Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,202 @@
+---
+base_model: vidore/colqwen2-base
+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 0.14.0

adapter_config.json

@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "vidore/colqwen2-base",
+  "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)",
+  "task_type": "FEATURE_EXTRACTION",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:147ef6d31e6dcf1e9f6c5d8fa2c88bc3a21aa715220ab29d4c04c204b3f0bf4f
+size 147983720

added_tokens.json

@@ -0,0 +1,16 @@
+{
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
+}

checkpoint-2310/README.md

@@ -0,0 +1,202 @@
+---
+base_model: ./models/base_models/colqwen2-base
+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 0.14.0

checkpoint-2310/adapter_config.json

@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "./models/base_models/colqwen2-base",
+  "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": "gaussian",
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": "(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)",
+  "task_type": "FEATURE_EXTRACTION",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-2310/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:147ef6d31e6dcf1e9f6c5d8fa2c88bc3a21aa715220ab29d4c04c204b3f0bf4f
+size 147983720

checkpoint-2310/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b647718cff3619615a417919e643c6471686e5c271f02f3b8e0ac1ae6f2bdc34
+size 296193520

checkpoint-2310/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf627896ea992a29f3737001f3e1489fd382a1a7a01dd6d6eb53291c96d48522
+size 14244

checkpoint-2310/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f79aaad2075ba282a87e2d2c96cae3c97a7b3716636483b1e8df7079915eb3db
+size 1064

checkpoint-2310/trainer_state.json

@@ -0,0 +1,1835 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 100,
+  "global_step": 2310,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.021645021645021644,
+      "grad_norm": 2.8125383853912354,
+      "learning_rate": 2e-05,
+      "loss": 0.9514,
+      "step": 10
+    },
+    {
+      "epoch": 0.04329004329004329,
+      "grad_norm": 1.2800142765045166,
+      "learning_rate": 4e-05,
+      "loss": 0.7706,
+      "step": 20
+    },
+    {
+      "epoch": 0.06493506493506493,
+      "grad_norm": 1.0814083814620972,
+      "learning_rate": 6e-05,
+      "loss": 0.6133,
+      "step": 30
+    },
+    {
+      "epoch": 0.08658008658008658,
+      "grad_norm": 0.7970395088195801,
+      "learning_rate": 8e-05,
+      "loss": 0.5192,
+      "step": 40
+    },
+    {
+      "epoch": 0.10822510822510822,
+      "grad_norm": 0.6642207503318787,
+      "learning_rate": 0.0001,
+      "loss": 0.4351,
+      "step": 50
+    },
+    {
+      "epoch": 0.12987012987012986,
+      "grad_norm": 0.5111215114593506,
+      "learning_rate": 0.00012,
+      "loss": 0.3603,
+      "step": 60
+    },
+    {
+      "epoch": 0.15151515151515152,
+      "grad_norm": 1.2057693004608154,
+      "learning_rate": 0.00014,
+      "loss": 0.3361,
+      "step": 70
+    },
+    {
+      "epoch": 0.17316017316017315,
+      "grad_norm": 0.4057958424091339,
+      "learning_rate": 0.00016,
+      "loss": 0.3053,
+      "step": 80
+    },
+    {
+      "epoch": 0.19480519480519481,
+      "grad_norm": 0.5448099374771118,
+      "learning_rate": 0.00018,
+      "loss": 0.3024,
+      "step": 90
+    },
+    {
+      "epoch": 0.21645021645021645,
+      "grad_norm": 0.6636802554130554,
+      "learning_rate": 0.0002,
+      "loss": 0.2595,
+      "step": 100
+    },
+    {
+      "epoch": 0.21645021645021645,
+      "eval_loss": 0.1525065004825592,
+      "eval_runtime": 43.7956,
+      "eval_samples_per_second": 11.417,
+      "eval_steps_per_second": 0.365,
+      "step": 100
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 0.6046785116195679,
+      "learning_rate": 0.0001990950226244344,
+      "loss": 0.2407,
+      "step": 110
+    },
+    {
+      "epoch": 0.2597402597402597,
+      "grad_norm": 0.5757384896278381,
+      "learning_rate": 0.0001981900452488688,
+      "loss": 0.2223,
+      "step": 120
+    },
+    {
+      "epoch": 0.2813852813852814,
+      "grad_norm": 0.5616349577903748,
+      "learning_rate": 0.00019728506787330318,
+      "loss": 0.224,
+      "step": 130
+    },
+    {
+      "epoch": 0.30303030303030304,
+      "grad_norm": 0.5419019460678101,
+      "learning_rate": 0.00019638009049773755,
+      "loss": 0.2306,
+      "step": 140
+    },
+    {
+      "epoch": 0.3246753246753247,
+      "grad_norm": 0.49299636483192444,
+      "learning_rate": 0.00019547511312217194,
+      "loss": 0.2214,
+      "step": 150
+    },
+    {
+      "epoch": 0.3463203463203463,
+      "grad_norm": 0.6301876306533813,
+      "learning_rate": 0.00019457013574660634,
+      "loss": 0.2296,
+      "step": 160
+    },
+    {
+      "epoch": 0.36796536796536794,
+      "grad_norm": 1.009000539779663,
+      "learning_rate": 0.00019366515837104074,
+      "loss": 0.2277,
+      "step": 170
+    },
+    {
+      "epoch": 0.38961038961038963,
+      "grad_norm": 0.5606774091720581,
+      "learning_rate": 0.00019276018099547514,
+      "loss": 0.2235,
+      "step": 180
+    },
+    {
+      "epoch": 0.41125541125541126,
+      "grad_norm": 0.47087350487709045,
+      "learning_rate": 0.0001918552036199095,
+      "loss": 0.2175,
+      "step": 190
+    },
+    {
+      "epoch": 0.4329004329004329,
+      "grad_norm": 0.566985011100769,
+      "learning_rate": 0.0001909502262443439,
+      "loss": 0.2124,
+      "step": 200
+    },
+    {
+      "epoch": 0.4329004329004329,
+      "eval_loss": 0.13294899463653564,
+      "eval_runtime": 23.5705,
+      "eval_samples_per_second": 21.213,
+      "eval_steps_per_second": 0.679,
+      "step": 200
+    },
+    {
+      "epoch": 0.45454545454545453,
+      "grad_norm": 0.7860066294670105,
+      "learning_rate": 0.00019004524886877827,
+      "loss": 0.2164,
+      "step": 210
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 0.47214949131011963,
+      "learning_rate": 0.00018914027149321267,
+      "loss": 0.2086,
+      "step": 220
+    },
+    {
+      "epoch": 0.49783549783549785,
+      "grad_norm": 0.5978089570999146,
+      "learning_rate": 0.00018823529411764707,
+      "loss": 0.2176,
+      "step": 230
+    },
+    {
+      "epoch": 0.5194805194805194,
+      "grad_norm": 0.5042030811309814,
+      "learning_rate": 0.00018733031674208147,
+      "loss": 0.2003,
+      "step": 240
+    },
+    {
+      "epoch": 0.5411255411255411,
+      "grad_norm": 0.5709463953971863,
+      "learning_rate": 0.00018642533936651584,
+      "loss": 0.1969,
+      "step": 250
+    },
+    {
+      "epoch": 0.5627705627705628,
+      "grad_norm": 0.5876468420028687,
+      "learning_rate": 0.00018552036199095024,
+      "loss": 0.1928,
+      "step": 260
+    },
+    {
+      "epoch": 0.5844155844155844,
+      "grad_norm": 0.5380046963691711,
+      "learning_rate": 0.00018461538461538463,
+      "loss": 0.1944,
+      "step": 270
+    },
+    {
+      "epoch": 0.6060606060606061,
+      "grad_norm": 0.3912442624568939,
+      "learning_rate": 0.000183710407239819,
+      "loss": 0.1999,
+      "step": 280
+    },
+    {
+      "epoch": 0.6277056277056277,
+      "grad_norm": 0.6305837631225586,
+      "learning_rate": 0.0001828054298642534,
+      "loss": 0.1976,
+      "step": 290
+    },
+    {
+      "epoch": 0.6493506493506493,
+      "grad_norm": 0.6630299091339111,
+      "learning_rate": 0.0001819004524886878,
+      "loss": 0.1881,
+      "step": 300
+    },
+    {
+      "epoch": 0.6493506493506493,
+      "eval_loss": 0.11815983802080154,
+      "eval_runtime": 23.4543,
+      "eval_samples_per_second": 21.318,
+      "eval_steps_per_second": 0.682,
+      "step": 300
+    },
+    {
+      "epoch": 0.670995670995671,
+      "grad_norm": 0.3871740996837616,
+      "learning_rate": 0.00018099547511312217,
+      "loss": 0.1918,
+      "step": 310
+    },
+    {
+      "epoch": 0.6926406926406926,
+      "grad_norm": 0.39743664860725403,
+      "learning_rate": 0.00018009049773755657,
+      "loss": 0.1875,
+      "step": 320
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 0.580439031124115,
+      "learning_rate": 0.00017918552036199096,
+      "loss": 0.1881,
+      "step": 330
+    },
+    {
+      "epoch": 0.7359307359307359,
+      "grad_norm": 0.39449596405029297,
+      "learning_rate": 0.00017828054298642536,
+      "loss": 0.189,
+      "step": 340
+    },
+    {
+      "epoch": 0.7575757575757576,
+      "grad_norm": 0.5583558678627014,
+      "learning_rate": 0.00017737556561085973,
+      "loss": 0.1912,
+      "step": 350
+    },
+    {
+      "epoch": 0.7792207792207793,
+      "grad_norm": 0.5783946514129639,
+      "learning_rate": 0.00017647058823529413,
+      "loss": 0.1891,
+      "step": 360
+    },
+    {
+      "epoch": 0.8008658008658008,
+      "grad_norm": 0.5039686560630798,
+      "learning_rate": 0.0001755656108597285,
+      "loss": 0.1859,
+      "step": 370
+    },
+    {
+      "epoch": 0.8225108225108225,
+      "grad_norm": 0.45908617973327637,
+      "learning_rate": 0.0001746606334841629,
+      "loss": 0.1868,
+      "step": 380
+    },
+    {
+      "epoch": 0.8441558441558441,
+      "grad_norm": 0.4211718738079071,
+      "learning_rate": 0.0001737556561085973,
+      "loss": 0.1961,
+      "step": 390
+    },
+    {
+      "epoch": 0.8658008658008658,
+      "grad_norm": 0.4308678209781647,
+      "learning_rate": 0.0001728506787330317,
+      "loss": 0.2057,
+      "step": 400
+    },
+    {
+      "epoch": 0.8658008658008658,
+      "eval_loss": 0.12011528760194778,
+      "eval_runtime": 21.7151,
+      "eval_samples_per_second": 23.025,
+      "eval_steps_per_second": 0.737,
+      "step": 400
+    },
+    {
+      "epoch": 0.8874458874458875,
+      "grad_norm": 0.5541667342185974,
+      "learning_rate": 0.0001719457013574661,
+      "loss": 0.1946,
+      "step": 410
+    },
+    {
+      "epoch": 0.9090909090909091,
+      "grad_norm": 0.782439112663269,
+      "learning_rate": 0.00017104072398190046,
+      "loss": 0.1746,
+      "step": 420
+    },
+    {
+      "epoch": 0.9307359307359307,
+      "grad_norm": 0.4267190396785736,
+      "learning_rate": 0.00017013574660633486,
+      "loss": 0.1905,
+      "step": 430
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.4720573127269745,
+      "learning_rate": 0.00016923076923076923,
+      "loss": 0.1925,
+      "step": 440
+    },
+    {
+      "epoch": 0.974025974025974,
+      "grad_norm": 0.49390169978141785,
+      "learning_rate": 0.00016832579185520363,
+      "loss": 0.1967,
+      "step": 450
+    },
+    {
+      "epoch": 0.9956709956709957,
+      "grad_norm": 0.4374224841594696,
+      "learning_rate": 0.00016742081447963802,
+      "loss": 0.1813,
+      "step": 460
+    },
+    {
+      "epoch": 1.0173160173160174,
+      "grad_norm": 0.4421156048774719,
+      "learning_rate": 0.00016651583710407242,
+      "loss": 0.1673,
+      "step": 470
+    },
+    {
+      "epoch": 1.0389610389610389,
+      "grad_norm": 0.35764458775520325,
+      "learning_rate": 0.0001656108597285068,
+      "loss": 0.1759,
+      "step": 480
+    },
+    {
+      "epoch": 1.0606060606060606,
+      "grad_norm": 0.36364632844924927,
+      "learning_rate": 0.0001647058823529412,
+      "loss": 0.1726,
+      "step": 490
+    },
+    {
+      "epoch": 1.0822510822510822,
+      "grad_norm": 0.37476664781570435,
+      "learning_rate": 0.00016380090497737556,
+      "loss": 0.1819,
+      "step": 500
+    },
+    {
+      "epoch": 1.0822510822510822,
+      "eval_loss": 0.12120614945888519,
+      "eval_runtime": 38.4695,
+      "eval_samples_per_second": 12.997,
+      "eval_steps_per_second": 0.416,
+      "step": 500
+    },
+    {
+      "epoch": 1.103896103896104,
+      "grad_norm": 0.4027290344238281,
+      "learning_rate": 0.00016289592760180996,
+      "loss": 0.1633,
+      "step": 510
+    },
+    {
+      "epoch": 1.1255411255411256,
+      "grad_norm": 0.3566421866416931,
+      "learning_rate": 0.00016199095022624435,
+      "loss": 0.1607,
+      "step": 520
+    },
+    {
+      "epoch": 1.1471861471861473,
+      "grad_norm": 0.4911403954029083,
+      "learning_rate": 0.00016108597285067875,
+      "loss": 0.1655,
+      "step": 530
+    },
+    {
+      "epoch": 1.1688311688311688,
+      "grad_norm": 0.3717828392982483,
+      "learning_rate": 0.00016018099547511315,
+      "loss": 0.1646,
+      "step": 540
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 0.4402911961078644,
+      "learning_rate": 0.00015927601809954752,
+      "loss": 0.1677,
+      "step": 550
+    },
+    {
+      "epoch": 1.2121212121212122,
+      "grad_norm": 0.4106089770793915,
+      "learning_rate": 0.0001583710407239819,
+      "loss": 0.1607,
+      "step": 560
+    },
+    {
+      "epoch": 1.2337662337662338,
+      "grad_norm": 0.4335547387599945,
+      "learning_rate": 0.0001574660633484163,
+      "loss": 0.1625,
+      "step": 570
+    },
+    {
+      "epoch": 1.2554112554112553,
+      "grad_norm": 0.3730684816837311,
+      "learning_rate": 0.00015656108597285069,
+      "loss": 0.1627,
+      "step": 580
+    },
+    {
+      "epoch": 1.277056277056277,
+      "grad_norm": 0.4024712145328522,
+      "learning_rate": 0.00015565610859728508,
+      "loss": 0.1575,
+      "step": 590
+    },
+    {
+      "epoch": 1.2987012987012987,
+      "grad_norm": 0.5081992745399475,
+      "learning_rate": 0.00015475113122171948,
+      "loss": 0.1654,
+      "step": 600
+    },
+    {
+      "epoch": 1.2987012987012987,
+      "eval_loss": 0.11448249965906143,
+      "eval_runtime": 25.9879,
+      "eval_samples_per_second": 19.24,
+      "eval_steps_per_second": 0.616,
+      "step": 600
+    },
+    {
+      "epoch": 1.3203463203463204,
+      "grad_norm": 0.5350440144538879,
+      "learning_rate": 0.00015384615384615385,
+      "loss": 0.158,
+      "step": 610
+    },
+    {
+      "epoch": 1.341991341991342,
+      "grad_norm": 0.44623634219169617,
+      "learning_rate": 0.00015294117647058822,
+      "loss": 0.1628,
+      "step": 620
+    },
+    {
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.4370821714401245,
+      "learning_rate": 0.00015203619909502262,
+      "loss": 0.1563,
+      "step": 630
+    },
+    {
+      "epoch": 1.3852813852813852,
+      "grad_norm": 0.35637155175209045,
+      "learning_rate": 0.00015113122171945702,
+      "loss": 0.1706,
+      "step": 640
+    },
+    {
+      "epoch": 1.406926406926407,
+      "grad_norm": 0.3456045687198639,
+      "learning_rate": 0.00015022624434389141,
+      "loss": 0.1695,
+      "step": 650
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.48406872153282166,
+      "learning_rate": 0.0001493212669683258,
+      "loss": 0.1561,
+      "step": 660
+    },
+    {
+      "epoch": 1.4502164502164503,
+      "grad_norm": 0.45908018946647644,
+      "learning_rate": 0.00014841628959276018,
+      "loss": 0.1584,
+      "step": 670
+    },
+    {
+      "epoch": 1.4718614718614718,
+      "grad_norm": 0.4222131073474884,
+      "learning_rate": 0.00014751131221719458,
+      "loss": 0.1524,
+      "step": 680
+    },
+    {
+      "epoch": 1.4935064935064934,
+      "grad_norm": 0.40552523732185364,
+      "learning_rate": 0.00014660633484162895,
+      "loss": 0.1664,
+      "step": 690
+    },
+    {
+      "epoch": 1.5151515151515151,
+      "grad_norm": 0.441180020570755,
+      "learning_rate": 0.00014570135746606335,
+      "loss": 0.1732,
+      "step": 700
+    },
+    {
+      "epoch": 1.5151515151515151,
+      "eval_loss": 0.10767688602209091,
+      "eval_runtime": 25.2267,
+      "eval_samples_per_second": 19.82,
+      "eval_steps_per_second": 0.634,
+      "step": 700
+    },
+    {
+      "epoch": 1.5367965367965368,
+      "grad_norm": 0.39050132036209106,
+      "learning_rate": 0.00014479638009049775,
+      "loss": 0.1544,
+      "step": 710
+    },
+    {
+      "epoch": 1.5584415584415585,
+      "grad_norm": 0.37475523352622986,
+      "learning_rate": 0.00014389140271493214,
+      "loss": 0.1534,
+      "step": 720
+    },
+    {
+      "epoch": 1.5800865800865802,
+      "grad_norm": 0.3851136267185211,
+      "learning_rate": 0.00014298642533936651,
+      "loss": 0.1623,
+      "step": 730
+    },
+    {
+      "epoch": 1.601731601731602,
+      "grad_norm": 0.37124335765838623,
+      "learning_rate": 0.0001420814479638009,
+      "loss": 0.152,
+      "step": 740
+    },
+    {
+      "epoch": 1.6233766233766234,
+      "grad_norm": 0.3956158459186554,
+      "learning_rate": 0.0001411764705882353,
+      "loss": 0.1619,
+      "step": 750
+    },
+    {
+      "epoch": 1.645021645021645,
+      "grad_norm": 0.34193211793899536,
+      "learning_rate": 0.00014027149321266968,
+      "loss": 0.158,
+      "step": 760
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.5510150194168091,
+      "learning_rate": 0.00013936651583710408,
+      "loss": 0.158,
+      "step": 770
+    },
+    {
+      "epoch": 1.6883116883116882,
+      "grad_norm": 0.6107664108276367,
+      "learning_rate": 0.00013846153846153847,
+      "loss": 0.1487,
+      "step": 780
+    },
+    {
+      "epoch": 1.70995670995671,
+      "grad_norm": 0.4111076295375824,
+      "learning_rate": 0.00013755656108597284,
+      "loss": 0.1588,
+      "step": 790
+    },
+    {
+      "epoch": 1.7316017316017316,
+      "grad_norm": 0.5250778198242188,
+      "learning_rate": 0.00013665158371040724,
+      "loss": 0.1653,
+      "step": 800
+    },
+    {
+      "epoch": 1.7316017316017316,
+      "eval_loss": 0.10399862378835678,
+      "eval_runtime": 23.9812,
+      "eval_samples_per_second": 20.85,
+      "eval_steps_per_second": 0.667,
+      "step": 800
+    },
+    {
+      "epoch": 1.7532467532467533,
+      "grad_norm": 0.4260198771953583,
+      "learning_rate": 0.00013574660633484164,
+      "loss": 0.1482,
+      "step": 810
+    },
+    {
+      "epoch": 1.774891774891775,
+      "grad_norm": 0.28968125581741333,
+      "learning_rate": 0.00013484162895927604,
+      "loss": 0.1541,
+      "step": 820
+    },
+    {
+      "epoch": 1.7965367965367967,
+      "grad_norm": 0.29932093620300293,
+      "learning_rate": 0.0001339366515837104,
+      "loss": 0.1625,
+      "step": 830
+    },
+    {
+      "epoch": 1.8181818181818183,
+      "grad_norm": 0.4358128011226654,
+      "learning_rate": 0.0001330316742081448,
+      "loss": 0.1658,
+      "step": 840
+    },
+    {
+      "epoch": 1.8398268398268398,
+      "grad_norm": 0.4161946773529053,
+      "learning_rate": 0.00013212669683257918,
+      "loss": 0.1472,
+      "step": 850
+    },
+    {
+      "epoch": 1.8614718614718615,
+      "grad_norm": 0.3558347225189209,
+      "learning_rate": 0.00013122171945701357,
+      "loss": 0.1671,
+      "step": 860
+    },
+    {
+      "epoch": 1.883116883116883,
+      "grad_norm": 0.41748765110969543,
+      "learning_rate": 0.00013031674208144797,
+      "loss": 0.1664,
+      "step": 870
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.5195390582084656,
+      "learning_rate": 0.00012941176470588237,
+      "loss": 0.1603,
+      "step": 880
+    },
+    {
+      "epoch": 1.9264069264069263,
+      "grad_norm": 0.344159334897995,
+      "learning_rate": 0.00012850678733031677,
+      "loss": 0.1589,
+      "step": 890
+    },
+    {
+      "epoch": 1.948051948051948,
+      "grad_norm": 0.4217064380645752,
+      "learning_rate": 0.00012760180995475114,
+      "loss": 0.1631,
+      "step": 900
+    },
+    {
+      "epoch": 1.948051948051948,
+      "eval_loss": 0.11024898290634155,
+      "eval_runtime": 26.0192,
+      "eval_samples_per_second": 19.217,
+      "eval_steps_per_second": 0.615,
+      "step": 900
+    },
+    {
+      "epoch": 1.9696969696969697,
+      "grad_norm": 0.377990186214447,
+      "learning_rate": 0.0001266968325791855,
+      "loss": 0.162,
+      "step": 910
+    },
+    {
+      "epoch": 1.9913419913419914,
+      "grad_norm": 0.3964829742908478,
+      "learning_rate": 0.0001257918552036199,
+      "loss": 0.1565,
+      "step": 920
+    },
+    {
+      "epoch": 2.012987012987013,
+      "grad_norm": 0.31516608595848083,
+      "learning_rate": 0.0001248868778280543,
+      "loss": 0.1356,
+      "step": 930
+    },
+    {
+      "epoch": 2.034632034632035,
+      "grad_norm": 0.4001309275627136,
+      "learning_rate": 0.0001239819004524887,
+      "loss": 0.1398,
+      "step": 940
+    },
+    {
+      "epoch": 2.0562770562770565,
+      "grad_norm": 0.2704612612724304,
+      "learning_rate": 0.0001230769230769231,
+      "loss": 0.139,
+      "step": 950
+    },
+    {
+      "epoch": 2.0779220779220777,
+      "grad_norm": 0.37967827916145325,
+      "learning_rate": 0.0001221719457013575,
+      "loss": 0.1474,
+      "step": 960
+    },
+    {
+      "epoch": 2.0995670995670994,
+      "grad_norm": 0.3611961603164673,
+      "learning_rate": 0.00012126696832579185,
+      "loss": 0.1394,
+      "step": 970
+    },
+    {
+      "epoch": 2.121212121212121,
+      "grad_norm": 0.38165083527565,
+      "learning_rate": 0.00012036199095022625,
+      "loss": 0.1509,
+      "step": 980
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 0.44539740681648254,
+      "learning_rate": 0.00011945701357466063,
+      "loss": 0.1335,
+      "step": 990
+    },
+    {
+      "epoch": 2.1645021645021645,
+      "grad_norm": 0.2796030044555664,
+      "learning_rate": 0.00011855203619909503,
+      "loss": 0.1354,
+      "step": 1000
+    },
+    {
+      "epoch": 2.1645021645021645,
+      "eval_loss": 0.11003410071134567,
+      "eval_runtime": 23.006,
+      "eval_samples_per_second": 21.733,
+      "eval_steps_per_second": 0.695,
+      "step": 1000
+    },
+    {
+      "epoch": 2.186147186147186,
+      "grad_norm": 0.39886558055877686,
+      "learning_rate": 0.00011764705882352942,
+      "loss": 0.1365,
+      "step": 1010
+    },
+    {
+      "epoch": 2.207792207792208,
+      "grad_norm": 0.27149999141693115,
+      "learning_rate": 0.00011674208144796381,
+      "loss": 0.1303,
+      "step": 1020
+    },
+    {
+      "epoch": 2.2294372294372296,
+      "grad_norm": 0.4485577940940857,
+      "learning_rate": 0.00011583710407239821,
+      "loss": 0.136,
+      "step": 1030
+    },
+    {
+      "epoch": 2.2510822510822512,
+      "grad_norm": 0.5114185214042664,
+      "learning_rate": 0.00011493212669683258,
+      "loss": 0.129,
+      "step": 1040
+    },
+    {
+      "epoch": 2.2727272727272725,
+      "grad_norm": 0.41526681184768677,
+      "learning_rate": 0.00011402714932126696,
+      "loss": 0.1352,
+      "step": 1050
+    },
+    {
+      "epoch": 2.2943722943722946,
+      "grad_norm": 0.4560807943344116,
+      "learning_rate": 0.00011312217194570136,
+      "loss": 0.1473,
+      "step": 1060
+    },
+    {
+      "epoch": 2.316017316017316,
+      "grad_norm": 0.4193613529205322,
+      "learning_rate": 0.00011221719457013576,
+      "loss": 0.1427,
+      "step": 1070
+    },
+    {
+      "epoch": 2.3376623376623376,
+      "grad_norm": 0.5285654664039612,
+      "learning_rate": 0.00011131221719457014,
+      "loss": 0.1395,
+      "step": 1080
+    },
+    {
+      "epoch": 2.3593073593073592,
+      "grad_norm": 0.3672701120376587,
+      "learning_rate": 0.00011040723981900454,
+      "loss": 0.1303,
+      "step": 1090
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.4193953573703766,
+      "learning_rate": 0.00010950226244343893,
+      "loss": 0.1353,
+      "step": 1100
+    },
+    {
+      "epoch": 2.380952380952381,
+      "eval_loss": 0.10145355015993118,
+      "eval_runtime": 20.0537,
+      "eval_samples_per_second": 24.933,
+      "eval_steps_per_second": 0.798,
+      "step": 1100
+    },
+    {
+      "epoch": 2.4025974025974026,
+      "grad_norm": 0.39170539379119873,
+      "learning_rate": 0.0001085972850678733,
+      "loss": 0.1344,
+      "step": 1110
+    },
+    {
+      "epoch": 2.4242424242424243,
+      "grad_norm": 0.4148479104042053,
+      "learning_rate": 0.0001076923076923077,
+      "loss": 0.1463,
+      "step": 1120
+    },
+    {
+      "epoch": 2.445887445887446,
+      "grad_norm": 0.3758697211742401,
+      "learning_rate": 0.00010678733031674209,
+      "loss": 0.1338,
+      "step": 1130
+    },
+    {
+      "epoch": 2.4675324675324677,
+      "grad_norm": 0.3020533621311188,
+      "learning_rate": 0.00010588235294117647,
+      "loss": 0.1369,
+      "step": 1140
+    },
+    {
+      "epoch": 2.4891774891774894,
+      "grad_norm": 0.3161577582359314,
+      "learning_rate": 0.00010497737556561087,
+      "loss": 0.1438,
+      "step": 1150
+    },
+    {
+      "epoch": 2.5108225108225106,
+      "grad_norm": 0.39273881912231445,
+      "learning_rate": 0.00010407239819004526,
+      "loss": 0.1404,
+      "step": 1160
+    },
+    {
+      "epoch": 2.5324675324675323,
+      "grad_norm": 0.3449176251888275,
+      "learning_rate": 0.00010316742081447965,
+      "loss": 0.1346,
+      "step": 1170
+    },
+    {
+      "epoch": 2.554112554112554,
+      "grad_norm": 0.261294960975647,
+      "learning_rate": 0.00010226244343891402,
+      "loss": 0.1244,
+      "step": 1180
+    },
+    {
+      "epoch": 2.5757575757575757,
+      "grad_norm": 0.5024470686912537,
+      "learning_rate": 0.00010135746606334842,
+      "loss": 0.1385,
+      "step": 1190
+    },
+    {
+      "epoch": 2.5974025974025974,
+      "grad_norm": 0.4657529294490814,
+      "learning_rate": 0.0001004524886877828,
+      "loss": 0.1458,
+      "step": 1200
+    },
+    {
+      "epoch": 2.5974025974025974,
+      "eval_loss": 0.09968920797109604,
+      "eval_runtime": 22.2909,
+      "eval_samples_per_second": 22.431,
+      "eval_steps_per_second": 0.718,
+      "step": 1200
+    },
+    {
+      "epoch": 2.619047619047619,
+      "grad_norm": 0.28800147771835327,
+      "learning_rate": 9.95475113122172e-05,
+      "loss": 0.1366,
+      "step": 1210
+    },
+    {
+      "epoch": 2.6406926406926408,
+      "grad_norm": 0.24663890898227692,
+      "learning_rate": 9.864253393665159e-05,
+      "loss": 0.1315,
+      "step": 1220
+    },
+    {
+      "epoch": 2.6623376623376624,
+      "grad_norm": 0.3126956522464752,
+      "learning_rate": 9.773755656108597e-05,
+      "loss": 0.1305,
+      "step": 1230
+    },
+    {
+      "epoch": 2.683982683982684,
+      "grad_norm": 0.34753990173339844,
+      "learning_rate": 9.683257918552037e-05,
+      "loss": 0.1326,
+      "step": 1240
+    },
+    {
+      "epoch": 2.7056277056277054,
+      "grad_norm": 0.4045655429363251,
+      "learning_rate": 9.592760180995475e-05,
+      "loss": 0.147,
+      "step": 1250
+    },
+    {
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.369632750749588,
+      "learning_rate": 9.502262443438914e-05,
+      "loss": 0.1339,
+      "step": 1260
+    },
+    {
+      "epoch": 2.7489177489177488,
+      "grad_norm": 0.3891206383705139,
+      "learning_rate": 9.411764705882353e-05,
+      "loss": 0.1326,
+      "step": 1270
+    },
+    {
+      "epoch": 2.7705627705627704,
+      "grad_norm": 0.4095381498336792,
+      "learning_rate": 9.321266968325792e-05,
+      "loss": 0.1305,
+      "step": 1280
+    },
+    {
+      "epoch": 2.792207792207792,
+      "grad_norm": 0.39312949776649475,
+      "learning_rate": 9.230769230769232e-05,
+      "loss": 0.1348,
+      "step": 1290
+    },
+    {
+      "epoch": 2.813852813852814,
+      "grad_norm": 0.4769635498523712,
+      "learning_rate": 9.14027149321267e-05,
+      "loss": 0.1292,
+      "step": 1300
+    },
+    {
+      "epoch": 2.813852813852814,
+      "eval_loss": 0.09773257374763489,
+      "eval_runtime": 21.4232,
+      "eval_samples_per_second": 23.339,
+      "eval_steps_per_second": 0.747,
+      "step": 1300
+    },
+    {
+      "epoch": 2.8354978354978355,
+      "grad_norm": 0.4419485628604889,
+      "learning_rate": 9.049773755656108e-05,
+      "loss": 0.1332,
+      "step": 1310
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 0.273708313703537,
+      "learning_rate": 8.959276018099548e-05,
+      "loss": 0.1351,
+      "step": 1320
+    },
+    {
+      "epoch": 2.878787878787879,
+      "grad_norm": 0.41094347834587097,
+      "learning_rate": 8.868778280542987e-05,
+      "loss": 0.1436,
+      "step": 1330
+    },
+    {
+      "epoch": 2.9004329004329006,
+      "grad_norm": 0.3220170736312866,
+      "learning_rate": 8.778280542986425e-05,
+      "loss": 0.1372,
+      "step": 1340
+    },
+    {
+      "epoch": 2.9220779220779223,
+      "grad_norm": 0.3632793128490448,
+      "learning_rate": 8.687782805429865e-05,
+      "loss": 0.1421,
+      "step": 1350
+    },
+    {
+      "epoch": 2.9437229437229435,
+      "grad_norm": 0.33830907940864563,
+      "learning_rate": 8.597285067873304e-05,
+      "loss": 0.1452,
+      "step": 1360
+    },
+    {
+      "epoch": 2.965367965367965,
+      "grad_norm": 0.3863072097301483,
+      "learning_rate": 8.506787330316743e-05,
+      "loss": 0.1424,
+      "step": 1370
+    },
+    {
+      "epoch": 2.987012987012987,
+      "grad_norm": 0.5766463279724121,
+      "learning_rate": 8.416289592760181e-05,
+      "loss": 0.1424,
+      "step": 1380
+    },
+    {
+      "epoch": 3.0086580086580086,
+      "grad_norm": 0.23465487360954285,
+      "learning_rate": 8.325791855203621e-05,
+      "loss": 0.1241,
+      "step": 1390
+    },
+    {
+      "epoch": 3.0303030303030303,
+      "grad_norm": 0.32335883378982544,
+      "learning_rate": 8.23529411764706e-05,
+      "loss": 0.1181,
+      "step": 1400
+    },
+    {
+      "epoch": 3.0303030303030303,
+      "eval_loss": 0.10378438234329224,
+      "eval_runtime": 24.145,
+      "eval_samples_per_second": 20.708,
+      "eval_steps_per_second": 0.663,
+      "step": 1400
+    },
+    {
+      "epoch": 3.051948051948052,
+      "grad_norm": 0.3104758560657501,
+      "learning_rate": 8.144796380090498e-05,
+      "loss": 0.1227,
+      "step": 1410
+    },
+    {
+      "epoch": 3.0735930735930737,
+      "grad_norm": 0.27100279927253723,
+      "learning_rate": 8.054298642533938e-05,
+      "loss": 0.1164,
+      "step": 1420
+    },
+    {
+      "epoch": 3.0952380952380953,
+      "grad_norm": 0.3317118287086487,
+      "learning_rate": 7.963800904977376e-05,
+      "loss": 0.1233,
+      "step": 1430
+    },
+    {
+      "epoch": 3.116883116883117,
+      "grad_norm": 0.41682133078575134,
+      "learning_rate": 7.873303167420814e-05,
+      "loss": 0.1213,
+      "step": 1440
+    },
+    {
+      "epoch": 3.1385281385281387,
+      "grad_norm": 0.31365829706192017,
+      "learning_rate": 7.782805429864254e-05,
+      "loss": 0.1183,
+      "step": 1450
+    },
+    {
+      "epoch": 3.16017316017316,
+      "grad_norm": 0.3687366247177124,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": 0.119,
+      "step": 1460
+    },
+    {
+      "epoch": 3.1818181818181817,
+      "grad_norm": 0.4626697599887848,
+      "learning_rate": 7.601809954751131e-05,
+      "loss": 0.1244,
+      "step": 1470
+    },
+    {
+      "epoch": 3.2034632034632033,
+      "grad_norm": 0.46320992708206177,
+      "learning_rate": 7.511312217194571e-05,
+      "loss": 0.1214,
+      "step": 1480
+    },
+    {
+      "epoch": 3.225108225108225,
+      "grad_norm": 0.3054867386817932,
+      "learning_rate": 7.420814479638009e-05,
+      "loss": 0.1276,
+      "step": 1490
+    },
+    {
+      "epoch": 3.2467532467532467,
+      "grad_norm": 0.26975077390670776,
+      "learning_rate": 7.330316742081448e-05,
+      "loss": 0.1235,
+      "step": 1500
+    },
+    {
+      "epoch": 3.2467532467532467,
+      "eval_loss": 0.10016042739152908,
+      "eval_runtime": 23.6078,
+      "eval_samples_per_second": 21.179,
+      "eval_steps_per_second": 0.678,
+      "step": 1500
+    },
+    {
+      "epoch": 3.2683982683982684,
+      "grad_norm": 0.33022555708885193,
+      "learning_rate": 7.239819004524887e-05,
+      "loss": 0.12,
+      "step": 1510
+    },
+    {
+      "epoch": 3.29004329004329,
+      "grad_norm": 0.4828197956085205,
+      "learning_rate": 7.149321266968326e-05,
+      "loss": 0.1299,
+      "step": 1520
+    },
+    {
+      "epoch": 3.311688311688312,
+      "grad_norm": 0.36965712904930115,
+      "learning_rate": 7.058823529411765e-05,
+      "loss": 0.1217,
+      "step": 1530
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.4338422417640686,
+      "learning_rate": 6.968325791855204e-05,
+      "loss": 0.1276,
+      "step": 1540
+    },
+    {
+      "epoch": 3.354978354978355,
+      "grad_norm": 0.40615785121917725,
+      "learning_rate": 6.877828054298642e-05,
+      "loss": 0.1214,
+      "step": 1550
+    },
+    {
+      "epoch": 3.3766233766233764,
+      "grad_norm": 0.2703055739402771,
+      "learning_rate": 6.787330316742082e-05,
+      "loss": 0.1179,
+      "step": 1560
+    },
+    {
+      "epoch": 3.398268398268398,
+      "grad_norm": 0.469108521938324,
+      "learning_rate": 6.69683257918552e-05,
+      "loss": 0.1189,
+      "step": 1570
+    },
+    {
+      "epoch": 3.41991341991342,
+      "grad_norm": 0.2688772976398468,
+      "learning_rate": 6.606334841628959e-05,
+      "loss": 0.1174,
+      "step": 1580
+    },
+    {
+      "epoch": 3.4415584415584415,
+      "grad_norm": 0.2989065647125244,
+      "learning_rate": 6.515837104072399e-05,
+      "loss": 0.111,
+      "step": 1590
+    },
+    {
+      "epoch": 3.463203463203463,
+      "grad_norm": 0.397502601146698,
+      "learning_rate": 6.425339366515838e-05,
+      "loss": 0.1189,
+      "step": 1600
+    },
+    {
+      "epoch": 3.463203463203463,
+      "eval_loss": 0.10202713310718536,
+      "eval_runtime": 25.0286,
+      "eval_samples_per_second": 19.977,
+      "eval_steps_per_second": 0.639,
+      "step": 1600
+    },
+    {
+      "epoch": 3.484848484848485,
+      "grad_norm": 0.3283541202545166,
+      "learning_rate": 6.334841628959275e-05,
+      "loss": 0.1216,
+      "step": 1610
+    },
+    {
+      "epoch": 3.5064935064935066,
+      "grad_norm": 0.3664211928844452,
+      "learning_rate": 6.244343891402715e-05,
+      "loss": 0.1171,
+      "step": 1620
+    },
+    {
+      "epoch": 3.5281385281385282,
+      "grad_norm": 0.33489325642585754,
+      "learning_rate": 6.153846153846155e-05,
+      "loss": 0.1222,
+      "step": 1630
+    },
+    {
+      "epoch": 3.54978354978355,
+      "grad_norm": 0.27272143959999084,
+      "learning_rate": 6.0633484162895926e-05,
+      "loss": 0.1089,
+      "step": 1640
+    },
+    {
+      "epoch": 3.571428571428571,
+      "grad_norm": 0.3384559750556946,
+      "learning_rate": 5.972850678733032e-05,
+      "loss": 0.1092,
+      "step": 1650
+    },
+    {
+      "epoch": 3.5930735930735933,
+      "grad_norm": 0.4882173240184784,
+      "learning_rate": 5.882352941176471e-05,
+      "loss": 0.1326,
+      "step": 1660
+    },
+    {
+      "epoch": 3.6147186147186146,
+      "grad_norm": 0.33967503905296326,
+      "learning_rate": 5.7918552036199105e-05,
+      "loss": 0.119,
+      "step": 1670
+    },
+    {
+      "epoch": 3.6363636363636362,
+      "grad_norm": 0.30732062458992004,
+      "learning_rate": 5.701357466063348e-05,
+      "loss": 0.1115,
+      "step": 1680
+    },
+    {
+      "epoch": 3.658008658008658,
+      "grad_norm": 0.46696531772613525,
+      "learning_rate": 5.610859728506788e-05,
+      "loss": 0.1246,
+      "step": 1690
+    },
+    {
+      "epoch": 3.6796536796536796,
+      "grad_norm": 0.4071387052536011,
+      "learning_rate": 5.520361990950227e-05,
+      "loss": 0.1305,
+      "step": 1700
+    },
+    {
+      "epoch": 3.6796536796536796,
+      "eval_loss": 0.09984961152076721,
+      "eval_runtime": 22.2364,
+      "eval_samples_per_second": 22.486,
+      "eval_steps_per_second": 0.72,
+      "step": 1700
+    },
+    {
+      "epoch": 3.7012987012987013,
+      "grad_norm": 0.34078463912010193,
+      "learning_rate": 5.429864253393665e-05,
+      "loss": 0.1227,
+      "step": 1710
+    },
+    {
+      "epoch": 3.722943722943723,
+      "grad_norm": 0.46931177377700806,
+      "learning_rate": 5.3393665158371045e-05,
+      "loss": 0.1222,
+      "step": 1720
+    },
+    {
+      "epoch": 3.7445887445887447,
+      "grad_norm": 0.28127625584602356,
+      "learning_rate": 5.2488687782805436e-05,
+      "loss": 0.1186,
+      "step": 1730
+    },
+    {
+      "epoch": 3.7662337662337664,
+      "grad_norm": 0.3582036793231964,
+      "learning_rate": 5.158371040723983e-05,
+      "loss": 0.1124,
+      "step": 1740
+    },
+    {
+      "epoch": 3.787878787878788,
+      "grad_norm": 0.31167203187942505,
+      "learning_rate": 5.067873303167421e-05,
+      "loss": 0.1179,
+      "step": 1750
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 0.391791969537735,
+      "learning_rate": 4.97737556561086e-05,
+      "loss": 0.127,
+      "step": 1760
+    },
+    {
+      "epoch": 3.8311688311688314,
+      "grad_norm": 0.3683635890483856,
+      "learning_rate": 4.8868778280542986e-05,
+      "loss": 0.1129,
+      "step": 1770
+    },
+    {
+      "epoch": 3.8528138528138527,
+      "grad_norm": 0.3955051302909851,
+      "learning_rate": 4.7963800904977377e-05,
+      "loss": 0.125,
+      "step": 1780
+    },
+    {
+      "epoch": 3.8744588744588744,
+      "grad_norm": 0.40321120619773865,
+      "learning_rate": 4.705882352941177e-05,
+      "loss": 0.1272,
+      "step": 1790
+    },
+    {
+      "epoch": 3.896103896103896,
+      "grad_norm": 0.31911927461624146,
+      "learning_rate": 4.615384615384616e-05,
+      "loss": 0.1072,
+      "step": 1800
+    },
+    {
+      "epoch": 3.896103896103896,
+      "eval_loss": 0.10081314295530319,
+      "eval_runtime": 21.9055,
+      "eval_samples_per_second": 22.825,
+      "eval_steps_per_second": 0.73,
+      "step": 1800
+    },
+    {
+      "epoch": 3.9177489177489178,
+      "grad_norm": 0.4627375900745392,
+      "learning_rate": 4.524886877828054e-05,
+      "loss": 0.1238,
+      "step": 1810
+    },
+    {
+      "epoch": 3.9393939393939394,
+      "grad_norm": 0.43323320150375366,
+      "learning_rate": 4.434389140271493e-05,
+      "loss": 0.1202,
+      "step": 1820
+    },
+    {
+      "epoch": 3.961038961038961,
+      "grad_norm": 0.4214964807033539,
+      "learning_rate": 4.3438914027149324e-05,
+      "loss": 0.1186,
+      "step": 1830
+    },
+    {
+      "epoch": 3.982683982683983,
+      "grad_norm": 0.302083283662796,
+      "learning_rate": 4.2533936651583714e-05,
+      "loss": 0.1226,
+      "step": 1840
+    },
+    {
+      "epoch": 4.004329004329004,
+      "grad_norm": 0.39333033561706543,
+      "learning_rate": 4.1628959276018105e-05,
+      "loss": 0.1249,
+      "step": 1850
+    },
+    {
+      "epoch": 4.025974025974026,
+      "grad_norm": 0.34783726930618286,
+      "learning_rate": 4.072398190045249e-05,
+      "loss": 0.1135,
+      "step": 1860
+    },
+    {
+      "epoch": 4.0476190476190474,
+      "grad_norm": 0.3140253722667694,
+      "learning_rate": 3.981900452488688e-05,
+      "loss": 0.1095,
+      "step": 1870
+    },
+    {
+      "epoch": 4.06926406926407,
+      "grad_norm": 0.3263933062553406,
+      "learning_rate": 3.891402714932127e-05,
+      "loss": 0.1136,
+      "step": 1880
+    },
+    {
+      "epoch": 4.090909090909091,
+      "grad_norm": 0.4250706434249878,
+      "learning_rate": 3.8009049773755655e-05,
+      "loss": 0.1056,
+      "step": 1890
+    },
+    {
+      "epoch": 4.112554112554113,
+      "grad_norm": 0.2250766158103943,
+      "learning_rate": 3.7104072398190046e-05,
+      "loss": 0.1079,
+      "step": 1900
+    },
+    {
+      "epoch": 4.112554112554113,
+      "eval_loss": 0.10422435402870178,
+      "eval_runtime": 22.3048,
+      "eval_samples_per_second": 22.417,
+      "eval_steps_per_second": 0.717,
+      "step": 1900
+    },
+    {
+      "epoch": 4.134199134199134,
+      "grad_norm": 0.35859718918800354,
+      "learning_rate": 3.6199095022624436e-05,
+      "loss": 0.1048,
+      "step": 1910
+    },
+    {
+      "epoch": 4.1558441558441555,
+      "grad_norm": 0.2876388728618622,
+      "learning_rate": 3.529411764705883e-05,
+      "loss": 0.1109,
+      "step": 1920
+    },
+    {
+      "epoch": 4.177489177489178,
+      "grad_norm": 0.2952074408531189,
+      "learning_rate": 3.438914027149321e-05,
+      "loss": 0.1118,
+      "step": 1930
+    },
+    {
+      "epoch": 4.199134199134199,
+      "grad_norm": 0.38420605659484863,
+      "learning_rate": 3.34841628959276e-05,
+      "loss": 0.1113,
+      "step": 1940
+    },
+    {
+      "epoch": 4.220779220779221,
+      "grad_norm": 0.35064896941185,
+      "learning_rate": 3.257918552036199e-05,
+      "loss": 0.108,
+      "step": 1950
+    },
+    {
+      "epoch": 4.242424242424242,
+      "grad_norm": 0.3417244553565979,
+      "learning_rate": 3.167420814479638e-05,
+      "loss": 0.1067,
+      "step": 1960
+    },
+    {
+      "epoch": 4.264069264069264,
+      "grad_norm": 0.2506988048553467,
+      "learning_rate": 3.0769230769230774e-05,
+      "loss": 0.116,
+      "step": 1970
+    },
+    {
+      "epoch": 4.285714285714286,
+      "grad_norm": 0.40218839049339294,
+      "learning_rate": 2.986425339366516e-05,
+      "loss": 0.111,
+      "step": 1980
+    },
+    {
+      "epoch": 4.307359307359308,
+      "grad_norm": 0.33777332305908203,
+      "learning_rate": 2.8959276018099553e-05,
+      "loss": 0.1081,
+      "step": 1990
+    },
+    {
+      "epoch": 4.329004329004329,
+      "grad_norm": 0.2849682569503784,
+      "learning_rate": 2.805429864253394e-05,
+      "loss": 0.1043,
+      "step": 2000
+    },
+    {
+      "epoch": 4.329004329004329,
+      "eval_loss": 0.10182041674852371,
+      "eval_runtime": 32.5102,
+      "eval_samples_per_second": 15.38,
+      "eval_steps_per_second": 0.492,
+      "step": 2000
+    },
+    {
+      "epoch": 4.35064935064935,
+      "grad_norm": 0.32567378878593445,
+      "learning_rate": 2.7149321266968324e-05,
+      "loss": 0.1065,
+      "step": 2010
+    },
+    {
+      "epoch": 4.372294372294372,
+      "grad_norm": 0.35031840205192566,
+      "learning_rate": 2.6244343891402718e-05,
+      "loss": 0.1133,
+      "step": 2020
+    },
+    {
+      "epoch": 4.393939393939394,
+      "grad_norm": 0.3233943581581116,
+      "learning_rate": 2.5339366515837106e-05,
+      "loss": 0.1034,
+      "step": 2030
+    },
+    {
+      "epoch": 4.415584415584416,
+      "grad_norm": 0.326164573431015,
+      "learning_rate": 2.4434389140271493e-05,
+      "loss": 0.1053,
+      "step": 2040
+    },
+    {
+      "epoch": 4.437229437229437,
+      "grad_norm": 0.3046077489852905,
+      "learning_rate": 2.3529411764705884e-05,
+      "loss": 0.1034,
+      "step": 2050
+    },
+    {
+      "epoch": 4.458874458874459,
+      "grad_norm": 0.3930221498012543,
+      "learning_rate": 2.262443438914027e-05,
+      "loss": 0.1132,
+      "step": 2060
+    },
+    {
+      "epoch": 4.48051948051948,
+      "grad_norm": 0.4203428626060486,
+      "learning_rate": 2.1719457013574662e-05,
+      "loss": 0.1123,
+      "step": 2070
+    },
+    {
+      "epoch": 4.5021645021645025,
+      "grad_norm": 0.23356233537197113,
+      "learning_rate": 2.0814479638009053e-05,
+      "loss": 0.1158,
+      "step": 2080
+    },
+    {
+      "epoch": 4.523809523809524,
+      "grad_norm": 0.3921778202056885,
+      "learning_rate": 1.990950226244344e-05,
+      "loss": 0.1017,
+      "step": 2090
+    },
+    {
+      "epoch": 4.545454545454545,
+      "grad_norm": 0.41227224469184875,
+      "learning_rate": 1.9004524886877827e-05,
+      "loss": 0.1095,
+      "step": 2100
+    },
+    {
+      "epoch": 4.545454545454545,
+      "eval_loss": 0.10190469026565552,
+      "eval_runtime": 22.5368,
+      "eval_samples_per_second": 22.186,
+      "eval_steps_per_second": 0.71,
+      "step": 2100
+    },
+    {
+      "epoch": 4.567099567099567,
+      "grad_norm": 0.46910232305526733,
+      "learning_rate": 1.8099547511312218e-05,
+      "loss": 0.1041,
+      "step": 2110
+    },
+    {
+      "epoch": 4.588744588744589,
+      "grad_norm": 0.2650741934776306,
+      "learning_rate": 1.7194570135746606e-05,
+      "loss": 0.1096,
+      "step": 2120
+    },
+    {
+      "epoch": 4.6103896103896105,
+      "grad_norm": 0.30446478724479675,
+      "learning_rate": 1.6289592760180996e-05,
+      "loss": 0.11,
+      "step": 2130
+    },
+    {
+      "epoch": 4.632034632034632,
+      "grad_norm": 0.30164214968681335,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 0.1009,
+      "step": 2140
+    },
+    {
+      "epoch": 4.653679653679654,
+      "grad_norm": 0.33579009771347046,
+      "learning_rate": 1.4479638009049776e-05,
+      "loss": 0.116,
+      "step": 2150
+    },
+    {
+      "epoch": 4.675324675324675,
+      "grad_norm": 0.2805331349372864,
+      "learning_rate": 1.3574660633484162e-05,
+      "loss": 0.1154,
+      "step": 2160
+    },
+    {
+      "epoch": 4.696969696969697,
+      "grad_norm": 0.2765107750892639,
+      "learning_rate": 1.2669683257918553e-05,
+      "loss": 0.1124,
+      "step": 2170
+    },
+    {
+      "epoch": 4.7186147186147185,
+      "grad_norm": 0.39715567231178284,
+      "learning_rate": 1.1764705882352942e-05,
+      "loss": 0.1078,
+      "step": 2180
+    },
+    {
+      "epoch": 4.740259740259741,
+      "grad_norm": 0.4515060782432556,
+      "learning_rate": 1.0859728506787331e-05,
+      "loss": 0.1135,
+      "step": 2190
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.3021312654018402,
+      "learning_rate": 9.95475113122172e-06,
+      "loss": 0.1069,
+      "step": 2200
+    },
+    {
+      "epoch": 4.761904761904762,
+      "eval_loss": 0.10030877590179443,
+      "eval_runtime": 22.0777,
+      "eval_samples_per_second": 22.647,
+      "eval_steps_per_second": 0.725,
+      "step": 2200
+    },
+    {
+      "epoch": 4.783549783549784,
+      "grad_norm": 0.41326892375946045,
+      "learning_rate": 9.049773755656109e-06,
+      "loss": 0.1131,
+      "step": 2210
+    },
+    {
+      "epoch": 4.805194805194805,
+      "grad_norm": 0.3018375039100647,
+      "learning_rate": 8.144796380090498e-06,
+      "loss": 0.1064,
+      "step": 2220
+    },
+    {
+      "epoch": 4.8268398268398265,
+      "grad_norm": 0.3863151967525482,
+      "learning_rate": 7.239819004524888e-06,
+      "loss": 0.1187,
+      "step": 2230
+    },
+    {
+      "epoch": 4.848484848484849,
+      "grad_norm": 0.47006744146347046,
+      "learning_rate": 6.334841628959276e-06,
+      "loss": 0.1073,
+      "step": 2240
+    },
+    {
+      "epoch": 4.87012987012987,
+      "grad_norm": 0.36669015884399414,
+      "learning_rate": 5.4298642533936655e-06,
+      "loss": 0.1098,
+      "step": 2250
+    },
+    {
+      "epoch": 4.891774891774892,
+      "grad_norm": 0.3454284965991974,
+      "learning_rate": 4.5248868778280546e-06,
+      "loss": 0.0987,
+      "step": 2260
+    },
+    {
+      "epoch": 4.913419913419913,
+      "grad_norm": 0.45213282108306885,
+      "learning_rate": 3.619909502262444e-06,
+      "loss": 0.1094,
+      "step": 2270
+    },
+    {
+      "epoch": 4.935064935064935,
+      "grad_norm": 0.3391339182853699,
+      "learning_rate": 2.7149321266968327e-06,
+      "loss": 0.1011,
+      "step": 2280
+    },
+    {
+      "epoch": 4.956709956709957,
+      "grad_norm": 0.3675156235694885,
+      "learning_rate": 1.809954751131222e-06,
+      "loss": 0.1143,
+      "step": 2290
+    },
+    {
+      "epoch": 4.978354978354979,
+      "grad_norm": 0.3254551589488983,
+      "learning_rate": 9.04977375565611e-07,
+      "loss": 0.1123,
+      "step": 2300
+    },
+    {
+      "epoch": 4.978354978354979,
+      "eval_loss": 0.09990464895963669,
+      "eval_runtime": 21.2785,
+      "eval_samples_per_second": 23.498,
+      "eval_steps_per_second": 0.752,
+      "step": 2300
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.34439313411712646,
+      "learning_rate": 0.0,
+      "loss": 0.1115,
+      "step": 2310
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2310,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 7.35957960819826e+18,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2310/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:643e95e5ac406cc2ed73fee78c57d46adfe11ecc231393466b4f17abf1f17fbb
+size 5368

git_hash.txt

@@ -0,0 +1 @@
+258a8d9a4ce209476594fb23c349a485eb5fa068

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 802816,
+  "merge_size": 2,
+  "min_pixels": 3136,
+  "patch_size": 14,
+  "processor_class": "ColQwen2Processor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 802816,
+    "shortest_edge": 3136
+  },
+  "temporal_patch_size": 2
+}

results.json

@@ -0,0 +1,588 @@
+{
+    "metadata": {
+        "timestamp": "2025-04-07T13:42:49.983941",
+        "vidore_benchmark_version": "5.0.1.dev6+g9e0da63"
+    },
+    "metrics": {
+        "./data_dir/eval_vidore/syntheticDocQA_artificial_intelligence_test": {
+            "ndcg_at_1": 1.0,
+            "ndcg_at_3": 1.0,
+            "ndcg_at_5": 1.0,
+            "ndcg_at_10": 1.0,
+            "ndcg_at_20": 1.0,
+            "ndcg_at_50": 1.0,
+            "ndcg_at_100": 1.0,
+            "map_at_1": 1.0,
+            "map_at_3": 1.0,
+            "map_at_5": 1.0,
+            "map_at_10": 1.0,
+            "map_at_20": 1.0,
+            "map_at_50": 1.0,
+            "map_at_100": 1.0,
+            "recall_at_1": 1.0,
+            "recall_at_3": 1.0,
+            "recall_at_5": 1.0,
+            "recall_at_10": 1.0,
+            "recall_at_20": 1.0,
+            "recall_at_50": 1.0,
+            "recall_at_100": 1.0,
+            "precision_at_1": 1.0,
+            "precision_at_3": 0.33333,
+            "precision_at_5": 0.2,
+            "precision_at_10": 0.1,
+            "precision_at_20": 0.05,
+            "precision_at_50": 0.02,
+            "precision_at_100": 0.01,
+            "mrr_at_1": 1.0,
+            "mrr_at_3": 1.0,
+            "mrr_at_5": 1.0,
+            "mrr_at_10": 1.0,
+            "mrr_at_20": 1.0,
+            "mrr_at_50": 1.0,
+            "mrr_at_100": 1.0,
+            "naucs_at_1_max": null,
+            "naucs_at_1_std": null,
+            "naucs_at_1_diff1": null,
+            "naucs_at_3_max": 1.0,
+            "naucs_at_3_std": 1.0,
+            "naucs_at_3_diff1": 1.0,
+            "naucs_at_5_max": 1.0,
+            "naucs_at_5_std": 1.0,
+            "naucs_at_5_diff1": 1.0,
+            "naucs_at_10_max": 1.0,
+            "naucs_at_10_std": 1.0,
+            "naucs_at_10_diff1": 1.0,
+            "naucs_at_20_max": 1.0,
+            "naucs_at_20_std": 1.0,
+            "naucs_at_20_diff1": 1.0,
+            "naucs_at_50_max": null,
+            "naucs_at_50_std": null,
+            "naucs_at_50_diff1": null,
+            "naucs_at_100_max": null,
+            "naucs_at_100_std": null,
+            "naucs_at_100_diff1": null
+        },
+        "./data_dir/eval_vidore/syntheticDocQA_energy_test": {
+            "ndcg_at_1": 0.96,
+            "ndcg_at_3": 0.96631,
+            "ndcg_at_5": 0.97062,
+            "ndcg_at_10": 0.97062,
+            "ndcg_at_20": 0.97602,
+            "ndcg_at_50": 0.97602,
+            "ndcg_at_100": 0.97602,
+            "map_at_1": 0.96,
+            "map_at_3": 0.965,
+            "map_at_5": 0.9675,
+            "map_at_10": 0.9675,
+            "map_at_20": 0.96917,
+            "map_at_50": 0.96917,
+            "map_at_100": 0.96917,
+            "recall_at_1": 0.96,
+            "recall_at_3": 0.97,
+            "recall_at_5": 0.98,
+            "recall_at_10": 0.98,
+            "recall_at_20": 1.0,
+            "recall_at_50": 1.0,
+            "recall_at_100": 1.0,
+            "precision_at_1": 0.96,
+            "precision_at_3": 0.32333,
+            "precision_at_5": 0.196,
+            "precision_at_10": 0.098,
+            "precision_at_20": 0.05,
+            "precision_at_50": 0.02,
+            "precision_at_100": 0.01,
+            "mrr_at_1": 0.96,
+            "mrr_at_3": 0.965,
+            "mrr_at_5": 0.9675,
+            "mrr_at_10": 0.9675,
+            "mrr_at_20": 0.9693181818181819,
+            "mrr_at_50": 0.9693181818181819,
+            "mrr_at_100": 0.9693181818181819,
+            "naucs_at_1_max": 0.7700746965452845,
+            "naucs_at_1_std": -0.8232959850606876,
+            "naucs_at_1_diff1": 0.96732026143791,
+            "naucs_at_3_max": 0.7860255213196378,
+            "naucs_at_3_std": -1.21708683473389,
+            "naucs_at_3_diff1": 1.0,
+            "naucs_at_5_max": 0.6790382819794609,
+            "naucs_at_5_std": -1.7399626517273414,
+            "naucs_at_5_diff1": 1.0,
+            "naucs_at_10_max": 0.6790382819794609,
+            "naucs_at_10_std": -1.7399626517273414,
+            "naucs_at_10_diff1": 1.0,
+            "naucs_at_20_max": 1.0,
+            "naucs_at_20_std": 1.0,
+            "naucs_at_20_diff1": 1.0,
+            "naucs_at_50_max": null,
+            "naucs_at_50_std": null,
+            "naucs_at_50_diff1": null,
+            "naucs_at_100_max": null,
+            "naucs_at_100_std": null,
+            "naucs_at_100_diff1": null
+        },
+        "./data_dir/eval_vidore/tatdqa_test": {
+            "ndcg_at_1": 0.70352,
+            "ndcg_at_3": 0.79986,
+            "ndcg_at_5": 0.81641,
+            "ndcg_at_10": 0.83016,
+            "ndcg_at_20": 0.83437,
+            "ndcg_at_50": 0.8377,
+            "ndcg_at_100": 0.83916,
+            "map_at_1": 0.70352,
+            "map_at_3": 0.77643,
+            "map_at_5": 0.78578,
+            "map_at_10": 0.79157,
+            "map_at_20": 0.79277,
+            "map_at_50": 0.79328,
+            "map_at_100": 0.7934,
+            "recall_at_1": 0.70352,
+            "recall_at_3": 0.86756,
+            "recall_at_5": 0.90705,
+            "recall_at_10": 0.94897,
+            "recall_at_20": 0.96537,
+            "recall_at_50": 0.98238,
+            "recall_at_100": 0.99149,
+            "precision_at_1": 0.70352,
+            "precision_at_3": 0.28919,
+            "precision_at_5": 0.18141,
+            "precision_at_10": 0.0949,
+            "precision_at_20": 0.04827,
+            "precision_at_50": 0.01965,
+            "precision_at_100": 0.00991,
+            "mrr_at_1": 0.7004860267314702,
+            "mrr_at_3": 0.7765289590927501,
+            "mrr_at_5": 0.7844268934791414,
+            "mrr_at_10": 0.790128787440452,
+            "mrr_at_20": 0.7913423674323303,
+            "mrr_at_50": 0.7919207273733465,
+            "mrr_at_100": 0.792037412007671,
+            "naucs_at_1_max": 0.30712069371146206,
+            "naucs_at_1_std": -0.14877076174901377,
+            "naucs_at_1_diff1": 0.8349485938382216,
+            "naucs_at_3_max": 0.3678958030568473,
+            "naucs_at_3_std": -0.02507864846676702,
+            "naucs_at_3_diff1": 0.7517064585275754,
+            "naucs_at_5_max": 0.44253146077029726,
+            "naucs_at_5_std": 0.11026281660700264,
+            "naucs_at_5_diff1": 0.7300454594518606,
+            "naucs_at_10_max": 0.5290095815470773,
+            "naucs_at_10_std": 0.33727621290447424,
+            "naucs_at_10_diff1": 0.665454536221219,
+            "naucs_at_20_max": 0.5525939398718661,
+            "naucs_at_20_std": 0.44184740259085264,
+            "naucs_at_20_diff1": 0.6237746503620547,
+            "naucs_at_50_max": 0.6332654079181073,
+            "naucs_at_50_std": 0.5148901808157141,
+            "naucs_at_50_diff1": 0.7220173231390383,
+            "naucs_at_100_max": 0.7592562052200388,
+            "naucs_at_100_std": 0.6908846813166498,
+            "naucs_at_100_diff1": 0.7670992011544225
+        },
+        "./data_dir/eval_vidore/shiftproject_test": {
+            "ndcg_at_1": 0.77,
+            "ndcg_at_3": 0.87702,
+            "ndcg_at_5": 0.8895,
+            "ndcg_at_10": 0.8895,
+            "ndcg_at_20": 0.8895,
+            "ndcg_at_50": 0.89367,
+            "ndcg_at_100": 0.89367,
+            "map_at_1": 0.77,
+            "map_at_3": 0.85167,
+            "map_at_5": 0.85867,
+            "map_at_10": 0.85867,
+            "map_at_20": 0.85867,
+            "map_at_50": 0.85942,
+            "map_at_100": 0.85942,
+            "recall_at_1": 0.77,
+            "recall_at_3": 0.95,
+            "recall_at_5": 0.98,
+            "recall_at_10": 0.98,
+            "recall_at_20": 0.98,
+            "recall_at_50": 1.0,
+            "recall_at_100": 1.0,
+            "precision_at_1": 0.77,
+            "precision_at_3": 0.31667,
+            "precision_at_5": 0.196,
+            "precision_at_10": 0.098,
+            "precision_at_20": 0.049,
+            "precision_at_50": 0.02,
+            "precision_at_100": 0.01,
+            "mrr_at_1": 0.77,
+            "mrr_at_3": 0.8516666666666667,
+            "mrr_at_5": 0.8586666666666666,
+            "mrr_at_10": 0.8586666666666666,
+            "mrr_at_20": 0.8586666666666666,
+            "mrr_at_50": 0.8594585921325052,
+            "mrr_at_100": 0.8594585921325052,
+            "naucs_at_1_max": -0.027170385622524204,
+            "naucs_at_1_std": -0.41825384391575815,
+            "naucs_at_1_diff1": 0.7566527159195185,
+            "naucs_at_3_max": -0.22203548085900757,
+            "naucs_at_3_std": -0.38478057889822304,
+            "naucs_at_3_diff1": 0.7399626517273589,
+            "naucs_at_5_max": -0.43534080298785716,
+            "naucs_at_5_std": -0.5088702147525547,
+            "naucs_at_5_diff1": 0.9346405228758136,
+            "naucs_at_10_max": -0.43534080298785716,
+            "naucs_at_10_std": -0.5088702147525547,
+            "naucs_at_10_diff1": 0.9346405228758136,
+            "naucs_at_20_max": -0.43534080298785716,
+            "naucs_at_20_std": -0.5088702147525547,
+            "naucs_at_20_diff1": 0.9346405228758136,
+            "naucs_at_50_max": null,
+            "naucs_at_50_std": null,
+            "naucs_at_50_diff1": null,
+            "naucs_at_100_max": null,
+            "naucs_at_100_std": null,
+            "naucs_at_100_diff1": null
+        },
+        "./data_dir/eval_vidore/syntheticDocQA_healthcare_industry_test": {
+            "ndcg_at_1": 0.97,
+            "ndcg_at_3": 0.98893,
+            "ndcg_at_5": 0.98893,
+            "ndcg_at_10": 0.98893,
+            "ndcg_at_20": 0.98893,
+            "ndcg_at_50": 0.98893,
+            "ndcg_at_100": 0.98893,
+            "map_at_1": 0.97,
+            "map_at_3": 0.985,
+            "map_at_5": 0.985,
+            "map_at_10": 0.985,
+            "map_at_20": 0.985,
+            "map_at_50": 0.985,
+            "map_at_100": 0.985,
+            "recall_at_1": 0.97,
+            "recall_at_3": 1.0,
+            "recall_at_5": 1.0,
+            "recall_at_10": 1.0,
+            "recall_at_20": 1.0,
+            "recall_at_50": 1.0,
+            "recall_at_100": 1.0,
+            "precision_at_1": 0.97,
+            "precision_at_3": 0.33333,
+            "precision_at_5": 0.2,
+            "precision_at_10": 0.1,
+            "precision_at_20": 0.05,
+            "precision_at_50": 0.02,
+            "precision_at_100": 0.01,
+            "mrr_at_1": 0.98,
+            "mrr_at_3": 0.99,
+            "mrr_at_5": 0.99,
+            "mrr_at_10": 0.99,
+            "mrr_at_20": 0.99,
+            "mrr_at_50": 0.99,
+            "mrr_at_100": 0.99,
+            "naucs_at_1_max": 0.8078120136943662,
+            "naucs_at_1_std": -0.10597572362278272,
+            "naucs_at_1_diff1": 0.9564270152505465,
+            "naucs_at_3_max": 1.0,
+            "naucs_at_3_std": 1.0,
+            "naucs_at_3_diff1": 1.0,
+            "naucs_at_5_max": 1.0,
+            "naucs_at_5_std": 1.0,
+            "naucs_at_5_diff1": 1.0,
+            "naucs_at_10_max": 1.0,
+            "naucs_at_10_std": 1.0,
+            "naucs_at_10_diff1": 1.0,
+            "naucs_at_20_max": 1.0,
+            "naucs_at_20_std": 1.0,
+            "naucs_at_20_diff1": 1.0,
+            "naucs_at_50_max": null,
+            "naucs_at_50_std": null,
+            "naucs_at_50_diff1": null,
+            "naucs_at_100_max": null,
+            "naucs_at_100_std": null,
+            "naucs_at_100_diff1": null
+        },
+        "./data_dir/eval_vidore/syntheticDocQA_government_reports_test": {
+            "ndcg_at_1": 0.94,
+            "ndcg_at_3": 0.96524,
+            "ndcg_at_5": 0.97385,
+            "ndcg_at_10": 0.97385,
+            "ndcg_at_20": 0.97385,
+            "ndcg_at_50": 0.97385,
+            "ndcg_at_100": 0.97385,
+            "map_at_1": 0.94,
+            "map_at_3": 0.96,
+            "map_at_5": 0.965,
+            "map_at_10": 0.965,
+            "map_at_20": 0.965,
+            "map_at_50": 0.965,
+            "map_at_100": 0.965,
+            "recall_at_1": 0.94,
+            "recall_at_3": 0.98,
+            "recall_at_5": 1.0,
+            "recall_at_10": 1.0,
+            "recall_at_20": 1.0,
+            "recall_at_50": 1.0,
+            "recall_at_100": 1.0,
+            "precision_at_1": 0.94,
+            "precision_at_3": 0.32667,
+            "precision_at_5": 0.2,
+            "precision_at_10": 0.1,
+            "precision_at_20": 0.05,
+            "precision_at_50": 0.02,
+            "precision_at_100": 0.01,
+            "mrr_at_1": 0.95,
+            "mrr_at_3": 0.965,
+            "mrr_at_5": 0.97,
+            "mrr_at_10": 0.97,
+            "mrr_at_20": 0.97,
+            "mrr_at_50": 0.97,
+            "mrr_at_100": 0.97,
+            "naucs_at_1_max": 0.7857142857142844,
+            "naucs_at_1_std": 0.5056022408963584,
+            "naucs_at_1_diff1": 0.9319172113289763,
+            "naucs_at_3_max": 0.8611111111111119,
+            "naucs_at_3_std": 0.5401493930905577,
+            "naucs_at_3_diff1": 1.0,
+            "naucs_at_5_max": 1.0,
+            "naucs_at_5_std": 1.0,
+            "naucs_at_5_diff1": 1.0,
+            "naucs_at_10_max": 1.0,
+            "naucs_at_10_std": 1.0,
+            "naucs_at_10_diff1": 1.0,
+            "naucs_at_20_max": 1.0,
+            "naucs_at_20_std": 1.0,
+            "naucs_at_20_diff1": 1.0,
+            "naucs_at_50_max": null,
+            "naucs_at_50_std": null,
+            "naucs_at_50_diff1": null,
+            "naucs_at_100_max": null,
+            "naucs_at_100_std": null,
+            "naucs_at_100_diff1": null
+        },
+        "./data_dir/eval_vidore/docvqa_test_subsampled": {
+            "ndcg_at_1": 0.54767,
+            "ndcg_at_3": 0.60458,
+            "ndcg_at_5": 0.62758,
+            "ndcg_at_10": 0.64623,
+            "ndcg_at_20": 0.65727,
+            "ndcg_at_50": 0.66313,
+            "ndcg_at_100": 0.66931,
+            "map_at_1": 0.54767,
+            "map_at_3": 0.59054,
+            "map_at_5": 0.6034,
+            "map_at_10": 0.6111,
+            "map_at_20": 0.61404,
+            "map_at_50": 0.61504,
+            "map_at_100": 0.6156,
+            "recall_at_1": 0.54767,
+            "recall_at_3": 0.64523,
+            "recall_at_5": 0.70067,
+            "recall_at_10": 0.75831,
+            "recall_at_20": 0.80266,
+            "recall_at_50": 0.83149,
+            "recall_at_100": 0.86918,
+            "precision_at_1": 0.54767,
+            "precision_at_3": 0.21508,
+            "precision_at_5": 0.14013,
+            "precision_at_10": 0.07583,
+            "precision_at_20": 0.04013,
+            "precision_at_50": 0.01663,
+            "precision_at_100": 0.00869,
+            "mrr_at_1": 0.549889135254989,
+            "mrr_at_3": 0.5916481892091647,
+            "mrr_at_5": 0.6031781226903179,
+            "mrr_at_10": 0.6112298947664802,
+            "mrr_at_20": 0.6144150798334311,
+            "mrr_at_50": 0.615365808840627,
+            "mrr_at_100": 0.6158082058718903,
+            "naucs_at_1_max": 0.26994818984174423,
+            "naucs_at_1_std": 0.5906240429505522,
+            "naucs_at_1_diff1": 0.9167196184748201,
+            "naucs_at_3_max": 0.23789664157644613,
+            "naucs_at_3_std": 0.7236619387134813,
+            "naucs_at_3_diff1": 0.8666053534752101,
+            "naucs_at_5_max": 0.11514001916616068,
+            "naucs_at_5_std": 0.7631042600804109,
+            "naucs_at_5_diff1": 0.8479060689502268,
+            "naucs_at_10_max": -0.043203523595493495,
+            "naucs_at_10_std": 0.862093103464622,
+            "naucs_at_10_diff1": 0.8564068404442808,
+            "naucs_at_20_max": -0.18212291788839394,
+            "naucs_at_20_std": 0.9249806242337382,
+            "naucs_at_20_diff1": 0.8845536894418387,
+            "naucs_at_50_max": -0.24037092450687947,
+            "naucs_at_50_std": 0.9368754047403369,
+            "naucs_at_50_diff1": 0.8812951566425166,
+            "naucs_at_100_max": -0.40683538353746845,
+            "naucs_at_100_std": 0.9294103270531832,
+            "naucs_at_100_diff1": 0.8702662400596464
+        },
+        "./data_dir/eval_vidore/infovqa_test_subsampled": {
+            "ndcg_at_1": 0.89069,
+            "ndcg_at_3": 0.91847,
+            "ndcg_at_5": 0.92536,
+            "ndcg_at_10": 0.92807,
+            "ndcg_at_20": 0.93014,
+            "ndcg_at_50": 0.93334,
+            "ndcg_at_100": 0.93435,
+            "map_at_1": 0.89069,
+            "map_at_3": 0.91194,
+            "map_at_5": 0.91589,
+            "map_at_10": 0.91706,
+            "map_at_20": 0.91765,
+            "map_at_50": 0.91815,
+            "map_at_100": 0.91825,
+            "recall_at_1": 0.89069,
+            "recall_at_3": 0.93725,
+            "recall_at_5": 0.95344,
+            "recall_at_10": 0.96154,
+            "recall_at_20": 0.96964,
+            "recall_at_50": 0.98583,
+            "recall_at_100": 0.9919,
+            "precision_at_1": 0.89069,
+            "precision_at_3": 0.31242,
+            "precision_at_5": 0.19069,
+            "precision_at_10": 0.09615,
+            "precision_at_20": 0.04848,
+            "precision_at_50": 0.01972,
+            "precision_at_100": 0.00992,
+            "mrr_at_1": 0.8866396761133604,
+            "mrr_at_3": 0.9089068825910931,
+            "mrr_at_5": 0.9132591093117408,
+            "mrr_at_10": 0.9144640447272027,
+            "mrr_at_20": 0.9150498220615949,
+            "mrr_at_50": 0.9156121077361352,
+            "mrr_at_100": 0.9156733347422276,
+            "naucs_at_1_max": 0.6888031159871755,
+            "naucs_at_1_std": -0.008505670520150402,
+            "naucs_at_1_diff1": 0.9449011245519812,
+            "naucs_at_3_max": 0.7723774804158382,
+            "naucs_at_3_std": 0.004652604931807598,
+            "naucs_at_3_diff1": 0.923100327986361,
+            "naucs_at_5_max": 0.905123471497624,
+            "naucs_at_5_std": 0.3134960067202525,
+            "naucs_at_5_diff1": 0.9368196598112606,
+            "naucs_at_10_max": 0.9232252314824217,
+            "naucs_at_10_std": 0.3879086911094102,
+            "naucs_at_10_diff1": 0.9372660570415042,
+            "naucs_at_20_max": 0.9455551744236316,
+            "naucs_at_20_std": 0.44590016086873197,
+            "naucs_at_20_diff1": 0.9564661819784096,
+            "naucs_at_50_max": 0.9813426494193118,
+            "naucs_at_50_std": 0.7183015996994482,
+            "naucs_at_50_diff1": 0.9626852988386237,
+            "naucs_at_100_max": 1.0,
+            "naucs_at_100_std": 0.8979159520443043,
+            "naucs_at_100_diff1": 0.9673496364838197
+        },
+        "./data_dir/eval_vidore/arxivqa_test_subsampled": {
+            "ndcg_at_1": 0.822,
+            "ndcg_at_3": 0.86707,
+            "ndcg_at_5": 0.87533,
+            "ndcg_at_10": 0.88497,
+            "ndcg_at_20": 0.89022,
+            "ndcg_at_50": 0.8951,
+            "ndcg_at_100": 0.8951,
+            "map_at_1": 0.822,
+            "map_at_3": 0.85633,
+            "map_at_5": 0.86093,
+            "map_at_10": 0.86488,
+            "map_at_20": 0.86642,
+            "map_at_50": 0.86725,
+            "map_at_100": 0.86725,
+            "recall_at_1": 0.822,
+            "recall_at_3": 0.898,
+            "recall_at_5": 0.918,
+            "recall_at_10": 0.948,
+            "recall_at_20": 0.968,
+            "recall_at_50": 0.992,
+            "recall_at_100": 0.992,
+            "precision_at_1": 0.822,
+            "precision_at_3": 0.29933,
+            "precision_at_5": 0.1836,
+            "precision_at_10": 0.0948,
+            "precision_at_20": 0.0484,
+            "precision_at_50": 0.01984,
+            "precision_at_100": 0.00992,
+            "mrr_at_1": 0.824,
+            "mrr_at_3": 0.858,
+            "mrr_at_5": 0.8622000000000001,
+            "mrr_at_10": 0.8664793650793651,
+            "mrr_at_20": 0.8681266741428506,
+            "mrr_at_50": 0.8688783518847116,
+            "mrr_at_100": 0.8688783518847116,
+            "naucs_at_1_max": 0.7534804527092446,
+            "naucs_at_1_std": -0.09711881948385728,
+            "naucs_at_1_diff1": 0.950227578762283,
+            "naucs_at_3_max": 0.8118793707898164,
+            "naucs_at_3_std": 0.022246287696627846,
+            "naucs_at_3_diff1": 0.907169471925993,
+            "naucs_at_5_max": 0.7977955409806211,
+            "naucs_at_5_std": -0.04402086037667595,
+            "naucs_at_5_diff1": 0.921887454168659,
+            "naucs_at_10_max": 0.9135423400129307,
+            "naucs_at_10_std": 0.09272426919485754,
+            "naucs_at_10_diff1": 0.9340120663650082,
+            "naucs_at_20_max": 0.9210725957049468,
+            "naucs_at_20_std": 0.19062208216619833,
+            "naucs_at_20_diff1": 0.9264705882352913,
+            "naucs_at_50_max": 0.9673202614378978,
+            "naucs_at_50_std": 0.20354808590102869,
+            "naucs_at_50_diff1": 0.9346405228758211,
+            "naucs_at_100_max": 0.9673202614378978,
+            "naucs_at_100_std": 0.20354808590102869,
+            "naucs_at_100_diff1": 0.9346405228758211
+        },
+        "./data_dir/eval_vidore/tabfquad_test_subsampled": {
+            "ndcg_at_1": 0.84286,
+            "ndcg_at_3": 0.89141,
+            "ndcg_at_5": 0.89847,
+            "ndcg_at_10": 0.90417,
+            "ndcg_at_20": 0.91057,
+            "ndcg_at_50": 0.9134,
+            "ndcg_at_100": 0.9134,
+            "map_at_1": 0.84286,
+            "map_at_3": 0.87976,
+            "map_at_5": 0.88351,
+            "map_at_10": 0.88582,
+            "map_at_20": 0.88761,
+            "map_at_50": 0.88806,
+            "map_at_100": 0.88806,
+            "recall_at_1": 0.84286,
+            "recall_at_3": 0.925,
+            "recall_at_5": 0.94286,
+            "recall_at_10": 0.96071,
+            "recall_at_20": 0.98571,
+            "recall_at_50": 1.0,
+            "recall_at_100": 1.0,
+            "precision_at_1": 0.84286,
+            "precision_at_3": 0.30833,
+            "precision_at_5": 0.18857,
+            "precision_at_10": 0.09607,
+            "precision_at_20": 0.04929,
+            "precision_at_50": 0.02,
+            "precision_at_100": 0.01,
+            "mrr_at_1": 0.8428571428571429,
+            "mrr_at_3": 0.8797619047619047,
+            "mrr_at_5": 0.8827976190476191,
+            "mrr_at_10": 0.8856746031746031,
+            "mrr_at_20": 0.887474271759986,
+            "mrr_at_50": 0.8879308740437567,
+            "mrr_at_100": 0.8879308740437567,
+            "naucs_at_1_max": 0.5389336957490425,
+            "naucs_at_1_std": 0.2014303538009097,
+            "naucs_at_1_diff1": 0.8885014704571788,
+            "naucs_at_3_max": 0.6380329909741672,
+            "naucs_at_3_std": 0.4563603219065409,
+            "naucs_at_3_diff1": 0.8615223867324734,
+            "naucs_at_5_max": 0.5759803921568608,
+            "naucs_at_5_std": 0.40143557422968984,
+            "naucs_at_5_diff1": 0.8345880018674129,
+            "naucs_at_10_max": 0.823444529326883,
+            "naucs_at_10_std": 0.626602156013919,
+            "naucs_at_10_diff1": 0.8355827179356624,
+            "naucs_at_20_max": 0.967320261437913,
+            "naucs_at_20_std": 0.8284313725490264,
+            "naucs_at_20_diff1": 0.9346405228758147,
+            "naucs_at_50_max": 1.0,
+            "naucs_at_50_std": 1.0,
+            "naucs_at_50_diff1": 1.0,
+            "naucs_at_100_max": 1.0,
+            "naucs_at_100_std": 1.0,
+            "naucs_at_100_diff1": 1.0
+        }
+    }
+}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091aa7594dc2fcfbfa06b9e3c22a5f0562ac14f30375c13af7309407a0e67b8a
+size 11420371

tokenizer_config.json

@@ -0,0 +1,146 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "max_num_visual_tokens": 1024,
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "processor_class": "ColQwen2Processor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

training_config.yml

@@ -0,0 +1,63 @@
+config:
+  (): colpali_engine.trainer.colmodel_training.ColModelTrainingConfig
+  output_dir: !path ../../../models/colqwen2-ba256-3e-0304
+  processor:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2Processor
+    pretrained_model_name_or_path:  "./models/base_models/colqwen2-base"
+    max_num_visual_tokens: 1024
+
+  model:
+    (): colpali_engine.utils.transformers_wrappers.AllPurposeWrapper
+    class_to_instanciate: !ext colpali_engine.models.ColQwen2
+    pretrained_model_name_or_path: "./models/base_models/colqwen2-base"
+    torch_dtype:  !ext torch.bfloat16
+    use_cache: false
+    attn_implementation: "flash_attention_2"
+
+  dataset_loading_func: !ext colpali_engine.utils.dataset_transformation.load_train_set
+  eval_dataset_loader: !import ../data/test_data.yaml
+
+  # max_length: 50
+  run_eval: true
+  loss_func:
+    (): colpali_engine.loss.late_interaction_losses.ColbertPairwiseCELoss
+  tr_args:
+    (): transformers.training_args.TrainingArguments
+    output_dir: null
+    overwrite_output_dir: true
+    num_train_epochs: 5
+    per_device_train_batch_size: 64
+    gradient_checkpointing: true
+    gradient_checkpointing_kwargs: { "use_reentrant": false }
+    # 6 x 8 gpus = 48 batch size
+    # gradient_accumulation_steps: 4
+    per_device_eval_batch_size: 8
+    eval_strategy: "steps"
+    dataloader_num_workers: 8
+    # bf16: true
+    save_steps: 500
+    logging_steps: 10
+    eval_steps: 100
+    warmup_steps: 100
+    learning_rate: 2e-4
+    save_total_limit: 1
+    # resume_from_checkpoint: true
+    # optim: "paged_adamw_8bit"
+    # wandb logging
+    # wandb_project: "colqwen2"
+    # run_name: "colqwen2-ba32-nolora"
+    report_to: "wandb"
+
+
+  peft_config:
+    (): peft.LoraConfig
+    r: 32
+    lora_alpha: 32
+    lora_dropout: 0.1
+    init_lora_weights: "gaussian"
+    bias: "none"
+    task_type: "FEATURE_EXTRACTION"
+    target_modules: '(.*(model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+    # target_modules: '(.*(language_model).*(down_proj|gate_proj|up_proj|k_proj|q_proj|v_proj|o_proj).*$|.*(custom_text_proj).*$)'
+