0.0002, "epoch": 4.413793103448276, "step": 6080}, {"loss": 0.512, "grad_norm": 1.0821926593780518, "learning_rate": 0.0002, "epoch": 4.421052631578947, "step": 6090}, {"loss": 0.5065, "grad_norm": 1.0533246994018555, "learning_rate": 0.0002, "epoch": 4.428312159709619, "step": 6100}, {"loss": 0.4577, "grad_norm": 0.9231932759284973, "learning_rate": 0.0002, "epoch": 4.43557168784029, "step": 6110}, {"loss": 0.583, "grad_norm": 0.9910260438919067, "learning_rate": 0.0002, "epoch": 4.442831215970962, "step": 6120}, {"loss": 0.4717, "grad_norm": 1.061949372291565, "learning_rate": 0.0002, "epoch": 4.450090744101633, "step": 6130}, {"loss": 0.5893, "grad_norm": 1.2927039861679077, "learning_rate": 0.0002, "epoch": 4.457350272232305, "step": 6140}, {"loss": 0.4684, "grad_norm": 1.3966081142425537, "learning_rate": 0.0002, "epoch": 4.464609800362976, "step": 6150}, {"loss": 0.5507, "grad_norm": 1.3835992813110352, "learning_rate": 0.0002, "epoch": 4.471869328493648, "step": 6160}, {"loss": 0.5911, "grad_norm": 1.0892692804336548, "learning_rate": 0.0002, "epoch": 4.479128856624319, "step": 6170}, {"loss": 0.478, "grad_norm": 1.0318800210952759, "learning_rate": 0.0002, "epoch": 4.486388384754991, "step": 6180}, {"loss": 0.5198, "grad_norm": 0.8174677491188049, "learning_rate": 0.0002, "epoch": 4.493647912885662, "step": 6190}, {"loss": 0.5387, "grad_norm": 1.4157509803771973, "learning_rate": 0.0002, "epoch": 4.500907441016334, "step": 6200}, {"loss": 0.5868, "grad_norm": 1.5244114398956299, "learning_rate": 0.0002, "epoch": 4.508166969147005, "step": 6210}, {"loss": 0.4642, "grad_norm": 0.8164850473403931, "learning_rate": 0.0002, "epoch": 4.5154264972776765, "step": 6220}, {"loss": 0.522, "grad_norm": 1.2904746532440186, "learning_rate": 0.0002, "epoch": 4.5226860254083485, "step": 6230}, {"loss": 0.5103, "grad_norm": 0.7987732887268066, "learning_rate": 0.0002, "epoch": 4.52994555353902, "step": 6240}, {"loss": 0.4615, "grad_norm": 0.831040620803833, "learning_rate": 0.0002, "epoch": 4.537205081669692, "step": 6250}, {"loss": 0.5065, "grad_norm": 0.9545485973358154, "learning_rate": 0.0002, "epoch": 4.544464609800363, "step": 6260}, {"loss": 0.5515, "grad_norm": 0.9291793704032898, "learning_rate": 0.0002, "epoch": 4.551724137931035, "step": 6270}, {"loss": 0.4535, "grad_norm": 0.8977208733558655, "learning_rate": 0.0002, "epoch": 4.558983666061706, "step": 6280}, {"loss": 0.544, "grad_norm": 1.1768537759780884, "learning_rate": 0.0002, "epoch": 4.566243194192378, "step": 6290}, {"loss": 0.5925, "grad_norm": 1.0688952207565308, "learning_rate": 0.0002, "epoch": 4.573502722323049, "step": 6300}, {"loss": 0.5207, "grad_norm": 0.8800966739654541, "learning_rate": 0.0002, "epoch": 4.580762250453721, "step": 6310}, {"loss": 0.6106, "grad_norm": 1.0911834239959717, "learning_rate": 0.0002, "epoch": 4.588021778584392, "step": 6320}, {"loss": 0.5109, "grad_norm": 1.1420872211456299, "learning_rate": 0.0002, "epoch": 4.595281306715064, "step": 6330}, {"loss": 0.5147, "grad_norm": 1.0215224027633667, "learning_rate": 0.0002, "epoch": 4.602540834845735, "step": 6340}, {"loss": 0.592, "grad_norm": 0.9685489535331726, "learning_rate": 0.0002, "epoch": 4.609800362976406, "step": 6350}, {"loss": 0.5775, "grad_norm": 1.12773597240448, "learning_rate": 0.0002, "epoch": 4.617059891107078, "step": 6360}, {"loss": 0.5966, "grad_norm": 1.0663973093032837, "learning_rate": 0.0002, "epoch": 4.624319419237749, "step": 6370}, {"loss": 0.512, "grad_norm": 1.1707262992858887, "learning_rate": 0.0002, "epoch": 4.631578947368421, "step": 6380}, {"loss": 0.5497, "grad_norm": 1.0672980546951294, "learning_rate": 0.0002, "epoch": 4.638838475499092, "step": 6390}, {"loss": 0.5699, "grad_norm": 1.1464333534240723, "learning_rate": 0.0002, "epoch": 4.646098003629764, "step": 6400}, {"loss": 0.5514, "grad_norm": 1.070230484008789, "learning_rate": 0.0002, "epoch": 4.653357531760435, "step": 6410}, {"loss": 0.5013, "grad_norm": 0.9673764109611511, "learning_rate": 0.0002, "epoch": 4.660617059891107, "step": 6420}, {"loss": 0.5901, "grad_norm": 1.0189043283462524, "learning_rate": 0.0002, "epoch": 4.6678765880217785, "step": 6430}, {"loss": 0.5193, "grad_norm": 1.185896396636963, "learning_rate": 0.0002, "epoch": 4.67513611615245, "step": 6440}, {"loss": 0.5318, "grad_norm": 1.0682812929153442, "learning_rate": 0.0002, "epoch": 4.682395644283122, "step": 6450}, {"loss": 0.5773, "grad_norm": 1.3586071729660034, "learning_rate": 0.0002, "epoch": 4.689655172413794, "step": 6460}, {"loss": 0.5482, "grad_norm": 0.6561792492866516, "learning_rate": 0.0002, "epoch": 4.696914700544465, "step": 6470}, {"loss": 0.5711, "grad_norm": 1.1394113302230835, "learning_rate": 0.0002, "epoch": 4.704174228675136, "step": 6480}, {"loss": 0.5325, "grad_norm": 0.9683151245117188, "learning_rate": 0.0002, "epoch": 4.711433756805808, "step": 6490}, {"loss": 0.5139, "grad_norm": 1.0247553586959839, "learning_rate": 0.0002, "epoch": 4.718693284936479, "step": 6500}, {"loss": 0.5794, "grad_norm": 0.8046169281005859, "learning_rate": 0.0002, "epoch": 4.725952813067151, "step": 6510}, {"loss": 0.5471, "grad_norm": 1.0710240602493286, "learning_rate": 0.0002, "epoch": 4.733212341197822, "step": 6520}, {"loss": 0.4805, "grad_norm": 0.9438924193382263, "learning_rate": 0.0002, "epoch": 4.740471869328494, "step": 6530}, {"loss": 0.5404, "grad_norm": 0.869162380695343, "learning_rate": 0.0002, "epoch": 4.747731397459165, "step": 6540}, {"loss": 0.6379, "grad_norm": 0.9776787161827087, "learning_rate": 0.0002, "epoch": 4.754990925589837, "step": 6550}, {"loss": 0.5288, "grad_norm": 1.1990505456924438, "learning_rate": 0.0002, "epoch": 4.762250453720508, "step": 6560}, {"loss": 0.5539, "grad_norm": 1.0582209825515747, "learning_rate": 0.0002, "epoch": 4.769509981851179, "step": 6570}, {"loss": 0.489, "grad_norm": 0.9966367483139038, "learning_rate": 0.0002, "epoch": 4.776769509981851, "step": 6580}, {"loss": 0.5514, "grad_norm": 0.9130612015724182, "learning_rate": 0.0002, "epoch": 4.784029038112522, "step": 6590}, {"loss": 0.5864, "grad_norm": 1.0950500965118408, "learning_rate": 0.0002, "epoch": 4.791288566243194, "step": 6600}, {"loss": 0.5266, "grad_norm": 1.108681321144104, "learning_rate": 0.0002, "epoch": 4.798548094373865, "step": 6610}, {"loss": 0.5875, "grad_norm": 1.1873763799667358, "learning_rate": 0.0002, "epoch": 4.805807622504537, "step": 6620}, {"loss": 0.5736, "grad_norm": 1.305367112159729, "learning_rate": 0.0002, "epoch": 4.8130671506352085, "step": 6630}, {"loss": 0.5636, "grad_norm": 1.2801482677459717, "learning_rate": 0.0002, "epoch": 4.8203266787658805, "step": 6640}, {"loss": 0.582, "grad_norm": 1.26764976978302, "learning_rate": 0.0002, "epoch": 4.827586206896552, "step": 6650}, {"loss": 0.5259, "grad_norm": 1.0018208026885986, "learning_rate": 0.0002, "epoch": 4.834845735027224, "step": 6660}, {"loss": 0.548, "grad_norm": 1.2326326370239258, "learning_rate": 0.0002, "epoch": 4.842105263157895, "step": 6670}, {"loss": 0.5933, "grad_norm": 0.9707282781600952, "learning_rate": 0.0002, "epoch": 4.849364791288567, "step": 6680}, {"loss": 0.5612, "grad_norm": 1.2772048711776733, "learning_rate": 0.0002, "epoch": 4.856624319419238, "step": 6690}, {"loss": 0.5346, "grad_norm": 2.6652262210845947, "learning_rate": 0.0002, "epoch": 4.863883847549909, "step": 6700}, {"loss": 0.5428, "grad_norm": 1.215828537940979, "learning_rate": 0.0002, "epoch": 4.871143375680581, "step": 6710}, {"loss": 0.6571, "grad_norm": 1.3704510927200317, "learning_rate": 0.0002, "epoch": 4.878402903811252, "step": 6720}, {"loss": 0.4963, "grad_norm": 0.7781757116317749, "learning_rate": 0.0002, "epoch": 4.885662431941924, "step": 6730}, {"loss": 0.5989, "grad_norm": 1.1883646249771118, "learning_rate": 0.0002, "epoch": 4.892921960072595, "step": 6740}, {"loss": 0.6067, "grad_norm": 0.9216066002845764, "learning_rate": 0.0002, "epoch": 4.900181488203267, "step": 6750}, {"loss": 0.5085, "grad_norm": 1.0558464527130127, "learning_rate": 0.0002, "epoch": 4.907441016333938, "step": 6760}, {"loss": 0.5216, "grad_norm": 1.032656192779541, "learning_rate": 0.0002, "epoch": 4.91470054446461, "step": 6770}, {"loss": 0.5426, "grad_norm": 1.1261441707611084, "learning_rate": 0.0002, "epoch": 4.921960072595281, "step": 6780}, {"loss": 0.5295, "grad_norm": 1.2178640365600586, "learning_rate": 0.0002, "epoch": 4.929219600725952, "step": 6790}, {"loss": 0.5476, "grad_norm": 1.5369361639022827, "learning_rate": 0.0002, "epoch": 4.936479128856624, "step": 6800}, {"loss": 0.5358, "grad_norm": 1.1188377141952515, "learning_rate": 0.0002, "epoch": 4.943738656987296, "step": 6810}, {"loss": 0.5483, "grad_norm": 1.2506113052368164, "learning_rate": 0.0002, "epoch": 4.950998185117967, "step": 6820}, {"loss": 0.567, "grad_norm": 0.8776047825813293, "learning_rate": 0.0002, "epoch": 4.9582577132486385, "step": 6830}, {"loss": 0.5764, "grad_norm": 0.9700555205345154, "learning_rate": 0.0002, "epoch": 4.9655172413793105, "step": 6840}, {"loss": 0.5396, "grad_norm": 1.2713534832000732, "learning_rate": 0.0002, "epoch": 4.972776769509982, "step": 6850}, {"loss": 0.5451, "grad_norm": 0.9855955243110657, "learning_rate": 0.0002, "epoch": 4.980036297640654, "step": 6860}, {"loss": 0.5884, "grad_norm": 0.8734853863716125, "learning_rate": 0.0002, "epoch": 4.987295825771325, "step": 6870}, {"loss": 0.5189, "grad_norm": 0.8065403699874878, "learning_rate": 0.0002, "epoch": 4.994555353901997, "step": 6880}]}
+{"epoch": 6.0, "step": 8265, "epoch_duration": 2057.1978249549866, "total_accumulated_duration": 18403.34801387787, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4936, "grad_norm": 0.46010470390319824, "learning_rate": 0.0002, "epoch": 0.007259528130671506, "step": 10}, {"loss": 1.8442, "grad_norm": 0.5103179216384888, "learning_rate": 0.0002, "epoch": 0.014519056261343012, "step": 20}, {"loss": 1.559, "grad_norm": 0.6282716989517212, "learning_rate": 0.0002, "epoch": 0.021778584392014518, "step": 30}, {"loss": 1.3618, "grad_norm": 1.2480497360229492, "learning_rate": 0.0002, "epoch": 0.029038112522686024, "step": 40}, {"loss": 1.3874, "grad_norm": 0.4114855229854584, "learning_rate": 0.0002, "epoch": 0.036297640653357534, "step": 50}, {"loss": 1.2836, "grad_norm": 0.49482840299606323, "learning_rate": 0.0002, "epoch": 0.043557168784029036, "step": 60}, {"loss": 1.1728, "grad_norm": 0.4536272883415222, "learning_rate": 0.0002, "epoch": 0.050816696914700546, "step": 70}, {"loss": 1.1125, "grad_norm": 0.32328274846076965, "learning_rate": 0.0002, "epoch": 0.05807622504537205, "step": 80}, {"loss": 1.2447, "grad_norm": 0.40990468859672546, "learning_rate": 0.0002, "epoch": 0.06533575317604355, "step": 90}, {"loss": 1.2216, "grad_norm": 0.37273502349853516, "learning_rate": 0.0002, "epoch": 0.07259528130671507, "step": 100}, {"loss": 1.2259, "grad_norm": 0.3903871476650238, "learning_rate": 0.0002, "epoch": 0.07985480943738657, "step": 110}, {"loss": 1.3718, "grad_norm": 0.3272787928581238, "learning_rate": 0.0002, "epoch": 0.08711433756805807, "step": 120}, {"loss": 1.1832, "grad_norm": 0.3622824251651764, "learning_rate": 0.0002, "epoch": 0.09437386569872959, "step": 130}, {"loss": 1.2966, "grad_norm": 0.3503916561603546, "learning_rate": 0.0002, "epoch": 0.10163339382940109, "step": 140}, {"loss": 1.0913, "grad_norm": 0.32787832617759705, "learning_rate": 0.0002, "epoch": 0.1088929219600726, "step": 150}, {"loss": 1.2143, "grad_norm": 0.5822657942771912, "learning_rate": 0.0002, "epoch": 0.1161524500907441, "step": 160}, {"loss": 1.2091, "grad_norm": 0.28028249740600586, "learning_rate": 0.0002, "epoch": 0.12341197822141561, "step": 170}, {"loss": 1.1731, "grad_norm": 0.33602750301361084, "learning_rate": 0.0002, "epoch": 0.1306715063520871, "step": 180}, {"loss": 1.127, "grad_norm": 0.29106274247169495, "learning_rate": 0.0002, "epoch": 0.13793103448275862, "step": 190}, {"loss": 1.3261, "grad_norm": 0.38753363490104675, "learning_rate": 0.0002, "epoch": 0.14519056261343014, "step": 200}, {"loss": 1.1536, "grad_norm": 0.361009418964386, "learning_rate": 0.0002, "epoch": 0.15245009074410162, "step": 210}, {"loss": 1.1483, "grad_norm": 0.6743836402893066, "learning_rate": 0.0002, "epoch": 0.15970961887477314, "step": 220}, {"loss": 1.264, "grad_norm": 0.3923613727092743, "learning_rate": 0.0002, "epoch": 0.16696914700544466, "step": 230}, {"loss": 1.0437, "grad_norm": 0.2809699773788452, "learning_rate": 0.0002, "epoch": 0.17422867513611615, "step": 240}, {"loss": 1.1826, "grad_norm": 0.3631494641304016, "learning_rate": 0.0002, "epoch": 0.18148820326678766, "step": 250}, {"loss": 1.1584, "grad_norm": 0.24658171832561493, "learning_rate": 0.0002, "epoch": 0.18874773139745918, "step": 260}, {"loss": 1.0065, "grad_norm": 0.5780664682388306, "learning_rate": 0.0002, "epoch": 0.19600725952813067, "step": 270}, {"loss": 1.1251, "grad_norm": 0.3056720495223999, "learning_rate": 0.0002, "epoch": 0.20326678765880218, "step": 280}, {"loss": 1.1503, "grad_norm": 0.2857084572315216, "learning_rate": 0.0002, "epoch": 0.21052631578947367, "step": 290}, {"loss": 1.1215, "grad_norm": 0.6645345687866211, "learning_rate": 0.0002, "epoch": 0.2177858439201452, "step": 300}, {"loss": 1.127, "grad_norm": 0.5966078639030457, "learning_rate": 0.0002, "epoch": 0.2250453720508167, "step": 310}, {"loss": 1.1974, "grad_norm": 0.40937140583992004, "learning_rate": 0.0002, "epoch": 0.2323049001814882, "step": 320}, {"loss": 1.2385, "grad_norm": 0.5642806887626648, "learning_rate": 0.0002, "epoch": 0.2395644283121597, "step": 330}, {"loss": 1.1703, "grad_norm": 0.2750748097896576, "learning_rate": 0.0002, "epoch": 0.24682395644283123, "step": 340}, {"loss": 1.2598, "grad_norm": 0.34350234270095825, "learning_rate": 0.0002, "epoch": 0.2540834845735027, "step": 350}, {"loss": 1.1942, "grad_norm": 0.6767239570617676, "learning_rate": 0.0002, "epoch": 0.2613430127041742, "step": 360}, {"loss": 1.1436, "grad_norm": 0.31006959080696106, "learning_rate": 0.0002, "epoch": 0.26860254083484575, "step": 370}, {"loss": 1.2084, "grad_norm": 0.3825474679470062, "learning_rate": 0.0002, "epoch": 0.27586206896551724, "step": 380}, {"loss": 1.1523, "grad_norm": 0.30241551995277405, "learning_rate": 0.0002, "epoch": 0.2831215970961887, "step": 390}, {"loss": 1.1298, "grad_norm": 0.2962397336959839, "learning_rate": 0.0002, "epoch": 0.29038112522686027, "step": 400}, {"loss": 1.1299, "grad_norm": 0.2600369155406952, "learning_rate": 0.0002, "epoch": 0.29764065335753176, "step": 410}, {"loss": 1.1366, "grad_norm": 0.3675060272216797, "learning_rate": 0.0002, "epoch": 0.30490018148820325, "step": 420}, {"loss": 1.156, "grad_norm": 0.3429498076438904, "learning_rate": 0.0002, "epoch": 0.3121597096188748, "step": 430}, {"loss": 1.2741, "grad_norm": 0.34311825037002563, "learning_rate": 0.0002, "epoch": 0.3194192377495463, "step": 440}, {"loss": 1.3523, "grad_norm": 0.37872210144996643, "learning_rate": 0.0002, "epoch": 0.32667876588021777, "step": 450}, {"loss": 1.1365, "grad_norm": 0.33271121978759766, "learning_rate": 0.0002, "epoch": 0.3339382940108893, "step": 460}, {"loss": 1.1572, "grad_norm": 0.34605276584625244, "learning_rate": 0.0002, "epoch": 0.3411978221415608, "step": 470}, {"loss": 1.2755, "grad_norm": 0.41050562262535095, "learning_rate": 0.0002, "epoch": 0.3484573502722323, "step": 480}, {"loss": 1.075, "grad_norm": 0.2066836953163147, "learning_rate": 0.0002, "epoch": 0.35571687840290384, "step": 490}, {"loss": 1.2072, "grad_norm": 0.2859014868736267, "learning_rate": 0.0002, "epoch": 0.3629764065335753, "step": 500}, {"loss": 1.1435, "grad_norm": 0.28763777017593384, "learning_rate": 0.0002, "epoch": 0.3702359346642468, "step": 510}, {"loss": 1.1341, "grad_norm": 0.2730471193790436, "learning_rate": 0.0002, "epoch": 0.37749546279491836, "step": 520}, {"loss": 1.348, "grad_norm": 0.3968936800956726, "learning_rate": 0.0002, "epoch": 0.38475499092558985, "step": 530}, {"loss": 1.2363, "grad_norm": 0.3624701201915741, "learning_rate": 0.0002, "epoch": 0.39201451905626133, "step": 540}, {"loss": 1.1206, "grad_norm": 0.3303608298301697, "learning_rate": 0.0002, "epoch": 0.3992740471869328, "step": 550}, {"loss": 1.203, "grad_norm": 0.33507466316223145, "learning_rate": 0.0002, "epoch": 0.40653357531760437, "step": 560}, {"loss": 1.2294, "grad_norm": 0.3297670781612396, "learning_rate": 0.0002, "epoch": 0.41379310344827586, "step": 570}, {"loss": 1.2, "grad_norm": 0.32334890961647034, "learning_rate": 0.0002, "epoch": 0.42105263157894735, "step": 580}, {"loss": 1.112, "grad_norm": 0.30281195044517517, "learning_rate": 0.0002, "epoch": 0.4283121597096189, "step": 590}, {"loss": 1.1352, "grad_norm": 0.5900027751922607, "learning_rate": 0.0002, "epoch": 0.4355716878402904, "step": 600}, {"loss": 1.1575, "grad_norm": 0.28437477350234985, "learning_rate": 0.0002, "epoch": 0.44283121597096187, "step": 610}, {"loss": 1.2009, "grad_norm": 0.39601704478263855, "learning_rate": 0.0002, "epoch": 0.4500907441016334, "step": 620}, {"loss": 1.2302, "grad_norm": 0.41971510648727417, "learning_rate": 0.0002, "epoch": 0.4573502722323049, "step": 630}, {"loss": 1.236, "grad_norm": 0.33814409375190735, "learning_rate": 0.0002, "epoch": 0.4646098003629764, "step": 640}, {"loss": 1.2471, "grad_norm": 0.575718104839325, "learning_rate": 0.0002, "epoch": 0.47186932849364793, "step": 650}, {"loss": 1.1012, "grad_norm": 0.37927401065826416, "learning_rate": 0.0002, "epoch": 0.4791288566243194, "step": 660}, {"loss": 1.1552, "grad_norm": 0.3224332630634308, "learning_rate": 0.0002, "epoch": 0.4863883847549909, "step": 670}, {"loss": 1.1354, "grad_norm": 0.32683515548706055, "learning_rate": 0.0002, "epoch": 0.49364791288566245, "step": 680}, {"loss": 1.207, "grad_norm": 0.4316163659095764, "learning_rate": 0.0002, "epoch": 0.5009074410163339, "step": 690}, {"loss": 1.2116, "grad_norm": 0.342602401971817, "learning_rate": 0.0002, "epoch": 0.5081669691470054, "step": 700}, {"loss": 1.0433, "grad_norm": 0.2794898748397827, "learning_rate": 0.0002, "epoch": 0.515426497277677, "step": 710}, {"loss": 1.2395, "grad_norm": 0.3322339951992035, "learning_rate": 0.0002, "epoch": 0.5226860254083484, "step": 720}, {"loss": 1.0762, "grad_norm": 0.3088509142398834, "learning_rate": 0.0002, "epoch": 0.52994555353902, "step": 730}, {"loss": 1.1495, "grad_norm": 0.24444378912448883, "learning_rate": 0.0002, "epoch": 0.5372050816696915, "step": 740}, {"loss": 1.2176, "grad_norm": 0.3483171761035919, "learning_rate": 0.0002, "epoch": 0.5444646098003629, "step": 750}, {"loss": 1.2248, "grad_norm": 0.3425690531730652, "learning_rate": 0.0002, "epoch": 0.5517241379310345, "step": 760}, {"loss": 1.1127, "grad_norm": 0.31841927766799927, "learning_rate": 0.0002, "epoch": 0.558983666061706, "step": 770}, {"loss": 1.1841, "grad_norm": 0.39423868060112, "learning_rate": 0.0002, "epoch": 0.5662431941923775, "step": 780}, {"loss": 1.0583, "grad_norm": 0.30328479409217834, "learning_rate": 0.0002, "epoch": 0.573502722323049, "step": 790}, {"loss": 1.0718, "grad_norm": 0.24475938081741333, "learning_rate": 0.0002, "epoch": 0.5807622504537205, "step": 800}, {"loss": 1.2006, "grad_norm": 0.37132805585861206, "learning_rate": 0.0002, "epoch": 0.588021778584392, "step": 810}, {"loss": 1.1544, "grad_norm": 0.32195979356765747, "learning_rate": 0.0002, "epoch": 0.5952813067150635, "step": 820}, {"loss": 0.9937, "grad_norm": 0.2848738729953766, "learning_rate": 0.0002, "epoch": 0.6025408348457351, "step": 830}, {"loss": 1.0937, "grad_norm": 0.28015264868736267, "learning_rate": 0.0002, "epoch": 0.6098003629764065, "step": 840}, {"loss": 1.1234, "grad_norm": 0.37796008586883545, "learning_rate": 0.0002, "epoch": 0.617059891107078, "step": 850}, {"loss": 1.1455, "grad_norm": 0.39311841130256653, "learning_rate": 0.0002, "epoch": 0.6243194192377496, "step": 860}, {"loss": 1.1066, "grad_norm": 0.2761685252189636, "learning_rate": 0.0002, "epoch": 0.631578947368421, "step": 870}, {"loss": 1.0859, "grad_norm": 0.3826720118522644, "learning_rate": 0.0002, "epoch": 0.6388384754990926, "step": 880}, {"loss": 1.1182, "grad_norm": 0.30076679587364197, "learning_rate": 0.0002, "epoch": 0.6460980036297641, "step": 890}, {"loss": 1.0927, "grad_norm": 0.21997687220573425, "learning_rate": 0.0002, "epoch": 0.6533575317604355, "step": 900}, {"loss": 1.1603, "grad_norm": 0.32593777775764465, "learning_rate": 0.0002, "epoch": 0.6606170598911071, "step": 910}, {"loss": 1.1578, "grad_norm": 0.30347898602485657, "learning_rate": 0.0002, "epoch": 0.6678765880217786, "step": 920}, {"loss": 1.1642, "grad_norm": 0.44173774123191833, "learning_rate": 0.0002, "epoch": 0.6751361161524501, "step": 930}, {"loss": 1.1832, "grad_norm": 0.2507467269897461, "learning_rate": 0.0002, "epoch": 0.6823956442831216, "step": 940}, {"loss": 1.3001, "grad_norm": 0.29463833570480347, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 950}, {"loss": 1.1651, "grad_norm": 0.9363154172897339, "learning_rate": 0.0002, "epoch": 0.6969147005444646, "step": 960}, {"loss": 1.158, "grad_norm": 0.3236212134361267, "learning_rate": 0.0002, "epoch": 0.7041742286751361, "step": 970}, {"loss": 1.1078, "grad_norm": 0.3123254179954529, "learning_rate": 0.0002, "epoch": 0.7114337568058077, "step": 980}, {"loss": 1.1578, "grad_norm": 0.3395805060863495, "learning_rate": 0.0002, "epoch": 0.7186932849364791, "step": 990}, {"loss": 1.0754, "grad_norm": 0.3240964412689209, "learning_rate": 0.0002, "epoch": 0.7259528130671506, "step": 1000}, {"loss": 1.0313, "grad_norm": 0.31902948021888733, "learning_rate": 0.0002, "epoch": 0.7332123411978222, "step": 1010}, {"loss": 1.1664, "grad_norm": 0.4848408102989197, "learning_rate": 0.0002, "epoch": 0.7404718693284936, "step": 1020}, {"loss": 1.0838, "grad_norm": 0.33006033301353455, "learning_rate": 0.0002, "epoch": 0.7477313974591652, "step": 1030}, {"loss": 1.0979, "grad_norm": 0.2928730547428131, "learning_rate": 0.0002, "epoch": 0.7549909255898367, "step": 1040}, {"loss": 1.1351, "grad_norm": 0.3529164791107178, "learning_rate": 0.0002, "epoch": 0.7622504537205081, "step": 1050}, {"loss": 1.1274, "grad_norm": 0.2736213803291321, "learning_rate": 0.0002, "epoch": 0.7695099818511797, "step": 1060}, {"loss": 1.1169, "grad_norm": 0.7200686931610107, "learning_rate": 0.0002, "epoch": 0.7767695099818511, "step": 1070}, {"loss": 1.1171, "grad_norm": 0.33396708965301514, "learning_rate": 0.0002, "epoch": 0.7840290381125227, "step": 1080}, {"loss": 1.2447, "grad_norm": 1.5760449171066284, "learning_rate": 0.0002, "epoch": 0.7912885662431942, "step": 1090}, {"loss": 1.1126, "grad_norm": 0.28138381242752075, "learning_rate": 0.0002, "epoch": 0.7985480943738656, "step": 1100}, {"loss": 1.2012, "grad_norm": 0.2597472369670868, "learning_rate": 0.0002, "epoch": 0.8058076225045372, "step": 1110}, {"loss": 1.1177, "grad_norm": 0.3305445611476898, "learning_rate": 0.0002, "epoch": 0.8130671506352087, "step": 1120}, {"loss": 1.0849, "grad_norm": 0.3934599459171295, "learning_rate": 0.0002, "epoch": 0.8203266787658802, "step": 1130}, {"loss": 1.0889, "grad_norm": 0.3472191393375397, "learning_rate": 0.0002, "epoch": 0.8275862068965517, "step": 1140}, {"loss": 1.265, "grad_norm": 0.2857365906238556, "learning_rate": 0.0002, "epoch": 0.8348457350272233, "step": 1150}, {"loss": 1.03, "grad_norm": 0.3207702934741974, "learning_rate": 0.0002, "epoch": 0.8421052631578947, "step": 1160}, {"loss": 1.1669, "grad_norm": 0.3176484704017639, "learning_rate": 0.0002, "epoch": 0.8493647912885662, "step": 1170}, {"loss": 1.1386, "grad_norm": 0.40685558319091797, "learning_rate": 0.0002, "epoch": 0.8566243194192378, "step": 1180}, {"loss": 1.1383, "grad_norm": 0.31125199794769287, "learning_rate": 0.0002, "epoch": 0.8638838475499092, "step": 1190}, {"loss": 1.1512, "grad_norm": 0.7361181378364563, "learning_rate": 0.0002, "epoch": 0.8711433756805808, "step": 1200}, {"loss": 0.9983, "grad_norm": 0.33699527382850647, "learning_rate": 0.0002, "epoch": 0.8784029038112523, "step": 1210}, {"loss": 1.11, "grad_norm": 0.3315220773220062, "learning_rate": 0.0002, "epoch": 0.8856624319419237, "step": 1220}, {"loss": 1.0266, "grad_norm": 0.6256054043769836, "learning_rate": 0.0002, "epoch": 0.8929219600725953, "step": 1230}, {"loss": 1.1738, "grad_norm": 0.3692137897014618, "learning_rate": 0.0002, "epoch": 0.9001814882032668, "step": 1240}, {"loss": 1.1112, "grad_norm": 0.3538484573364258, "learning_rate": 0.0002, "epoch": 0.9074410163339383, "step": 1250}, {"loss": 1.0878, "grad_norm": 0.27863040566444397, "learning_rate": 0.0002, "epoch": 0.9147005444646098, "step": 1260}, {"loss": 1.0826, "grad_norm": 0.3322528302669525, "learning_rate": 0.0002, "epoch": 0.9219600725952813, "step": 1270}, {"loss": 1.0158, "grad_norm": 0.7553173303604126, "learning_rate": 0.0002, "epoch": 0.9292196007259528, "step": 1280}, {"loss": 1.198, "grad_norm": 0.4856191575527191, "learning_rate": 0.0002, "epoch": 0.9364791288566243, "step": 1290}, {"loss": 1.1678, "grad_norm": 0.3668074905872345, "learning_rate": 0.0002, "epoch": 0.9437386569872959, "step": 1300}, {"loss": 1.1809, "grad_norm": 0.29851067066192627, "learning_rate": 0.0002, "epoch": 0.9509981851179673, "step": 1310}, {"loss": 1.1299, "grad_norm": 0.276664674282074, "learning_rate": 0.0002, "epoch": 0.9582577132486388, "step": 1320}, {"loss": 1.0597, "grad_norm": 0.2941018342971802, "learning_rate": 0.0002, "epoch": 0.9655172413793104, "step": 1330}, {"loss": 1.1731, "grad_norm": 0.3505859076976776, "learning_rate": 0.0002, "epoch": 0.9727767695099818, "step": 1340}, {"loss": 1.1809, "grad_norm": 0.3067687451839447, "learning_rate": 0.0002, "epoch": 0.9800362976406534, "step": 1350}, {"loss": 1.1511, "grad_norm": 0.27151066064834595, "learning_rate": 0.0002, "epoch": 0.9872958257713249, "step": 1360}, {"loss": 1.1919, "grad_norm": 0.36370083689689636, "learning_rate": 0.0002, "epoch": 0.9945553539019963, "step": 1370}, {"eval_loss": 1.1381088495254517, "eval_runtime": 96.0848, "eval_samples_per_second": 4.538, "eval_steps_per_second": 0.572, "epoch": 0.9996370235934664, "step": 1377}, {"loss": 1.0082, "grad_norm": 0.27980583906173706, "learning_rate": 0.0002, "epoch": 1.0018148820326678, "step": 1380}, {"loss": 1.0565, "grad_norm": 0.26713913679122925, "learning_rate": 0.0002, "epoch": 1.0090744101633393, "step": 1390}, {"loss": 1.0241, "grad_norm": 0.3089541494846344, "learning_rate": 0.0002, "epoch": 1.0163339382940109, "step": 1400}, {"loss": 0.9649, "grad_norm": 0.4188242256641388, "learning_rate": 0.0002, "epoch": 1.0235934664246824, "step": 1410}, {"loss": 0.9821, "grad_norm": 0.5246463418006897, "learning_rate": 0.0002, "epoch": 1.030852994555354, "step": 1420}, {"loss": 1.0143, "grad_norm": 0.2728777825832367, "learning_rate": 0.0002, "epoch": 1.0381125226860255, "step": 1430}, {"loss": 1.075, "grad_norm": 0.38167616724967957, "learning_rate": 0.0002, "epoch": 1.0453720508166968, "step": 1440}, {"loss": 1.0229, "grad_norm": 0.4439380168914795, "learning_rate": 0.0002, "epoch": 1.0526315789473684, "step": 1450}, {"loss": 1.0451, "grad_norm": 0.30954182147979736, "learning_rate": 0.0002, "epoch": 1.05989110707804, "step": 1460}, {"loss": 1.0762, "grad_norm": 0.4022280275821686, "learning_rate": 0.0002, "epoch": 1.0671506352087115, "step": 1470}, {"loss": 1.0754, "grad_norm": 0.7390811443328857, "learning_rate": 0.0002, "epoch": 1.074410163339383, "step": 1480}, {"loss": 1.1378, "grad_norm": 0.3885486423969269, "learning_rate": 0.0002, "epoch": 1.0816696914700545, "step": 1490}, {"loss": 1.005, "grad_norm": 0.5275560617446899, "learning_rate": 0.0002, "epoch": 1.0889292196007259, "step": 1500}, {"loss": 1.0147, "grad_norm": 0.35112282633781433, "learning_rate": 0.0002, "epoch": 1.0961887477313974, "step": 1510}, {"loss": 1.0331, "grad_norm": 0.33714351058006287, "learning_rate": 0.0002, "epoch": 1.103448275862069, "step": 1520}, {"loss": 0.8723, "grad_norm": 0.31221693754196167, "learning_rate": 0.0002, "epoch": 1.1107078039927405, "step": 1530}, {"loss": 1.0701, "grad_norm": 0.27549654245376587, "learning_rate": 0.0002, "epoch": 1.117967332123412, "step": 1540}, {"loss": 1.0114, "grad_norm": 0.8465521335601807, "learning_rate": 0.0002, "epoch": 1.1252268602540836, "step": 1550}, {"loss": 0.9643, "grad_norm": 0.36125949025154114, "learning_rate": 0.0002, "epoch": 1.132486388384755, "step": 1560}, {"loss": 1.0744, "grad_norm": 0.37420371174812317, "learning_rate": 0.0002, "epoch": 1.1397459165154264, "step": 1570}, {"loss": 1.165, "grad_norm": 0.3294760584831238, "learning_rate": 0.0002, "epoch": 1.147005444646098, "step": 1580}, {"loss": 0.9207, "grad_norm": 0.3881238102912903, "learning_rate": 0.0002, "epoch": 1.1542649727767695, "step": 1590}, {"loss": 1.0983, "grad_norm": 0.4766491651535034, "learning_rate": 0.0002, "epoch": 1.161524500907441, "step": 1600}, {"loss": 1.0092, "grad_norm": 0.395530104637146, "learning_rate": 0.0002, "epoch": 1.1687840290381124, "step": 1610}, {"loss": 0.9895, "grad_norm": 0.3297106623649597, "learning_rate": 0.0002, "epoch": 1.176043557168784, "step": 1620}, {"loss": 1.1121, "grad_norm": 0.39528271555900574, "learning_rate": 0.0002, "epoch": 1.1833030852994555, "step": 1630}, {"loss": 0.9202, "grad_norm": 0.3370221257209778, "learning_rate": 0.0002, "epoch": 1.190562613430127, "step": 1640}, {"loss": 1.102, "grad_norm": 0.31922030448913574, "learning_rate": 0.0002, "epoch": 1.1978221415607986, "step": 1650}, {"loss": 1.0565, "grad_norm": 0.6142027378082275, "learning_rate": 0.0002, "epoch": 1.2050816696914701, "step": 1660}, {"loss": 0.9541, "grad_norm": 0.44769710302352905, "learning_rate": 0.0002, "epoch": 1.2123411978221417, "step": 1670}, {"loss": 1.2501, "grad_norm": 0.41233646869659424, "learning_rate": 0.0002, "epoch": 1.219600725952813, "step": 1680}, {"loss": 1.0147, "grad_norm": 0.2928866147994995, "learning_rate": 0.0002, "epoch": 1.2268602540834845, "step": 1690}, {"loss": 0.9074, "grad_norm": 0.36913734674453735, "learning_rate": 0.0002, "epoch": 1.234119782214156, "step": 1700}, {"loss": 1.0847, "grad_norm": 0.5281891226768494, "learning_rate": 0.0002, "epoch": 1.2413793103448276, "step": 1710}, {"loss": 0.9482, "grad_norm": 0.3374697268009186, "learning_rate": 0.0002, "epoch": 1.2486388384754992, "step": 1720}, {"loss": 0.9826, "grad_norm": 0.3802020847797394, "learning_rate": 0.0002, "epoch": 1.2558983666061705, "step": 1730}, {"loss": 1.0191, "grad_norm": 0.38048651814460754, "learning_rate": 0.0002, "epoch": 1.263157894736842, "step": 1740}, {"loss": 1.0142, "grad_norm": 0.6676169633865356, "learning_rate": 0.0002, "epoch": 1.2704174228675136, "step": 1750}, {"loss": 0.992, "grad_norm": 0.4075961410999298, "learning_rate": 0.0002, "epoch": 1.2776769509981851, "step": 1760}, {"loss": 1.0301, "grad_norm": 0.4374721348285675, "learning_rate": 0.0002, "epoch": 1.2849364791288567, "step": 1770}, {"loss": 1.0601, "grad_norm": 0.4638824164867401, "learning_rate": 0.0002, "epoch": 1.2921960072595282, "step": 1780}, {"loss": 0.9593, "grad_norm": 0.38631564378738403, "learning_rate": 0.0002, "epoch": 1.2994555353901998, "step": 1790}, {"loss": 1.0042, "grad_norm": 0.35873809456825256, "learning_rate": 0.0002, "epoch": 1.306715063520871, "step": 1800}, {"loss": 0.9835, "grad_norm": 0.33602237701416016, "learning_rate": 0.0002, "epoch": 1.3139745916515426, "step": 1810}, {"loss": 1.1032, "grad_norm": 0.46696463227272034, "learning_rate": 0.0002, "epoch": 1.3212341197822142, "step": 1820}, {"loss": 1.0123, "grad_norm": 0.368958979845047, "learning_rate": 0.0002, "epoch": 1.3284936479128857, "step": 1830}, {"loss": 1.0456, "grad_norm": 0.3160957396030426, "learning_rate": 0.0002, "epoch": 1.335753176043557, "step": 1840}, {"loss": 1.0264, "grad_norm": 0.4511511027812958, "learning_rate": 0.0002, "epoch": 1.3430127041742286, "step": 1850}, {"loss": 1.0179, "grad_norm": 0.6769845485687256, "learning_rate": 0.0002, "epoch": 1.3502722323049001, "step": 1860}, {"loss": 1.011, "grad_norm": 0.3749309480190277, "learning_rate": 0.0002, "epoch": 1.3575317604355717, "step": 1870}, {"loss": 1.0223, "grad_norm": 0.553798496723175, "learning_rate": 0.0002, "epoch": 1.3647912885662432, "step": 1880}, {"loss": 0.9634, "grad_norm": 0.3538985550403595, "learning_rate": 0.0002, "epoch": 1.3720508166969148, "step": 1890}, {"loss": 1.0275, "grad_norm": 0.5501534938812256, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1900}, {"loss": 1.0461, "grad_norm": 0.4432051181793213, "learning_rate": 0.0002, "epoch": 1.3865698729582578, "step": 1910}, {"loss": 0.9082, "grad_norm": 0.41755786538124084, "learning_rate": 0.0002, "epoch": 1.3938294010889292, "step": 1920}, {"loss": 1.0858, "grad_norm": 0.5626114010810852, "learning_rate": 0.0002, "epoch": 1.4010889292196007, "step": 1930}, {"loss": 1.0687, "grad_norm": 0.44038185477256775, "learning_rate": 0.0002, "epoch": 1.4083484573502723, "step": 1940}, {"loss": 0.9454, "grad_norm": 0.3397001624107361, "learning_rate": 0.0002, "epoch": 1.4156079854809438, "step": 1950}, {"loss": 0.943, "grad_norm": 0.4325368106365204, "learning_rate": 0.0002, "epoch": 1.4228675136116151, "step": 1960}, {"loss": 1.0488, "grad_norm": 0.3900907039642334, "learning_rate": 0.0002, "epoch": 1.4301270417422867, "step": 1970}, {"loss": 0.9699, "grad_norm": 0.369612455368042, "learning_rate": 0.0002, "epoch": 1.4373865698729582, "step": 1980}, {"loss": 1.0609, "grad_norm": 0.4389338791370392, "learning_rate": 0.0002, "epoch": 1.4446460980036298, "step": 1990}, {"loss": 1.042, "grad_norm": 1.694450855255127, "learning_rate": 0.0002, "epoch": 1.4519056261343013, "step": 2000}, {"loss": 0.997, "grad_norm": 0.516957700252533, "learning_rate": 0.0002, "epoch": 1.4591651542649728, "step": 2010}, {"loss": 1.0979, "grad_norm": 0.45515501499176025, "learning_rate": 0.0002, "epoch": 1.4664246823956444, "step": 2020}, {"loss": 1.0188, "grad_norm": 0.4153139591217041, "learning_rate": 0.0002, "epoch": 1.4736842105263157, "step": 2030}, {"loss": 1.1521, "grad_norm": 0.44353052973747253, "learning_rate": 0.0002, "epoch": 1.4809437386569873, "step": 2040}, {"loss": 0.9653, "grad_norm": 0.570554256439209, "learning_rate": 0.0002, "epoch": 1.4882032667876588, "step": 2050}, {"loss": 1.0352, "grad_norm": 0.5742740035057068, "learning_rate": 0.0002, "epoch": 1.4954627949183303, "step": 2060}, {"loss": 1.0963, "grad_norm": 0.5890517830848694, "learning_rate": 0.0002, "epoch": 1.5027223230490017, "step": 2070}, {"loss": 1.0639, "grad_norm": 0.4162650406360626, "learning_rate": 0.0002, "epoch": 1.5099818511796732, "step": 2080}, {"loss": 1.071, "grad_norm": 0.7334717512130737, "learning_rate": 0.0002, "epoch": 1.5172413793103448, "step": 2090}, {"loss": 0.8957, "grad_norm": 0.2963249683380127, "learning_rate": 0.0002, "epoch": 1.5245009074410163, "step": 2100}, {"loss": 1.0446, "grad_norm": 0.30676454305648804, "learning_rate": 0.0002, "epoch": 1.5317604355716878, "step": 2110}, {"loss": 0.9967, "grad_norm": 0.35984641313552856, "learning_rate": 0.0002, "epoch": 1.5390199637023594, "step": 2120}, {"loss": 1.0352, "grad_norm": 0.3384549617767334, "learning_rate": 0.0002, "epoch": 1.546279491833031, "step": 2130}, {"loss": 0.9874, "grad_norm": 0.4725518226623535, "learning_rate": 0.0002, "epoch": 1.5535390199637025, "step": 2140}, {"loss": 1.1166, "grad_norm": 0.3252685070037842, "learning_rate": 0.0002, "epoch": 1.560798548094374, "step": 2150}, {"loss": 0.9127, "grad_norm": 0.45043081045150757, "learning_rate": 0.0002, "epoch": 1.5680580762250453, "step": 2160}, {"loss": 1.0767, "grad_norm": 0.374208003282547, "learning_rate": 0.0002, "epoch": 1.5753176043557169, "step": 2170}, {"loss": 1.0255, "grad_norm": 0.5118404030799866, "learning_rate": 0.0002, "epoch": 1.5825771324863884, "step": 2180}, {"loss": 1.0062, "grad_norm": 0.482170969247818, "learning_rate": 0.0002, "epoch": 1.5898366606170597, "step": 2190}, {"loss": 1.0872, "grad_norm": 0.5337533950805664, "learning_rate": 0.0002, "epoch": 1.5970961887477313, "step": 2200}, {"loss": 1.0405, "grad_norm": 0.5195064544677734, "learning_rate": 0.0002, "epoch": 1.6043557168784028, "step": 2210}, {"loss": 1.0454, "grad_norm": 0.30807098746299744, "learning_rate": 0.0002, "epoch": 1.6116152450090744, "step": 2220}, {"loss": 1.0293, "grad_norm": 0.3962925672531128, "learning_rate": 0.0002, "epoch": 1.618874773139746, "step": 2230}, {"loss": 1.0137, "grad_norm": 0.7636962532997131, "learning_rate": 0.0002, "epoch": 1.6261343012704175, "step": 2240}, {"loss": 0.991, "grad_norm": 0.32380592823028564, "learning_rate": 0.0002, "epoch": 1.633393829401089, "step": 2250}, {"loss": 0.9471, "grad_norm": 0.5767741799354553, "learning_rate": 0.0002, "epoch": 1.6406533575317606, "step": 2260}, {"loss": 0.9524, "grad_norm": 0.39964812994003296, "learning_rate": 0.0002, "epoch": 1.647912885662432, "step": 2270}, {"loss": 0.9866, "grad_norm": 0.622629702091217, "learning_rate": 0.0002, "epoch": 1.6551724137931034, "step": 2280}, {"loss": 0.9609, "grad_norm": 0.40202152729034424, "learning_rate": 0.0002, "epoch": 1.662431941923775, "step": 2290}, {"loss": 0.9859, "grad_norm": 0.4467349052429199, "learning_rate": 0.0002, "epoch": 1.6696914700544465, "step": 2300}, {"loss": 1.0312, "grad_norm": 0.5026949048042297, "learning_rate": 0.0002, "epoch": 1.6769509981851178, "step": 2310}, {"loss": 1.0776, "grad_norm": 0.43754541873931885, "learning_rate": 0.0002, "epoch": 1.6842105263157894, "step": 2320}, {"loss": 0.9786, "grad_norm": 0.42869430780410767, "learning_rate": 0.0002, "epoch": 1.691470054446461, "step": 2330}, {"loss": 1.0747, "grad_norm": 0.4192679524421692, "learning_rate": 0.0002, "epoch": 1.6987295825771325, "step": 2340}, {"loss": 1.0054, "grad_norm": 0.3243155777454376, "learning_rate": 0.0002, "epoch": 1.705989110707804, "step": 2350}, {"loss": 0.9165, "grad_norm": 1.0514075756072998, "learning_rate": 0.0002, "epoch": 1.7132486388384756, "step": 2360}, {"loss": 1.1353, "grad_norm": 0.4719122648239136, "learning_rate": 0.0002, "epoch": 1.720508166969147, "step": 2370}, {"loss": 0.9913, "grad_norm": 0.3846144676208496, "learning_rate": 0.0002, "epoch": 1.7277676950998186, "step": 2380}, {"loss": 0.9935, "grad_norm": 0.7266581058502197, "learning_rate": 0.0002, "epoch": 1.73502722323049, "step": 2390}, {"loss": 0.9509, "grad_norm": 0.6183241605758667, "learning_rate": 0.0002, "epoch": 1.7422867513611615, "step": 2400}, {"loss": 0.9713, "grad_norm": 0.3658260405063629, "learning_rate": 0.0002, "epoch": 1.749546279491833, "step": 2410}, {"loss": 1.0033, "grad_norm": 0.6036322712898254, "learning_rate": 0.0002, "epoch": 1.7568058076225044, "step": 2420}, {"loss": 1.0526, "grad_norm": 0.7872936129570007, "learning_rate": 0.0002, "epoch": 1.764065335753176, "step": 2430}, {"loss": 0.99, "grad_norm": 0.35946124792099, "learning_rate": 0.0002, "epoch": 1.7713248638838475, "step": 2440}, {"loss": 0.8845, "grad_norm": 0.3740338981151581, "learning_rate": 0.0002, "epoch": 1.778584392014519, "step": 2450}, {"loss": 0.9333, "grad_norm": 0.6150230169296265, "learning_rate": 0.0002, "epoch": 1.7858439201451906, "step": 2460}, {"loss": 1.0373, "grad_norm": 0.4726075530052185, "learning_rate": 0.0002, "epoch": 1.793103448275862, "step": 2470}, {"loss": 0.974, "grad_norm": 0.31292253732681274, "learning_rate": 0.0002, "epoch": 1.8003629764065336, "step": 2480}, {"loss": 1.0205, "grad_norm": 0.4463104009628296, "learning_rate": 0.0002, "epoch": 1.8076225045372052, "step": 2490}, {"loss": 1.0771, "grad_norm": 0.7848200798034668, "learning_rate": 0.0002, "epoch": 1.8148820326678767, "step": 2500}, {"loss": 0.9508, "grad_norm": 0.5562082529067993, "learning_rate": 0.0002, "epoch": 1.822141560798548, "step": 2510}, {"loss": 0.9343, "grad_norm": 0.39892756938934326, "learning_rate": 0.0002, "epoch": 1.8294010889292196, "step": 2520}, {"loss": 1.0489, "grad_norm": 1.2923320531845093, "learning_rate": 0.0002, "epoch": 1.8366606170598911, "step": 2530}, {"loss": 1.1203, "grad_norm": 0.6316490769386292, "learning_rate": 0.0002, "epoch": 1.8439201451905625, "step": 2540}, {"loss": 0.9131, "grad_norm": 0.46100497245788574, "learning_rate": 0.0002, "epoch": 1.851179673321234, "step": 2550}, {"loss": 1.0977, "grad_norm": 0.7902987003326416, "learning_rate": 0.0002, "epoch": 1.8584392014519056, "step": 2560}, {"loss": 0.9702, "grad_norm": 0.4596365690231323, "learning_rate": 0.0002, "epoch": 1.865698729582577, "step": 2570}, {"loss": 0.9322, "grad_norm": 0.6592172384262085, "learning_rate": 0.0002, "epoch": 1.8729582577132486, "step": 2580}, {"loss": 0.9549, "grad_norm": 0.5764662027359009, "learning_rate": 0.0002, "epoch": 1.8802177858439202, "step": 2590}, {"loss": 0.9485, "grad_norm": 0.8421637415885925, "learning_rate": 0.0002, "epoch": 1.8874773139745917, "step": 2600}, {"loss": 1.012, "grad_norm": 0.5635305047035217, "learning_rate": 0.0002, "epoch": 1.8947368421052633, "step": 2610}, {"loss": 0.8907, "grad_norm": 0.46231237053871155, "learning_rate": 0.0002, "epoch": 1.9019963702359348, "step": 2620}, {"loss": 0.9543, "grad_norm": 0.3944607079029083, "learning_rate": 0.0002, "epoch": 1.9092558983666061, "step": 2630}, {"loss": 0.9964, "grad_norm": 0.4753907322883606, "learning_rate": 0.0002, "epoch": 1.9165154264972777, "step": 2640}, {"loss": 1.0217, "grad_norm": 0.4151090979576111, "learning_rate": 0.0002, "epoch": 1.9237749546279492, "step": 2650}, {"loss": 1.0397, "grad_norm": 0.6793725490570068, "learning_rate": 0.0002, "epoch": 1.9310344827586206, "step": 2660}, {"loss": 1.0401, "grad_norm": 0.339755117893219, "learning_rate": 0.0002, "epoch": 1.938294010889292, "step": 2670}, {"loss": 1.082, "grad_norm": 0.40789374709129333, "learning_rate": 0.0002, "epoch": 1.9455535390199636, "step": 2680}, {"loss": 0.9973, "grad_norm": 0.3750005066394806, "learning_rate": 0.0002, "epoch": 1.9528130671506352, "step": 2690}, {"loss": 0.9837, "grad_norm": 0.39684441685676575, "learning_rate": 0.0002, "epoch": 1.9600725952813067, "step": 2700}, {"loss": 1.1204, "grad_norm": 0.378287672996521, "learning_rate": 0.0002, "epoch": 1.9673321234119783, "step": 2710}, {"loss": 0.9284, "grad_norm": 0.3668482005596161, "learning_rate": 0.0002, "epoch": 1.9745916515426498, "step": 2720}, {"loss": 1.0512, "grad_norm": 0.49997565150260925, "learning_rate": 0.0002, "epoch": 1.9818511796733214, "step": 2730}, {"loss": 0.9311, "grad_norm": 0.36852124333381653, "learning_rate": 0.0002, "epoch": 1.989110707803993, "step": 2740}, {"loss": 1.0898, "grad_norm": 0.5203380584716797, "learning_rate": 0.0002, "epoch": 1.9963702359346642, "step": 2750}, {"eval_loss": 1.1086540222167969, "eval_runtime": 95.6866, "eval_samples_per_second": 4.557, "eval_steps_per_second": 0.575, "epoch": 2.0, "step": 2755}, {"loss": 0.9791, "grad_norm": 0.35921573638916016, "learning_rate": 0.0002, "epoch": 2.0036297640653356, "step": 2760}, {"loss": 0.8968, "grad_norm": 1.013861894607544, "learning_rate": 0.0002, "epoch": 2.010889292196007, "step": 2770}, {"loss": 0.7873, "grad_norm": 0.4425240159034729, "learning_rate": 0.0002, "epoch": 2.0181488203266786, "step": 2780}, {"loss": 0.8758, "grad_norm": 0.60174161195755, "learning_rate": 0.0002, "epoch": 2.02540834845735, "step": 2790}, {"loss": 0.7989, "grad_norm": 0.47582098841667175, "learning_rate": 0.0002, "epoch": 2.0326678765880217, "step": 2800}, {"loss": 0.8898, "grad_norm": 0.6012811660766602, "learning_rate": 0.0002, "epoch": 2.0399274047186933, "step": 2810}, {"loss": 0.8109, "grad_norm": 0.4444408118724823, "learning_rate": 0.0002, "epoch": 2.047186932849365, "step": 2820}, {"loss": 0.8621, "grad_norm": 0.4864003360271454, "learning_rate": 0.0002, "epoch": 2.0544464609800364, "step": 2830}, {"loss": 0.8817, "grad_norm": 0.5104215741157532, "learning_rate": 0.0002, "epoch": 2.061705989110708, "step": 2840}, {"loss": 0.7789, "grad_norm": 0.6218489408493042, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 2850}, {"loss": 0.8778, "grad_norm": 0.705784261226654, "learning_rate": 0.0002, "epoch": 2.076225045372051, "step": 2860}, {"loss": 0.7851, "grad_norm": 0.48091503977775574, "learning_rate": 0.0002, "epoch": 2.0834845735027225, "step": 2870}, {"loss": 0.8444, "grad_norm": 0.5062456727027893, "learning_rate": 0.0002, "epoch": 2.0907441016333936, "step": 2880}, {"loss": 0.7898, "grad_norm": 0.4862022399902344, "learning_rate": 0.0002, "epoch": 2.098003629764065, "step": 2890}, {"loss": 0.8593, "grad_norm": 0.48264044523239136, "learning_rate": 0.0002, "epoch": 2.1052631578947367, "step": 2900}, {"loss": 0.8325, "grad_norm": 0.43744346499443054, "learning_rate": 0.0002, "epoch": 2.1125226860254083, "step": 2910}, {"loss": 0.9099, "grad_norm": 0.5480492115020752, "learning_rate": 0.0002, "epoch": 2.11978221415608, "step": 2920}, {"loss": 0.7727, "grad_norm": 0.5068560838699341, "learning_rate": 0.0002, "epoch": 2.1270417422867514, "step": 2930}, {"loss": 0.8403, "grad_norm": 0.4650852680206299, "learning_rate": 0.0002, "epoch": 2.134301270417423, "step": 2940}, {"loss": 0.8467, "grad_norm": 0.4929981231689453, "learning_rate": 0.0002, "epoch": 2.1415607985480944, "step": 2950}, {"loss": 0.8747, "grad_norm": 0.6537389755249023, "learning_rate": 0.0002, "epoch": 2.148820326678766, "step": 2960}, {"loss": 0.7663, "grad_norm": 0.8032940626144409, "learning_rate": 0.0002, "epoch": 2.1560798548094375, "step": 2970}, {"loss": 0.7604, "grad_norm": 0.7131643891334534, "learning_rate": 0.0002, "epoch": 2.163339382940109, "step": 2980}, {"loss": 0.8424, "grad_norm": 0.6034275889396667, "learning_rate": 0.0002, "epoch": 2.1705989110707806, "step": 2990}, {"loss": 0.8838, "grad_norm": 0.6081095933914185, "learning_rate": 0.0002, "epoch": 2.1778584392014517, "step": 3000}, {"loss": 0.9013, "grad_norm": 0.5706912875175476, "learning_rate": 0.0002, "epoch": 2.1851179673321233, "step": 3010}, {"loss": 0.8473, "grad_norm": 0.6742380261421204, "learning_rate": 0.0002, "epoch": 2.192377495462795, "step": 3020}, {"loss": 0.8372, "grad_norm": 0.4847496449947357, "learning_rate": 0.0002, "epoch": 2.1996370235934664, "step": 3030}, {"loss": 0.8487, "grad_norm": 0.5529342889785767, "learning_rate": 0.0002, "epoch": 2.206896551724138, "step": 3040}, {"loss": 0.9057, "grad_norm": 0.6108783483505249, "learning_rate": 0.0002, "epoch": 2.2141560798548094, "step": 3050}, {"loss": 0.9659, "grad_norm": 0.8841571807861328, "learning_rate": 0.0002, "epoch": 2.221415607985481, "step": 3060}, {"loss": 0.8538, "grad_norm": 0.4227530360221863, "learning_rate": 0.0002, "epoch": 2.2286751361161525, "step": 3070}, {"loss": 0.7925, "grad_norm": 0.461935818195343, "learning_rate": 0.0002, "epoch": 2.235934664246824, "step": 3080}, {"loss": 0.8542, "grad_norm": 0.5407412648200989, "learning_rate": 0.0002, "epoch": 2.2431941923774956, "step": 3090}, {"loss": 0.8392, "grad_norm": 0.5057998895645142, "learning_rate": 0.0002, "epoch": 2.250453720508167, "step": 3100}, {"loss": 0.7898, "grad_norm": 0.530057966709137, "learning_rate": 0.0002, "epoch": 2.2577132486388383, "step": 3110}, {"loss": 0.8507, "grad_norm": 0.5066842436790466, "learning_rate": 0.0002, "epoch": 2.26497277676951, "step": 3120}, {"loss": 0.7903, "grad_norm": 0.5069178342819214, "learning_rate": 0.0002, "epoch": 2.2722323049001814, "step": 3130}, {"loss": 0.8458, "grad_norm": 0.6095499396324158, "learning_rate": 0.0002, "epoch": 2.279491833030853, "step": 3140}, {"loss": 0.8171, "grad_norm": 0.49524766206741333, "learning_rate": 0.0002, "epoch": 2.2867513611615244, "step": 3150}, {"loss": 0.7568, "grad_norm": 0.5334409475326538, "learning_rate": 0.0002, "epoch": 2.294010889292196, "step": 3160}, {"loss": 0.7533, "grad_norm": 1.681748867034912, "learning_rate": 0.0002, "epoch": 2.3012704174228675, "step": 3170}, {"loss": 0.8155, "grad_norm": 0.7225565314292908, "learning_rate": 0.0002, "epoch": 2.308529945553539, "step": 3180}, {"loss": 0.8034, "grad_norm": 0.5379388928413391, "learning_rate": 0.0002, "epoch": 2.3157894736842106, "step": 3190}, {"loss": 0.8833, "grad_norm": 0.45770326256752014, "learning_rate": 0.0002, "epoch": 2.323049001814882, "step": 3200}, {"loss": 0.8419, "grad_norm": 0.6984533071517944, "learning_rate": 0.0002, "epoch": 2.3303085299455537, "step": 3210}, {"loss": 0.7414, "grad_norm": 0.6725744605064392, "learning_rate": 0.0002, "epoch": 2.337568058076225, "step": 3220}, {"loss": 0.8657, "grad_norm": 1.1247471570968628, "learning_rate": 0.0002, "epoch": 2.344827586206897, "step": 3230}, {"loss": 0.7782, "grad_norm": 1.0240263938903809, "learning_rate": 0.0002, "epoch": 2.352087114337568, "step": 3240}, {"loss": 0.826, "grad_norm": 0.5608096122741699, "learning_rate": 0.0002, "epoch": 2.3593466424682394, "step": 3250}, {"loss": 0.8191, "grad_norm": 0.8294990062713623, "learning_rate": 0.0002, "epoch": 2.366606170598911, "step": 3260}, {"loss": 0.9845, "grad_norm": 0.6734224557876587, "learning_rate": 0.0002, "epoch": 2.3738656987295825, "step": 3270}, {"loss": 0.7921, "grad_norm": 0.6862800717353821, "learning_rate": 0.0002, "epoch": 2.381125226860254, "step": 3280}, {"loss": 0.7752, "grad_norm": 0.5442930459976196, "learning_rate": 0.0002, "epoch": 2.3883847549909256, "step": 3290}, {"loss": 0.8543, "grad_norm": 0.5745394229888916, "learning_rate": 0.0002, "epoch": 2.395644283121597, "step": 3300}, {"loss": 0.833, "grad_norm": 0.6257799863815308, "learning_rate": 0.0002, "epoch": 2.4029038112522687, "step": 3310}, {"loss": 0.8191, "grad_norm": 0.5608420968055725, "learning_rate": 0.0002, "epoch": 2.4101633393829403, "step": 3320}, {"loss": 0.8745, "grad_norm": 0.5512017011642456, "learning_rate": 0.0002, "epoch": 2.417422867513612, "step": 3330}, {"loss": 0.8642, "grad_norm": 0.7452999353408813, "learning_rate": 0.0002, "epoch": 2.4246823956442833, "step": 3340}, {"loss": 0.8417, "grad_norm": 0.4604301452636719, "learning_rate": 0.0002, "epoch": 2.4319419237749544, "step": 3350}, {"loss": 0.9148, "grad_norm": 0.8225823640823364, "learning_rate": 0.0002, "epoch": 2.439201451905626, "step": 3360}, {"loss": 0.8144, "grad_norm": 0.8604981303215027, "learning_rate": 0.0002, "epoch": 2.4464609800362975, "step": 3370}, {"loss": 0.9276, "grad_norm": 0.6620925664901733, "learning_rate": 0.0002, "epoch": 2.453720508166969, "step": 3380}, {"loss": 0.8381, "grad_norm": 0.4750158488750458, "learning_rate": 0.0002, "epoch": 2.4609800362976406, "step": 3390}, {"loss": 0.8249, "grad_norm": 0.6061418056488037, "learning_rate": 0.0002, "epoch": 2.468239564428312, "step": 3400}, {"loss": 0.8944, "grad_norm": 0.5977247357368469, "learning_rate": 0.0002, "epoch": 2.4754990925589837, "step": 3410}, {"loss": 0.8366, "grad_norm": 0.5004227757453918, "learning_rate": 0.0002, "epoch": 2.4827586206896552, "step": 3420}, {"loss": 0.9133, "grad_norm": 0.46428972482681274, "learning_rate": 0.0002, "epoch": 2.490018148820327, "step": 3430}, {"loss": 0.8534, "grad_norm": 0.7658395171165466, "learning_rate": 0.0002, "epoch": 2.4972776769509983, "step": 3440}, {"loss": 0.9066, "grad_norm": 0.621819794178009, "learning_rate": 0.0002, "epoch": 2.50453720508167, "step": 3450}, {"loss": 0.8481, "grad_norm": 0.4556088149547577, "learning_rate": 0.0002, "epoch": 2.511796733212341, "step": 3460}, {"loss": 0.8522, "grad_norm": 0.6124140024185181, "learning_rate": 0.0002, "epoch": 2.519056261343013, "step": 3470}, {"loss": 0.8158, "grad_norm": 0.6256231665611267, "learning_rate": 0.0002, "epoch": 2.526315789473684, "step": 3480}, {"loss": 0.9097, "grad_norm": 0.5464141964912415, "learning_rate": 0.0002, "epoch": 2.5335753176043556, "step": 3490}, {"loss": 0.8708, "grad_norm": 0.51471346616745, "learning_rate": 0.0002, "epoch": 2.540834845735027, "step": 3500}, {"loss": 0.9042, "grad_norm": 0.5326165556907654, "learning_rate": 0.0002, "epoch": 2.5480943738656987, "step": 3510}, {"loss": 0.8749, "grad_norm": 0.4750378429889679, "learning_rate": 0.0002, "epoch": 2.5553539019963702, "step": 3520}, {"loss": 0.8248, "grad_norm": 0.5292279124259949, "learning_rate": 0.0002, "epoch": 2.562613430127042, "step": 3530}, {"loss": 0.8808, "grad_norm": 0.6145227551460266, "learning_rate": 0.0002, "epoch": 2.5698729582577133, "step": 3540}, {"loss": 0.846, "grad_norm": 0.8275189399719238, "learning_rate": 0.0002, "epoch": 2.577132486388385, "step": 3550}, {"loss": 0.875, "grad_norm": 0.5037438273429871, "learning_rate": 0.0002, "epoch": 2.5843920145190564, "step": 3560}, {"loss": 0.8324, "grad_norm": 0.5838707685470581, "learning_rate": 0.0002, "epoch": 2.5916515426497275, "step": 3570}, {"loss": 0.8716, "grad_norm": 0.5398710370063782, "learning_rate": 0.0002, "epoch": 2.5989110707803995, "step": 3580}, {"loss": 0.7843, "grad_norm": 0.6115376949310303, "learning_rate": 0.0002, "epoch": 2.6061705989110706, "step": 3590}, {"loss": 0.8142, "grad_norm": 1.118809461593628, "learning_rate": 0.0002, "epoch": 2.613430127041742, "step": 3600}, {"loss": 0.8282, "grad_norm": 0.7811771631240845, "learning_rate": 0.0002, "epoch": 2.6206896551724137, "step": 3610}, {"loss": 0.7997, "grad_norm": 0.5753175020217896, "learning_rate": 0.0002, "epoch": 2.6279491833030852, "step": 3620}, {"loss": 0.7705, "grad_norm": 0.550829291343689, "learning_rate": 0.0002, "epoch": 2.635208711433757, "step": 3630}, {"loss": 0.8564, "grad_norm": 0.5360019207000732, "learning_rate": 0.0002, "epoch": 2.6424682395644283, "step": 3640}, {"loss": 0.7648, "grad_norm": 0.63050377368927, "learning_rate": 0.0002, "epoch": 2.6497277676951, "step": 3650}, {"loss": 0.9156, "grad_norm": 0.5833110213279724, "learning_rate": 0.0002, "epoch": 2.6569872958257714, "step": 3660}, {"loss": 0.8033, "grad_norm": 0.5543047189712524, "learning_rate": 0.0002, "epoch": 2.664246823956443, "step": 3670}, {"loss": 0.8867, "grad_norm": 0.6842212080955505, "learning_rate": 0.0002, "epoch": 2.671506352087114, "step": 3680}, {"loss": 0.8569, "grad_norm": 0.6859333515167236, "learning_rate": 0.0002, "epoch": 2.678765880217786, "step": 3690}, {"loss": 0.8081, "grad_norm": 0.7038410902023315, "learning_rate": 0.0002, "epoch": 2.686025408348457, "step": 3700}, {"loss": 0.8686, "grad_norm": 0.447233647108078, "learning_rate": 0.0002, "epoch": 2.6932849364791287, "step": 3710}, {"loss": 0.8951, "grad_norm": 0.6453872919082642, "learning_rate": 0.0002, "epoch": 2.7005444646098002, "step": 3720}, {"loss": 0.8292, "grad_norm": 0.8025672435760498, "learning_rate": 0.0002, "epoch": 2.707803992740472, "step": 3730}, {"loss": 0.8488, "grad_norm": 0.5997087955474854, "learning_rate": 0.0002, "epoch": 2.7150635208711433, "step": 3740}, {"loss": 0.8308, "grad_norm": 0.6901142001152039, "learning_rate": 0.0002, "epoch": 2.722323049001815, "step": 3750}, {"loss": 0.8517, "grad_norm": 1.036145567893982, "learning_rate": 0.0002, "epoch": 2.7295825771324864, "step": 3760}, {"loss": 0.8455, "grad_norm": 0.7207072377204895, "learning_rate": 0.0002, "epoch": 2.736842105263158, "step": 3770}, {"loss": 0.8009, "grad_norm": 1.0452989339828491, "learning_rate": 0.0002, "epoch": 2.7441016333938295, "step": 3780}, {"loss": 0.8868, "grad_norm": 0.5615278482437134, "learning_rate": 0.0002, "epoch": 2.751361161524501, "step": 3790}, {"loss": 0.9047, "grad_norm": 0.46439215540885925, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 3800}, {"loss": 0.7847, "grad_norm": 0.7134917974472046, "learning_rate": 0.0002, "epoch": 2.7658802177858437, "step": 3810}, {"loss": 0.8315, "grad_norm": 0.5139115452766418, "learning_rate": 0.0002, "epoch": 2.7731397459165157, "step": 3820}, {"loss": 0.8995, "grad_norm": 0.8595781326293945, "learning_rate": 0.0002, "epoch": 2.780399274047187, "step": 3830}, {"loss": 0.9079, "grad_norm": 0.544614851474762, "learning_rate": 0.0002, "epoch": 2.7876588021778583, "step": 3840}, {"loss": 0.8322, "grad_norm": 0.6073850393295288, "learning_rate": 0.0002, "epoch": 2.79491833030853, "step": 3850}, {"loss": 0.8277, "grad_norm": 0.8224069476127625, "learning_rate": 0.0002, "epoch": 2.8021778584392014, "step": 3860}, {"loss": 0.815, "grad_norm": 0.5347970128059387, "learning_rate": 0.0002, "epoch": 2.809437386569873, "step": 3870}, {"loss": 0.8219, "grad_norm": 0.6799601912498474, "learning_rate": 0.0002, "epoch": 2.8166969147005445, "step": 3880}, {"loss": 0.8796, "grad_norm": 0.5219197869300842, "learning_rate": 0.0002, "epoch": 2.823956442831216, "step": 3890}, {"loss": 0.8649, "grad_norm": 0.5710130333900452, "learning_rate": 0.0002, "epoch": 2.8312159709618876, "step": 3900}, {"loss": 0.8741, "grad_norm": 0.5857213139533997, "learning_rate": 0.0002, "epoch": 2.838475499092559, "step": 3910}, {"loss": 0.8154, "grad_norm": 0.5206644535064697, "learning_rate": 0.0002, "epoch": 2.8457350272232302, "step": 3920}, {"loss": 0.9328, "grad_norm": 1.2902015447616577, "learning_rate": 0.0002, "epoch": 2.8529945553539022, "step": 3930}, {"loss": 0.8192, "grad_norm": 0.5252797603607178, "learning_rate": 0.0002, "epoch": 2.8602540834845733, "step": 3940}, {"loss": 0.9047, "grad_norm": 0.5925108790397644, "learning_rate": 0.0002, "epoch": 2.867513611615245, "step": 3950}, {"loss": 0.8611, "grad_norm": 0.9719610810279846, "learning_rate": 0.0002, "epoch": 2.8747731397459164, "step": 3960}, {"loss": 0.8531, "grad_norm": 0.6834747195243835, "learning_rate": 0.0002, "epoch": 2.882032667876588, "step": 3970}, {"loss": 0.8124, "grad_norm": 0.6952353119850159, "learning_rate": 0.0002, "epoch": 2.8892921960072595, "step": 3980}, {"loss": 0.8365, "grad_norm": 0.49889910221099854, "learning_rate": 0.0002, "epoch": 2.896551724137931, "step": 3990}, {"loss": 0.8197, "grad_norm": 0.5007026791572571, "learning_rate": 0.0002, "epoch": 2.9038112522686026, "step": 4000}, {"loss": 0.7752, "grad_norm": 0.5474239587783813, "learning_rate": 0.0002, "epoch": 2.911070780399274, "step": 4010}, {"loss": 0.8579, "grad_norm": 0.6617428660392761, "learning_rate": 0.0002, "epoch": 2.9183303085299457, "step": 4020}, {"loss": 0.8583, "grad_norm": 0.6097776293754578, "learning_rate": 0.0002, "epoch": 2.925589836660617, "step": 4030}, {"loss": 0.8296, "grad_norm": 0.5985828638076782, "learning_rate": 0.0002, "epoch": 2.9328493647912888, "step": 4040}, {"loss": 0.9134, "grad_norm": 0.769488513469696, "learning_rate": 0.0002, "epoch": 2.94010889292196, "step": 4050}, {"loss": 0.8148, "grad_norm": 0.5167055130004883, "learning_rate": 0.0002, "epoch": 2.9473684210526314, "step": 4060}, {"loss": 0.8665, "grad_norm": 0.6265496015548706, "learning_rate": 0.0002, "epoch": 2.954627949183303, "step": 4070}, {"loss": 0.9218, "grad_norm": 1.2644082307815552, "learning_rate": 0.0002, "epoch": 2.9618874773139745, "step": 4080}, {"loss": 0.8026, "grad_norm": 0.6007561087608337, "learning_rate": 0.0002, "epoch": 2.969147005444646, "step": 4090}, {"loss": 0.8262, "grad_norm": 0.47984927892684937, "learning_rate": 0.0002, "epoch": 2.9764065335753176, "step": 4100}, {"loss": 0.7897, "grad_norm": 1.128198504447937, "learning_rate": 0.0002, "epoch": 2.983666061705989, "step": 4110}, {"loss": 0.7883, "grad_norm": 0.526292085647583, "learning_rate": 0.0002, "epoch": 2.9909255898366607, "step": 4120}, {"loss": 0.7801, "grad_norm": 0.5462674498558044, "learning_rate": 0.0002, "epoch": 2.9981851179673322, "step": 4130}, {"eval_loss": 1.1093357801437378, "eval_runtime": 46.2498, "eval_samples_per_second": 9.427, "eval_steps_per_second": 1.189, "epoch": 2.9996370235934666, "step": 4132}, {"loss": 0.721, "grad_norm": 0.48288026452064514, "learning_rate": 0.0002, "epoch": 3.0054446460980038, "step": 4140}, {"loss": 0.6769, "grad_norm": 1.0181782245635986, "learning_rate": 0.0002, "epoch": 3.0127041742286753, "step": 4150}, {"loss": 0.7185, "grad_norm": 0.7718019485473633, "learning_rate": 0.0002, "epoch": 3.019963702359347, "step": 4160}, {"loss": 0.6552, "grad_norm": 0.7492219805717468, "learning_rate": 0.0002, "epoch": 3.027223230490018, "step": 4170}, {"loss": 0.6678, "grad_norm": 0.9363632798194885, "learning_rate": 0.0002, "epoch": 3.0344827586206895, "step": 4180}, {"loss": 0.7187, "grad_norm": 0.6888533234596252, "learning_rate": 0.0002, "epoch": 3.041742286751361, "step": 4190}, {"loss": 0.6469, "grad_norm": 0.7072834968566895, "learning_rate": 0.0002, "epoch": 3.0490018148820326, "step": 4200}, {"loss": 0.6387, "grad_norm": 0.7182047963142395, "learning_rate": 0.0002, "epoch": 3.056261343012704, "step": 4210}, {"loss": 0.6385, "grad_norm": 0.7194355130195618, "learning_rate": 0.0002, "epoch": 3.0635208711433757, "step": 4220}, {"loss": 0.5812, "grad_norm": 0.9454023838043213, "learning_rate": 0.0002, "epoch": 3.0707803992740472, "step": 4230}, {"loss": 0.6036, "grad_norm": 0.838657557964325, "learning_rate": 0.0002, "epoch": 3.0780399274047188, "step": 4240}, {"loss": 0.646, "grad_norm": 0.740113377571106, "learning_rate": 0.0002, "epoch": 3.0852994555353903, "step": 4250}, {"loss": 0.604, "grad_norm": 0.6616561412811279, "learning_rate": 0.0002, "epoch": 3.092558983666062, "step": 4260}, {"loss": 0.6462, "grad_norm": 0.8846506476402283, "learning_rate": 0.0002, "epoch": 3.0998185117967334, "step": 4270}, {"loss": 0.6037, "grad_norm": 0.6322125792503357, "learning_rate": 0.0002, "epoch": 3.107078039927405, "step": 4280}, {"loss": 0.5953, "grad_norm": 0.7461467385292053, "learning_rate": 0.0002, "epoch": 3.114337568058076, "step": 4290}, {"loss": 0.6099, "grad_norm": 0.8251287341117859, "learning_rate": 0.0002, "epoch": 3.1215970961887476, "step": 4300}, {"loss": 0.6284, "grad_norm": 0.8767673373222351, "learning_rate": 0.0002, "epoch": 3.128856624319419, "step": 4310}, {"loss": 0.7535, "grad_norm": 0.7758759260177612, "learning_rate": 0.0002, "epoch": 3.1361161524500907, "step": 4320}, {"loss": 0.6624, "grad_norm": 1.1056879758834839, "learning_rate": 0.0002, "epoch": 3.143375680580762, "step": 4330}, {"loss": 0.691, "grad_norm": 0.8259835243225098, "learning_rate": 0.0002, "epoch": 3.1506352087114338, "step": 4340}, {"loss": 0.6635, "grad_norm": 0.6607027053833008, "learning_rate": 0.0002, "epoch": 3.1578947368421053, "step": 4350}, {"loss": 0.5911, "grad_norm": 0.7983301281929016, "learning_rate": 0.0002, "epoch": 3.165154264972777, "step": 4360}, {"loss": 0.6496, "grad_norm": 0.6725239157676697, "learning_rate": 0.0002, "epoch": 3.1724137931034484, "step": 4370}, {"loss": 0.5966, "grad_norm": 0.9052095413208008, "learning_rate": 0.0002, "epoch": 3.17967332123412, "step": 4380}, {"loss": 0.6877, "grad_norm": 0.8131307363510132, "learning_rate": 0.0002, "epoch": 3.1869328493647915, "step": 4390}, {"loss": 0.6384, "grad_norm": 0.6435626149177551, "learning_rate": 0.0002, "epoch": 3.1941923774954626, "step": 4400}, {"loss": 0.5819, "grad_norm": 0.84367436170578, "learning_rate": 0.0002, "epoch": 3.201451905626134, "step": 4410}, {"loss": 0.6104, "grad_norm": 1.5018867254257202, "learning_rate": 0.0002, "epoch": 3.2087114337568057, "step": 4420}, {"loss": 0.6838, "grad_norm": 0.7019091844558716, "learning_rate": 0.0002, "epoch": 3.215970961887477, "step": 4430}, {"loss": 0.6153, "grad_norm": 0.9164197444915771, "learning_rate": 0.0002, "epoch": 3.2232304900181488, "step": 4440}, {"loss": 0.6618, "grad_norm": 0.7890861630439758, "learning_rate": 0.0002, "epoch": 3.2304900181488203, "step": 4450}, {"loss": 0.6401, "grad_norm": 0.6517660617828369, "learning_rate": 0.0002, "epoch": 3.237749546279492, "step": 4460}, {"loss": 0.6699, "grad_norm": 1.10188889503479, "learning_rate": 0.0002, "epoch": 3.2450090744101634, "step": 4470}, {"loss": 0.6356, "grad_norm": 0.8158330917358398, "learning_rate": 0.0002, "epoch": 3.252268602540835, "step": 4480}, {"loss": 0.7757, "grad_norm": 0.7663109302520752, "learning_rate": 0.0002, "epoch": 3.2595281306715065, "step": 4490}, {"loss": 0.6539, "grad_norm": 0.8473444581031799, "learning_rate": 0.0002, "epoch": 3.266787658802178, "step": 4500}, {"loss": 0.6511, "grad_norm": 0.9724768996238708, "learning_rate": 0.0002, "epoch": 3.274047186932849, "step": 4510}, {"loss": 0.5464, "grad_norm": 0.8516759276390076, "learning_rate": 0.0002, "epoch": 3.281306715063521, "step": 4520}, {"loss": 0.6534, "grad_norm": 0.7543437480926514, "learning_rate": 0.0002, "epoch": 3.288566243194192, "step": 4530}, {"loss": 0.6095, "grad_norm": 1.0472029447555542, "learning_rate": 0.0002, "epoch": 3.2958257713248638, "step": 4540}, {"loss": 0.6216, "grad_norm": 0.6240826845169067, "learning_rate": 0.0002, "epoch": 3.3030852994555353, "step": 4550}, {"loss": 0.6223, "grad_norm": 0.9957774877548218, "learning_rate": 0.0002, "epoch": 3.310344827586207, "step": 4560}, {"loss": 0.618, "grad_norm": 0.6448912620544434, "learning_rate": 0.0002, "epoch": 3.3176043557168784, "step": 4570}, {"loss": 0.6188, "grad_norm": 0.7519692778587341, "learning_rate": 0.0002, "epoch": 3.32486388384755, "step": 4580}, {"loss": 0.6672, "grad_norm": 0.7367453575134277, "learning_rate": 0.0002, "epoch": 3.3321234119782215, "step": 4590}, {"loss": 0.6517, "grad_norm": 0.8064960837364197, "learning_rate": 0.0002, "epoch": 3.339382940108893, "step": 4600}, {"loss": 0.6062, "grad_norm": 0.7664631009101868, "learning_rate": 0.0002, "epoch": 3.3466424682395646, "step": 4610}, {"loss": 0.6834, "grad_norm": 0.7803396582603455, "learning_rate": 0.0002, "epoch": 3.353901996370236, "step": 4620}, {"loss": 0.6961, "grad_norm": 0.9141599535942078, "learning_rate": 0.0002, "epoch": 3.3611615245009077, "step": 4630}, {"loss": 0.6889, "grad_norm": 0.9719856381416321, "learning_rate": 0.0002, "epoch": 3.3684210526315788, "step": 4640}, {"loss": 0.6914, "grad_norm": 0.9223218560218811, "learning_rate": 0.0002, "epoch": 3.3756805807622503, "step": 4650}, {"loss": 0.5981, "grad_norm": 0.7289277911186218, "learning_rate": 0.0002, "epoch": 3.382940108892922, "step": 4660}, {"loss": 0.595, "grad_norm": 1.039724349975586, "learning_rate": 0.0002, "epoch": 3.3901996370235934, "step": 4670}, {"loss": 0.8121, "grad_norm": 1.397438883781433, "learning_rate": 0.0002, "epoch": 3.397459165154265, "step": 4680}, {"loss": 0.6334, "grad_norm": 1.0069999694824219, "learning_rate": 0.0002, "epoch": 3.4047186932849365, "step": 4690}, {"loss": 0.6598, "grad_norm": 0.816291332244873, "learning_rate": 0.0002, "epoch": 3.411978221415608, "step": 4700}, {"loss": 0.6748, "grad_norm": 1.2831530570983887, "learning_rate": 0.0002, "epoch": 3.4192377495462796, "step": 4710}, {"loss": 0.6625, "grad_norm": 0.9573889970779419, "learning_rate": 0.0002, "epoch": 3.426497277676951, "step": 4720}, {"loss": 0.7279, "grad_norm": 0.7685632705688477, "learning_rate": 0.0002, "epoch": 3.4337568058076227, "step": 4730}, {"loss": 0.6104, "grad_norm": 0.7019195556640625, "learning_rate": 0.0002, "epoch": 3.441016333938294, "step": 4740}, {"loss": 0.7606, "grad_norm": 0.7244833707809448, "learning_rate": 0.0002, "epoch": 3.4482758620689653, "step": 4750}, {"loss": 0.6951, "grad_norm": 1.3468551635742188, "learning_rate": 0.0002, "epoch": 3.455535390199637, "step": 4760}, {"loss": 0.6945, "grad_norm": 0.822846531867981, "learning_rate": 0.0002, "epoch": 3.4627949183303084, "step": 4770}, {"loss": 0.6431, "grad_norm": 0.7311608195304871, "learning_rate": 0.0002, "epoch": 3.47005444646098, "step": 4780}, {"loss": 0.7019, "grad_norm": 0.9466770887374878, "learning_rate": 0.0002, "epoch": 3.4773139745916515, "step": 4790}, {"loss": 0.7767, "grad_norm": 1.1527155637741089, "learning_rate": 0.0002, "epoch": 3.484573502722323, "step": 4800}, {"loss": 0.6882, "grad_norm": 1.1288906335830688, "learning_rate": 0.0002, "epoch": 3.4918330308529946, "step": 4810}, {"loss": 0.6564, "grad_norm": 0.9096164107322693, "learning_rate": 0.0002, "epoch": 3.499092558983666, "step": 4820}, {"loss": 0.6127, "grad_norm": 0.7988565564155579, "learning_rate": 0.0002, "epoch": 3.5063520871143377, "step": 4830}, {"loss": 0.7004, "grad_norm": 0.7183415293693542, "learning_rate": 0.0002, "epoch": 3.513611615245009, "step": 4840}, {"loss": 0.74, "grad_norm": 0.6614915132522583, "learning_rate": 0.0002, "epoch": 3.5208711433756807, "step": 4850}, {"loss": 0.7271, "grad_norm": 0.8609521985054016, "learning_rate": 0.0002, "epoch": 3.528130671506352, "step": 4860}, {"loss": 0.6664, "grad_norm": 0.86552894115448, "learning_rate": 0.0002, "epoch": 3.535390199637024, "step": 4870}, {"loss": 0.6432, "grad_norm": 0.6926496028900146, "learning_rate": 0.0002, "epoch": 3.542649727767695, "step": 4880}, {"loss": 0.7117, "grad_norm": 0.8157467246055603, "learning_rate": 0.0002, "epoch": 3.5499092558983665, "step": 4890}, {"loss": 0.6201, "grad_norm": 0.9085357189178467, "learning_rate": 0.0002, "epoch": 3.557168784029038, "step": 4900}, {"loss": 0.6521, "grad_norm": 0.6322644948959351, "learning_rate": 0.0002, "epoch": 3.5644283121597096, "step": 4910}, {"loss": 0.6607, "grad_norm": 1.263205885887146, "learning_rate": 0.0002, "epoch": 3.571687840290381, "step": 4920}, {"loss": 0.6657, "grad_norm": 0.8901070356369019, "learning_rate": 0.0002, "epoch": 3.5789473684210527, "step": 4930}, {"loss": 0.6434, "grad_norm": 0.7983952164649963, "learning_rate": 0.0002, "epoch": 3.586206896551724, "step": 4940}, {"loss": 0.6861, "grad_norm": 0.9887813925743103, "learning_rate": 0.0002, "epoch": 3.5934664246823957, "step": 4950}, {"loss": 0.6502, "grad_norm": 0.7895187735557556, "learning_rate": 0.0002, "epoch": 3.6007259528130673, "step": 4960}, {"loss": 0.7111, "grad_norm": 0.9685819745063782, "learning_rate": 0.0002, "epoch": 3.6079854809437384, "step": 4970}, {"loss": 0.6915, "grad_norm": 0.6576591730117798, "learning_rate": 0.0002, "epoch": 3.6152450090744104, "step": 4980}, {"loss": 0.6195, "grad_norm": 0.856985330581665, "learning_rate": 0.0002, "epoch": 3.6225045372050815, "step": 4990}, {"loss": 0.6318, "grad_norm": 0.7230252623558044, "learning_rate": 0.0002, "epoch": 3.629764065335753, "step": 5000}, {"loss": 0.742, "grad_norm": 0.8260893821716309, "learning_rate": 0.0002, "epoch": 3.6370235934664246, "step": 5010}, {"loss": 0.7223, "grad_norm": 0.7635950446128845, "learning_rate": 0.0002, "epoch": 3.644283121597096, "step": 5020}, {"loss": 0.6837, "grad_norm": 0.7060768604278564, "learning_rate": 0.0002, "epoch": 3.6515426497277677, "step": 5030}, {"loss": 0.6921, "grad_norm": 0.8020303249359131, "learning_rate": 0.0002, "epoch": 3.658802177858439, "step": 5040}, {"loss": 0.6446, "grad_norm": 0.8530341386795044, "learning_rate": 0.0002, "epoch": 3.6660617059891107, "step": 5050}, {"loss": 0.7222, "grad_norm": 0.6667101979255676, "learning_rate": 0.0002, "epoch": 3.6733212341197823, "step": 5060}, {"loss": 0.7081, "grad_norm": 0.7385406494140625, "learning_rate": 0.0002, "epoch": 3.680580762250454, "step": 5070}, {"loss": 0.7035, "grad_norm": 0.7753380537033081, "learning_rate": 0.0002, "epoch": 3.6878402903811254, "step": 5080}, {"loss": 0.6491, "grad_norm": 0.7516207098960876, "learning_rate": 0.0002, "epoch": 3.695099818511797, "step": 5090}, {"loss": 0.672, "grad_norm": 0.8171586394309998, "learning_rate": 0.0002, "epoch": 3.702359346642468, "step": 5100}, {"loss": 0.7459, "grad_norm": 1.0796279907226562, "learning_rate": 0.0002, "epoch": 3.70961887477314, "step": 5110}, {"loss": 0.5948, "grad_norm": 0.6957688927650452, "learning_rate": 0.0002, "epoch": 3.716878402903811, "step": 5120}, {"loss": 0.7515, "grad_norm": 0.8550161719322205, "learning_rate": 0.0002, "epoch": 3.7241379310344827, "step": 5130}, {"loss": 0.7286, "grad_norm": 0.9396728277206421, "learning_rate": 0.0002, "epoch": 3.731397459165154, "step": 5140}, {"loss": 0.7594, "grad_norm": 1.4264805316925049, "learning_rate": 0.0002, "epoch": 3.7386569872958257, "step": 5150}, {"loss": 0.6575, "grad_norm": 0.8725108504295349, "learning_rate": 0.0002, "epoch": 3.7459165154264973, "step": 5160}, {"loss": 0.6778, "grad_norm": 1.0346195697784424, "learning_rate": 0.0002, "epoch": 3.753176043557169, "step": 5170}, {"loss": 0.6371, "grad_norm": 0.5395554304122925, "learning_rate": 0.0002, "epoch": 3.7604355716878404, "step": 5180}, {"loss": 0.7308, "grad_norm": 1.3153616189956665, "learning_rate": 0.0002, "epoch": 3.767695099818512, "step": 5190}, {"loss": 0.78, "grad_norm": 0.9879828691482544, "learning_rate": 0.0002, "epoch": 3.7749546279491835, "step": 5200}, {"loss": 0.7068, "grad_norm": 0.8876672983169556, "learning_rate": 0.0002, "epoch": 3.7822141560798546, "step": 5210}, {"loss": 0.6283, "grad_norm": 0.8363267779350281, "learning_rate": 0.0002, "epoch": 3.7894736842105265, "step": 5220}, {"loss": 0.6255, "grad_norm": 0.637294590473175, "learning_rate": 0.0002, "epoch": 3.7967332123411976, "step": 5230}, {"loss": 0.6685, "grad_norm": 1.1408970355987549, "learning_rate": 0.0002, "epoch": 3.803992740471869, "step": 5240}, {"loss": 0.6761, "grad_norm": 1.0128360986709595, "learning_rate": 0.0002, "epoch": 3.8112522686025407, "step": 5250}, {"loss": 0.6764, "grad_norm": 0.8061144351959229, "learning_rate": 0.0002, "epoch": 3.8185117967332123, "step": 5260}, {"loss": 0.7254, "grad_norm": 0.9626626968383789, "learning_rate": 0.0002, "epoch": 3.825771324863884, "step": 5270}, {"loss": 0.7367, "grad_norm": 0.9013627171516418, "learning_rate": 0.0002, "epoch": 3.8330308529945554, "step": 5280}, {"loss": 0.6806, "grad_norm": 0.8411344289779663, "learning_rate": 0.0002, "epoch": 3.840290381125227, "step": 5290}, {"loss": 0.6818, "grad_norm": 0.7426059246063232, "learning_rate": 0.0002, "epoch": 3.8475499092558985, "step": 5300}, {"loss": 0.6748, "grad_norm": 1.003413438796997, "learning_rate": 0.0002, "epoch": 3.85480943738657, "step": 5310}, {"loss": 0.8554, "grad_norm": 0.7527840733528137, "learning_rate": 0.0002, "epoch": 3.862068965517241, "step": 5320}, {"loss": 0.7521, "grad_norm": 0.738610565662384, "learning_rate": 0.0002, "epoch": 3.869328493647913, "step": 5330}, {"loss": 0.7266, "grad_norm": 0.7277999520301819, "learning_rate": 0.0002, "epoch": 3.876588021778584, "step": 5340}, {"loss": 0.7503, "grad_norm": 0.5951359272003174, "learning_rate": 0.0002, "epoch": 3.8838475499092557, "step": 5350}, {"loss": 0.7447, "grad_norm": 1.043884038925171, "learning_rate": 0.0002, "epoch": 3.8911070780399273, "step": 5360}, {"loss": 0.6862, "grad_norm": 0.8436498045921326, "learning_rate": 0.0002, "epoch": 3.898366606170599, "step": 5370}, {"loss": 0.665, "grad_norm": 0.5603365302085876, "learning_rate": 0.0002, "epoch": 3.9056261343012704, "step": 5380}, {"loss": 0.7098, "grad_norm": 1.0128886699676514, "learning_rate": 0.0002, "epoch": 3.912885662431942, "step": 5390}, {"loss": 0.6707, "grad_norm": 0.7970930337905884, "learning_rate": 0.0002, "epoch": 3.9201451905626135, "step": 5400}, {"loss": 0.637, "grad_norm": 0.7699369192123413, "learning_rate": 0.0002, "epoch": 3.927404718693285, "step": 5410}, {"loss": 0.6742, "grad_norm": 0.800561249256134, "learning_rate": 0.0002, "epoch": 3.9346642468239565, "step": 5420}, {"loss": 0.7208, "grad_norm": 0.8020331859588623, "learning_rate": 0.0002, "epoch": 3.941923774954628, "step": 5430}, {"loss": 0.7294, "grad_norm": 0.7461140155792236, "learning_rate": 0.0002, "epoch": 3.9491833030852996, "step": 5440}, {"loss": 0.7013, "grad_norm": 0.8346918821334839, "learning_rate": 0.0002, "epoch": 3.9564428312159707, "step": 5450}, {"loss": 0.6289, "grad_norm": 0.9723302125930786, "learning_rate": 0.0002, "epoch": 3.9637023593466427, "step": 5460}, {"loss": 0.8029, "grad_norm": 0.6809740662574768, "learning_rate": 0.0002, "epoch": 3.970961887477314, "step": 5470}, {"loss": 0.6896, "grad_norm": 0.7353498339653015, "learning_rate": 0.0002, "epoch": 3.9782214156079854, "step": 5480}, {"loss": 0.6722, "grad_norm": 0.748009443283081, "learning_rate": 0.0002, "epoch": 3.985480943738657, "step": 5490}, {"loss": 0.6866, "grad_norm": 1.3656195402145386, "learning_rate": 0.0002, "epoch": 3.9927404718693285, "step": 5500}, {"loss": 0.7368, "grad_norm": 0.8402108550071716, "learning_rate": 0.0002, "epoch": 4.0, "step": 5510}, {"eval_loss": 1.17229425907135, "eval_runtime": 46.2554, "eval_samples_per_second": 9.426, "eval_steps_per_second": 1.189, "epoch": 4.0, "step": 5510}, {"loss": 0.4637, "grad_norm": 0.8601235747337341, "learning_rate": 0.0002, "epoch": 4.007259528130671, "step": 5520}, {"loss": 0.4717, "grad_norm": 1.2635200023651123, "learning_rate": 0.0002, "epoch": 4.014519056261343, "step": 5530}, {"loss": 0.503, "grad_norm": 1.0257477760314941, "learning_rate": 0.0002, "epoch": 4.021778584392014, "step": 5540}, {"loss": 0.4547, "grad_norm": 0.9436745047569275, "learning_rate": 0.0002, "epoch": 4.029038112522686, "step": 5550}, {"loss": 0.459, "grad_norm": 0.9443606734275818, "learning_rate": 0.0002, "epoch": 4.036297640653357, "step": 5560}, {"loss": 0.5386, "grad_norm": 1.3965742588043213, "learning_rate": 0.0002, "epoch": 4.043557168784029, "step": 5570}, {"loss": 0.4248, "grad_norm": 0.8973520398139954, "learning_rate": 0.0002, "epoch": 4.0508166969147, "step": 5580}, {"loss": 0.4111, "grad_norm": 0.9998409748077393, "learning_rate": 0.0002, "epoch": 4.058076225045372, "step": 5590}, {"loss": 0.4828, "grad_norm": 1.1213387250900269, "learning_rate": 0.0002, "epoch": 4.0653357531760435, "step": 5600}, {"loss": 0.439, "grad_norm": 0.7064558863639832, "learning_rate": 0.0002, "epoch": 4.072595281306715, "step": 5610}, {"loss": 0.4607, "grad_norm": 1.2390803098678589, "learning_rate": 0.0002, "epoch": 4.0798548094373865, "step": 5620}, {"loss": 0.5014, "grad_norm": 1.123469591140747, "learning_rate": 0.0002, "epoch": 4.087114337568058, "step": 5630}, {"loss": 0.513, "grad_norm": 1.229573369026184, "learning_rate": 0.0002, "epoch": 4.09437386569873, "step": 5640}, {"loss": 0.5258, "grad_norm": 1.7182831764221191, "learning_rate": 0.0002, "epoch": 4.101633393829401, "step": 5650}, {"loss": 0.5371, "grad_norm": 0.894903302192688, "learning_rate": 0.0002, "epoch": 4.108892921960073, "step": 5660}, {"loss": 0.4813, "grad_norm": 0.8754552006721497, "learning_rate": 0.0002, "epoch": 4.116152450090744, "step": 5670}, {"loss": 0.491, "grad_norm": 1.2401553392410278, "learning_rate": 0.0002, "epoch": 4.123411978221416, "step": 5680}, {"loss": 0.4549, "grad_norm": 0.8631148934364319, "learning_rate": 0.0002, "epoch": 4.130671506352087, "step": 5690}, {"loss": 0.487, "grad_norm": 1.1798022985458374, "learning_rate": 0.0002, "epoch": 4.137931034482759, "step": 5700}, {"loss": 0.4522, "grad_norm": 0.8344549536705017, "learning_rate": 0.0002, "epoch": 4.14519056261343, "step": 5710}, {"loss": 0.4559, "grad_norm": 1.2342697381973267, "learning_rate": 0.0002, "epoch": 4.152450090744102, "step": 5720}, {"loss": 0.53, "grad_norm": 1.1601094007492065, "learning_rate": 0.0002, "epoch": 4.159709618874773, "step": 5730}, {"loss": 0.4755, "grad_norm": 1.2925703525543213, "learning_rate": 0.0002, "epoch": 4.166969147005445, "step": 5740}, {"loss": 0.4973, "grad_norm": 1.0870997905731201, "learning_rate": 0.0002, "epoch": 4.174228675136116, "step": 5750}, {"loss": 0.5184, "grad_norm": 0.9077792763710022, "learning_rate": 0.0002, "epoch": 4.181488203266787, "step": 5760}, {"loss": 0.4606, "grad_norm": 1.009273886680603, "learning_rate": 0.0002, "epoch": 4.188747731397459, "step": 5770}, {"loss": 0.5383, "grad_norm": 1.2465530633926392, "learning_rate": 0.0002, "epoch": 4.19600725952813, "step": 5780}, {"loss": 0.4938, "grad_norm": 1.2261253595352173, "learning_rate": 0.0002, "epoch": 4.203266787658802, "step": 5790}, {"loss": 0.5255, "grad_norm": 1.1498041152954102, "learning_rate": 0.0002, "epoch": 4.2105263157894735, "step": 5800}, {"loss": 0.5269, "grad_norm": 1.1966725587844849, "learning_rate": 0.0002, "epoch": 4.217785843920145, "step": 5810}, {"loss": 0.5626, "grad_norm": 1.2651296854019165, "learning_rate": 0.0002, "epoch": 4.2250453720508165, "step": 5820}, {"loss": 0.5213, "grad_norm": 1.0388574600219727, "learning_rate": 0.0002, "epoch": 4.2323049001814885, "step": 5830}, {"loss": 0.4965, "grad_norm": 1.3042771816253662, "learning_rate": 0.0002, "epoch": 4.23956442831216, "step": 5840}, {"loss": 0.5116, "grad_norm": 1.1127727031707764, "learning_rate": 0.0002, "epoch": 4.246823956442832, "step": 5850}, {"loss": 0.5197, "grad_norm": 0.9653958082199097, "learning_rate": 0.0002, "epoch": 4.254083484573503, "step": 5860}, {"loss": 0.4414, "grad_norm": 1.0500504970550537, "learning_rate": 0.0002, "epoch": 4.261343012704174, "step": 5870}, {"loss": 0.544, "grad_norm": 1.1476165056228638, "learning_rate": 0.0002, "epoch": 4.268602540834846, "step": 5880}, {"loss": 0.4667, "grad_norm": 0.9424414038658142, "learning_rate": 0.0002, "epoch": 4.275862068965517, "step": 5890}, {"loss": 0.5039, "grad_norm": 1.3309166431427002, "learning_rate": 0.0002, "epoch": 4.283121597096189, "step": 5900}, {"loss": 0.5472, "grad_norm": 1.3025873899459839, "learning_rate": 0.0002, "epoch": 4.29038112522686, "step": 5910}, {"loss": 0.4644, "grad_norm": 1.1442325115203857, "learning_rate": 0.0002, "epoch": 4.297640653357532, "step": 5920}, {"loss": 0.5066, "grad_norm": 0.9820859432220459, "learning_rate": 0.0002, "epoch": 4.304900181488203, "step": 5930}, {"loss": 0.5224, "grad_norm": 0.9615740180015564, "learning_rate": 0.0002, "epoch": 4.312159709618875, "step": 5940}, {"loss": 0.5665, "grad_norm": 1.1627109050750732, "learning_rate": 0.0002, "epoch": 4.319419237749546, "step": 5950}, {"loss": 0.4852, "grad_norm": 0.9381322860717773, "learning_rate": 0.0002, "epoch": 4.326678765880218, "step": 5960}, {"loss": 0.4532, "grad_norm": 0.8154335618019104, "learning_rate": 0.0002, "epoch": 4.333938294010889, "step": 5970}, {"loss": 0.5237, "grad_norm": 0.877671480178833, "learning_rate": 0.0002, "epoch": 4.341197822141561, "step": 5980}, {"loss": 0.6113, "grad_norm": 1.1742031574249268, "learning_rate": 0.0002, "epoch": 4.348457350272232, "step": 5990}, {"loss": 0.5704, "grad_norm": 1.0352917909622192, "learning_rate": 0.0002, "epoch": 4.3557168784029034, "step": 6000}, {"loss": 0.4996, "grad_norm": 0.9963878989219666, "learning_rate": 0.0002, "epoch": 4.362976406533575, "step": 6010}, {"loss": 0.4813, "grad_norm": 1.1892237663269043, "learning_rate": 0.0002, "epoch": 4.3702359346642465, "step": 6020}, {"loss": 0.5224, "grad_norm": 1.2516111135482788, "learning_rate": 0.0002, "epoch": 4.3774954627949185, "step": 6030}, {"loss": 0.5952, "grad_norm": 1.2111951112747192, "learning_rate": 0.0002, "epoch": 4.38475499092559, "step": 6040}, {"loss": 0.4275, "grad_norm": 1.0820083618164062, "learning_rate": 0.0002, "epoch": 4.392014519056262, "step": 6050}, {"loss": 0.5117, "grad_norm": 1.033915638923645, "learning_rate": 0.0002, "epoch": 4.399274047186933, "step": 6060}, {"loss": 0.5431, "grad_norm": 1.0635870695114136, "learning_rate": 0.0002, "epoch": 4.406533575317605, "step": 6070}, {"loss": 0.5341, "grad_norm": 1.0520414113998413, "learning_rate": 0.0002, "epoch": 4.413793103448276, "step": 6080}, {"loss": 0.512, "grad_norm": 1.0821926593780518, "learning_rate": 0.0002, "epoch": 4.421052631578947, "step": 6090}, {"loss": 0.5065, "grad_norm": 1.0533246994018555, "learning_rate": 0.0002, "epoch": 4.428312159709619, "step": 6100}, {"loss": 0.4577, "grad_norm": 0.9231932759284973, "learning_rate": 0.0002, "epoch": 4.43557168784029, "step": 6110}, {"loss": 0.583, "grad_norm": 0.9910260438919067, "learning_rate": 0.0002, "epoch": 4.442831215970962, "step": 6120}, {"loss": 0.4717, "grad_norm": 1.061949372291565, "learning_rate": 0.0002, "epoch": 4.450090744101633, "step": 6130}, {"loss": 0.5893, "grad_norm": 1.2927039861679077, "learning_rate": 0.0002, "epoch": 4.457350272232305, "step": 6140}, {"loss": 0.4684, "grad_norm": 1.3966081142425537, "learning_rate": 0.0002, "epoch": 4.464609800362976, "step": 6150}, {"loss": 0.5507, "grad_norm": 1.3835992813110352, "learning_rate": 0.0002, "epoch": 4.471869328493648, "step": 6160}, {"loss": 0.5911, "grad_norm": 1.0892692804336548, "learning_rate": 0.0002, "epoch": 4.479128856624319, "step": 6170}, {"loss": 0.478, "grad_norm": 1.0318800210952759, "learning_rate": 0.0002, "epoch": 4.486388384754991, "step": 6180}, {"loss": 0.5198, "grad_norm": 0.8174677491188049, "learning_rate": 0.0002, "epoch": 4.493647912885662, "step": 6190}, {"loss": 0.5387, "grad_norm": 1.4157509803771973, "learning_rate": 0.0002, "epoch": 4.500907441016334, "step": 6200}, {"loss": 0.5868, "grad_norm": 1.5244114398956299, "learning_rate": 0.0002, "epoch": 4.508166969147005, "step": 6210}, {"loss": 0.4642, "grad_norm": 0.8164850473403931, "learning_rate": 0.0002, "epoch": 4.5154264972776765, "step": 6220}, {"loss": 0.522, "grad_norm": 1.2904746532440186, "learning_rate": 0.0002, "epoch": 4.5226860254083485, "step": 6230}, {"loss": 0.5103, "grad_norm": 0.7987732887268066, "learning_rate": 0.0002, "epoch": 4.52994555353902, "step": 6240}, {"loss": 0.4615, "grad_norm": 0.831040620803833, "learning_rate": 0.0002, "epoch": 4.537205081669692, "step": 6250}, {"loss": 0.5065, "grad_norm": 0.9545485973358154, "learning_rate": 0.0002, "epoch": 4.544464609800363, "step": 6260}, {"loss": 0.5515, "grad_norm": 0.9291793704032898, "learning_rate": 0.0002, "epoch": 4.551724137931035, "step": 6270}, {"loss": 0.4535, "grad_norm": 0.8977208733558655, "learning_rate": 0.0002, "epoch": 4.558983666061706, "step": 6280}, {"loss": 0.544, "grad_norm": 1.1768537759780884, "learning_rate": 0.0002, "epoch": 4.566243194192378, "step": 6290}, {"loss": 0.5925, "grad_norm": 1.0688952207565308, "learning_rate": 0.0002, "epoch": 4.573502722323049, "step": 6300}, {"loss": 0.5207, "grad_norm": 0.8800966739654541, "learning_rate": 0.0002, "epoch": 4.580762250453721, "step": 6310}, {"loss": 0.6106, "grad_norm": 1.0911834239959717, "learning_rate": 0.0002, "epoch": 4.588021778584392, "step": 6320}, {"loss": 0.5109, "grad_norm": 1.1420872211456299, "learning_rate": 0.0002, "epoch": 4.595281306715064, "step": 6330}, {"loss": 0.5147, "grad_norm": 1.0215224027633667, "learning_rate": 0.0002, "epoch": 4.602540834845735, "step": 6340}, {"loss": 0.592, "grad_norm": 0.9685489535331726, "learning_rate": 0.0002, "epoch": 4.609800362976406, "step": 6350}, {"loss": 0.5775, "grad_norm": 1.12773597240448, "learning_rate": 0.0002, "epoch": 4.617059891107078, "step": 6360}, {"loss": 0.5966, "grad_norm": 1.0663973093032837, "learning_rate": 0.0002, "epoch": 4.624319419237749, "step": 6370}, {"loss": 0.512, "grad_norm": 1.1707262992858887, "learning_rate": 0.0002, "epoch": 4.631578947368421, "step": 6380}, {"loss": 0.5497, "grad_norm": 1.0672980546951294, "learning_rate": 0.0002, "epoch": 4.638838475499092, "step": 6390}, {"loss": 0.5699, "grad_norm": 1.1464333534240723, "learning_rate": 0.0002, "epoch": 4.646098003629764, "step": 6400}, {"loss": 0.5514, "grad_norm": 1.070230484008789, "learning_rate": 0.0002, "epoch": 4.653357531760435, "step": 6410}, {"loss": 0.5013, "grad_norm": 0.9673764109611511, "learning_rate": 0.0002, "epoch": 4.660617059891107, "step": 6420}, {"loss": 0.5901, "grad_norm": 1.0189043283462524, "learning_rate": 0.0002, "epoch": 4.6678765880217785, "step": 6430}, {"loss": 0.5193, "grad_norm": 1.185896396636963, "learning_rate": 0.0002, "epoch": 4.67513611615245, "step": 6440}, {"loss": 0.5318, "grad_norm": 1.0682812929153442, "learning_rate": 0.0002, "epoch": 4.682395644283122, "step": 6450}, {"loss": 0.5773, "grad_norm": 1.3586071729660034, "learning_rate": 0.0002, "epoch": 4.689655172413794, "step": 6460}, {"loss": 0.5482, "grad_norm": 0.6561792492866516, "learning_rate": 0.0002, "epoch": 4.696914700544465, "step": 6470}, {"loss": 0.5711, "grad_norm": 1.1394113302230835, "learning_rate": 0.0002, "epoch": 4.704174228675136, "step": 6480}, {"loss": 0.5325, "grad_norm": 0.9683151245117188, "learning_rate": 0.0002, "epoch": 4.711433756805808, "step": 6490}, {"loss": 0.5139, "grad_norm": 1.0247553586959839, "learning_rate": 0.0002, "epoch": 4.718693284936479, "step": 6500}, {"loss": 0.5794, "grad_norm": 0.8046169281005859, "learning_rate": 0.0002, "epoch": 4.725952813067151, "step": 6510}, {"loss": 0.5471, "grad_norm": 1.0710240602493286, "learning_rate": 0.0002, "epoch": 4.733212341197822, "step": 6520}, {"loss": 0.4805, "grad_norm": 0.9438924193382263, "learning_rate": 0.0002, "epoch": 4.740471869328494, "step": 6530}, {"loss": 0.5404, "grad_norm": 0.869162380695343, "learning_rate": 0.0002, "epoch": 4.747731397459165, "step": 6540}, {"loss": 0.6379, "grad_norm": 0.9776787161827087, "learning_rate": 0.0002, "epoch": 4.754990925589837, "step": 6550}, {"loss": 0.5288, "grad_norm": 1.1990505456924438, "learning_rate": 0.0002, "epoch": 4.762250453720508, "step": 6560}, {"loss": 0.5539, "grad_norm": 1.0582209825515747, "learning_rate": 0.0002, "epoch": 4.769509981851179, "step": 6570}, {"loss": 0.489, "grad_norm": 0.9966367483139038, "learning_rate": 0.0002, "epoch": 4.776769509981851, "step": 6580}, {"loss": 0.5514, "grad_norm": 0.9130612015724182, "learning_rate": 0.0002, "epoch": 4.784029038112522, "step": 6590}, {"loss": 0.5864, "grad_norm": 1.0950500965118408, "learning_rate": 0.0002, "epoch": 4.791288566243194, "step": 6600}, {"loss": 0.5266, "grad_norm": 1.108681321144104, "learning_rate": 0.0002, "epoch": 4.798548094373865, "step": 6610}, {"loss": 0.5875, "grad_norm": 1.1873763799667358, "learning_rate": 0.0002, "epoch": 4.805807622504537, "step": 6620}, {"loss": 0.5736, "grad_norm": 1.305367112159729, "learning_rate": 0.0002, "epoch": 4.8130671506352085, "step": 6630}, {"loss": 0.5636, "grad_norm": 1.2801482677459717, "learning_rate": 0.0002, "epoch": 4.8203266787658805, "step": 6640}, {"loss": 0.582, "grad_norm": 1.26764976978302, "learning_rate": 0.0002, "epoch": 4.827586206896552, "step": 6650}, {"loss": 0.5259, "grad_norm": 1.0018208026885986, "learning_rate": 0.0002, "epoch": 4.834845735027224, "step": 6660}, {"loss": 0.548, "grad_norm": 1.2326326370239258, "learning_rate": 0.0002, "epoch": 4.842105263157895, "step": 6670}, {"loss": 0.5933, "grad_norm": 0.9707282781600952, "learning_rate": 0.0002, "epoch": 4.849364791288567, "step": 6680}, {"loss": 0.5612, "grad_norm": 1.2772048711776733, "learning_rate": 0.0002, "epoch": 4.856624319419238, "step": 6690}, {"loss": 0.5346, "grad_norm": 2.6652262210845947, "learning_rate": 0.0002, "epoch": 4.863883847549909, "step": 6700}, {"loss": 0.5428, "grad_norm": 1.215828537940979, "learning_rate": 0.0002, "epoch": 4.871143375680581, "step": 6710}, {"loss": 0.6571, "grad_norm": 1.3704510927200317, "learning_rate": 0.0002, "epoch": 4.878402903811252, "step": 6720}, {"loss": 0.4963, "grad_norm": 0.7781757116317749, "learning_rate": 0.0002, "epoch": 4.885662431941924, "step": 6730}, {"loss": 0.5989, "grad_norm": 1.1883646249771118, "learning_rate": 0.0002, "epoch": 4.892921960072595, "step": 6740}, {"loss": 0.6067, "grad_norm": 0.9216066002845764, "learning_rate": 0.0002, "epoch": 4.900181488203267, "step": 6750}, {"loss": 0.5085, "grad_norm": 1.0558464527130127, "learning_rate": 0.0002, "epoch": 4.907441016333938, "step": 6760}, {"loss": 0.5216, "grad_norm": 1.032656192779541, "learning_rate": 0.0002, "epoch": 4.91470054446461, "step": 6770}, {"loss": 0.5426, "grad_norm": 1.1261441707611084, "learning_rate": 0.0002, "epoch": 4.921960072595281, "step": 6780}, {"loss": 0.5295, "grad_norm": 1.2178640365600586, "learning_rate": 0.0002, "epoch": 4.929219600725952, "step": 6790}, {"loss": 0.5476, "grad_norm": 1.5369361639022827, "learning_rate": 0.0002, "epoch": 4.936479128856624, "step": 6800}, {"loss": 0.5358, "grad_norm": 1.1188377141952515, "learning_rate": 0.0002, "epoch": 4.943738656987296, "step": 6810}, {"loss": 0.5483, "grad_norm": 1.2506113052368164, "learning_rate": 0.0002, "epoch": 4.950998185117967, "step": 6820}, {"loss": 0.567, "grad_norm": 0.8776047825813293, "learning_rate": 0.0002, "epoch": 4.9582577132486385, "step": 6830}, {"loss": 0.5764, "grad_norm": 0.9700555205345154, "learning_rate": 0.0002, "epoch": 4.9655172413793105, "step": 6840}, {"loss": 0.5396, "grad_norm": 1.2713534832000732, "learning_rate": 0.0002, "epoch": 4.972776769509982, "step": 6850}, {"loss": 0.5451, "grad_norm": 0.9855955243110657, "learning_rate": 0.0002, "epoch": 4.980036297640654, "step": 6860}, {"loss": 0.5884, "grad_norm": 0.8734853863716125, "learning_rate": 0.0002, "epoch": 4.987295825771325, "step": 6870}, {"loss": 0.5189, "grad_norm": 0.8065403699874878, "learning_rate": 0.0002, "epoch": 4.994555353901997, "step": 6880}, {"eval_loss": 1.3302682638168335, "eval_runtime": 46.2496, "eval_samples_per_second": 9.427, "eval_steps_per_second": 1.189, "epoch": 4.999637023593467, "step": 6887}, {"loss": 0.4889, "grad_norm": 0.5163813829421997, "learning_rate": 0.0002, "epoch": 5.001814882032668, "step": 6890}, {"loss": 0.3545, "grad_norm": 1.1496137380599976, "learning_rate": 0.0002, "epoch": 5.00907441016334, "step": 6900}, {"loss": 0.39, "grad_norm": 1.0133885145187378, "learning_rate": 0.0002, "epoch": 5.016333938294011, "step": 6910}, {"loss": 0.3693, "grad_norm": 0.9479621052742004, "learning_rate": 0.0002, "epoch": 5.023593466424682, "step": 6920}, {"loss": 0.4012, "grad_norm": 0.8587583303451538, "learning_rate": 0.0002, "epoch": 5.030852994555354, "step": 6930}, {"loss": 0.3428, "grad_norm": 1.3314697742462158, "learning_rate": 0.0002, "epoch": 5.038112522686025, "step": 6940}, {"loss": 0.3909, "grad_norm": 1.195448875427246, "learning_rate": 0.0002, "epoch": 5.045372050816697, "step": 6950}, {"loss": 0.3322, "grad_norm": 1.2482256889343262, "learning_rate": 0.0002, "epoch": 5.052631578947368, "step": 6960}, {"loss": 0.3893, "grad_norm": 1.2011528015136719, "learning_rate": 0.0002, "epoch": 5.05989110707804, "step": 6970}, {"loss": 0.3265, "grad_norm": 1.3997188806533813, "learning_rate": 0.0002, "epoch": 5.067150635208711, "step": 6980}, {"loss": 0.3716, "grad_norm": 1.2147513628005981, "learning_rate": 0.0002, "epoch": 5.074410163339383, "step": 6990}, {"loss": 0.4053, "grad_norm": 1.6030137538909912, "learning_rate": 0.0002, "epoch": 5.081669691470054, "step": 7000}, {"loss": 0.3665, "grad_norm": 0.9466970562934875, "learning_rate": 0.0002, "epoch": 5.088929219600726, "step": 7010}, {"loss": 0.3451, "grad_norm": 1.4593411684036255, "learning_rate": 0.0002, "epoch": 5.096188747731397, "step": 7020}, {"loss": 0.3843, "grad_norm": 1.2196033000946045, "learning_rate": 0.0002, "epoch": 5.103448275862069, "step": 7030}, {"loss": 0.3896, "grad_norm": 1.1341328620910645, "learning_rate": 0.0002, "epoch": 5.1107078039927405, "step": 7040}, {"loss": 0.3627, "grad_norm": 1.2248976230621338, "learning_rate": 0.0002, "epoch": 5.117967332123412, "step": 7050}, {"loss": 0.3784, "grad_norm": 1.1620593070983887, "learning_rate": 0.0002, "epoch": 5.125226860254084, "step": 7060}, {"loss": 0.3678, "grad_norm": 0.9300723671913147, "learning_rate": 0.0002, "epoch": 5.132486388384755, "step": 7070}, {"loss": 0.3756, "grad_norm": 1.2265169620513916, "learning_rate": 0.0002, "epoch": 5.139745916515427, "step": 7080}, {"loss": 0.3595, "grad_norm": 1.4430373907089233, "learning_rate": 0.0002, "epoch": 5.147005444646098, "step": 7090}, {"loss": 0.3788, "grad_norm": 1.0821576118469238, "learning_rate": 0.0002, "epoch": 5.15426497277677, "step": 7100}, {"loss": 0.383, "grad_norm": 1.2574739456176758, "learning_rate": 0.0002, "epoch": 5.161524500907441, "step": 7110}, {"loss": 0.3692, "grad_norm": 1.1806069612503052, "learning_rate": 0.0002, "epoch": 5.168784029038113, "step": 7120}, {"loss": 0.3978, "grad_norm": 0.9900956153869629, "learning_rate": 0.0002, "epoch": 5.176043557168784, "step": 7130}, {"loss": 0.4358, "grad_norm": 1.2414425611495972, "learning_rate": 0.0002, "epoch": 5.183303085299456, "step": 7140}, {"loss": 0.3485, "grad_norm": 0.8220699429512024, "learning_rate": 0.0002, "epoch": 5.190562613430127, "step": 7150}, {"loss": 0.3517, "grad_norm": 1.29408860206604, "learning_rate": 0.0002, "epoch": 5.197822141560798, "step": 7160}, {"loss": 0.3405, "grad_norm": 0.8510639071464539, "learning_rate": 0.0002, "epoch": 5.20508166969147, "step": 7170}, {"loss": 0.4233, "grad_norm": 1.3448902368545532, "learning_rate": 0.0002, "epoch": 5.212341197822141, "step": 7180}, {"loss": 0.3808, "grad_norm": 1.054451584815979, "learning_rate": 0.0002, "epoch": 5.219600725952813, "step": 7190}, {"loss": 0.368, "grad_norm": 1.3752713203430176, "learning_rate": 0.0002, "epoch": 5.226860254083484, "step": 7200}, {"loss": 0.3844, "grad_norm": 1.4848095178604126, "learning_rate": 0.0002, "epoch": 5.234119782214156, "step": 7210}, {"loss": 0.4187, "grad_norm": 1.428842544555664, "learning_rate": 0.0002, "epoch": 5.241379310344827, "step": 7220}, {"loss": 0.3778, "grad_norm": 1.1703591346740723, "learning_rate": 0.0002, "epoch": 5.248638838475499, "step": 7230}, {"loss": 0.417, "grad_norm": 1.2180451154708862, "learning_rate": 0.0002, "epoch": 5.2558983666061705, "step": 7240}, {"loss": 0.3656, "grad_norm": 1.094045877456665, "learning_rate": 0.0002, "epoch": 5.2631578947368425, "step": 7250}, {"loss": 0.4331, "grad_norm": 0.9545766115188599, "learning_rate": 0.0002, "epoch": 5.270417422867514, "step": 7260}, {"loss": 0.3642, "grad_norm": 0.8356652855873108, "learning_rate": 0.0002, "epoch": 5.277676950998185, "step": 7270}, {"loss": 0.3576, "grad_norm": 1.148160457611084, "learning_rate": 0.0002, "epoch": 5.284936479128857, "step": 7280}, {"loss": 0.4178, "grad_norm": 1.2009977102279663, "learning_rate": 0.0002, "epoch": 5.292196007259528, "step": 7290}, {"loss": 0.3977, "grad_norm": 1.3283873796463013, "learning_rate": 0.0002, "epoch": 5.2994555353902, "step": 7300}, {"loss": 0.3853, "grad_norm": 0.9850481748580933, "learning_rate": 0.0002, "epoch": 5.306715063520871, "step": 7310}, {"loss": 0.3645, "grad_norm": 1.367550015449524, "learning_rate": 0.0002, "epoch": 5.313974591651543, "step": 7320}, {"loss": 0.3898, "grad_norm": 0.8602936863899231, "learning_rate": 0.0002, "epoch": 5.321234119782214, "step": 7330}, {"loss": 0.4173, "grad_norm": 1.1130679845809937, "learning_rate": 0.0002, "epoch": 5.328493647912886, "step": 7340}, {"loss": 0.3642, "grad_norm": 1.3002253770828247, "learning_rate": 0.0002, "epoch": 5.335753176043557, "step": 7350}, {"loss": 0.4138, "grad_norm": 1.6235289573669434, "learning_rate": 0.0002, "epoch": 5.343012704174229, "step": 7360}, {"loss": 0.4779, "grad_norm": 1.156379222869873, "learning_rate": 0.0002, "epoch": 5.3502722323049, "step": 7370}, {"loss": 0.3222, "grad_norm": 1.0569308996200562, "learning_rate": 0.0002, "epoch": 5.357531760435572, "step": 7380}, {"loss": 0.3573, "grad_norm": 1.6674021482467651, "learning_rate": 0.0002, "epoch": 5.364791288566243, "step": 7390}, {"loss": 0.4325, "grad_norm": 1.2962018251419067, "learning_rate": 0.0002, "epoch": 5.372050816696914, "step": 7400}, {"loss": 0.3809, "grad_norm": 1.1904195547103882, "learning_rate": 0.0002, "epoch": 5.379310344827586, "step": 7410}, {"loss": 0.3728, "grad_norm": 1.316245675086975, "learning_rate": 0.0002, "epoch": 5.386569872958257, "step": 7420}, {"loss": 0.4096, "grad_norm": 1.127570390701294, "learning_rate": 0.0002, "epoch": 5.393829401088929, "step": 7430}, {"loss": 0.3933, "grad_norm": 1.3895777463912964, "learning_rate": 0.0002, "epoch": 5.4010889292196005, "step": 7440}, {"loss": 0.4085, "grad_norm": 1.626830816268921, "learning_rate": 0.0002, "epoch": 5.4083484573502725, "step": 7450}, {"loss": 0.4186, "grad_norm": 1.3703926801681519, "learning_rate": 0.0002, "epoch": 5.415607985480944, "step": 7460}, {"loss": 0.3517, "grad_norm": 1.3854840993881226, "learning_rate": 0.0002, "epoch": 5.422867513611616, "step": 7470}, {"loss": 0.3714, "grad_norm": 1.107065200805664, "learning_rate": 0.0002, "epoch": 5.430127041742287, "step": 7480}, {"loss": 0.3855, "grad_norm": 0.7843456268310547, "learning_rate": 0.0002, "epoch": 5.437386569872959, "step": 7490}, {"loss": 0.4159, "grad_norm": 1.6692372560501099, "learning_rate": 0.0002, "epoch": 5.44464609800363, "step": 7500}, {"loss": 0.4185, "grad_norm": 1.2583858966827393, "learning_rate": 0.0002, "epoch": 5.451905626134302, "step": 7510}, {"loss": 0.4401, "grad_norm": 1.6827000379562378, "learning_rate": 0.0002, "epoch": 5.459165154264973, "step": 7520}, {"loss": 0.397, "grad_norm": 1.6680560111999512, "learning_rate": 0.0002, "epoch": 5.466424682395644, "step": 7530}, {"loss": 0.4193, "grad_norm": 1.3696072101593018, "learning_rate": 0.0002, "epoch": 5.473684210526316, "step": 7540}, {"loss": 0.4244, "grad_norm": 1.4523496627807617, "learning_rate": 0.0002, "epoch": 5.480943738656987, "step": 7550}, {"loss": 0.3609, "grad_norm": 1.3432692289352417, "learning_rate": 0.0002, "epoch": 5.488203266787659, "step": 7560}, {"loss": 0.3675, "grad_norm": 1.363818645477295, "learning_rate": 0.0002, "epoch": 5.49546279491833, "step": 7570}, {"loss": 0.3726, "grad_norm": 1.0176721811294556, "learning_rate": 0.0002, "epoch": 5.502722323049002, "step": 7580}, {"loss": 0.3751, "grad_norm": 1.1625547409057617, "learning_rate": 0.0002, "epoch": 5.509981851179673, "step": 7590}, {"loss": 0.433, "grad_norm": 1.2480388879776, "learning_rate": 0.0002, "epoch": 5.517241379310345, "step": 7600}, {"loss": 0.4511, "grad_norm": 1.341509222984314, "learning_rate": 0.0002, "epoch": 5.524500907441016, "step": 7610}, {"loss": 0.4642, "grad_norm": 1.7048436403274536, "learning_rate": 0.0002, "epoch": 5.531760435571687, "step": 7620}, {"loss": 0.4509, "grad_norm": 1.1435480117797852, "learning_rate": 0.0002, "epoch": 5.539019963702359, "step": 7630}, {"loss": 0.4528, "grad_norm": 1.2381842136383057, "learning_rate": 0.0002, "epoch": 5.5462794918330305, "step": 7640}, {"loss": 0.4496, "grad_norm": 1.50786292552948, "learning_rate": 0.0002, "epoch": 5.5535390199637025, "step": 7650}, {"loss": 0.4242, "grad_norm": 1.2263519763946533, "learning_rate": 0.0002, "epoch": 5.560798548094374, "step": 7660}, {"loss": 0.418, "grad_norm": 1.2864696979522705, "learning_rate": 0.0002, "epoch": 5.568058076225046, "step": 7670}, {"loss": 0.3832, "grad_norm": 1.4443191289901733, "learning_rate": 0.0002, "epoch": 5.575317604355717, "step": 7680}, {"loss": 0.3964, "grad_norm": 1.3360971212387085, "learning_rate": 0.0002, "epoch": 5.582577132486389, "step": 7690}, {"loss": 0.4639, "grad_norm": 1.391828179359436, "learning_rate": 0.0002, "epoch": 5.58983666061706, "step": 7700}, {"loss": 0.4722, "grad_norm": 1.3699384927749634, "learning_rate": 0.0002, "epoch": 5.597096188747732, "step": 7710}, {"loss": 0.4302, "grad_norm": 1.3778468370437622, "learning_rate": 0.0002, "epoch": 5.604355716878403, "step": 7720}, {"loss": 0.4179, "grad_norm": 1.1009501218795776, "learning_rate": 0.0002, "epoch": 5.611615245009075, "step": 7730}, {"loss": 0.4104, "grad_norm": 1.0410021543502808, "learning_rate": 0.0002, "epoch": 5.618874773139746, "step": 7740}, {"loss": 0.4489, "grad_norm": 1.1012226343154907, "learning_rate": 0.0002, "epoch": 5.626134301270417, "step": 7750}, {"loss": 0.4544, "grad_norm": 1.3246384859085083, "learning_rate": 0.0002, "epoch": 5.633393829401089, "step": 7760}, {"loss": 0.4381, "grad_norm": 1.4301716089248657, "learning_rate": 0.0002, "epoch": 5.64065335753176, "step": 7770}, {"loss": 0.4297, "grad_norm": 1.1368978023529053, "learning_rate": 0.0002, "epoch": 5.647912885662432, "step": 7780}, {"loss": 0.4063, "grad_norm": 1.3493064641952515, "learning_rate": 0.0002, "epoch": 5.655172413793103, "step": 7790}, {"loss": 0.4562, "grad_norm": 1.3328721523284912, "learning_rate": 0.0002, "epoch": 5.662431941923775, "step": 7800}, {"loss": 0.4075, "grad_norm": 1.3235671520233154, "learning_rate": 0.0002, "epoch": 5.669691470054446, "step": 7810}, {"loss": 0.4589, "grad_norm": 1.1961841583251953, "learning_rate": 0.0002, "epoch": 5.676950998185118, "step": 7820}, {"loss": 0.4503, "grad_norm": 1.4189636707305908, "learning_rate": 0.0002, "epoch": 5.684210526315789, "step": 7830}, {"loss": 0.4452, "grad_norm": 1.3551312685012817, "learning_rate": 0.0002, "epoch": 5.691470054446461, "step": 7840}, {"loss": 0.4268, "grad_norm": 1.449987769126892, "learning_rate": 0.0002, "epoch": 5.6987295825771325, "step": 7850}, {"loss": 0.4141, "grad_norm": 1.1225156784057617, "learning_rate": 0.0002, "epoch": 5.7059891107078045, "step": 7860}, {"loss": 0.41, "grad_norm": 1.4734594821929932, "learning_rate": 0.0002, "epoch": 5.713248638838476, "step": 7870}, {"loss": 0.4013, "grad_norm": 1.3793359994888306, "learning_rate": 0.0002, "epoch": 5.720508166969147, "step": 7880}, {"loss": 0.4065, "grad_norm": 1.2431834936141968, "learning_rate": 0.0002, "epoch": 5.727767695099819, "step": 7890}, {"loss": 0.4595, "grad_norm": 1.1158313751220703, "learning_rate": 0.0002, "epoch": 5.73502722323049, "step": 7900}, {"loss": 0.4342, "grad_norm": 1.212248682975769, "learning_rate": 0.0002, "epoch": 5.742286751361162, "step": 7910}, {"loss": 0.4611, "grad_norm": 1.5259995460510254, "learning_rate": 0.0002, "epoch": 5.749546279491833, "step": 7920}, {"loss": 0.4483, "grad_norm": 1.3909121751785278, "learning_rate": 0.0002, "epoch": 5.756805807622505, "step": 7930}, {"loss": 0.4325, "grad_norm": 1.2511249780654907, "learning_rate": 0.0002, "epoch": 5.764065335753176, "step": 7940}, {"loss": 0.4048, "grad_norm": 1.2511906623840332, "learning_rate": 0.0002, "epoch": 5.771324863883848, "step": 7950}, {"loss": 0.3715, "grad_norm": 1.1489921808242798, "learning_rate": 0.0002, "epoch": 5.778584392014519, "step": 7960}, {"loss": 0.4196, "grad_norm": 1.028943419456482, "learning_rate": 0.0002, "epoch": 5.78584392014519, "step": 7970}, {"loss": 0.4334, "grad_norm": 1.0820423364639282, "learning_rate": 0.0002, "epoch": 5.793103448275862, "step": 7980}, {"loss": 0.3917, "grad_norm": 1.296520471572876, "learning_rate": 0.0002, "epoch": 5.800362976406533, "step": 7990}, {"loss": 0.4509, "grad_norm": 1.3597749471664429, "learning_rate": 0.0002, "epoch": 5.807622504537205, "step": 8000}, {"loss": 0.4535, "grad_norm": 0.8741790652275085, "learning_rate": 0.0002, "epoch": 5.814882032667876, "step": 8010}, {"loss": 0.4239, "grad_norm": 1.1471822261810303, "learning_rate": 0.0002, "epoch": 5.822141560798548, "step": 8020}, {"loss": 0.5042, "grad_norm": 1.2997334003448486, "learning_rate": 0.0002, "epoch": 5.829401088929219, "step": 8030}, {"loss": 0.4758, "grad_norm": 1.1027175188064575, "learning_rate": 0.0002, "epoch": 5.836660617059891, "step": 8040}, {"loss": 0.4192, "grad_norm": 1.2695307731628418, "learning_rate": 0.0002, "epoch": 5.8439201451905625, "step": 8050}, {"loss": 0.5173, "grad_norm": 1.5275461673736572, "learning_rate": 0.0002, "epoch": 5.8511796733212345, "step": 8060}, {"loss": 0.5012, "grad_norm": 1.3059501647949219, "learning_rate": 0.0002, "epoch": 5.8584392014519056, "step": 8070}, {"loss": 0.4425, "grad_norm": 1.57442045211792, "learning_rate": 0.0002, "epoch": 5.8656987295825775, "step": 8080}, {"loss": 0.4261, "grad_norm": 1.119564414024353, "learning_rate": 0.0002, "epoch": 5.872958257713249, "step": 8090}, {"loss": 0.465, "grad_norm": 1.6517373323440552, "learning_rate": 0.0002, "epoch": 5.88021778584392, "step": 8100}, {"loss": 0.4406, "grad_norm": 1.4093554019927979, "learning_rate": 0.0002, "epoch": 5.887477313974592, "step": 8110}, {"loss": 0.4433, "grad_norm": 1.278843641281128, "learning_rate": 0.0002, "epoch": 5.894736842105263, "step": 8120}, {"loss": 0.4007, "grad_norm": 1.2042944431304932, "learning_rate": 0.0002, "epoch": 5.901996370235935, "step": 8130}, {"loss": 0.3972, "grad_norm": 1.1788326501846313, "learning_rate": 0.0002, "epoch": 5.909255898366606, "step": 8140}, {"loss": 0.4506, "grad_norm": 1.4364569187164307, "learning_rate": 0.0002, "epoch": 5.916515426497278, "step": 8150}, {"loss": 0.4651, "grad_norm": 1.1704283952713013, "learning_rate": 0.0002, "epoch": 5.923774954627949, "step": 8160}, {"loss": 0.3972, "grad_norm": 1.040814995765686, "learning_rate": 0.0002, "epoch": 5.931034482758621, "step": 8170}, {"loss": 0.4038, "grad_norm": 1.1367416381835938, "learning_rate": 0.0002, "epoch": 5.938294010889292, "step": 8180}, {"loss": 0.4387, "grad_norm": 1.3401511907577515, "learning_rate": 0.0002, "epoch": 5.945553539019964, "step": 8190}, {"loss": 0.4396, "grad_norm": 1.1154041290283203, "learning_rate": 0.0002, "epoch": 5.952813067150635, "step": 8200}, {"loss": 0.4744, "grad_norm": 1.426089882850647, "learning_rate": 0.0002, "epoch": 5.960072595281307, "step": 8210}, {"loss": 0.4105, "grad_norm": 1.3170222043991089, "learning_rate": 0.0002, "epoch": 5.967332123411978, "step": 8220}, {"loss": 0.4137, "grad_norm": 1.1960029602050781, "learning_rate": 0.0002, "epoch": 5.974591651542649, "step": 8230}, {"loss": 0.423, "grad_norm": 1.0843931436538696, "learning_rate": 0.0002, "epoch": 5.981851179673321, "step": 8240}, {"loss": 0.459, "grad_norm": 1.050421118736267, "learning_rate": 0.0002, "epoch": 5.9891107078039925, "step": 8250}, {"loss": 0.3993, "grad_norm": 1.0183138847351074, "learning_rate": 0.0002, "epoch": 5.9963702359346644, "step": 8260}]}
+{"epoch": 6.999637023593467, "step": 9642, "epoch_duration": 2054.809324026108, "total_accumulated_duration": 20458.157337903976, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7887.97119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4936, "grad_norm": 0.46010470390319824, "learning_rate": 0.0002, "epoch": 0.007259528130671506, "step": 10}, {"loss": 1.8442, "grad_norm": 0.5103179216384888, "learning_rate": 0.0002, "epoch": 0.014519056261343012, "step": 20}, {"loss": 1.559, "grad_norm": 0.6282716989517212, "learning_rate": 0.0002, "epoch": 0.021778584392014518, "step": 30}, {"loss": 1.3618, "grad_norm": 1.2480497360229492, "learning_rate": 0.0002, "epoch": 0.029038112522686024, "step": 40}, {"loss": 1.3874, "grad_norm": 0.4114855229854584, "learning_rate": 0.0002, "epoch": 0.036297640653357534, "step": 50}, {"loss": 1.2836, "grad_norm": 0.49482840299606323, "learning_rate": 0.0002, "epoch": 0.043557168784029036, "step": 60}, {"loss": 1.1728, "grad_norm": 0.4536272883415222, "learning_rate": 0.0002, "epoch": 0.050816696914700546, "step": 70}, {"loss": 1.1125, "grad_norm": 0.32328274846076965, "learning_rate": 0.0002, "epoch": 0.05807622504537205, "step": 80}, {"loss": 1.2447, "grad_norm": 0.40990468859672546, "learning_rate": 0.0002, "epoch": 0.06533575317604355, "step": 90}, {"loss": 1.2216, "grad_norm": 0.37273502349853516, "learning_rate": 0.0002, "epoch": 0.07259528130671507, "step": 100}, {"loss": 1.2259, "grad_norm": 0.3903871476650238, "learning_rate": 0.0002, "epoch": 0.07985480943738657, "step": 110}, {"loss": 1.3718, "grad_norm": 0.3272787928581238, "learning_rate": 0.0002, "epoch": 0.08711433756805807, "step": 120}, {"loss": 1.1832, "grad_norm": 0.3622824251651764, "learning_rate": 0.0002, "epoch": 0.09437386569872959, "step": 130}, {"loss": 1.2966, "grad_norm": 0.3503916561603546, "learning_rate": 0.0002, "epoch": 0.10163339382940109, "step": 140}, {"loss": 1.0913, "grad_norm": 0.32787832617759705, "learning_rate": 0.0002, "epoch": 0.1088929219600726, "step": 150}, {"loss": 1.2143, "grad_norm": 0.5822657942771912, "learning_rate": 0.0002, "epoch": 0.1161524500907441, "step": 160}, {"loss": 1.2091, "grad_norm": 0.28028249740600586, "learning_rate": 0.0002, "epoch": 0.12341197822141561, "step": 170}, {"loss": 1.1731, "grad_norm": 0.33602750301361084, "learning_rate": 0.0002, "epoch": 0.1306715063520871, "step": 180}, {"loss": 1.127, "grad_norm": 0.29106274247169495, "learning_rate": 0.0002, "epoch": 0.13793103448275862, "step": 190}, {"loss": 1.3261, "grad_norm": 0.38753363490104675, "learning_rate": 0.0002, "epoch": 0.14519056261343014, "step": 200}, {"loss": 1.1536, "grad_norm": 0.361009418964386, "learning_rate": 0.0002, "epoch": 0.15245009074410162, "step": 210}, {"loss": 1.1483, "grad_norm": 0.6743836402893066, "learning_rate": 0.0002, "epoch": 0.15970961887477314, "step": 220}, {"loss": 1.264, "grad_norm": 0.3923613727092743, "learning_rate": 0.0002, "epoch": 0.16696914700544466, "step": 230}, {"loss": 1.0437, "grad_norm": 0.2809699773788452, "learning_rate": 0.0002, "epoch": 0.17422867513611615, "step": 240}, {"loss": 1.1826, "grad_norm": 0.3631494641304016, "learning_rate": 0.0002, "epoch": 0.18148820326678766, "step": 250}, {"loss": 1.1584, "grad_norm": 0.24658171832561493, "learning_rate": 0.0002, "epoch": 0.18874773139745918, "step": 260}, {"loss": 1.0065, "grad_norm": 0.5780664682388306, "learning_rate": 0.0002, "epoch": 0.19600725952813067, "step": 270}, {"loss": 1.1251, "grad_norm": 0.3056720495223999, "learning_rate": 0.0002, "epoch": 0.20326678765880218, "step": 280}, {"loss": 1.1503, "grad_norm": 0.2857084572315216, "learning_rate": 0.0002, "epoch": 0.21052631578947367, "step": 290}, {"loss": 1.1215, "grad_norm": 0.6645345687866211, "learning_rate": 0.0002, "epoch": 0.2177858439201452, "step": 300}, {"loss": 1.127, "grad_norm": 0.5966078639030457, "learning_rate": 0.0002, "epoch": 0.2250453720508167, "step": 310}, {"loss": 1.1974, "grad_norm": 0.40937140583992004, "learning_rate": 0.0002, "epoch": 0.2323049001814882, "step": 320}, {"loss": 1.2385, "grad_norm": 0.5642806887626648, "learning_rate": 0.0002, "epoch": 0.2395644283121597, "step": 330}, {"loss": 1.1703, "grad_norm": 0.2750748097896576, "learning_rate": 0.0002, "epoch": 0.24682395644283123, "step": 340}, {"loss": 1.2598, "grad_norm": 0.34350234270095825, "learning_rate": 0.0002, "epoch": 0.2540834845735027, "step": 350}, {"loss": 1.1942, "grad_norm": 0.6767239570617676, "learning_rate": 0.0002, "epoch": 0.2613430127041742, "step": 360}, {"loss": 1.1436, "grad_norm": 0.31006959080696106, "learning_rate": 0.0002, "epoch": 0.26860254083484575, "step": 370}, {"loss": 1.2084, "grad_norm": 0.3825474679470062, "learning_rate": 0.0002, "epoch": 0.27586206896551724, "step": 380}, {"loss": 1.1523, "grad_norm": 0.30241551995277405, "learning_rate": 0.0002, "epoch": 0.2831215970961887, "step": 390}, {"loss": 1.1298, "grad_norm": 0.2962397336959839, "learning_rate": 0.0002, "epoch": 0.29038112522686027, "step": 400}, {"loss": 1.1299, "grad_norm": 0.2600369155406952, "learning_rate": 0.0002, "epoch": 0.29764065335753176, "step": 410}, {"loss": 1.1366, "grad_norm": 0.3675060272216797, "learning_rate": 0.0002, "epoch": 0.30490018148820325, "step": 420}, {"loss": 1.156, "grad_norm": 0.3429498076438904, "learning_rate": 0.0002, "epoch": 0.3121597096188748, "step": 430}, {"loss": 1.2741, "grad_norm": 0.34311825037002563, "learning_rate": 0.0002, "epoch": 0.3194192377495463, "step": 440}, {"loss": 1.3523, "grad_norm": 0.37872210144996643, "learning_rate": 0.0002, "epoch": 0.32667876588021777, "step": 450}, {"loss": 1.1365, "grad_norm": 0.33271121978759766, "learning_rate": 0.0002, "epoch": 0.3339382940108893, "step": 460}, {"loss": 1.1572, "grad_norm": 0.34605276584625244, "learning_rate": 0.0002, "epoch": 0.3411978221415608, "step": 470}, {"loss": 1.2755, "grad_norm": 0.41050562262535095, "learning_rate": 0.0002, "epoch": 0.3484573502722323, "step": 480}, {"loss": 1.075, "grad_norm": 0.2066836953163147, "learning_rate": 0.0002, "epoch": 0.35571687840290384, "step": 490}, {"loss": 1.2072, "grad_norm": 0.2859014868736267, "learning_rate": 0.0002, "epoch": 0.3629764065335753, "step": 500}, {"loss": 1.1435, "grad_norm": 0.28763777017593384, "learning_rate": 0.0002, "epoch": 0.3702359346642468, "step": 510}, {"loss": 1.1341, "grad_norm": 0.2730471193790436, "learning_rate": 0.0002, "epoch": 0.37749546279491836, "step": 520}, {"loss": 1.348, "grad_norm": 0.3968936800956726, "learning_rate": 0.0002, "epoch": 0.38475499092558985, "step": 530}, {"loss": 1.2363, "grad_norm": 0.3624701201915741, "learning_rate": 0.0002, "epoch": 0.39201451905626133, "step": 540}, {"loss": 1.1206, "grad_norm": 0.3303608298301697, "learning_rate": 0.0002, "epoch": 0.3992740471869328, "step": 550}, {"loss": 1.203, "grad_norm": 0.33507466316223145, "learning_rate": 0.0002, "epoch": 0.40653357531760437, "step": 560}, {"loss": 1.2294, "grad_norm": 0.3297670781612396, "learning_rate": 0.0002, "epoch": 0.41379310344827586, "step": 570}, {"loss": 1.2, "grad_norm": 0.32334890961647034, "learning_rate": 0.0002, "epoch": 0.42105263157894735, "step": 580}, {"loss": 1.112, "grad_norm": 0.30281195044517517, "learning_rate": 0.0002, "epoch": 0.4283121597096189, "step": 590}, {"loss": 1.1352, "grad_norm": 0.5900027751922607, "learning_rate": 0.0002, "epoch": 0.4355716878402904, "step": 600}, {"loss": 1.1575, "grad_norm": 0.28437477350234985, "learning_rate": 0.0002, "epoch": 0.44283121597096187, "step": 610}, {"loss": 1.2009, "grad_norm": 0.39601704478263855, "learning_rate": 0.0002, "epoch": 0.4500907441016334, "step": 620}, {"loss": 1.2302, "grad_norm": 0.41971510648727417, "learning_rate": 0.0002, "epoch": 0.4573502722323049, "step": 630}, {"loss": 1.236, "grad_norm": 0.33814409375190735, "learning_rate": 0.0002, "epoch": 0.4646098003629764, "step": 640}, {"loss": 1.2471, "grad_norm": 0.575718104839325, "learning_rate": 0.0002, "epoch": 0.47186932849364793, "step": 650}, {"loss": 1.1012, "grad_norm": 0.37927401065826416, "learning_rate": 0.0002, "epoch": 0.4791288566243194, "step": 660}, {"loss": 1.1552, "grad_norm": 0.3224332630634308, "learning_rate": 0.0002, "epoch": 0.4863883847549909, "step": 670}, {"loss": 1.1354, "grad_norm": 0.32683515548706055, "learning_rate": 0.0002, "epoch": 0.49364791288566245, "step": 680}, {"loss": 1.207, "grad_norm": 0.4316163659095764, "learning_rate": 0.0002, "epoch": 0.5009074410163339, "step": 690}, {"loss": 1.2116, "grad_norm": 0.342602401971817, "learning_rate": 0.0002, "epoch": 0.5081669691470054, "step": 700}, {"loss": 1.0433, "grad_norm": 0.2794898748397827, "learning_rate": 0.0002, "epoch": 0.515426497277677, "step": 710}, {"loss": 1.2395, "grad_norm": 0.3322339951992035, "learning_rate": 0.0002, "epoch": 0.5226860254083484, "step": 720}, {"loss": 1.0762, "grad_norm": 0.3088509142398834, "learning_rate": 0.0002, "epoch": 0.52994555353902, "step": 730}, {"loss": 1.1495, "grad_norm": 0.24444378912448883, "learning_rate": 0.0002, "epoch": 0.5372050816696915, "step": 740}, {"loss": 1.2176, "grad_norm": 0.3483171761035919, "learning_rate": 0.0002, "epoch": 0.5444646098003629, "step": 750}, {"loss": 1.2248, "grad_norm": 0.3425690531730652, "learning_rate": 0.0002, "epoch": 0.5517241379310345, "step": 760}, {"loss": 1.1127, "grad_norm": 0.31841927766799927, "learning_rate": 0.0002, "epoch": 0.558983666061706, "step": 770}, {"loss": 1.1841, "grad_norm": 0.39423868060112, "learning_rate": 0.0002, "epoch": 0.5662431941923775, "step": 780}, {"loss": 1.0583, "grad_norm": 0.30328479409217834, "learning_rate": 0.0002, "epoch": 0.573502722323049, "step": 790}, {"loss": 1.0718, "grad_norm": 0.24475938081741333, "learning_rate": 0.0002, "epoch": 0.5807622504537205, "step": 800}, {"loss": 1.2006, "grad_norm": 0.37132805585861206, "learning_rate": 0.0002, "epoch": 0.588021778584392, "step": 810}, {"loss": 1.1544, "grad_norm": 0.32195979356765747, "learning_rate": 0.0002, "epoch": 0.5952813067150635, "step": 820}, {"loss": 0.9937, "grad_norm": 0.2848738729953766, "learning_rate": 0.0002, "epoch": 0.6025408348457351, "step": 830}, {"loss": 1.0937, "grad_norm": 0.28015264868736267, "learning_rate": 0.0002, "epoch": 0.6098003629764065, "step": 840}, {"loss": 1.1234, "grad_norm": 0.37796008586883545, "learning_rate": 0.0002, "epoch": 0.617059891107078, "step": 850}, {"loss": 1.1455, "grad_norm": 0.39311841130256653, "learning_rate": 0.0002, "epoch": 0.6243194192377496, "step": 860}, {"loss": 1.1066, "grad_norm": 0.2761685252189636, "learning_rate": 0.0002, "epoch": 0.631578947368421, "step": 870}, {"loss": 1.0859, "grad_norm": 0.3826720118522644, "learning_rate": 0.0002, "epoch": 0.6388384754990926, "step": 880}, {"loss": 1.1182, "grad_norm": 0.30076679587364197, "learning_rate": 0.0002, "epoch": 0.6460980036297641, "step": 890}, {"loss": 1.0927, "grad_norm": 0.21997687220573425, "learning_rate": 0.0002, "epoch": 0.6533575317604355, "step": 900}, {"loss": 1.1603, "grad_norm": 0.32593777775764465, "learning_rate": 0.0002, "epoch": 0.6606170598911071, "step": 910}, {"loss": 1.1578, "grad_norm": 0.30347898602485657, "learning_rate": 0.0002, "epoch": 0.6678765880217786, "step": 920}, {"loss": 1.1642, "grad_norm": 0.44173774123191833, "learning_rate": 0.0002, "epoch": 0.6751361161524501, "step": 930}, {"loss": 1.1832, "grad_norm": 0.2507467269897461, "learning_rate": 0.0002, "epoch": 0.6823956442831216, "step": 940}, {"loss": 1.3001, "grad_norm": 0.29463833570480347, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 950}, {"loss": 1.1651, "grad_norm": 0.9363154172897339, "learning_rate": 0.0002, "epoch": 0.6969147005444646, "step": 960}, {"loss": 1.158, "grad_norm": 0.3236212134361267, "learning_rate": 0.0002, "epoch": 0.7041742286751361, "step": 970}, {"loss": 1.1078, "grad_norm": 0.3123254179954529, "learning_rate": 0.0002, "epoch": 0.7114337568058077, "step": 980}, {"loss": 1.1578, "grad_norm": 0.3395805060863495, "learning_rate": 0.0002, "epoch": 0.7186932849364791, "step": 990}, {"loss": 1.0754, "grad_norm": 0.3240964412689209, "learning_rate": 0.0002, "epoch": 0.7259528130671506, "step": 1000}, {"loss": 1.0313, "grad_norm": 0.31902948021888733, "learning_rate": 0.0002, "epoch": 0.7332123411978222, "step": 1010}, {"loss": 1.1664, "grad_norm": 0.4848408102989197, "learning_rate": 0.0002, "epoch": 0.7404718693284936, "step": 1020}, {"loss": 1.0838, "grad_norm": 0.33006033301353455, "learning_rate": 0.0002, "epoch": 0.7477313974591652, "step": 1030}, {"loss": 1.0979, "grad_norm": 0.2928730547428131, "learning_rate": 0.0002, "epoch": 0.7549909255898367, "step": 1040}, {"loss": 1.1351, "grad_norm": 0.3529164791107178, "learning_rate": 0.0002, "epoch": 0.7622504537205081, "step": 1050}, {"loss": 1.1274, "grad_norm": 0.2736213803291321, "learning_rate": 0.0002, "epoch": 0.7695099818511797, "step": 1060}, {"loss": 1.1169, "grad_norm": 0.7200686931610107, "learning_rate": 0.0002, "epoch": 0.7767695099818511, "step": 1070}, {"loss": 1.1171, "grad_norm": 0.33396708965301514, "learning_rate": 0.0002, "epoch": 0.7840290381125227, "step": 1080}, {"loss": 1.2447, "grad_norm": 1.5760449171066284, "learning_rate": 0.0002, "epoch": 0.7912885662431942, "step": 1090}, {"loss": 1.1126, "grad_norm": 0.28138381242752075, "learning_rate": 0.0002, "epoch": 0.7985480943738656, "step": 1100}, {"loss": 1.2012, "grad_norm": 0.2597472369670868, "learning_rate": 0.0002, "epoch": 0.8058076225045372, "step": 1110}, {"loss": 1.1177, "grad_norm": 0.3305445611476898, "learning_rate": 0.0002, "epoch": 0.8130671506352087, "step": 1120}, {"loss": 1.0849, "grad_norm": 0.3934599459171295, "learning_rate": 0.0002, "epoch": 0.8203266787658802, "step": 1130}, {"loss": 1.0889, "grad_norm": 0.3472191393375397, "learning_rate": 0.0002, "epoch": 0.8275862068965517, "step": 1140}, {"loss": 1.265, "grad_norm": 0.2857365906238556, "learning_rate": 0.0002, "epoch": 0.8348457350272233, "step": 1150}, {"loss": 1.03, "grad_norm": 0.3207702934741974, "learning_rate": 0.0002, "epoch": 0.8421052631578947, "step": 1160}, {"loss": 1.1669, "grad_norm": 0.3176484704017639, "learning_rate": 0.0002, "epoch": 0.8493647912885662, "step": 1170}, {"loss": 1.1386, "grad_norm": 0.40685558319091797, "learning_rate": 0.0002, "epoch": 0.8566243194192378, "step": 1180}, {"loss": 1.1383, "grad_norm": 0.31125199794769287, "learning_rate": 0.0002, "epoch": 0.8638838475499092, "step": 1190}, {"loss": 1.1512, "grad_norm": 0.7361181378364563, "learning_rate": 0.0002, "epoch": 0.8711433756805808, "step": 1200}, {"loss": 0.9983, "grad_norm": 0.33699527382850647, "learning_rate": 0.0002, "epoch": 0.8784029038112523, "step": 1210}, {"loss": 1.11, "grad_norm": 0.3315220773220062, "learning_rate": 0.0002, "epoch": 0.8856624319419237, "step": 1220}, {"loss": 1.0266, "grad_norm": 0.6256054043769836, "learning_rate": 0.0002, "epoch": 0.8929219600725953, "step": 1230}, {"loss": 1.1738, "grad_norm": 0.3692137897014618, "learning_rate": 0.0002, "epoch": 0.9001814882032668, "step": 1240}, {"loss": 1.1112, "grad_norm": 0.3538484573364258, "learning_rate": 0.0002, "epoch": 0.9074410163339383, "step": 1250}, {"loss": 1.0878, "grad_norm": 0.27863040566444397, "learning_rate": 0.0002, "epoch": 0.9147005444646098, "step": 1260}, {"loss": 1.0826, "grad_norm": 0.3322528302669525, "learning_rate": 0.0002, "epoch": 0.9219600725952813, "step": 1270}, {"loss": 1.0158, "grad_norm": 0.7553173303604126, "learning_rate": 0.0002, "epoch": 0.9292196007259528, "step": 1280}, {"loss": 1.198, "grad_norm": 0.4856191575527191, "learning_rate": 0.0002, "epoch": 0.9364791288566243, "step": 1290}, {"loss": 1.1678, "grad_norm": 0.3668074905872345, "learning_rate": 0.0002, "epoch": 0.9437386569872959, "step": 1300}, {"loss": 1.1809, "grad_norm": 0.29851067066192627, "learning_rate": 0.0002, "epoch": 0.9509981851179673, "step": 1310}, {"loss": 1.1299, "grad_norm": 0.276664674282074, "learning_rate": 0.0002, "epoch": 0.9582577132486388, "step": 1320}, {"loss": 1.0597, "grad_norm": 0.2941018342971802, "learning_rate": 0.0002, "epoch": 0.9655172413793104, "step": 1330}, {"loss": 1.1731, "grad_norm": 0.3505859076976776, "learning_rate": 0.0002, "epoch": 0.9727767695099818, "step": 1340}, {"loss": 1.1809, "grad_norm": 0.3067687451839447, "learning_rate": 0.0002, "epoch": 0.9800362976406534, "step": 1350}, {"loss": 1.1511, "grad_norm": 0.27151066064834595, "learning_rate": 0.0002, "epoch": 0.9872958257713249, "step": 1360}, {"loss": 1.1919, "grad_norm": 0.36370083689689636, "learning_rate": 0.0002, "epoch": 0.9945553539019963, "step": 1370}, {"eval_loss": 1.1381088495254517, "eval_runtime": 96.0848, "eval_samples_per_second": 4.538, "eval_steps_per_second": 0.572, "epoch": 0.9996370235934664, "step": 1377}, {"loss": 1.0082, "grad_norm": 0.27980583906173706, "learning_rate": 0.0002, "epoch": 1.0018148820326678, "step": 1380}, {"loss": 1.0565, "grad_norm": 0.26713913679122925, "learning_rate": 0.0002, "epoch": 1.0090744101633393, "step": 1390}, {"loss": 1.0241, "grad_norm": 0.3089541494846344, "learning_rate": 0.0002, "epoch": 1.0163339382940109, "step": 1400}, {"loss": 0.9649, "grad_norm": 0.4188242256641388, "learning_rate": 0.0002, "epoch": 1.0235934664246824, "step": 1410}, {"loss": 0.9821, "grad_norm": 0.5246463418006897, "learning_rate": 0.0002, "epoch": 1.030852994555354, "step": 1420}, {"loss": 1.0143, "grad_norm": 0.2728777825832367, "learning_rate": 0.0002, "epoch": 1.0381125226860255, "step": 1430}, {"loss": 1.075, "grad_norm": 0.38167616724967957, "learning_rate": 0.0002, "epoch": 1.0453720508166968, "step": 1440}, {"loss": 1.0229, "grad_norm": 0.4439380168914795, "learning_rate": 0.0002, "epoch": 1.0526315789473684, "step": 1450}, {"loss": 1.0451, "grad_norm": 0.30954182147979736, "learning_rate": 0.0002, "epoch": 1.05989110707804, "step": 1460}, {"loss": 1.0762, "grad_norm": 0.4022280275821686, "learning_rate": 0.0002, "epoch": 1.0671506352087115, "step": 1470}, {"loss": 1.0754, "grad_norm": 0.7390811443328857, "learning_rate": 0.0002, "epoch": 1.074410163339383, "step": 1480}, {"loss": 1.1378, "grad_norm": 0.3885486423969269, "learning_rate": 0.0002, "epoch": 1.0816696914700545, "step": 1490}, {"loss": 1.005, "grad_norm": 0.5275560617446899, "learning_rate": 0.0002, "epoch": 1.0889292196007259, "step": 1500}, {"loss": 1.0147, "grad_norm": 0.35112282633781433, "learning_rate": 0.0002, "epoch": 1.0961887477313974, "step": 1510}, {"loss": 1.0331, "grad_norm": 0.33714351058006287, "learning_rate": 0.0002, "epoch": 1.103448275862069, "step": 1520}, {"loss": 0.8723, "grad_norm": 0.31221693754196167, "learning_rate": 0.0002, "epoch": 1.1107078039927405, "step": 1530}, {"loss": 1.0701, "grad_norm": 0.27549654245376587, "learning_rate": 0.0002, "epoch": 1.117967332123412, "step": 1540}, {"loss": 1.0114, "grad_norm": 0.8465521335601807, "learning_rate": 0.0002, "epoch": 1.1252268602540836, "step": 1550}, {"loss": 0.9643, "grad_norm": 0.36125949025154114, "learning_rate": 0.0002, "epoch": 1.132486388384755, "step": 1560}, {"loss": 1.0744, "grad_norm": 0.37420371174812317, "learning_rate": 0.0002, "epoch": 1.1397459165154264, "step": 1570}, {"loss": 1.165, "grad_norm": 0.3294760584831238, "learning_rate": 0.0002, "epoch": 1.147005444646098, "step": 1580}, {"loss": 0.9207, "grad_norm": 0.3881238102912903, "learning_rate": 0.0002, "epoch": 1.1542649727767695, "step": 1590}, {"loss": 1.0983, "grad_norm": 0.4766491651535034, "learning_rate": 0.0002, "epoch": 1.161524500907441, "step": 1600}, {"loss": 1.0092, "grad_norm": 0.395530104637146, "learning_rate": 0.0002, "epoch": 1.1687840290381124, "step": 1610}, {"loss": 0.9895, "grad_norm": 0.3297106623649597, "learning_rate": 0.0002, "epoch": 1.176043557168784, "step": 1620}, {"loss": 1.1121, "grad_norm": 0.39528271555900574, "learning_rate": 0.0002, "epoch": 1.1833030852994555, "step": 1630}, {"loss": 0.9202, "grad_norm": 0.3370221257209778, "learning_rate": 0.0002, "epoch": 1.190562613430127, "step": 1640}, {"loss": 1.102, "grad_norm": 0.31922030448913574, "learning_rate": 0.0002, "epoch": 1.1978221415607986, "step": 1650}, {"loss": 1.0565, "grad_norm": 0.6142027378082275, "learning_rate": 0.0002, "epoch": 1.2050816696914701, "step": 1660}, {"loss": 0.9541, "grad_norm": 0.44769710302352905, "learning_rate": 0.0002, "epoch": 1.2123411978221417, "step": 1670}, {"loss": 1.2501, "grad_norm": 0.41233646869659424, "learning_rate": 0.0002, "epoch": 1.219600725952813, "step": 1680}, {"loss": 1.0147, "grad_norm": 0.2928866147994995, "learning_rate": 0.0002, "epoch": 1.2268602540834845, "step": 1690}, {"loss": 0.9074, "grad_norm": 0.36913734674453735, "learning_rate": 0.0002, "epoch": 1.234119782214156, "step": 1700}, {"loss": 1.0847, "grad_norm": 0.5281891226768494, "learning_rate": 0.0002, "epoch": 1.2413793103448276, "step": 1710}, {"loss": 0.9482, "grad_norm": 0.3374697268009186, "learning_rate": 0.0002, "epoch": 1.2486388384754992, "step": 1720}, {"loss": 0.9826, "grad_norm": 0.3802020847797394, "learning_rate": 0.0002, "epoch": 1.2558983666061705, "step": 1730}, {"loss": 1.0191, "grad_norm": 0.38048651814460754, "learning_rate": 0.0002, "epoch": 1.263157894736842, "step": 1740}, {"loss": 1.0142, "grad_norm": 0.6676169633865356, "learning_rate": 0.0002, "epoch": 1.2704174228675136, "step": 1750}, {"loss": 0.992, "grad_norm": 0.4075961410999298, "learning_rate": 0.0002, "epoch": 1.2776769509981851, "step": 1760}, {"loss": 1.0301, "grad_norm": 0.4374721348285675, "learning_rate": 0.0002, "epoch": 1.2849364791288567, "step": 1770}, {"loss": 1.0601, "grad_norm": 0.4638824164867401, "learning_rate": 0.0002, "epoch": 1.2921960072595282, "step": 1780}, {"loss": 0.9593, "grad_norm": 0.38631564378738403, "learning_rate": 0.0002, "epoch": 1.2994555353901998, "step": 1790}, {"loss": 1.0042, "grad_norm": 0.35873809456825256, "learning_rate": 0.0002, "epoch": 1.306715063520871, "step": 1800}, {"loss": 0.9835, "grad_norm": 0.33602237701416016, "learning_rate": 0.0002, "epoch": 1.3139745916515426, "step": 1810}, {"loss": 1.1032, "grad_norm": 0.46696463227272034, "learning_rate": 0.0002, "epoch": 1.3212341197822142, "step": 1820}, {"loss": 1.0123, "grad_norm": 0.368958979845047, "learning_rate": 0.0002, "epoch": 1.3284936479128857, "step": 1830}, {"loss": 1.0456, "grad_norm": 0.3160957396030426, "learning_rate": 0.0002, "epoch": 1.335753176043557, "step": 1840}, {"loss": 1.0264, "grad_norm": 0.4511511027812958, "learning_rate": 0.0002, "epoch": 1.3430127041742286, "step": 1850}, {"loss": 1.0179, "grad_norm": 0.6769845485687256, "learning_rate": 0.0002, "epoch": 1.3502722323049001, "step": 1860}, {"loss": 1.011, "grad_norm": 0.3749309480190277, "learning_rate": 0.0002, "epoch": 1.3575317604355717, "step": 1870}, {"loss": 1.0223, "grad_norm": 0.553798496723175, "learning_rate": 0.0002, "epoch": 1.3647912885662432, "step": 1880}, {"loss": 0.9634, "grad_norm": 0.3538985550403595, "learning_rate": 0.0002, "epoch": 1.3720508166969148, "step": 1890}, {"loss": 1.0275, "grad_norm": 0.5501534938812256, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1900}, {"loss": 1.0461, "grad_norm": 0.4432051181793213, "learning_rate": 0.0002, "epoch": 1.3865698729582578, "step": 1910}, {"loss": 0.9082, "grad_norm": 0.41755786538124084, "learning_rate": 0.0002, "epoch": 1.3938294010889292, "step": 1920}, {"loss": 1.0858, "grad_norm": 0.5626114010810852, "learning_rate": 0.0002, "epoch": 1.4010889292196007, "step": 1930}, {"loss": 1.0687, "grad_norm": 0.44038185477256775, "learning_rate": 0.0002, "epoch": 1.4083484573502723, "step": 1940}, {"loss": 0.9454, "grad_norm": 0.3397001624107361, "learning_rate": 0.0002, "epoch": 1.4156079854809438, "step": 1950}, {"loss": 0.943, "grad_norm": 0.4325368106365204, "learning_rate": 0.0002, "epoch": 1.4228675136116151, "step": 1960}, {"loss": 1.0488, "grad_norm": 0.3900907039642334, "learning_rate": 0.0002, "epoch": 1.4301270417422867, "step": 1970}, {"loss": 0.9699, "grad_norm": 0.369612455368042, "learning_rate": 0.0002, "epoch": 1.4373865698729582, "step": 1980}, {"loss": 1.0609, "grad_norm": 0.4389338791370392, "learning_rate": 0.0002, "epoch": 1.4446460980036298, "step": 1990}, {"loss": 1.042, "grad_norm": 1.694450855255127, "learning_rate": 0.0002, "epoch": 1.4519056261343013, "step": 2000}, {"loss": 0.997, "grad_norm": 0.516957700252533, "learning_rate": 0.0002, "epoch": 1.4591651542649728, "step": 2010}, {"loss": 1.0979, "grad_norm": 0.45515501499176025, "learning_rate": 0.0002, "epoch": 1.4664246823956444, "step": 2020}, {"loss": 1.0188, "grad_norm": 0.4153139591217041, "learning_rate": 0.0002, "epoch": 1.4736842105263157, "step": 2030}, {"loss": 1.1521, "grad_norm": 0.44353052973747253, "learning_rate": 0.0002, "epoch": 1.4809437386569873, "step": 2040}, {"loss": 0.9653, "grad_norm": 0.570554256439209, "learning_rate": 0.0002, "epoch": 1.4882032667876588, "step": 2050}, {"loss": 1.0352, "grad_norm": 0.5742740035057068, "learning_rate": 0.0002, "epoch": 1.4954627949183303, "step": 2060}, {"loss": 1.0963, "grad_norm": 0.5890517830848694, "learning_rate": 0.0002, "epoch": 1.5027223230490017, "step": 2070}, {"loss": 1.0639, "grad_norm": 0.4162650406360626, "learning_rate": 0.0002, "epoch": 1.5099818511796732, "step": 2080}, {"loss": 1.071, "grad_norm": 0.7334717512130737, "learning_rate": 0.0002, "epoch": 1.5172413793103448, "step": 2090}, {"loss": 0.8957, "grad_norm": 0.2963249683380127, "learning_rate": 0.0002, "epoch": 1.5245009074410163, "step": 2100}, {"loss": 1.0446, "grad_norm": 0.30676454305648804, "learning_rate": 0.0002, "epoch": 1.5317604355716878, "step": 2110}, {"loss": 0.9967, "grad_norm": 0.35984641313552856, "learning_rate": 0.0002, "epoch": 1.5390199637023594, "step": 2120}, {"loss": 1.0352, "grad_norm": 0.3384549617767334, "learning_rate": 0.0002, "epoch": 1.546279491833031, "step": 2130}, {"loss": 0.9874, "grad_norm": 0.4725518226623535, "learning_rate": 0.0002, "epoch": 1.5535390199637025, "step": 2140}, {"loss": 1.1166, "grad_norm": 0.3252685070037842, "learning_rate": 0.0002, "epoch": 1.560798548094374, "step": 2150}, {"loss": 0.9127, "grad_norm": 0.45043081045150757, "learning_rate": 0.0002, "epoch": 1.5680580762250453, "step": 2160}, {"loss": 1.0767, "grad_norm": 0.374208003282547, "learning_rate": 0.0002, "epoch": 1.5753176043557169, "step": 2170}, {"loss": 1.0255, "grad_norm": 0.5118404030799866, "learning_rate": 0.0002, "epoch": 1.5825771324863884, "step": 2180}, {"loss": 1.0062, "grad_norm": 0.482170969247818, "learning_rate": 0.0002, "epoch": 1.5898366606170597, "step": 2190}, {"loss": 1.0872, "grad_norm": 0.5337533950805664, "learning_rate": 0.0002, "epoch": 1.5970961887477313, "step": 2200}, {"loss": 1.0405, "grad_norm": 0.5195064544677734, "learning_rate": 0.0002, "epoch": 1.6043557168784028, "step": 2210}, {"loss": 1.0454, "grad_norm": 0.30807098746299744, "learning_rate": 0.0002, "epoch": 1.6116152450090744, "step": 2220}, {"loss": 1.0293, "grad_norm": 0.3962925672531128, "learning_rate": 0.0002, "epoch": 1.618874773139746, "step": 2230}, {"loss": 1.0137, "grad_norm": 0.7636962532997131, "learning_rate": 0.0002, "epoch": 1.6261343012704175, "step": 2240}, {"loss": 0.991, "grad_norm": 0.32380592823028564, "learning_rate": 0.0002, "epoch": 1.633393829401089, "step": 2250}, {"loss": 0.9471, "grad_norm": 0.5767741799354553, "learning_rate": 0.0002, "epoch": 1.6406533575317606, "step": 2260}, {"loss": 0.9524, "grad_norm": 0.39964812994003296, "learning_rate": 0.0002, "epoch": 1.647912885662432, "step": 2270}, {"loss": 0.9866, "grad_norm": 0.622629702091217, "learning_rate": 0.0002, "epoch": 1.6551724137931034, "step": 2280}, {"loss": 0.9609, "grad_norm": 0.40202152729034424, "learning_rate": 0.0002, "epoch": 1.662431941923775, "step": 2290}, {"loss": 0.9859, "grad_norm": 0.4467349052429199, "learning_rate": 0.0002, "epoch": 1.6696914700544465, "step": 2300}, {"loss": 1.0312, "grad_norm": 0.5026949048042297, "learning_rate": 0.0002, "epoch": 1.6769509981851178, "step": 2310}, {"loss": 1.0776, "grad_norm": 0.43754541873931885, "learning_rate": 0.0002, "epoch": 1.6842105263157894, "step": 2320}, {"loss": 0.9786, "grad_norm": 0.42869430780410767, "learning_rate": 0.0002, "epoch": 1.691470054446461, "step": 2330}, {"loss": 1.0747, "grad_norm": 0.4192679524421692, "learning_rate": 0.0002, "epoch": 1.6987295825771325, "step": 2340}, {"loss": 1.0054, "grad_norm": 0.3243155777454376, "learning_rate": 0.0002, "epoch": 1.705989110707804, "step": 2350}, {"loss": 0.9165, "grad_norm": 1.0514075756072998, "learning_rate": 0.0002, "epoch": 1.7132486388384756, "step": 2360}, {"loss": 1.1353, "grad_norm": 0.4719122648239136, "learning_rate": 0.0002, "epoch": 1.720508166969147, "step": 2370}, {"loss": 0.9913, "grad_norm": 0.3846144676208496, "learning_rate": 0.0002, "epoch": 1.7277676950998186, "step": 2380}, {"loss": 0.9935, "grad_norm": 0.7266581058502197, "learning_rate": 0.0002, "epoch": 1.73502722323049, "step": 2390}, {"loss": 0.9509, "grad_norm": 0.6183241605758667, "learning_rate": 0.0002, "epoch": 1.7422867513611615, "step": 2400}, {"loss": 0.9713, "grad_norm": 0.3658260405063629, "learning_rate": 0.0002, "epoch": 1.749546279491833, "step": 2410}, {"loss": 1.0033, "grad_norm": 0.6036322712898254, "learning_rate": 0.0002, "epoch": 1.7568058076225044, "step": 2420}, {"loss": 1.0526, "grad_norm": 0.7872936129570007, "learning_rate": 0.0002, "epoch": 1.764065335753176, "step": 2430}, {"loss": 0.99, "grad_norm": 0.35946124792099, "learning_rate": 0.0002, "epoch": 1.7713248638838475, "step": 2440}, {"loss": 0.8845, "grad_norm": 0.3740338981151581, "learning_rate": 0.0002, "epoch": 1.778584392014519, "step": 2450}, {"loss": 0.9333, "grad_norm": 0.6150230169296265, "learning_rate": 0.0002, "epoch": 1.7858439201451906, "step": 2460}, {"loss": 1.0373, "grad_norm": 0.4726075530052185, "learning_rate": 0.0002, "epoch": 1.793103448275862, "step": 2470}, {"loss": 0.974, "grad_norm": 0.31292253732681274, "learning_rate": 0.0002, "epoch": 1.8003629764065336, "step": 2480}, {"loss": 1.0205, "grad_norm": 0.4463104009628296, "learning_rate": 0.0002, "epoch": 1.8076225045372052, "step": 2490}, {"loss": 1.0771, "grad_norm": 0.7848200798034668, "learning_rate": 0.0002, "epoch": 1.8148820326678767, "step": 2500}, {"loss": 0.9508, "grad_norm": 0.5562082529067993, "learning_rate": 0.0002, "epoch": 1.822141560798548, "step": 2510}, {"loss": 0.9343, "grad_norm": 0.39892756938934326, "learning_rate": 0.0002, "epoch": 1.8294010889292196, "step": 2520}, {"loss": 1.0489, "grad_norm": 1.2923320531845093, "learning_rate": 0.0002, "epoch": 1.8366606170598911, "step": 2530}, {"loss": 1.1203, "grad_norm": 0.6316490769386292, "learning_rate": 0.0002, "epoch": 1.8439201451905625, "step": 2540}, {"loss": 0.9131, "grad_norm": 0.46100497245788574, "learning_rate": 0.0002, "epoch": 1.851179673321234, "step": 2550}, {"loss": 1.0977, "grad_norm": 0.7902987003326416, "learning_rate": 0.0002, "epoch": 1.8584392014519056, "step": 2560}, {"loss": 0.9702, "grad_norm": 0.4596365690231323, "learning_rate": 0.0002, "epoch": 1.865698729582577, "step": 2570}, {"loss": 0.9322, "grad_norm": 0.6592172384262085, "learning_rate": 0.0002, "epoch": 1.8729582577132486, "step": 2580}, {"loss": 0.9549, "grad_norm": 0.5764662027359009, "learning_rate": 0.0002, "epoch": 1.8802177858439202, "step": 2590}, {"loss": 0.9485, "grad_norm": 0.8421637415885925, "learning_rate": 0.0002, "epoch": 1.8874773139745917, "step": 2600}, {"loss": 1.012, "grad_norm": 0.5635305047035217, "learning_rate": 0.0002, "epoch": 1.8947368421052633, "step": 2610}, {"loss": 0.8907, "grad_norm": 0.46231237053871155, "learning_rate": 0.0002, "epoch": 1.9019963702359348, "step": 2620}, {"loss": 0.9543, "grad_norm": 0.3944607079029083, "learning_rate": 0.0002, "epoch": 1.9092558983666061, "step": 2630}, {"loss": 0.9964, "grad_norm": 0.4753907322883606, "learning_rate": 0.0002, "epoch": 1.9165154264972777, "step": 2640}, {"loss": 1.0217, "grad_norm": 0.4151090979576111, "learning_rate": 0.0002, "epoch": 1.9237749546279492, "step": 2650}, {"loss": 1.0397, "grad_norm": 0.6793725490570068, "learning_rate": 0.0002, "epoch": 1.9310344827586206, "step": 2660}, {"loss": 1.0401, "grad_norm": 0.339755117893219, "learning_rate": 0.0002, "epoch": 1.938294010889292, "step": 2670}, {"loss": 1.082, "grad_norm": 0.40789374709129333, "learning_rate": 0.0002, "epoch": 1.9455535390199636, "step": 2680}, {"loss": 0.9973, "grad_norm": 0.3750005066394806, "learning_rate": 0.0002, "epoch": 1.9528130671506352, "step": 2690}, {"loss": 0.9837, "grad_norm": 0.39684441685676575, "learning_rate": 0.0002, "epoch": 1.9600725952813067, "step": 2700}, {"loss": 1.1204, "grad_norm": 0.378287672996521, "learning_rate": 0.0002, "epoch": 1.9673321234119783, "step": 2710}, {"loss": 0.9284, "grad_norm": 0.3668482005596161, "learning_rate": 0.0002, "epoch": 1.9745916515426498, "step": 2720}, {"loss": 1.0512, "grad_norm": 0.49997565150260925, "learning_rate": 0.0002, "epoch": 1.9818511796733214, "step": 2730}, {"loss": 0.9311, "grad_norm": 0.36852124333381653, "learning_rate": 0.0002, "epoch": 1.989110707803993, "step": 2740}, {"loss": 1.0898, "grad_norm": 0.5203380584716797, "learning_rate": 0.0002, "epoch": 1.9963702359346642, "step": 2750}, {"eval_loss": 1.1086540222167969, "eval_runtime": 95.6866, "eval_samples_per_second": 4.557, "eval_steps_per_second": 0.575, "epoch": 2.0, "step": 2755}, {"loss": 0.9791, "grad_norm": 0.35921573638916016, "learning_rate": 0.0002, "epoch": 2.0036297640653356, "step": 2760}, {"loss": 0.8968, "grad_norm": 1.013861894607544, "learning_rate": 0.0002, "epoch": 2.010889292196007, "step": 2770}, {"loss": 0.7873, "grad_norm": 0.4425240159034729, "learning_rate": 0.0002, "epoch": 2.0181488203266786, "step": 2780}, {"loss": 0.8758, "grad_norm": 0.60174161195755, "learning_rate": 0.0002, "epoch": 2.02540834845735, "step": 2790}, {"loss": 0.7989, "grad_norm": 0.47582098841667175, "learning_rate": 0.0002, "epoch": 2.0326678765880217, "step": 2800}, {"loss": 0.8898, "grad_norm": 0.6012811660766602, "learning_rate": 0.0002, "epoch": 2.0399274047186933, "step": 2810}, {"loss": 0.8109, "grad_norm": 0.4444408118724823, "learning_rate": 0.0002, "epoch": 2.047186932849365, "step": 2820}, {"loss": 0.8621, "grad_norm": 0.4864003360271454, "learning_rate": 0.0002, "epoch": 2.0544464609800364, "step": 2830}, {"loss": 0.8817, "grad_norm": 0.5104215741157532, "learning_rate": 0.0002, "epoch": 2.061705989110708, "step": 2840}, {"loss": 0.7789, "grad_norm": 0.6218489408493042, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 2850}, {"loss": 0.8778, "grad_norm": 0.705784261226654, "learning_rate": 0.0002, "epoch": 2.076225045372051, "step": 2860}, {"loss": 0.7851, "grad_norm": 0.48091503977775574, "learning_rate": 0.0002, "epoch": 2.0834845735027225, "step": 2870}, {"loss": 0.8444, "grad_norm": 0.5062456727027893, "learning_rate": 0.0002, "epoch": 2.0907441016333936, "step": 2880}, {"loss": 0.7898, "grad_norm": 0.4862022399902344, "learning_rate": 0.0002, "epoch": 2.098003629764065, "step": 2890}, {"loss": 0.8593, "grad_norm": 0.48264044523239136, "learning_rate": 0.0002, "epoch": 2.1052631578947367, "step": 2900}, {"loss": 0.8325, "grad_norm": 0.43744346499443054, "learning_rate": 0.0002, "epoch": 2.1125226860254083, "step": 2910}, {"loss": 0.9099, "grad_norm": 0.5480492115020752, "learning_rate": 0.0002, "epoch": 2.11978221415608, "step": 2920}, {"loss": 0.7727, "grad_norm": 0.5068560838699341, "learning_rate": 0.0002, "epoch": 2.1270417422867514, "step": 2930}, {"loss": 0.8403, "grad_norm": 0.4650852680206299, "learning_rate": 0.0002, "epoch": 2.134301270417423, "step": 2940}, {"loss": 0.8467, "grad_norm": 0.4929981231689453, "learning_rate": 0.0002, "epoch": 2.1415607985480944, "step": 2950}, {"loss": 0.8747, "grad_norm": 0.6537389755249023, "learning_rate": 0.0002, "epoch": 2.148820326678766, "step": 2960}, {"loss": 0.7663, "grad_norm": 0.8032940626144409, "learning_rate": 0.0002, "epoch": 2.1560798548094375, "step": 2970}, {"loss": 0.7604, "grad_norm": 0.7131643891334534, "learning_rate": 0.0002, "epoch": 2.163339382940109, "step": 2980}, {"loss": 0.8424, "grad_norm": 0.6034275889396667, "learning_rate": 0.0002, "epoch": 2.1705989110707806, "step": 2990}, {"loss": 0.8838, "grad_norm": 0.6081095933914185, "learning_rate": 0.0002, "epoch": 2.1778584392014517, "step": 3000}, {"loss": 0.9013, "grad_norm": 0.5706912875175476, "learning_rate": 0.0002, "epoch": 2.1851179673321233, "step": 3010}, {"loss": 0.8473, "grad_norm": 0.6742380261421204, "learning_rate": 0.0002, "epoch": 2.192377495462795, "step": 3020}, {"loss": 0.8372, "grad_norm": 0.4847496449947357, "learning_rate": 0.0002, "epoch": 2.1996370235934664, "step": 3030}, {"loss": 0.8487, "grad_norm": 0.5529342889785767, "learning_rate": 0.0002, "epoch": 2.206896551724138, "step": 3040}, {"loss": 0.9057, "grad_norm": 0.6108783483505249, "learning_rate": 0.0002, "epoch": 2.2141560798548094, "step": 3050}, {"loss": 0.9659, "grad_norm": 0.8841571807861328, "learning_rate": 0.0002, "epoch": 2.221415607985481, "step": 3060}, {"loss": 0.8538, "grad_norm": 0.4227530360221863, "learning_rate": 0.0002, "epoch": 2.2286751361161525, "step": 3070}, {"loss": 0.7925, "grad_norm": 0.461935818195343, "learning_rate": 0.0002, "epoch": 2.235934664246824, "step": 3080}, {"loss": 0.8542, "grad_norm": 0.5407412648200989, "learning_rate": 0.0002, "epoch": 2.2431941923774956, "step": 3090}, {"loss": 0.8392, "grad_norm": 0.5057998895645142, "learning_rate": 0.0002, "epoch": 2.250453720508167, "step": 3100}, {"loss": 0.7898, "grad_norm": 0.530057966709137, "learning_rate": 0.0002, "epoch": 2.2577132486388383, "step": 3110}, {"loss": 0.8507, "grad_norm": 0.5066842436790466, "learning_rate": 0.0002, "epoch": 2.26497277676951, "step": 3120}, {"loss": 0.7903, "grad_norm": 0.5069178342819214, "learning_rate": 0.0002, "epoch": 2.2722323049001814, "step": 3130}, {"loss": 0.8458, "grad_norm": 0.6095499396324158, "learning_rate": 0.0002, "epoch": 2.279491833030853, "step": 3140}, {"loss": 0.8171, "grad_norm": 0.49524766206741333, "learning_rate": 0.0002, "epoch": 2.2867513611615244, "step": 3150}, {"loss": 0.7568, "grad_norm": 0.5334409475326538, "learning_rate": 0.0002, "epoch": 2.294010889292196, "step": 3160}, {"loss": 0.7533, "grad_norm": 1.681748867034912, "learning_rate": 0.0002, "epoch": 2.3012704174228675, "step": 3170}, {"loss": 0.8155, "grad_norm": 0.7225565314292908, "learning_rate": 0.0002, "epoch": 2.308529945553539, "step": 3180}, {"loss": 0.8034, "grad_norm": 0.5379388928413391, "learning_rate": 0.0002, "epoch": 2.3157894736842106, "step": 3190}, {"loss": 0.8833, "grad_norm": 0.45770326256752014, "learning_rate": 0.0002, "epoch": 2.323049001814882, "step": 3200}, {"loss": 0.8419, "grad_norm": 0.6984533071517944, "learning_rate": 0.0002, "epoch": 2.3303085299455537, "step": 3210}, {"loss": 0.7414, "grad_norm": 0.6725744605064392, "learning_rate": 0.0002, "epoch": 2.337568058076225, "step": 3220}, {"loss": 0.8657, "grad_norm": 1.1247471570968628, "learning_rate": 0.0002, "epoch": 2.344827586206897, "step": 3230}, {"loss": 0.7782, "grad_norm": 1.0240263938903809, "learning_rate": 0.0002, "epoch": 2.352087114337568, "step": 3240}, {"loss": 0.826, "grad_norm": 0.5608096122741699, "learning_rate": 0.0002, "epoch": 2.3593466424682394, "step": 3250}, {"loss": 0.8191, "grad_norm": 0.8294990062713623, "learning_rate": 0.0002, "epoch": 2.366606170598911, "step": 3260}, {"loss": 0.9845, "grad_norm": 0.6734224557876587, "learning_rate": 0.0002, "epoch": 2.3738656987295825, "step": 3270}, {"loss": 0.7921, "grad_norm": 0.6862800717353821, "learning_rate": 0.0002, "epoch": 2.381125226860254, "step": 3280}, {"loss": 0.7752, "grad_norm": 0.5442930459976196, "learning_rate": 0.0002, "epoch": 2.3883847549909256, "step": 3290}, {"loss": 0.8543, "grad_norm": 0.5745394229888916, "learning_rate": 0.0002, "epoch": 2.395644283121597, "step": 3300}, {"loss": 0.833, "grad_norm": 0.6257799863815308, "learning_rate": 0.0002, "epoch": 2.4029038112522687, "step": 3310}, {"loss": 0.8191, "grad_norm": 0.5608420968055725, "learning_rate": 0.0002, "epoch": 2.4101633393829403, "step": 3320}, {"loss": 0.8745, "grad_norm": 0.5512017011642456, "learning_rate": 0.0002, "epoch": 2.417422867513612, "step": 3330}, {"loss": 0.8642, "grad_norm": 0.7452999353408813, "learning_rate": 0.0002, "epoch": 2.4246823956442833, "step": 3340}, {"loss": 0.8417, "grad_norm": 0.4604301452636719, "learning_rate": 0.0002, "epoch": 2.4319419237749544, "step": 3350}, {"loss": 0.9148, "grad_norm": 0.8225823640823364, "learning_rate": 0.0002, "epoch": 2.439201451905626, "step": 3360}, {"loss": 0.8144, "grad_norm": 0.8604981303215027, "learning_rate": 0.0002, "epoch": 2.4464609800362975, "step": 3370}, {"loss": 0.9276, "grad_norm": 0.6620925664901733, "learning_rate": 0.0002, "epoch": 2.453720508166969, "step": 3380}, {"loss": 0.8381, "grad_norm": 0.4750158488750458, "learning_rate": 0.0002, "epoch": 2.4609800362976406, "step": 3390}, {"loss": 0.8249, "grad_norm": 0.6061418056488037, "learning_rate": 0.0002, "epoch": 2.468239564428312, "step": 3400}, {"loss": 0.8944, "grad_norm": 0.5977247357368469, "learning_rate": 0.0002, "epoch": 2.4754990925589837, "step": 3410}, {"loss": 0.8366, "grad_norm": 0.5004227757453918, "learning_rate": 0.0002, "epoch": 2.4827586206896552, "step": 3420}, {"loss": 0.9133, "grad_norm": 0.46428972482681274, "learning_rate": 0.0002, "epoch": 2.490018148820327, "step": 3430}, {"loss": 0.8534, "grad_norm": 0.7658395171165466, "learning_rate": 0.0002, "epoch": 2.4972776769509983, "step": 3440}, {"loss": 0.9066, "grad_norm": 0.621819794178009, "learning_rate": 0.0002, "epoch": 2.50453720508167, "step": 3450}, {"loss": 0.8481, "grad_norm": 0.4556088149547577, "learning_rate": 0.0002, "epoch": 2.511796733212341, "step": 3460}, {"loss": 0.8522, "grad_norm": 0.6124140024185181, "learning_rate": 0.0002, "epoch": 2.519056261343013, "step": 3470}, {"loss": 0.8158, "grad_norm": 0.6256231665611267, "learning_rate": 0.0002, "epoch": 2.526315789473684, "step": 3480}, {"loss": 0.9097, "grad_norm": 0.5464141964912415, "learning_rate": 0.0002, "epoch": 2.5335753176043556, "step": 3490}, {"loss": 0.8708, "grad_norm": 0.51471346616745, "learning_rate": 0.0002, "epoch": 2.540834845735027, "step": 3500}, {"loss": 0.9042, "grad_norm": 0.5326165556907654, "learning_rate": 0.0002, "epoch": 2.5480943738656987, "step": 3510}, {"loss": 0.8749, "grad_norm": 0.4750378429889679, "learning_rate": 0.0002, "epoch": 2.5553539019963702, "step": 3520}, {"loss": 0.8248, "grad_norm": 0.5292279124259949, "learning_rate": 0.0002, "epoch": 2.562613430127042, "step": 3530}, {"loss": 0.8808, "grad_norm": 0.6145227551460266, "learning_rate": 0.0002, "epoch": 2.5698729582577133, "step": 3540}, {"loss": 0.846, "grad_norm": 0.8275189399719238, "learning_rate": 0.0002, "epoch": 2.577132486388385, "step": 3550}, {"loss": 0.875, "grad_norm": 0.5037438273429871, "learning_rate": 0.0002, "epoch": 2.5843920145190564, "step": 3560}, {"loss": 0.8324, "grad_norm": 0.5838707685470581, "learning_rate": 0.0002, "epoch": 2.5916515426497275, "step": 3570}, {"loss": 0.8716, "grad_norm": 0.5398710370063782, "learning_rate": 0.0002, "epoch": 2.5989110707803995, "step": 3580}, {"loss": 0.7843, "grad_norm": 0.6115376949310303, "learning_rate": 0.0002, "epoch": 2.6061705989110706, "step": 3590}, {"loss": 0.8142, "grad_norm": 1.118809461593628, "learning_rate": 0.0002, "epoch": 2.613430127041742, "step": 3600}, {"loss": 0.8282, "grad_norm": 0.7811771631240845, "learning_rate": 0.0002, "epoch": 2.6206896551724137, "step": 3610}, {"loss": 0.7997, "grad_norm": 0.5753175020217896, "learning_rate": 0.0002, "epoch": 2.6279491833030852, "step": 3620}, {"loss": 0.7705, "grad_norm": 0.550829291343689, "learning_rate": 0.0002, "epoch": 2.635208711433757, "step": 3630}, {"loss": 0.8564, "grad_norm": 0.5360019207000732, "learning_rate": 0.0002, "epoch": 2.6424682395644283, "step": 3640}, {"loss": 0.7648, "grad_norm": 0.63050377368927, "learning_rate": 0.0002, "epoch": 2.6497277676951, "step": 3650}, {"loss": 0.9156, "grad_norm": 0.5833110213279724, "learning_rate": 0.0002, "epoch": 2.6569872958257714, "step": 3660}, {"loss": 0.8033, "grad_norm": 0.5543047189712524, "learning_rate": 0.0002, "epoch": 2.664246823956443, "step": 3670}, {"loss": 0.8867, "grad_norm": 0.6842212080955505, "learning_rate": 0.0002, "epoch": 2.671506352087114, "step": 3680}, {"loss": 0.8569, "grad_norm": 0.6859333515167236, "learning_rate": 0.0002, "epoch": 2.678765880217786, "step": 3690}, {"loss": 0.8081, "grad_norm": 0.7038410902023315, "learning_rate": 0.0002, "epoch": 2.686025408348457, "step": 3700}, {"loss": 0.8686, "grad_norm": 0.447233647108078, "learning_rate": 0.0002, "epoch": 2.6932849364791287, "step": 3710}, {"loss": 0.8951, "grad_norm": 0.6453872919082642, "learning_rate": 0.0002, "epoch": 2.7005444646098002, "step": 3720}, {"loss": 0.8292, "grad_norm": 0.8025672435760498, "learning_rate": 0.0002, "epoch": 2.707803992740472, "step": 3730}, {"loss": 0.8488, "grad_norm": 0.5997087955474854, "learning_rate": 0.0002, "epoch": 2.7150635208711433, "step": 3740}, {"loss": 0.8308, "grad_norm": 0.6901142001152039, "learning_rate": 0.0002, "epoch": 2.722323049001815, "step": 3750}, {"loss": 0.8517, "grad_norm": 1.036145567893982, "learning_rate": 0.0002, "epoch": 2.7295825771324864, "step": 3760}, {"loss": 0.8455, "grad_norm": 0.7207072377204895, "learning_rate": 0.0002, "epoch": 2.736842105263158, "step": 3770}, {"loss": 0.8009, "grad_norm": 1.0452989339828491, "learning_rate": 0.0002, "epoch": 2.7441016333938295, "step": 3780}, {"loss": 0.8868, "grad_norm": 0.5615278482437134, "learning_rate": 0.0002, "epoch": 2.751361161524501, "step": 3790}, {"loss": 0.9047, "grad_norm": 0.46439215540885925, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 3800}, {"loss": 0.7847, "grad_norm": 0.7134917974472046, "learning_rate": 0.0002, "epoch": 2.7658802177858437, "step": 3810}, {"loss": 0.8315, "grad_norm": 0.5139115452766418, "learning_rate": 0.0002, "epoch": 2.7731397459165157, "step": 3820}, {"loss": 0.8995, "grad_norm": 0.8595781326293945, "learning_rate": 0.0002, "epoch": 2.780399274047187, "step": 3830}, {"loss": 0.9079, "grad_norm": 0.544614851474762, "learning_rate": 0.0002, "epoch": 2.7876588021778583, "step": 3840}, {"loss": 0.8322, "grad_norm": 0.6073850393295288, "learning_rate": 0.0002, "epoch": 2.79491833030853, "step": 3850}, {"loss": 0.8277, "grad_norm": 0.8224069476127625, "learning_rate": 0.0002, "epoch": 2.8021778584392014, "step": 3860}, {"loss": 0.815, "grad_norm": 0.5347970128059387, "learning_rate": 0.0002, "epoch": 2.809437386569873, "step": 3870}, {"loss": 0.8219, "grad_norm": 0.6799601912498474, "learning_rate": 0.0002, "epoch": 2.8166969147005445, "step": 3880}, {"loss": 0.8796, "grad_norm": 0.5219197869300842, "learning_rate": 0.0002, "epoch": 2.823956442831216, "step": 3890}, {"loss": 0.8649, "grad_norm": 0.5710130333900452, "learning_rate": 0.0002, "epoch": 2.8312159709618876, "step": 3900}, {"loss": 0.8741, "grad_norm": 0.5857213139533997, "learning_rate": 0.0002, "epoch": 2.838475499092559, "step": 3910}, {"loss": 0.8154, "grad_norm": 0.5206644535064697, "learning_rate": 0.0002, "epoch": 2.8457350272232302, "step": 3920}, {"loss": 0.9328, "grad_norm": 1.2902015447616577, "learning_rate": 0.0002, "epoch": 2.8529945553539022, "step": 3930}, {"loss": 0.8192, "grad_norm": 0.5252797603607178, "learning_rate": 0.0002, "epoch": 2.8602540834845733, "step": 3940}, {"loss": 0.9047, "grad_norm": 0.5925108790397644, "learning_rate": 0.0002, "epoch": 2.867513611615245, "step": 3950}, {"loss": 0.8611, "grad_norm": 0.9719610810279846, "learning_rate": 0.0002, "epoch": 2.8747731397459164, "step": 3960}, {"loss": 0.8531, "grad_norm": 0.6834747195243835, "learning_rate": 0.0002, "epoch": 2.882032667876588, "step": 3970}, {"loss": 0.8124, "grad_norm": 0.6952353119850159, "learning_rate": 0.0002, "epoch": 2.8892921960072595, "step": 3980}, {"loss": 0.8365, "grad_norm": 0.49889910221099854, "learning_rate": 0.0002, "epoch": 2.896551724137931, "step": 3990}, {"loss": 0.8197, "grad_norm": 0.5007026791572571, "learning_rate": 0.0002, "epoch": 2.9038112522686026, "step": 4000}, {"loss": 0.7752, "grad_norm": 0.5474239587783813, "learning_rate": 0.0002, "epoch": 2.911070780399274, "step": 4010}, {"loss": 0.8579, "grad_norm": 0.6617428660392761, "learning_rate": 0.0002, "epoch": 2.9183303085299457, "step": 4020}, {"loss": 0.8583, "grad_norm": 0.6097776293754578, "learning_rate": 0.0002, "epoch": 2.925589836660617, "step": 4030}, {"loss": 0.8296, "grad_norm": 0.5985828638076782, "learning_rate": 0.0002, "epoch": 2.9328493647912888, "step": 4040}, {"loss": 0.9134, "grad_norm": 0.769488513469696, "learning_rate": 0.0002, "epoch": 2.94010889292196, "step": 4050}, {"loss": 0.8148, "grad_norm": 0.5167055130004883, "learning_rate": 0.0002, "epoch": 2.9473684210526314, "step": 4060}, {"loss": 0.8665, "grad_norm": 0.6265496015548706, "learning_rate": 0.0002, "epoch": 2.954627949183303, "step": 4070}, {"loss": 0.9218, "grad_norm": 1.2644082307815552, "learning_rate": 0.0002, "epoch": 2.9618874773139745, "step": 4080}, {"loss": 0.8026, "grad_norm": 0.6007561087608337, "learning_rate": 0.0002, "epoch": 2.969147005444646, "step": 4090}, {"loss": 0.8262, "grad_norm": 0.47984927892684937, "learning_rate": 0.0002, "epoch": 2.9764065335753176, "step": 4100}, {"loss": 0.7897, "grad_norm": 1.128198504447937, "learning_rate": 0.0002, "epoch": 2.983666061705989, "step": 4110}, {"loss": 0.7883, "grad_norm": 0.526292085647583, "learning_rate": 0.0002, "epoch": 2.9909255898366607, "step": 4120}, {"loss": 0.7801, "grad_norm": 0.5462674498558044, "learning_rate": 0.0002, "epoch": 2.9981851179673322, "step": 4130}, {"eval_loss": 1.1093357801437378, "eval_runtime": 46.2498, "eval_samples_per_second": 9.427, "eval_steps_per_second": 1.189, "epoch": 2.9996370235934666, "step": 4132}, {"loss": 0.721, "grad_norm": 0.48288026452064514, "learning_rate": 0.0002, "epoch": 3.0054446460980038, "step": 4140}, {"loss": 0.6769, "grad_norm": 1.0181782245635986, "learning_rate": 0.0002, "epoch": 3.0127041742286753, "step": 4150}, {"loss": 0.7185, "grad_norm": 0.7718019485473633, "learning_rate": 0.0002, "epoch": 3.019963702359347, "step": 4160}, {"loss": 0.6552, "grad_norm": 0.7492219805717468, "learning_rate": 0.0002, "epoch": 3.027223230490018, "step": 4170}, {"loss": 0.6678, "grad_norm": 0.9363632798194885, "learning_rate": 0.0002, "epoch": 3.0344827586206895, "step": 4180}, {"loss": 0.7187, "grad_norm": 0.6888533234596252, "learning_rate": 0.0002, "epoch": 3.041742286751361, "step": 4190}, {"loss": 0.6469, "grad_norm": 0.7072834968566895, "learning_rate": 0.0002, "epoch": 3.0490018148820326, "step": 4200}, {"loss": 0.6387, "grad_norm": 0.7182047963142395, "learning_rate": 0.0002, "epoch": 3.056261343012704, "step": 4210}, {"loss": 0.6385, "grad_norm": 0.7194355130195618, "learning_rate": 0.0002, "epoch": 3.0635208711433757, "step": 4220}, {"loss": 0.5812, "grad_norm": 0.9454023838043213, "learning_rate": 0.0002, "epoch": 3.0707803992740472, "step": 4230}, {"loss": 0.6036, "grad_norm": 0.838657557964325, "learning_rate": 0.0002, "epoch": 3.0780399274047188, "step": 4240}, {"loss": 0.646, "grad_norm": 0.740113377571106, "learning_rate": 0.0002, "epoch": 3.0852994555353903, "step": 4250}, {"loss": 0.604, "grad_norm": 0.6616561412811279, "learning_rate": 0.0002, "epoch": 3.092558983666062, "step": 4260}, {"loss": 0.6462, "grad_norm": 0.8846506476402283, "learning_rate": 0.0002, "epoch": 3.0998185117967334, "step": 4270}, {"loss": 0.6037, "grad_norm": 0.6322125792503357, "learning_rate": 0.0002, "epoch": 3.107078039927405, "step": 4280}, {"loss": 0.5953, "grad_norm": 0.7461467385292053, "learning_rate": 0.0002, "epoch": 3.114337568058076, "step": 4290}, {"loss": 0.6099, "grad_norm": 0.8251287341117859, "learning_rate": 0.0002, "epoch": 3.1215970961887476, "step": 4300}, {"loss": 0.6284, "grad_norm": 0.8767673373222351, "learning_rate": 0.0002, "epoch": 3.128856624319419, "step": 4310}, {"loss": 0.7535, "grad_norm": 0.7758759260177612, "learning_rate": 0.0002, "epoch": 3.1361161524500907, "step": 4320}, {"loss": 0.6624, "grad_norm": 1.1056879758834839, "learning_rate": 0.0002, "epoch": 3.143375680580762, "step": 4330}, {"loss": 0.691, "grad_norm": 0.8259835243225098, "learning_rate": 0.0002, "epoch": 3.1506352087114338, "step": 4340}, {"loss": 0.6635, "grad_norm": 0.6607027053833008, "learning_rate": 0.0002, "epoch": 3.1578947368421053, "step": 4350}, {"loss": 0.5911, "grad_norm": 0.7983301281929016, "learning_rate": 0.0002, "epoch": 3.165154264972777, "step": 4360}, {"loss": 0.6496, "grad_norm": 0.6725239157676697, "learning_rate": 0.0002, "epoch": 3.1724137931034484, "step": 4370}, {"loss": 0.5966, "grad_norm": 0.9052095413208008, "learning_rate": 0.0002, "epoch": 3.17967332123412, "step": 4380}, {"loss": 0.6877, "grad_norm": 0.8131307363510132, "learning_rate": 0.0002, "epoch": 3.1869328493647915, "step": 4390}, {"loss": 0.6384, "grad_norm": 0.6435626149177551, "learning_rate": 0.0002, "epoch": 3.1941923774954626, "step": 4400}, {"loss": 0.5819, "grad_norm": 0.84367436170578, "learning_rate": 0.0002, "epoch": 3.201451905626134, "step": 4410}, {"loss": 0.6104, "grad_norm": 1.5018867254257202, "learning_rate": 0.0002, "epoch": 3.2087114337568057, "step": 4420}, {"loss": 0.6838, "grad_norm": 0.7019091844558716, "learning_rate": 0.0002, "epoch": 3.215970961887477, "step": 4430}, {"loss": 0.6153, "grad_norm": 0.9164197444915771, "learning_rate": 0.0002, "epoch": 3.2232304900181488, "step": 4440}, {"loss": 0.6618, "grad_norm": 0.7890861630439758, "learning_rate": 0.0002, "epoch": 3.2304900181488203, "step": 4450}, {"loss": 0.6401, "grad_norm": 0.6517660617828369, "learning_rate": 0.0002, "epoch": 3.237749546279492, "step": 4460}, {"loss": 0.6699, "grad_norm": 1.10188889503479, "learning_rate": 0.0002, "epoch": 3.2450090744101634, "step": 4470}, {"loss": 0.6356, "grad_norm": 0.8158330917358398, "learning_rate": 0.0002, "epoch": 3.252268602540835, "step": 4480}, {"loss": 0.7757, "grad_norm": 0.7663109302520752, "learning_rate": 0.0002, "epoch": 3.2595281306715065, "step": 4490}, {"loss": 0.6539, "grad_norm": 0.8473444581031799, "learning_rate": 0.0002, "epoch": 3.266787658802178, "step": 4500}, {"loss": 0.6511, "grad_norm": 0.9724768996238708, "learning_rate": 0.0002, "epoch": 3.274047186932849, "step": 4510}, {"loss": 0.5464, "grad_norm": 0.8516759276390076, "learning_rate": 0.0002, "epoch": 3.281306715063521, "step": 4520}, {"loss": 0.6534, "grad_norm": 0.7543437480926514, "learning_rate": 0.0002, "epoch": 3.288566243194192, "step": 4530}, {"loss": 0.6095, "grad_norm": 1.0472029447555542, "learning_rate": 0.0002, "epoch": 3.2958257713248638, "step": 4540}, {"loss": 0.6216, "grad_norm": 0.6240826845169067, "learning_rate": 0.0002, "epoch": 3.3030852994555353, "step": 4550}, {"loss": 0.6223, "grad_norm": 0.9957774877548218, "learning_rate": 0.0002, "epoch": 3.310344827586207, "step": 4560}, {"loss": 0.618, "grad_norm": 0.6448912620544434, "learning_rate": 0.0002, "epoch": 3.3176043557168784, "step": 4570}, {"loss": 0.6188, "grad_norm": 0.7519692778587341, "learning_rate": 0.0002, "epoch": 3.32486388384755, "step": 4580}, {"loss": 0.6672, "grad_norm": 0.7367453575134277, "learning_rate": 0.0002, "epoch": 3.3321234119782215, "step": 4590}, {"loss": 0.6517, "grad_norm": 0.8064960837364197, "learning_rate": 0.0002, "epoch": 3.339382940108893, "step": 4600}, {"loss": 0.6062, "grad_norm": 0.7664631009101868, "learning_rate": 0.0002, "epoch": 3.3466424682395646, "step": 4610}, {"loss": 0.6834, "grad_norm": 0.7803396582603455, "learning_rate": 0.0002, "epoch": 3.353901996370236, "step": 4620}, {"loss": 0.6961, "grad_norm": 0.9141599535942078, "learning_rate": 0.0002, "epoch": 3.3611615245009077, "step": 4630}, {"loss": 0.6889, "grad_norm": 0.9719856381416321, "learning_rate": 0.0002, "epoch": 3.3684210526315788, "step": 4640}, {"loss": 0.6914, "grad_norm": 0.9223218560218811, "learning_rate": 0.0002, "epoch": 3.3756805807622503, "step": 4650}, {"loss": 0.5981, "grad_norm": 0.7289277911186218, "learning_rate": 0.0002, "epoch": 3.382940108892922, "step": 4660}, {"loss": 0.595, "grad_norm": 1.039724349975586, "learning_rate": 0.0002, "epoch": 3.3901996370235934, "step": 4670}, {"loss": 0.8121, "grad_norm": 1.397438883781433, "learning_rate": 0.0002, "epoch": 3.397459165154265, "step": 4680}, {"loss": 0.6334, "grad_norm": 1.0069999694824219, "learning_rate": 0.0002, "epoch": 3.4047186932849365, "step": 4690}, {"loss": 0.6598, "grad_norm": 0.816291332244873, "learning_rate": 0.0002, "epoch": 3.411978221415608, "step": 4700}, {"loss": 0.6748, "grad_norm": 1.2831530570983887, "learning_rate": 0.0002, "epoch": 3.4192377495462796, "step": 4710}, {"loss": 0.6625, "grad_norm": 0.9573889970779419, "learning_rate": 0.0002, "epoch": 3.426497277676951, "step": 4720}, {"loss": 0.7279, "grad_norm": 0.7685632705688477, "learning_rate": 0.0002, "epoch": 3.4337568058076227, "step": 4730}, {"loss": 0.6104, "grad_norm": 0.7019195556640625, "learning_rate": 0.0002, "epoch": 3.441016333938294, "step": 4740}, {"loss": 0.7606, "grad_norm": 0.7244833707809448, "learning_rate": 0.0002, "epoch": 3.4482758620689653, "step": 4750}, {"loss": 0.6951, "grad_norm": 1.3468551635742188, "learning_rate": 0.0002, "epoch": 3.455535390199637, "step": 4760}, {"loss": 0.6945, "grad_norm": 0.822846531867981, "learning_rate": 0.0002, "epoch": 3.4627949183303084, "step": 4770}, {"loss": 0.6431, "grad_norm": 0.7311608195304871, "learning_rate": 0.0002, "epoch": 3.47005444646098, "step": 4780}, {"loss": 0.7019, "grad_norm": 0.9466770887374878, "learning_rate": 0.0002, "epoch": 3.4773139745916515, "step": 4790}, {"loss": 0.7767, "grad_norm": 1.1527155637741089, "learning_rate": 0.0002, "epoch": 3.484573502722323, "step": 4800}, {"loss": 0.6882, "grad_norm": 1.1288906335830688, "learning_rate": 0.0002, "epoch": 3.4918330308529946, "step": 4810}, {"loss": 0.6564, "grad_norm": 0.9096164107322693, "learning_rate": 0.0002, "epoch": 3.499092558983666, "step": 4820}, {"loss": 0.6127, "grad_norm": 0.7988565564155579, "learning_rate": 0.0002, "epoch": 3.5063520871143377, "step": 4830}, {"loss": 0.7004, "grad_norm": 0.7183415293693542, "learning_rate": 0.0002, "epoch": 3.513611615245009, "step": 4840}, {"loss": 0.74, "grad_norm": 0.6614915132522583, "learning_rate": 0.0002, "epoch": 3.5208711433756807, "step": 4850}, {"loss": 0.7271, "grad_norm": 0.8609521985054016, "learning_rate": 0.0002, "epoch": 3.528130671506352, "step": 4860}, {"loss": 0.6664, "grad_norm": 0.86552894115448, "learning_rate": 0.0002, "epoch": 3.535390199637024, "step": 4870}, {"loss": 0.6432, "grad_norm": 0.6926496028900146, "learning_rate": 0.0002, "epoch": 3.542649727767695, "step": 4880}, {"loss": 0.7117, "grad_norm": 0.8157467246055603, "learning_rate": 0.0002, "epoch": 3.5499092558983665, "step": 4890}, {"loss": 0.6201, "grad_norm": 0.9085357189178467, "learning_rate": 0.0002, "epoch": 3.557168784029038, "step": 4900}, {"loss": 0.6521, "grad_norm": 0.6322644948959351, "learning_rate": 0.0002, "epoch": 3.5644283121597096, "step": 4910}, {"loss": 0.6607, "grad_norm": 1.263205885887146, "learning_rate": 0.0002, "epoch": 3.571687840290381, "step": 4920}, {"loss": 0.6657, "grad_norm": 0.8901070356369019, "learning_rate": 0.0002, "epoch": 3.5789473684210527, "step": 4930}, {"loss": 0.6434, "grad_norm": 0.7983952164649963, "learning_rate": 0.0002, "epoch": 3.586206896551724, "step": 4940}, {"loss": 0.6861, "grad_norm": 0.9887813925743103, "learning_rate": 0.0002, "epoch": 3.5934664246823957, "step": 4950}, {"loss": 0.6502, "grad_norm": 0.7895187735557556, "learning_rate": 0.0002, "epoch": 3.6007259528130673, "step": 4960}, {"loss": 0.7111, "grad_norm": 0.9685819745063782, "learning_rate": 0.0002, "epoch": 3.6079854809437384, "step": 4970}, {"loss": 0.6915, "grad_norm": 0.6576591730117798, "learning_rate": 0.0002, "epoch": 3.6152450090744104, "step": 4980}, {"loss": 0.6195, "grad_norm": 0.856985330581665, "learning_rate": 0.0002, "epoch": 3.6225045372050815, "step": 4990}, {"loss": 0.6318, "grad_norm": 0.7230252623558044, "learning_rate": 0.0002, "epoch": 3.629764065335753, "step": 5000}, {"loss": 0.742, "grad_norm": 0.8260893821716309, "learning_rate": 0.0002, "epoch": 3.6370235934664246, "step": 5010}, {"loss": 0.7223, "grad_norm": 0.7635950446128845, "learning_rate": 0.0002, "epoch": 3.644283121597096, "step": 5020}, {"loss": 0.6837, "grad_norm": 0.7060768604278564, "learning_rate": 0.0002, "epoch": 3.6515426497277677, "step": 5030}, {"loss": 0.6921, "grad_norm": 0.8020303249359131, "learning_rate": 0.0002, "epoch": 3.658802177858439, "step": 5040}, {"loss": 0.6446, "grad_norm": 0.8530341386795044, "learning_rate": 0.0002, "epoch": 3.6660617059891107, "step": 5050}, {"loss": 0.7222, "grad_norm": 0.6667101979255676, "learning_rate": 0.0002, "epoch": 3.6733212341197823, "step": 5060}, {"loss": 0.7081, "grad_norm": 0.7385406494140625, "learning_rate": 0.0002, "epoch": 3.680580762250454, "step": 5070}, {"loss": 0.7035, "grad_norm": 0.7753380537033081, "learning_rate": 0.0002, "epoch": 3.6878402903811254, "step": 5080}, {"loss": 0.6491, "grad_norm": 0.7516207098960876, "learning_rate": 0.0002, "epoch": 3.695099818511797, "step": 5090}, {"loss": 0.672, "grad_norm": 0.8171586394309998, "learning_rate": 0.0002, "epoch": 3.702359346642468, "step": 5100}, {"loss": 0.7459, "grad_norm": 1.0796279907226562, "learning_rate": 0.0002, "epoch": 3.70961887477314, "step": 5110}, {"loss": 0.5948, "grad_norm": 0.6957688927650452, "learning_rate": 0.0002, "epoch": 3.716878402903811, "step": 5120}, {"loss": 0.7515, "grad_norm": 0.8550161719322205, "learning_rate": 0.0002, "epoch": 3.7241379310344827, "step": 5130}, {"loss": 0.7286, "grad_norm": 0.9396728277206421, "learning_rate": 0.0002, "epoch": 3.731397459165154, "step": 5140}, {"loss": 0.7594, "grad_norm": 1.4264805316925049, "learning_rate": 0.0002, "epoch": 3.7386569872958257, "step": 5150}, {"loss": 0.6575, "grad_norm": 0.8725108504295349, "learning_rate": 0.0002, "epoch": 3.7459165154264973, "step": 5160}, {"loss": 0.6778, "grad_norm": 1.0346195697784424, "learning_rate": 0.0002, "epoch": 3.753176043557169, "step": 5170}, {"loss": 0.6371, "grad_norm": 0.5395554304122925, "learning_rate": 0.0002, "epoch": 3.7604355716878404, "step": 5180}, {"loss": 0.7308, "grad_norm": 1.3153616189956665, "learning_rate": 0.0002, "epoch": 3.767695099818512, "step": 5190}, {"loss": 0.78, "grad_norm": 0.9879828691482544, "learning_rate": 0.0002, "epoch": 3.7749546279491835, "step": 5200}, {"loss": 0.7068, "grad_norm": 0.8876672983169556, "learning_rate": 0.0002, "epoch": 3.7822141560798546, "step": 5210}, {"loss": 0.6283, "grad_norm": 0.8363267779350281, "learning_rate": 0.0002, "epoch": 3.7894736842105265, "step": 5220}, {"loss": 0.6255, "grad_norm": 0.637294590473175, "learning_rate": 0.0002, "epoch": 3.7967332123411976, "step": 5230}, {"loss": 0.6685, "grad_norm": 1.1408970355987549, "learning_rate": 0.0002, "epoch": 3.803992740471869, "step": 5240}, {"loss": 0.6761, "grad_norm": 1.0128360986709595, "learning_rate": 0.0002, "epoch": 3.8112522686025407, "step": 5250}, {"loss": 0.6764, "grad_norm": 0.8061144351959229, "learning_rate": 0.0002, "epoch": 3.8185117967332123, "step": 5260}, {"loss": 0.7254, "grad_norm": 0.9626626968383789, "learning_rate": 0.0002, "epoch": 3.825771324863884, "step": 5270}, {"loss": 0.7367, "grad_norm": 0.9013627171516418, "learning_rate": 0.0002, "epoch": 3.8330308529945554, "step": 5280}, {"loss": 0.6806, "grad_norm": 0.8411344289779663, "learning_rate": 0.0002, "epoch": 3.840290381125227, "step": 5290}, {"loss": 0.6818, "grad_norm": 0.7426059246063232, "learning_rate": 0.0002, "epoch": 3.8475499092558985, "step": 5300}, {"loss": 0.6748, "grad_norm": 1.003413438796997, "learning_rate": 0.0002, "epoch": 3.85480943738657, "step": 5310}, {"loss": 0.8554, "grad_norm": 0.7527840733528137, "learning_rate": 0.0002, "epoch": 3.862068965517241, "step": 5320}, {"loss": 0.7521, "grad_norm": 0.738610565662384, "learning_rate": 0.0002, "epoch": 3.869328493647913, "step": 5330}, {"loss": 0.7266, "grad_norm": 0.7277999520301819, "learning_rate": 0.0002, "epoch": 3.876588021778584, "step": 5340}, {"loss": 0.7503, "grad_norm": 0.5951359272003174, "learning_rate": 0.0002, "epoch": 3.8838475499092557, "step": 5350}, {"loss": 0.7447, "grad_norm": 1.043884038925171, "learning_rate": 0.0002, "epoch": 3.8911070780399273, "step": 5360}, {"loss": 0.6862, "grad_norm": 0.8436498045921326, "learning_rate": 0.0002, "epoch": 3.898366606170599, "step": 5370}, {"loss": 0.665, "grad_norm": 0.5603365302085876, "learning_rate": 0.0002, "epoch": 3.9056261343012704, "step": 5380}, {"loss": 0.7098, "grad_norm": 1.0128886699676514, "learning_rate": 0.0002, "epoch": 3.912885662431942, "step": 5390}, {"loss": 0.6707, "grad_norm": 0.7970930337905884, "learning_rate": 0.0002, "epoch": 3.9201451905626135, "step": 5400}, {"loss": 0.637, "grad_norm": 0.7699369192123413, "learning_rate": 0.0002, "epoch": 3.927404718693285, "step": 5410}, {"loss": 0.6742, "grad_norm": 0.800561249256134, "learning_rate": 0.0002, "epoch": 3.9346642468239565, "step": 5420}, {"loss": 0.7208, "grad_norm": 0.8020331859588623, "learning_rate": 0.0002, "epoch": 3.941923774954628, "step": 5430}, {"loss": 0.7294, "grad_norm": 0.7461140155792236, "learning_rate": 0.0002, "epoch": 3.9491833030852996, "step": 5440}, {"loss": 0.7013, "grad_norm": 0.8346918821334839, "learning_rate": 0.0002, "epoch": 3.9564428312159707, "step": 5450}, {"loss": 0.6289, "grad_norm": 0.9723302125930786, "learning_rate": 0.0002, "epoch": 3.9637023593466427, "step": 5460}, {"loss": 0.8029, "grad_norm": 0.6809740662574768, "learning_rate": 0.0002, "epoch": 3.970961887477314, "step": 5470}, {"loss": 0.6896, "grad_norm": 0.7353498339653015, "learning_rate": 0.0002, "epoch": 3.9782214156079854, "step": 5480}, {"loss": 0.6722, "grad_norm": 0.748009443283081, "learning_rate": 0.0002, "epoch": 3.985480943738657, "step": 5490}, {"loss": 0.6866, "grad_norm": 1.3656195402145386, "learning_rate": 0.0002, "epoch": 3.9927404718693285, "step": 5500}, {"loss": 0.7368, "grad_norm": 0.8402108550071716, "learning_rate": 0.0002, "epoch": 4.0, "step": 5510}, {"eval_loss": 1.17229425907135, "eval_runtime": 46.2554, "eval_samples_per_second": 9.426, "eval_steps_per_second": 1.189, "epoch": 4.0, "step": 5510}, {"loss": 0.4637, "grad_norm": 0.8601235747337341, "learning_rate": 0.0002, "epoch": 4.007259528130671, "step": 5520}, {"loss": 0.4717, "grad_norm": 1.2635200023651123, "learning_rate": 0.0002, "epoch": 4.014519056261343, "step": 5530}, {"loss": 0.503, "grad_norm": 1.0257477760314941, "learning_rate": 0.0002, "epoch": 4.021778584392014, "step": 5540}, {"loss": 0.4547, "grad_norm": 0.9436745047569275, "learning_rate": 0.0002, "epoch": 4.029038112522686, "step": 5550}, {"loss": 0.459, "grad_norm": 0.9443606734275818, "learning_rate": 0.0002, "epoch": 4.036297640653357, "step": 5560}, {"loss": 0.5386, "grad_norm": 1.3965742588043213, "learning_rate": 0.0002, "epoch": 4.043557168784029, "step": 5570}, {"loss": 0.4248, "grad_norm": 0.8973520398139954, "learning_rate": 0.0002, "epoch": 4.0508166969147, "step": 5580}, {"loss": 0.4111, "grad_norm": 0.9998409748077393, "learning_rate": 0.0002, "epoch": 4.058076225045372, "step": 5590}, {"loss": 0.4828, "grad_norm": 1.1213387250900269, "learning_rate": 0.0002, "epoch": 4.0653357531760435, "step": 5600}, {"loss": 0.439, "grad_norm": 0.7064558863639832, "learning_rate": 0.0002, "epoch": 4.072595281306715, "step": 5610}, {"loss": 0.4607, "grad_norm": 1.2390803098678589, "learning_rate": 0.0002, "epoch": 4.0798548094373865, "step": 5620}, {"loss": 0.5014, "grad_norm": 1.123469591140747, "learning_rate": 0.0002, "epoch": 4.087114337568058, "step": 5630}, {"loss": 0.513, "grad_norm": 1.229573369026184, "learning_rate": 0.0002, "epoch": 4.09437386569873, "step": 5640}, {"loss": 0.5258, "grad_norm": 1.7182831764221191, "learning_rate": 0.0002, "epoch": 4.101633393829401, "step": 5650}, {"loss": 0.5371, "grad_norm": 0.894903302192688, "learning_rate": 0.0002, "epoch": 4.108892921960073, "step": 5660}, {"loss": 0.4813, "grad_norm": 0.8754552006721497, "learning_rate": 0.0002, "epoch": 4.116152450090744, "step": 5670}, {"loss": 0.491, "grad_norm": 1.2401553392410278, "learning_rate": 0.0002, "epoch": 4.123411978221416, "step": 5680}, {"loss": 0.4549, "grad_norm": 0.8631148934364319, "learning_rate": 0.0002, "epoch": 4.130671506352087, "step": 5690}, {"loss": 0.487, "grad_norm": 1.1798022985458374, "learning_rate": 0.0002, "epoch": 4.137931034482759, "step": 5700}, {"loss": 0.4522, "grad_norm": 0.8344549536705017, "learning_rate": 0.0002, "epoch": 4.14519056261343, "step": 5710}, {"loss": 0.4559, "grad_norm": 1.2342697381973267, "learning_rate": 0.0002, "epoch": 4.152450090744102, "step": 5720}, {"loss": 0.53, "grad_norm": 1.1601094007492065, "learning_rate": 0.0002, "epoch": 4.159709618874773, "step": 5730}, {"loss": 0.4755, "grad_norm": 1.2925703525543213, "learning_rate": 0.0002, "epoch": 4.166969147005445, "step": 5740}, {"loss": 0.4973, "grad_norm": 1.0870997905731201, "learning_rate": 0.0002, "epoch": 4.174228675136116, "step": 5750}, {"loss": 0.5184, "grad_norm": 0.9077792763710022, "learning_rate": 0.0002, "epoch": 4.181488203266787, "step": 5760}, {"loss": 0.4606, "grad_norm": 1.009273886680603, "learning_rate": 0.0002, "epoch": 4.188747731397459, "step": 5770}, {"loss": 0.5383, "grad_norm": 1.2465530633926392, "learning_rate": 0.0002, "epoch": 4.19600725952813, "step": 5780}, {"loss": 0.4938, "grad_norm": 1.2261253595352173, "learning_rate": 0.0002, "epoch": 4.203266787658802, "step": 5790}, {"loss": 0.5255, "grad_norm": 1.1498041152954102, "learning_rate": 0.0002, "epoch": 4.2105263157894735, "step": 5800}, {"loss": 0.5269, "grad_norm": 1.1966725587844849, "learning_rate": 0.0002, "epoch": 4.217785843920145, "step": 5810}, {"loss": 0.5626, "grad_norm": 1.2651296854019165, "learning_rate": 0.0002, "epoch": 4.2250453720508165, "step": 5820}, {"loss": 0.5213, "grad_norm": 1.0388574600219727, "learning_rate": 0.0002, "epoch": 4.2323049001814885, "step": 5830}, {"loss": 0.4965, "grad_norm": 1.3042771816253662, "learning_rate": 0.0002, "epoch": 4.23956442831216, "step": 5840}, {"loss": 0.5116, "grad_norm": 1.1127727031707764, "learning_rate": 0.0002, "epoch": 4.246823956442832, "step": 5850}, {"loss": 0.5197, "grad_norm": 0.9653958082199097, "learning_rate": 0.0002, "epoch": 4.254083484573503, "step": 5860}, {"loss": 0.4414, "grad_norm": 1.0500504970550537, "learning_rate": 0.0002, "epoch": 4.261343012704174, "step": 5870}, {"loss": 0.544, "grad_norm": 1.1476165056228638, "learning_rate": 0.0002, "epoch": 4.268602540834846, "step": 5880}, {"loss": 0.4667, "grad_norm": 0.9424414038658142, "learning_rate": 0.0002, "epoch": 4.275862068965517, "step": 5890}, {"loss": 0.5039, "grad_norm": 1.3309166431427002, "learning_rate": 0.0002, "epoch": 4.283121597096189, "step": 5900}, {"loss": 0.5472, "grad_norm": 1.3025873899459839, "learning_rate": 0.0002, "epoch": 4.29038112522686, "step": 5910}, {"loss": 0.4644, "grad_norm": 1.1442325115203857, "learning_rate": 0.0002, "epoch": 4.297640653357532, "step": 5920}, {"loss": 0.5066, "grad_norm": 0.9820859432220459, "learning_rate": 0.0002, "epoch": 4.304900181488203, "step": 5930}, {"loss": 0.5224, "grad_norm": 0.9615740180015564, "learning_rate": 0.0002, "epoch": 4.312159709618875, "step": 5940}, {"loss": 0.5665, "grad_norm": 1.1627109050750732, "learning_rate": 0.0002, "epoch": 4.319419237749546, "step": 5950}, {"loss": 0.4852, "grad_norm": 0.9381322860717773, "learning_rate": 0.0002, "epoch": 4.326678765880218, "step": 5960}, {"loss": 0.4532, "grad_norm": 0.8154335618019104, "learning_rate": 0.0002, "epoch": 4.333938294010889, "step": 5970}, {"loss": 0.5237, "grad_norm": 0.877671480178833, "learning_rate": 0.0002, "epoch": 4.341197822141561, "step": 5980}, {"loss": 0.6113, "grad_norm": 1.1742031574249268, "learning_rate": 0.0002, "epoch": 4.348457350272232, "step": 5990}, {"loss": 0.5704, "grad_norm": 1.0352917909622192, "learning_rate": 0.0002, "epoch": 4.3557168784029034, "step": 6000}, {"loss": 0.4996, "grad_norm": 0.9963878989219666, "learning_rate": 0.0002, "epoch": 4.362976406533575, "step": 6010}, {"loss": 0.4813, "grad_norm": 1.1892237663269043, "learning_rate": 0.0002, "epoch": 4.3702359346642465, "step": 6020}, {"loss": 0.5224, "grad_norm": 1.2516111135482788, "learning_rate": 0.0002, "epoch": 4.3774954627949185, "step": 6030}, {"loss": 0.5952, "grad_norm": 1.2111951112747192, "learning_rate": 0.0002, "epoch": 4.38475499092559, "step": 6040}, {"loss": 0.4275, "grad_norm": 1.0820083618164062, "learning_rate": 0.0002, "epoch": 4.392014519056262, "step": 6050}, {"loss": 0.5117, "grad_norm": 1.033915638923645, "learning_rate": 0.0002, "epoch": 4.399274047186933, "step": 6060}, {"loss": 0.5431, "grad_norm": 1.0635870695114136, "learning_rate": 0.0002, "epoch": 4.406533575317605, "step": 6070}, {"loss": 0.5341, "grad_norm": 1.0520414113998413, "learning_rate": 0.0002, "epoch": 4.413793103448276, "step": 6080}, {"loss": 0.512, "grad_norm": 1.0821926593780518, "learning_rate": 0.0002, "epoch": 4.421052631578947, "step": 6090}, {"loss": 0.5065, "grad_norm": 1.0533246994018555, "learning_rate": 0.0002, "epoch": 4.428312159709619, "step": 6100}, {"loss": 0.4577, "grad_norm": 0.9231932759284973, "learning_rate": 0.0002, "epoch": 4.43557168784029, "step": 6110}, {"loss": 0.583, "grad_norm": 0.9910260438919067, "learning_rate": 0.0002, "epoch": 4.442831215970962, "step": 6120}, {"loss": 0.4717, "grad_norm": 1.061949372291565, "learning_rate": 0.0002, "epoch": 4.450090744101633, "step": 6130}, {"loss": 0.5893, "grad_norm": 1.2927039861679077, "learning_rate": 0.0002, "epoch": 4.457350272232305, "step": 6140}, {"loss": 0.4684, "grad_norm": 1.3966081142425537, "learning_rate": 0.0002, "epoch": 4.464609800362976, "step": 6150}, {"loss": 0.5507, "grad_norm": 1.3835992813110352, "learning_rate": 0.0002, "epoch": 4.471869328493648, "step": 6160}, {"loss": 0.5911, "grad_norm": 1.0892692804336548, "learning_rate": 0.0002, "epoch": 4.479128856624319, "step": 6170}, {"loss": 0.478, "grad_norm": 1.0318800210952759, "learning_rate": 0.0002, "epoch": 4.486388384754991, "step": 6180}, {"loss": 0.5198, "grad_norm": 0.8174677491188049, "learning_rate": 0.0002, "epoch": 4.493647912885662, "step": 6190}, {"loss": 0.5387, "grad_norm": 1.4157509803771973, "learning_rate": 0.0002, "epoch": 4.500907441016334, "step": 6200}, {"loss": 0.5868, "grad_norm": 1.5244114398956299, "learning_rate": 0.0002, "epoch": 4.508166969147005, "step": 6210}, {"loss": 0.4642, "grad_norm": 0.8164850473403931, "learning_rate": 0.0002, "epoch": 4.5154264972776765, "step": 6220}, {"loss": 0.522, "grad_norm": 1.2904746532440186, "learning_rate": 0.0002, "epoch": 4.5226860254083485, "step": 6230}, {"loss": 0.5103, "grad_norm": 0.7987732887268066, "learning_rate": 0.0002, "epoch": 4.52994555353902, "step": 6240}, {"loss": 0.4615, "grad_norm": 0.831040620803833, "learning_rate": 0.0002, "epoch": 4.537205081669692, "step": 6250}, {"loss": 0.5065, "grad_norm": 0.9545485973358154, "learning_rate": 0.0002, "epoch": 4.544464609800363, "step": 6260}, {"loss": 0.5515, "grad_norm": 0.9291793704032898, "learning_rate": 0.0002, "epoch": 4.551724137931035, "step": 6270}, {"loss": 0.4535, "grad_norm": 0.8977208733558655, "learning_rate": 0.0002, "epoch": 4.558983666061706, "step": 6280}, {"loss": 0.544, "grad_norm": 1.1768537759780884, "learning_rate": 0.0002, "epoch": 4.566243194192378, "step": 6290}, {"loss": 0.5925, "grad_norm": 1.0688952207565308, "learning_rate": 0.0002, "epoch": 4.573502722323049, "step": 6300}, {"loss": 0.5207, "grad_norm": 0.8800966739654541, "learning_rate": 0.0002, "epoch": 4.580762250453721, "step": 6310}, {"loss": 0.6106, "grad_norm": 1.0911834239959717, "learning_rate": 0.0002, "epoch": 4.588021778584392, "step": 6320}, {"loss": 0.5109, "grad_norm": 1.1420872211456299, "learning_rate": 0.0002, "epoch": 4.595281306715064, "step": 6330}, {"loss": 0.5147, "grad_norm": 1.0215224027633667, "learning_rate": 0.0002, "epoch": 4.602540834845735, "step": 6340}, {"loss": 0.592, "grad_norm": 0.9685489535331726, "learning_rate": 0.0002, "epoch": 4.609800362976406, "step": 6350}, {"loss": 0.5775, "grad_norm": 1.12773597240448, "learning_rate": 0.0002, "epoch": 4.617059891107078, "step": 6360}, {"loss": 0.5966, "grad_norm": 1.0663973093032837, "learning_rate": 0.0002, "epoch": 4.624319419237749, "step": 6370}, {"loss": 0.512, "grad_norm": 1.1707262992858887, "learning_rate": 0.0002, "epoch": 4.631578947368421, "step": 6380}, {"loss": 0.5497, "grad_norm": 1.0672980546951294, "learning_rate": 0.0002, "epoch": 4.638838475499092, "step": 6390}, {"loss": 0.5699, "grad_norm": 1.1464333534240723, "learning_rate": 0.0002, "epoch": 4.646098003629764, "step": 6400}, {"loss": 0.5514, "grad_norm": 1.070230484008789, "learning_rate": 0.0002, "epoch": 4.653357531760435, "step": 6410}, {"loss": 0.5013, "grad_norm": 0.9673764109611511, "learning_rate": 0.0002, "epoch": 4.660617059891107, "step": 6420}, {"loss": 0.5901, "grad_norm": 1.0189043283462524, "learning_rate": 0.0002, "epoch": 4.6678765880217785, "step": 6430}, {"loss": 0.5193, "grad_norm": 1.185896396636963, "learning_rate": 0.0002, "epoch": 4.67513611615245, "step": 6440}, {"loss": 0.5318, "grad_norm": 1.0682812929153442, "learning_rate": 0.0002, "epoch": 4.682395644283122, "step": 6450}, {"loss": 0.5773, "grad_norm": 1.3586071729660034, "learning_rate": 0.0002, "epoch": 4.689655172413794, "step": 6460}, {"loss": 0.5482, "grad_norm": 0.6561792492866516, "learning_rate": 0.0002, "epoch": 4.696914700544465, "step": 6470}, {"loss": 0.5711, "grad_norm": 1.1394113302230835, "learning_rate": 0.0002, "epoch": 4.704174228675136, "step": 6480}, {"loss": 0.5325, "grad_norm": 0.9683151245117188, "learning_rate": 0.0002, "epoch": 4.711433756805808, "step": 6490}, {"loss": 0.5139, "grad_norm": 1.0247553586959839, "learning_rate": 0.0002, "epoch": 4.718693284936479, "step": 6500}, {"loss": 0.5794, "grad_norm": 0.8046169281005859, "learning_rate": 0.0002, "epoch": 4.725952813067151, "step": 6510}, {"loss": 0.5471, "grad_norm": 1.0710240602493286, "learning_rate": 0.0002, "epoch": 4.733212341197822, "step": 6520}, {"loss": 0.4805, "grad_norm": 0.9438924193382263, "learning_rate": 0.0002, "epoch": 4.740471869328494, "step": 6530}, {"loss": 0.5404, "grad_norm": 0.869162380695343, "learning_rate": 0.0002, "epoch": 4.747731397459165, "step": 6540}, {"loss": 0.6379, "grad_norm": 0.9776787161827087, "learning_rate": 0.0002, "epoch": 4.754990925589837, "step": 6550}, {"loss": 0.5288, "grad_norm": 1.1990505456924438, "learning_rate": 0.0002, "epoch": 4.762250453720508, "step": 6560}, {"loss": 0.5539, "grad_norm": 1.0582209825515747, "learning_rate": 0.0002, "epoch": 4.769509981851179, "step": 6570}, {"loss": 0.489, "grad_norm": 0.9966367483139038, "learning_rate": 0.0002, "epoch": 4.776769509981851, "step": 6580}, {"loss": 0.5514, "grad_norm": 0.9130612015724182, "learning_rate": 0.0002, "epoch": 4.784029038112522, "step": 6590}, {"loss": 0.5864, "grad_norm": 1.0950500965118408, "learning_rate": 0.0002, "epoch": 4.791288566243194, "step": 6600}, {"loss": 0.5266, "grad_norm": 1.108681321144104, "learning_rate": 0.0002, "epoch": 4.798548094373865, "step": 6610}, {"loss": 0.5875, "grad_norm": 1.1873763799667358, "learning_rate": 0.0002, "epoch": 4.805807622504537, "step": 6620}, {"loss": 0.5736, "grad_norm": 1.305367112159729, "learning_rate": 0.0002, "epoch": 4.8130671506352085, "step": 6630}, {"loss": 0.5636, "grad_norm": 1.2801482677459717, "learning_rate": 0.0002, "epoch": 4.8203266787658805, "step": 6640}, {"loss": 0.582, "grad_norm": 1.26764976978302, "learning_rate": 0.0002, "epoch": 4.827586206896552, "step": 6650}, {"loss": 0.5259, "grad_norm": 1.0018208026885986, "learning_rate": 0.0002, "epoch": 4.834845735027224, "step": 6660}, {"loss": 0.548, "grad_norm": 1.2326326370239258, "learning_rate": 0.0002, "epoch": 4.842105263157895, "step": 6670}, {"loss": 0.5933, "grad_norm": 0.9707282781600952, "learning_rate": 0.0002, "epoch": 4.849364791288567, "step": 6680}, {"loss": 0.5612, "grad_norm": 1.2772048711776733, "learning_rate": 0.0002, "epoch": 4.856624319419238, "step": 6690}, {"loss": 0.5346, "grad_norm": 2.6652262210845947, "learning_rate": 0.0002, "epoch": 4.863883847549909, "step": 6700}, {"loss": 0.5428, "grad_norm": 1.215828537940979, "learning_rate": 0.0002, "epoch": 4.871143375680581, "step": 6710}, {"loss": 0.6571, "grad_norm": 1.3704510927200317, "learning_rate": 0.0002, "epoch": 4.878402903811252, "step": 6720}, {"loss": 0.4963, "grad_norm": 0.7781757116317749, "learning_rate": 0.0002, "epoch": 4.885662431941924, "step": 6730}, {"loss": 0.5989, "grad_norm": 1.1883646249771118, "learning_rate": 0.0002, "epoch": 4.892921960072595, "step": 6740}, {"loss": 0.6067, "grad_norm": 0.9216066002845764, "learning_rate": 0.0002, "epoch": 4.900181488203267, "step": 6750}, {"loss": 0.5085, "grad_norm": 1.0558464527130127, "learning_rate": 0.0002, "epoch": 4.907441016333938, "step": 6760}, {"loss": 0.5216, "grad_norm": 1.032656192779541, "learning_rate": 0.0002, "epoch": 4.91470054446461, "step": 6770}, {"loss": 0.5426, "grad_norm": 1.1261441707611084, "learning_rate": 0.0002, "epoch": 4.921960072595281, "step": 6780}, {"loss": 0.5295, "grad_norm": 1.2178640365600586, "learning_rate": 0.0002, "epoch": 4.929219600725952, "step": 6790}, {"loss": 0.5476, "grad_norm": 1.5369361639022827, "learning_rate": 0.0002, "epoch": 4.936479128856624, "step": 6800}, {"loss": 0.5358, "grad_norm": 1.1188377141952515, "learning_rate": 0.0002, "epoch": 4.943738656987296, "step": 6810}, {"loss": 0.5483, "grad_norm": 1.2506113052368164, "learning_rate": 0.0002, "epoch": 4.950998185117967, "step": 6820}, {"loss": 0.567, "grad_norm": 0.8776047825813293, "learning_rate": 0.0002, "epoch": 4.9582577132486385, "step": 6830}, {"loss": 0.5764, "grad_norm": 0.9700555205345154, "learning_rate": 0.0002, "epoch": 4.9655172413793105, "step": 6840}, {"loss": 0.5396, "grad_norm": 1.2713534832000732, "learning_rate": 0.0002, "epoch": 4.972776769509982, "step": 6850}, {"loss": 0.5451, "grad_norm": 0.9855955243110657, "learning_rate": 0.0002, "epoch": 4.980036297640654, "step": 6860}, {"loss": 0.5884, "grad_norm": 0.8734853863716125, "learning_rate": 0.0002, "epoch": 4.987295825771325, "step": 6870}, {"loss": 0.5189, "grad_norm": 0.8065403699874878, "learning_rate": 0.0002, "epoch": 4.994555353901997, "step": 6880}, {"eval_loss": 1.3302682638168335, "eval_runtime": 46.2496, "eval_samples_per_second": 9.427, "eval_steps_per_second": 1.189, "epoch": 4.999637023593467, "step": 6887}, {"loss": 0.4889, "grad_norm": 0.5163813829421997, "learning_rate": 0.0002, "epoch": 5.001814882032668, "step": 6890}, {"loss": 0.3545, "grad_norm": 1.1496137380599976, "learning_rate": 0.0002, "epoch": 5.00907441016334, "step": 6900}, {"loss": 0.39, "grad_norm": 1.0133885145187378, "learning_rate": 0.0002, "epoch": 5.016333938294011, "step": 6910}, {"loss": 0.3693, "grad_norm": 0.9479621052742004, "learning_rate": 0.0002, "epoch": 5.023593466424682, "step": 6920}, {"loss": 0.4012, "grad_norm": 0.8587583303451538, "learning_rate": 0.0002, "epoch": 5.030852994555354, "step": 6930}, {"loss": 0.3428, "grad_norm": 1.3314697742462158, "learning_rate": 0.0002, "epoch": 5.038112522686025, "step": 6940}, {"loss": 0.3909, "grad_norm": 1.195448875427246, "learning_rate": 0.0002, "epoch": 5.045372050816697, "step": 6950}, {"loss": 0.3322, "grad_norm": 1.2482256889343262, "learning_rate": 0.0002, "epoch": 5.052631578947368, "step": 6960}, {"loss": 0.3893, "grad_norm": 1.2011528015136719, "learning_rate": 0.0002, "epoch": 5.05989110707804, "step": 6970}, {"loss": 0.3265, "grad_norm": 1.3997188806533813, "learning_rate": 0.0002, "epoch": 5.067150635208711, "step": 6980}, {"loss": 0.3716, "grad_norm": 1.2147513628005981, "learning_rate": 0.0002, "epoch": 5.074410163339383, "step": 6990}, {"loss": 0.4053, "grad_norm": 1.6030137538909912, "learning_rate": 0.0002, "epoch": 5.081669691470054, "step": 7000}, {"loss": 0.3665, "grad_norm": 0.9466970562934875, "learning_rate": 0.0002, "epoch": 5.088929219600726, "step": 7010}, {"loss": 0.3451, "grad_norm": 1.4593411684036255, "learning_rate": 0.0002, "epoch": 5.096188747731397, "step": 7020}, {"loss": 0.3843, "grad_norm": 1.2196033000946045, "learning_rate": 0.0002, "epoch": 5.103448275862069, "step": 7030}, {"loss": 0.3896, "grad_norm": 1.1341328620910645, "learning_rate": 0.0002, "epoch": 5.1107078039927405, "step": 7040}, {"loss": 0.3627, "grad_norm": 1.2248976230621338, "learning_rate": 0.0002, "epoch": 5.117967332123412, "step": 7050}, {"loss": 0.3784, "grad_norm": 1.1620593070983887, "learning_rate": 0.0002, "epoch": 5.125226860254084, "step": 7060}, {"loss": 0.3678, "grad_norm": 0.9300723671913147, "learning_rate": 0.0002, "epoch": 5.132486388384755, "step": 7070}, {"loss": 0.3756, "grad_norm": 1.2265169620513916, "learning_rate": 0.0002, "epoch": 5.139745916515427, "step": 7080}, {"loss": 0.3595, "grad_norm": 1.4430373907089233, "learning_rate": 0.0002, "epoch": 5.147005444646098, "step": 7090}, {"loss": 0.3788, "grad_norm": 1.0821576118469238, "learning_rate": 0.0002, "epoch": 5.15426497277677, "step": 7100}, {"loss": 0.383, "grad_norm": 1.2574739456176758, "learning_rate": 0.0002, "epoch": 5.161524500907441, "step": 7110}, {"loss": 0.3692, "grad_norm": 1.1806069612503052, "learning_rate": 0.0002, "epoch": 5.168784029038113, "step": 7120}, {"loss": 0.3978, "grad_norm": 0.9900956153869629, "learning_rate": 0.0002, "epoch": 5.176043557168784, "step": 7130}, {"loss": 0.4358, "grad_norm": 1.2414425611495972, "learning_rate": 0.0002, "epoch": 5.183303085299456, "step": 7140}, {"loss": 0.3485, "grad_norm": 0.8220699429512024, "learning_rate": 0.0002, "epoch": 5.190562613430127, "step": 7150}, {"loss": 0.3517, "grad_norm": 1.29408860206604, "learning_rate": 0.0002, "epoch": 5.197822141560798, "step": 7160}, {"loss": 0.3405, "grad_norm": 0.8510639071464539, "learning_rate": 0.0002, "epoch": 5.20508166969147, "step": 7170}, {"loss": 0.4233, "grad_norm": 1.3448902368545532, "learning_rate": 0.0002, "epoch": 5.212341197822141, "step": 7180}, {"loss": 0.3808, "grad_norm": 1.054451584815979, "learning_rate": 0.0002, "epoch": 5.219600725952813, "step": 7190}, {"loss": 0.368, "grad_norm": 1.3752713203430176, "learning_rate": 0.0002, "epoch": 5.226860254083484, "step": 7200}, {"loss": 0.3844, "grad_norm": 1.4848095178604126, "learning_rate": 0.0002, "epoch": 5.234119782214156, "step": 7210}, {"loss": 0.4187, "grad_norm": 1.428842544555664, "learning_rate": 0.0002, "epoch": 5.241379310344827, "step": 7220}, {"loss": 0.3778, "grad_norm": 1.1703591346740723, "learning_rate": 0.0002, "epoch": 5.248638838475499, "step": 7230}, {"loss": 0.417, "grad_norm": 1.2180451154708862, "learning_rate": 0.0002, "epoch": 5.2558983666061705, "step": 7240}, {"loss": 0.3656, "grad_norm": 1.094045877456665, "learning_rate": 0.0002, "epoch": 5.2631578947368425, "step": 7250}, {"loss": 0.4331, "grad_norm": 0.9545766115188599, "learning_rate": 0.0002, "epoch": 5.270417422867514, "step": 7260}, {"loss": 0.3642, "grad_norm": 0.8356652855873108, "learning_rate": 0.0002, "epoch": 5.277676950998185, "step": 7270}, {"loss": 0.3576, "grad_norm": 1.148160457611084, "learning_rate": 0.0002, "epoch": 5.284936479128857, "step": 7280}, {"loss": 0.4178, "grad_norm": 1.2009977102279663, "learning_rate": 0.0002, "epoch": 5.292196007259528, "step": 7290}, {"loss": 0.3977, "grad_norm": 1.3283873796463013, "learning_rate": 0.0002, "epoch": 5.2994555353902, "step": 7300}, {"loss": 0.3853, "grad_norm": 0.9850481748580933, "learning_rate": 0.0002, "epoch": 5.306715063520871, "step": 7310}, {"loss": 0.3645, "grad_norm": 1.367550015449524, "learning_rate": 0.0002, "epoch": 5.313974591651543, "step": 7320}, {"loss": 0.3898, "grad_norm": 0.8602936863899231, "learning_rate": 0.0002, "epoch": 5.321234119782214, "step": 7330}, {"loss": 0.4173, "grad_norm": 1.1130679845809937, "learning_rate": 0.0002, "epoch": 5.328493647912886, "step": 7340}, {"loss": 0.3642, "grad_norm": 1.3002253770828247, "learning_rate": 0.0002, "epoch": 5.335753176043557, "step": 7350}, {"loss": 0.4138, "grad_norm": 1.6235289573669434, "learning_rate": 0.0002, "epoch": 5.343012704174229, "step": 7360}, {"loss": 0.4779, "grad_norm": 1.156379222869873, "learning_rate": 0.0002, "epoch": 5.3502722323049, "step": 7370}, {"loss": 0.3222, "grad_norm": 1.0569308996200562, "learning_rate": 0.0002, "epoch": 5.357531760435572, "step": 7380}, {"loss": 0.3573, "grad_norm": 1.6674021482467651, "learning_rate": 0.0002, "epoch": 5.364791288566243, "step": 7390}, {"loss": 0.4325, "grad_norm": 1.2962018251419067, "learning_rate": 0.0002, "epoch": 5.372050816696914, "step": 7400}, {"loss": 0.3809, "grad_norm": 1.1904195547103882, "learning_rate": 0.0002, "epoch": 5.379310344827586, "step": 7410}, {"loss": 0.3728, "grad_norm": 1.316245675086975, "learning_rate": 0.0002, "epoch": 5.386569872958257, "step": 7420}, {"loss": 0.4096, "grad_norm": 1.127570390701294, "learning_rate": 0.0002, "epoch": 5.393829401088929, "step": 7430}, {"loss": 0.3933, "grad_norm": 1.3895777463912964, "learning_rate": 0.0002, "epoch": 5.4010889292196005, "step": 7440}, {"loss": 0.4085, "grad_norm": 1.626830816268921, "learning_rate": 0.0002, "epoch": 5.4083484573502725, "step": 7450}, {"loss": 0.4186, "grad_norm": 1.3703926801681519, "learning_rate": 0.0002, "epoch": 5.415607985480944, "step": 7460}, {"loss": 0.3517, "grad_norm": 1.3854840993881226, "learning_rate": 0.0002, "epoch": 5.422867513611616, "step": 7470}, {"loss": 0.3714, "grad_norm": 1.107065200805664, "learning_rate": 0.0002, "epoch": 5.430127041742287, "step": 7480}, {"loss": 0.3855, "grad_norm": 0.7843456268310547, "learning_rate": 0.0002, "epoch": 5.437386569872959, "step": 7490}, {"loss": 0.4159, "grad_norm": 1.6692372560501099, "learning_rate": 0.0002, "epoch": 5.44464609800363, "step": 7500}, {"loss": 0.4185, "grad_norm": 1.2583858966827393, "learning_rate": 0.0002, "epoch": 5.451905626134302, "step": 7510}, {"loss": 0.4401, "grad_norm": 1.6827000379562378, "learning_rate": 0.0002, "epoch": 5.459165154264973, "step": 7520}, {"loss": 0.397, "grad_norm": 1.6680560111999512, "learning_rate": 0.0002, "epoch": 5.466424682395644, "step": 7530}, {"loss": 0.4193, "grad_norm": 1.3696072101593018, "learning_rate": 0.0002, "epoch": 5.473684210526316, "step": 7540}, {"loss": 0.4244, "grad_norm": 1.4523496627807617, "learning_rate": 0.0002, "epoch": 5.480943738656987, "step": 7550}, {"loss": 0.3609, "grad_norm": 1.3432692289352417, "learning_rate": 0.0002, "epoch": 5.488203266787659, "step": 7560}, {"loss": 0.3675, "grad_norm": 1.363818645477295, "learning_rate": 0.0002, "epoch": 5.49546279491833, "step": 7570}, {"loss": 0.3726, "grad_norm": 1.0176721811294556, "learning_rate": 0.0002, "epoch": 5.502722323049002, "step": 7580}, {"loss": 0.3751, "grad_norm": 1.1625547409057617, "learning_rate": 0.0002, "epoch": 5.509981851179673, "step": 7590}, {"loss": 0.433, "grad_norm": 1.2480388879776, "learning_rate": 0.0002, "epoch": 5.517241379310345, "step": 7600}, {"loss": 0.4511, "grad_norm": 1.341509222984314, "learning_rate": 0.0002, "epoch": 5.524500907441016, "step": 7610}, {"loss": 0.4642, "grad_norm": 1.7048436403274536, "learning_rate": 0.0002, "epoch": 5.531760435571687, "step": 7620}, {"loss": 0.4509, "grad_norm": 1.1435480117797852, "learning_rate": 0.0002, "epoch": 5.539019963702359, "step": 7630}, {"loss": 0.4528, "grad_norm": 1.2381842136383057, "learning_rate": 0.0002, "epoch": 5.5462794918330305, "step": 7640}, {"loss": 0.4496, "grad_norm": 1.50786292552948, "learning_rate": 0.0002, "epoch": 5.5535390199637025, "step": 7650}, {"loss": 0.4242, "grad_norm": 1.2263519763946533, "learning_rate": 0.0002, "epoch": 5.560798548094374, "step": 7660}, {"loss": 0.418, "grad_norm": 1.2864696979522705, "learning_rate": 0.0002, "epoch": 5.568058076225046, "step": 7670}, {"loss": 0.3832, "grad_norm": 1.4443191289901733, "learning_rate": 0.0002, "epoch": 5.575317604355717, "step": 7680}, {"loss": 0.3964, "grad_norm": 1.3360971212387085, "learning_rate": 0.0002, "epoch": 5.582577132486389, "step": 7690}, {"loss": 0.4639, "grad_norm": 1.391828179359436, "learning_rate": 0.0002, "epoch": 5.58983666061706, "step": 7700}, {"loss": 0.4722, "grad_norm": 1.3699384927749634, "learning_rate": 0.0002, "epoch": 5.597096188747732, "step": 7710}, {"loss": 0.4302, "grad_norm": 1.3778468370437622, "learning_rate": 0.0002, "epoch": 5.604355716878403, "step": 7720}, {"loss": 0.4179, "grad_norm": 1.1009501218795776, "learning_rate": 0.0002, "epoch": 5.611615245009075, "step": 7730}, {"loss": 0.4104, "grad_norm": 1.0410021543502808, "learning_rate": 0.0002, "epoch": 5.618874773139746, "step": 7740}, {"loss": 0.4489, "grad_norm": 1.1012226343154907, "learning_rate": 0.0002, "epoch": 5.626134301270417, "step": 7750}, {"loss": 0.4544, "grad_norm": 1.3246384859085083, "learning_rate": 0.0002, "epoch": 5.633393829401089, "step": 7760}, {"loss": 0.4381, "grad_norm": 1.4301716089248657, "learning_rate": 0.0002, "epoch": 5.64065335753176, "step": 7770}, {"loss": 0.4297, "grad_norm": 1.1368978023529053, "learning_rate": 0.0002, "epoch": 5.647912885662432, "step": 7780}, {"loss": 0.4063, "grad_norm": 1.3493064641952515, "learning_rate": 0.0002, "epoch": 5.655172413793103, "step": 7790}, {"loss": 0.4562, "grad_norm": 1.3328721523284912, "learning_rate": 0.0002, "epoch": 5.662431941923775, "step": 7800}, {"loss": 0.4075, "grad_norm": 1.3235671520233154, "learning_rate": 0.0002, "epoch": 5.669691470054446, "step": 7810}, {"loss": 0.4589, "grad_norm": 1.1961841583251953, "learning_rate": 0.0002, "epoch": 5.676950998185118, "step": 7820}, {"loss": 0.4503, "grad_norm": 1.4189636707305908, "learning_rate": 0.0002, "epoch": 5.684210526315789, "step": 7830}, {"loss": 0.4452, "grad_norm": 1.3551312685012817, "learning_rate": 0.0002, "epoch": 5.691470054446461, "step": 7840}, {"loss": 0.4268, "grad_norm": 1.449987769126892, "learning_rate": 0.0002, "epoch": 5.6987295825771325, "step": 7850}, {"loss": 0.4141, "grad_norm": 1.1225156784057617, "learning_rate": 0.0002, "epoch": 5.7059891107078045, "step": 7860}, {"loss": 0.41, "grad_norm": 1.4734594821929932, "learning_rate": 0.0002, "epoch": 5.713248638838476, "step": 7870}, {"loss": 0.4013, "grad_norm": 1.3793359994888306, "learning_rate": 0.0002, "epoch": 5.720508166969147, "step": 7880}, {"loss": 0.4065, "grad_norm": 1.2431834936141968, "learning_rate": 0.0002, "epoch": 5.727767695099819, "step": 7890}, {"loss": 0.4595, "grad_norm": 1.1158313751220703, "learning_rate": 0.0002, "epoch": 5.73502722323049, "step": 7900}, {"loss": 0.4342, "grad_norm": 1.212248682975769, "learning_rate": 0.0002, "epoch": 5.742286751361162, "step": 7910}, {"loss": 0.4611, "grad_norm": 1.5259995460510254, "learning_rate": 0.0002, "epoch": 5.749546279491833, "step": 7920}, {"loss": 0.4483, "grad_norm": 1.3909121751785278, "learning_rate": 0.0002, "epoch": 5.756805807622505, "step": 7930}, {"loss": 0.4325, "grad_norm": 1.2511249780654907, "learning_rate": 0.0002, "epoch": 5.764065335753176, "step": 7940}, {"loss": 0.4048, "grad_norm": 1.2511906623840332, "learning_rate": 0.0002, "epoch": 5.771324863883848, "step": 7950}, {"loss": 0.3715, "grad_norm": 1.1489921808242798, "learning_rate": 0.0002, "epoch": 5.778584392014519, "step": 7960}, {"loss": 0.4196, "grad_norm": 1.028943419456482, "learning_rate": 0.0002, "epoch": 5.78584392014519, "step": 7970}, {"loss": 0.4334, "grad_norm": 1.0820423364639282, "learning_rate": 0.0002, "epoch": 5.793103448275862, "step": 7980}, {"loss": 0.3917, "grad_norm": 1.296520471572876, "learning_rate": 0.0002, "epoch": 5.800362976406533, "step": 7990}, {"loss": 0.4509, "grad_norm": 1.3597749471664429, "learning_rate": 0.0002, "epoch": 5.807622504537205, "step": 8000}, {"loss": 0.4535, "grad_norm": 0.8741790652275085, "learning_rate": 0.0002, "epoch": 5.814882032667876, "step": 8010}, {"loss": 0.4239, "grad_norm": 1.1471822261810303, "learning_rate": 0.0002, "epoch": 5.822141560798548, "step": 8020}, {"loss": 0.5042, "grad_norm": 1.2997334003448486, "learning_rate": 0.0002, "epoch": 5.829401088929219, "step": 8030}, {"loss": 0.4758, "grad_norm": 1.1027175188064575, "learning_rate": 0.0002, "epoch": 5.836660617059891, "step": 8040}, {"loss": 0.4192, "grad_norm": 1.2695307731628418, "learning_rate": 0.0002, "epoch": 5.8439201451905625, "step": 8050}, {"loss": 0.5173, "grad_norm": 1.5275461673736572, "learning_rate": 0.0002, "epoch": 5.8511796733212345, "step": 8060}, {"loss": 0.5012, "grad_norm": 1.3059501647949219, "learning_rate": 0.0002, "epoch": 5.8584392014519056, "step": 8070}, {"loss": 0.4425, "grad_norm": 1.57442045211792, "learning_rate": 0.0002, "epoch": 5.8656987295825775, "step": 8080}, {"loss": 0.4261, "grad_norm": 1.119564414024353, "learning_rate": 0.0002, "epoch": 5.872958257713249, "step": 8090}, {"loss": 0.465, "grad_norm": 1.6517373323440552, "learning_rate": 0.0002, "epoch": 5.88021778584392, "step": 8100}, {"loss": 0.4406, "grad_norm": 1.4093554019927979, "learning_rate": 0.0002, "epoch": 5.887477313974592, "step": 8110}, {"loss": 0.4433, "grad_norm": 1.278843641281128, "learning_rate": 0.0002, "epoch": 5.894736842105263, "step": 8120}, {"loss": 0.4007, "grad_norm": 1.2042944431304932, "learning_rate": 0.0002, "epoch": 5.901996370235935, "step": 8130}, {"loss": 0.3972, "grad_norm": 1.1788326501846313, "learning_rate": 0.0002, "epoch": 5.909255898366606, "step": 8140}, {"loss": 0.4506, "grad_norm": 1.4364569187164307, "learning_rate": 0.0002, "epoch": 5.916515426497278, "step": 8150}, {"loss": 0.4651, "grad_norm": 1.1704283952713013, "learning_rate": 0.0002, "epoch": 5.923774954627949, "step": 8160}, {"loss": 0.3972, "grad_norm": 1.040814995765686, "learning_rate": 0.0002, "epoch": 5.931034482758621, "step": 8170}, {"loss": 0.4038, "grad_norm": 1.1367416381835938, "learning_rate": 0.0002, "epoch": 5.938294010889292, "step": 8180}, {"loss": 0.4387, "grad_norm": 1.3401511907577515, "learning_rate": 0.0002, "epoch": 5.945553539019964, "step": 8190}, {"loss": 0.4396, "grad_norm": 1.1154041290283203, "learning_rate": 0.0002, "epoch": 5.952813067150635, "step": 8200}, {"loss": 0.4744, "grad_norm": 1.426089882850647, "learning_rate": 0.0002, "epoch": 5.960072595281307, "step": 8210}, {"loss": 0.4105, "grad_norm": 1.3170222043991089, "learning_rate": 0.0002, "epoch": 5.967332123411978, "step": 8220}, {"loss": 0.4137, "grad_norm": 1.1960029602050781, "learning_rate": 0.0002, "epoch": 5.974591651542649, "step": 8230}, {"loss": 0.423, "grad_norm": 1.0843931436538696, "learning_rate": 0.0002, "epoch": 5.981851179673321, "step": 8240}, {"loss": 0.459, "grad_norm": 1.050421118736267, "learning_rate": 0.0002, "epoch": 5.9891107078039925, "step": 8250}, {"loss": 0.3993, "grad_norm": 1.0183138847351074, "learning_rate": 0.0002, "epoch": 5.9963702359346644, "step": 8260}, {"eval_loss": 1.4677470922470093, "eval_runtime": 46.2504, "eval_samples_per_second": 9.427, "eval_steps_per_second": 1.189, "epoch": 6.0, "step": 8265}, {"loss": 0.3947, "grad_norm": 1.1702998876571655, "learning_rate": 0.0002, "epoch": 6.0036297640653356, "step": 8270}, {"loss": 0.2854, "grad_norm": 1.5389727354049683, "learning_rate": 0.0002, "epoch": 6.0108892921960075, "step": 8280}, {"loss": 0.2603, "grad_norm": 1.502568244934082, "learning_rate": 0.0002, "epoch": 6.018148820326679, "step": 8290}, {"loss": 0.3329, "grad_norm": 1.3846043348312378, "learning_rate": 0.0002, "epoch": 6.025408348457351, "step": 8300}, {"loss": 0.2651, "grad_norm": 1.173553228378296, "learning_rate": 0.0002, "epoch": 6.032667876588022, "step": 8310}, {"loss": 0.3142, "grad_norm": 1.5325932502746582, "learning_rate": 0.0002, "epoch": 6.039927404718694, "step": 8320}, {"loss": 0.2511, "grad_norm": 1.303783655166626, "learning_rate": 0.0002, "epoch": 6.047186932849365, "step": 8330}, {"loss": 0.2352, "grad_norm": 0.9408994913101196, "learning_rate": 0.0002, "epoch": 6.054446460980036, "step": 8340}, {"loss": 0.2548, "grad_norm": 1.5430388450622559, "learning_rate": 0.0002, "epoch": 6.061705989110708, "step": 8350}, {"loss": 0.2682, "grad_norm": 0.8765342235565186, "learning_rate": 0.0002, "epoch": 6.068965517241379, "step": 8360}, {"loss": 0.2614, "grad_norm": 1.2363157272338867, "learning_rate": 0.0002, "epoch": 6.076225045372051, "step": 8370}, {"loss": 0.294, "grad_norm": 1.21284818649292, "learning_rate": 0.0002, "epoch": 6.083484573502722, "step": 8380}, {"loss": 0.2498, "grad_norm": 1.3261712789535522, "learning_rate": 0.0002, "epoch": 6.090744101633394, "step": 8390}, {"loss": 0.2649, "grad_norm": 1.077317714691162, "learning_rate": 0.0002, "epoch": 6.098003629764065, "step": 8400}, {"loss": 0.269, "grad_norm": 0.9873808026313782, "learning_rate": 0.0002, "epoch": 6.105263157894737, "step": 8410}, {"loss": 0.2736, "grad_norm": 1.032258152961731, "learning_rate": 0.0002, "epoch": 6.112522686025408, "step": 8420}, {"loss": 0.2854, "grad_norm": 1.1014811992645264, "learning_rate": 0.0002, "epoch": 6.11978221415608, "step": 8430}, {"loss": 0.2924, "grad_norm": 1.4264203310012817, "learning_rate": 0.0002, "epoch": 6.127041742286751, "step": 8440}, {"loss": 0.3388, "grad_norm": 1.4086531400680542, "learning_rate": 0.0002, "epoch": 6.1343012704174225, "step": 8450}, {"loss": 0.2786, "grad_norm": 1.3842453956604004, "learning_rate": 0.0002, "epoch": 6.1415607985480944, "step": 8460}, {"loss": 0.3201, "grad_norm": 1.4356757402420044, "learning_rate": 0.0002, "epoch": 6.1488203266787655, "step": 8470}, {"loss": 0.2908, "grad_norm": 1.193315029144287, "learning_rate": 0.0002, "epoch": 6.1560798548094375, "step": 8480}, {"loss": 0.342, "grad_norm": 1.0623924732208252, "learning_rate": 0.0002, "epoch": 6.163339382940109, "step": 8490}, {"loss": 0.3257, "grad_norm": 1.5484434366226196, "learning_rate": 0.0002, "epoch": 6.170598911070781, "step": 8500}, {"loss": 0.2861, "grad_norm": 1.3520029783248901, "learning_rate": 0.0002, "epoch": 6.177858439201452, "step": 8510}, {"loss": 0.3242, "grad_norm": 1.2773103713989258, "learning_rate": 0.0002, "epoch": 6.185117967332124, "step": 8520}, {"loss": 0.3108, "grad_norm": 1.4675105810165405, "learning_rate": 0.0002, "epoch": 6.192377495462795, "step": 8530}, {"loss": 0.3044, "grad_norm": 1.2118732929229736, "learning_rate": 0.0002, "epoch": 6.199637023593467, "step": 8540}, {"loss": 0.2726, "grad_norm": 1.264024257659912, "learning_rate": 0.0002, "epoch": 6.206896551724138, "step": 8550}, {"loss": 0.306, "grad_norm": 1.406931757926941, "learning_rate": 0.0002, "epoch": 6.21415607985481, "step": 8560}, {"loss": 0.2904, "grad_norm": 1.385459542274475, "learning_rate": 0.0002, "epoch": 6.221415607985481, "step": 8570}, {"loss": 0.3413, "grad_norm": 1.9336168766021729, "learning_rate": 0.0002, "epoch": 6.228675136116152, "step": 8580}, {"loss": 0.2769, "grad_norm": 0.9880136847496033, "learning_rate": 0.0002, "epoch": 6.235934664246824, "step": 8590}, {"loss": 0.3035, "grad_norm": 1.3870339393615723, "learning_rate": 0.0002, "epoch": 6.243194192377495, "step": 8600}, {"loss": 0.286, "grad_norm": 1.2303647994995117, "learning_rate": 0.0002, "epoch": 6.250453720508167, "step": 8610}, {"loss": 0.3113, "grad_norm": 1.5406211614608765, "learning_rate": 0.0002, "epoch": 6.257713248638838, "step": 8620}, {"loss": 0.292, "grad_norm": 1.2436790466308594, "learning_rate": 0.0002, "epoch": 6.26497277676951, "step": 8630}, {"loss": 0.3102, "grad_norm": 0.8844212293624878, "learning_rate": 0.0002, "epoch": 6.272232304900181, "step": 8640}, {"loss": 0.3373, "grad_norm": 1.2846336364746094, "learning_rate": 0.0002, "epoch": 6.279491833030853, "step": 8650}, {"loss": 0.3535, "grad_norm": 1.593814730644226, "learning_rate": 0.0002, "epoch": 6.286751361161524, "step": 8660}, {"loss": 0.3413, "grad_norm": 1.2277469635009766, "learning_rate": 0.0002, "epoch": 6.2940108892921955, "step": 8670}, {"loss": 0.2958, "grad_norm": 1.2574384212493896, "learning_rate": 0.0002, "epoch": 6.3012704174228675, "step": 8680}, {"loss": 0.3251, "grad_norm": 1.335150957107544, "learning_rate": 0.0002, "epoch": 6.308529945553539, "step": 8690}, {"loss": 0.3009, "grad_norm": 1.3140437602996826, "learning_rate": 0.0002, "epoch": 6.315789473684211, "step": 8700}, {"loss": 0.2783, "grad_norm": 1.1689209938049316, "learning_rate": 0.0002, "epoch": 6.323049001814882, "step": 8710}, {"loss": 0.3476, "grad_norm": 1.6448503732681274, "learning_rate": 0.0002, "epoch": 6.330308529945554, "step": 8720}, {"loss": 0.2934, "grad_norm": 0.9944145679473877, "learning_rate": 0.0002, "epoch": 6.337568058076225, "step": 8730}, {"loss": 0.3315, "grad_norm": 1.1775634288787842, "learning_rate": 0.0002, "epoch": 6.344827586206897, "step": 8740}, {"loss": 0.3514, "grad_norm": 1.8438639640808105, "learning_rate": 0.0002, "epoch": 6.352087114337568, "step": 8750}, {"loss": 0.303, "grad_norm": 1.062495470046997, "learning_rate": 0.0002, "epoch": 6.35934664246824, "step": 8760}, {"loss": 0.2737, "grad_norm": 1.3224315643310547, "learning_rate": 0.0002, "epoch": 6.366606170598911, "step": 8770}, {"loss": 0.3445, "grad_norm": 1.399844765663147, "learning_rate": 0.0002, "epoch": 6.373865698729583, "step": 8780}, {"loss": 0.3277, "grad_norm": 1.0409915447235107, "learning_rate": 0.0002, "epoch": 6.381125226860254, "step": 8790}, {"loss": 0.3218, "grad_norm": 1.5657726526260376, "learning_rate": 0.0002, "epoch": 6.388384754990925, "step": 8800}, {"loss": 0.3031, "grad_norm": 1.4098644256591797, "learning_rate": 0.0002, "epoch": 6.395644283121597, "step": 8810}, {"loss": 0.3133, "grad_norm": 1.5154732465744019, "learning_rate": 0.0002, "epoch": 6.402903811252268, "step": 8820}, {"loss": 0.3111, "grad_norm": 1.1139698028564453, "learning_rate": 0.0002, "epoch": 6.41016333938294, "step": 8830}, {"loss": 0.3553, "grad_norm": 1.4149729013442993, "learning_rate": 0.0002, "epoch": 6.417422867513611, "step": 8840}, {"loss": 0.287, "grad_norm": 1.2632299661636353, "learning_rate": 0.0002, "epoch": 6.424682395644283, "step": 8850}, {"loss": 0.3198, "grad_norm": 1.6636109352111816, "learning_rate": 0.0002, "epoch": 6.431941923774954, "step": 8860}, {"loss": 0.3749, "grad_norm": 1.4149386882781982, "learning_rate": 0.0002, "epoch": 6.439201451905626, "step": 8870}, {"loss": 0.3504, "grad_norm": 1.1396206617355347, "learning_rate": 0.0002, "epoch": 6.4464609800362975, "step": 8880}, {"loss": 0.3328, "grad_norm": 1.2188775539398193, "learning_rate": 0.0002, "epoch": 6.4537205081669695, "step": 8890}, {"loss": 0.3427, "grad_norm": 0.9740369319915771, "learning_rate": 0.0002, "epoch": 6.460980036297641, "step": 8900}, {"loss": 0.3223, "grad_norm": 1.228569746017456, "learning_rate": 0.0002, "epoch": 6.468239564428313, "step": 8910}, {"loss": 0.3151, "grad_norm": 1.5019789934158325, "learning_rate": 0.0002, "epoch": 6.475499092558984, "step": 8920}, {"loss": 0.2916, "grad_norm": 1.3320101499557495, "learning_rate": 0.0002, "epoch": 6.482758620689655, "step": 8930}, {"loss": 0.298, "grad_norm": 1.5551502704620361, "learning_rate": 0.0002, "epoch": 6.490018148820327, "step": 8940}, {"loss": 0.3238, "grad_norm": 1.470131754875183, "learning_rate": 0.0002, "epoch": 6.497277676950998, "step": 8950}, {"loss": 0.2808, "grad_norm": 1.1803025007247925, "learning_rate": 0.0002, "epoch": 6.50453720508167, "step": 8960}, {"loss": 0.3025, "grad_norm": 1.3505640029907227, "learning_rate": 0.0002, "epoch": 6.511796733212341, "step": 8970}, {"loss": 0.3124, "grad_norm": 1.13093900680542, "learning_rate": 0.0002, "epoch": 6.519056261343013, "step": 8980}, {"loss": 0.3454, "grad_norm": 1.347386121749878, "learning_rate": 0.0002, "epoch": 6.526315789473684, "step": 8990}, {"loss": 0.3532, "grad_norm": 1.7879165410995483, "learning_rate": 0.0002, "epoch": 6.533575317604356, "step": 9000}, {"loss": 0.3382, "grad_norm": 1.2168169021606445, "learning_rate": 0.0002, "epoch": 6.540834845735027, "step": 9010}, {"loss": 0.3413, "grad_norm": 1.1758877038955688, "learning_rate": 0.0002, "epoch": 6.548094373865698, "step": 9020}, {"loss": 0.2806, "grad_norm": 1.7366445064544678, "learning_rate": 0.0002, "epoch": 6.55535390199637, "step": 9030}, {"loss": 0.3437, "grad_norm": 1.5919222831726074, "learning_rate": 0.0002, "epoch": 6.562613430127042, "step": 9040}, {"loss": 0.3261, "grad_norm": 1.336863398551941, "learning_rate": 0.0002, "epoch": 6.569872958257713, "step": 9050}, {"loss": 0.3103, "grad_norm": 1.1769421100616455, "learning_rate": 0.0002, "epoch": 6.577132486388384, "step": 9060}, {"loss": 0.3295, "grad_norm": 1.0048751831054688, "learning_rate": 0.0002, "epoch": 6.584392014519056, "step": 9070}, {"loss": 0.3156, "grad_norm": 1.5268515348434448, "learning_rate": 0.0002, "epoch": 6.5916515426497275, "step": 9080}, {"loss": 0.3752, "grad_norm": 1.434610366821289, "learning_rate": 0.0002, "epoch": 6.5989110707803995, "step": 9090}, {"loss": 0.3375, "grad_norm": 1.1151410341262817, "learning_rate": 0.0002, "epoch": 6.606170598911071, "step": 9100}, {"loss": 0.363, "grad_norm": 1.6690642833709717, "learning_rate": 0.0002, "epoch": 6.613430127041743, "step": 9110}, {"loss": 0.3703, "grad_norm": 1.4495552778244019, "learning_rate": 0.0002, "epoch": 6.620689655172414, "step": 9120}, {"loss": 0.3648, "grad_norm": 1.377621054649353, "learning_rate": 0.0002, "epoch": 6.627949183303086, "step": 9130}, {"loss": 0.3766, "grad_norm": 1.5459434986114502, "learning_rate": 0.0002, "epoch": 6.635208711433757, "step": 9140}, {"loss": 0.3196, "grad_norm": 1.0920850038528442, "learning_rate": 0.0002, "epoch": 6.642468239564428, "step": 9150}, {"loss": 0.3505, "grad_norm": 1.6708381175994873, "learning_rate": 0.0002, "epoch": 6.6497277676951, "step": 9160}, {"loss": 0.368, "grad_norm": 1.747514009475708, "learning_rate": 0.0002, "epoch": 6.656987295825771, "step": 9170}, {"loss": 0.3099, "grad_norm": 1.133466362953186, "learning_rate": 0.0002, "epoch": 6.664246823956443, "step": 9180}, {"loss": 0.3175, "grad_norm": 1.394358515739441, "learning_rate": 0.0002, "epoch": 6.671506352087114, "step": 9190}, {"loss": 0.2981, "grad_norm": 0.9258374571800232, "learning_rate": 0.0002, "epoch": 6.678765880217786, "step": 9200}, {"loss": 0.3723, "grad_norm": 1.3750739097595215, "learning_rate": 0.0002, "epoch": 6.686025408348457, "step": 9210}, {"loss": 0.3441, "grad_norm": 0.8604967594146729, "learning_rate": 0.0002, "epoch": 6.693284936479129, "step": 9220}, {"loss": 0.3775, "grad_norm": 1.6074559688568115, "learning_rate": 0.0002, "epoch": 6.7005444646098, "step": 9230}, {"loss": 0.3139, "grad_norm": 0.9576877355575562, "learning_rate": 0.0002, "epoch": 6.707803992740472, "step": 9240}, {"loss": 0.3633, "grad_norm": 1.7193048000335693, "learning_rate": 0.0002, "epoch": 6.715063520871143, "step": 9250}, {"loss": 0.3139, "grad_norm": 1.3131844997406006, "learning_rate": 0.0002, "epoch": 6.722323049001815, "step": 9260}, {"loss": 0.3121, "grad_norm": 1.2978184223175049, "learning_rate": 0.0002, "epoch": 6.729582577132486, "step": 9270}, {"loss": 0.3534, "grad_norm": 1.4792617559432983, "learning_rate": 0.0002, "epoch": 6.7368421052631575, "step": 9280}, {"loss": 0.3429, "grad_norm": 1.1265567541122437, "learning_rate": 0.0002, "epoch": 6.7441016333938295, "step": 9290}, {"loss": 0.3526, "grad_norm": 1.8553377389907837, "learning_rate": 0.0002, "epoch": 6.751361161524501, "step": 9300}, {"loss": 0.3666, "grad_norm": 1.3602519035339355, "learning_rate": 0.0002, "epoch": 6.758620689655173, "step": 9310}, {"loss": 0.2922, "grad_norm": 1.2874794006347656, "learning_rate": 0.0002, "epoch": 6.765880217785844, "step": 9320}, {"loss": 0.3816, "grad_norm": 1.4834712743759155, "learning_rate": 0.0002, "epoch": 6.773139745916516, "step": 9330}, {"loss": 0.3557, "grad_norm": 2.0824034214019775, "learning_rate": 0.0002, "epoch": 6.780399274047187, "step": 9340}, {"loss": 0.3174, "grad_norm": 1.2267698049545288, "learning_rate": 0.0002, "epoch": 6.787658802177859, "step": 9350}, {"loss": 0.3665, "grad_norm": 1.4485498666763306, "learning_rate": 0.0002, "epoch": 6.79491833030853, "step": 9360}, {"loss": 0.3676, "grad_norm": 1.3199396133422852, "learning_rate": 0.0002, "epoch": 6.802177858439201, "step": 9370}, {"loss": 0.298, "grad_norm": 1.2552456855773926, "learning_rate": 0.0002, "epoch": 6.809437386569873, "step": 9380}, {"loss": 0.3152, "grad_norm": 1.3895127773284912, "learning_rate": 0.0002, "epoch": 6.816696914700545, "step": 9390}, {"loss": 0.3375, "grad_norm": 1.7637823820114136, "learning_rate": 0.0002, "epoch": 6.823956442831216, "step": 9400}, {"loss": 0.3234, "grad_norm": 1.6004475355148315, "learning_rate": 0.0002, "epoch": 6.831215970961887, "step": 9410}, {"loss": 0.3364, "grad_norm": 1.4133695363998413, "learning_rate": 0.0002, "epoch": 6.838475499092559, "step": 9420}, {"loss": 0.3656, "grad_norm": 1.1583502292633057, "learning_rate": 0.0002, "epoch": 6.84573502722323, "step": 9430}, {"loss": 0.3499, "grad_norm": 1.3769075870513916, "learning_rate": 0.0002, "epoch": 6.852994555353902, "step": 9440}, {"loss": 0.3333, "grad_norm": 1.1831218004226685, "learning_rate": 0.0002, "epoch": 6.860254083484573, "step": 9450}, {"loss": 0.3501, "grad_norm": 1.6092621088027954, "learning_rate": 0.0002, "epoch": 6.867513611615245, "step": 9460}, {"loss": 0.3933, "grad_norm": 1.3850210905075073, "learning_rate": 0.0002, "epoch": 6.874773139745916, "step": 9470}, {"loss": 0.3868, "grad_norm": 1.4119619131088257, "learning_rate": 0.0002, "epoch": 6.882032667876588, "step": 9480}, {"loss": 0.3939, "grad_norm": 1.3494242429733276, "learning_rate": 0.0002, "epoch": 6.8892921960072595, "step": 9490}, {"loss": 0.3217, "grad_norm": 1.3130041360855103, "learning_rate": 0.0002, "epoch": 6.896551724137931, "step": 9500}, {"loss": 0.3738, "grad_norm": 1.169256329536438, "learning_rate": 0.0002, "epoch": 6.903811252268603, "step": 9510}, {"loss": 0.408, "grad_norm": 1.7475035190582275, "learning_rate": 0.0002, "epoch": 6.911070780399274, "step": 9520}, {"loss": 0.3407, "grad_norm": 1.440434217453003, "learning_rate": 0.0002, "epoch": 6.918330308529946, "step": 9530}, {"loss": 0.3707, "grad_norm": 1.6768704652786255, "learning_rate": 0.0002, "epoch": 6.925589836660617, "step": 9540}, {"loss": 0.3283, "grad_norm": 1.3720577955245972, "learning_rate": 0.0002, "epoch": 6.932849364791289, "step": 9550}, {"loss": 0.3257, "grad_norm": 1.8140523433685303, "learning_rate": 0.0002, "epoch": 6.94010889292196, "step": 9560}, {"loss": 0.3308, "grad_norm": 1.1828241348266602, "learning_rate": 0.0002, "epoch": 6.947368421052632, "step": 9570}, {"loss": 0.3536, "grad_norm": 1.2755135297775269, "learning_rate": 0.0002, "epoch": 6.954627949183303, "step": 9580}, {"loss": 0.3711, "grad_norm": 1.622009038925171, "learning_rate": 0.0002, "epoch": 6.961887477313975, "step": 9590}, {"loss": 0.3529, "grad_norm": 1.1543664932250977, "learning_rate": 0.0002, "epoch": 6.969147005444646, "step": 9600}, {"loss": 0.416, "grad_norm": 1.6755319833755493, "learning_rate": 0.0002, "epoch": 6.976406533575318, "step": 9610}, {"loss": 0.339, "grad_norm": 1.3726437091827393, "learning_rate": 0.0002, "epoch": 6.983666061705989, "step": 9620}, {"loss": 0.3709, "grad_norm": 1.1605958938598633, "learning_rate": 0.0002, "epoch": 6.99092558983666, "step": 9630}, {"loss": 0.371, "grad_norm": 1.5371781587600708, "learning_rate": 0.0002, "epoch": 6.998185117967332, "step": 9640}]}
+{"epoch": 7.997096188747731, "step": 11016, "epoch_duration": 2048.3427917957306, "total_accumulated_duration": 22506.500129699707, "gpu_info": {"GPU_0": "NVIDIA A100-PCIE-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7751.47119140625}, "peak_memory_usage": {"GPU_0": 19996.724609375}, "avg_memory_reserved": {"GPU_0": 24486.0}, "peak_memory_reserved": {"GPU_0": 24752.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/gemma-2-9b-it_int4_mmlu-routerbench-0shot-full-by-task_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.9-num-8828-sd-42/checkpoint-2755", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.4936, "grad_norm": 0.46010470390319824, "learning_rate": 0.0002, "epoch": 0.007259528130671506, "step": 10}, {"loss": 1.8442, "grad_norm": 0.5103179216384888, "learning_rate": 0.0002, "epoch": 0.014519056261343012, "step": 20}, {"loss": 1.559, "grad_norm": 0.6282716989517212, "learning_rate": 0.0002, "epoch": 0.021778584392014518, "step": 30}, {"loss": 1.3618, "grad_norm": 1.2480497360229492, "learning_rate": 0.0002, "epoch": 0.029038112522686024, "step": 40}, {"loss": 1.3874, "grad_norm": 0.4114855229854584, "learning_rate": 0.0002, "epoch": 0.036297640653357534, "step": 50}, {"loss": 1.2836, "grad_norm": 0.49482840299606323, "learning_rate": 0.0002, "epoch": 0.043557168784029036, "step": 60}, {"loss": 1.1728, "grad_norm": 0.4536272883415222, "learning_rate": 0.0002, "epoch": 0.050816696914700546, "step": 70}, {"loss": 1.1125, "grad_norm": 0.32328274846076965, "learning_rate": 0.0002, "epoch": 0.05807622504537205, "step": 80}, {"loss": 1.2447, "grad_norm": 0.40990468859672546, "learning_rate": 0.0002, "epoch": 0.06533575317604355, "step": 90}, {"loss": 1.2216, "grad_norm": 0.37273502349853516, "learning_rate": 0.0002, "epoch": 0.07259528130671507, "step": 100}, {"loss": 1.2259, "grad_norm": 0.3903871476650238, "learning_rate": 0.0002, "epoch": 0.07985480943738657, "step": 110}, {"loss": 1.3718, "grad_norm": 0.3272787928581238, "learning_rate": 0.0002, "epoch": 0.08711433756805807, "step": 120}, {"loss": 1.1832, "grad_norm": 0.3622824251651764, "learning_rate": 0.0002, "epoch": 0.09437386569872959, "step": 130}, {"loss": 1.2966, "grad_norm": 0.3503916561603546, "learning_rate": 0.0002, "epoch": 0.10163339382940109, "step": 140}, {"loss": 1.0913, "grad_norm": 0.32787832617759705, "learning_rate": 0.0002, "epoch": 0.1088929219600726, "step": 150}, {"loss": 1.2143, "grad_norm": 0.5822657942771912, "learning_rate": 0.0002, "epoch": 0.1161524500907441, "step": 160}, {"loss": 1.2091, "grad_norm": 0.28028249740600586, "learning_rate": 0.0002, "epoch": 0.12341197822141561, "step": 170}, {"loss": 1.1731, "grad_norm": 0.33602750301361084, "learning_rate": 0.0002, "epoch": 0.1306715063520871, "step": 180}, {"loss": 1.127, "grad_norm": 0.29106274247169495, "learning_rate": 0.0002, "epoch": 0.13793103448275862, "step": 190}, {"loss": 1.3261, "grad_norm": 0.38753363490104675, "learning_rate": 0.0002, "epoch": 0.14519056261343014, "step": 200}, {"loss": 1.1536, "grad_norm": 0.361009418964386, "learning_rate": 0.0002, "epoch": 0.15245009074410162, "step": 210}, {"loss": 1.1483, "grad_norm": 0.6743836402893066, "learning_rate": 0.0002, "epoch": 0.15970961887477314, "step": 220}, {"loss": 1.264, "grad_norm": 0.3923613727092743, "learning_rate": 0.0002, "epoch": 0.16696914700544466, "step": 230}, {"loss": 1.0437, "grad_norm": 0.2809699773788452, "learning_rate": 0.0002, "epoch": 0.17422867513611615, "step": 240}, {"loss": 1.1826, "grad_norm": 0.3631494641304016, "learning_rate": 0.0002, "epoch": 0.18148820326678766, "step": 250}, {"loss": 1.1584, "grad_norm": 0.24658171832561493, "learning_rate": 0.0002, "epoch": 0.18874773139745918, "step": 260}, {"loss": 1.0065, "grad_norm": 0.5780664682388306, "learning_rate": 0.0002, "epoch": 0.19600725952813067, "step": 270}, {"loss": 1.1251, "grad_norm": 0.3056720495223999, "learning_rate": 0.0002, "epoch": 0.20326678765880218, "step": 280}, {"loss": 1.1503, "grad_norm": 0.2857084572315216, "learning_rate": 0.0002, "epoch": 0.21052631578947367, "step": 290}, {"loss": 1.1215, "grad_norm": 0.6645345687866211, "learning_rate": 0.0002, "epoch": 0.2177858439201452, "step": 300}, {"loss": 1.127, "grad_norm": 0.5966078639030457, "learning_rate": 0.0002, "epoch": 0.2250453720508167, "step": 310}, {"loss": 1.1974, "grad_norm": 0.40937140583992004, "learning_rate": 0.0002, "epoch": 0.2323049001814882, "step": 320}, {"loss": 1.2385, "grad_norm": 0.5642806887626648, "learning_rate": 0.0002, "epoch": 0.2395644283121597, "step": 330}, {"loss": 1.1703, "grad_norm": 0.2750748097896576, "learning_rate": 0.0002, "epoch": 0.24682395644283123, "step": 340}, {"loss": 1.2598, "grad_norm": 0.34350234270095825, "learning_rate": 0.0002, "epoch": 0.2540834845735027, "step": 350}, {"loss": 1.1942, "grad_norm": 0.6767239570617676, "learning_rate": 0.0002, "epoch": 0.2613430127041742, "step": 360}, {"loss": 1.1436, "grad_norm": 0.31006959080696106, "learning_rate": 0.0002, "epoch": 0.26860254083484575, "step": 370}, {"loss": 1.2084, "grad_norm": 0.3825474679470062, "learning_rate": 0.0002, "epoch": 0.27586206896551724, "step": 380}, {"loss": 1.1523, "grad_norm": 0.30241551995277405, "learning_rate": 0.0002, "epoch": 0.2831215970961887, "step": 390}, {"loss": 1.1298, "grad_norm": 0.2962397336959839, "learning_rate": 0.0002, "epoch": 0.29038112522686027, "step": 400}, {"loss": 1.1299, "grad_norm": 0.2600369155406952, "learning_rate": 0.0002, "epoch": 0.29764065335753176, "step": 410}, {"loss": 1.1366, "grad_norm": 0.3675060272216797, "learning_rate": 0.0002, "epoch": 0.30490018148820325, "step": 420}, {"loss": 1.156, "grad_norm": 0.3429498076438904, "learning_rate": 0.0002, "epoch": 0.3121597096188748, "step": 430}, {"loss": 1.2741, "grad_norm": 0.34311825037002563, "learning_rate": 0.0002, "epoch": 0.3194192377495463, "step": 440}, {"loss": 1.3523, "grad_norm": 0.37872210144996643, "learning_rate": 0.0002, "epoch": 0.32667876588021777, "step": 450}, {"loss": 1.1365, "grad_norm": 0.33271121978759766, "learning_rate": 0.0002, "epoch": 0.3339382940108893, "step": 460}, {"loss": 1.1572, "grad_norm": 0.34605276584625244, "learning_rate": 0.0002, "epoch": 0.3411978221415608, "step": 470}, {"loss": 1.2755, "grad_norm": 0.41050562262535095, "learning_rate": 0.0002, "epoch": 0.3484573502722323, "step": 480}, {"loss": 1.075, "grad_norm": 0.2066836953163147, "learning_rate": 0.0002, "epoch": 0.35571687840290384, "step": 490}, {"loss": 1.2072, "grad_norm": 0.2859014868736267, "learning_rate": 0.0002, "epoch": 0.3629764065335753, "step": 500}, {"loss": 1.1435, "grad_norm": 0.28763777017593384, "learning_rate": 0.0002, "epoch": 0.3702359346642468, "step": 510}, {"loss": 1.1341, "grad_norm": 0.2730471193790436, "learning_rate": 0.0002, "epoch": 0.37749546279491836, "step": 520}, {"loss": 1.348, "grad_norm": 0.3968936800956726, "learning_rate": 0.0002, "epoch": 0.38475499092558985, "step": 530}, {"loss": 1.2363, "grad_norm": 0.3624701201915741, "learning_rate": 0.0002, "epoch": 0.39201451905626133, "step": 540}, {"loss": 1.1206, "grad_norm": 0.3303608298301697, "learning_rate": 0.0002, "epoch": 0.3992740471869328, "step": 550}, {"loss": 1.203, "grad_norm": 0.33507466316223145, "learning_rate": 0.0002, "epoch": 0.40653357531760437, "step": 560}, {"loss": 1.2294, "grad_norm": 0.3297670781612396, "learning_rate": 0.0002, "epoch": 0.41379310344827586, "step": 570}, {"loss": 1.2, "grad_norm": 0.32334890961647034, "learning_rate": 0.0002, "epoch": 0.42105263157894735, "step": 580}, {"loss": 1.112, "grad_norm": 0.30281195044517517, "learning_rate": 0.0002, "epoch": 0.4283121597096189, "step": 590}, {"loss": 1.1352, "grad_norm": 0.5900027751922607, "learning_rate": 0.0002, "epoch": 0.4355716878402904, "step": 600}, {"loss": 1.1575, "grad_norm": 0.28437477350234985, "learning_rate": 0.0002, "epoch": 0.44283121597096187, "step": 610}, {"loss": 1.2009, "grad_norm": 0.39601704478263855, "learning_rate": 0.0002, "epoch": 0.4500907441016334, "step": 620}, {"loss": 1.2302, "grad_norm": 0.41971510648727417, "learning_rate": 0.0002, "epoch": 0.4573502722323049, "step": 630}, {"loss": 1.236, "grad_norm": 0.33814409375190735, "learning_rate": 0.0002, "epoch": 0.4646098003629764, "step": 640}, {"loss": 1.2471, "grad_norm": 0.575718104839325, "learning_rate": 0.0002, "epoch": 0.47186932849364793, "step": 650}, {"loss": 1.1012, "grad_norm": 0.37927401065826416, "learning_rate": 0.0002, "epoch": 0.4791288566243194, "step": 660}, {"loss": 1.1552, "grad_norm": 0.3224332630634308, "learning_rate": 0.0002, "epoch": 0.4863883847549909, "step": 670}, {"loss": 1.1354, "grad_norm": 0.32683515548706055, "learning_rate": 0.0002, "epoch": 0.49364791288566245, "step": 680}, {"loss": 1.207, "grad_norm": 0.4316163659095764, "learning_rate": 0.0002, "epoch": 0.5009074410163339, "step": 690}, {"loss": 1.2116, "grad_norm": 0.342602401971817, "learning_rate": 0.0002, "epoch": 0.5081669691470054, "step": 700}, {"loss": 1.0433, "grad_norm": 0.2794898748397827, "learning_rate": 0.0002, "epoch": 0.515426497277677, "step": 710}, {"loss": 1.2395, "grad_norm": 0.3322339951992035, "learning_rate": 0.0002, "epoch": 0.5226860254083484, "step": 720}, {"loss": 1.0762, "grad_norm": 0.3088509142398834, "learning_rate": 0.0002, "epoch": 0.52994555353902, "step": 730}, {"loss": 1.1495, "grad_norm": 0.24444378912448883, "learning_rate": 0.0002, "epoch": 0.5372050816696915, "step": 740}, {"loss": 1.2176, "grad_norm": 0.3483171761035919, "learning_rate": 0.0002, "epoch": 0.5444646098003629, "step": 750}, {"loss": 1.2248, "grad_norm": 0.3425690531730652, "learning_rate": 0.0002, "epoch": 0.5517241379310345, "step": 760}, {"loss": 1.1127, "grad_norm": 0.31841927766799927, "learning_rate": 0.0002, "epoch": 0.558983666061706, "step": 770}, {"loss": 1.1841, "grad_norm": 0.39423868060112, "learning_rate": 0.0002, "epoch": 0.5662431941923775, "step": 780}, {"loss": 1.0583, "grad_norm": 0.30328479409217834, "learning_rate": 0.0002, "epoch": 0.573502722323049, "step": 790}, {"loss": 1.0718, "grad_norm": 0.24475938081741333, "learning_rate": 0.0002, "epoch": 0.5807622504537205, "step": 800}, {"loss": 1.2006, "grad_norm": 0.37132805585861206, "learning_rate": 0.0002, "epoch": 0.588021778584392, "step": 810}, {"loss": 1.1544, "grad_norm": 0.32195979356765747, "learning_rate": 0.0002, "epoch": 0.5952813067150635, "step": 820}, {"loss": 0.9937, "grad_norm": 0.2848738729953766, "learning_rate": 0.0002, "epoch": 0.6025408348457351, "step": 830}, {"loss": 1.0937, "grad_norm": 0.28015264868736267, "learning_rate": 0.0002, "epoch": 0.6098003629764065, "step": 840}, {"loss": 1.1234, "grad_norm": 0.37796008586883545, "learning_rate": 0.0002, "epoch": 0.617059891107078, "step": 850}, {"loss": 1.1455, "grad_norm": 0.39311841130256653, "learning_rate": 0.0002, "epoch": 0.6243194192377496, "step": 860}, {"loss": 1.1066, "grad_norm": 0.2761685252189636, "learning_rate": 0.0002, "epoch": 0.631578947368421, "step": 870}, {"loss": 1.0859, "grad_norm": 0.3826720118522644, "learning_rate": 0.0002, "epoch": 0.6388384754990926, "step": 880}, {"loss": 1.1182, "grad_norm": 0.30076679587364197, "learning_rate": 0.0002, "epoch": 0.6460980036297641, "step": 890}, {"loss": 1.0927, "grad_norm": 0.21997687220573425, "learning_rate": 0.0002, "epoch": 0.6533575317604355, "step": 900}, {"loss": 1.1603, "grad_norm": 0.32593777775764465, "learning_rate": 0.0002, "epoch": 0.6606170598911071, "step": 910}, {"loss": 1.1578, "grad_norm": 0.30347898602485657, "learning_rate": 0.0002, "epoch": 0.6678765880217786, "step": 920}, {"loss": 1.1642, "grad_norm": 0.44173774123191833, "learning_rate": 0.0002, "epoch": 0.6751361161524501, "step": 930}, {"loss": 1.1832, "grad_norm": 0.2507467269897461, "learning_rate": 0.0002, "epoch": 0.6823956442831216, "step": 940}, {"loss": 1.3001, "grad_norm": 0.29463833570480347, "learning_rate": 0.0002, "epoch": 0.6896551724137931, "step": 950}, {"loss": 1.1651, "grad_norm": 0.9363154172897339, "learning_rate": 0.0002, "epoch": 0.6969147005444646, "step": 960}, {"loss": 1.158, "grad_norm": 0.3236212134361267, "learning_rate": 0.0002, "epoch": 0.7041742286751361, "step": 970}, {"loss": 1.1078, "grad_norm": 0.3123254179954529, "learning_rate": 0.0002, "epoch": 0.7114337568058077, "step": 980}, {"loss": 1.1578, "grad_norm": 0.3395805060863495, "learning_rate": 0.0002, "epoch": 0.7186932849364791, "step": 990}, {"loss": 1.0754, "grad_norm": 0.3240964412689209, "learning_rate": 0.0002, "epoch": 0.7259528130671506, "step": 1000}, {"loss": 1.0313, "grad_norm": 0.31902948021888733, "learning_rate": 0.0002, "epoch": 0.7332123411978222, "step": 1010}, {"loss": 1.1664, "grad_norm": 0.4848408102989197, "learning_rate": 0.0002, "epoch": 0.7404718693284936, "step": 1020}, {"loss": 1.0838, "grad_norm": 0.33006033301353455, "learning_rate": 0.0002, "epoch": 0.7477313974591652, "step": 1030}, {"loss": 1.0979, "grad_norm": 0.2928730547428131, "learning_rate": 0.0002, "epoch": 0.7549909255898367, "step": 1040}, {"loss": 1.1351, "grad_norm": 0.3529164791107178, "learning_rate": 0.0002, "epoch": 0.7622504537205081, "step": 1050}, {"loss": 1.1274, "grad_norm": 0.2736213803291321, "learning_rate": 0.0002, "epoch": 0.7695099818511797, "step": 1060}, {"loss": 1.1169, "grad_norm": 0.7200686931610107, "learning_rate": 0.0002, "epoch": 0.7767695099818511, "step": 1070}, {"loss": 1.1171, "grad_norm": 0.33396708965301514, "learning_rate": 0.0002, "epoch": 0.7840290381125227, "step": 1080}, {"loss": 1.2447, "grad_norm": 1.5760449171066284, "learning_rate": 0.0002, "epoch": 0.7912885662431942, "step": 1090}, {"loss": 1.1126, "grad_norm": 0.28138381242752075, "learning_rate": 0.0002, "epoch": 0.7985480943738656, "step": 1100}, {"loss": 1.2012, "grad_norm": 0.2597472369670868, "learning_rate": 0.0002, "epoch": 0.8058076225045372, "step": 1110}, {"loss": 1.1177, "grad_norm": 0.3305445611476898, "learning_rate": 0.0002, "epoch": 0.8130671506352087, "step": 1120}, {"loss": 1.0849, "grad_norm": 0.3934599459171295, "learning_rate": 0.0002, "epoch": 0.8203266787658802, "step": 1130}, {"loss": 1.0889, "grad_norm": 0.3472191393375397, "learning_rate": 0.0002, "epoch": 0.8275862068965517, "step": 1140}, {"loss": 1.265, "grad_norm": 0.2857365906238556, "learning_rate": 0.0002, "epoch": 0.8348457350272233, "step": 1150}, {"loss": 1.03, "grad_norm": 0.3207702934741974, "learning_rate": 0.0002, "epoch": 0.8421052631578947, "step": 1160}, {"loss": 1.1669, "grad_norm": 0.3176484704017639, "learning_rate": 0.0002, "epoch": 0.8493647912885662, "step": 1170}, {"loss": 1.1386, "grad_norm": 0.40685558319091797, "learning_rate": 0.0002, "epoch": 0.8566243194192378, "step": 1180}, {"loss": 1.1383, "grad_norm": 0.31125199794769287, "learning_rate": 0.0002, "epoch": 0.8638838475499092, "step": 1190}, {"loss": 1.1512, "grad_norm": 0.7361181378364563, "learning_rate": 0.0002, "epoch": 0.8711433756805808, "step": 1200}, {"loss": 0.9983, "grad_norm": 0.33699527382850647, "learning_rate": 0.0002, "epoch": 0.8784029038112523, "step": 1210}, {"loss": 1.11, "grad_norm": 0.3315220773220062, "learning_rate": 0.0002, "epoch": 0.8856624319419237, "step": 1220}, {"loss": 1.0266, "grad_norm": 0.6256054043769836, "learning_rate": 0.0002, "epoch": 0.8929219600725953, "step": 1230}, {"loss": 1.1738, "grad_norm": 0.3692137897014618, "learning_rate": 0.0002, "epoch": 0.9001814882032668, "step": 1240}, {"loss": 1.1112, "grad_norm": 0.3538484573364258, "learning_rate": 0.0002, "epoch": 0.9074410163339383, "step": 1250}, {"loss": 1.0878, "grad_norm": 0.27863040566444397, "learning_rate": 0.0002, "epoch": 0.9147005444646098, "step": 1260}, {"loss": 1.0826, "grad_norm": 0.3322528302669525, "learning_rate": 0.0002, "epoch": 0.9219600725952813, "step": 1270}, {"loss": 1.0158, "grad_norm": 0.7553173303604126, "learning_rate": 0.0002, "epoch": 0.9292196007259528, "step": 1280}, {"loss": 1.198, "grad_norm": 0.4856191575527191, "learning_rate": 0.0002, "epoch": 0.9364791288566243, "step": 1290}, {"loss": 1.1678, "grad_norm": 0.3668074905872345, "learning_rate": 0.0002, "epoch": 0.9437386569872959, "step": 1300}, {"loss": 1.1809, "grad_norm": 0.29851067066192627, "learning_rate": 0.0002, "epoch": 0.9509981851179673, "step": 1310}, {"loss": 1.1299, "grad_norm": 0.276664674282074, "learning_rate": 0.0002, "epoch": 0.9582577132486388, "step": 1320}, {"loss": 1.0597, "grad_norm": 0.2941018342971802, "learning_rate": 0.0002, "epoch": 0.9655172413793104, "step": 1330}, {"loss": 1.1731, "grad_norm": 0.3505859076976776, "learning_rate": 0.0002, "epoch": 0.9727767695099818, "step": 1340}, {"loss": 1.1809, "grad_norm": 0.3067687451839447, "learning_rate": 0.0002, "epoch": 0.9800362976406534, "step": 1350}, {"loss": 1.1511, "grad_norm": 0.27151066064834595, "learning_rate": 0.0002, "epoch": 0.9872958257713249, "step": 1360}, {"loss": 1.1919, "grad_norm": 0.36370083689689636, "learning_rate": 0.0002, "epoch": 0.9945553539019963, "step": 1370}, {"eval_loss": 1.1381088495254517, "eval_runtime": 96.0848, "eval_samples_per_second": 4.538, "eval_steps_per_second": 0.572, "epoch": 0.9996370235934664, "step": 1377}, {"loss": 1.0082, "grad_norm": 0.27980583906173706, "learning_rate": 0.0002, "epoch": 1.0018148820326678, "step": 1380}, {"loss": 1.0565, "grad_norm": 0.26713913679122925, "learning_rate": 0.0002, "epoch": 1.0090744101633393, "step": 1390}, {"loss": 1.0241, "grad_norm": 0.3089541494846344, "learning_rate": 0.0002, "epoch": 1.0163339382940109, "step": 1400}, {"loss": 0.9649, "grad_norm": 0.4188242256641388, "learning_rate": 0.0002, "epoch": 1.0235934664246824, "step": 1410}, {"loss": 0.9821, "grad_norm": 0.5246463418006897, "learning_rate": 0.0002, "epoch": 1.030852994555354, "step": 1420}, {"loss": 1.0143, "grad_norm": 0.2728777825832367, "learning_rate": 0.0002, "epoch": 1.0381125226860255, "step": 1430}, {"loss": 1.075, "grad_norm": 0.38167616724967957, "learning_rate": 0.0002, "epoch": 1.0453720508166968, "step": 1440}, {"loss": 1.0229, "grad_norm": 0.4439380168914795, "learning_rate": 0.0002, "epoch": 1.0526315789473684, "step": 1450}, {"loss": 1.0451, "grad_norm": 0.30954182147979736, "learning_rate": 0.0002, "epoch": 1.05989110707804, "step": 1460}, {"loss": 1.0762, "grad_norm": 0.4022280275821686, "learning_rate": 0.0002, "epoch": 1.0671506352087115, "step": 1470}, {"loss": 1.0754, "grad_norm": 0.7390811443328857, "learning_rate": 0.0002, "epoch": 1.074410163339383, "step": 1480}, {"loss": 1.1378, "grad_norm": 0.3885486423969269, "learning_rate": 0.0002, "epoch": 1.0816696914700545, "step": 1490}, {"loss": 1.005, "grad_norm": 0.5275560617446899, "learning_rate": 0.0002, "epoch": 1.0889292196007259, "step": 1500}, {"loss": 1.0147, "grad_norm": 0.35112282633781433, "learning_rate": 0.0002, "epoch": 1.0961887477313974, "step": 1510}, {"loss": 1.0331, "grad_norm": 0.33714351058006287, "learning_rate": 0.0002, "epoch": 1.103448275862069, "step": 1520}, {"loss": 0.8723, "grad_norm": 0.31221693754196167, "learning_rate": 0.0002, "epoch": 1.1107078039927405, "step": 1530}, {"loss": 1.0701, "grad_norm": 0.27549654245376587, "learning_rate": 0.0002, "epoch": 1.117967332123412, "step": 1540}, {"loss": 1.0114, "grad_norm": 0.8465521335601807, "learning_rate": 0.0002, "epoch": 1.1252268602540836, "step": 1550}, {"loss": 0.9643, "grad_norm": 0.36125949025154114, "learning_rate": 0.0002, "epoch": 1.132486388384755, "step": 1560}, {"loss": 1.0744, "grad_norm": 0.37420371174812317, "learning_rate": 0.0002, "epoch": 1.1397459165154264, "step": 1570}, {"loss": 1.165, "grad_norm": 0.3294760584831238, "learning_rate": 0.0002, "epoch": 1.147005444646098, "step": 1580}, {"loss": 0.9207, "grad_norm": 0.3881238102912903, "learning_rate": 0.0002, "epoch": 1.1542649727767695, "step": 1590}, {"loss": 1.0983, "grad_norm": 0.4766491651535034, "learning_rate": 0.0002, "epoch": 1.161524500907441, "step": 1600}, {"loss": 1.0092, "grad_norm": 0.395530104637146, "learning_rate": 0.0002, "epoch": 1.1687840290381124, "step": 1610}, {"loss": 0.9895, "grad_norm": 0.3297106623649597, "learning_rate": 0.0002, "epoch": 1.176043557168784, "step": 1620}, {"loss": 1.1121, "grad_norm": 0.39528271555900574, "learning_rate": 0.0002, "epoch": 1.1833030852994555, "step": 1630}, {"loss": 0.9202, "grad_norm": 0.3370221257209778, "learning_rate": 0.0002, "epoch": 1.190562613430127, "step": 1640}, {"loss": 1.102, "grad_norm": 0.31922030448913574, "learning_rate": 0.0002, "epoch": 1.1978221415607986, "step": 1650}, {"loss": 1.0565, "grad_norm": 0.6142027378082275, "learning_rate": 0.0002, "epoch": 1.2050816696914701, "step": 1660}, {"loss": 0.9541, "grad_norm": 0.44769710302352905, "learning_rate": 0.0002, "epoch": 1.2123411978221417, "step": 1670}, {"loss": 1.2501, "grad_norm": 0.41233646869659424, "learning_rate": 0.0002, "epoch": 1.219600725952813, "step": 1680}, {"loss": 1.0147, "grad_norm": 0.2928866147994995, "learning_rate": 0.0002, "epoch": 1.2268602540834845, "step": 1690}, {"loss": 0.9074, "grad_norm": 0.36913734674453735, "learning_rate": 0.0002, "epoch": 1.234119782214156, "step": 1700}, {"loss": 1.0847, "grad_norm": 0.5281891226768494, "learning_rate": 0.0002, "epoch": 1.2413793103448276, "step": 1710}, {"loss": 0.9482, "grad_norm": 0.3374697268009186, "learning_rate": 0.0002, "epoch": 1.2486388384754992, "step": 1720}, {"loss": 0.9826, "grad_norm": 0.3802020847797394, "learning_rate": 0.0002, "epoch": 1.2558983666061705, "step": 1730}, {"loss": 1.0191, "grad_norm": 0.38048651814460754, "learning_rate": 0.0002, "epoch": 1.263157894736842, "step": 1740}, {"loss": 1.0142, "grad_norm": 0.6676169633865356, "learning_rate": 0.0002, "epoch": 1.2704174228675136, "step": 1750}, {"loss": 0.992, "grad_norm": 0.4075961410999298, "learning_rate": 0.0002, "epoch": 1.2776769509981851, "step": 1760}, {"loss": 1.0301, "grad_norm": 0.4374721348285675, "learning_rate": 0.0002, "epoch": 1.2849364791288567, "step": 1770}, {"loss": 1.0601, "grad_norm": 0.4638824164867401, "learning_rate": 0.0002, "epoch": 1.2921960072595282, "step": 1780}, {"loss": 0.9593, "grad_norm": 0.38631564378738403, "learning_rate": 0.0002, "epoch": 1.2994555353901998, "step": 1790}, {"loss": 1.0042, "grad_norm": 0.35873809456825256, "learning_rate": 0.0002, "epoch": 1.306715063520871, "step": 1800}, {"loss": 0.9835, "grad_norm": 0.33602237701416016, "learning_rate": 0.0002, "epoch": 1.3139745916515426, "step": 1810}, {"loss": 1.1032, "grad_norm": 0.46696463227272034, "learning_rate": 0.0002, "epoch": 1.3212341197822142, "step": 1820}, {"loss": 1.0123, "grad_norm": 0.368958979845047, "learning_rate": 0.0002, "epoch": 1.3284936479128857, "step": 1830}, {"loss": 1.0456, "grad_norm": 0.3160957396030426, "learning_rate": 0.0002, "epoch": 1.335753176043557, "step": 1840}, {"loss": 1.0264, "grad_norm": 0.4511511027812958, "learning_rate": 0.0002, "epoch": 1.3430127041742286, "step": 1850}, {"loss": 1.0179, "grad_norm": 0.6769845485687256, "learning_rate": 0.0002, "epoch": 1.3502722323049001, "step": 1860}, {"loss": 1.011, "grad_norm": 0.3749309480190277, "learning_rate": 0.0002, "epoch": 1.3575317604355717, "step": 1870}, {"loss": 1.0223, "grad_norm": 0.553798496723175, "learning_rate": 0.0002, "epoch": 1.3647912885662432, "step": 1880}, {"loss": 0.9634, "grad_norm": 0.3538985550403595, "learning_rate": 0.0002, "epoch": 1.3720508166969148, "step": 1890}, {"loss": 1.0275, "grad_norm": 0.5501534938812256, "learning_rate": 0.0002, "epoch": 1.3793103448275863, "step": 1900}, {"loss": 1.0461, "grad_norm": 0.4432051181793213, "learning_rate": 0.0002, "epoch": 1.3865698729582578, "step": 1910}, {"loss": 0.9082, "grad_norm": 0.41755786538124084, "learning_rate": 0.0002, "epoch": 1.3938294010889292, "step": 1920}, {"loss": 1.0858, "grad_norm": 0.5626114010810852, "learning_rate": 0.0002, "epoch": 1.4010889292196007, "step": 1930}, {"loss": 1.0687, "grad_norm": 0.44038185477256775, "learning_rate": 0.0002, "epoch": 1.4083484573502723, "step": 1940}, {"loss": 0.9454, "grad_norm": 0.3397001624107361, "learning_rate": 0.0002, "epoch": 1.4156079854809438, "step": 1950}, {"loss": 0.943, "grad_norm": 0.4325368106365204, "learning_rate": 0.0002, "epoch": 1.4228675136116151, "step": 1960}, {"loss": 1.0488, "grad_norm": 0.3900907039642334, "learning_rate": 0.0002, "epoch": 1.4301270417422867, "step": 1970}, {"loss": 0.9699, "grad_norm": 0.369612455368042, "learning_rate": 0.0002, "epoch": 1.4373865698729582, "step": 1980}, {"loss": 1.0609, "grad_norm": 0.4389338791370392, "learning_rate": 0.0002, "epoch": 1.4446460980036298, "step": 1990}, {"loss": 1.042, "grad_norm": 1.694450855255127, "learning_rate": 0.0002, "epoch": 1.4519056261343013, "step": 2000}, {"loss": 0.997, "grad_norm": 0.516957700252533, "learning_rate": 0.0002, "epoch": 1.4591651542649728, "step": 2010}, {"loss": 1.0979, "grad_norm": 0.45515501499176025, "learning_rate": 0.0002, "epoch": 1.4664246823956444, "step": 2020}, {"loss": 1.0188, "grad_norm": 0.4153139591217041, "learning_rate": 0.0002, "epoch": 1.4736842105263157, "step": 2030}, {"loss": 1.1521, "grad_norm": 0.44353052973747253, "learning_rate": 0.0002, "epoch": 1.4809437386569873, "step": 2040}, {"loss": 0.9653, "grad_norm": 0.570554256439209, "learning_rate": 0.0002, "epoch": 1.4882032667876588, "step": 2050}, {"loss": 1.0352, "grad_norm": 0.5742740035057068, "learning_rate": 0.0002, "epoch": 1.4954627949183303, "step": 2060}, {"loss": 1.0963, "grad_norm": 0.5890517830848694, "learning_rate": 0.0002, "epoch": 1.5027223230490017, "step": 2070}, {"loss": 1.0639, "grad_norm": 0.4162650406360626, "learning_rate": 0.0002, "epoch": 1.5099818511796732, "step": 2080}, {"loss": 1.071, "grad_norm": 0.7334717512130737, "learning_rate": 0.0002, "epoch": 1.5172413793103448, "step": 2090}, {"loss": 0.8957, "grad_norm": 0.2963249683380127, "learning_rate": 0.0002, "epoch": 1.5245009074410163, "step": 2100}, {"loss": 1.0446, "grad_norm": 0.30676454305648804, "learning_rate": 0.0002, "epoch": 1.5317604355716878, "step": 2110}, {"loss": 0.9967, "grad_norm": 0.35984641313552856, "learning_rate": 0.0002, "epoch": 1.5390199637023594, "step": 2120}, {"loss": 1.0352, "grad_norm": 0.3384549617767334, "learning_rate": 0.0002, "epoch": 1.546279491833031, "step": 2130}, {"loss": 0.9874, "grad_norm": 0.4725518226623535, "learning_rate": 0.0002, "epoch": 1.5535390199637025, "step": 2140}, {"loss": 1.1166, "grad_norm": 0.3252685070037842, "learning_rate": 0.0002, "epoch": 1.560798548094374, "step": 2150}, {"loss": 0.9127, "grad_norm": 0.45043081045150757, "learning_rate": 0.0002, "epoch": 1.5680580762250453, "step": 2160}, {"loss": 1.0767, "grad_norm": 0.374208003282547, "learning_rate": 0.0002, "epoch": 1.5753176043557169, "step": 2170}, {"loss": 1.0255, "grad_norm": 0.5118404030799866, "learning_rate": 0.0002, "epoch": 1.5825771324863884, "step": 2180}, {"loss": 1.0062, "grad_norm": 0.482170969247818, "learning_rate": 0.0002, "epoch": 1.5898366606170597, "step": 2190}, {"loss": 1.0872, "grad_norm": 0.5337533950805664, "learning_rate": 0.0002, "epoch": 1.5970961887477313, "step": 2200}, {"loss": 1.0405, "grad_norm": 0.5195064544677734, "learning_rate": 0.0002, "epoch": 1.6043557168784028, "step": 2210}, {"loss": 1.0454, "grad_norm": 0.30807098746299744, "learning_rate": 0.0002, "epoch": 1.6116152450090744, "step": 2220}, {"loss": 1.0293, "grad_norm": 0.3962925672531128, "learning_rate": 0.0002, "epoch": 1.618874773139746, "step": 2230}, {"loss": 1.0137, "grad_norm": 0.7636962532997131, "learning_rate": 0.0002, "epoch": 1.6261343012704175, "step": 2240}, {"loss": 0.991, "grad_norm": 0.32380592823028564, "learning_rate": 0.0002, "epoch": 1.633393829401089, "step": 2250}, {"loss": 0.9471, "grad_norm": 0.5767741799354553, "learning_rate": 0.0002, "epoch": 1.6406533575317606, "step": 2260}, {"loss": 0.9524, "grad_norm": 0.39964812994003296, "learning_rate": 0.0002, "epoch": 1.647912885662432, "step": 2270}, {"loss": 0.9866, "grad_norm": 0.622629702091217, "learning_rate": 0.0002, "epoch": 1.6551724137931034, "step": 2280}, {"loss": 0.9609, "grad_norm": 0.40202152729034424, "learning_rate": 0.0002, "epoch": 1.662431941923775, "step": 2290}, {"loss": 0.9859, "grad_norm": 0.4467349052429199, "learning_rate": 0.0002, "epoch": 1.6696914700544465, "step": 2300}, {"loss": 1.0312, "grad_norm": 0.5026949048042297, "learning_rate": 0.0002, "epoch": 1.6769509981851178, "step": 2310}, {"loss": 1.0776, "grad_norm": 0.43754541873931885, "learning_rate": 0.0002, "epoch": 1.6842105263157894, "step": 2320}, {"loss": 0.9786, "grad_norm": 0.42869430780410767, "learning_rate": 0.0002, "epoch": 1.691470054446461, "step": 2330}, {"loss": 1.0747, "grad_norm": 0.4192679524421692, "learning_rate": 0.0002, "epoch": 1.6987295825771325, "step": 2340}, {"loss": 1.0054, "grad_norm": 0.3243155777454376, "learning_rate": 0.0002, "epoch": 1.705989110707804, "step": 2350}, {"loss": 0.9165, "grad_norm": 1.0514075756072998, "learning_rate": 0.0002, "epoch": 1.7132486388384756, "step": 2360}, {"loss": 1.1353, "grad_norm": 0.4719122648239136, "learning_rate": 0.0002, "epoch": 1.720508166969147, "step": 2370}, {"loss": 0.9913, "grad_norm": 0.3846144676208496, "learning_rate": 0.0002, "epoch": 1.7277676950998186, "step": 2380}, {"loss": 0.9935, "grad_norm": 0.7266581058502197, "learning_rate": 0.0002, "epoch": 1.73502722323049, "step": 2390}, {"loss": 0.9509, "grad_norm": 0.6183241605758667, "learning_rate": 0.0002, "epoch": 1.7422867513611615, "step": 2400}, {"loss": 0.9713, "grad_norm": 0.3658260405063629, "learning_rate": 0.0002, "epoch": 1.749546279491833, "step": 2410}, {"loss": 1.0033, "grad_norm": 0.6036322712898254, "learning_rate": 0.0002, "epoch": 1.7568058076225044, "step": 2420}, {"loss": 1.0526, "grad_norm": 0.7872936129570007, "learning_rate": 0.0002, "epoch": 1.764065335753176, "step": 2430}, {"loss": 0.99, "grad_norm": 0.35946124792099, "learning_rate": 0.0002, "epoch": 1.7713248638838475, "step": 2440}, {"loss": 0.8845, "grad_norm": 0.3740338981151581, "learning_rate": 0.0002, "epoch": 1.778584392014519, "step": 2450}, {"loss": 0.9333, "grad_norm": 0.6150230169296265, "learning_rate": 0.0002, "epoch": 1.7858439201451906, "step": 2460}, {"loss": 1.0373, "grad_norm": 0.4726075530052185, "learning_rate": 0.0002, "epoch": 1.793103448275862, "step": 2470}, {"loss": 0.974, "grad_norm": 0.31292253732681274, "learning_rate": 0.0002, "epoch": 1.8003629764065336, "step": 2480}, {"loss": 1.0205, "grad_norm": 0.4463104009628296, "learning_rate": 0.0002, "epoch": 1.8076225045372052, "step": 2490}, {"loss": 1.0771, "grad_norm": 0.7848200798034668, "learning_rate": 0.0002, "epoch": 1.8148820326678767, "step": 2500}, {"loss": 0.9508, "grad_norm": 0.5562082529067993, "learning_rate": 0.0002, "epoch": 1.822141560798548, "step": 2510}, {"loss": 0.9343, "grad_norm": 0.39892756938934326, "learning_rate": 0.0002, "epoch": 1.8294010889292196, "step": 2520}, {"loss": 1.0489, "grad_norm": 1.2923320531845093, "learning_rate": 0.0002, "epoch": 1.8366606170598911, "step": 2530}, {"loss": 1.1203, "grad_norm": 0.6316490769386292, "learning_rate": 0.0002, "epoch": 1.8439201451905625, "step": 2540}, {"loss": 0.9131, "grad_norm": 0.46100497245788574, "learning_rate": 0.0002, "epoch": 1.851179673321234, "step": 2550}, {"loss": 1.0977, "grad_norm": 0.7902987003326416, "learning_rate": 0.0002, "epoch": 1.8584392014519056, "step": 2560}, {"loss": 0.9702, "grad_norm": 0.4596365690231323, "learning_rate": 0.0002, "epoch": 1.865698729582577, "step": 2570}, {"loss": 0.9322, "grad_norm": 0.6592172384262085, "learning_rate": 0.0002, "epoch": 1.8729582577132486, "step": 2580}, {"loss": 0.9549, "grad_norm": 0.5764662027359009, "learning_rate": 0.0002, "epoch": 1.8802177858439202, "step": 2590}, {"loss": 0.9485, "grad_norm": 0.8421637415885925, "learning_rate": 0.0002, "epoch": 1.8874773139745917, "step": 2600}, {"loss": 1.012, "grad_norm": 0.5635305047035217, "learning_rate": 0.0002, "epoch": 1.8947368421052633, "step": 2610}, {"loss": 0.8907, "grad_norm": 0.46231237053871155, "learning_rate": 0.0002, "epoch": 1.9019963702359348, "step": 2620}, {"loss": 0.9543, "grad_norm": 0.3944607079029083, "learning_rate": 0.0002, "epoch": 1.9092558983666061, "step": 2630}, {"loss": 0.9964, "grad_norm": 0.4753907322883606, "learning_rate": 0.0002, "epoch": 1.9165154264972777, "step": 2640}, {"loss": 1.0217, "grad_norm": 0.4151090979576111, "learning_rate": 0.0002, "epoch": 1.9237749546279492, "step": 2650}, {"loss": 1.0397, "grad_norm": 0.6793725490570068, "learning_rate": 0.0002, "epoch": 1.9310344827586206, "step": 2660}, {"loss": 1.0401, "grad_norm": 0.339755117893219, "learning_rate": 0.0002, "epoch": 1.938294010889292, "step": 2670}, {"loss": 1.082, "grad_norm": 0.40789374709129333, "learning_rate": 0.0002, "epoch": 1.9455535390199636, "step": 2680}, {"loss": 0.9973, "grad_norm": 0.3750005066394806, "learning_rate": 0.0002, "epoch": 1.9528130671506352, "step": 2690}, {"loss": 0.9837, "grad_norm": 0.39684441685676575, "learning_rate": 0.0002, "epoch": 1.9600725952813067, "step": 2700}, {"loss": 1.1204, "grad_norm": 0.378287672996521, "learning_rate": 0.0002, "epoch": 1.9673321234119783, "step": 2710}, {"loss": 0.9284, "grad_norm": 0.3668482005596161, "learning_rate": 0.0002, "epoch": 1.9745916515426498, "step": 2720}, {"loss": 1.0512, "grad_norm": 0.49997565150260925, "learning_rate": 0.0002, "epoch": 1.9818511796733214, "step": 2730}, {"loss": 0.9311, "grad_norm": 0.36852124333381653, "learning_rate": 0.0002, "epoch": 1.989110707803993, "step": 2740}, {"loss": 1.0898, "grad_norm": 0.5203380584716797, "learning_rate": 0.0002, "epoch": 1.9963702359346642, "step": 2750}, {"eval_loss": 1.1086540222167969, "eval_runtime": 95.6866, "eval_samples_per_second": 4.557, "eval_steps_per_second": 0.575, "epoch": 2.0, "step": 2755}, {"loss": 0.9791, "grad_norm": 0.35921573638916016, "learning_rate": 0.0002, "epoch": 2.0036297640653356, "step": 2760}, {"loss": 0.8968, "grad_norm": 1.013861894607544, "learning_rate": 0.0002, "epoch": 2.010889292196007, "step": 2770}, {"loss": 0.7873, "grad_norm": 0.4425240159034729, "learning_rate": 0.0002, "epoch": 2.0181488203266786, "step": 2780}, {"loss": 0.8758, "grad_norm": 0.60174161195755, "learning_rate": 0.0002, "epoch": 2.02540834845735, "step": 2790}, {"loss": 0.7989, "grad_norm": 0.47582098841667175, "learning_rate": 0.0002, "epoch": 2.0326678765880217, "step": 2800}, {"loss": 0.8898, "grad_norm": 0.6012811660766602, "learning_rate": 0.0002, "epoch": 2.0399274047186933, "step": 2810}, {"loss": 0.8109, "grad_norm": 0.4444408118724823, "learning_rate": 0.0002, "epoch": 2.047186932849365, "step": 2820}, {"loss": 0.8621, "grad_norm": 0.4864003360271454, "learning_rate": 0.0002, "epoch": 2.0544464609800364, "step": 2830}, {"loss": 0.8817, "grad_norm": 0.5104215741157532, "learning_rate": 0.0002, "epoch": 2.061705989110708, "step": 2840}, {"loss": 0.7789, "grad_norm": 0.6218489408493042, "learning_rate": 0.0002, "epoch": 2.0689655172413794, "step": 2850}, {"loss": 0.8778, "grad_norm": 0.705784261226654, "learning_rate": 0.0002, "epoch": 2.076225045372051, "step": 2860}, {"loss": 0.7851, "grad_norm": 0.48091503977775574, "learning_rate": 0.0002, "epoch": 2.0834845735027225, "step": 2870}, {"loss": 0.8444, "grad_norm": 0.5062456727027893, "learning_rate": 0.0002, "epoch": 2.0907441016333936, "step": 2880}, {"loss": 0.7898, "grad_norm": 0.4862022399902344, "learning_rate": 0.0002, "epoch": 2.098003629764065, "step": 2890}, {"loss": 0.8593, "grad_norm": 0.48264044523239136, "learning_rate": 0.0002, "epoch": 2.1052631578947367, "step": 2900}, {"loss": 0.8325, "grad_norm": 0.43744346499443054, "learning_rate": 0.0002, "epoch": 2.1125226860254083, "step": 2910}, {"loss": 0.9099, "grad_norm": 0.5480492115020752, "learning_rate": 0.0002, "epoch": 2.11978221415608, "step": 2920}, {"loss": 0.7727, "grad_norm": 0.5068560838699341, "learning_rate": 0.0002, "epoch": 2.1270417422867514, "step": 2930}, {"loss": 0.8403, "grad_norm": 0.4650852680206299, "learning_rate": 0.0002, "epoch": 2.134301270417423, "step": 2940}, {"loss": 0.8467, "grad_norm": 0.4929981231689453, "learning_rate": 0.0002, "epoch": 2.1415607985480944, "step": 2950}, {"loss": 0.8747, "grad_norm": 0.6537389755249023, "learning_rate": 0.0002, "epoch": 2.148820326678766, "step": 2960}, {"loss": 0.7663, "grad_norm": 0.8032940626144409, "learning_rate": 0.0002, "epoch": 2.1560798548094375, "step": 2970}, {"loss": 0.7604, "grad_norm": 0.7131643891334534, "learning_rate": 0.0002, "epoch": 2.163339382940109, "step": 2980}, {"loss": 0.8424, "grad_norm": 0.6034275889396667, "learning_rate": 0.0002, "epoch": 2.1705989110707806, "step": 2990}, {"loss": 0.8838, "grad_norm": 0.6081095933914185, "learning_rate": 0.0002, "epoch": 2.1778584392014517, "step": 3000}, {"loss": 0.9013, "grad_norm": 0.5706912875175476, "learning_rate": 0.0002, "epoch": 2.1851179673321233, "step": 3010}, {"loss": 0.8473, "grad_norm": 0.6742380261421204, "learning_rate": 0.0002, "epoch": 2.192377495462795, "step": 3020}, {"loss": 0.8372, "grad_norm": 0.4847496449947357, "learning_rate": 0.0002, "epoch": 2.1996370235934664, "step": 3030}, {"loss": 0.8487, "grad_norm": 0.5529342889785767, "learning_rate": 0.0002, "epoch": 2.206896551724138, "step": 3040}, {"loss": 0.9057, "grad_norm": 0.6108783483505249, "learning_rate": 0.0002, "epoch": 2.2141560798548094, "step": 3050}, {"loss": 0.9659, "grad_norm": 0.8841571807861328, "learning_rate": 0.0002, "epoch": 2.221415607985481, "step": 3060}, {"loss": 0.8538, "grad_norm": 0.4227530360221863, "learning_rate": 0.0002, "epoch": 2.2286751361161525, "step": 3070}, {"loss": 0.7925, "grad_norm": 0.461935818195343, "learning_rate": 0.0002, "epoch": 2.235934664246824, "step": 3080}, {"loss": 0.8542, "grad_norm": 0.5407412648200989, "learning_rate": 0.0002, "epoch": 2.2431941923774956, "step": 3090}, {"loss": 0.8392, "grad_norm": 0.5057998895645142, "learning_rate": 0.0002, "epoch": 2.250453720508167, "step": 3100}, {"loss": 0.7898, "grad_norm": 0.530057966709137, "learning_rate": 0.0002, "epoch": 2.2577132486388383, "step": 3110}, {"loss": 0.8507, "grad_norm": 0.5066842436790466, "learning_rate": 0.0002, "epoch": 2.26497277676951, "step": 3120}, {"loss": 0.7903, "grad_norm": 0.5069178342819214, "learning_rate": 0.0002, "epoch": 2.2722323049001814, "step": 3130}, {"loss": 0.8458, "grad_norm": 0.6095499396324158, "learning_rate": 0.0002, "epoch": 2.279491833030853, "step": 3140}, {"loss": 0.8171, "grad_norm": 0.49524766206741333, "learning_rate": 0.0002, "epoch": 2.2867513611615244, "step": 3150}, {"loss": 0.7568, "grad_norm": 0.5334409475326538, "learning_rate": 0.0002, "epoch": 2.294010889292196, "step": 3160}, {"loss": 0.7533, "grad_norm": 1.681748867034912, "learning_rate": 0.0002, "epoch": 2.3012704174228675, "step": 3170}, {"loss": 0.8155, "grad_norm": 0.7225565314292908, "learning_rate": 0.0002, "epoch": 2.308529945553539, "step": 3180}, {"loss": 0.8034, "grad_norm": 0.5379388928413391, "learning_rate": 0.0002, "epoch": 2.3157894736842106, "step": 3190}, {"loss": 0.8833, "grad_norm": 0.45770326256752014, "learning_rate": 0.0002, "epoch": 2.323049001814882, "step": 3200}, {"loss": 0.8419, "grad_norm": 0.6984533071517944, "learning_rate": 0.0002, "epoch": 2.3303085299455537, "step": 3210}, {"loss": 0.7414, "grad_norm": 0.6725744605064392, "learning_rate": 0.0002, "epoch": 2.337568058076225, "step": 3220}, {"loss": 0.8657, "grad_norm": 1.1247471570968628, "learning_rate": 0.0002, "epoch": 2.344827586206897, "step": 3230}, {"loss": 0.7782, "grad_norm": 1.0240263938903809, "learning_rate": 0.0002, "epoch": 2.352087114337568, "step": 3240}, {"loss": 0.826, "grad_norm": 0.5608096122741699, "learning_rate": 0.0002, "epoch": 2.3593466424682394, "step": 3250}, {"loss": 0.8191, "grad_norm": 0.8294990062713623, "learning_rate": 0.0002, "epoch": 2.366606170598911, "step": 3260}, {"loss": 0.9845, "grad_norm": 0.6734224557876587, "learning_rate": 0.0002, "epoch": 2.3738656987295825, "step": 3270}, {"loss": 0.7921, "grad_norm": 0.6862800717353821, "learning_rate": 0.0002, "epoch": 2.381125226860254, "step": 3280}, {"loss": 0.7752, "grad_norm": 0.5442930459976196, "learning_rate": 0.0002, "epoch": 2.3883847549909256, "step": 3290}, {"loss": 0.8543, "grad_norm": 0.5745394229888916, "learning_rate": 0.0002, "epoch": 2.395644283121597, "step": 3300}, {"loss": 0.833, "grad_norm": 0.6257799863815308, "learning_rate": 0.0002, "epoch": 2.4029038112522687, "step": 3310}, {"loss": 0.8191, "grad_norm": 0.5608420968055725, "learning_rate": 0.0002, "epoch": 2.4101633393829403, "step": 3320}, {"loss": 0.8745, "grad_norm": 0.5512017011642456, "learning_rate": 0.0002, "epoch": 2.417422867513612, "step": 3330}, {"loss": 0.8642, "grad_norm": 0.7452999353408813, "learning_rate": 0.0002, "epoch": 2.4246823956442833, "step": 3340}, {"loss": 0.8417, "grad_norm": 0.4604301452636719, "learning_rate": 0.0002, "epoch": 2.4319419237749544, "step": 3350}, {"loss": 0.9148, "grad_norm": 0.8225823640823364, "learning_rate": 0.0002, "epoch": 2.439201451905626, "step": 3360}, {"loss": 0.8144, "grad_norm": 0.8604981303215027, "learning_rate": 0.0002, "epoch": 2.4464609800362975, "step": 3370}, {"loss": 0.9276, "grad_norm": 0.6620925664901733, "learning_rate": 0.0002, "epoch": 2.453720508166969, "step": 3380}, {"loss": 0.8381, "grad_norm": 0.4750158488750458, "learning_rate": 0.0002, "epoch": 2.4609800362976406, "step": 3390}, {"loss": 0.8249, "grad_norm": 0.6061418056488037, "learning_rate": 0.0002, "epoch": 2.468239564428312, "step": 3400}, {"loss": 0.8944, "grad_norm": 0.5977247357368469, "learning_rate": 0.0002, "epoch": 2.4754990925589837, "step": 3410}, {"loss": 0.8366, "grad_norm": 0.5004227757453918, "learning_rate": 0.0002, "epoch": 2.4827586206896552, "step": 3420}, {"loss": 0.9133, "grad_norm": 0.46428972482681274, "learning_rate": 0.0002, "epoch": 2.490018148820327, "step": 3430}, {"loss": 0.8534, "grad_norm": 0.7658395171165466, "learning_rate": 0.0002, "epoch": 2.4972776769509983, "step": 3440}, {"loss": 0.9066, "grad_norm": 0.621819794178009, "learning_rate": 0.0002, "epoch": 2.50453720508167, "step": 3450}, {"loss": 0.8481, "grad_norm": 0.4556088149547577, "learning_rate": 0.0002, "epoch": 2.511796733212341, "step": 3460}, {"loss": 0.8522, "grad_norm": 0.6124140024185181, "learning_rate": 0.0002, "epoch": 2.519056261343013, "step": 3470}, {"loss": 0.8158, "grad_norm": 0.6256231665611267, "learning_rate": 0.0002, "epoch": 2.526315789473684, "step": 3480}, {"loss": 0.9097, "grad_norm": 0.5464141964912415, "learning_rate": 0.0002, "epoch": 2.5335753176043556, "step": 3490}, {"loss": 0.8708, "grad_norm": 0.51471346616745, "learning_rate": 0.0002, "epoch": 2.540834845735027, "step": 3500}, {"loss": 0.9042, "grad_norm": 0.5326165556907654, "learning_rate": 0.0002, "epoch": 2.5480943738656987, "step": 3510}, {"loss": 0.8749, "grad_norm": 0.4750378429889679, "learning_rate": 0.0002, "epoch": 2.5553539019963702, "step": 3520}, {"loss": 0.8248, "grad_norm": 0.5292279124259949, "learning_rate": 0.0002, "epoch": 2.562613430127042, "step": 3530}, {"loss": 0.8808, "grad_norm": 0.6145227551460266, "learning_rate": 0.0002, "epoch": 2.5698729582577133, "step": 3540}, {"loss": 0.846, "grad_norm": 0.8275189399719238, "learning_rate": 0.0002, "epoch": 2.577132486388385, "step": 3550}, {"loss": 0.875, "grad_norm": 0.5037438273429871, "learning_rate": 0.0002, "epoch": 2.5843920145190564, "step": 3560}, {"loss": 0.8324, "grad_norm": 0.5838707685470581, "learning_rate": 0.0002, "epoch": 2.5916515426497275, "step": 3570}, {"loss": 0.8716, "grad_norm": 0.5398710370063782, "learning_rate": 0.0002, "epoch": 2.5989110707803995, "step": 3580}, {"loss": 0.7843, "grad_norm": 0.6115376949310303, "learning_rate": 0.0002, "epoch": 2.6061705989110706, "step": 3590}, {"loss": 0.8142, "grad_norm": 1.118809461593628, "learning_rate": 0.0002, "epoch": 2.613430127041742, "step": 3600}, {"loss": 0.8282, "grad_norm": 0.7811771631240845, "learning_rate": 0.0002, "epoch": 2.6206896551724137, "step": 3610}, {"loss": 0.7997, "grad_norm": 0.5753175020217896, "learning_rate": 0.0002, "epoch": 2.6279491833030852, "step": 3620}, {"loss": 0.7705, "grad_norm": 0.550829291343689, "learning_rate": 0.0002, "epoch": 2.635208711433757, "step": 3630}, {"loss": 0.8564, "grad_norm": 0.5360019207000732, "learning_rate": 0.0002, "epoch": 2.6424682395644283, "step": 3640}, {"loss": 0.7648, "grad_norm": 0.63050377368927, "learning_rate": 0.0002, "epoch": 2.6497277676951, "step": 3650}, {"loss": 0.9156, "grad_norm": 0.5833110213279724, "learning_rate": 0.0002, "epoch": 2.6569872958257714, "step": 3660}, {"loss": 0.8033, "grad_norm": 0.5543047189712524, "learning_rate": 0.0002, "epoch": 2.664246823956443, "step": 3670}, {"loss": 0.8867, "grad_norm": 0.6842212080955505, "learning_rate": 0.0002, "epoch": 2.671506352087114, "step": 3680}, {"loss": 0.8569, "grad_norm": 0.6859333515167236, "learning_rate": 0.0002, "epoch": 2.678765880217786, "step": 3690}, {"loss": 0.8081, "grad_norm": 0.7038410902023315, "learning_rate": 0.0002, "epoch": 2.686025408348457, "step": 3700}, {"loss": 0.8686, "grad_norm": 0.447233647108078, "learning_rate": 0.0002, "epoch": 2.6932849364791287, "step": 3710}, {"loss": 0.8951, "grad_norm": 0.6453872919082642, "learning_rate": 0.0002, "epoch": 2.7005444646098002, "step": 3720}, {"loss": 0.8292, "grad_norm": 0.8025672435760498, "learning_rate": 0.0002, "epoch": 2.707803992740472, "step": 3730}, {"loss": 0.8488, "grad_norm": 0.5997087955474854, "learning_rate": 0.0002, "epoch": 2.7150635208711433, "step": 3740}, {"loss": 0.8308, "grad_norm": 0.6901142001152039, "learning_rate": 0.0002, "epoch": 2.722323049001815, "step": 3750}, {"loss": 0.8517, "grad_norm": 1.036145567893982, "learning_rate": 0.0002, "epoch": 2.7295825771324864, "step": 3760}, {"loss": 0.8455, "grad_norm": 0.7207072377204895, "learning_rate": 0.0002, "epoch": 2.736842105263158, "step": 3770}, {"loss": 0.8009, "grad_norm": 1.0452989339828491, "learning_rate": 0.0002, "epoch": 2.7441016333938295, "step": 3780}, {"loss": 0.8868, "grad_norm": 0.5615278482437134, "learning_rate": 0.0002, "epoch": 2.751361161524501, "step": 3790}, {"loss": 0.9047, "grad_norm": 0.46439215540885925, "learning_rate": 0.0002, "epoch": 2.7586206896551726, "step": 3800}, {"loss": 0.7847, "grad_norm": 0.7134917974472046, "learning_rate": 0.0002, "epoch": 2.7658802177858437, "step": 3810}, {"loss": 0.8315, "grad_norm": 0.5139115452766418, "learning_rate": 0.0002, "epoch": 2.7731397459165157, "step": 3820}, {"loss": 0.8995, "grad_norm": 0.8595781326293945, "learning_rate": 0.0002, "epoch": 2.780399274047187, "step": 3830}, {"loss": 0.9079, "grad_norm": 0.544614851474762, "learning_rate": 0.0002, "epoch": 2.7876588021778583, "step": 3840}, {"loss": 0.8322, "grad_norm": 0.6073850393295288, "learning_rate": 0.0002, "epoch": 2.79491833030853, "step": 3850}, {"loss": 0.8277, "grad_norm": 0.8224069476127625, "learning_rate": 0.0002, "epoch": 2.8021778584392014, "step": 3860}, {"loss": 0.815, "grad_norm": 0.5347970128059387, "learning_rate": 0.0002, "epoch": 2.809437386569873, "step": 3870}, {"loss": 0.8219, "grad_norm": 0.6799601912498474, "learning_rate": 0.0002, "epoch": 2.8166969147005445, "step": 3880}, {"loss": 0.8796, "grad_norm": 0.5219197869300842, "learning_rate": 0.0002, "epoch": 2.823956442831216, "step": 3890}, {"loss": 0.8649, "grad_norm": 0.5710130333900452, "learning_rate": 0.0002, "epoch": 2.8312159709618876, "step": 3900}, {"loss": 0.8741, "grad_norm": 0.5857213139533997, "learning_rate": 0.0002, "epoch": 2.838475499092559, "step": 3910}, {"loss": 0.8154, "grad_norm": 0.5206644535064697, "learning_rate": 0.0002, "epoch": 2.8457350272232302, "step": 3920}, {"loss": 0.9328, "grad_norm": 1.2902015447616577, "learning_rate": 0.0002, "epoch": 2.8529945553539022, "step": 3930}, {"loss": 0.8192, "grad_norm": 0.5252797603607178, "learning_rate": 0.0002, "epoch": 2.8602540834845733, "step": 3940}, {"loss": 0.9047, "grad_norm": 0.5925108790397644, "learning_rate": 0.0002, "epoch": 2.867513611615245, "step": 3950}, {"loss": 0.8611, "grad_norm": 0.9719610810279846, "learning_rate": 0.0002, "epoch": 2.8747731397459164, "step": 3960}, {"loss": 0.8531, "grad_norm": 0.6834747195243835, "learning_rate": 0.0002, "epoch": 2.882032667876588, "step": 3970}, {"loss": 0.8124, "grad_norm": 0.6952353119850159, "learning_rate": 0.0002, "epoch": 2.8892921960072595, "step": 3980}, {"loss": 0.8365, "grad_norm": 0.49889910221099854, "learning_rate": 0.0002, "epoch": 2.896551724137931, "step": 3990}, {"loss": 0.8197, "grad_norm": 0.5007026791572571, "learning_rate": 0.0002, "epoch": 2.9038112522686026, "step": 4000}, {"loss": 0.7752, "grad_norm": 0.5474239587783813, "learning_rate": 0.0002, "epoch": 2.911070780399274, "step": 4010}, {"loss": 0.8579, "grad_norm": 0.6617428660392761, "learning_rate": 0.0002, "epoch": 2.9183303085299457, "step": 4020}, {"loss": 0.8583, "grad_norm": 0.6097776293754578, "learning_rate": 0.0002, "epoch": 2.925589836660617, "step": 4030}, {"loss": 0.8296, "grad_norm": 0.5985828638076782, "learning_rate": 0.0002, "epoch": 2.9328493647912888, "step": 4040}, {"loss": 0.9134, "grad_norm": 0.769488513469696, "learning_rate": 0.0002, "epoch": 2.94010889292196, "step": 4050}, {"loss": 0.8148, "grad_norm": 0.5167055130004883, "learning_rate": 0.0002, "epoch": 2.9473684210526314, "step": 4060}, {"loss": 0.8665, "grad_norm": 0.6265496015548706, "learning_rate": 0.0002, "epoch": 2.954627949183303, "step": 4070}, {"loss": 0.9218, "grad_norm": 1.2644082307815552, "learning_rate": 0.0002, "epoch": 2.9618874773139745, "step": 4080}, {"loss": 0.8026, "grad_norm": 0.6007561087608337, "learning_rate": 0.0002, "epoch": 2.969147005444646, "step": 4090}, {"loss": 0.8262, "grad_norm": 0.47984927892684937, "learning_rate": 0.0002, "epoch": 2.9764065335753176, "step": 4100}, {"loss": 0.7897, "grad_norm": 1.128198504447937, "learning_rate": 0.0002, "epoch": 2.983666061705989, "step": 4110}, {"loss": 0.7883, "grad_norm": 0.526292085647583, "learning_rate": 0.0002, "epoch": 2.9909255898366607, "step": 4120}, {"loss": 0.7801, "grad_norm": 0.5462674498558044, "learning_rate": 0.0002, "epoch": 2.9981851179673322, "step": 4130}, {"eval_loss": 1.1093357801437378, "eval_runtime": 46.2498, "eval_samples_per_second": 9.427, "eval_steps_per_second": 1.189, "epoch": 2.9996370235934666, "step": 4132}, {"loss": 0.721, "grad_norm": 0.48288026452064514, "learning_rate": 0.0002, "epoch": 3.0054446460980038, "step": 4140}, {"loss": 0.6769, "grad_norm": 1.0181782245635986, "learning_rate": 0.0002, "epoch": 3.0127041742286753, "step": 4150}, {"loss": 0.7185, "grad_norm": 0.7718019485473633, "learning_rate": 0.0002, "epoch": 3.019963702359347, "step": 4160}, {"loss": 0.6552, "grad_norm": 0.7492219805717468, "learning_rate": 0.0002, "epoch": 3.027223230490018, "step": 4170}, {"loss": 0.6678, "grad_norm": 0.9363632798194885, "learning_rate": 0.0002, "epoch": 3.0344827586206895, "step": 4180}, {"loss": 0.7187, "grad_norm": 0.6888533234596252, "learning_rate": 0.0002, "epoch": 3.041742286751361, "step": 4190}, {"loss": 0.6469, "grad_norm": 0.7072834968566895, "learning_rate": 0.0002, "epoch": 3.0490018148820326, "step": 4200}, {"loss": 0.6387, "grad_norm": 0.7182047963142395, "learning_rate": 0.0002, "epoch": 3.056261343012704, "step": 4210}, {"loss": 0.6385, "grad_norm": 0.7194355130195618, "learning_rate": 0.0002, "epoch": 3.0635208711433757, "step": 4220}, {"loss": 0.5812, "grad_norm": 0.9454023838043213, "learning_rate": 0.0002, "epoch": 3.0707803992740472, "step": 4230}, {"loss": 0.6036, "grad_norm": 0.838657557964325, "learning_rate": 0.0002, "epoch": 3.0780399274047188, "step": 4240}, {"loss": 0.646, "grad_norm": 0.740113377571106, "learning_rate": 0.0002, "epoch": 3.0852994555353903, "step": 4250}, {"loss": 0.604, "grad_norm": 0.6616561412811279, "learning_rate": 0.0002, "epoch": 3.092558983666062, "step": 4260}, {"loss": 0.6462, "grad_norm": 0.8846506476402283, "learning_rate": 0.0002, "epoch": 3.0998185117967334, "step": 4270}, {"loss": 0.6037, "grad_norm": 0.6322125792503357, "learning_rate": 0.0002, "epoch": 3.107078039927405, "step": 4280}, {"loss": 0.5953, "grad_norm": 0.7461467385292053, "learning_rate": 0.0002, "epoch": 3.114337568058076, "step": 4290}, {"loss": 0.6099, "grad_norm": 0.8251287341117859, "learning_rate": 0.0002, "epoch": 3.1215970961887476, "step": 4300}, {"loss": 0.6284, "grad_norm": 0.8767673373222351, "learning_rate": 0.0002, "epoch": 3.128856624319419, "step": 4310}, {"loss": 0.7535, "grad_norm": 0.7758759260177612, "learning_rate": 0.0002, "epoch": 3.1361161524500907, "step": 4320}, {"loss": 0.6624, "grad_norm": 1.1056879758834839, "learning_rate": 0.0002, "epoch": 3.143375680580762, "step": 4330}, {"loss": 0.691, "grad_norm": 0.8259835243225098, "learning_rate": 0.0002, "epoch": 3.1506352087114338, "step": 4340}, {"loss": 0.6635, "grad_norm": 0.6607027053833008, "learning_rate": 0.0002, "epoch": 3.1578947368421053, "step": 4350}, {"loss": 0.5911, "grad_norm": 0.7983301281929016, "learning_rate": 0.0002, "epoch": 3.165154264972777, "step": 4360}, {"loss": 0.6496, "grad_norm": 0.6725239157676697, "learning_rate": 0.0002, "epoch": 3.1724137931034484, "step": 4370}, {"loss": 0.5966, "grad_norm": 0.9052095413208008, "learning_rate": 0.0002, "epoch": 3.17967332123412, "step": 4380}, {"loss": 0.6877, "grad_norm": 0.8131307363510132, "learning_rate": 0.0002, "epoch": 3.1869328493647915, "step": 4390}, {"loss": 0.6384, "grad_norm": 0.6435626149177551, "learning_rate": 0.0002, "epoch": 3.1941923774954626, "step": 4400}, {"loss": 0.5819, "grad_norm": 0.84367436170578, "learning_rate": 0.0002, "epoch": 3.201451905626134, "step": 4410}, {"loss": 0.6104, "grad_norm": 1.5018867254257202, "learning_rate": 0.0002, "epoch": 3.2087114337568057, "step": 4420}, {"loss": 0.6838, "grad_norm": 0.7019091844558716, "learning_rate": 0.0002, "epoch": 3.215970961887477, "step": 4430}, {"loss": 0.6153, "grad_norm": 0.9164197444915771, "learning_rate": 0.0002, "epoch": 3.2232304900181488, "step": 4440}, {"loss": 0.6618, "grad_norm": 0.7890861630439758, "learning_rate": 0.0002, "epoch": 3.2304900181488203, "step": 4450}, {"loss": 0.6401, "grad_norm": 0.6517660617828369, "learning_rate": 0.0002, "epoch": 3.237749546279492, "step": 4460}, {"loss": 0.6699, "grad_norm": 1.10188889503479, "learning_rate": 0.0002, "epoch": 3.2450090744101634, "step": 4470}, {"loss": 0.6356, "grad_norm": 0.8158330917358398, "learning_rate": 0.0002, "epoch": 3.252268602540835, "step": 4480}, {"loss": 0.7757, "grad_norm": 0.7663109302520752, "learning_rate": 0.0002, "epoch": 3.2595281306715065, "step": 4490}, {"loss": 0.6539, "grad_norm": 0.8473444581031799, "learning_rate": 0.0002, "epoch": 3.266787658802178, "step": 4500}, {"loss": 0.6511, "grad_norm": 0.9724768996238708, "learning_rate": 0.0002, "epoch": 3.274047186932849, "step": 4510}, {"loss": 0.5464, "grad_norm": 0.8516759276390076, "learning_rate": 0.0002, "epoch": 3.281306715063521, "step": 4520}, {"loss": 0.6534, "grad_norm": 0.7543437480926514, "learning_rate": 0.0002, "epoch": 3.288566243194192, "step": 4530}, {"loss": 0.6095, "grad_norm": 1.0472029447555542, "learning_rate": 0.0002, "epoch": 3.2958257713248638, "step": 4540}, {"loss": 0.6216, "grad_norm": 0.6240826845169067, "learning_rate": 0.0002, "epoch": 3.3030852994555353, "step": 4550}, {"loss": 0.6223, "grad_norm": 0.9957774877548218, "learning_rate": 0.0002, "epoch": 3.310344827586207, "step": 4560}, {"loss": 0.618, "grad_norm": 0.6448912620544434, "learning_rate": 0.0002, "epoch": 3.3176043557168784, "step": 4570}, {"loss": 0.6188, "grad_norm": 0.7519692778587341, "learning_rate": 0.0002, "epoch": 3.32486388384755, "step": 4580}, {"loss": 0.6672, "grad_norm": 0.7367453575134277, "learning_rate": 0.0002, "epoch": 3.3321234119782215, "step": 4590}, {"loss": 0.6517, "grad_norm": 0.8064960837364197, "learning_rate": 0.0002, "epoch": 3.339382940108893, "step": 4600}, {"loss": 0.6062, "grad_norm": 0.7664631009101868, "learning_rate": 0.0002, "epoch": 3.3466424682395646, "step": 4610}, {"loss": 0.6834, "grad_norm": 0.7803396582603455, "learning_rate": 0.0002, "epoch": 3.353901996370236, "step": 4620}, {"loss": 0.6961, "grad_norm": 0.9141599535942078, "learning_rate": 0.0002, "epoch": 3.3611615245009077, "step": 4630}, {"loss": 0.6889, "grad_norm": 0.9719856381416321, "learning_rate": 0.0002, "epoch": 3.3684210526315788, "step": 4640}, {"loss": 0.6914, "grad_norm": 0.9223218560218811, "learning_rate": 0.0002, "epoch": 3.3756805807622503, "step": 4650}, {"loss": 0.5981, "grad_norm": 0.7289277911186218, "learning_rate": 0.0002, "epoch": 3.382940108892922, "step": 4660}, {"loss": 0.595, "grad_norm": 1.039724349975586, "learning_rate": 0.0002, "epoch": 3.3901996370235934, "step": 4670}, {"loss": 0.8121, "grad_norm": 1.397438883781433, "learning_rate": 0.0002, "epoch": 3.397459165154265, "step": 4680}, {"loss": 0.6334, "grad_norm": 1.0069999694824219, "learning_rate": 0.0002, "epoch": 3.4047186932849365, "step": 4690}, {"loss": 0.6598, "grad_norm": 0.816291332244873, "learning_rate": 0.0002, "epoch": 3.411978221415608, "step": 4700}, {"loss": 0.6748, "grad_norm": 1.2831530570983887, "learning_rate": 0.0002, "epoch": 3.4192377495462796, "step": 4710}, {"loss": 0.6625, "grad_norm": 0.9573889970779419, "learning_rate": 0.0002, "epoch": 3.426497277676951, "step": 4720}, {"loss": 0.7279, "grad_norm": 0.7685632705688477, "learning_rate": 0.0002, "epoch": 3.4337568058076227, "step": 4730}, {"loss": 0.6104, "grad_norm": 0.7019195556640625, "learning_rate": 0.0002, "epoch": 3.441016333938294, "step": 4740}, {"loss": 0.7606, "grad_norm": 0.7244833707809448, "learning_rate": 0.0002, "epoch": 3.4482758620689653, "step": 4750}, {"loss": 0.6951, "grad_norm": 1.3468551635742188, "learning_rate": 0.0002, "epoch": 3.455535390199637, "step": 4760}, {"loss": 0.6945, "grad_norm": 0.822846531867981, "learning_rate": 0.0002, "epoch": 3.4627949183303084, "step": 4770}, {"loss": 0.6431, "grad_norm": 0.7311608195304871, "learning_rate": 0.0002, "epoch": 3.47005444646098, "step": 4780}, {"loss": 0.7019, "grad_norm": 0.9466770887374878, "learning_rate": 0.0002, "epoch": 3.4773139745916515, "step": 4790}, {"loss": 0.7767, "grad_norm": 1.1527155637741089, "learning_rate": 0.0002, "epoch": 3.484573502722323, "step": 4800}, {"loss": 0.6882, "grad_norm": 1.1288906335830688, "learning_rate": 0.0002, "epoch": 3.4918330308529946, "step": 4810}, {"loss": 0.6564, "grad_norm": 0.9096164107322693, "learning_rate": 0.0002, "epoch": 3.499092558983666, "step": 4820}, {"loss": 0.6127, "grad_norm": 0.7988565564155579, "learning_rate": 0.0002, "epoch": 3.5063520871143377, "step": 4830}, {"loss": 0.7004, "grad_norm": 0.7183415293693542, "learning_rate": 0.0002, "epoch": 3.513611615245009, "step": 4840}, {"loss": 0.74, "grad_norm": 0.6614915132522583, "learning_rate": 0.0002, "epoch": 3.5208711433756807, "step": 4850}, {"loss": 0.7271, "grad_norm": 0.8609521985054016, "learning_rate": 0.0002, "epoch": 3.528130671506352, "step": 4860}, {"loss": 0.6664, "grad_norm": 0.86552894115448, "learning_rate": 0.0002, "epoch": 3.535390199637024, "step": 4870}, {"loss": 0.6432, "grad_norm": 0.6926496028900146, "learning_rate": 0.0002, "epoch": 3.542649727767695, "step": 4880}, {"loss": 0.7117, "grad_norm": 0.8157467246055603, "learning_rate": 0.0002, "epoch": 3.5499092558983665, "step": 4890}, {"loss": 0.6201, "grad_norm": 0.9085357189178467, "learning_rate": 0.0002, "epoch": 3.557168784029038, "step": 4900}, {"loss": 0.6521, "grad_norm": 0.6322644948959351, "learning_rate": 0.0002, "epoch": 3.5644283121597096, "step": 4910}, {"loss": 0.6607, "grad_norm": 1.263205885887146, "learning_rate": 0.0002, "epoch": 3.571687840290381, "step": 4920}, {"loss": 0.6657, "grad_norm": 0.8901070356369019, "learning_rate": 0.0002, "epoch": 3.5789473684210527, "step": 4930}, {"loss": 0.6434, "grad_norm": 0.7983952164649963, "learning_rate": 0.0002, "epoch": 3.586206896551724, "step": 4940}, {"loss": 0.6861, "grad_norm": 0.9887813925743103, "learning_rate": 0.0002, "epoch": 3.5934664246823957, "step": 4950}, {"loss": 0.6502, "grad_norm": 0.7895187735557556, "learning_rate": 0.0002, "epoch": 3.6007259528130673, "step": 4960}, {"loss": 0.7111, "grad_norm": 0.9685819745063782, "learning_rate": 0.0002, "epoch": 3.6079854809437384, "step": 4970}, {"loss": 0.6915, "grad_norm": 0.6576591730117798, "learning_rate": 0.0002, "epoch": 3.6152450090744104, "step": 4980}, {"loss": 0.6195, "grad_norm": 0.856985330581665, "learning_rate": 0.0002, "epoch": 3.6225045372050815, "step": 4990}, {"loss": 0.6318, "grad_norm": 0.7230252623558044, "learning_rate": 0.0002, "epoch": 3.629764065335753, "step": 5000}, {"loss": 0.742, "grad_norm": 0.8260893821716309, "learning_rate": 0.0002, "epoch": 3.6370235934664246, "step": 5010}, {"loss": 0.7223, "grad_norm": 0.7635950446128845, "learning_rate": 0.0002, "epoch": 3.644283121597096, "step": 5020}, {"loss": 0.6837, "grad_norm": 0.7060768604278564, "learning_rate": 0.0002, "epoch": 3.6515426497277677, "step": 5030}, {"loss": 0.6921, "grad_norm": 0.8020303249359131, "learning_rate": 0.0002, "epoch": 3.658802177858439, "step": 5040}, {"loss": 0.6446, "grad_norm": 0.8530341386795044, "learning_rate": 0.0002, "epoch": 3.6660617059891107, "step": 5050}, {"loss": 0.7222, "grad_norm": 0.6667101979255676, "learning_rate": 0.0002, "epoch": 3.6733212341197823, "step": 5060}, {"loss": 0.7081, "grad_norm": 0.7385406494140625, "learning_rate": 0.0002, "epoch": 3.680580762250454, "step": 5070}, {"loss": 0.7035, "grad_norm": 0.7753380537033081, "learning_rate": 0.0002, "epoch": 3.6878402903811254, "step": 5080}, {"loss": 0.6491, "grad_norm": 0.7516207098960876, "learning_rate": 0.0002, "epoch": 3.695099818511797, "step": 5090}, {"loss": 0.672, "grad_norm": 0.8171586394309998, "learning_rate": 0.0002, "epoch": 3.702359346642468, "step": 5100}, {"loss": 0.7459, "grad_norm": 1.0796279907226562, "learning_rate": 0.0002, "epoch": 3.70961887477314, "step": 5110}, {"loss": 0.5948, "grad_norm": 0.6957688927650452, "learning_rate": 0.0002, "epoch": 3.716878402903811, "step": 5120}, {"loss": 0.7515, "grad_norm": 0.8550161719322205, "learning_rate": 0.0002, "epoch": 3.7241379310344827, "step": 5130}, {"loss": 0.7286, "grad_norm": 0.9396728277206421, "learning_rate": 0.0002, "epoch": 3.731397459165154, "step": 5140}, {"loss": 0.7594, "grad_norm": 1.4264805316925049, "learning_rate": 0.0002, "epoch": 3.7386569872958257, "step": 5150}, {"loss": 0.6575, "grad_norm": 0.8725108504295349, "learning_rate": 0.0002, "epoch": 3.7459165154264973, "step": 5160}, {"loss": 0.6778, "grad_norm": 1.0346195697784424, "learning_rate": 0.0002, "epoch": 3.753176043557169, "step": 5170}, {"loss": 0.6371, "grad_norm": 0.5395554304122925, "learning_rate": 0.0002, "epoch": 3.7604355716878404, "step": 5180}, {"loss": 0.7308, "grad_norm": 1.3153616189956665, "learning_rate": 0.0002, "epoch": 3.767695099818512, "step": 5190}, {"loss": 0.78, "grad_norm": 0.9879828691482544, "learning_rate": 0.0002, "epoch": 3.7749546279491835, "step": 5200}, {"loss": 0.7068, "grad_norm": 0.8876672983169556, "learning_rate": 0.0002, "epoch": 3.7822141560798546, "step": 5210}, {"loss": 0.6283, "grad_norm": 0.8363267779350281, "learning_rate": 0.0002, "epoch": 3.7894736842105265, "step": 5220}, {"loss": 0.6255, "grad_norm": 0.637294590473175, "learning_rate": 0.0002, "epoch": 3.7967332123411976, "step": 5230}, {"loss": 0.6685, "grad_norm": 1.1408970355987549, "learning_rate": 0.0002, "epoch": 3.803992740471869, "step": 5240}, {"loss": 0.6761, "grad_norm": 1.0128360986709595, "learning_rate": 0.0002, "epoch": 3.8112522686025407, "step": 5250}, {"loss": 0.6764, "grad_norm": 0.8061144351959229, "learning_rate": 0.0002, "epoch": 3.8185117967332123, "step": 5260}, {"loss": 0.7254, "grad_norm": 0.9626626968383789, "learning_rate": 0.0002, "epoch": 3.825771324863884, "step": 5270}, {"loss": 0.7367, "grad_norm": 0.9013627171516418, "learning_rate": 0.0002, "epoch": 3.8330308529945554, "step": 5280}, {"loss": 0.6806, "grad_norm": 0.8411344289779663, "learning_rate": 0.0002, "epoch": 3.840290381125227, "step": 5290}, {"loss": 0.6818, "grad_norm": 0.7426059246063232, "learning_rate": 0.0002, "epoch": 3.8475499092558985, "step": 5300}, {"loss": 0.6748, "grad_norm": 1.003413438796997, "learning_rate": 0.0002, "epoch": 3.85480943738657, "step": 5310}, {"loss": 0.8554, "grad_norm": 0.7527840733528137, "learning_rate": 0.0002, "epoch": 3.862068965517241, "step": 5320}, {"loss": 0.7521, "grad_norm": 0.738610565662384, "learning_rate": 0.0002, "epoch": 3.869328493647913, "step": 5330}, {"loss": 0.7266, "grad_norm": 0.7277999520301819, "learning_rate": 0.0002, "epoch": 3.876588021778584, "step": 5340}, {"loss": 0.7503, "grad_norm": 0.5951359272003174, "learning_rate": 0.0002, "epoch": 3.8838475499092557, "step": 5350}, {"loss": 0.7447, "grad_norm": 1.043884038925171, "learning_rate": 0.0002, "epoch": 3.8911070780399273, "step": 5360}, {"loss": 0.6862, "grad_norm": 0.8436498045921326, "learning_rate": 0.0002, "epoch": 3.898366606170599, "step": 5370}, {"loss": 0.665, "grad_norm": 0.5603365302085876, "learning_rate": 0.0002, "epoch": 3.9056261343012704, "step": 5380}, {"loss": 0.7098, "grad_norm": 1.0128886699676514, "learning_rate": 0.0002, "epoch": 3.912885662431942, "step": 5390}, {"loss": 0.6707, "grad_norm": 0.7970930337905884, "learning_rate": 0.0002, "epoch": 3.9201451905626135, "step": 5400}, {"loss": 0.637, "grad_norm": 0.7699369192123413, "learning_rate": 0.0002, "epoch": 3.927404718693285, "step": 5410}, {"loss": 0.6742, "grad_norm": 0.800561249256134, "learning_rate": 0.0002, "epoch": 3.9346642468239565, "step": 5420}, {"loss": 0.7208, "grad_norm": 0.8020331859588623, "learning_rate": 0.0002, "epoch": 3.941923774954628, "step": 5430}, {"loss": 0.7294, "grad_norm": 0.7461140155792236, "learning_rate": 0.0002, "epoch": 3.9491833030852996, "step": 5440}, {"loss": 0.7013, "grad_norm": 0.8346918821334839, "learning_rate": 0.0002, "epoch": 3.9564428312159707, "step": 5450}, {"loss": 0.6289, "grad_norm": 0.9723302125930786, "learning_rate": 0.0002, "epoch": 3.9637023593466427, "step": 5460}, {"loss": 0.8029, "grad_norm": 0.6809740662574768, "learning_rate": 0.0002, "epoch": 3.970961887477314, "step": 5470}, {"loss": 0.6896, "grad_norm": 0.7353498339653015, "learning_rate": 0.0002, "epoch": 3.9782214156079854, "step": 5480}, {"loss": 0.6722, "grad_norm": 0.748009443283081, "learning_rate": 0.0002, "epoch": 3.985480943738657, "step": 5490}, {"loss": 0.6866, "grad_norm": 1.3656195402145386, "learning_rate": 0.0002, "epoch": 3.9927404718693285, "step": 5500}, {"loss": 0.7368, "grad_norm": 0.8402108550071716, "learning_rate": 0.0002, "epoch": 4.0, "step": 5510}, {"eval_loss": 1.17229425907135, "eval_runtime": 46.2554, "eval_samples_per_second": 9.426, "eval_steps_per_second": 1.189, "epoch": 4.0, "step": 5510}, {"loss": 0.4637, "grad_norm": 0.8601235747337341, "learning_rate": 0.0002, "epoch": 4.007259528130671, "step": 5520}, {"loss": 0.4717, "grad_norm": 1.2635200023651123, "learning_rate": 0.0002, "epoch": 4.014519056261343, "step": 5530}, {"loss": 0.503, "grad_norm": 1.0257477760314941, "learning_rate": 0.0002, "epoch": 4.021778584392014, "step": 5540}, {"loss": 0.4547, "grad_norm": 0.9436745047569275, "learning_rate": 0.0002, "epoch": 4.029038112522686, "step": 5550}, {"loss": 0.459, "grad_norm": 0.9443606734275818, "learning_rate": 0.0002, "epoch": 4.036297640653357, "step": 5560}, {"loss": 0.5386, "grad_norm": 1.3965742588043213, "learning_rate": 0.0002, "epoch": 4.043557168784029, "step": 5570}, {"loss": 0.4248, "grad_norm": 0.8973520398139954, "learning_rate": 0.0002, "epoch": 4.0508166969147, "step": 5580}, {"loss": 0.4111, "grad_norm": 0.9998409748077393, "learning_rate": 0.0002, "epoch": 4.058076225045372, "step": 5590}, {"loss": 0.4828, "grad_norm": 1.1213387250900269, "learning_rate": 0.0002, "epoch": 4.0653357531760435, "step": 5600}, {"loss": 0.439, "grad_norm": 0.7064558863639832, "learning_rate": 0.0002, "epoch": 4.072595281306715, "step": 5610}, {"loss": 0.4607, "grad_norm": 1.2390803098678589, "learning_rate": 0.0002, "epoch": 4.0798548094373865, "step": 5620}, {"loss": 0.5014, "grad_norm": 1.123469591140747, "learning_rate": 0.0002, "epoch": 4.087114337568058, "step": 5630}, {"loss": 0.513, "grad_norm": 1.229573369026184, "learning_rate": 0.0002, "epoch": 4.09437386569873, "step": 5640}, {"loss": 0.5258, "grad_norm": 1.7182831764221191, "learning_rate": 0.0002, "epoch": 4.101633393829401, "step": 5650}, {"loss": 0.5371, "grad_norm": 0.894903302192688, "learning_rate": 0.0002, "epoch": 4.108892921960073, "step": 5660}, {"loss": 0.4813, "grad_norm": 0.8754552006721497, "learning_rate": 0.0002, "epoch": 4.116152450090744, "step": 5670}, {"loss": 0.491, "grad_norm": 1.2401553392410278, "learning_rate": 0.0002, "epoch": 4.123411978221416, "step": 5680}, {"loss": 0.4549, "grad_norm": 0.8631148934364319, "learning_rate": 0.0002, "epoch": 4.130671506352087, "step": 5690}, {"loss": 0.487, "grad_norm": 1.1798022985458374, "learning_rate": 0.0002, "epoch": 4.137931034482759, "step": 5700}, {"loss": 0.4522, "grad_norm": 0.8344549536705017, "learning_rate": 0.0002, "epoch": 4.14519056261343, "step": 5710}, {"loss": 0.4559, "grad_norm": 1.2342697381973267, "learning_rate": 0.0002, "epoch": 4.152450090744102, "step": 5720}, {"loss": 0.53, "grad_norm": 1.1601094007492065, "learning_rate": 0.0002, "epoch": 4.159709618874773, "step": 5730}, {"loss": 0.4755, "grad_norm": 1.2925703525543213, "learning_rate": 0.0002, "epoch": 4.166969147005445, "step": 5740}, {"loss": 0.4973, "grad_norm": 1.0870997905731201, "learning_rate": 0.0002, "epoch": 4.174228675136116, "step": 5750}, {"loss": 0.5184, "grad_norm": 0.9077792763710022, "learning_rate": 0.0002, "epoch": 4.181488203266787, "step": 5760}, {"loss": 0.4606, "grad_norm": 1.009273886680603, "learning_rate": 0.0002, "epoch": 4.188747731397459, "step": 5770}, {"loss": 0.5383, "grad_norm": 1.2465530633926392, "learning_rate": 0.0002, "epoch": 4.19600725952813, "step": 5780}, {"loss": 0.4938, "grad_norm": 1.2261253595352173, "learning_rate": 0.0002, "epoch": 4.203266787658802, "step": 5790}, {"loss": 0.5255, "grad_norm": 1.1498041152954102, "learning_rate": 0.0002, "epoch": 4.2105263157894735, "step": 5800}, {"loss": 0.5269, "grad_norm": 1.1966725587844849, "learning_rate": 0.0002, "epoch": 4.217785843920145, "step": 5810}, {"loss": 0.5626, "grad_norm": 1.2651296854019165, "learning_rate": 0.0002, "epoch": 4.2250453720508165, "step": 5820}, {"loss": 0.5213, "grad_norm": 1.0388574600219727, "learning_rate": 0.0002, "epoch": 4.2323049001814885, "step": 5830}, {"loss": 0.4965, "grad_norm": 1.3042771816253662, "learning_rate": 0.0002, "epoch": 4.23956442831216, "step": 5840}, {"loss": 0.5116, "grad_norm": 1.1127727031707764, "learning_rate": 0.0002, "epoch": 4.246823956442832, "step": 5850}, {"loss": 0.5197, "grad_norm": 0.9653958082199097, "learning_rate": 0.0002, "epoch": 4.254083484573503, "step": 5860}, {"loss": 0.4414, "grad_norm": 1.0500504970550537, "learning_rate": 0.0002, "epoch": 4.261343012704174, "step": 5870}, {"loss": 0.544, "grad_norm": 1.1476165056228638, "learning_rate": 0.0002, "epoch": 4.268602540834846, "step": 5880}, {"loss": 0.4667, "grad_norm": 0.9424414038658142, "learning_rate": 0.0002, "epoch": 4.275862068965517, "step": 5890}, {"loss": 0.5039, "grad_norm": 1.3309166431427002, "learning_rate": 0.0002, "epoch": 4.283121597096189, "step": 5900}, {"loss": 0.5472, "grad_norm": 1.3025873899459839, "learning_rate": 0.0002, "epoch": 4.29038112522686, "step": 5910}, {"loss": 0.4644, "grad_norm": 1.1442325115203857, "learning_rate": 0.0002, "epoch": 4.297640653357532, "step": 5920}, {"loss": 0.5066, "grad_norm": 0.9820859432220459, "learning_rate": 0.0002, "epoch": 4.304900181488203, "step": 5930}, {"loss": 0.5224, "grad_norm": 0.9615740180015564, "learning_rate": 0.0002, "epoch": 4.312159709618875, "step": 5940}, {"loss": 0.5665, "grad_norm": 1.1627109050750732, "learning_rate": 0.0002, "epoch": 4.319419237749546, "step": 5950}, {"loss": 0.4852, "grad_norm": 0.9381322860717773, "learning_rate": 0.0002, "epoch": 4.326678765880218, "step": 5960}, {"loss": 0.4532, "grad_norm": 0.8154335618019104, "learning_rate": 0.0002, "epoch": 4.333938294010889, "step": 5970}, {"loss": 0.5237, "grad_norm": 0.877671480178833, "learning_rate": 0.0002, "epoch": 4.341197822141561, "step": 5980}, {"loss": 0.6113, "grad_norm": 1.1742031574249268, "learning_rate": 0.0002, "epoch": 4.348457350272232, "step": 5990}, {"loss": 0.5704, "grad_norm": 1.0352917909622192, "learning_rate": 0.0002, "epoch": 4.3557168784029034, "step": 6000}, {"loss": 0.4996, "grad_norm": 0.9963878989219666, "learning_rate": 0.0002, "epoch": 4.362976406533575, "step": 6010}, {"loss": 0.4813, "grad_norm": 1.1892237663269043, "learning_rate": 0.0002, "epoch": 4.3702359346642465, "step": 6020}, {"loss": 0.5224, "grad_norm": 1.2516111135482788, "learning_rate": 0.0002, "epoch": 4.3774954627949185, "step": 6030}, {"loss": 0.5952, "grad_norm": 1.2111951112747192, "learning_rate": 0.0002, "epoch": 4.38475499092559, "step": 6040}, {"loss": 0.4275, "grad_norm": 1.0820083618164062, "learning_rate": 0.0002, "epoch": 4.392014519056262, "step": 6050}, {"loss": 0.5117, "grad_norm": 1.033915638923645, "learning_rate": 0.0002, "epoch": 4.399274047186933, "step": 6060}, {"loss": 0.5431, "grad_norm": 1.0635870695114136, "learning_rate": 0.0002, "epoch": 4.406533575317605, "step": 6070}, {"loss": 0.5341, "grad_norm": 1.0520414113998413, "learning_rate": 0.0002, "epoch": 4.413793103448276, "step": 6080}, {"loss": 0.512, "grad_norm": 1.0821926593780518, "learning_rate": 0.0002, "epoch": 4.421052631578947, "step": 6090}, {"loss": 0.5065, "grad_norm": 1.0533246994018555, "learning_rate": 0.0002, "epoch": 4.428312159709619, "step": 6100}, {"loss": 0.4577, "grad_norm": 0.9231932759284973, "learning_rate": 0.0002, "epoch": 4.43557168784029, "step": 6110}, {"loss": 0.583, "grad_norm": 0.9910260438919067, "learning_rate": 0.0002, "epoch": 4.442831215970962, "step": 6120}, {"loss": 0.4717, "grad_norm": 1.061949372291565, "learning_rate": 0.0002, "epoch": 4.450090744101633, "step": 6130}, {"loss": 0.5893, "grad_norm": 1.2927039861679077, "learning_rate": 0.0002, "epoch": 4.457350272232305, "step": 6140}, {"loss": 0.4684, "grad_norm": 1.3966081142425537, "learning_rate": 0.0002, "epoch": 4.464609800362976, "step": 6150}, {"loss": 0.5507, "grad_norm": 1.3835992813110352, "learning_rate": 0.0002, "epoch": 4.471869328493648, "step": 6160}, {"loss": 0.5911, "grad_norm": 1.0892692804336548, "learning_rate": 0.0002, "epoch": 4.479128856624319, "step": 6170}, {"loss": 0.478, "grad_norm": 1.0318800210952759, "learning_rate": 0.0002, "epoch": 4.486388384754991, "step": 6180}, {"loss": 0.5198, "grad_norm": 0.8174677491188049, "learning_rate": 0.0002, "epoch": 4.493647912885662, "step": 6190}, {"loss": 0.5387, "grad_norm": 1.4157509803771973, "learning_rate": 0.0002, "epoch": 4.500907441016334, "step": 6200}, {"loss": 0.5868, "grad_norm": 1.5244114398956299, "learning_rate": 0.0002, "epoch": 4.508166969147005, "step": 6210}, {"loss": 0.4642, "grad_norm": 0.8164850473403931, "learning_rate": 0.0002, "epoch": 4.5154264972776765, "step": 6220}, {"loss": 0.522, "grad_norm": 1.2904746532440186, "learning_rate": 0.0002, "epoch": 4.5226860254083485, "step": 6230}, {"loss": 0.5103, "grad_norm": 0.7987732887268066, "learning_rate": 0.0002, "epoch": 4.52994555353902, "step": 6240}, {"loss": 0.4615, "grad_norm": 0.831040620803833, "learning_rate": 0.0002, "epoch": 4.537205081669692, "step": 6250}, {"loss": 0.5065, "grad_norm": 0.9545485973358154, "learning_rate": 0.0002, "epoch": 4.544464609800363, "step": 6260}, {"loss": 0.5515, "grad_norm": 0.9291793704032898, "learning_rate": 0.0002, "epoch": 4.551724137931035, "step": 6270}, {"loss": 0.4535, "grad_norm": 0.8977208733558655, "learning_rate": 0.0002, "epoch": 4.558983666061706, "step": 6280}, {"loss": 0.544, "grad_norm": 1.1768537759780884, "learning_rate": 0.0002, "epoch": 4.566243194192378, "step": 6290}, {"loss": 0.5925, "grad_norm": 1.0688952207565308, "learning_rate": 0.0002, "epoch": 4.573502722323049, "step": 6300}, {"loss": 0.5207, "grad_norm": 0.8800966739654541, "learning_rate": 0.0002, "epoch": 4.580762250453721, "step": 6310}, {"loss": 0.6106, "grad_norm": 1.0911834239959717, "learning_rate": 0.0002, "epoch": 4.588021778584392, "step": 6320}, {"loss": 0.5109, "grad_norm": 1.1420872211456299, "learning_rate": 0.0002, "epoch": 4.595281306715064, "step": 6330}, {"loss": 0.5147, "grad_norm": 1.0215224027633667, "learning_rate": 0.0002, "epoch": 4.602540834845735, "step": 6340}, {"loss": 0.592, "grad_norm": 0.9685489535331726, "learning_rate": 0.0002, "epoch": 4.609800362976406, "step": 6350}, {"loss": 0.5775, "grad_norm": 1.12773597240448, "learning_rate": 0.0002, "epoch": 4.617059891107078, "step": 6360}, {"loss": 0.5966, "grad_norm": 1.0663973093032837, "learning_rate": 0.0002, "epoch": 4.624319419237749, "step": 6370}, {"loss": 0.512, "grad_norm": 1.1707262992858887, "learning_rate": 0.0002, "epoch": 4.631578947368421, "step": 6380}, {"loss": 0.5497, "grad_norm": 1.0672980546951294, "learning_rate": 0.0002, "epoch": 4.638838475499092, "step": 6390}, {"loss": 0.5699, "grad_norm": 1.1464333534240723, "learning_rate": 0.0002, "epoch": 4.646098003629764, "step": 6400}, {"loss": 0.5514, "grad_norm": 1.070230484008789, "learning_rate": 0.0002, "epoch": 4.653357531760435, "step": 6410}, {"loss": 0.5013, "grad_norm": 0.9673764109611511, "learning_rate": 0.0002, "epoch": 4.660617059891107, "step": 6420}, {"loss": 0.5901, "grad_norm": 1.0189043283462524, "learning_rate": 0.0002, "epoch": 4.6678765880217785, "step": 6430}, {"loss": 0.5193, "grad_norm": 1.185896396636963, "learning_rate": 0.0002, "epoch": 4.67513611615245, "step": 6440}, {"loss": 0.5318, "grad_norm": 1.0682812929153442, "learning_rate": 0.0002, "epoch": 4.682395644283122, "step": 6450}, {"loss": 0.5773, "grad_norm": 1.3586071729660034, "learning_rate": 0.0002, "epoch": 4.689655172413794, "step": 6460}, {"loss": 0.5482, "grad_norm": 0.6561792492866516, "learning_rate": 0.0002, "epoch": 4.696914700544465, "step": 6470}, {"loss": 0.5711, "grad_norm": 1.1394113302230835, "learning_rate": 0.0002, "epoch": 4.704174228675136, "step": 6480}, {"loss": 0.5325, "grad_norm": 0.9683151245117188, "learning_rate": 0.0002, "epoch": 4.711433756805808, "step": 6490}, {"loss": 0.5139, "grad_norm": 1.0247553586959839, "learning_rate": 0.0002, "epoch": 4.718693284936479, "step": 6500}, {"loss": 0.5794, "grad_norm": 0.8046169281005859, "learning_rate": 0.0002, "epoch": 4.725952813067151, "step": 6510}, {"loss": 0.5471, "grad_norm": 1.0710240602493286, "learning_rate": 0.0002, "epoch": 4.733212341197822, "step": 6520}, {"loss": 0.4805, "grad_norm": 0.9438924193382263, "learning_rate": 0.0002, "epoch": 4.740471869328494, "step": 6530}, {"loss": 0.5404, "grad_norm": 0.869162380695343, "learning_rate": 0.0002, "epoch": 4.747731397459165, "step": 6540}, {"loss": 0.6379, "grad_norm": 0.9776787161827087, "learning_rate": 0.0002, "epoch": 4.754990925589837, "step": 6550}, {"loss": 0.5288, "grad_norm": 1.1990505456924438, "learning_rate": 0.0002, "epoch": 4.762250453720508, "step": 6560}, {"loss": 0.5539, "grad_norm": 1.0582209825515747, "learning_rate": 0.0002, "epoch": 4.769509981851179, "step": 6570}, {"loss": 0.489, "grad_norm": 0.9966367483139038, "learning_rate": 0.0002, "epoch": 4.776769509981851, "step": 6580}, {"loss": 0.5514, "grad_norm": 0.9130612015724182, "learning_rate": 0.0002, "epoch": 4.784029038112522, "step": 6590}, {"loss": 0.5864, "grad_norm": 1.0950500965118408, "learning_rate": 0.0002, "epoch": 4.791288566243194, "step": 6600}, {"loss": 0.5266, "grad_norm": 1.108681321144104, "learning_rate": 0.0002, "epoch": 4.798548094373865, "step": 6610}, {"loss": 0.5875, "grad_norm": 1.1873763799667358, "learning_rate": 0.0002, "epoch": 4.805807622504537, "step": 6620}, {"loss": 0.5736, "grad_norm": 1.305367112159729, "learning_rate": 0.0002, "epoch": 4.8130671506352085, "step": 6630}, {"loss": 0.5636, "grad_norm": 1.2801482677459717, "learning_rate": 0.0002, "epoch": 4.8203266787658805, "step": 6640}, {"loss": 0.582, "grad_norm": 1.26764976978302, "learning_rate": 0.0002, "epoch": 4.827586206896552, "step": 6650}, {"loss": 0.5259, "grad_norm": 1.0018208026885986, "learning_rate": 0.0002, "epoch": 4.834845735027224, "step": 6660}, {"loss": 0.548, "grad_norm": 1.2326326370239258, "learning_rate": 0.0002, "epoch": 4.842105263157895, "step": 6670}, {"loss": 0.5933, "grad_norm": 0.9707282781600952, "learning_rate": 0.0002, "epoch": 4.849364791288567, "step": 6680}, {"loss": 0.5612, "grad_norm": 1.2772048711776733, "learning_rate": 0.0002, "epoch": 4.856624319419238, "step": 6690}, {"loss": 0.5346, "grad_norm": 2.6652262210845947, "learning_rate": 0.0002, "epoch": 4.863883847549909, "step": 6700}, {"loss": 0.5428, "grad_norm": 1.215828537940979, "learning_rate": 0.0002, "epoch": 4.871143375680581, "step": 6710}, {"loss": 0.6571, "grad_norm": 1.3704510927200317, "learning_rate": 0.0002, "epoch": 4.878402903811252, "step": 6720}, {"loss": 0.4963, "grad_norm": 0.7781757116317749, "learning_rate": 0.0002, "epoch": 4.885662431941924, "step": 6730}, {"loss": 0.5989, "grad_norm": 1.1883646249771118, "learning_rate": 0.0002, "epoch": 4.892921960072595, "step": 6740}, {"loss": 0.6067, "grad_norm": 0.9216066002845764, "learning_rate": 0.0002, "epoch": 4.900181488203267, "step": 6750}, {"loss": 0.5085, "grad_norm": 1.0558464527130127, "learning_rate": 0.0002, "epoch": 4.907441016333938, "step": 6760}, {"loss": 0.5216, "grad_norm": 1.032656192779541, "learning_rate": 0.0002, "epoch": 4.91470054446461, "step": 6770}, {"loss": 0.5426, "grad_norm": 1.1261441707611084, "learning_rate": 0.0002, "epoch": 4.921960072595281, "step": 6780}, {"loss": 0.5295, "grad_norm": 1.2178640365600586, "learning_rate": 0.0002, "epoch": 4.929219600725952, "step": 6790}, {"loss": 0.5476, "grad_norm": 1.5369361639022827, "learning_rate": 0.0002, "epoch": 4.936479128856624, "step": 6800}, {"loss": 0.5358, "grad_norm": 1.1188377141952515, "learning_rate": 0.0002, "epoch": 4.943738656987296, "step": 6810}, {"loss": 0.5483, "grad_norm": 1.2506113052368164, "learning_rate": 0.0002, "epoch": 4.950998185117967, "step": 6820}, {"loss": 0.567, "grad_norm": 0.8776047825813293, "learning_rate": 0.0002, "epoch": 4.9582577132486385, "step": 6830}, {"loss": 0.5764, "grad_norm": 0.9700555205345154, "learning_rate": 0.0002, "epoch": 4.9655172413793105, "step": 6840}, {"loss": 0.5396, "grad_norm": 1.2713534832000732, "learning_rate": 0.0002, "epoch": 4.972776769509982, "step": 6850}, {"loss": 0.5451, "grad_norm": 0.9855955243110657, "learning_rate": 0.0002, "epoch": 4.980036297640654, "step": 6860}, {"loss": 0.5884, "grad_norm": 0.8734853863716125, "learning_rate": 0.0002, "epoch": 4.987295825771325, "step": 6870}, {"loss": 0.5189, "grad_norm": 0.8065403699874878, "learning_rate": 0.0002, "epoch": 4.994555353901997, "step": 6880}, {"eval_loss": 1.3302682638168335, "eval_runtime": 46.2496, "eval_samples_per_second": 9.427, "eval_steps_per_second": 1.189, "epoch": 4.999637023593467, "step": 6887}, {"loss": 0.4889, "grad_norm": 0.5163813829421997, "learning_rate": 0.0002, "epoch": 5.001814882032668, "step": 6890}, {"loss": 0.3545, "grad_norm": 1.1496137380599976, "learning_rate": 0.0002, "epoch": 5.00907441016334, "step": 6900}, {"loss": 0.39, "grad_norm": 1.0133885145187378, "learning_rate": 0.0002, "epoch": 5.016333938294011, "step": 6910}, {"loss": 0.3693, "grad_norm": 0.9479621052742004, "learning_rate": 0.0002, "epoch": 5.023593466424682, "step": 6920}, {"loss": 0.4012, "grad_norm": 0.8587583303451538, "learning_rate": 0.0002, "epoch": 5.030852994555354, "step": 6930}, {"loss": 0.3428, "grad_norm": 1.3314697742462158, "learning_rate": 0.0002, "epoch": 5.038112522686025, "step": 6940}, {"loss": 0.3909, "grad_norm": 1.195448875427246, "learning_rate": 0.0002, "epoch": 5.045372050816697, "step": 6950}, {"loss": 0.3322, "grad_norm": 1.2482256889343262, "learning_rate": 0.0002, "epoch": 5.052631578947368, "step": 6960}, {"loss": 0.3893, "grad_norm": 1.2011528015136719, "learning_rate": 0.0002, "epoch": 5.05989110707804, "step": 6970}, {"loss": 0.3265, "grad_norm": 1.3997188806533813, "learning_rate": 0.0002, "epoch": 5.067150635208711, "step": 6980}, {"loss": 0.3716, "grad_norm": 1.2147513628005981, "learning_rate": 0.0002, "epoch": 5.074410163339383, "step": 6990}, {"loss": 0.4053, "grad_norm": 1.6030137538909912, "learning_rate": 0.0002, "epoch": 5.081669691470054, "step": 7000}, {"loss": 0.3665, "grad_norm": 0.9466970562934875, "learning_rate": 0.0002, "epoch": 5.088929219600726, "step": 7010}, {"loss": 0.3451, "grad_norm": 1.4593411684036255, "learning_rate": 0.0002, "epoch": 5.096188747731397, "step": 7020}, {"loss": 0.3843, "grad_norm": 1.2196033000946045, "learning_rate": 0.0002, "epoch": 5.103448275862069, "step": 7030}, {"loss": 0.3896, "grad_norm": 1.1341328620910645, "learning_rate": 0.0002, "epoch": 5.1107078039927405, "step": 7040}, {"loss": 0.3627, "grad_norm": 1.2248976230621338, "learning_rate": 0.0002, "epoch": 5.117967332123412, "step": 7050}, {"loss": 0.3784, "grad_norm": 1.1620593070983887, "learning_rate": 0.0002, "epoch": 5.125226860254084, "step": 7060}, {"loss": 0.3678, "grad_norm": 0.9300723671913147, "learning_rate": 0.0002, "epoch": 5.132486388384755, "step": 7070}, {"loss": 0.3756, "grad_norm": 1.2265169620513916, "learning_rate": 0.0002, "epoch": 5.139745916515427, "step": 7080}, {"loss": 0.3595, "grad_norm": 1.4430373907089233, "learning_rate": 0.0002, "epoch": 5.147005444646098, "step": 7090}, {"loss": 0.3788, "grad_norm": 1.0821576118469238, "learning_rate": 0.0002, "epoch": 5.15426497277677, "step": 7100}, {"loss": 0.383, "grad_norm": 1.2574739456176758, "learning_rate": 0.0002, "epoch": 5.161524500907441, "step": 7110}, {"loss": 0.3692, "grad_norm": 1.1806069612503052, "learning_rate": 0.0002, "epoch": 5.168784029038113, "step": 7120}, {"loss": 0.3978, "grad_norm": 0.9900956153869629, "learning_rate": 0.0002, "epoch": 5.176043557168784, "step": 7130}, {"loss": 0.4358, "grad_norm": 1.2414425611495972, "learning_rate": 0.0002, "epoch": 5.183303085299456, "step": 7140}, {"loss": 0.3485, "grad_norm": 0.8220699429512024, "learning_rate": 0.0002, "epoch": 5.190562613430127, "step": 7150}, {"loss": 0.3517, "grad_norm": 1.29408860206604, "learning_rate": 0.0002, "epoch": 5.197822141560798, "step": 7160}, {"loss": 0.3405, "grad_norm": 0.8510639071464539, "learning_rate": 0.0002, "epoch": 5.20508166969147, "step": 7170}, {"loss": 0.4233, "grad_norm": 1.3448902368545532, "learning_rate": 0.0002, "epoch": 5.212341197822141, "step": 7180}, {"loss": 0.3808, "grad_norm": 1.054451584815979, "learning_rate": 0.0002, "epoch": 5.219600725952813, "step": 7190}, {"loss": 0.368, "grad_norm": 1.3752713203430176, "learning_rate": 0.0002, "epoch": 5.226860254083484, "step": 7200}, {"loss": 0.3844, "grad_norm": 1.4848095178604126, "learning_rate": 0.0002, "epoch": 5.234119782214156, "step": 7210}, {"loss": 0.4187, "grad_norm": 1.428842544555664, "learning_rate": 0.0002, "epoch": 5.241379310344827, "step": 7220}, {"loss": 0.3778, "grad_norm": 1.1703591346740723, "learning_rate": 0.0002, "epoch": 5.248638838475499, "step": 7230}, {"loss": 0.417, "grad_norm": 1.2180451154708862, "learning_rate": 0.0002, "epoch": 5.2558983666061705, "step": 7240}, {"loss": 0.3656, "grad_norm": 1.094045877456665, "learning_rate": 0.0002, "epoch": 5.2631578947368425, "step": 7250}, {"loss": 0.4331, "grad_norm": 0.9545766115188599, "learning_rate": 0.0002, "epoch": 5.270417422867514, "step": 7260}, {"loss": 0.3642, "grad_norm": 0.8356652855873108, "learning_rate": 0.0002, "epoch": 5.277676950998185, "step": 7270}, {"loss": 0.3576, "grad_norm": 1.148160457611084, "learning_rate": 0.0002, "epoch": 5.284936479128857, "step": 7280}, {"loss": 0.4178, "grad_norm": 1.2009977102279663, "learning_rate": 0.0002, "epoch": 5.292196007259528, "step": 7290}, {"loss": 0.3977, "grad_norm": 1.3283873796463013, "learning_rate": 0.0002, "epoch": 5.2994555353902, "step": 7300}, {"loss": 0.3853, "grad_norm": 0.9850481748580933, "learning_rate": 0.0002, "epoch": 5.306715063520871, "step": 7310}, {"loss": 0.3645, "grad_norm": 1.367550015449524, "learning_rate": 0.0002, "epoch": 5.313974591651543, "step": 7320}, {"loss": 0.3898, "grad_norm": 0.8602936863899231, "learning_rate": 0.0002, "epoch": 5.321234119782214, "step": 7330}, {"loss": 0.4173, "grad_norm": 1.1130679845809937, "learning_rate": 0.0002, "epoch": 5.328493647912886, "step": 7340}, {"loss": 0.3642, "grad_norm": 1.3002253770828247, "learning_rate": 0.0002, "epoch": 5.335753176043557, "step": 7350}, {"loss": 0.4138, "grad_norm": 1.6235289573669434, "learning_rate": 0.0002, "epoch": 5.343012704174229, "step": 7360}, {"loss": 0.4779, "grad_norm": 1.156379222869873, "learning_rate": 0.0002, "epoch": 5.3502722323049, "step": 7370}, {"loss": 0.3222, "grad_norm": 1.0569308996200562, "learning_rate": 0.0002, "epoch": 5.357531760435572, "step": 7380}, {"loss": 0.3573, "grad_norm": 1.6674021482467651, "learning_rate": 0.0002, "epoch": 5.364791288566243, "step": 7390}, {"loss": 0.4325, "grad_norm": 1.2962018251419067, "learning_rate": 0.0002, "epoch": 5.372050816696914, "step": 7400}, {"loss": 0.3809, "grad_norm": 1.1904195547103882, "learning_rate": 0.0002, "epoch": 5.379310344827586, "step": 7410}, {"loss": 0.3728, "grad_norm": 1.316245675086975, "learning_rate": 0.0002, "epoch": 5.386569872958257, "step": 7420}, {"loss": 0.4096, "grad_norm": 1.127570390701294, "learning_rate": 0.0002, "epoch": 5.393829401088929, "step": 7430}, {"loss": 0.3933, "grad_norm": 1.3895777463912964, "learning_rate": 0.0002, "epoch": 5.4010889292196005, "step": 7440}, {"loss": 0.4085, "grad_norm": 1.626830816268921, "learning_rate": 0.0002, "epoch": 5.4083484573502725, "step": 7450}, {"loss": 0.4186, "grad_norm": 1.3703926801681519, "learning_rate": 0.0002, "epoch": 5.415607985480944, "step": 7460}, {"loss": 0.3517, "grad_norm": 1.3854840993881226, "learning_rate": 0.0002, "epoch": 5.422867513611616, "step": 7470}, {"loss": 0.3714, "grad_norm": 1.107065200805664, "learning_rate": 0.0002, "epoch": 5.430127041742287, "step": 7480}, {"loss": 0.3855, "grad_norm": 0.7843456268310547, "learning_rate": 0.0002, "epoch": 5.437386569872959, "step": 7490}, {"loss": 0.4159, "grad_norm": 1.6692372560501099, "learning_rate": 0.0002, "epoch": 5.44464609800363, "step": 7500}, {"loss": 0.4185, "grad_norm": 1.2583858966827393, "learning_rate": 0.0002, "epoch": 5.451905626134302, "step": 7510}, {"loss": 0.4401, "grad_norm": 1.6827000379562378, "learning_rate": 0.0002, "epoch": 5.459165154264973, "step": 7520}, {"loss": 0.397, "grad_norm": 1.6680560111999512, "learning_rate": 0.0002, "epoch": 5.466424682395644, "step": 7530}, {"loss": 0.4193, "grad_norm": 1.3696072101593018, "learning_rate": 0.0002, "epoch": 5.473684210526316, "step": 7540}, {"loss": 0.4244, "grad_norm": 1.4523496627807617, "learning_rate": 0.0002, "epoch": 5.480943738656987, "step": 7550}, {"loss": 0.3609, "grad_norm": 1.3432692289352417, "learning_rate": 0.0002, "epoch": 5.488203266787659, "step": 7560}, {"loss": 0.3675, "grad_norm": 1.363818645477295, "learning_rate": 0.0002, "epoch": 5.49546279491833, "step": 7570}, {"loss": 0.3726, "grad_norm": 1.0176721811294556, "learning_rate": 0.0002, "epoch": 5.502722323049002, "step": 7580}, {"loss": 0.3751, "grad_norm": 1.1625547409057617, "learning_rate": 0.0002, "epoch": 5.509981851179673, "step": 7590}, {"loss": 0.433, "grad_norm": 1.2480388879776, "learning_rate": 0.0002, "epoch": 5.517241379310345, "step": 7600}, {"loss": 0.4511, "grad_norm": 1.341509222984314, "learning_rate": 0.0002, "epoch": 5.524500907441016, "step": 7610}, {"loss": 0.4642, "grad_norm": 1.7048436403274536, "learning_rate": 0.0002, "epoch": 5.531760435571687, "step": 7620}, {"loss": 0.4509, "grad_norm": 1.1435480117797852, "learning_rate": 0.0002, "epoch": 5.539019963702359, "step": 7630}, {"loss": 0.4528, "grad_norm": 1.2381842136383057, "learning_rate": 0.0002, "epoch": 5.5462794918330305, "step": 7640}, {"loss": 0.4496, "grad_norm": 1.50786292552948, "learning_rate": 0.0002, "epoch": 5.5535390199637025, "step": 7650}, {"loss": 0.4242, "grad_norm": 1.2263519763946533, "learning_rate": 0.0002, "epoch": 5.560798548094374, "step": 7660}, {"loss": 0.418, "grad_norm": 1.2864696979522705, "learning_rate": 0.0002, "epoch": 5.568058076225046, "step": 7670}, {"loss": 0.3832, "grad_norm": 1.4443191289901733, "learning_rate": 0.0002, "epoch": 5.575317604355717, "step": 7680}, {"loss": 0.3964, "grad_norm": 1.3360971212387085, "learning_rate": 0.0002, "epoch": 5.582577132486389, "step": 7690}, {"loss": 0.4639, "grad_norm": 1.391828179359436, "learning_rate": 0.0002, "epoch": 5.58983666061706, "step": 7700}, {"loss": 0.4722, "grad_norm": 1.3699384927749634, "learning_rate": 0.0002, "epoch": 5.597096188747732, "step": 7710}, {"loss": 0.4302, "grad_norm": 1.3778468370437622, "learning_rate": 0.0002, "epoch": 5.604355716878403, "step": 7720}, {"loss": 0.4179, "grad_norm": 1.1009501218795776, "learning_rate": 0.0002, "epoch": 5.611615245009075, "step": 7730}, {"loss": 0.4104, "grad_norm": 1.0410021543502808, "learning_rate": 0.0002, "epoch": 5.618874773139746, "step": 7740}, {"loss": 0.4489, "grad_norm": 1.1012226343154907, "learning_rate": 0.0002, "epoch": 5.626134301270417, "step": 7750}, {"loss": 0.4544, "grad_norm": 1.3246384859085083, "learning_rate": 0.0002, "epoch": 5.633393829401089, "step": 7760}, {"loss": 0.4381, "grad_norm": 1.4301716089248657, "learning_rate": 0.0002, "epoch": 5.64065335753176, "step": 7770}, {"loss": 0.4297, "grad_norm": 1.1368978023529053, "learning_rate": 0.0002, "epoch": 5.647912885662432, "step": 7780}, {"loss": 0.4063, "grad_norm": 1.3493064641952515, "learning_rate": 0.0002, "epoch": 5.655172413793103, "step": 7790}, {"loss": 0.4562, "grad_norm": 1.3328721523284912, "learning_rate": 0.0002, "epoch": 5.662431941923775, "step": 7800}, {"loss": 0.4075, "grad_norm": 1.3235671520233154, "learning_rate": 0.0002, "epoch": 5.669691470054446, "step": 7810}, {"loss": 0.4589, "grad_norm": 1.1961841583251953, "learning_rate": 0.0002, "epoch": 5.676950998185118, "step": 7820}, {"loss": 0.4503, "grad_norm": 1.4189636707305908, "learning_rate": 0.0002, "epoch": 5.684210526315789, "step": 7830}, {"loss": 0.4452, "grad_norm": 1.3551312685012817, "learning_rate": 0.0002, "epoch": 5.691470054446461, "step": 7840}, {"loss": 0.4268, "grad_norm": 1.449987769126892, "learning_rate": 0.0002, "epoch": 5.6987295825771325, "step": 7850}, {"loss": 0.4141, "grad_norm": 1.1225156784057617, "learning_rate": 0.0002, "epoch": 5.7059891107078045, "step": 7860}, {"loss": 0.41, "grad_norm": 1.4734594821929932, "learning_rate": 0.0002, "epoch": 5.713248638838476, "step": 7870}, {"loss": 0.4013, "grad_norm": 1.3793359994888306, "learning_rate": 0.0002, "epoch": 5.720508166969147, "step": 7880}, {"loss": 0.4065, "grad_norm": 1.2431834936141968, "learning_rate": 0.0002, "epoch": 5.727767695099819, "step": 7890}, {"loss": 0.4595, "grad_norm": 1.1158313751220703, "learning_rate": 0.0002, "epoch": 5.73502722323049, "step": 7900}, {"loss": 0.4342, "grad_norm": 1.212248682975769, "learning_rate": 0.0002, "epoch": 5.742286751361162, "step": 7910}, {"loss": 0.4611, "grad_norm": 1.5259995460510254, "learning_rate": 0.0002, "epoch": 5.749546279491833, "step": 7920}, {"loss": 0.4483, "grad_norm": 1.3909121751785278, "learning_rate": 0.0002, "epoch": 5.756805807622505, "step": 7930}, {"loss": 0.4325, "grad_norm": 1.2511249780654907, "learning_rate": 0.0002, "epoch": 5.764065335753176, "step": 7940}, {"loss": 0.4048, "grad_norm": 1.2511906623840332, "learning_rate": 0.0002, "epoch": 5.771324863883848, "step": 7950}, {"loss": 0.3715, "grad_norm": 1.1489921808242798, "learning_rate": 0.0002, "epoch": 5.778584392014519, "step": 7960}, {"loss": 0.4196, "grad_norm": 1.028943419456482, "learning_rate": 0.0002, "epoch": 5.78584392014519, "step": 7970}, {"loss": 0.4334, "grad_norm": 1.0820423364639282, "learning_rate": 0.0002, "epoch": 5.793103448275862, "step": 7980}, {"loss": 0.3917, "grad_norm": 1.296520471572876, "learning_rate": 0.0002, "epoch": 5.800362976406533, "step": 7990}, {"loss": 0.4509, "grad_norm": 1.3597749471664429, "learning_rate": 0.0002, "epoch": 5.807622504537205, "step": 8000}, {"loss": 0.4535, "grad_norm": 0.8741790652275085, "learning_rate": 0.0002, "epoch": 5.814882032667876, "step": 8010}, {"loss": 0.4239, "grad_norm": 1.1471822261810303, "learning_rate": 0.0002, "epoch": 5.822141560798548, "step": 8020}, {"loss": 0.5042, "grad_norm": 1.2997334003448486, "learning_rate": 0.0002, "epoch": 5.829401088929219, "step": 8030}, {"loss": 0.4758, "grad_norm": 1.1027175188064575, "learning_rate": 0.0002, "epoch": 5.836660617059891, "step": 8040}, {"loss": 0.4192, "grad_norm": 1.2695307731628418, "learning_rate": 0.0002, "epoch": 5.8439201451905625, "step": 8050}, {"loss": 0.5173, "grad_norm": 1.5275461673736572, "learning_rate": 0.0002, "epoch": 5.8511796733212345, "step": 8060}, {"loss": 0.5012, "grad_norm": 1.3059501647949219, "learning_rate": 0.0002, "epoch": 5.8584392014519056, "step": 8070}, {"loss": 0.4425, "grad_norm": 1.57442045211792, "learning_rate": 0.0002, "epoch": 5.8656987295825775, "step": 8080}, {"loss": 0.4261, "grad_norm": 1.119564414024353, "learning_rate": 0.0002, "epoch": 5.872958257713249, "step": 8090}, {"loss": 0.465, "grad_norm": 1.6517373323440552, "learning_rate": 0.0002, "epoch": 5.88021778584392, "step": 8100}, {"loss": 0.4406, "grad_norm": 1.4093554019927979, "learning_rate": 0.0002, "epoch": 5.887477313974592, "step": 8110}, {"loss": 0.4433, "grad_norm": 1.278843641281128, "learning_rate": 0.0002, "epoch": 5.894736842105263, "step": 8120}, {"loss": 0.4007, "grad_norm": 1.2042944431304932, "learning_rate": 0.0002, "epoch": 5.901996370235935, "step": 8130}, {"loss": 0.3972, "grad_norm": 1.1788326501846313, "learning_rate": 0.0002, "epoch": 5.909255898366606, "step": 8140}, {"loss": 0.4506, "grad_norm": 1.4364569187164307, "learning_rate": 0.0002, "epoch": 5.916515426497278, "step": 8150}, {"loss": 0.4651, "grad_norm": 1.1704283952713013, "learning_rate": 0.0002, "epoch": 5.923774954627949, "step": 8160}, {"loss": 0.3972, "grad_norm": 1.040814995765686, "learning_rate": 0.0002, "epoch": 5.931034482758621, "step": 8170}, {"loss": 0.4038, "grad_norm": 1.1367416381835938, "learning_rate": 0.0002, "epoch": 5.938294010889292, "step": 8180}, {"loss": 0.4387, "grad_norm": 1.3401511907577515, "learning_rate": 0.0002, "epoch": 5.945553539019964, "step": 8190}, {"loss": 0.4396, "grad_norm": 1.1154041290283203, "learning_rate": 0.0002, "epoch": 5.952813067150635, "step": 8200}, {"loss": 0.4744, "grad_norm": 1.426089882850647, "learning_rate": 0.0002, "epoch": 5.960072595281307, "step": 8210}, {"loss": 0.4105, "grad_norm": 1.3170222043991089, "learning_rate": 0.0002, "epoch": 5.967332123411978, "step": 8220}, {"loss": 0.4137, "grad_norm": 1.1960029602050781, "learning_rate": 0.0002, "epoch": 5.974591651542649, "step": 8230}, {"loss": 0.423, "grad_norm": 1.0843931436538696, "learning_rate": 0.0002, "epoch": 5.981851179673321, "step": 8240}, {"loss": 0.459, "grad_norm": 1.050421118736267, "learning_rate": 0.0002, "epoch": 5.9891107078039925, "step": 8250}, {"loss": 0.3993, "grad_norm": 1.0183138847351074, "learning_rate": 0.0002, "epoch": 5.9963702359346644, "step": 8260}, {"eval_loss": 1.4677470922470093, "eval_runtime": 46.2504, "eval_samples_per_second": 9.427, "eval_steps_per_second": 1.189, "epoch": 6.0, "step": 8265}, {"loss": 0.3947, "grad_norm": 1.1702998876571655, "learning_rate": 0.0002, "epoch": 6.0036297640653356, "step": 8270}, {"loss": 0.2854, "grad_norm": 1.5389727354049683, "learning_rate": 0.0002, "epoch": 6.0108892921960075, "step": 8280}, {"loss": 0.2603, "grad_norm": 1.502568244934082, "learning_rate": 0.0002, "epoch": 6.018148820326679, "step": 8290}, {"loss": 0.3329, "grad_norm": 1.3846043348312378, "learning_rate": 0.0002, "epoch": 6.025408348457351, "step": 8300}, {"loss": 0.2651, "grad_norm": 1.173553228378296, "learning_rate": 0.0002, "epoch": 6.032667876588022, "step": 8310}, {"loss": 0.3142, "grad_norm": 1.5325932502746582, "learning_rate": 0.0002, "epoch": 6.039927404718694, "step": 8320}, {"loss": 0.2511, "grad_norm": 1.303783655166626, "learning_rate": 0.0002, "epoch": 6.047186932849365, "step": 8330}, {"loss": 0.2352, "grad_norm": 0.9408994913101196, "learning_rate": 0.0002, "epoch": 6.054446460980036, "step": 8340}, {"loss": 0.2548, "grad_norm": 1.5430388450622559, "learning_rate": 0.0002, "epoch": 6.061705989110708, "step": 8350}, {"loss": 0.2682, "grad_norm": 0.8765342235565186, "learning_rate": 0.0002, "epoch": 6.068965517241379, "step": 8360}, {"loss": 0.2614, "grad_norm": 1.2363157272338867, "learning_rate": 0.0002, "epoch": 6.076225045372051, "step": 8370}, {"loss": 0.294, "grad_norm": 1.21284818649292, "learning_rate": 0.0002, "epoch": 6.083484573502722, "step": 8380}, {"loss": 0.2498, "grad_norm": 1.3261712789535522, "learning_rate": 0.0002, "epoch": 6.090744101633394, "step": 8390}, {"loss": 0.2649, "grad_norm": 1.077317714691162, "learning_rate": 0.0002, "epoch": 6.098003629764065, "step": 8400}, {"loss": 0.269, "grad_norm": 0.9873808026313782, "learning_rate": 0.0002, "epoch": 6.105263157894737, "step": 8410}, {"loss": 0.2736, "grad_norm": 1.032258152961731, "learning_rate": 0.0002, "epoch": 6.112522686025408, "step": 8420}, {"loss": 0.2854, "grad_norm": 1.1014811992645264, "learning_rate": 0.0002, "epoch": 6.11978221415608, "step": 8430}, {"loss": 0.2924, "grad_norm": 1.4264203310012817, "learning_rate": 0.0002, "epoch": 6.127041742286751, "step": 8440}, {"loss": 0.3388, "grad_norm": 1.4086531400680542, "learning_rate": 0.0002, "epoch": 6.1343012704174225, "step": 8450}, {"loss": 0.2786, "grad_norm": 1.3842453956604004, "learning_rate": 0.0002, "epoch": 6.1415607985480944, "step": 8460}, {"loss": 0.3201, "grad_norm": 1.4356757402420044, "learning_rate": 0.0002, "epoch": 6.1488203266787655, "step": 8470}, {"loss": 0.2908, "grad_norm": 1.193315029144287, "learning_rate": 0.0002, "epoch": 6.1560798548094375, "step": 8480}, {"loss": 0.342, "grad_norm": 1.0623924732208252, "learning_rate": 0.0002, "epoch": 6.163339382940109, "step": 8490}, {"loss": 0.3257, "grad_norm": 1.5484434366226196, "learning_rate": 0.0002, "epoch": 6.170598911070781, "step": 8500}, {"loss": 0.2861, "grad_norm": 1.3520029783248901, "learning_rate": 0.0002, "epoch": 6.177858439201452, "step": 8510}, {"loss": 0.3242, "grad_norm": 1.2773103713989258, "learning_rate": 0.0002, "epoch": 6.185117967332124, "step": 8520}, {"loss": 0.3108, "grad_norm": 1.4675105810165405, "learning_rate": 0.0002, "epoch": 6.192377495462795, "step": 8530}, {"loss": 0.3044, "grad_norm": 1.2118732929229736, "learning_rate": 0.0002, "epoch": 6.199637023593467, "step": 8540}, {"loss": 0.2726, "grad_norm": 1.264024257659912, "learning_rate": 0.0002, "epoch": 6.206896551724138, "step": 8550}, {"loss": 0.306, "grad_norm": 1.406931757926941, "learning_rate": 0.0002, "epoch": 6.21415607985481, "step": 8560}, {"loss": 0.2904, "grad_norm": 1.385459542274475, "learning_rate": 0.0002, "epoch": 6.221415607985481, "step": 8570}, {"loss": 0.3413, "grad_norm": 1.9336168766021729, "learning_rate": 0.0002, "epoch": 6.228675136116152, "step": 8580}, {"loss": 0.2769, "grad_norm": 0.9880136847496033, "learning_rate": 0.0002, "epoch": 6.235934664246824, "step": 8590}, {"loss": 0.3035, "grad_norm": 1.3870339393615723, "learning_rate": 0.0002, "epoch": 6.243194192377495, "step": 8600}, {"loss": 0.286, "grad_norm": 1.2303647994995117, "learning_rate": 0.0002, "epoch": 6.250453720508167, "step": 8610}, {"loss": 0.3113, "grad_norm": 1.5406211614608765, "learning_rate": 0.0002, "epoch": 6.257713248638838, "step": 8620}, {"loss": 0.292, "grad_norm": 1.2436790466308594, "learning_rate": 0.0002, "epoch": 6.26497277676951, "step": 8630}, {"loss": 0.3102, "grad_norm": 0.8844212293624878, "learning_rate": 0.0002, "epoch": 6.272232304900181, "step": 8640}, {"loss": 0.3373, "grad_norm": 1.2846336364746094, "learning_rate": 0.0002, "epoch": 6.279491833030853, "step": 8650}, {"loss": 0.3535, "grad_norm": 1.593814730644226, "learning_rate": 0.0002, "epoch": 6.286751361161524, "step": 8660}, {"loss": 0.3413, "grad_norm": 1.2277469635009766, "learning_rate": 0.0002, "epoch": 6.2940108892921955, "step": 8670}, {"loss": 0.2958, "grad_norm": 1.2574384212493896, "learning_rate": 0.0002, "epoch": 6.3012704174228675, "step": 8680}, {"loss": 0.3251, "grad_norm": 1.335150957107544, "learning_rate": 0.0002, "epoch": 6.308529945553539, "step": 8690}, {"loss": 0.3009, "grad_norm": 1.3140437602996826, "learning_rate": 0.0002, "epoch": 6.315789473684211, "step": 8700}, {"loss": 0.2783, "grad_norm": 1.1689209938049316, "learning_rate": 0.0002, "epoch": 6.323049001814882, "step": 8710}, {"loss": 0.3476, "grad_norm": 1.6448503732681274, "learning_rate": 0.0002, "epoch": 6.330308529945554, "step": 8720}, {"loss": 0.2934, "grad_norm": 0.9944145679473877, "learning_rate": 0.0002, "epoch": 6.337568058076225, "step": 8730}, {"loss": 0.3315, "grad_norm": 1.1775634288787842, "learning_rate": 0.0002, "epoch": 6.344827586206897, "step": 8740}, {"loss": 0.3514, "grad_norm": 1.8438639640808105, "learning_rate": 0.0002, "epoch": 6.352087114337568, "step": 8750}, {"loss": 0.303, "grad_norm": 1.062495470046997, "learning_rate": 0.0002, "epoch": 6.35934664246824, "step": 8760}, {"loss": 0.2737, "grad_norm": 1.3224315643310547, "learning_rate": 0.0002, "epoch": 6.366606170598911, "step": 8770}, {"loss": 0.3445, "grad_norm": 1.399844765663147, "learning_rate": 0.0002, "epoch": 6.373865698729583, "step": 8780}, {"loss": 0.3277, "grad_norm": 1.0409915447235107, "learning_rate": 0.0002, "epoch": 6.381125226860254, "step": 8790}, {"loss": 0.3218, "grad_norm": 1.5657726526260376, "learning_rate": 0.0002, "epoch": 6.388384754990925, "step": 8800}, {"loss": 0.3031, "grad_norm": 1.4098644256591797, "learning_rate": 0.0002, "epoch": 6.395644283121597, "step": 8810}, {"loss": 0.3133, "grad_norm": 1.5154732465744019, "learning_rate": 0.0002, "epoch": 6.402903811252268, "step": 8820}, {"loss": 0.3111, "grad_norm": 1.1139698028564453, "learning_rate": 0.0002, "epoch": 6.41016333938294, "step": 8830}, {"loss": 0.3553, "grad_norm": 1.4149729013442993, "learning_rate": 0.0002, "epoch": 6.417422867513611, "step": 8840}, {"loss": 0.287, "grad_norm": 1.2632299661636353, "learning_rate": 0.0002, "epoch": 6.424682395644283, "step": 8850}, {"loss": 0.3198, "grad_norm": 1.6636109352111816, "learning_rate": 0.0002, "epoch": 6.431941923774954, "step": 8860}, {"loss": 0.3749, "grad_norm": 1.4149386882781982, "learning_rate": 0.0002, "epoch": 6.439201451905626, "step": 8870}, {"loss": 0.3504, "grad_norm": 1.1396206617355347, "learning_rate": 0.0002, "epoch": 6.4464609800362975, "step": 8880}, {"loss": 0.3328, "grad_norm": 1.2188775539398193, "learning_rate": 0.0002, "epoch": 6.4537205081669695, "step": 8890}, {"loss": 0.3427, "grad_norm": 0.9740369319915771, "learning_rate": 0.0002, "epoch": 6.460980036297641, "step": 8900}, {"loss": 0.3223, "grad_norm": 1.228569746017456, "learning_rate": 0.0002, "epoch": 6.468239564428313, "step": 8910}, {"loss": 0.3151, "grad_norm": 1.5019789934158325, "learning_rate": 0.0002, "epoch": 6.475499092558984, "step": 8920}, {"loss": 0.2916, "grad_norm": 1.3320101499557495, "learning_rate": 0.0002, "epoch": 6.482758620689655, "step": 8930}, {"loss": 0.298, "grad_norm": 1.5551502704620361, "learning_rate": 0.0002, "epoch": 6.490018148820327, "step": 8940}, {"loss": 0.3238, "grad_norm": 1.470131754875183, "learning_rate": 0.0002, "epoch": 6.497277676950998, "step": 8950}, {"loss": 0.2808, "grad_norm": 1.1803025007247925, "learning_rate": 0.0002, "epoch": 6.50453720508167, "step": 8960}, {"loss": 0.3025, "grad_norm": 1.3505640029907227, "learning_rate": 0.0002, "epoch": 6.511796733212341, "step": 8970}, {"loss": 0.3124, "grad_norm": 1.13093900680542, "learning_rate": 0.0002, "epoch": 6.519056261343013, "step": 8980}, {"loss": 0.3454, "grad_norm": 1.347386121749878, "learning_rate": 0.0002, "epoch": 6.526315789473684, "step": 8990}, {"loss": 0.3532, "grad_norm": 1.7879165410995483, "learning_rate": 0.0002, "epoch": 6.533575317604356, "step": 9000}, {"loss": 0.3382, "grad_norm": 1.2168169021606445, "learning_rate": 0.0002, "epoch": 6.540834845735027, "step": 9010}, {"loss": 0.3413, "grad_norm": 1.1758877038955688, "learning_rate": 0.0002, "epoch": 6.548094373865698, "step": 9020}, {"loss": 0.2806, "grad_norm": 1.7366445064544678, "learning_rate": 0.0002, "epoch": 6.55535390199637, "step": 9030}, {"loss": 0.3437, "grad_norm": 1.5919222831726074, "learning_rate": 0.0002, "epoch": 6.562613430127042, "step": 9040}, {"loss": 0.3261, "grad_norm": 1.336863398551941, "learning_rate": 0.0002, "epoch": 6.569872958257713, "step": 9050}, {"loss": 0.3103, "grad_norm": 1.1769421100616455, "learning_rate": 0.0002, "epoch": 6.577132486388384, "step": 9060}, {"loss": 0.3295, "grad_norm": 1.0048751831054688, "learning_rate": 0.0002, "epoch": 6.584392014519056, "step": 9070}, {"loss": 0.3156, "grad_norm": 1.5268515348434448, "learning_rate": 0.0002, "epoch": 6.5916515426497275, "step": 9080}, {"loss": 0.3752, "grad_norm": 1.434610366821289, "learning_rate": 0.0002, "epoch": 6.5989110707803995, "step": 9090}, {"loss": 0.3375, "grad_norm": 1.1151410341262817, "learning_rate": 0.0002, "epoch": 6.606170598911071, "step": 9100}, {"loss": 0.363, "grad_norm": 1.6690642833709717, "learning_rate": 0.0002, "epoch": 6.613430127041743, "step": 9110}, {"loss": 0.3703, "grad_norm": 1.4495552778244019, "learning_rate": 0.0002, "epoch": 6.620689655172414, "step": 9120}, {"loss": 0.3648, "grad_norm": 1.377621054649353, "learning_rate": 0.0002, "epoch": 6.627949183303086, "step": 9130}, {"loss": 0.3766, "grad_norm": 1.5459434986114502, "learning_rate": 0.0002, "epoch": 6.635208711433757, "step": 9140}, {"loss": 0.3196, "grad_norm": 1.0920850038528442, "learning_rate": 0.0002, "epoch": 6.642468239564428, "step": 9150}, {"loss": 0.3505, "grad_norm": 1.6708381175994873, "learning_rate": 0.0002, "epoch": 6.6497277676951, "step": 9160}, {"loss": 0.368, "grad_norm": 1.747514009475708, "learning_rate": 0.0002, "epoch": 6.656987295825771, "step": 9170}, {"loss": 0.3099, "grad_norm": 1.133466362953186, "learning_rate": 0.0002, "epoch": 6.664246823956443, "step": 9180}, {"loss": 0.3175, "grad_norm": 1.394358515739441, "learning_rate": 0.0002, "epoch": 6.671506352087114, "step": 9190}, {"loss": 0.2981, "grad_norm": 0.9258374571800232, "learning_rate": 0.0002, "epoch": 6.678765880217786, "step": 9200}, {"loss": 0.3723, "grad_norm": 1.3750739097595215, "learning_rate": 0.0002, "epoch": 6.686025408348457, "step": 9210}, {"loss": 0.3441, "grad_norm": 0.8604967594146729, "learning_rate": 0.0002, "epoch": 6.693284936479129, "step": 9220}, {"loss": 0.3775, "grad_norm": 1.6074559688568115, "learning_rate": 0.0002, "epoch": 6.7005444646098, "step": 9230}, {"loss": 0.3139, "grad_norm": 0.9576877355575562, "learning_rate": 0.0002, "epoch": 6.707803992740472, "step": 9240}, {"loss": 0.3633, "grad_norm": 1.7193048000335693, "learning_rate": 0.0002, "epoch": 6.715063520871143, "step": 9250}, {"loss": 0.3139, "grad_norm": 1.3131844997406006, "learning_rate": 0.0002, "epoch": 6.722323049001815, "step": 9260}, {"loss": 0.3121, "grad_norm": 1.2978184223175049, "learning_rate": 0.0002, "epoch": 6.729582577132486, "step": 9270}, {"loss": 0.3534, "grad_norm": 1.4792617559432983, "learning_rate": 0.0002, "epoch": 6.7368421052631575, "step": 9280}, {"loss": 0.3429, "grad_norm": 1.1265567541122437, "learning_rate": 0.0002, "epoch": 6.7441016333938295, "step": 9290}, {"loss": 0.3526, "grad_norm": 1.8553377389907837, "learning_rate": 0.0002, "epoch": 6.751361161524501, "step": 9300}, {"loss": 0.3666, "grad_norm": 1.3602519035339355, "learning_rate": 0.0002, "epoch": 6.758620689655173, "step": 9310}, {"loss": 0.2922, "grad_norm": 1.2874794006347656, "learning_rate": 0.0002, "epoch": 6.765880217785844, "step": 9320}, {"loss": 0.3816, "grad_norm": 1.4834712743759155, "learning_rate": 0.0002, "epoch": 6.773139745916516, "step": 9330}, {"loss": 0.3557, "grad_norm": 2.0824034214019775, "learning_rate": 0.0002, "epoch": 6.780399274047187, "step": 9340}, {"loss": 0.3174, "grad_norm": 1.2267698049545288, "learning_rate": 0.0002, "epoch": 6.787658802177859, "step": 9350}, {"loss": 0.3665, "grad_norm": 1.4485498666763306, "learning_rate": 0.0002, "epoch": 6.79491833030853, "step": 9360}, {"loss": 0.3676, "grad_norm": 1.3199396133422852, "learning_rate": 0.0002, "epoch": 6.802177858439201, "step": 9370}, {"loss": 0.298, "grad_norm": 1.2552456855773926, "learning_rate": 0.0002, "epoch": 6.809437386569873, "step": 9380}, {"loss": 0.3152, "grad_norm": 1.3895127773284912, "learning_rate": 0.0002, "epoch": 6.816696914700545, "step": 9390}, {"loss": 0.3375, "grad_norm": 1.7637823820114136, "learning_rate": 0.0002, "epoch": 6.823956442831216, "step": 9400}, {"loss": 0.3234, "grad_norm": 1.6004475355148315, "learning_rate": 0.0002, "epoch": 6.831215970961887, "step": 9410}, {"loss": 0.3364, "grad_norm": 1.4133695363998413, "learning_rate": 0.0002, "epoch": 6.838475499092559, "step": 9420}, {"loss": 0.3656, "grad_norm": 1.1583502292633057, "learning_rate": 0.0002, "epoch": 6.84573502722323, "step": 9430}, {"loss": 0.3499, "grad_norm": 1.3769075870513916, "learning_rate": 0.0002, "epoch": 6.852994555353902, "step": 9440}, {"loss": 0.3333, "grad_norm": 1.1831218004226685, "learning_rate": 0.0002, "epoch": 6.860254083484573, "step": 9450}, {"loss": 0.3501, "grad_norm": 1.6092621088027954, "learning_rate": 0.0002, "epoch": 6.867513611615245, "step": 9460}, {"loss": 0.3933, "grad_norm": 1.3850210905075073, "learning_rate": 0.0002, "epoch": 6.874773139745916, "step": 9470}, {"loss": 0.3868, "grad_norm": 1.4119619131088257, "learning_rate": 0.0002, "epoch": 6.882032667876588, "step": 9480}, {"loss": 0.3939, "grad_norm": 1.3494242429733276, "learning_rate": 0.0002, "epoch": 6.8892921960072595, "step": 9490}, {"loss": 0.3217, "grad_norm": 1.3130041360855103, "learning_rate": 0.0002, "epoch": 6.896551724137931, "step": 9500}, {"loss": 0.3738, "grad_norm": 1.169256329536438, "learning_rate": 0.0002, "epoch": 6.903811252268603, "step": 9510}, {"loss": 0.408, "grad_norm": 1.7475035190582275, "learning_rate": 0.0002, "epoch": 6.911070780399274, "step": 9520}, {"loss": 0.3407, "grad_norm": 1.440434217453003, "learning_rate": 0.0002, "epoch": 6.918330308529946, "step": 9530}, {"loss": 0.3707, "grad_norm": 1.6768704652786255, "learning_rate": 0.0002, "epoch": 6.925589836660617, "step": 9540}, {"loss": 0.3283, "grad_norm": 1.3720577955245972, "learning_rate": 0.0002, "epoch": 6.932849364791289, "step": 9550}, {"loss": 0.3257, "grad_norm": 1.8140523433685303, "learning_rate": 0.0002, "epoch": 6.94010889292196, "step": 9560}, {"loss": 0.3308, "grad_norm": 1.1828241348266602, "learning_rate": 0.0002, "epoch": 6.947368421052632, "step": 9570}, {"loss": 0.3536, "grad_norm": 1.2755135297775269, "learning_rate": 0.0002, "epoch": 6.954627949183303, "step": 9580}, {"loss": 0.3711, "grad_norm": 1.622009038925171, "learning_rate": 0.0002, "epoch": 6.961887477313975, "step": 9590}, {"loss": 0.3529, "grad_norm": 1.1543664932250977, "learning_rate": 0.0002, "epoch": 6.969147005444646, "step": 9600}, {"loss": 0.416, "grad_norm": 1.6755319833755493, "learning_rate": 0.0002, "epoch": 6.976406533575318, "step": 9610}, {"loss": 0.339, "grad_norm": 1.3726437091827393, "learning_rate": 0.0002, "epoch": 6.983666061705989, "step": 9620}, {"loss": 0.3709, "grad_norm": 1.1605958938598633, "learning_rate": 0.0002, "epoch": 6.99092558983666, "step": 9630}, {"loss": 0.371, "grad_norm": 1.5371781587600708, "learning_rate": 0.0002, "epoch": 6.998185117967332, "step": 9640}, {"eval_loss": 1.6280181407928467, "eval_runtime": 46.1964, "eval_samples_per_second": 9.438, "eval_steps_per_second": 1.191, "epoch": 6.999637023593467, "step": 9642}, {"loss": 0.2449, "grad_norm": 1.1645569801330566, "learning_rate": 0.0002, "epoch": 7.005444646098003, "step": 9650}, {"loss": 0.251, "grad_norm": 0.7663792967796326, "learning_rate": 0.0002, "epoch": 7.012704174228675, "step": 9660}, {"loss": 0.2553, "grad_norm": 1.5808782577514648, "learning_rate": 0.0002, "epoch": 7.019963702359346, "step": 9670}, {"loss": 0.2349, "grad_norm": 1.046607255935669, "learning_rate": 0.0002, "epoch": 7.027223230490018, "step": 9680}, {"loss": 0.2078, "grad_norm": 1.2008668184280396, "learning_rate": 0.0002, "epoch": 7.0344827586206895, "step": 9690}, {"loss": 0.2519, "grad_norm": 1.9596126079559326, "learning_rate": 0.0002, "epoch": 7.0417422867513615, "step": 9700}, {"loss": 0.2275, "grad_norm": 1.0400182008743286, "learning_rate": 0.0002, "epoch": 7.049001814882033, "step": 9710}, {"loss": 0.2136, "grad_norm": 1.3162504434585571, "learning_rate": 0.0002, "epoch": 7.056261343012705, "step": 9720}, {"loss": 0.2227, "grad_norm": 1.180074691772461, "learning_rate": 0.0002, "epoch": 7.063520871143376, "step": 9730}, {"loss": 0.2206, "grad_norm": 1.2093719244003296, "learning_rate": 0.0002, "epoch": 7.070780399274047, "step": 9740}, {"loss": 0.2387, "grad_norm": 1.4838900566101074, "learning_rate": 0.0002, "epoch": 7.078039927404719, "step": 9750}, {"loss": 0.2391, "grad_norm": 1.2319235801696777, "learning_rate": 0.0002, "epoch": 7.08529945553539, "step": 9760}, {"loss": 0.2624, "grad_norm": 1.2346558570861816, "learning_rate": 0.0002, "epoch": 7.092558983666062, "step": 9770}, {"loss": 0.2321, "grad_norm": 1.0748975276947021, "learning_rate": 0.0002, "epoch": 7.099818511796733, "step": 9780}, {"loss": 0.2751, "grad_norm": 1.0162630081176758, "learning_rate": 0.0002, "epoch": 7.107078039927405, "step": 9790}, {"loss": 0.2142, "grad_norm": 1.0014166831970215, "learning_rate": 0.0002, "epoch": 7.114337568058076, "step": 9800}, {"loss": 0.2439, "grad_norm": 1.0928411483764648, "learning_rate": 0.0002, "epoch": 7.121597096188748, "step": 9810}, {"loss": 0.2261, "grad_norm": 1.181496500968933, "learning_rate": 0.0002, "epoch": 7.128856624319419, "step": 9820}, {"loss": 0.2345, "grad_norm": 1.5846176147460938, "learning_rate": 0.0002, "epoch": 7.136116152450091, "step": 9830}, {"loss": 0.2282, "grad_norm": 0.8734912872314453, "learning_rate": 0.0002, "epoch": 7.143375680580762, "step": 9840}, {"loss": 0.2127, "grad_norm": 1.1599528789520264, "learning_rate": 0.0002, "epoch": 7.150635208711433, "step": 9850}, {"loss": 0.2614, "grad_norm": 1.168256402015686, "learning_rate": 0.0002, "epoch": 7.157894736842105, "step": 9860}, {"loss": 0.2418, "grad_norm": 1.4439860582351685, "learning_rate": 0.0002, "epoch": 7.165154264972776, "step": 9870}, {"loss": 0.2183, "grad_norm": 1.3615007400512695, "learning_rate": 0.0002, "epoch": 7.172413793103448, "step": 9880}, {"loss": 0.2366, "grad_norm": 1.1908115148544312, "learning_rate": 0.0002, "epoch": 7.1796733212341195, "step": 9890}, {"loss": 0.2338, "grad_norm": 1.452515959739685, "learning_rate": 0.0002, "epoch": 7.1869328493647915, "step": 9900}, {"loss": 0.2203, "grad_norm": 0.8387667536735535, "learning_rate": 0.0002, "epoch": 7.194192377495463, "step": 9910}, {"loss": 0.2117, "grad_norm": 1.3990435600280762, "learning_rate": 0.0002, "epoch": 7.201451905626135, "step": 9920}, {"loss": 0.2188, "grad_norm": 1.057800531387329, "learning_rate": 0.0002, "epoch": 7.208711433756806, "step": 9930}, {"loss": 0.2516, "grad_norm": 1.3718253374099731, "learning_rate": 0.0002, "epoch": 7.215970961887478, "step": 9940}, {"loss": 0.2084, "grad_norm": 1.2011432647705078, "learning_rate": 0.0002, "epoch": 7.223230490018149, "step": 9950}, {"loss": 0.2063, "grad_norm": 1.1608737707138062, "learning_rate": 0.0002, "epoch": 7.230490018148821, "step": 9960}, {"loss": 0.2275, "grad_norm": 1.7522791624069214, "learning_rate": 0.0002, "epoch": 7.237749546279492, "step": 9970}, {"loss": 0.2358, "grad_norm": 1.0787912607192993, "learning_rate": 0.0002, "epoch": 7.245009074410163, "step": 9980}, {"loss": 0.2361, "grad_norm": 1.8227689266204834, "learning_rate": 0.0002, "epoch": 7.252268602540835, "step": 9990}, {"loss": 0.2217, "grad_norm": 1.1438913345336914, "learning_rate": 0.0002, "epoch": 7.259528130671506, "step": 10000}, {"loss": 0.2438, "grad_norm": 1.331770420074463, "learning_rate": 0.0002, "epoch": 7.266787658802178, "step": 10010}, {"loss": 0.2622, "grad_norm": 1.2809056043624878, "learning_rate": 0.0002, "epoch": 7.274047186932849, "step": 10020}, {"loss": 0.2634, "grad_norm": 1.2245303392410278, "learning_rate": 0.0002, "epoch": 7.281306715063521, "step": 10030}, {"loss": 0.2706, "grad_norm": 1.2359435558319092, "learning_rate": 0.0002, "epoch": 7.288566243194192, "step": 10040}, {"loss": 0.2427, "grad_norm": 1.3707170486450195, "learning_rate": 0.0002, "epoch": 7.295825771324864, "step": 10050}, {"loss": 0.2497, "grad_norm": 1.7405836582183838, "learning_rate": 0.0002, "epoch": 7.303085299455535, "step": 10060}, {"loss": 0.2628, "grad_norm": 1.446069359779358, "learning_rate": 0.0002, "epoch": 7.310344827586207, "step": 10070}, {"loss": 0.2457, "grad_norm": 1.48823082447052, "learning_rate": 0.0002, "epoch": 7.317604355716878, "step": 10080}, {"loss": 0.2391, "grad_norm": 1.1720311641693115, "learning_rate": 0.0002, "epoch": 7.3248638838475495, "step": 10090}, {"loss": 0.2701, "grad_norm": 1.5485225915908813, "learning_rate": 0.0002, "epoch": 7.3321234119782215, "step": 10100}, {"loss": 0.2855, "grad_norm": 1.6018894910812378, "learning_rate": 0.0002, "epoch": 7.339382940108893, "step": 10110}, {"loss": 0.2662, "grad_norm": 1.4753694534301758, "learning_rate": 0.0002, "epoch": 7.346642468239565, "step": 10120}, {"loss": 0.2323, "grad_norm": 1.3604710102081299, "learning_rate": 0.0002, "epoch": 7.353901996370236, "step": 10130}, {"loss": 0.2664, "grad_norm": 1.5755873918533325, "learning_rate": 0.0002, "epoch": 7.361161524500908, "step": 10140}, {"loss": 0.215, "grad_norm": 0.9421692490577698, "learning_rate": 0.0002, "epoch": 7.368421052631579, "step": 10150}, {"loss": 0.297, "grad_norm": 1.3055956363677979, "learning_rate": 0.0002, "epoch": 7.375680580762251, "step": 10160}, {"loss": 0.2385, "grad_norm": 1.4764302968978882, "learning_rate": 0.0002, "epoch": 7.382940108892922, "step": 10170}, {"loss": 0.2724, "grad_norm": 1.3726946115493774, "learning_rate": 0.0002, "epoch": 7.390199637023594, "step": 10180}, {"loss": 0.2599, "grad_norm": 1.446473240852356, "learning_rate": 0.0002, "epoch": 7.397459165154265, "step": 10190}, {"loss": 0.2837, "grad_norm": 1.489094614982605, "learning_rate": 0.0002, "epoch": 7.404718693284936, "step": 10200}, {"loss": 0.2786, "grad_norm": 1.247572898864746, "learning_rate": 0.0002, "epoch": 7.411978221415608, "step": 10210}, {"loss": 0.2498, "grad_norm": 1.2741918563842773, "learning_rate": 0.0002, "epoch": 7.419237749546279, "step": 10220}, {"loss": 0.2649, "grad_norm": 1.0347636938095093, "learning_rate": 0.0002, "epoch": 7.426497277676951, "step": 10230}, {"loss": 0.2419, "grad_norm": 1.3295499086380005, "learning_rate": 0.0002, "epoch": 7.433756805807622, "step": 10240}, {"loss": 0.226, "grad_norm": 1.6056840419769287, "learning_rate": 0.0002, "epoch": 7.441016333938294, "step": 10250}, {"loss": 0.2571, "grad_norm": 1.4824398756027222, "learning_rate": 0.0002, "epoch": 7.448275862068965, "step": 10260}, {"loss": 0.2939, "grad_norm": 1.6259359121322632, "learning_rate": 0.0002, "epoch": 7.455535390199637, "step": 10270}, {"loss": 0.2873, "grad_norm": 1.5065499544143677, "learning_rate": 0.0002, "epoch": 7.462794918330308, "step": 10280}, {"loss": 0.2614, "grad_norm": 1.3505364656448364, "learning_rate": 0.0002, "epoch": 7.47005444646098, "step": 10290}, {"loss": 0.2862, "grad_norm": 1.4457359313964844, "learning_rate": 0.0002, "epoch": 7.4773139745916515, "step": 10300}, {"loss": 0.233, "grad_norm": 1.0782662630081177, "learning_rate": 0.0002, "epoch": 7.4845735027223235, "step": 10310}, {"loss": 0.2506, "grad_norm": 1.4209016561508179, "learning_rate": 0.0002, "epoch": 7.491833030852995, "step": 10320}, {"loss": 0.2741, "grad_norm": 1.4511336088180542, "learning_rate": 0.0002, "epoch": 7.499092558983666, "step": 10330}, {"loss": 0.2736, "grad_norm": 1.306691288948059, "learning_rate": 0.0002, "epoch": 7.506352087114338, "step": 10340}, {"loss": 0.2528, "grad_norm": 1.0647870302200317, "learning_rate": 0.0002, "epoch": 7.513611615245009, "step": 10350}, {"loss": 0.2402, "grad_norm": 1.0374330282211304, "learning_rate": 0.0002, "epoch": 7.520871143375681, "step": 10360}, {"loss": 0.3057, "grad_norm": 0.8428803086280823, "learning_rate": 0.0002, "epoch": 7.528130671506352, "step": 10370}, {"loss": 0.2446, "grad_norm": 1.3868707418441772, "learning_rate": 0.0002, "epoch": 7.535390199637024, "step": 10380}, {"loss": 0.2851, "grad_norm": 1.4324088096618652, "learning_rate": 0.0002, "epoch": 7.542649727767695, "step": 10390}, {"loss": 0.292, "grad_norm": 1.6413776874542236, "learning_rate": 0.0002, "epoch": 7.549909255898367, "step": 10400}, {"loss": 0.2585, "grad_norm": 1.4302188158035278, "learning_rate": 0.0002, "epoch": 7.557168784029038, "step": 10410}, {"loss": 0.306, "grad_norm": 1.3648983240127563, "learning_rate": 0.0002, "epoch": 7.564428312159709, "step": 10420}, {"loss": 0.2927, "grad_norm": 1.4480061531066895, "learning_rate": 0.0002, "epoch": 7.571687840290381, "step": 10430}, {"loss": 0.2711, "grad_norm": 1.0944541692733765, "learning_rate": 0.0002, "epoch": 7.578947368421053, "step": 10440}, {"loss": 0.2698, "grad_norm": 1.4632091522216797, "learning_rate": 0.0002, "epoch": 7.586206896551724, "step": 10450}, {"loss": 0.2994, "grad_norm": 1.562364935874939, "learning_rate": 0.0002, "epoch": 7.593466424682395, "step": 10460}, {"loss": 0.271, "grad_norm": 1.2138582468032837, "learning_rate": 0.0002, "epoch": 7.600725952813067, "step": 10470}, {"loss": 0.2638, "grad_norm": 1.467578411102295, "learning_rate": 0.0002, "epoch": 7.607985480943738, "step": 10480}, {"loss": 0.2949, "grad_norm": 1.3470213413238525, "learning_rate": 0.0002, "epoch": 7.61524500907441, "step": 10490}, {"loss": 0.2518, "grad_norm": 1.5385268926620483, "learning_rate": 0.0002, "epoch": 7.6225045372050815, "step": 10500}, {"loss": 0.2527, "grad_norm": 1.1245018243789673, "learning_rate": 0.0002, "epoch": 7.6297640653357535, "step": 10510}, {"loss": 0.267, "grad_norm": 1.3161317110061646, "learning_rate": 0.0002, "epoch": 7.637023593466425, "step": 10520}, {"loss": 0.2803, "grad_norm": 1.0402427911758423, "learning_rate": 0.0002, "epoch": 7.6442831215970966, "step": 10530}, {"loss": 0.2676, "grad_norm": 1.2699987888336182, "learning_rate": 0.0002, "epoch": 7.651542649727768, "step": 10540}, {"loss": 0.2835, "grad_norm": 1.47243332862854, "learning_rate": 0.0002, "epoch": 7.658802177858439, "step": 10550}, {"loss": 0.2624, "grad_norm": 1.1261144876480103, "learning_rate": 0.0002, "epoch": 7.666061705989111, "step": 10560}, {"loss": 0.282, "grad_norm": 1.5402237176895142, "learning_rate": 0.0002, "epoch": 7.673321234119782, "step": 10570}, {"loss": 0.2866, "grad_norm": 1.1316986083984375, "learning_rate": 0.0002, "epoch": 7.680580762250454, "step": 10580}, {"loss": 0.2593, "grad_norm": 1.2155439853668213, "learning_rate": 0.0002, "epoch": 7.687840290381125, "step": 10590}, {"loss": 0.274, "grad_norm": 1.566380500793457, "learning_rate": 0.0002, "epoch": 7.695099818511797, "step": 10600}, {"loss": 0.2664, "grad_norm": 1.7367318868637085, "learning_rate": 0.0002, "epoch": 7.702359346642468, "step": 10610}, {"loss": 0.3115, "grad_norm": 1.5213567018508911, "learning_rate": 0.0002, "epoch": 7.70961887477314, "step": 10620}, {"loss": 0.2863, "grad_norm": 1.3955585956573486, "learning_rate": 0.0002, "epoch": 7.716878402903811, "step": 10630}, {"loss": 0.2683, "grad_norm": 1.321916937828064, "learning_rate": 0.0002, "epoch": 7.724137931034483, "step": 10640}, {"loss": 0.306, "grad_norm": 1.8494919538497925, "learning_rate": 0.0002, "epoch": 7.731397459165154, "step": 10650}, {"loss": 0.2544, "grad_norm": 1.5309828519821167, "learning_rate": 0.0002, "epoch": 7.738656987295826, "step": 10660}, {"loss": 0.2693, "grad_norm": 1.3796069622039795, "learning_rate": 0.0002, "epoch": 7.745916515426497, "step": 10670}, {"loss": 0.2872, "grad_norm": 1.2416858673095703, "learning_rate": 0.0002, "epoch": 7.753176043557168, "step": 10680}, {"loss": 0.2729, "grad_norm": 1.4447332620620728, "learning_rate": 0.0002, "epoch": 7.76043557168784, "step": 10690}, {"loss": 0.2736, "grad_norm": 1.2003352642059326, "learning_rate": 0.0002, "epoch": 7.7676950998185115, "step": 10700}, {"loss": 0.2771, "grad_norm": 1.3607908487319946, "learning_rate": 0.0002, "epoch": 7.7749546279491835, "step": 10710}, {"loss": 0.2739, "grad_norm": 1.1789227724075317, "learning_rate": 0.0002, "epoch": 7.782214156079855, "step": 10720}, {"loss": 0.2927, "grad_norm": 1.2998148202896118, "learning_rate": 0.0002, "epoch": 7.7894736842105265, "step": 10730}, {"loss": 0.2825, "grad_norm": 1.8224656581878662, "learning_rate": 0.0002, "epoch": 7.796733212341198, "step": 10740}, {"loss": 0.2864, "grad_norm": 1.2510570287704468, "learning_rate": 0.0002, "epoch": 7.80399274047187, "step": 10750}, {"loss": 0.3007, "grad_norm": 1.065926194190979, "learning_rate": 0.0002, "epoch": 7.811252268602541, "step": 10760}, {"loss": 0.277, "grad_norm": 1.0313589572906494, "learning_rate": 0.0002, "epoch": 7.818511796733213, "step": 10770}, {"loss": 0.2954, "grad_norm": 1.10769784450531, "learning_rate": 0.0002, "epoch": 7.825771324863884, "step": 10780}, {"loss": 0.2893, "grad_norm": 1.4168727397918701, "learning_rate": 0.0002, "epoch": 7.833030852994556, "step": 10790}, {"loss": 0.2903, "grad_norm": 1.8239266872406006, "learning_rate": 0.0002, "epoch": 7.840290381125227, "step": 10800}, {"loss": 0.2573, "grad_norm": 1.5748721361160278, "learning_rate": 0.0002, "epoch": 7.847549909255898, "step": 10810}, {"loss": 0.3282, "grad_norm": 1.5762766599655151, "learning_rate": 0.0002, "epoch": 7.85480943738657, "step": 10820}, {"loss": 0.2981, "grad_norm": 1.1119135618209839, "learning_rate": 0.0002, "epoch": 7.862068965517241, "step": 10830}, {"loss": 0.3037, "grad_norm": 1.478314995765686, "learning_rate": 0.0002, "epoch": 7.869328493647913, "step": 10840}, {"loss": 0.2866, "grad_norm": 1.2225514650344849, "learning_rate": 0.0002, "epoch": 7.876588021778584, "step": 10850}, {"loss": 0.2795, "grad_norm": 1.503473162651062, "learning_rate": 0.0002, "epoch": 7.883847549909256, "step": 10860}, {"loss": 0.2732, "grad_norm": 1.0334484577178955, "learning_rate": 0.0002, "epoch": 7.891107078039927, "step": 10870}, {"loss": 0.3206, "grad_norm": 1.2068367004394531, "learning_rate": 0.0002, "epoch": 7.898366606170599, "step": 10880}, {"loss": 0.2936, "grad_norm": 1.3105504512786865, "learning_rate": 0.0002, "epoch": 7.90562613430127, "step": 10890}, {"loss": 0.3063, "grad_norm": 1.2941272258758545, "learning_rate": 0.0002, "epoch": 7.9128856624319415, "step": 10900}, {"loss": 0.2862, "grad_norm": 1.2809823751449585, "learning_rate": 0.0002, "epoch": 7.9201451905626135, "step": 10910}, {"loss": 0.3202, "grad_norm": 1.5362727642059326, "learning_rate": 0.0002, "epoch": 7.927404718693285, "step": 10920}, {"loss": 0.3005, "grad_norm": 1.5019413232803345, "learning_rate": 0.0002, "epoch": 7.9346642468239565, "step": 10930}, {"loss": 0.2976, "grad_norm": 1.5947920083999634, "learning_rate": 0.0002, "epoch": 7.941923774954628, "step": 10940}, {"loss": 0.2976, "grad_norm": 1.9482372999191284, "learning_rate": 0.0002, "epoch": 7.9491833030853, "step": 10950}, {"loss": 0.3297, "grad_norm": 1.8445630073547363, "learning_rate": 0.0002, "epoch": 7.956442831215971, "step": 10960}, {"loss": 0.2843, "grad_norm": 1.4342153072357178, "learning_rate": 0.0002, "epoch": 7.963702359346643, "step": 10970}, {"loss": 0.3066, "grad_norm": 1.3202505111694336, "learning_rate": 0.0002, "epoch": 7.970961887477314, "step": 10980}, {"loss": 0.2785, "grad_norm": 1.186015009880066, "learning_rate": 0.0002, "epoch": 7.978221415607986, "step": 10990}, {"loss": 0.324, "grad_norm": 1.2714571952819824, "learning_rate": 0.0002, "epoch": 7.985480943738657, "step": 11000}, {"loss": 0.2795, "grad_norm": 0.9723673462867737, "learning_rate": 0.0002, "epoch": 7.992740471869329, "step": 11010}]}