Hugging Face's logo

AabirDey
/
QA-OpenLlama-test-1

PEFT
TensorBoard
Safetensors
llama
Generated from Trainer
8-bit precision
bitsandbytes
Model card Files Metrics Community
AabirDey commited on
Commit
844340e
·
verified ·
1 Parent(s): 55f65f4

Upload 16 files

Browse files
Files changed (16) hide show
  1. README.md +144 -3
  2. adapter_config.json +33 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-527/README.md +202 -0
  5. checkpoint-527/adapter_config.json +33 -0
  6. checkpoint-527/adapter_model.safetensors +3 -0
  7. checkpoint-527/optimizer.pt +3 -0
  8. checkpoint-527/rng_state.pth +3 -0
  9. checkpoint-527/scheduler.pt +3 -0
  10. checkpoint-527/trainer_state.json +3742 -0
  11. checkpoint-527/training_args.bin +3 -0
  12. config.json +43 -0
  13. runs/Mar09_05-23-07_9b5078085e9b/events.out.tfevents.1709961787.9b5078085e9b.1472.0 +3 -0
  14. special_tokens_map.json +24 -0
  15. tokenizer.model +3 -0
  16. tokenizer_config.json +43 -0
README.md CHANGED
@@ -1,3 +1,144 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - generated_from_trainer
6
+ base_model: openlm-research/open_llama_3b_v2
7
+ model-index:
8
+ - name: lora-out
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ [<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
16
+ <details><summary>See axolotl config</summary>
17
+
18
+ axolotl version: `0.4.0`
19
+ ```yaml
20
+ base_model: openlm-research/open_llama_3b_v2
21
+ model_type: LlamaForCausalLM
22
+ tokenizer_type: LlamaTokenizer
23
+ load_in_8bit: true
24
+ load_in_4bit: false
25
+ strict: false
26
+ push_dataset_to_hub:
27
+ datasets:
28
+ - path: AabirDey/job-queries-and-customer-service
29
+ type: alpaca
30
+ dataset_prepared_path:
31
+ val_set_size: 0.02
32
+ adapter: lora
33
+ lora_model_dir:
34
+ sequence_len: 1024
35
+ sample_packing: true
36
+ lora_r: 8
37
+ lora_alpha: 16
38
+ lora_dropout: 0.0
39
+ lora_target_modules:
40
+ - gate_proj
41
+ - down_proj
42
+ - up_proj
43
+ - q_proj
44
+ - v_proj
45
+ - k_proj
46
+ - o_proj
47
+ lora_fan_in_fan_out:
48
+ wandb_project:
49
+ wandb_entity:
50
+ wandb_watch:
51
+ wandb_name:
52
+ wandb_log_model:
53
+ output_dir: ./lora-out
54
+ gradient_accumulation_steps: 8
55
+ micro_batch_size: 1
56
+ num_epochs: 1
57
+ optimizer: adamw_bnb_8bit
58
+ torchdistx_path:
59
+ lr_scheduler: cosine
60
+ learning_rate: 0.0002
61
+ train_on_inputs: false
62
+ group_by_length: false
63
+ bf16: false
64
+ fp16: true
65
+ tf32: false
66
+ gradient_checkpointing: true
67
+ early_stopping_patience:
68
+ resume_from_checkpoint:
69
+ local_rank:
70
+ logging_steps: 1
71
+ xformers_attention:
72
+ flash_attention: true
73
+ gptq_groupsize:
74
+ s2_attention:
75
+ gptq_model_v1:
76
+ warmup_steps: 20
77
+ evals_per_epoch: 4
78
+ saves_per_epoch: 1
79
+ debug:
80
+ deepspeed:
81
+ weight_decay: 0.1
82
+ fsdp:
83
+ fsdp_config:
84
+ special_tokens:
85
+ bos_token: "<s>"
86
+ eos_token: "</s>"
87
+ unk_token: "<unk>"
88
+
89
+ ```
90
+
91
+ </details><br>
92
+
93
+ # lora-out
94
+
95
+ This model is a fine-tuned version of [openlm-research/open_llama_3b_v2](https://huggingface.co/openlm-research/open_llama_3b_v2) on the None dataset.
96
+ It achieves the following results on the evaluation set:
97
+ - Loss: 0.6555
98
+
99
+ ## Model description
100
+
101
+ More information needed
102
+
103
+ ## Intended uses & limitations
104
+
105
+ More information needed
106
+
107
+ ## Training and evaluation data
108
+
109
+ More information needed
110
+
111
+ ## Training procedure
112
+
113
+ ### Training hyperparameters
114
+
115
+ The following hyperparameters were used during training:
116
+ - learning_rate: 0.0002
117
+ - train_batch_size: 1
118
+ - eval_batch_size: 1
119
+ - seed: 42
120
+ - gradient_accumulation_steps: 8
121
+ - total_train_batch_size: 8
122
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
123
+ - lr_scheduler_type: cosine
124
+ - lr_scheduler_warmup_steps: 20
125
+ - num_epochs: 1
126
+ - mixed_precision_training: Native AMP
127
+
128
+ ### Training results
129
+
130
+ | Training Loss | Epoch | Step | Validation Loss |
131
+ |:-------------:|:-----:|:----:|:---------------:|
132
+ | 2.0783 | 0.0 | 1 | 2.0402 |
133
+ | 0.8144 | 0.25 | 132 | 0.8365 |
134
+ | 0.7159 | 0.5 | 264 | 0.7109 |
135
+ | 0.6664 | 0.75 | 396 | 0.6555 |
136
+
137
+
138
+ ### Framework versions
139
+
140
+ - PEFT 0.9.0
141
+ - Transformers 4.38.2
142
+ - Pytorch 2.2.0+cu118
143
+ - Datasets 2.18.0
144
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openlm-research/open_llama_3b_v2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 16,
13
+ "lora_dropout": 0.0,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "up_proj",
23
+ "k_proj",
24
+ "o_proj",
25
+ "gate_proj",
26
+ "q_proj",
27
+ "down_proj",
28
+ "v_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70c52922453b15fec25d5828641908edf859b7a77b86b4a20a0a77c2be5a7f59
3
+ size 50982842
checkpoint-527/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: openlm-research/open_llama_3b_v2
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- 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. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ 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).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.9.0
checkpoint-527/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openlm-research/open_llama_3b_v2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 16,
13
+ "lora_dropout": 0.0,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "up_proj",
23
+ "k_proj",
24
+ "o_proj",
25
+ "gate_proj",
26
+ "q_proj",
27
+ "down_proj",
28
+ "v_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
checkpoint-527/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8295bcd57cd2e395157ae145d8ed4e37b19438151a56f46ab611c050802dacb7
3
+ size 50899792
checkpoint-527/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6bbf9f0256d3259a97ee0a6a827ac3ad2f9d5eda4141c61b208d5a73e75882d4
3
+ size 25871876
checkpoint-527/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b3ee827a7a00012c0a116546df467feee35e70376d81a7a85b1a70eb90414d3
3
+ size 14244
checkpoint-527/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c657142c52d00cceef4d96f4405b39cd8ef2c1725c2a0d85b5a5d6841cabbb43
3
+ size 1064
checkpoint-527/trainer_state.json ADDED
@@ -0,0 +1,3742 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9997628645956841,
5
+ "eval_steps": 132,
6
+ "global_step": 527,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "grad_norm": 0.5150582194328308,
14
+ "learning_rate": 1e-05,
15
+ "loss": 2.0783,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0,
20
+ "eval_loss": 2.0402109622955322,
21
+ "eval_runtime": 14.6647,
22
+ "eval_samples_per_second": 32.186,
23
+ "eval_steps_per_second": 32.186,
24
+ "step": 1
25
+ },
26
+ {
27
+ "epoch": 0.0,
28
+ "grad_norm": 0.3772180378437042,
29
+ "learning_rate": 2e-05,
30
+ "loss": 1.9188,
31
+ "step": 2
32
+ },
33
+ {
34
+ "epoch": 0.01,
35
+ "grad_norm": 0.43772926926612854,
36
+ "learning_rate": 3e-05,
37
+ "loss": 2.015,
38
+ "step": 3
39
+ },
40
+ {
41
+ "epoch": 0.01,
42
+ "grad_norm": 0.40814539790153503,
43
+ "learning_rate": 4e-05,
44
+ "loss": 1.971,
45
+ "step": 4
46
+ },
47
+ {
48
+ "epoch": 0.01,
49
+ "grad_norm": 0.3195774555206299,
50
+ "learning_rate": 5e-05,
51
+ "loss": 1.8184,
52
+ "step": 5
53
+ },
54
+ {
55
+ "epoch": 0.01,
56
+ "grad_norm": 0.4951046109199524,
57
+ "learning_rate": 6e-05,
58
+ "loss": 2.0699,
59
+ "step": 6
60
+ },
61
+ {
62
+ "epoch": 0.01,
63
+ "grad_norm": 0.5869188904762268,
64
+ "learning_rate": 7e-05,
65
+ "loss": 2.1714,
66
+ "step": 7
67
+ },
68
+ {
69
+ "epoch": 0.02,
70
+ "grad_norm": 0.3898499608039856,
71
+ "learning_rate": 8e-05,
72
+ "loss": 1.8471,
73
+ "step": 8
74
+ },
75
+ {
76
+ "epoch": 0.02,
77
+ "grad_norm": 0.4388677179813385,
78
+ "learning_rate": 9e-05,
79
+ "loss": 1.7862,
80
+ "step": 9
81
+ },
82
+ {
83
+ "epoch": 0.02,
84
+ "grad_norm": 0.6679179072380066,
85
+ "learning_rate": 0.0001,
86
+ "loss": 1.664,
87
+ "step": 10
88
+ },
89
+ {
90
+ "epoch": 0.02,
91
+ "grad_norm": 0.5442260503768921,
92
+ "learning_rate": 0.00011000000000000002,
93
+ "loss": 1.8824,
94
+ "step": 11
95
+ },
96
+ {
97
+ "epoch": 0.02,
98
+ "grad_norm": 0.7661851644515991,
99
+ "learning_rate": 0.00012,
100
+ "loss": 2.0104,
101
+ "step": 12
102
+ },
103
+ {
104
+ "epoch": 0.02,
105
+ "grad_norm": 0.47241243720054626,
106
+ "learning_rate": 0.00013000000000000002,
107
+ "loss": 1.5412,
108
+ "step": 13
109
+ },
110
+ {
111
+ "epoch": 0.03,
112
+ "grad_norm": 0.852836012840271,
113
+ "learning_rate": 0.00014,
114
+ "loss": 1.6664,
115
+ "step": 14
116
+ },
117
+ {
118
+ "epoch": 0.03,
119
+ "grad_norm": 0.833349883556366,
120
+ "learning_rate": 0.00015000000000000001,
121
+ "loss": 1.5698,
122
+ "step": 15
123
+ },
124
+ {
125
+ "epoch": 0.03,
126
+ "grad_norm": 0.7689324021339417,
127
+ "learning_rate": 0.00016,
128
+ "loss": 1.6071,
129
+ "step": 16
130
+ },
131
+ {
132
+ "epoch": 0.03,
133
+ "grad_norm": 0.8255270719528198,
134
+ "learning_rate": 0.00017,
135
+ "loss": 1.5949,
136
+ "step": 17
137
+ },
138
+ {
139
+ "epoch": 0.03,
140
+ "grad_norm": 0.6706920266151428,
141
+ "learning_rate": 0.00018,
142
+ "loss": 1.477,
143
+ "step": 18
144
+ },
145
+ {
146
+ "epoch": 0.04,
147
+ "grad_norm": 0.6380877494812012,
148
+ "learning_rate": 0.00019,
149
+ "loss": 1.3816,
150
+ "step": 19
151
+ },
152
+ {
153
+ "epoch": 0.04,
154
+ "grad_norm": 1.2489427328109741,
155
+ "learning_rate": 0.0002,
156
+ "loss": 1.3171,
157
+ "step": 20
158
+ },
159
+ {
160
+ "epoch": 0.04,
161
+ "grad_norm": 0.7026221752166748,
162
+ "learning_rate": 0.0001999980802156745,
163
+ "loss": 1.5759,
164
+ "step": 21
165
+ },
166
+ {
167
+ "epoch": 0.04,
168
+ "grad_norm": 0.5911536812782288,
169
+ "learning_rate": 0.00019999232093640933,
170
+ "loss": 1.4688,
171
+ "step": 22
172
+ },
173
+ {
174
+ "epoch": 0.04,
175
+ "grad_norm": 0.576220691204071,
176
+ "learning_rate": 0.00019998272238333606,
177
+ "loss": 1.2945,
178
+ "step": 23
179
+ },
180
+ {
181
+ "epoch": 0.05,
182
+ "grad_norm": 0.5969688892364502,
183
+ "learning_rate": 0.0001999692849249977,
184
+ "loss": 1.3766,
185
+ "step": 24
186
+ },
187
+ {
188
+ "epoch": 0.05,
189
+ "grad_norm": 0.5913142561912537,
190
+ "learning_rate": 0.00019995200907733468,
191
+ "loss": 1.3836,
192
+ "step": 25
193
+ },
194
+ {
195
+ "epoch": 0.05,
196
+ "grad_norm": 0.7268506288528442,
197
+ "learning_rate": 0.000199930895503665,
198
+ "loss": 1.3697,
199
+ "step": 26
200
+ },
201
+ {
202
+ "epoch": 0.05,
203
+ "grad_norm": 0.5529304146766663,
204
+ "learning_rate": 0.00019990594501465884,
205
+ "loss": 1.221,
206
+ "step": 27
207
+ },
208
+ {
209
+ "epoch": 0.05,
210
+ "grad_norm": 0.5636203289031982,
211
+ "learning_rate": 0.0001998771585683074,
212
+ "loss": 1.341,
213
+ "step": 28
214
+ },
215
+ {
216
+ "epoch": 0.06,
217
+ "grad_norm": 0.5537460446357727,
218
+ "learning_rate": 0.000199844537269886,
219
+ "loss": 1.212,
220
+ "step": 29
221
+ },
222
+ {
223
+ "epoch": 0.06,
224
+ "grad_norm": 0.5263828039169312,
225
+ "learning_rate": 0.00019980808237191178,
226
+ "loss": 1.1306,
227
+ "step": 30
228
+ },
229
+ {
230
+ "epoch": 0.06,
231
+ "grad_norm": 0.5496954917907715,
232
+ "learning_rate": 0.0001997677952740956,
233
+ "loss": 1.3379,
234
+ "step": 31
235
+ },
236
+ {
237
+ "epoch": 0.06,
238
+ "grad_norm": 0.6726649403572083,
239
+ "learning_rate": 0.00019972367752328824,
240
+ "loss": 1.2297,
241
+ "step": 32
242
+ },
243
+ {
244
+ "epoch": 0.06,
245
+ "grad_norm": 0.5256686806678772,
246
+ "learning_rate": 0.00019967573081342103,
247
+ "loss": 1.1823,
248
+ "step": 33
249
+ },
250
+ {
251
+ "epoch": 0.06,
252
+ "grad_norm": 0.5452485084533691,
253
+ "learning_rate": 0.00019962395698544077,
254
+ "loss": 1.1396,
255
+ "step": 34
256
+ },
257
+ {
258
+ "epoch": 0.07,
259
+ "grad_norm": 0.5839616060256958,
260
+ "learning_rate": 0.00019956835802723916,
261
+ "loss": 1.1203,
262
+ "step": 35
263
+ },
264
+ {
265
+ "epoch": 0.07,
266
+ "grad_norm": 0.4983978867530823,
267
+ "learning_rate": 0.00019950893607357636,
268
+ "loss": 1.1478,
269
+ "step": 36
270
+ },
271
+ {
272
+ "epoch": 0.07,
273
+ "grad_norm": 0.6289787292480469,
274
+ "learning_rate": 0.00019944569340599912,
275
+ "loss": 1.1202,
276
+ "step": 37
277
+ },
278
+ {
279
+ "epoch": 0.07,
280
+ "grad_norm": 0.7068483233451843,
281
+ "learning_rate": 0.00019937863245275304,
282
+ "loss": 1.2123,
283
+ "step": 38
284
+ },
285
+ {
286
+ "epoch": 0.07,
287
+ "grad_norm": 0.655548632144928,
288
+ "learning_rate": 0.00019930775578868947,
289
+ "loss": 1.1072,
290
+ "step": 39
291
+ },
292
+ {
293
+ "epoch": 0.08,
294
+ "grad_norm": 0.6124732494354248,
295
+ "learning_rate": 0.0001992330661351665,
296
+ "loss": 1.2471,
297
+ "step": 40
298
+ },
299
+ {
300
+ "epoch": 0.08,
301
+ "grad_norm": 0.6181141138076782,
302
+ "learning_rate": 0.0001991545663599448,
303
+ "loss": 1.0698,
304
+ "step": 41
305
+ },
306
+ {
307
+ "epoch": 0.08,
308
+ "grad_norm": 0.5633079409599304,
309
+ "learning_rate": 0.00019907225947707704,
310
+ "loss": 1.0341,
311
+ "step": 42
312
+ },
313
+ {
314
+ "epoch": 0.08,
315
+ "grad_norm": 0.5386618375778198,
316
+ "learning_rate": 0.0001989861486467925,
317
+ "loss": 1.0025,
318
+ "step": 43
319
+ },
320
+ {
321
+ "epoch": 0.08,
322
+ "grad_norm": 0.7523472309112549,
323
+ "learning_rate": 0.00019889623717537564,
324
+ "loss": 1.1307,
325
+ "step": 44
326
+ },
327
+ {
328
+ "epoch": 0.09,
329
+ "grad_norm": 0.6467829942703247,
330
+ "learning_rate": 0.00019880252851503915,
331
+ "loss": 0.9503,
332
+ "step": 45
333
+ },
334
+ {
335
+ "epoch": 0.09,
336
+ "grad_norm": 0.6346300840377808,
337
+ "learning_rate": 0.00019870502626379127,
338
+ "loss": 1.1203,
339
+ "step": 46
340
+ },
341
+ {
342
+ "epoch": 0.09,
343
+ "grad_norm": 0.724151611328125,
344
+ "learning_rate": 0.00019860373416529802,
345
+ "loss": 1.1548,
346
+ "step": 47
347
+ },
348
+ {
349
+ "epoch": 0.09,
350
+ "grad_norm": 0.5642977356910706,
351
+ "learning_rate": 0.00019849865610873898,
352
+ "loss": 1.0182,
353
+ "step": 48
354
+ },
355
+ {
356
+ "epoch": 0.09,
357
+ "grad_norm": 0.6579070091247559,
358
+ "learning_rate": 0.0001983897961286583,
359
+ "loss": 1.0998,
360
+ "step": 49
361
+ },
362
+ {
363
+ "epoch": 0.09,
364
+ "grad_norm": 0.6359585523605347,
365
+ "learning_rate": 0.0001982771584048096,
366
+ "loss": 1.0401,
367
+ "step": 50
368
+ },
369
+ {
370
+ "epoch": 0.1,
371
+ "grad_norm": 0.6426641941070557,
372
+ "learning_rate": 0.00019816074726199565,
373
+ "loss": 1.0209,
374
+ "step": 51
375
+ },
376
+ {
377
+ "epoch": 0.1,
378
+ "grad_norm": 0.7192591428756714,
379
+ "learning_rate": 0.0001980405671699022,
380
+ "loss": 1.061,
381
+ "step": 52
382
+ },
383
+ {
384
+ "epoch": 0.1,
385
+ "grad_norm": 0.6428646445274353,
386
+ "learning_rate": 0.00019791662274292637,
387
+ "loss": 0.9462,
388
+ "step": 53
389
+ },
390
+ {
391
+ "epoch": 0.1,
392
+ "grad_norm": 0.7455520033836365,
393
+ "learning_rate": 0.00019778891873999954,
394
+ "loss": 1.0747,
395
+ "step": 54
396
+ },
397
+ {
398
+ "epoch": 0.1,
399
+ "grad_norm": 0.8657999038696289,
400
+ "learning_rate": 0.00019765746006440455,
401
+ "loss": 1.38,
402
+ "step": 55
403
+ },
404
+ {
405
+ "epoch": 0.11,
406
+ "grad_norm": 0.8126336336135864,
407
+ "learning_rate": 0.00019752225176358757,
408
+ "loss": 1.0305,
409
+ "step": 56
410
+ },
411
+ {
412
+ "epoch": 0.11,
413
+ "grad_norm": 0.5417783260345459,
414
+ "learning_rate": 0.00019738329902896403,
415
+ "loss": 1.0167,
416
+ "step": 57
417
+ },
418
+ {
419
+ "epoch": 0.11,
420
+ "grad_norm": 0.8542747497558594,
421
+ "learning_rate": 0.00019724060719571962,
422
+ "loss": 1.2036,
423
+ "step": 58
424
+ },
425
+ {
426
+ "epoch": 0.11,
427
+ "grad_norm": 0.7181089520454407,
428
+ "learning_rate": 0.0001970941817426052,
429
+ "loss": 1.06,
430
+ "step": 59
431
+ },
432
+ {
433
+ "epoch": 0.11,
434
+ "grad_norm": 0.6730545163154602,
435
+ "learning_rate": 0.00019694402829172663,
436
+ "loss": 0.9376,
437
+ "step": 60
438
+ },
439
+ {
440
+ "epoch": 0.12,
441
+ "grad_norm": 0.7448925971984863,
442
+ "learning_rate": 0.00019679015260832872,
443
+ "loss": 1.016,
444
+ "step": 61
445
+ },
446
+ {
447
+ "epoch": 0.12,
448
+ "grad_norm": 0.8230142593383789,
449
+ "learning_rate": 0.00019663256060057393,
450
+ "loss": 1.0177,
451
+ "step": 62
452
+ },
453
+ {
454
+ "epoch": 0.12,
455
+ "grad_norm": 0.6832039952278137,
456
+ "learning_rate": 0.0001964712583193156,
457
+ "loss": 0.9556,
458
+ "step": 63
459
+ },
460
+ {
461
+ "epoch": 0.12,
462
+ "grad_norm": 0.6915066242218018,
463
+ "learning_rate": 0.00019630625195786558,
464
+ "loss": 0.9383,
465
+ "step": 64
466
+ },
467
+ {
468
+ "epoch": 0.12,
469
+ "grad_norm": 0.9440682530403137,
470
+ "learning_rate": 0.0001961375478517564,
471
+ "loss": 1.1566,
472
+ "step": 65
473
+ },
474
+ {
475
+ "epoch": 0.13,
476
+ "grad_norm": 0.695676863193512,
477
+ "learning_rate": 0.000195965152478498,
478
+ "loss": 0.983,
479
+ "step": 66
480
+ },
481
+ {
482
+ "epoch": 0.13,
483
+ "grad_norm": 0.6894365549087524,
484
+ "learning_rate": 0.0001957890724573291,
485
+ "loss": 0.9134,
486
+ "step": 67
487
+ },
488
+ {
489
+ "epoch": 0.13,
490
+ "grad_norm": 0.6659775376319885,
491
+ "learning_rate": 0.00019560931454896298,
492
+ "loss": 0.8653,
493
+ "step": 68
494
+ },
495
+ {
496
+ "epoch": 0.13,
497
+ "grad_norm": 0.618377685546875,
498
+ "learning_rate": 0.00019542588565532799,
499
+ "loss": 0.9735,
500
+ "step": 69
501
+ },
502
+ {
503
+ "epoch": 0.13,
504
+ "grad_norm": 0.6974190473556519,
505
+ "learning_rate": 0.00019523879281930235,
506
+ "loss": 1.0011,
507
+ "step": 70
508
+ },
509
+ {
510
+ "epoch": 0.13,
511
+ "grad_norm": 0.6658157110214233,
512
+ "learning_rate": 0.000195048043224444,
513
+ "loss": 0.974,
514
+ "step": 71
515
+ },
516
+ {
517
+ "epoch": 0.14,
518
+ "grad_norm": 0.6543557643890381,
519
+ "learning_rate": 0.00019485364419471454,
520
+ "loss": 0.9153,
521
+ "step": 72
522
+ },
523
+ {
524
+ "epoch": 0.14,
525
+ "grad_norm": 0.8609347939491272,
526
+ "learning_rate": 0.00019465560319419824,
527
+ "loss": 0.9097,
528
+ "step": 73
529
+ },
530
+ {
531
+ "epoch": 0.14,
532
+ "grad_norm": 0.5851992964744568,
533
+ "learning_rate": 0.00019445392782681522,
534
+ "loss": 0.8875,
535
+ "step": 74
536
+ },
537
+ {
538
+ "epoch": 0.14,
539
+ "grad_norm": 0.6669757962226868,
540
+ "learning_rate": 0.00019424862583602965,
541
+ "loss": 0.9028,
542
+ "step": 75
543
+ },
544
+ {
545
+ "epoch": 0.14,
546
+ "grad_norm": 0.9252307415008545,
547
+ "learning_rate": 0.00019403970510455248,
548
+ "loss": 0.8723,
549
+ "step": 76
550
+ },
551
+ {
552
+ "epoch": 0.15,
553
+ "grad_norm": 0.6840870380401611,
554
+ "learning_rate": 0.00019382717365403854,
555
+ "loss": 0.9596,
556
+ "step": 77
557
+ },
558
+ {
559
+ "epoch": 0.15,
560
+ "grad_norm": 0.8419337272644043,
561
+ "learning_rate": 0.00019361103964477883,
562
+ "loss": 0.9653,
563
+ "step": 78
564
+ },
565
+ {
566
+ "epoch": 0.15,
567
+ "grad_norm": 0.7405250072479248,
568
+ "learning_rate": 0.00019339131137538696,
569
+ "loss": 0.7997,
570
+ "step": 79
571
+ },
572
+ {
573
+ "epoch": 0.15,
574
+ "grad_norm": 0.759087085723877,
575
+ "learning_rate": 0.00019316799728248075,
576
+ "loss": 1.0548,
577
+ "step": 80
578
+ },
579
+ {
580
+ "epoch": 0.15,
581
+ "grad_norm": 0.6998041868209839,
582
+ "learning_rate": 0.00019294110594035804,
583
+ "loss": 0.8764,
584
+ "step": 81
585
+ },
586
+ {
587
+ "epoch": 0.16,
588
+ "grad_norm": 0.7717428803443909,
589
+ "learning_rate": 0.0001927106460606677,
590
+ "loss": 0.9785,
591
+ "step": 82
592
+ },
593
+ {
594
+ "epoch": 0.16,
595
+ "grad_norm": 1.446915626525879,
596
+ "learning_rate": 0.0001924766264920751,
597
+ "loss": 0.8841,
598
+ "step": 83
599
+ },
600
+ {
601
+ "epoch": 0.16,
602
+ "grad_norm": 0.8436324596405029,
603
+ "learning_rate": 0.00019223905621992206,
604
+ "loss": 1.0682,
605
+ "step": 84
606
+ },
607
+ {
608
+ "epoch": 0.16,
609
+ "grad_norm": 0.767516553401947,
610
+ "learning_rate": 0.00019199794436588243,
611
+ "loss": 0.9833,
612
+ "step": 85
613
+ },
614
+ {
615
+ "epoch": 0.16,
616
+ "grad_norm": 0.8140639066696167,
617
+ "learning_rate": 0.0001917533001876113,
618
+ "loss": 0.9773,
619
+ "step": 86
620
+ },
621
+ {
622
+ "epoch": 0.17,
623
+ "grad_norm": 0.6604956984519958,
624
+ "learning_rate": 0.00019150513307838988,
625
+ "loss": 1.0004,
626
+ "step": 87
627
+ },
628
+ {
629
+ "epoch": 0.17,
630
+ "grad_norm": 0.9023972153663635,
631
+ "learning_rate": 0.00019125345256676467,
632
+ "loss": 0.9804,
633
+ "step": 88
634
+ },
635
+ {
636
+ "epoch": 0.17,
637
+ "grad_norm": 0.6216576099395752,
638
+ "learning_rate": 0.0001909982683161817,
639
+ "loss": 0.8837,
640
+ "step": 89
641
+ },
642
+ {
643
+ "epoch": 0.17,
644
+ "grad_norm": 0.7384454607963562,
645
+ "learning_rate": 0.00019073959012461545,
646
+ "loss": 0.8443,
647
+ "step": 90
648
+ },
649
+ {
650
+ "epoch": 0.17,
651
+ "grad_norm": 0.8240922093391418,
652
+ "learning_rate": 0.00019047742792419268,
653
+ "loss": 0.9069,
654
+ "step": 91
655
+ },
656
+ {
657
+ "epoch": 0.17,
658
+ "grad_norm": 0.629971981048584,
659
+ "learning_rate": 0.00019021179178081105,
660
+ "loss": 0.9381,
661
+ "step": 92
662
+ },
663
+ {
664
+ "epoch": 0.18,
665
+ "grad_norm": 1.0065981149673462,
666
+ "learning_rate": 0.00018994269189375268,
667
+ "loss": 0.989,
668
+ "step": 93
669
+ },
670
+ {
671
+ "epoch": 0.18,
672
+ "grad_norm": 0.7697680592536926,
673
+ "learning_rate": 0.00018967013859529246,
674
+ "loss": 0.9805,
675
+ "step": 94
676
+ },
677
+ {
678
+ "epoch": 0.18,
679
+ "grad_norm": 0.6588563919067383,
680
+ "learning_rate": 0.00018939414235030134,
681
+ "loss": 0.9206,
682
+ "step": 95
683
+ },
684
+ {
685
+ "epoch": 0.18,
686
+ "grad_norm": 0.7448359727859497,
687
+ "learning_rate": 0.00018911471375584468,
688
+ "loss": 0.9082,
689
+ "step": 96
690
+ },
691
+ {
692
+ "epoch": 0.18,
693
+ "grad_norm": 0.787029504776001,
694
+ "learning_rate": 0.0001888318635407752,
695
+ "loss": 0.9051,
696
+ "step": 97
697
+ },
698
+ {
699
+ "epoch": 0.19,
700
+ "grad_norm": 0.6427372694015503,
701
+ "learning_rate": 0.000188545602565321,
702
+ "loss": 0.8041,
703
+ "step": 98
704
+ },
705
+ {
706
+ "epoch": 0.19,
707
+ "grad_norm": 0.633730411529541,
708
+ "learning_rate": 0.00018825594182066886,
709
+ "loss": 0.8297,
710
+ "step": 99
711
+ },
712
+ {
713
+ "epoch": 0.19,
714
+ "grad_norm": 1.0125346183776855,
715
+ "learning_rate": 0.0001879628924285419,
716
+ "loss": 0.9292,
717
+ "step": 100
718
+ },
719
+ {
720
+ "epoch": 0.19,
721
+ "grad_norm": 0.851612389087677,
722
+ "learning_rate": 0.00018766646564077265,
723
+ "loss": 0.8787,
724
+ "step": 101
725
+ },
726
+ {
727
+ "epoch": 0.19,
728
+ "grad_norm": 0.9300311803817749,
729
+ "learning_rate": 0.00018736667283887116,
730
+ "loss": 0.933,
731
+ "step": 102
732
+ },
733
+ {
734
+ "epoch": 0.2,
735
+ "grad_norm": 0.9016942977905273,
736
+ "learning_rate": 0.00018706352553358793,
737
+ "loss": 0.9506,
738
+ "step": 103
739
+ },
740
+ {
741
+ "epoch": 0.2,
742
+ "grad_norm": 0.8613467216491699,
743
+ "learning_rate": 0.00018675703536447178,
744
+ "loss": 0.8779,
745
+ "step": 104
746
+ },
747
+ {
748
+ "epoch": 0.2,
749
+ "grad_norm": 0.7087985277175903,
750
+ "learning_rate": 0.00018644721409942323,
751
+ "loss": 0.9122,
752
+ "step": 105
753
+ },
754
+ {
755
+ "epoch": 0.2,
756
+ "grad_norm": 1.0127161741256714,
757
+ "learning_rate": 0.00018613407363424238,
758
+ "loss": 0.9366,
759
+ "step": 106
760
+ },
761
+ {
762
+ "epoch": 0.2,
763
+ "grad_norm": 0.786028265953064,
764
+ "learning_rate": 0.00018581762599217242,
765
+ "loss": 0.8335,
766
+ "step": 107
767
+ },
768
+ {
769
+ "epoch": 0.2,
770
+ "grad_norm": 1.2019354104995728,
771
+ "learning_rate": 0.00018549788332343777,
772
+ "loss": 0.9577,
773
+ "step": 108
774
+ },
775
+ {
776
+ "epoch": 0.21,
777
+ "grad_norm": 0.8300893902778625,
778
+ "learning_rate": 0.0001851748579047777,
779
+ "loss": 0.9094,
780
+ "step": 109
781
+ },
782
+ {
783
+ "epoch": 0.21,
784
+ "grad_norm": 0.7978622913360596,
785
+ "learning_rate": 0.00018484856213897498,
786
+ "loss": 0.8753,
787
+ "step": 110
788
+ },
789
+ {
790
+ "epoch": 0.21,
791
+ "grad_norm": 0.77290278673172,
792
+ "learning_rate": 0.0001845190085543795,
793
+ "loss": 0.8746,
794
+ "step": 111
795
+ },
796
+ {
797
+ "epoch": 0.21,
798
+ "grad_norm": 0.6986845135688782,
799
+ "learning_rate": 0.00018418620980442736,
800
+ "loss": 0.8153,
801
+ "step": 112
802
+ },
803
+ {
804
+ "epoch": 0.21,
805
+ "grad_norm": 0.8883097767829895,
806
+ "learning_rate": 0.00018385017866715507,
807
+ "loss": 0.9012,
808
+ "step": 113
809
+ },
810
+ {
811
+ "epoch": 0.22,
812
+ "grad_norm": 1.0097709894180298,
813
+ "learning_rate": 0.00018351092804470885,
814
+ "loss": 0.9251,
815
+ "step": 114
816
+ },
817
+ {
818
+ "epoch": 0.22,
819
+ "grad_norm": 0.7271776795387268,
820
+ "learning_rate": 0.00018316847096284917,
821
+ "loss": 0.8207,
822
+ "step": 115
823
+ },
824
+ {
825
+ "epoch": 0.22,
826
+ "grad_norm": 0.6595386862754822,
827
+ "learning_rate": 0.00018282282057045088,
828
+ "loss": 0.8365,
829
+ "step": 116
830
+ },
831
+ {
832
+ "epoch": 0.22,
833
+ "grad_norm": 0.9244414567947388,
834
+ "learning_rate": 0.00018247399013899805,
835
+ "loss": 0.901,
836
+ "step": 117
837
+ },
838
+ {
839
+ "epoch": 0.22,
840
+ "grad_norm": 0.8177185654640198,
841
+ "learning_rate": 0.00018212199306207456,
842
+ "loss": 0.9121,
843
+ "step": 118
844
+ },
845
+ {
846
+ "epoch": 0.23,
847
+ "grad_norm": 0.6657998561859131,
848
+ "learning_rate": 0.00018176684285484983,
849
+ "loss": 0.8596,
850
+ "step": 119
851
+ },
852
+ {
853
+ "epoch": 0.23,
854
+ "grad_norm": 0.7159414291381836,
855
+ "learning_rate": 0.0001814085531535599,
856
+ "loss": 0.7837,
857
+ "step": 120
858
+ },
859
+ {
860
+ "epoch": 0.23,
861
+ "grad_norm": 0.7221715450286865,
862
+ "learning_rate": 0.00018104713771498383,
863
+ "loss": 0.8141,
864
+ "step": 121
865
+ },
866
+ {
867
+ "epoch": 0.23,
868
+ "grad_norm": 0.7190296053886414,
869
+ "learning_rate": 0.00018068261041591548,
870
+ "loss": 0.7683,
871
+ "step": 122
872
+ },
873
+ {
874
+ "epoch": 0.23,
875
+ "grad_norm": 0.9818452596664429,
876
+ "learning_rate": 0.00018031498525263072,
877
+ "loss": 1.0075,
878
+ "step": 123
879
+ },
880
+ {
881
+ "epoch": 0.24,
882
+ "grad_norm": 0.7468921542167664,
883
+ "learning_rate": 0.00017994427634035015,
884
+ "loss": 0.8308,
885
+ "step": 124
886
+ },
887
+ {
888
+ "epoch": 0.24,
889
+ "grad_norm": 0.8131096959114075,
890
+ "learning_rate": 0.00017957049791269685,
891
+ "loss": 0.9214,
892
+ "step": 125
893
+ },
894
+ {
895
+ "epoch": 0.24,
896
+ "grad_norm": 0.7757013440132141,
897
+ "learning_rate": 0.00017919366432115024,
898
+ "loss": 0.8861,
899
+ "step": 126
900
+ },
901
+ {
902
+ "epoch": 0.24,
903
+ "grad_norm": 0.6586887240409851,
904
+ "learning_rate": 0.00017881379003449472,
905
+ "loss": 0.8654,
906
+ "step": 127
907
+ },
908
+ {
909
+ "epoch": 0.24,
910
+ "grad_norm": 0.6687777042388916,
911
+ "learning_rate": 0.00017843088963826435,
912
+ "loss": 0.7934,
913
+ "step": 128
914
+ },
915
+ {
916
+ "epoch": 0.24,
917
+ "grad_norm": 0.7160292267799377,
918
+ "learning_rate": 0.00017804497783418266,
919
+ "loss": 0.8842,
920
+ "step": 129
921
+ },
922
+ {
923
+ "epoch": 0.25,
924
+ "grad_norm": 0.7248404622077942,
925
+ "learning_rate": 0.00017765606943959833,
926
+ "loss": 0.8668,
927
+ "step": 130
928
+ },
929
+ {
930
+ "epoch": 0.25,
931
+ "grad_norm": 0.953946053981781,
932
+ "learning_rate": 0.00017726417938691619,
933
+ "loss": 0.8918,
934
+ "step": 131
935
+ },
936
+ {
937
+ "epoch": 0.25,
938
+ "grad_norm": 0.8473754525184631,
939
+ "learning_rate": 0.0001768693227230238,
940
+ "loss": 0.8144,
941
+ "step": 132
942
+ },
943
+ {
944
+ "epoch": 0.25,
945
+ "eval_loss": 0.8364565372467041,
946
+ "eval_runtime": 14.7047,
947
+ "eval_samples_per_second": 32.099,
948
+ "eval_steps_per_second": 32.099,
949
+ "step": 132
950
+ },
951
+ {
952
+ "epoch": 0.25,
953
+ "grad_norm": 0.7996974587440491,
954
+ "learning_rate": 0.00017647151460871386,
955
+ "loss": 0.9233,
956
+ "step": 133
957
+ },
958
+ {
959
+ "epoch": 0.25,
960
+ "grad_norm": 0.728558361530304,
961
+ "learning_rate": 0.00017607077031810202,
962
+ "loss": 0.766,
963
+ "step": 134
964
+ },
965
+ {
966
+ "epoch": 0.26,
967
+ "grad_norm": 0.8249170184135437,
968
+ "learning_rate": 0.00017566710523804043,
969
+ "loss": 0.8565,
970
+ "step": 135
971
+ },
972
+ {
973
+ "epoch": 0.26,
974
+ "grad_norm": 0.6535347104072571,
975
+ "learning_rate": 0.00017526053486752695,
976
+ "loss": 0.7229,
977
+ "step": 136
978
+ },
979
+ {
980
+ "epoch": 0.26,
981
+ "grad_norm": 0.7984380722045898,
982
+ "learning_rate": 0.00017485107481711012,
983
+ "loss": 0.7271,
984
+ "step": 137
985
+ },
986
+ {
987
+ "epoch": 0.26,
988
+ "grad_norm": 0.7088860869407654,
989
+ "learning_rate": 0.00017443874080828962,
990
+ "loss": 0.7622,
991
+ "step": 138
992
+ },
993
+ {
994
+ "epoch": 0.26,
995
+ "grad_norm": 0.8559077978134155,
996
+ "learning_rate": 0.0001740235486729128,
997
+ "loss": 0.9194,
998
+ "step": 139
999
+ },
1000
+ {
1001
+ "epoch": 0.27,
1002
+ "grad_norm": 0.9430374503135681,
1003
+ "learning_rate": 0.00017360551435256674,
1004
+ "loss": 0.8032,
1005
+ "step": 140
1006
+ },
1007
+ {
1008
+ "epoch": 0.27,
1009
+ "grad_norm": 0.8738976716995239,
1010
+ "learning_rate": 0.00017318465389796613,
1011
+ "loss": 0.8433,
1012
+ "step": 141
1013
+ },
1014
+ {
1015
+ "epoch": 0.27,
1016
+ "grad_norm": 1.5377166271209717,
1017
+ "learning_rate": 0.00017276098346833712,
1018
+ "loss": 0.7793,
1019
+ "step": 142
1020
+ },
1021
+ {
1022
+ "epoch": 0.27,
1023
+ "grad_norm": 0.8574996590614319,
1024
+ "learning_rate": 0.00017233451933079664,
1025
+ "loss": 0.8422,
1026
+ "step": 143
1027
+ },
1028
+ {
1029
+ "epoch": 0.27,
1030
+ "grad_norm": 0.7897395491600037,
1031
+ "learning_rate": 0.00017190527785972807,
1032
+ "loss": 0.9424,
1033
+ "step": 144
1034
+ },
1035
+ {
1036
+ "epoch": 0.28,
1037
+ "grad_norm": 0.8188284039497375,
1038
+ "learning_rate": 0.0001714732755361523,
1039
+ "loss": 0.6568,
1040
+ "step": 145
1041
+ },
1042
+ {
1043
+ "epoch": 0.28,
1044
+ "grad_norm": 0.6334033012390137,
1045
+ "learning_rate": 0.00017103852894709517,
1046
+ "loss": 0.7111,
1047
+ "step": 146
1048
+ },
1049
+ {
1050
+ "epoch": 0.28,
1051
+ "grad_norm": 0.9169749021530151,
1052
+ "learning_rate": 0.00017060105478495044,
1053
+ "loss": 0.8087,
1054
+ "step": 147
1055
+ },
1056
+ {
1057
+ "epoch": 0.28,
1058
+ "grad_norm": 0.8317080140113831,
1059
+ "learning_rate": 0.00017016086984683888,
1060
+ "loss": 0.7404,
1061
+ "step": 148
1062
+ },
1063
+ {
1064
+ "epoch": 0.28,
1065
+ "grad_norm": 0.6163865327835083,
1066
+ "learning_rate": 0.00016971799103396334,
1067
+ "loss": 0.6276,
1068
+ "step": 149
1069
+ },
1070
+ {
1071
+ "epoch": 0.28,
1072
+ "grad_norm": 0.8176054358482361,
1073
+ "learning_rate": 0.00016927243535095997,
1074
+ "loss": 0.9162,
1075
+ "step": 150
1076
+ },
1077
+ {
1078
+ "epoch": 0.29,
1079
+ "grad_norm": 0.6680843830108643,
1080
+ "learning_rate": 0.00016882421990524498,
1081
+ "loss": 0.7782,
1082
+ "step": 151
1083
+ },
1084
+ {
1085
+ "epoch": 0.29,
1086
+ "grad_norm": 0.8415772914886475,
1087
+ "learning_rate": 0.00016837336190635824,
1088
+ "loss": 0.759,
1089
+ "step": 152
1090
+ },
1091
+ {
1092
+ "epoch": 0.29,
1093
+ "grad_norm": 0.8282654285430908,
1094
+ "learning_rate": 0.00016791987866530202,
1095
+ "loss": 0.7986,
1096
+ "step": 153
1097
+ },
1098
+ {
1099
+ "epoch": 0.29,
1100
+ "grad_norm": 0.7016776204109192,
1101
+ "learning_rate": 0.00016746378759387675,
1102
+ "loss": 0.6493,
1103
+ "step": 154
1104
+ },
1105
+ {
1106
+ "epoch": 0.29,
1107
+ "grad_norm": 0.5957922339439392,
1108
+ "learning_rate": 0.00016700510620401224,
1109
+ "loss": 0.7463,
1110
+ "step": 155
1111
+ },
1112
+ {
1113
+ "epoch": 0.3,
1114
+ "grad_norm": 0.7017227411270142,
1115
+ "learning_rate": 0.00016654385210709531,
1116
+ "loss": 0.7806,
1117
+ "step": 156
1118
+ },
1119
+ {
1120
+ "epoch": 0.3,
1121
+ "grad_norm": 0.77492356300354,
1122
+ "learning_rate": 0.00016608004301329365,
1123
+ "loss": 0.7959,
1124
+ "step": 157
1125
+ },
1126
+ {
1127
+ "epoch": 0.3,
1128
+ "grad_norm": 0.6599026918411255,
1129
+ "learning_rate": 0.00016561369673087588,
1130
+ "loss": 0.764,
1131
+ "step": 158
1132
+ },
1133
+ {
1134
+ "epoch": 0.3,
1135
+ "grad_norm": 0.7865548133850098,
1136
+ "learning_rate": 0.00016514483116552762,
1137
+ "loss": 0.7801,
1138
+ "step": 159
1139
+ },
1140
+ {
1141
+ "epoch": 0.3,
1142
+ "grad_norm": 0.6737894415855408,
1143
+ "learning_rate": 0.00016467346431966413,
1144
+ "loss": 0.7365,
1145
+ "step": 160
1146
+ },
1147
+ {
1148
+ "epoch": 0.31,
1149
+ "grad_norm": 0.7904422879219055,
1150
+ "learning_rate": 0.0001641996142917391,
1151
+ "loss": 0.8118,
1152
+ "step": 161
1153
+ },
1154
+ {
1155
+ "epoch": 0.31,
1156
+ "grad_norm": 0.760714590549469,
1157
+ "learning_rate": 0.0001637232992755496,
1158
+ "loss": 0.6932,
1159
+ "step": 162
1160
+ },
1161
+ {
1162
+ "epoch": 0.31,
1163
+ "grad_norm": 0.6482570767402649,
1164
+ "learning_rate": 0.00016324453755953773,
1165
+ "loss": 0.5669,
1166
+ "step": 163
1167
+ },
1168
+ {
1169
+ "epoch": 0.31,
1170
+ "grad_norm": 0.8352275490760803,
1171
+ "learning_rate": 0.00016276334752608822,
1172
+ "loss": 0.7958,
1173
+ "step": 164
1174
+ },
1175
+ {
1176
+ "epoch": 0.31,
1177
+ "grad_norm": 0.6860224604606628,
1178
+ "learning_rate": 0.00016227974765082274,
1179
+ "loss": 0.8383,
1180
+ "step": 165
1181
+ },
1182
+ {
1183
+ "epoch": 0.31,
1184
+ "grad_norm": 0.8616229891777039,
1185
+ "learning_rate": 0.00016179375650189048,
1186
+ "loss": 0.7639,
1187
+ "step": 166
1188
+ },
1189
+ {
1190
+ "epoch": 0.32,
1191
+ "grad_norm": 0.6623161435127258,
1192
+ "learning_rate": 0.0001613053927392553,
1193
+ "loss": 0.661,
1194
+ "step": 167
1195
+ },
1196
+ {
1197
+ "epoch": 0.32,
1198
+ "grad_norm": 0.8096372485160828,
1199
+ "learning_rate": 0.0001608146751139791,
1200
+ "loss": 0.7507,
1201
+ "step": 168
1202
+ },
1203
+ {
1204
+ "epoch": 0.32,
1205
+ "grad_norm": 0.7616962790489197,
1206
+ "learning_rate": 0.00016032162246750197,
1207
+ "loss": 0.7798,
1208
+ "step": 169
1209
+ },
1210
+ {
1211
+ "epoch": 0.32,
1212
+ "grad_norm": 0.7718486189842224,
1213
+ "learning_rate": 0.00015982625373091875,
1214
+ "loss": 0.8187,
1215
+ "step": 170
1216
+ },
1217
+ {
1218
+ "epoch": 0.32,
1219
+ "grad_norm": 0.6922149658203125,
1220
+ "learning_rate": 0.0001593285879242522,
1221
+ "loss": 0.7562,
1222
+ "step": 171
1223
+ },
1224
+ {
1225
+ "epoch": 0.33,
1226
+ "grad_norm": 0.7328513860702515,
1227
+ "learning_rate": 0.0001588286441557226,
1228
+ "loss": 0.7315,
1229
+ "step": 172
1230
+ },
1231
+ {
1232
+ "epoch": 0.33,
1233
+ "grad_norm": 0.6606435775756836,
1234
+ "learning_rate": 0.00015832644162101417,
1235
+ "loss": 0.5997,
1236
+ "step": 173
1237
+ },
1238
+ {
1239
+ "epoch": 0.33,
1240
+ "grad_norm": 1.2101995944976807,
1241
+ "learning_rate": 0.000157821999602538,
1242
+ "loss": 0.8754,
1243
+ "step": 174
1244
+ },
1245
+ {
1246
+ "epoch": 0.33,
1247
+ "grad_norm": 0.744756817817688,
1248
+ "learning_rate": 0.00015731533746869163,
1249
+ "loss": 0.8369,
1250
+ "step": 175
1251
+ },
1252
+ {
1253
+ "epoch": 0.33,
1254
+ "grad_norm": 0.6933184266090393,
1255
+ "learning_rate": 0.00015680647467311557,
1256
+ "loss": 0.7291,
1257
+ "step": 176
1258
+ },
1259
+ {
1260
+ "epoch": 0.34,
1261
+ "grad_norm": 0.8668520450592041,
1262
+ "learning_rate": 0.00015629543075394625,
1263
+ "loss": 0.6565,
1264
+ "step": 177
1265
+ },
1266
+ {
1267
+ "epoch": 0.34,
1268
+ "grad_norm": 1.06217360496521,
1269
+ "learning_rate": 0.0001557822253330657,
1270
+ "loss": 0.7949,
1271
+ "step": 178
1272
+ },
1273
+ {
1274
+ "epoch": 0.34,
1275
+ "grad_norm": 0.7296217679977417,
1276
+ "learning_rate": 0.00015526687811534838,
1277
+ "loss": 0.7218,
1278
+ "step": 179
1279
+ },
1280
+ {
1281
+ "epoch": 0.34,
1282
+ "grad_norm": 0.679236114025116,
1283
+ "learning_rate": 0.00015474940888790455,
1284
+ "loss": 0.7323,
1285
+ "step": 180
1286
+ },
1287
+ {
1288
+ "epoch": 0.34,
1289
+ "grad_norm": 0.7786648273468018,
1290
+ "learning_rate": 0.0001542298375193204,
1291
+ "loss": 0.7032,
1292
+ "step": 181
1293
+ },
1294
+ {
1295
+ "epoch": 0.35,
1296
+ "grad_norm": 1.0958847999572754,
1297
+ "learning_rate": 0.00015370818395889536,
1298
+ "loss": 0.856,
1299
+ "step": 182
1300
+ },
1301
+ {
1302
+ "epoch": 0.35,
1303
+ "grad_norm": 0.6277094483375549,
1304
+ "learning_rate": 0.00015318446823587596,
1305
+ "loss": 0.7215,
1306
+ "step": 183
1307
+ },
1308
+ {
1309
+ "epoch": 0.35,
1310
+ "grad_norm": 0.7913120985031128,
1311
+ "learning_rate": 0.000152658710458687,
1312
+ "loss": 0.7673,
1313
+ "step": 184
1314
+ },
1315
+ {
1316
+ "epoch": 0.35,
1317
+ "grad_norm": 1.0577659606933594,
1318
+ "learning_rate": 0.0001521309308141592,
1319
+ "loss": 0.7168,
1320
+ "step": 185
1321
+ },
1322
+ {
1323
+ "epoch": 0.35,
1324
+ "grad_norm": 0.7179178595542908,
1325
+ "learning_rate": 0.00015160114956675434,
1326
+ "loss": 0.7872,
1327
+ "step": 186
1328
+ },
1329
+ {
1330
+ "epoch": 0.35,
1331
+ "grad_norm": 1.1850095987319946,
1332
+ "learning_rate": 0.00015106938705778712,
1333
+ "loss": 0.7967,
1334
+ "step": 187
1335
+ },
1336
+ {
1337
+ "epoch": 0.36,
1338
+ "grad_norm": 0.7346929311752319,
1339
+ "learning_rate": 0.00015053566370464415,
1340
+ "loss": 0.7788,
1341
+ "step": 188
1342
+ },
1343
+ {
1344
+ "epoch": 0.36,
1345
+ "grad_norm": 0.8880440592765808,
1346
+ "learning_rate": 0.00015000000000000001,
1347
+ "loss": 0.7466,
1348
+ "step": 189
1349
+ },
1350
+ {
1351
+ "epoch": 0.36,
1352
+ "grad_norm": 0.6921265125274658,
1353
+ "learning_rate": 0.00014946241651103034,
1354
+ "loss": 0.8424,
1355
+ "step": 190
1356
+ },
1357
+ {
1358
+ "epoch": 0.36,
1359
+ "grad_norm": 0.8322516679763794,
1360
+ "learning_rate": 0.00014892293387862223,
1361
+ "loss": 0.7648,
1362
+ "step": 191
1363
+ },
1364
+ {
1365
+ "epoch": 0.36,
1366
+ "grad_norm": 0.8475639820098877,
1367
+ "learning_rate": 0.00014838157281658177,
1368
+ "loss": 0.674,
1369
+ "step": 192
1370
+ },
1371
+ {
1372
+ "epoch": 0.37,
1373
+ "grad_norm": 0.8423508405685425,
1374
+ "learning_rate": 0.00014783835411083854,
1375
+ "loss": 0.8193,
1376
+ "step": 193
1377
+ },
1378
+ {
1379
+ "epoch": 0.37,
1380
+ "grad_norm": 0.9078856110572815,
1381
+ "learning_rate": 0.0001472932986186477,
1382
+ "loss": 0.8027,
1383
+ "step": 194
1384
+ },
1385
+ {
1386
+ "epoch": 0.37,
1387
+ "grad_norm": 0.6911562085151672,
1388
+ "learning_rate": 0.00014674642726778906,
1389
+ "loss": 0.7821,
1390
+ "step": 195
1391
+ },
1392
+ {
1393
+ "epoch": 0.37,
1394
+ "grad_norm": 0.7079100608825684,
1395
+ "learning_rate": 0.0001461977610557635,
1396
+ "loss": 0.6792,
1397
+ "step": 196
1398
+ },
1399
+ {
1400
+ "epoch": 0.37,
1401
+ "grad_norm": 0.7966701984405518,
1402
+ "learning_rate": 0.000145647321048987,
1403
+ "loss": 0.8002,
1404
+ "step": 197
1405
+ },
1406
+ {
1407
+ "epoch": 0.38,
1408
+ "grad_norm": 0.9254695177078247,
1409
+ "learning_rate": 0.00014509512838198148,
1410
+ "loss": 0.8354,
1411
+ "step": 198
1412
+ },
1413
+ {
1414
+ "epoch": 0.38,
1415
+ "grad_norm": 0.6664696335792542,
1416
+ "learning_rate": 0.00014454120425656342,
1417
+ "loss": 0.7546,
1418
+ "step": 199
1419
+ },
1420
+ {
1421
+ "epoch": 0.38,
1422
+ "grad_norm": 1.0904691219329834,
1423
+ "learning_rate": 0.00014398556994102996,
1424
+ "loss": 0.7356,
1425
+ "step": 200
1426
+ },
1427
+ {
1428
+ "epoch": 0.38,
1429
+ "grad_norm": 0.8222538232803345,
1430
+ "learning_rate": 0.00014342824676934202,
1431
+ "loss": 0.6181,
1432
+ "step": 201
1433
+ },
1434
+ {
1435
+ "epoch": 0.38,
1436
+ "grad_norm": 0.848257839679718,
1437
+ "learning_rate": 0.00014286925614030542,
1438
+ "loss": 0.8353,
1439
+ "step": 202
1440
+ },
1441
+ {
1442
+ "epoch": 0.39,
1443
+ "grad_norm": 0.6590856313705444,
1444
+ "learning_rate": 0.00014230861951674913,
1445
+ "loss": 0.6996,
1446
+ "step": 203
1447
+ },
1448
+ {
1449
+ "epoch": 0.39,
1450
+ "grad_norm": 1.1949280500411987,
1451
+ "learning_rate": 0.00014174635842470118,
1452
+ "loss": 0.7901,
1453
+ "step": 204
1454
+ },
1455
+ {
1456
+ "epoch": 0.39,
1457
+ "grad_norm": 0.7490679621696472,
1458
+ "learning_rate": 0.00014118249445256223,
1459
+ "loss": 0.6008,
1460
+ "step": 205
1461
+ },
1462
+ {
1463
+ "epoch": 0.39,
1464
+ "grad_norm": 1.1541824340820312,
1465
+ "learning_rate": 0.00014061704925027652,
1466
+ "loss": 0.6705,
1467
+ "step": 206
1468
+ },
1469
+ {
1470
+ "epoch": 0.39,
1471
+ "grad_norm": 0.9292888045310974,
1472
+ "learning_rate": 0.00014005004452850083,
1473
+ "loss": 0.8084,
1474
+ "step": 207
1475
+ },
1476
+ {
1477
+ "epoch": 0.39,
1478
+ "grad_norm": 0.9962819218635559,
1479
+ "learning_rate": 0.0001394815020577707,
1480
+ "loss": 0.7381,
1481
+ "step": 208
1482
+ },
1483
+ {
1484
+ "epoch": 0.4,
1485
+ "grad_norm": 0.6327970027923584,
1486
+ "learning_rate": 0.00013891144366766456,
1487
+ "loss": 0.7034,
1488
+ "step": 209
1489
+ },
1490
+ {
1491
+ "epoch": 0.4,
1492
+ "grad_norm": 1.0281709432601929,
1493
+ "learning_rate": 0.00013833989124596572,
1494
+ "loss": 0.759,
1495
+ "step": 210
1496
+ },
1497
+ {
1498
+ "epoch": 0.4,
1499
+ "grad_norm": 0.6905353665351868,
1500
+ "learning_rate": 0.00013776686673782175,
1501
+ "loss": 0.7074,
1502
+ "step": 211
1503
+ },
1504
+ {
1505
+ "epoch": 0.4,
1506
+ "grad_norm": 0.7964909672737122,
1507
+ "learning_rate": 0.00013719239214490204,
1508
+ "loss": 0.8094,
1509
+ "step": 212
1510
+ },
1511
+ {
1512
+ "epoch": 0.4,
1513
+ "grad_norm": 0.8168321251869202,
1514
+ "learning_rate": 0.00013661648952455291,
1515
+ "loss": 0.7632,
1516
+ "step": 213
1517
+ },
1518
+ {
1519
+ "epoch": 0.41,
1520
+ "grad_norm": 0.6741600036621094,
1521
+ "learning_rate": 0.00013603918098895092,
1522
+ "loss": 0.7473,
1523
+ "step": 214
1524
+ },
1525
+ {
1526
+ "epoch": 0.41,
1527
+ "grad_norm": 1.1006839275360107,
1528
+ "learning_rate": 0.00013546048870425356,
1529
+ "loss": 0.6841,
1530
+ "step": 215
1531
+ },
1532
+ {
1533
+ "epoch": 0.41,
1534
+ "grad_norm": 1.0017565488815308,
1535
+ "learning_rate": 0.00013488043488974844,
1536
+ "loss": 0.8087,
1537
+ "step": 216
1538
+ },
1539
+ {
1540
+ "epoch": 0.41,
1541
+ "grad_norm": 0.7672898769378662,
1542
+ "learning_rate": 0.0001342990418169999,
1543
+ "loss": 0.787,
1544
+ "step": 217
1545
+ },
1546
+ {
1547
+ "epoch": 0.41,
1548
+ "grad_norm": 0.7265097498893738,
1549
+ "learning_rate": 0.00013371633180899416,
1550
+ "loss": 0.7716,
1551
+ "step": 218
1552
+ },
1553
+ {
1554
+ "epoch": 0.42,
1555
+ "grad_norm": 0.8181336522102356,
1556
+ "learning_rate": 0.000133132327239282,
1557
+ "loss": 0.7527,
1558
+ "step": 219
1559
+ },
1560
+ {
1561
+ "epoch": 0.42,
1562
+ "grad_norm": 0.7809070944786072,
1563
+ "learning_rate": 0.0001325470505311198,
1564
+ "loss": 0.6298,
1565
+ "step": 220
1566
+ },
1567
+ {
1568
+ "epoch": 0.42,
1569
+ "grad_norm": 0.8382383584976196,
1570
+ "learning_rate": 0.00013196052415660856,
1571
+ "loss": 0.723,
1572
+ "step": 221
1573
+ },
1574
+ {
1575
+ "epoch": 0.42,
1576
+ "grad_norm": 0.6496840119361877,
1577
+ "learning_rate": 0.0001313727706358311,
1578
+ "loss": 0.7228,
1579
+ "step": 222
1580
+ },
1581
+ {
1582
+ "epoch": 0.42,
1583
+ "grad_norm": 0.657471776008606,
1584
+ "learning_rate": 0.00013078381253598732,
1585
+ "loss": 0.6453,
1586
+ "step": 223
1587
+ },
1588
+ {
1589
+ "epoch": 0.42,
1590
+ "grad_norm": 0.8327584862709045,
1591
+ "learning_rate": 0.0001301936724705278,
1592
+ "loss": 0.7631,
1593
+ "step": 224
1594
+ },
1595
+ {
1596
+ "epoch": 0.43,
1597
+ "grad_norm": 0.7766168713569641,
1598
+ "learning_rate": 0.0001296023730982855,
1599
+ "loss": 0.705,
1600
+ "step": 225
1601
+ },
1602
+ {
1603
+ "epoch": 0.43,
1604
+ "grad_norm": 0.8591150045394897,
1605
+ "learning_rate": 0.0001290099371226058,
1606
+ "loss": 0.6703,
1607
+ "step": 226
1608
+ },
1609
+ {
1610
+ "epoch": 0.43,
1611
+ "grad_norm": 0.7799991369247437,
1612
+ "learning_rate": 0.00012841638729047463,
1613
+ "loss": 0.7068,
1614
+ "step": 227
1615
+ },
1616
+ {
1617
+ "epoch": 0.43,
1618
+ "grad_norm": 0.7640935182571411,
1619
+ "learning_rate": 0.0001278217463916453,
1620
+ "loss": 0.7237,
1621
+ "step": 228
1622
+ },
1623
+ {
1624
+ "epoch": 0.43,
1625
+ "grad_norm": 0.6747919321060181,
1626
+ "learning_rate": 0.00012722603725776329,
1627
+ "loss": 0.6774,
1628
+ "step": 229
1629
+ },
1630
+ {
1631
+ "epoch": 0.44,
1632
+ "grad_norm": 0.9026936292648315,
1633
+ "learning_rate": 0.00012662928276148985,
1634
+ "loss": 0.7422,
1635
+ "step": 230
1636
+ },
1637
+ {
1638
+ "epoch": 0.44,
1639
+ "grad_norm": 0.816842794418335,
1640
+ "learning_rate": 0.0001260315058156235,
1641
+ "loss": 0.7338,
1642
+ "step": 231
1643
+ },
1644
+ {
1645
+ "epoch": 0.44,
1646
+ "grad_norm": 1.1454386711120605,
1647
+ "learning_rate": 0.00012543272937222044,
1648
+ "loss": 0.8056,
1649
+ "step": 232
1650
+ },
1651
+ {
1652
+ "epoch": 0.44,
1653
+ "grad_norm": 0.7503964900970459,
1654
+ "learning_rate": 0.00012483297642171333,
1655
+ "loss": 0.6559,
1656
+ "step": 233
1657
+ },
1658
+ {
1659
+ "epoch": 0.44,
1660
+ "grad_norm": 0.6204442381858826,
1661
+ "learning_rate": 0.00012423226999202838,
1662
+ "loss": 0.5569,
1663
+ "step": 234
1664
+ },
1665
+ {
1666
+ "epoch": 0.45,
1667
+ "grad_norm": 0.7909918427467346,
1668
+ "learning_rate": 0.00012363063314770135,
1669
+ "loss": 0.7276,
1670
+ "step": 235
1671
+ },
1672
+ {
1673
+ "epoch": 0.45,
1674
+ "grad_norm": 0.5152371525764465,
1675
+ "learning_rate": 0.00012302808898899197,
1676
+ "loss": 0.499,
1677
+ "step": 236
1678
+ },
1679
+ {
1680
+ "epoch": 0.45,
1681
+ "grad_norm": 0.9286866188049316,
1682
+ "learning_rate": 0.00012242466065099685,
1683
+ "loss": 0.7141,
1684
+ "step": 237
1685
+ },
1686
+ {
1687
+ "epoch": 0.45,
1688
+ "grad_norm": 0.66729736328125,
1689
+ "learning_rate": 0.00012182037130276125,
1690
+ "loss": 0.6765,
1691
+ "step": 238
1692
+ },
1693
+ {
1694
+ "epoch": 0.45,
1695
+ "grad_norm": 0.6844175457954407,
1696
+ "learning_rate": 0.00012121524414638959,
1697
+ "loss": 0.7087,
1698
+ "step": 239
1699
+ },
1700
+ {
1701
+ "epoch": 0.46,
1702
+ "grad_norm": 0.7366169095039368,
1703
+ "learning_rate": 0.0001206093024161544,
1704
+ "loss": 0.6645,
1705
+ "step": 240
1706
+ },
1707
+ {
1708
+ "epoch": 0.46,
1709
+ "grad_norm": 0.691806435585022,
1710
+ "learning_rate": 0.00012000256937760445,
1711
+ "loss": 0.6785,
1712
+ "step": 241
1713
+ },
1714
+ {
1715
+ "epoch": 0.46,
1716
+ "grad_norm": 0.6430802941322327,
1717
+ "learning_rate": 0.00011939506832667128,
1718
+ "loss": 0.6785,
1719
+ "step": 242
1720
+ },
1721
+ {
1722
+ "epoch": 0.46,
1723
+ "grad_norm": 0.7784428000450134,
1724
+ "learning_rate": 0.00011878682258877478,
1725
+ "loss": 0.7017,
1726
+ "step": 243
1727
+ },
1728
+ {
1729
+ "epoch": 0.46,
1730
+ "grad_norm": 0.6317816972732544,
1731
+ "learning_rate": 0.00011817785551792766,
1732
+ "loss": 0.6515,
1733
+ "step": 244
1734
+ },
1735
+ {
1736
+ "epoch": 0.46,
1737
+ "grad_norm": 0.8226771950721741,
1738
+ "learning_rate": 0.00011756819049583861,
1739
+ "loss": 0.6753,
1740
+ "step": 245
1741
+ },
1742
+ {
1743
+ "epoch": 0.47,
1744
+ "grad_norm": 0.7601309418678284,
1745
+ "learning_rate": 0.00011695785093101475,
1746
+ "loss": 0.6568,
1747
+ "step": 246
1748
+ },
1749
+ {
1750
+ "epoch": 0.47,
1751
+ "grad_norm": 0.6727234721183777,
1752
+ "learning_rate": 0.00011634686025786264,
1753
+ "loss": 0.5909,
1754
+ "step": 247
1755
+ },
1756
+ {
1757
+ "epoch": 0.47,
1758
+ "grad_norm": 0.7579910755157471,
1759
+ "learning_rate": 0.00011573524193578863,
1760
+ "loss": 0.7882,
1761
+ "step": 248
1762
+ },
1763
+ {
1764
+ "epoch": 0.47,
1765
+ "grad_norm": 0.680234432220459,
1766
+ "learning_rate": 0.00011512301944829809,
1767
+ "loss": 0.6865,
1768
+ "step": 249
1769
+ },
1770
+ {
1771
+ "epoch": 0.47,
1772
+ "grad_norm": 0.8539860248565674,
1773
+ "learning_rate": 0.00011451021630209371,
1774
+ "loss": 0.7303,
1775
+ "step": 250
1776
+ },
1777
+ {
1778
+ "epoch": 0.48,
1779
+ "grad_norm": 0.7847228646278381,
1780
+ "learning_rate": 0.00011389685602617301,
1781
+ "loss": 0.5941,
1782
+ "step": 251
1783
+ },
1784
+ {
1785
+ "epoch": 0.48,
1786
+ "grad_norm": 0.7232929468154907,
1787
+ "learning_rate": 0.00011328296217092485,
1788
+ "loss": 0.7054,
1789
+ "step": 252
1790
+ },
1791
+ {
1792
+ "epoch": 0.48,
1793
+ "grad_norm": 0.8790189623832703,
1794
+ "learning_rate": 0.00011266855830722523,
1795
+ "loss": 0.741,
1796
+ "step": 253
1797
+ },
1798
+ {
1799
+ "epoch": 0.48,
1800
+ "grad_norm": 0.6331459879875183,
1801
+ "learning_rate": 0.0001120536680255323,
1802
+ "loss": 0.6574,
1803
+ "step": 254
1804
+ },
1805
+ {
1806
+ "epoch": 0.48,
1807
+ "grad_norm": 0.7007669806480408,
1808
+ "learning_rate": 0.0001114383149349806,
1809
+ "loss": 0.7269,
1810
+ "step": 255
1811
+ },
1812
+ {
1813
+ "epoch": 0.49,
1814
+ "grad_norm": 0.648810088634491,
1815
+ "learning_rate": 0.00011082252266247442,
1816
+ "loss": 0.6414,
1817
+ "step": 256
1818
+ },
1819
+ {
1820
+ "epoch": 0.49,
1821
+ "grad_norm": 0.7953597903251648,
1822
+ "learning_rate": 0.00011020631485178083,
1823
+ "loss": 0.7068,
1824
+ "step": 257
1825
+ },
1826
+ {
1827
+ "epoch": 0.49,
1828
+ "grad_norm": 0.7083666920661926,
1829
+ "learning_rate": 0.00010958971516262177,
1830
+ "loss": 0.6948,
1831
+ "step": 258
1832
+ },
1833
+ {
1834
+ "epoch": 0.49,
1835
+ "grad_norm": 1.0010278224945068,
1836
+ "learning_rate": 0.00010897274726976561,
1837
+ "loss": 0.7326,
1838
+ "step": 259
1839
+ },
1840
+ {
1841
+ "epoch": 0.49,
1842
+ "grad_norm": 0.682121217250824,
1843
+ "learning_rate": 0.00010835543486211815,
1844
+ "loss": 0.5291,
1845
+ "step": 260
1846
+ },
1847
+ {
1848
+ "epoch": 0.5,
1849
+ "grad_norm": 0.7658283710479736,
1850
+ "learning_rate": 0.00010773780164181305,
1851
+ "loss": 0.6827,
1852
+ "step": 261
1853
+ },
1854
+ {
1855
+ "epoch": 0.5,
1856
+ "grad_norm": 0.7295763492584229,
1857
+ "learning_rate": 0.00010711987132330181,
1858
+ "loss": 0.6928,
1859
+ "step": 262
1860
+ },
1861
+ {
1862
+ "epoch": 0.5,
1863
+ "grad_norm": 0.7689676284790039,
1864
+ "learning_rate": 0.0001065016676324433,
1865
+ "loss": 0.669,
1866
+ "step": 263
1867
+ },
1868
+ {
1869
+ "epoch": 0.5,
1870
+ "grad_norm": 0.7863064408302307,
1871
+ "learning_rate": 0.00010588321430559252,
1872
+ "loss": 0.7159,
1873
+ "step": 264
1874
+ },
1875
+ {
1876
+ "epoch": 0.5,
1877
+ "eval_loss": 0.7109076380729675,
1878
+ "eval_runtime": 14.6602,
1879
+ "eval_samples_per_second": 32.196,
1880
+ "eval_steps_per_second": 32.196,
1881
+ "step": 264
1882
+ },
1883
+ {
1884
+ "epoch": 0.5,
1885
+ "grad_norm": 0.6780825257301331,
1886
+ "learning_rate": 0.00010526453508868961,
1887
+ "loss": 0.7589,
1888
+ "step": 265
1889
+ },
1890
+ {
1891
+ "epoch": 0.5,
1892
+ "grad_norm": 0.7262443900108337,
1893
+ "learning_rate": 0.00010464565373634782,
1894
+ "loss": 0.6978,
1895
+ "step": 266
1896
+ },
1897
+ {
1898
+ "epoch": 0.51,
1899
+ "grad_norm": 1.149004578590393,
1900
+ "learning_rate": 0.00010402659401094152,
1901
+ "loss": 0.7782,
1902
+ "step": 267
1903
+ },
1904
+ {
1905
+ "epoch": 0.51,
1906
+ "grad_norm": 0.7568680047988892,
1907
+ "learning_rate": 0.00010340737968169389,
1908
+ "loss": 0.6883,
1909
+ "step": 268
1910
+ },
1911
+ {
1912
+ "epoch": 0.51,
1913
+ "grad_norm": 0.7424808144569397,
1914
+ "learning_rate": 0.00010278803452376416,
1915
+ "loss": 0.6757,
1916
+ "step": 269
1917
+ },
1918
+ {
1919
+ "epoch": 0.51,
1920
+ "grad_norm": 1.1873571872711182,
1921
+ "learning_rate": 0.00010216858231733488,
1922
+ "loss": 0.6569,
1923
+ "step": 270
1924
+ },
1925
+ {
1926
+ "epoch": 0.51,
1927
+ "grad_norm": 0.7609845399856567,
1928
+ "learning_rate": 0.00010154904684669877,
1929
+ "loss": 0.7123,
1930
+ "step": 271
1931
+ },
1932
+ {
1933
+ "epoch": 0.52,
1934
+ "grad_norm": 0.9140599966049194,
1935
+ "learning_rate": 0.00010092945189934558,
1936
+ "loss": 0.6424,
1937
+ "step": 272
1938
+ },
1939
+ {
1940
+ "epoch": 0.52,
1941
+ "grad_norm": 0.7356762290000916,
1942
+ "learning_rate": 0.0001003098212650486,
1943
+ "loss": 0.7502,
1944
+ "step": 273
1945
+ },
1946
+ {
1947
+ "epoch": 0.52,
1948
+ "grad_norm": 0.7270318269729614,
1949
+ "learning_rate": 9.969017873495143e-05,
1950
+ "loss": 0.7053,
1951
+ "step": 274
1952
+ },
1953
+ {
1954
+ "epoch": 0.52,
1955
+ "grad_norm": 0.7932801842689514,
1956
+ "learning_rate": 9.907054810065446e-05,
1957
+ "loss": 0.6494,
1958
+ "step": 275
1959
+ },
1960
+ {
1961
+ "epoch": 0.52,
1962
+ "grad_norm": 0.9142821431159973,
1963
+ "learning_rate": 9.845095315330123e-05,
1964
+ "loss": 0.6471,
1965
+ "step": 276
1966
+ },
1967
+ {
1968
+ "epoch": 0.53,
1969
+ "grad_norm": 1.4982346296310425,
1970
+ "learning_rate": 9.783141768266511e-05,
1971
+ "loss": 0.592,
1972
+ "step": 277
1973
+ },
1974
+ {
1975
+ "epoch": 0.53,
1976
+ "grad_norm": 0.7968485951423645,
1977
+ "learning_rate": 9.721196547623584e-05,
1978
+ "loss": 0.6657,
1979
+ "step": 278
1980
+ },
1981
+ {
1982
+ "epoch": 0.53,
1983
+ "grad_norm": 1.2386119365692139,
1984
+ "learning_rate": 9.659262031830612e-05,
1985
+ "loss": 0.7317,
1986
+ "step": 279
1987
+ },
1988
+ {
1989
+ "epoch": 0.53,
1990
+ "grad_norm": 0.7291132807731628,
1991
+ "learning_rate": 9.597340598905852e-05,
1992
+ "loss": 0.6723,
1993
+ "step": 280
1994
+ },
1995
+ {
1996
+ "epoch": 0.53,
1997
+ "grad_norm": 1.0461961030960083,
1998
+ "learning_rate": 9.53543462636522e-05,
1999
+ "loss": 0.7715,
2000
+ "step": 281
2001
+ },
2002
+ {
2003
+ "epoch": 0.53,
2004
+ "grad_norm": 0.74568110704422,
2005
+ "learning_rate": 9.473546491131041e-05,
2006
+ "loss": 0.7389,
2007
+ "step": 282
2008
+ },
2009
+ {
2010
+ "epoch": 0.54,
2011
+ "grad_norm": 0.7399306893348694,
2012
+ "learning_rate": 9.411678569440752e-05,
2013
+ "loss": 0.6715,
2014
+ "step": 283
2015
+ },
2016
+ {
2017
+ "epoch": 0.54,
2018
+ "grad_norm": 0.7953692674636841,
2019
+ "learning_rate": 9.349833236755674e-05,
2020
+ "loss": 0.6911,
2021
+ "step": 284
2022
+ },
2023
+ {
2024
+ "epoch": 0.54,
2025
+ "grad_norm": 0.7449805736541748,
2026
+ "learning_rate": 9.28801286766982e-05,
2027
+ "loss": 0.7159,
2028
+ "step": 285
2029
+ },
2030
+ {
2031
+ "epoch": 0.54,
2032
+ "grad_norm": 0.7682590484619141,
2033
+ "learning_rate": 9.226219835818699e-05,
2034
+ "loss": 0.7643,
2035
+ "step": 286
2036
+ },
2037
+ {
2038
+ "epoch": 0.54,
2039
+ "grad_norm": 0.8503318428993225,
2040
+ "learning_rate": 9.164456513788186e-05,
2041
+ "loss": 0.6482,
2042
+ "step": 287
2043
+ },
2044
+ {
2045
+ "epoch": 0.55,
2046
+ "grad_norm": 1.5629699230194092,
2047
+ "learning_rate": 9.10272527302344e-05,
2048
+ "loss": 0.564,
2049
+ "step": 288
2050
+ },
2051
+ {
2052
+ "epoch": 0.55,
2053
+ "grad_norm": 0.5936041474342346,
2054
+ "learning_rate": 9.041028483737825e-05,
2055
+ "loss": 0.6121,
2056
+ "step": 289
2057
+ },
2058
+ {
2059
+ "epoch": 0.55,
2060
+ "grad_norm": 0.8084244728088379,
2061
+ "learning_rate": 8.979368514821916e-05,
2062
+ "loss": 0.7165,
2063
+ "step": 290
2064
+ },
2065
+ {
2066
+ "epoch": 0.55,
2067
+ "grad_norm": 0.9212177395820618,
2068
+ "learning_rate": 8.917747733752564e-05,
2069
+ "loss": 0.7041,
2070
+ "step": 291
2071
+ },
2072
+ {
2073
+ "epoch": 0.55,
2074
+ "grad_norm": 0.6928449273109436,
2075
+ "learning_rate": 8.856168506501944e-05,
2076
+ "loss": 0.5556,
2077
+ "step": 292
2078
+ },
2079
+ {
2080
+ "epoch": 0.56,
2081
+ "grad_norm": 0.7461227774620056,
2082
+ "learning_rate": 8.79463319744677e-05,
2083
+ "loss": 0.7686,
2084
+ "step": 293
2085
+ },
2086
+ {
2087
+ "epoch": 0.56,
2088
+ "grad_norm": 0.7020424604415894,
2089
+ "learning_rate": 8.733144169277481e-05,
2090
+ "loss": 0.6741,
2091
+ "step": 294
2092
+ },
2093
+ {
2094
+ "epoch": 0.56,
2095
+ "grad_norm": 1.0011357069015503,
2096
+ "learning_rate": 8.671703782907518e-05,
2097
+ "loss": 0.7413,
2098
+ "step": 295
2099
+ },
2100
+ {
2101
+ "epoch": 0.56,
2102
+ "grad_norm": 0.6851476430892944,
2103
+ "learning_rate": 8.610314397382701e-05,
2104
+ "loss": 0.6345,
2105
+ "step": 296
2106
+ },
2107
+ {
2108
+ "epoch": 0.56,
2109
+ "grad_norm": 0.7206119894981384,
2110
+ "learning_rate": 8.548978369790631e-05,
2111
+ "loss": 0.693,
2112
+ "step": 297
2113
+ },
2114
+ {
2115
+ "epoch": 0.57,
2116
+ "grad_norm": 0.7812450528144836,
2117
+ "learning_rate": 8.487698055170192e-05,
2118
+ "loss": 0.6601,
2119
+ "step": 298
2120
+ },
2121
+ {
2122
+ "epoch": 0.57,
2123
+ "grad_norm": 0.9316333532333374,
2124
+ "learning_rate": 8.426475806421138e-05,
2125
+ "loss": 0.67,
2126
+ "step": 299
2127
+ },
2128
+ {
2129
+ "epoch": 0.57,
2130
+ "grad_norm": 0.7167903780937195,
2131
+ "learning_rate": 8.365313974213737e-05,
2132
+ "loss": 0.7628,
2133
+ "step": 300
2134
+ },
2135
+ {
2136
+ "epoch": 0.57,
2137
+ "grad_norm": 1.1033985614776611,
2138
+ "learning_rate": 8.304214906898526e-05,
2139
+ "loss": 0.4936,
2140
+ "step": 301
2141
+ },
2142
+ {
2143
+ "epoch": 0.57,
2144
+ "grad_norm": 0.8084871172904968,
2145
+ "learning_rate": 8.243180950416141e-05,
2146
+ "loss": 0.7081,
2147
+ "step": 302
2148
+ },
2149
+ {
2150
+ "epoch": 0.57,
2151
+ "grad_norm": 0.7424312233924866,
2152
+ "learning_rate": 8.182214448207239e-05,
2153
+ "loss": 0.6034,
2154
+ "step": 303
2155
+ },
2156
+ {
2157
+ "epoch": 0.58,
2158
+ "grad_norm": 0.878241240978241,
2159
+ "learning_rate": 8.121317741122525e-05,
2160
+ "loss": 0.7169,
2161
+ "step": 304
2162
+ },
2163
+ {
2164
+ "epoch": 0.58,
2165
+ "grad_norm": 0.8637294173240662,
2166
+ "learning_rate": 8.060493167332874e-05,
2167
+ "loss": 0.7805,
2168
+ "step": 305
2169
+ },
2170
+ {
2171
+ "epoch": 0.58,
2172
+ "grad_norm": 0.6890109777450562,
2173
+ "learning_rate": 7.999743062239557e-05,
2174
+ "loss": 0.6538,
2175
+ "step": 306
2176
+ },
2177
+ {
2178
+ "epoch": 0.58,
2179
+ "grad_norm": 0.7422961592674255,
2180
+ "learning_rate": 7.939069758384562e-05,
2181
+ "loss": 0.6648,
2182
+ "step": 307
2183
+ },
2184
+ {
2185
+ "epoch": 0.58,
2186
+ "grad_norm": 0.7701572775840759,
2187
+ "learning_rate": 7.878475585361045e-05,
2188
+ "loss": 0.7211,
2189
+ "step": 308
2190
+ },
2191
+ {
2192
+ "epoch": 0.59,
2193
+ "grad_norm": 0.887014627456665,
2194
+ "learning_rate": 7.817962869723876e-05,
2195
+ "loss": 0.7689,
2196
+ "step": 309
2197
+ },
2198
+ {
2199
+ "epoch": 0.59,
2200
+ "grad_norm": 0.6782895922660828,
2201
+ "learning_rate": 7.757533934900316e-05,
2202
+ "loss": 0.7301,
2203
+ "step": 310
2204
+ },
2205
+ {
2206
+ "epoch": 0.59,
2207
+ "grad_norm": 0.7926852107048035,
2208
+ "learning_rate": 7.697191101100801e-05,
2209
+ "loss": 0.7256,
2210
+ "step": 311
2211
+ },
2212
+ {
2213
+ "epoch": 0.59,
2214
+ "grad_norm": 0.8395385146141052,
2215
+ "learning_rate": 7.636936685229863e-05,
2216
+ "loss": 0.7125,
2217
+ "step": 312
2218
+ },
2219
+ {
2220
+ "epoch": 0.59,
2221
+ "grad_norm": 0.874583899974823,
2222
+ "learning_rate": 7.576773000797166e-05,
2223
+ "loss": 0.6143,
2224
+ "step": 313
2225
+ },
2226
+ {
2227
+ "epoch": 0.6,
2228
+ "grad_norm": 0.6823002099990845,
2229
+ "learning_rate": 7.516702357828672e-05,
2230
+ "loss": 0.6415,
2231
+ "step": 314
2232
+ },
2233
+ {
2234
+ "epoch": 0.6,
2235
+ "grad_norm": 0.6727049946784973,
2236
+ "learning_rate": 7.456727062777958e-05,
2237
+ "loss": 0.6053,
2238
+ "step": 315
2239
+ },
2240
+ {
2241
+ "epoch": 0.6,
2242
+ "grad_norm": 0.7634817361831665,
2243
+ "learning_rate": 7.396849418437652e-05,
2244
+ "loss": 0.5891,
2245
+ "step": 316
2246
+ },
2247
+ {
2248
+ "epoch": 0.6,
2249
+ "grad_norm": 0.766338050365448,
2250
+ "learning_rate": 7.337071723851017e-05,
2251
+ "loss": 0.6727,
2252
+ "step": 317
2253
+ },
2254
+ {
2255
+ "epoch": 0.6,
2256
+ "grad_norm": 0.8024521470069885,
2257
+ "learning_rate": 7.277396274223671e-05,
2258
+ "loss": 0.6873,
2259
+ "step": 318
2260
+ },
2261
+ {
2262
+ "epoch": 0.61,
2263
+ "grad_norm": 0.701219916343689,
2264
+ "learning_rate": 7.217825360835473e-05,
2265
+ "loss": 0.5562,
2266
+ "step": 319
2267
+ },
2268
+ {
2269
+ "epoch": 0.61,
2270
+ "grad_norm": 0.8103075623512268,
2271
+ "learning_rate": 7.15836127095254e-05,
2272
+ "loss": 0.5965,
2273
+ "step": 320
2274
+ },
2275
+ {
2276
+ "epoch": 0.61,
2277
+ "grad_norm": 0.8132802248001099,
2278
+ "learning_rate": 7.09900628773942e-05,
2279
+ "loss": 0.7483,
2280
+ "step": 321
2281
+ },
2282
+ {
2283
+ "epoch": 0.61,
2284
+ "grad_norm": 0.7442184686660767,
2285
+ "learning_rate": 7.039762690171447e-05,
2286
+ "loss": 0.7469,
2287
+ "step": 322
2288
+ },
2289
+ {
2290
+ "epoch": 0.61,
2291
+ "grad_norm": 0.7334067821502686,
2292
+ "learning_rate": 6.98063275294722e-05,
2293
+ "loss": 0.5311,
2294
+ "step": 323
2295
+ },
2296
+ {
2297
+ "epoch": 0.61,
2298
+ "grad_norm": 0.6543470621109009,
2299
+ "learning_rate": 6.921618746401273e-05,
2300
+ "loss": 0.6516,
2301
+ "step": 324
2302
+ },
2303
+ {
2304
+ "epoch": 0.62,
2305
+ "grad_norm": 0.6865355372428894,
2306
+ "learning_rate": 6.862722936416897e-05,
2307
+ "loss": 0.6957,
2308
+ "step": 325
2309
+ },
2310
+ {
2311
+ "epoch": 0.62,
2312
+ "grad_norm": 0.8854186534881592,
2313
+ "learning_rate": 6.803947584339148e-05,
2314
+ "loss": 0.6739,
2315
+ "step": 326
2316
+ },
2317
+ {
2318
+ "epoch": 0.62,
2319
+ "grad_norm": 0.7788788676261902,
2320
+ "learning_rate": 6.745294946888023e-05,
2321
+ "loss": 0.6183,
2322
+ "step": 327
2323
+ },
2324
+ {
2325
+ "epoch": 0.62,
2326
+ "grad_norm": 0.7810282707214355,
2327
+ "learning_rate": 6.686767276071803e-05,
2328
+ "loss": 0.6436,
2329
+ "step": 328
2330
+ },
2331
+ {
2332
+ "epoch": 0.62,
2333
+ "grad_norm": 0.7473210096359253,
2334
+ "learning_rate": 6.628366819100585e-05,
2335
+ "loss": 0.6759,
2336
+ "step": 329
2337
+ },
2338
+ {
2339
+ "epoch": 0.63,
2340
+ "grad_norm": 0.8050934076309204,
2341
+ "learning_rate": 6.570095818300012e-05,
2342
+ "loss": 0.6867,
2343
+ "step": 330
2344
+ },
2345
+ {
2346
+ "epoch": 0.63,
2347
+ "grad_norm": 0.7596076726913452,
2348
+ "learning_rate": 6.511956511025157e-05,
2349
+ "loss": 0.6552,
2350
+ "step": 331
2351
+ },
2352
+ {
2353
+ "epoch": 0.63,
2354
+ "grad_norm": 0.8490655422210693,
2355
+ "learning_rate": 6.453951129574644e-05,
2356
+ "loss": 0.7111,
2357
+ "step": 332
2358
+ },
2359
+ {
2360
+ "epoch": 0.63,
2361
+ "grad_norm": 0.7241820096969604,
2362
+ "learning_rate": 6.396081901104909e-05,
2363
+ "loss": 0.6154,
2364
+ "step": 333
2365
+ },
2366
+ {
2367
+ "epoch": 0.63,
2368
+ "grad_norm": 0.6183046102523804,
2369
+ "learning_rate": 6.338351047544707e-05,
2370
+ "loss": 0.5064,
2371
+ "step": 334
2372
+ },
2373
+ {
2374
+ "epoch": 0.64,
2375
+ "grad_norm": 0.8574573993682861,
2376
+ "learning_rate": 6.280760785509801e-05,
2377
+ "loss": 0.7799,
2378
+ "step": 335
2379
+ },
2380
+ {
2381
+ "epoch": 0.64,
2382
+ "grad_norm": 1.0641992092132568,
2383
+ "learning_rate": 6.223313326217828e-05,
2384
+ "loss": 0.5082,
2385
+ "step": 336
2386
+ },
2387
+ {
2388
+ "epoch": 0.64,
2389
+ "grad_norm": 0.9125005602836609,
2390
+ "learning_rate": 6.166010875403429e-05,
2391
+ "loss": 0.7545,
2392
+ "step": 337
2393
+ },
2394
+ {
2395
+ "epoch": 0.64,
2396
+ "grad_norm": 0.8739343285560608,
2397
+ "learning_rate": 6.108855633233546e-05,
2398
+ "loss": 0.5929,
2399
+ "step": 338
2400
+ },
2401
+ {
2402
+ "epoch": 0.64,
2403
+ "grad_norm": 0.6968967914581299,
2404
+ "learning_rate": 6.0518497942229325e-05,
2405
+ "loss": 0.6475,
2406
+ "step": 339
2407
+ },
2408
+ {
2409
+ "epoch": 0.65,
2410
+ "grad_norm": 0.8221940994262695,
2411
+ "learning_rate": 5.9949955471499186e-05,
2412
+ "loss": 0.579,
2413
+ "step": 340
2414
+ },
2415
+ {
2416
+ "epoch": 0.65,
2417
+ "grad_norm": 0.96427983045578,
2418
+ "learning_rate": 5.93829507497235e-05,
2419
+ "loss": 0.7766,
2420
+ "step": 341
2421
+ },
2422
+ {
2423
+ "epoch": 0.65,
2424
+ "grad_norm": 0.8502172231674194,
2425
+ "learning_rate": 5.881750554743779e-05,
2426
+ "loss": 0.7176,
2427
+ "step": 342
2428
+ },
2429
+ {
2430
+ "epoch": 0.65,
2431
+ "grad_norm": 1.1693928241729736,
2432
+ "learning_rate": 5.82536415752988e-05,
2433
+ "loss": 0.6447,
2434
+ "step": 343
2435
+ },
2436
+ {
2437
+ "epoch": 0.65,
2438
+ "grad_norm": 0.7575929760932922,
2439
+ "learning_rate": 5.769138048325087e-05,
2440
+ "loss": 0.7019,
2441
+ "step": 344
2442
+ },
2443
+ {
2444
+ "epoch": 0.65,
2445
+ "grad_norm": 0.9739773869514465,
2446
+ "learning_rate": 5.713074385969457e-05,
2447
+ "loss": 0.7178,
2448
+ "step": 345
2449
+ },
2450
+ {
2451
+ "epoch": 0.66,
2452
+ "grad_norm": 0.7964323163032532,
2453
+ "learning_rate": 5.657175323065802e-05,
2454
+ "loss": 0.6736,
2455
+ "step": 346
2456
+ },
2457
+ {
2458
+ "epoch": 0.66,
2459
+ "grad_norm": 0.8762109875679016,
2460
+ "learning_rate": 5.6014430058970114e-05,
2461
+ "loss": 0.6314,
2462
+ "step": 347
2463
+ },
2464
+ {
2465
+ "epoch": 0.66,
2466
+ "grad_norm": 0.8361276984214783,
2467
+ "learning_rate": 5.545879574343661e-05,
2468
+ "loss": 0.7155,
2469
+ "step": 348
2470
+ },
2471
+ {
2472
+ "epoch": 0.66,
2473
+ "grad_norm": 0.8512857556343079,
2474
+ "learning_rate": 5.490487161801854e-05,
2475
+ "loss": 0.6984,
2476
+ "step": 349
2477
+ },
2478
+ {
2479
+ "epoch": 0.66,
2480
+ "grad_norm": 0.8726841807365417,
2481
+ "learning_rate": 5.435267895101302e-05,
2482
+ "loss": 0.5749,
2483
+ "step": 350
2484
+ },
2485
+ {
2486
+ "epoch": 0.67,
2487
+ "grad_norm": 0.7646642923355103,
2488
+ "learning_rate": 5.3802238944236505e-05,
2489
+ "loss": 0.7181,
2490
+ "step": 351
2491
+ },
2492
+ {
2493
+ "epoch": 0.67,
2494
+ "grad_norm": 0.762646496295929,
2495
+ "learning_rate": 5.325357273221099e-05,
2496
+ "loss": 0.6197,
2497
+ "step": 352
2498
+ },
2499
+ {
2500
+ "epoch": 0.67,
2501
+ "grad_norm": 1.0159621238708496,
2502
+ "learning_rate": 5.270670138135234e-05,
2503
+ "loss": 0.5808,
2504
+ "step": 353
2505
+ },
2506
+ {
2507
+ "epoch": 0.67,
2508
+ "grad_norm": 1.1280730962753296,
2509
+ "learning_rate": 5.216164588916148e-05,
2510
+ "loss": 0.6669,
2511
+ "step": 354
2512
+ },
2513
+ {
2514
+ "epoch": 0.67,
2515
+ "grad_norm": 0.7981805205345154,
2516
+ "learning_rate": 5.161842718341825e-05,
2517
+ "loss": 0.6309,
2518
+ "step": 355
2519
+ },
2520
+ {
2521
+ "epoch": 0.68,
2522
+ "grad_norm": 0.5816403031349182,
2523
+ "learning_rate": 5.107706612137776e-05,
2524
+ "loss": 0.5003,
2525
+ "step": 356
2526
+ },
2527
+ {
2528
+ "epoch": 0.68,
2529
+ "grad_norm": 0.9182299971580505,
2530
+ "learning_rate": 5.0537583488969734e-05,
2531
+ "loss": 0.6022,
2532
+ "step": 357
2533
+ },
2534
+ {
2535
+ "epoch": 0.68,
2536
+ "grad_norm": 0.9387081265449524,
2537
+ "learning_rate": 5.000000000000002e-05,
2538
+ "loss": 0.623,
2539
+ "step": 358
2540
+ },
2541
+ {
2542
+ "epoch": 0.68,
2543
+ "grad_norm": 0.8042228817939758,
2544
+ "learning_rate": 4.9464336295355854e-05,
2545
+ "loss": 0.6668,
2546
+ "step": 359
2547
+ },
2548
+ {
2549
+ "epoch": 0.68,
2550
+ "grad_norm": 0.6955826878547668,
2551
+ "learning_rate": 4.8930612942212916e-05,
2552
+ "loss": 0.6922,
2553
+ "step": 360
2554
+ },
2555
+ {
2556
+ "epoch": 0.68,
2557
+ "grad_norm": 0.7190336585044861,
2558
+ "learning_rate": 4.83988504332457e-05,
2559
+ "loss": 0.5405,
2560
+ "step": 361
2561
+ },
2562
+ {
2563
+ "epoch": 0.69,
2564
+ "grad_norm": 0.6942302584648132,
2565
+ "learning_rate": 4.786906918584083e-05,
2566
+ "loss": 0.7027,
2567
+ "step": 362
2568
+ },
2569
+ {
2570
+ "epoch": 0.69,
2571
+ "grad_norm": 0.7399347424507141,
2572
+ "learning_rate": 4.734128954131304e-05,
2573
+ "loss": 0.6277,
2574
+ "step": 363
2575
+ },
2576
+ {
2577
+ "epoch": 0.69,
2578
+ "grad_norm": 0.8382598161697388,
2579
+ "learning_rate": 4.6815531764124045e-05,
2580
+ "loss": 0.6038,
2581
+ "step": 364
2582
+ },
2583
+ {
2584
+ "epoch": 0.69,
2585
+ "grad_norm": 0.7830659747123718,
2586
+ "learning_rate": 4.629181604110464e-05,
2587
+ "loss": 0.6611,
2588
+ "step": 365
2589
+ },
2590
+ {
2591
+ "epoch": 0.69,
2592
+ "grad_norm": 0.7205358743667603,
2593
+ "learning_rate": 4.5770162480679624e-05,
2594
+ "loss": 0.5469,
2595
+ "step": 366
2596
+ },
2597
+ {
2598
+ "epoch": 0.7,
2599
+ "grad_norm": 0.6976193189620972,
2600
+ "learning_rate": 4.525059111209548e-05,
2601
+ "loss": 0.6208,
2602
+ "step": 367
2603
+ },
2604
+ {
2605
+ "epoch": 0.7,
2606
+ "grad_norm": 0.7750869393348694,
2607
+ "learning_rate": 4.4733121884651664e-05,
2608
+ "loss": 0.6032,
2609
+ "step": 368
2610
+ },
2611
+ {
2612
+ "epoch": 0.7,
2613
+ "grad_norm": 0.7206342816352844,
2614
+ "learning_rate": 4.421777466693434e-05,
2615
+ "loss": 0.6763,
2616
+ "step": 369
2617
+ },
2618
+ {
2619
+ "epoch": 0.7,
2620
+ "grad_norm": 0.8639499545097351,
2621
+ "learning_rate": 4.3704569246053805e-05,
2622
+ "loss": 0.6836,
2623
+ "step": 370
2624
+ },
2625
+ {
2626
+ "epoch": 0.7,
2627
+ "grad_norm": 0.9039255976676941,
2628
+ "learning_rate": 4.3193525326884435e-05,
2629
+ "loss": 0.5905,
2630
+ "step": 371
2631
+ },
2632
+ {
2633
+ "epoch": 0.71,
2634
+ "grad_norm": 0.9770961403846741,
2635
+ "learning_rate": 4.2684662531308386e-05,
2636
+ "loss": 0.7598,
2637
+ "step": 372
2638
+ },
2639
+ {
2640
+ "epoch": 0.71,
2641
+ "grad_norm": 0.9292394518852234,
2642
+ "learning_rate": 4.217800039746206e-05,
2643
+ "loss": 0.741,
2644
+ "step": 373
2645
+ },
2646
+ {
2647
+ "epoch": 0.71,
2648
+ "grad_norm": 0.6980279088020325,
2649
+ "learning_rate": 4.167355837898584e-05,
2650
+ "loss": 0.6229,
2651
+ "step": 374
2652
+ },
2653
+ {
2654
+ "epoch": 0.71,
2655
+ "grad_norm": 0.8407570123672485,
2656
+ "learning_rate": 4.1171355844277394e-05,
2657
+ "loss": 0.6891,
2658
+ "step": 375
2659
+ },
2660
+ {
2661
+ "epoch": 0.71,
2662
+ "grad_norm": 0.7971334457397461,
2663
+ "learning_rate": 4.0671412075747816e-05,
2664
+ "loss": 0.6462,
2665
+ "step": 376
2666
+ },
2667
+ {
2668
+ "epoch": 0.72,
2669
+ "grad_norm": 0.7770645022392273,
2670
+ "learning_rate": 4.017374626908125e-05,
2671
+ "loss": 0.6123,
2672
+ "step": 377
2673
+ },
2674
+ {
2675
+ "epoch": 0.72,
2676
+ "grad_norm": 0.793257474899292,
2677
+ "learning_rate": 3.967837753249804e-05,
2678
+ "loss": 0.612,
2679
+ "step": 378
2680
+ },
2681
+ {
2682
+ "epoch": 0.72,
2683
+ "grad_norm": 0.7542336583137512,
2684
+ "learning_rate": 3.918532488602094e-05,
2685
+ "loss": 0.643,
2686
+ "step": 379
2687
+ },
2688
+ {
2689
+ "epoch": 0.72,
2690
+ "grad_norm": 1.2672817707061768,
2691
+ "learning_rate": 3.869460726074474e-05,
2692
+ "loss": 0.5942,
2693
+ "step": 380
2694
+ },
2695
+ {
2696
+ "epoch": 0.72,
2697
+ "grad_norm": 0.7728064060211182,
2698
+ "learning_rate": 3.820624349810954e-05,
2699
+ "loss": 0.5941,
2700
+ "step": 381
2701
+ },
2702
+ {
2703
+ "epoch": 0.72,
2704
+ "grad_norm": 0.6640836000442505,
2705
+ "learning_rate": 3.772025234917728e-05,
2706
+ "loss": 0.587,
2707
+ "step": 382
2708
+ },
2709
+ {
2710
+ "epoch": 0.73,
2711
+ "grad_norm": 0.7824575304985046,
2712
+ "learning_rate": 3.7236652473911814e-05,
2713
+ "loss": 0.7506,
2714
+ "step": 383
2715
+ },
2716
+ {
2717
+ "epoch": 0.73,
2718
+ "grad_norm": 0.7651994824409485,
2719
+ "learning_rate": 3.675546244046228e-05,
2720
+ "loss": 0.6337,
2721
+ "step": 384
2722
+ },
2723
+ {
2724
+ "epoch": 0.73,
2725
+ "grad_norm": 0.7056587934494019,
2726
+ "learning_rate": 3.6276700724450384e-05,
2727
+ "loss": 0.6715,
2728
+ "step": 385
2729
+ },
2730
+ {
2731
+ "epoch": 0.73,
2732
+ "grad_norm": 0.8649887442588806,
2733
+ "learning_rate": 3.580038570826093e-05,
2734
+ "loss": 0.7218,
2735
+ "step": 386
2736
+ },
2737
+ {
2738
+ "epoch": 0.73,
2739
+ "grad_norm": 0.8019347786903381,
2740
+ "learning_rate": 3.532653568033587e-05,
2741
+ "loss": 0.6234,
2742
+ "step": 387
2743
+ },
2744
+ {
2745
+ "epoch": 0.74,
2746
+ "grad_norm": 0.9476731419563293,
2747
+ "learning_rate": 3.485516883447239e-05,
2748
+ "loss": 0.5199,
2749
+ "step": 388
2750
+ },
2751
+ {
2752
+ "epoch": 0.74,
2753
+ "grad_norm": 0.8756290078163147,
2754
+ "learning_rate": 3.438630326912414e-05,
2755
+ "loss": 0.6379,
2756
+ "step": 389
2757
+ },
2758
+ {
2759
+ "epoch": 0.74,
2760
+ "grad_norm": 0.8843635320663452,
2761
+ "learning_rate": 3.391995698670638e-05,
2762
+ "loss": 0.6983,
2763
+ "step": 390
2764
+ },
2765
+ {
2766
+ "epoch": 0.74,
2767
+ "grad_norm": 0.8399759531021118,
2768
+ "learning_rate": 3.345614789290472e-05,
2769
+ "loss": 0.677,
2770
+ "step": 391
2771
+ },
2772
+ {
2773
+ "epoch": 0.74,
2774
+ "grad_norm": 0.7677200436592102,
2775
+ "learning_rate": 3.2994893795987766e-05,
2776
+ "loss": 0.5998,
2777
+ "step": 392
2778
+ },
2779
+ {
2780
+ "epoch": 0.75,
2781
+ "grad_norm": 1.0266742706298828,
2782
+ "learning_rate": 3.253621240612326e-05,
2783
+ "loss": 0.6444,
2784
+ "step": 393
2785
+ },
2786
+ {
2787
+ "epoch": 0.75,
2788
+ "grad_norm": 0.7664735913276672,
2789
+ "learning_rate": 3.208012133469799e-05,
2790
+ "loss": 0.643,
2791
+ "step": 394
2792
+ },
2793
+ {
2794
+ "epoch": 0.75,
2795
+ "grad_norm": 0.6599723100662231,
2796
+ "learning_rate": 3.162663809364178e-05,
2797
+ "loss": 0.5806,
2798
+ "step": 395
2799
+ },
2800
+ {
2801
+ "epoch": 0.75,
2802
+ "grad_norm": 0.7938780188560486,
2803
+ "learning_rate": 3.117578009475503e-05,
2804
+ "loss": 0.6664,
2805
+ "step": 396
2806
+ },
2807
+ {
2808
+ "epoch": 0.75,
2809
+ "eval_loss": 0.6555132269859314,
2810
+ "eval_runtime": 14.6967,
2811
+ "eval_samples_per_second": 32.116,
2812
+ "eval_steps_per_second": 32.116,
2813
+ "step": 396
2814
+ },
2815
+ {
2816
+ "epoch": 0.75,
2817
+ "grad_norm": 0.6883028149604797,
2818
+ "learning_rate": 3.072756464904006e-05,
2819
+ "loss": 0.6681,
2820
+ "step": 397
2821
+ },
2822
+ {
2823
+ "epoch": 0.76,
2824
+ "grad_norm": 0.7418582439422607,
2825
+ "learning_rate": 3.0282008966036646e-05,
2826
+ "loss": 0.619,
2827
+ "step": 398
2828
+ },
2829
+ {
2830
+ "epoch": 0.76,
2831
+ "grad_norm": 0.7900329232215881,
2832
+ "learning_rate": 2.9839130153161154e-05,
2833
+ "loss": 0.6181,
2834
+ "step": 399
2835
+ },
2836
+ {
2837
+ "epoch": 0.76,
2838
+ "grad_norm": 0.5973490476608276,
2839
+ "learning_rate": 2.9398945215049567e-05,
2840
+ "loss": 0.5325,
2841
+ "step": 400
2842
+ },
2843
+ {
2844
+ "epoch": 0.76,
2845
+ "grad_norm": 0.7216522693634033,
2846
+ "learning_rate": 2.8961471052904852e-05,
2847
+ "loss": 0.6496,
2848
+ "step": 401
2849
+ },
2850
+ {
2851
+ "epoch": 0.76,
2852
+ "grad_norm": 0.6355336308479309,
2853
+ "learning_rate": 2.8526724463847722e-05,
2854
+ "loss": 0.669,
2855
+ "step": 402
2856
+ },
2857
+ {
2858
+ "epoch": 0.76,
2859
+ "grad_norm": 0.849144697189331,
2860
+ "learning_rate": 2.809472214027199e-05,
2861
+ "loss": 0.6917,
2862
+ "step": 403
2863
+ },
2864
+ {
2865
+ "epoch": 0.77,
2866
+ "grad_norm": 1.0107152462005615,
2867
+ "learning_rate": 2.766548066920338e-05,
2868
+ "loss": 0.7632,
2869
+ "step": 404
2870
+ },
2871
+ {
2872
+ "epoch": 0.77,
2873
+ "grad_norm": 0.9305949807167053,
2874
+ "learning_rate": 2.7239016531662887e-05,
2875
+ "loss": 0.6732,
2876
+ "step": 405
2877
+ },
2878
+ {
2879
+ "epoch": 0.77,
2880
+ "grad_norm": 0.6983330249786377,
2881
+ "learning_rate": 2.6815346102033877e-05,
2882
+ "loss": 0.5301,
2883
+ "step": 406
2884
+ },
2885
+ {
2886
+ "epoch": 0.77,
2887
+ "grad_norm": 0.7181352376937866,
2888
+ "learning_rate": 2.6394485647433277e-05,
2889
+ "loss": 0.6498,
2890
+ "step": 407
2891
+ },
2892
+ {
2893
+ "epoch": 0.77,
2894
+ "grad_norm": 0.7354316711425781,
2895
+ "learning_rate": 2.5976451327087204e-05,
2896
+ "loss": 0.5822,
2897
+ "step": 408
2898
+ },
2899
+ {
2900
+ "epoch": 0.78,
2901
+ "grad_norm": 0.7557486891746521,
2902
+ "learning_rate": 2.5561259191710407e-05,
2903
+ "loss": 0.5257,
2904
+ "step": 409
2905
+ },
2906
+ {
2907
+ "epoch": 0.78,
2908
+ "grad_norm": 0.7786766290664673,
2909
+ "learning_rate": 2.514892518288988e-05,
2910
+ "loss": 0.596,
2911
+ "step": 410
2912
+ },
2913
+ {
2914
+ "epoch": 0.78,
2915
+ "grad_norm": 0.8994741439819336,
2916
+ "learning_rate": 2.4739465132473016e-05,
2917
+ "loss": 0.6946,
2918
+ "step": 411
2919
+ },
2920
+ {
2921
+ "epoch": 0.78,
2922
+ "grad_norm": 0.8272367119789124,
2923
+ "learning_rate": 2.4332894761959603e-05,
2924
+ "loss": 0.5883,
2925
+ "step": 412
2926
+ },
2927
+ {
2928
+ "epoch": 0.78,
2929
+ "grad_norm": 0.7980152368545532,
2930
+ "learning_rate": 2.3929229681898003e-05,
2931
+ "loss": 0.6563,
2932
+ "step": 413
2933
+ },
2934
+ {
2935
+ "epoch": 0.79,
2936
+ "grad_norm": 0.7788495421409607,
2937
+ "learning_rate": 2.3528485391286147e-05,
2938
+ "loss": 0.6374,
2939
+ "step": 414
2940
+ },
2941
+ {
2942
+ "epoch": 0.79,
2943
+ "grad_norm": 0.7941696643829346,
2944
+ "learning_rate": 2.3130677276976232e-05,
2945
+ "loss": 0.7056,
2946
+ "step": 415
2947
+ },
2948
+ {
2949
+ "epoch": 0.79,
2950
+ "grad_norm": 0.8505434989929199,
2951
+ "learning_rate": 2.2735820613083834e-05,
2952
+ "loss": 0.6537,
2953
+ "step": 416
2954
+ },
2955
+ {
2956
+ "epoch": 0.79,
2957
+ "grad_norm": 1.4972716569900513,
2958
+ "learning_rate": 2.234393056040166e-05,
2959
+ "loss": 0.6924,
2960
+ "step": 417
2961
+ },
2962
+ {
2963
+ "epoch": 0.79,
2964
+ "grad_norm": 0.7153500914573669,
2965
+ "learning_rate": 2.195502216581734e-05,
2966
+ "loss": 0.5352,
2967
+ "step": 418
2968
+ },
2969
+ {
2970
+ "epoch": 0.79,
2971
+ "grad_norm": 0.7720330953598022,
2972
+ "learning_rate": 2.1569110361735677e-05,
2973
+ "loss": 0.5763,
2974
+ "step": 419
2975
+ },
2976
+ {
2977
+ "epoch": 0.8,
2978
+ "grad_norm": 0.8194432258605957,
2979
+ "learning_rate": 2.118620996550529e-05,
2980
+ "loss": 0.6945,
2981
+ "step": 420
2982
+ },
2983
+ {
2984
+ "epoch": 0.8,
2985
+ "grad_norm": 0.9006208181381226,
2986
+ "learning_rate": 2.0806335678849765e-05,
2987
+ "loss": 0.675,
2988
+ "step": 421
2989
+ },
2990
+ {
2991
+ "epoch": 0.8,
2992
+ "grad_norm": 0.8461016416549683,
2993
+ "learning_rate": 2.0429502087303164e-05,
2994
+ "loss": 0.6308,
2995
+ "step": 422
2996
+ },
2997
+ {
2998
+ "epoch": 0.8,
2999
+ "grad_norm": 4.719590187072754,
3000
+ "learning_rate": 2.0055723659649904e-05,
3001
+ "loss": 0.7509,
3002
+ "step": 423
3003
+ },
3004
+ {
3005
+ "epoch": 0.8,
3006
+ "grad_norm": 0.8160830736160278,
3007
+ "learning_rate": 1.968501474736929e-05,
3008
+ "loss": 0.6519,
3009
+ "step": 424
3010
+ },
3011
+ {
3012
+ "epoch": 0.81,
3013
+ "grad_norm": 0.9067161083221436,
3014
+ "learning_rate": 1.9317389584084568e-05,
3015
+ "loss": 0.7138,
3016
+ "step": 425
3017
+ },
3018
+ {
3019
+ "epoch": 0.81,
3020
+ "grad_norm": 0.8717969655990601,
3021
+ "learning_rate": 1.8952862285016194e-05,
3022
+ "loss": 0.7402,
3023
+ "step": 426
3024
+ },
3025
+ {
3026
+ "epoch": 0.81,
3027
+ "grad_norm": 1.1476932764053345,
3028
+ "learning_rate": 1.8591446846440097e-05,
3029
+ "loss": 0.7367,
3030
+ "step": 427
3031
+ },
3032
+ {
3033
+ "epoch": 0.81,
3034
+ "grad_norm": 0.7396849393844604,
3035
+ "learning_rate": 1.823315714515018e-05,
3036
+ "loss": 0.6647,
3037
+ "step": 428
3038
+ },
3039
+ {
3040
+ "epoch": 0.81,
3041
+ "grad_norm": 0.7775653004646301,
3042
+ "learning_rate": 1.787800693792545e-05,
3043
+ "loss": 0.6635,
3044
+ "step": 429
3045
+ },
3046
+ {
3047
+ "epoch": 0.82,
3048
+ "grad_norm": 0.7602683901786804,
3049
+ "learning_rate": 1.7526009861001956e-05,
3050
+ "loss": 0.6498,
3051
+ "step": 430
3052
+ },
3053
+ {
3054
+ "epoch": 0.82,
3055
+ "grad_norm": 0.7206698656082153,
3056
+ "learning_rate": 1.717717942954914e-05,
3057
+ "loss": 0.5803,
3058
+ "step": 431
3059
+ },
3060
+ {
3061
+ "epoch": 0.82,
3062
+ "grad_norm": 0.97299724817276,
3063
+ "learning_rate": 1.6831529037150827e-05,
3064
+ "loss": 0.721,
3065
+ "step": 432
3066
+ },
3067
+ {
3068
+ "epoch": 0.82,
3069
+ "grad_norm": 1.1216304302215576,
3070
+ "learning_rate": 1.648907195529117e-05,
3071
+ "loss": 0.6673,
3072
+ "step": 433
3073
+ },
3074
+ {
3075
+ "epoch": 0.82,
3076
+ "grad_norm": 0.8716909885406494,
3077
+ "learning_rate": 1.614982133284495e-05,
3078
+ "loss": 0.6804,
3079
+ "step": 434
3080
+ },
3081
+ {
3082
+ "epoch": 0.83,
3083
+ "grad_norm": 0.775269627571106,
3084
+ "learning_rate": 1.5813790195572674e-05,
3085
+ "loss": 0.6865,
3086
+ "step": 435
3087
+ },
3088
+ {
3089
+ "epoch": 0.83,
3090
+ "grad_norm": 0.807775616645813,
3091
+ "learning_rate": 1.5480991445620542e-05,
3092
+ "loss": 0.6331,
3093
+ "step": 436
3094
+ },
3095
+ {
3096
+ "epoch": 0.83,
3097
+ "grad_norm": 0.7059893012046814,
3098
+ "learning_rate": 1.515143786102503e-05,
3099
+ "loss": 0.5684,
3100
+ "step": 437
3101
+ },
3102
+ {
3103
+ "epoch": 0.83,
3104
+ "grad_norm": 0.5673685669898987,
3105
+ "learning_rate": 1.482514209522231e-05,
3106
+ "loss": 0.5452,
3107
+ "step": 438
3108
+ },
3109
+ {
3110
+ "epoch": 0.83,
3111
+ "grad_norm": 0.849505603313446,
3112
+ "learning_rate": 1.4502116676562261e-05,
3113
+ "loss": 0.6149,
3114
+ "step": 439
3115
+ },
3116
+ {
3117
+ "epoch": 0.83,
3118
+ "grad_norm": 0.7047016024589539,
3119
+ "learning_rate": 1.4182374007827603e-05,
3120
+ "loss": 0.6779,
3121
+ "step": 440
3122
+ },
3123
+ {
3124
+ "epoch": 0.84,
3125
+ "grad_norm": 0.9966226816177368,
3126
+ "learning_rate": 1.3865926365757643e-05,
3127
+ "loss": 0.706,
3128
+ "step": 441
3129
+ },
3130
+ {
3131
+ "epoch": 0.84,
3132
+ "grad_norm": 0.8300743103027344,
3133
+ "learning_rate": 1.3552785900576792e-05,
3134
+ "loss": 0.6306,
3135
+ "step": 442
3136
+ },
3137
+ {
3138
+ "epoch": 0.84,
3139
+ "grad_norm": 0.6334244012832642,
3140
+ "learning_rate": 1.324296463552821e-05,
3141
+ "loss": 0.539,
3142
+ "step": 443
3143
+ },
3144
+ {
3145
+ "epoch": 0.84,
3146
+ "grad_norm": 0.8060217499732971,
3147
+ "learning_rate": 1.2936474466412085e-05,
3148
+ "loss": 0.6693,
3149
+ "step": 444
3150
+ },
3151
+ {
3152
+ "epoch": 0.84,
3153
+ "grad_norm": 0.8065308332443237,
3154
+ "learning_rate": 1.263332716112885e-05,
3155
+ "loss": 0.678,
3156
+ "step": 445
3157
+ },
3158
+ {
3159
+ "epoch": 0.85,
3160
+ "grad_norm": 0.7400084137916565,
3161
+ "learning_rate": 1.2333534359227384e-05,
3162
+ "loss": 0.5707,
3163
+ "step": 446
3164
+ },
3165
+ {
3166
+ "epoch": 0.85,
3167
+ "grad_norm": 0.7766801118850708,
3168
+ "learning_rate": 1.203710757145815e-05,
3169
+ "loss": 0.6258,
3170
+ "step": 447
3171
+ },
3172
+ {
3173
+ "epoch": 0.85,
3174
+ "grad_norm": 0.7389542460441589,
3175
+ "learning_rate": 1.1744058179331175e-05,
3176
+ "loss": 0.6995,
3177
+ "step": 448
3178
+ },
3179
+ {
3180
+ "epoch": 0.85,
3181
+ "grad_norm": 0.8142557144165039,
3182
+ "learning_rate": 1.1454397434679021e-05,
3183
+ "loss": 0.6823,
3184
+ "step": 449
3185
+ },
3186
+ {
3187
+ "epoch": 0.85,
3188
+ "grad_norm": 1.016276240348816,
3189
+ "learning_rate": 1.1168136459224842e-05,
3190
+ "loss": 0.7201,
3191
+ "step": 450
3192
+ },
3193
+ {
3194
+ "epoch": 0.86,
3195
+ "grad_norm": 0.7469731569290161,
3196
+ "learning_rate": 1.0885286244155345e-05,
3197
+ "loss": 0.6593,
3198
+ "step": 451
3199
+ },
3200
+ {
3201
+ "epoch": 0.86,
3202
+ "grad_norm": 0.8244340419769287,
3203
+ "learning_rate": 1.0605857649698669e-05,
3204
+ "loss": 0.6278,
3205
+ "step": 452
3206
+ },
3207
+ {
3208
+ "epoch": 0.86,
3209
+ "grad_norm": 0.8314517736434937,
3210
+ "learning_rate": 1.0329861404707564e-05,
3211
+ "loss": 0.6895,
3212
+ "step": 453
3213
+ },
3214
+ {
3215
+ "epoch": 0.86,
3216
+ "grad_norm": 0.7985360026359558,
3217
+ "learning_rate": 1.0057308106247332e-05,
3218
+ "loss": 0.6304,
3219
+ "step": 454
3220
+ },
3221
+ {
3222
+ "epoch": 0.86,
3223
+ "grad_norm": 0.7921756505966187,
3224
+ "learning_rate": 9.788208219188932e-06,
3225
+ "loss": 0.6864,
3226
+ "step": 455
3227
+ },
3228
+ {
3229
+ "epoch": 0.87,
3230
+ "grad_norm": 0.8282172679901123,
3231
+ "learning_rate": 9.522572075807334e-06,
3232
+ "loss": 0.6364,
3233
+ "step": 456
3234
+ },
3235
+ {
3236
+ "epoch": 0.87,
3237
+ "grad_norm": 0.9883142113685608,
3238
+ "learning_rate": 9.260409875384568e-06,
3239
+ "loss": 0.5466,
3240
+ "step": 457
3241
+ },
3242
+ {
3243
+ "epoch": 0.87,
3244
+ "grad_norm": 0.8472033143043518,
3245
+ "learning_rate": 9.001731683818337e-06,
3246
+ "loss": 0.5954,
3247
+ "step": 458
3248
+ },
3249
+ {
3250
+ "epoch": 0.87,
3251
+ "grad_norm": 0.804810643196106,
3252
+ "learning_rate": 8.746547433235364e-06,
3253
+ "loss": 0.6904,
3254
+ "step": 459
3255
+ },
3256
+ {
3257
+ "epoch": 0.87,
3258
+ "grad_norm": 0.716760516166687,
3259
+ "learning_rate": 8.494866921610133e-06,
3260
+ "loss": 0.5326,
3261
+ "step": 460
3262
+ },
3263
+ {
3264
+ "epoch": 0.87,
3265
+ "grad_norm": 1.048392653465271,
3266
+ "learning_rate": 8.246699812388714e-06,
3267
+ "loss": 0.6277,
3268
+ "step": 461
3269
+ },
3270
+ {
3271
+ "epoch": 0.88,
3272
+ "grad_norm": 0.7103719711303711,
3273
+ "learning_rate": 8.002055634117578e-06,
3274
+ "loss": 0.6877,
3275
+ "step": 462
3276
+ },
3277
+ {
3278
+ "epoch": 0.88,
3279
+ "grad_norm": 0.9400436878204346,
3280
+ "learning_rate": 7.760943780077933e-06,
3281
+ "loss": 0.6194,
3282
+ "step": 463
3283
+ },
3284
+ {
3285
+ "epoch": 0.88,
3286
+ "grad_norm": 0.7876155376434326,
3287
+ "learning_rate": 7.523373507924947e-06,
3288
+ "loss": 0.7067,
3289
+ "step": 464
3290
+ },
3291
+ {
3292
+ "epoch": 0.88,
3293
+ "grad_norm": 0.7430306077003479,
3294
+ "learning_rate": 7.289353939332288e-06,
3295
+ "loss": 0.7236,
3296
+ "step": 465
3297
+ },
3298
+ {
3299
+ "epoch": 0.88,
3300
+ "grad_norm": 0.7089613676071167,
3301
+ "learning_rate": 7.058894059641963e-06,
3302
+ "loss": 0.6644,
3303
+ "step": 466
3304
+ },
3305
+ {
3306
+ "epoch": 0.89,
3307
+ "grad_norm": 2.6532530784606934,
3308
+ "learning_rate": 6.8320027175192706e-06,
3309
+ "loss": 0.8418,
3310
+ "step": 467
3311
+ },
3312
+ {
3313
+ "epoch": 0.89,
3314
+ "grad_norm": 0.7472643852233887,
3315
+ "learning_rate": 6.608688624613057e-06,
3316
+ "loss": 0.6375,
3317
+ "step": 468
3318
+ },
3319
+ {
3320
+ "epoch": 0.89,
3321
+ "grad_norm": 0.8089672327041626,
3322
+ "learning_rate": 6.388960355221207e-06,
3323
+ "loss": 0.6339,
3324
+ "step": 469
3325
+ },
3326
+ {
3327
+ "epoch": 0.89,
3328
+ "grad_norm": 0.9210778474807739,
3329
+ "learning_rate": 6.1728263459614796e-06,
3330
+ "loss": 0.6573,
3331
+ "step": 470
3332
+ },
3333
+ {
3334
+ "epoch": 0.89,
3335
+ "grad_norm": 0.7312812209129333,
3336
+ "learning_rate": 5.960294895447549e-06,
3337
+ "loss": 0.6352,
3338
+ "step": 471
3339
+ },
3340
+ {
3341
+ "epoch": 0.9,
3342
+ "grad_norm": 0.7618116140365601,
3343
+ "learning_rate": 5.751374163970347e-06,
3344
+ "loss": 0.4996,
3345
+ "step": 472
3346
+ },
3347
+ {
3348
+ "epoch": 0.9,
3349
+ "grad_norm": 0.7650373578071594,
3350
+ "learning_rate": 5.546072173184791e-06,
3351
+ "loss": 0.6693,
3352
+ "step": 473
3353
+ },
3354
+ {
3355
+ "epoch": 0.9,
3356
+ "grad_norm": 0.8264601826667786,
3357
+ "learning_rate": 5.344396805801766e-06,
3358
+ "loss": 0.5748,
3359
+ "step": 474
3360
+ },
3361
+ {
3362
+ "epoch": 0.9,
3363
+ "grad_norm": 0.6858938932418823,
3364
+ "learning_rate": 5.146355805285452e-06,
3365
+ "loss": 0.7102,
3366
+ "step": 475
3367
+ },
3368
+ {
3369
+ "epoch": 0.9,
3370
+ "grad_norm": 1.2532672882080078,
3371
+ "learning_rate": 4.951956775555999e-06,
3372
+ "loss": 0.5724,
3373
+ "step": 476
3374
+ },
3375
+ {
3376
+ "epoch": 0.9,
3377
+ "grad_norm": 0.8648524284362793,
3378
+ "learning_rate": 4.7612071806976575e-06,
3379
+ "loss": 0.6986,
3380
+ "step": 477
3381
+ },
3382
+ {
3383
+ "epoch": 0.91,
3384
+ "grad_norm": 0.6834442615509033,
3385
+ "learning_rate": 4.574114344672042e-06,
3386
+ "loss": 0.6115,
3387
+ "step": 478
3388
+ },
3389
+ {
3390
+ "epoch": 0.91,
3391
+ "grad_norm": 1.053343415260315,
3392
+ "learning_rate": 4.390685451037025e-06,
3393
+ "loss": 0.6634,
3394
+ "step": 479
3395
+ },
3396
+ {
3397
+ "epoch": 0.91,
3398
+ "grad_norm": 0.6925364136695862,
3399
+ "learning_rate": 4.210927542670917e-06,
3400
+ "loss": 0.5087,
3401
+ "step": 480
3402
+ },
3403
+ {
3404
+ "epoch": 0.91,
3405
+ "grad_norm": 0.9825167059898376,
3406
+ "learning_rate": 4.034847521502028e-06,
3407
+ "loss": 0.5799,
3408
+ "step": 481
3409
+ },
3410
+ {
3411
+ "epoch": 0.91,
3412
+ "grad_norm": 0.7158229947090149,
3413
+ "learning_rate": 3.862452148243622e-06,
3414
+ "loss": 0.6089,
3415
+ "step": 482
3416
+ },
3417
+ {
3418
+ "epoch": 0.92,
3419
+ "grad_norm": 1.0506715774536133,
3420
+ "learning_rate": 3.693748042134415e-06,
3421
+ "loss": 0.6954,
3422
+ "step": 483
3423
+ },
3424
+ {
3425
+ "epoch": 0.92,
3426
+ "grad_norm": 0.75848788022995,
3427
+ "learning_rate": 3.528741680684411e-06,
3428
+ "loss": 0.6844,
3429
+ "step": 484
3430
+ },
3431
+ {
3432
+ "epoch": 0.92,
3433
+ "grad_norm": 0.8962730169296265,
3434
+ "learning_rate": 3.367439399426087e-06,
3435
+ "loss": 0.6411,
3436
+ "step": 485
3437
+ },
3438
+ {
3439
+ "epoch": 0.92,
3440
+ "grad_norm": 0.7715455293655396,
3441
+ "learning_rate": 3.2098473916712946e-06,
3442
+ "loss": 0.5918,
3443
+ "step": 486
3444
+ },
3445
+ {
3446
+ "epoch": 0.92,
3447
+ "grad_norm": 0.6626971364021301,
3448
+ "learning_rate": 3.055971708273375e-06,
3449
+ "loss": 0.4995,
3450
+ "step": 487
3451
+ },
3452
+ {
3453
+ "epoch": 0.93,
3454
+ "grad_norm": 0.9594528675079346,
3455
+ "learning_rate": 2.905818257394799e-06,
3456
+ "loss": 0.6345,
3457
+ "step": 488
3458
+ },
3459
+ {
3460
+ "epoch": 0.93,
3461
+ "grad_norm": 0.9263612627983093,
3462
+ "learning_rate": 2.759392804280414e-06,
3463
+ "loss": 0.6999,
3464
+ "step": 489
3465
+ },
3466
+ {
3467
+ "epoch": 0.93,
3468
+ "grad_norm": 0.7584457993507385,
3469
+ "learning_rate": 2.616700971036001e-06,
3470
+ "loss": 0.6901,
3471
+ "step": 490
3472
+ },
3473
+ {
3474
+ "epoch": 0.93,
3475
+ "grad_norm": 0.9710139036178589,
3476
+ "learning_rate": 2.4777482364124695e-06,
3477
+ "loss": 0.7086,
3478
+ "step": 491
3479
+ },
3480
+ {
3481
+ "epoch": 0.93,
3482
+ "grad_norm": 0.7744004726409912,
3483
+ "learning_rate": 2.342539935595445e-06,
3484
+ "loss": 0.6366,
3485
+ "step": 492
3486
+ },
3487
+ {
3488
+ "epoch": 0.94,
3489
+ "grad_norm": 0.7367916703224182,
3490
+ "learning_rate": 2.2110812600004694e-06,
3491
+ "loss": 0.5976,
3492
+ "step": 493
3493
+ },
3494
+ {
3495
+ "epoch": 0.94,
3496
+ "grad_norm": 0.9458748698234558,
3497
+ "learning_rate": 2.0833772570736375e-06,
3498
+ "loss": 0.6082,
3499
+ "step": 494
3500
+ },
3501
+ {
3502
+ "epoch": 0.94,
3503
+ "grad_norm": 0.8267062902450562,
3504
+ "learning_rate": 1.959432830097807e-06,
3505
+ "loss": 0.4377,
3506
+ "step": 495
3507
+ },
3508
+ {
3509
+ "epoch": 0.94,
3510
+ "grad_norm": 0.8359901905059814,
3511
+ "learning_rate": 1.83925273800436e-06,
3512
+ "loss": 0.6793,
3513
+ "step": 496
3514
+ },
3515
+ {
3516
+ "epoch": 0.94,
3517
+ "grad_norm": 0.6290463805198669,
3518
+ "learning_rate": 1.7228415951904165e-06,
3519
+ "loss": 0.579,
3520
+ "step": 497
3521
+ },
3522
+ {
3523
+ "epoch": 0.94,
3524
+ "grad_norm": 1.215668797492981,
3525
+ "learning_rate": 1.61020387134172e-06,
3526
+ "loss": 0.617,
3527
+ "step": 498
3528
+ },
3529
+ {
3530
+ "epoch": 0.95,
3531
+ "grad_norm": 0.8642238974571228,
3532
+ "learning_rate": 1.50134389126102e-06,
3533
+ "loss": 0.6548,
3534
+ "step": 499
3535
+ },
3536
+ {
3537
+ "epoch": 0.95,
3538
+ "grad_norm": 0.8044393062591553,
3539
+ "learning_rate": 1.396265834701982e-06,
3540
+ "loss": 0.6726,
3541
+ "step": 500
3542
+ },
3543
+ {
3544
+ "epoch": 0.95,
3545
+ "grad_norm": 0.786636233329773,
3546
+ "learning_rate": 1.2949737362087156e-06,
3547
+ "loss": 0.5501,
3548
+ "step": 501
3549
+ },
3550
+ {
3551
+ "epoch": 0.95,
3552
+ "grad_norm": 0.8221418261528015,
3553
+ "learning_rate": 1.1974714849608882e-06,
3554
+ "loss": 0.6528,
3555
+ "step": 502
3556
+ },
3557
+ {
3558
+ "epoch": 0.95,
3559
+ "grad_norm": 0.8847975134849548,
3560
+ "learning_rate": 1.103762824624377e-06,
3561
+ "loss": 0.6465,
3562
+ "step": 503
3563
+ },
3564
+ {
3565
+ "epoch": 0.96,
3566
+ "grad_norm": 0.9078425168991089,
3567
+ "learning_rate": 1.0138513532075067e-06,
3568
+ "loss": 0.5795,
3569
+ "step": 504
3570
+ },
3571
+ {
3572
+ "epoch": 0.96,
3573
+ "grad_norm": 0.8821399211883545,
3574
+ "learning_rate": 9.277405229229708e-07,
3575
+ "loss": 0.6663,
3576
+ "step": 505
3577
+ },
3578
+ {
3579
+ "epoch": 0.96,
3580
+ "grad_norm": 0.8557557463645935,
3581
+ "learning_rate": 8.454336400552154e-07,
3582
+ "loss": 0.6063,
3583
+ "step": 506
3584
+ },
3585
+ {
3586
+ "epoch": 0.96,
3587
+ "grad_norm": 0.8125676512718201,
3588
+ "learning_rate": 7.669338648334856e-07,
3589
+ "loss": 0.6687,
3590
+ "step": 507
3591
+ },
3592
+ {
3593
+ "epoch": 0.96,
3594
+ "grad_norm": 0.698814868927002,
3595
+ "learning_rate": 6.922442113105665e-07,
3596
+ "loss": 0.6499,
3597
+ "step": 508
3598
+ },
3599
+ {
3600
+ "epoch": 0.97,
3601
+ "grad_norm": 0.7289395928382874,
3602
+ "learning_rate": 6.21367547246976e-07,
3603
+ "loss": 0.6431,
3604
+ "step": 509
3605
+ },
3606
+ {
3607
+ "epoch": 0.97,
3608
+ "grad_norm": 0.6379066705703735,
3609
+ "learning_rate": 5.543065940008862e-07,
3610
+ "loss": 0.633,
3611
+ "step": 510
3612
+ },
3613
+ {
3614
+ "epoch": 0.97,
3615
+ "grad_norm": 0.9233940839767456,
3616
+ "learning_rate": 4.910639264236294e-07,
3617
+ "loss": 0.6173,
3618
+ "step": 511
3619
+ },
3620
+ {
3621
+ "epoch": 0.97,
3622
+ "grad_norm": 0.8638243079185486,
3623
+ "learning_rate": 4.316419727608434e-07,
3624
+ "loss": 0.7366,
3625
+ "step": 512
3626
+ },
3627
+ {
3628
+ "epoch": 0.97,
3629
+ "grad_norm": 0.7939671277999878,
3630
+ "learning_rate": 3.760430145592575e-07,
3631
+ "loss": 0.645,
3632
+ "step": 513
3633
+ },
3634
+ {
3635
+ "epoch": 0.98,
3636
+ "grad_norm": 0.8512327671051025,
3637
+ "learning_rate": 3.2426918657900704e-07,
3638
+ "loss": 0.6733,
3639
+ "step": 514
3640
+ },
3641
+ {
3642
+ "epoch": 0.98,
3643
+ "grad_norm": 0.7165814638137817,
3644
+ "learning_rate": 2.7632247671177667e-07,
3645
+ "loss": 0.5626,
3646
+ "step": 515
3647
+ },
3648
+ {
3649
+ "epoch": 0.98,
3650
+ "grad_norm": 0.7893707752227783,
3651
+ "learning_rate": 2.3220472590440579e-07,
3652
+ "loss": 0.7017,
3653
+ "step": 516
3654
+ },
3655
+ {
3656
+ "epoch": 0.98,
3657
+ "grad_norm": 0.6842373013496399,
3658
+ "learning_rate": 1.919176280882229e-07,
3659
+ "loss": 0.6092,
3660
+ "step": 517
3661
+ },
3662
+ {
3663
+ "epoch": 0.98,
3664
+ "grad_norm": 0.6570567488670349,
3665
+ "learning_rate": 1.554627301140199e-07,
3666
+ "loss": 0.518,
3667
+ "step": 518
3668
+ },
3669
+ {
3670
+ "epoch": 0.98,
3671
+ "grad_norm": 0.9338929653167725,
3672
+ "learning_rate": 1.2284143169261075e-07,
3673
+ "loss": 0.6498,
3674
+ "step": 519
3675
+ },
3676
+ {
3677
+ "epoch": 0.99,
3678
+ "grad_norm": 0.7087960243225098,
3679
+ "learning_rate": 9.405498534115209e-08,
3680
+ "loss": 0.6872,
3681
+ "step": 520
3682
+ },
3683
+ {
3684
+ "epoch": 0.99,
3685
+ "grad_norm": 0.9282191395759583,
3686
+ "learning_rate": 6.910449633501514e-08,
3687
+ "loss": 0.7439,
3688
+ "step": 521
3689
+ },
3690
+ {
3691
+ "epoch": 0.99,
3692
+ "grad_norm": 0.8347114324569702,
3693
+ "learning_rate": 4.799092266535299e-08,
3694
+ "loss": 0.6346,
3695
+ "step": 522
3696
+ },
3697
+ {
3698
+ "epoch": 0.99,
3699
+ "grad_norm": 1.1854585409164429,
3700
+ "learning_rate": 3.071507500231885e-08,
3701
+ "loss": 0.7587,
3702
+ "step": 523
3703
+ },
3704
+ {
3705
+ "epoch": 0.99,
3706
+ "grad_norm": 0.7577679753303528,
3707
+ "learning_rate": 1.727761666394656e-08,
3708
+ "loss": 0.7224,
3709
+ "step": 524
3710
+ },
3711
+ {
3712
+ "epoch": 1.0,
3713
+ "grad_norm": 0.9748638868331909,
3714
+ "learning_rate": 7.679063590670942e-09,
3715
+ "loss": 0.59,
3716
+ "step": 525
3717
+ },
3718
+ {
3719
+ "epoch": 1.0,
3720
+ "grad_norm": 1.2704914808273315,
3721
+ "learning_rate": 1.919784325521423e-09,
3722
+ "loss": 0.6941,
3723
+ "step": 526
3724
+ },
3725
+ {
3726
+ "epoch": 1.0,
3727
+ "grad_norm": 0.7523401379585266,
3728
+ "learning_rate": 0.0,
3729
+ "loss": 0.6226,
3730
+ "step": 527
3731
+ }
3732
+ ],
3733
+ "logging_steps": 1,
3734
+ "max_steps": 527,
3735
+ "num_input_tokens_seen": 0,
3736
+ "num_train_epochs": 1,
3737
+ "save_steps": 500,
3738
+ "total_flos": 7.975101399588864e+16,
3739
+ "train_batch_size": 1,
3740
+ "trial_name": null,
3741
+ "trial_params": null
3742
+ }
checkpoint-527/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc063ac87d7d75d2424e6f7bc5d3a41cf0a4d792bb021c6147681f06a6f84ac3
3
+ size 5624
config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openlm-research/open_llama_3b_v2",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 3200,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8640,
14
+ "max_position_embeddings": 2048,
15
+ "model_type": "llama",
16
+ "num_attention_heads": 32,
17
+ "num_hidden_layers": 26,
18
+ "num_key_value_heads": 32,
19
+ "pad_token_id": 0,
20
+ "pretraining_tp": 1,
21
+ "quantization_config": {
22
+ "_load_in_4bit": false,
23
+ "_load_in_8bit": true,
24
+ "bnb_4bit_compute_dtype": "float32",
25
+ "bnb_4bit_quant_type": "fp4",
26
+ "bnb_4bit_use_double_quant": false,
27
+ "llm_int8_enable_fp32_cpu_offload": false,
28
+ "llm_int8_has_fp16_weight": false,
29
+ "llm_int8_skip_modules": null,
30
+ "llm_int8_threshold": 6.0,
31
+ "load_in_4bit": false,
32
+ "load_in_8bit": true,
33
+ "quant_method": "bitsandbytes"
34
+ },
35
+ "rms_norm_eps": 1e-06,
36
+ "rope_scaling": null,
37
+ "rope_theta": 10000.0,
38
+ "tie_word_embeddings": false,
39
+ "torch_dtype": "float16",
40
+ "transformers_version": "4.38.2",
41
+ "use_cache": false,
42
+ "vocab_size": 32000
43
+ }
runs/Mar09_05-23-07_9b5078085e9b/events.out.tfevents.1709961787.9b5078085e9b.1472.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c97ce298bd3fe8ca2eeab1fa207f792f1031b0b4e5d068582a0c35e09553c17
3
+ size 117940
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91b289e85fa20fd375d8b33dc12f77616f18abc6359804471d1fafcb425fecb8
3
+ size 511574
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "legacy": true,
35
+ "model_max_length": 2048,
36
+ "pad_token": "</s>",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "unk_token": "<unk>",
41
+ "use_default_system_prompt": false,
42
+ "use_fast": true
43
+ }