Hugging Face's logo

Volko76
/
test

PEFT
TensorBoard
Safetensors
llama
Generated from Trainer
4-bit precision
bitsandbytes
Model card Files
xet
Metrics Community
Volko76 commited on
Commit
1eb0aee
·
verified ·
1 Parent(s): 20f9746

Upload folder using huggingface_hub

Browse files
Files changed (19) hide show
  1. README.md +135 -0
  2. adapter_config.json +34 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-426/README.md +202 -0
  5. checkpoint-426/adapter_config.json +34 -0
  6. checkpoint-426/adapter_model.safetensors +3 -0
  7. checkpoint-426/optimizer.pt +3 -0
  8. checkpoint-426/rng_state.pth +3 -0
  9. checkpoint-426/scheduler.pt +3 -0
  10. checkpoint-426/special_tokens_map.json +24 -0
  11. checkpoint-426/tokenizer.model +3 -0
  12. checkpoint-426/tokenizer_config.json +43 -0
  13. checkpoint-426/trainer_state.json +3035 -0
  14. checkpoint-426/training_args.bin +3 -0
  15. config.json +44 -0
  16. runs/Apr10_21-24-13_89272c674ade/events.out.tfevents.1712784253.89272c674ade.474.0 +3 -0
  17. special_tokens_map.json +24 -0
  18. tokenizer.model +3 -0
  19. tokenizer_config.json +43 -0
README.md ADDED
@@ -0,0 +1,135 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: qlora-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: false
24
+ load_in_4bit: true
25
+ strict: false
26
+ push_dataset_to_hub:
27
+ datasets:
28
+ - path: mhenrichsen/alpaca_2k_test
29
+ type: alpaca
30
+ dataset_prepared_path:
31
+ val_set_size: 0.05
32
+ adapter: qlora
33
+ lora_model_dir:
34
+ sequence_len: 1024
35
+ sample_packing: true
36
+ lora_r: 8
37
+ lora_alpha: 32
38
+ lora_dropout: 0.05
39
+ lora_target_modules:
40
+ lora_target_linear: true
41
+ lora_fan_in_fan_out:
42
+ wandb_project:
43
+ wandb_entity:
44
+ wandb_watch:
45
+ wandb_name:
46
+ wandb_log_model:
47
+ output_dir: ./qlora-out
48
+ gradient_accumulation_steps: 1
49
+ micro_batch_size: 1
50
+ num_epochs: 1
51
+ optimizer: paged_adamw_32bit
52
+ torchdistx_path:
53
+ lr_scheduler: cosine
54
+ learning_rate: 0.0002
55
+ train_on_inputs: false
56
+ group_by_length: false
57
+ bf16: false
58
+ fp16: true
59
+ tf32: false
60
+ gradient_checkpointing: true
61
+ early_stopping_patience:
62
+ resume_from_checkpoint:
63
+ local_rank:
64
+ logging_steps: 1
65
+ xformers_attention:
66
+ flash_attention: true
67
+ gptq_groupsize:
68
+ gptq_model_v1:
69
+ warmup_steps: 20
70
+ evals_per_epoch: 4
71
+ saves_per_epoch: 1
72
+ debug:
73
+ deepspeed:
74
+ weight_decay: 0.1
75
+ fsdp:
76
+ fsdp_config:
77
+ special_tokens:
78
+ bos_token: "<s>"
79
+ eos_token: "</s>"
80
+ unk_token: "<unk>"
81
+
82
+ ```
83
+
84
+ </details><br>
85
+
86
+ # qlora-out
87
+
88
+ 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.
89
+ It achieves the following results on the evaluation set:
90
+ - Loss: 1.1110
91
+
92
+ ## Model description
93
+
94
+ More information needed
95
+
96
+ ## Intended uses & limitations
97
+
98
+ More information needed
99
+
100
+ ## Training and evaluation data
101
+
102
+ More information needed
103
+
104
+ ## Training procedure
105
+
106
+ ### Training hyperparameters
107
+
108
+ The following hyperparameters were used during training:
109
+ - learning_rate: 0.0002
110
+ - train_batch_size: 1
111
+ - eval_batch_size: 1
112
+ - seed: 42
113
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
114
+ - lr_scheduler_type: cosine
115
+ - lr_scheduler_warmup_steps: 20
116
+ - num_epochs: 1
117
+ - mixed_precision_training: Native AMP
118
+
119
+ ### Training results
120
+
121
+ | Training Loss | Epoch | Step | Validation Loss |
122
+ |:-------------:|:-----:|:----:|:---------------:|
123
+ | 1.4577 | 0.0 | 1 | 1.3470 |
124
+ | 1.3192 | 0.25 | 107 | 1.1376 |
125
+ | 1.3095 | 0.5 | 214 | 1.1225 |
126
+ | 1.3178 | 0.75 | 321 | 1.1110 |
127
+
128
+
129
+ ### Framework versions
130
+
131
+ - PEFT 0.10.0
132
+ - Transformers 4.40.0.dev0
133
+ - Pytorch 2.1.2+cu118
134
+ - Datasets 2.18.0
135
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 32,
14
+ "lora_dropout": 0.05,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 8,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "down_proj",
24
+ "k_proj",
25
+ "up_proj",
26
+ "gate_proj",
27
+ "v_proj",
28
+ "q_proj",
29
+ "o_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:657691f9cb33f9d9f9aa25cbb349e6c5e413d7ff0efaa644da1a965f54ca7607
3
+ size 50982842
checkpoint-426/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 0.10.0
checkpoint-426/adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 32,
14
+ "lora_dropout": 0.05,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 8,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "down_proj",
24
+ "k_proj",
25
+ "up_proj",
26
+ "gate_proj",
27
+ "v_proj",
28
+ "q_proj",
29
+ "o_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }
checkpoint-426/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:06cb57b30fc3ea05f7090d9e3e25b54006785b0872748ad67958b9a4c00a1f64
3
+ size 50899792
checkpoint-426/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c1f6c9b943679ca9a70bf56145ba7e783afb2ca4699e33847eaf1b43e035051
3
+ size 101919290
checkpoint-426/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:663f136d6f5874f47794514754c8bc188e713f7b35e6aff342851b5525f6fa76
3
+ size 14244
checkpoint-426/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fca02f23aa183f137360b6f00ec0358b364aebd03cc5dd5fdcc96a563d6326e
3
+ size 1064
checkpoint-426/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
+ }
checkpoint-426/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91b289e85fa20fd375d8b33dc12f77616f18abc6359804471d1fafcb425fecb8
3
+ size 511574
checkpoint-426/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
+ }
checkpoint-426/trainer_state.json ADDED
@@ -0,0 +1,3035 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 107,
6
+ "global_step": 426,
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": 0.7947345972061157,
14
+ "learning_rate": 1e-05,
15
+ "loss": 1.4577,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0,
20
+ "eval_loss": 1.346970796585083,
21
+ "eval_runtime": 5.5779,
22
+ "eval_samples_per_second": 17.928,
23
+ "eval_steps_per_second": 17.928,
24
+ "step": 1
25
+ },
26
+ {
27
+ "epoch": 0.0,
28
+ "grad_norm": 0.780886709690094,
29
+ "learning_rate": 2e-05,
30
+ "loss": 1.6627,
31
+ "step": 2
32
+ },
33
+ {
34
+ "epoch": 0.01,
35
+ "grad_norm": 0.8595380187034607,
36
+ "learning_rate": 3e-05,
37
+ "loss": 1.2373,
38
+ "step": 3
39
+ },
40
+ {
41
+ "epoch": 0.01,
42
+ "grad_norm": 1.386765956878662,
43
+ "learning_rate": 4e-05,
44
+ "loss": 1.4095,
45
+ "step": 4
46
+ },
47
+ {
48
+ "epoch": 0.01,
49
+ "grad_norm": 0.8326601386070251,
50
+ "learning_rate": 5e-05,
51
+ "loss": 1.385,
52
+ "step": 5
53
+ },
54
+ {
55
+ "epoch": 0.01,
56
+ "grad_norm": 1.0141901969909668,
57
+ "learning_rate": 6e-05,
58
+ "loss": 1.311,
59
+ "step": 6
60
+ },
61
+ {
62
+ "epoch": 0.02,
63
+ "grad_norm": 0.9190280437469482,
64
+ "learning_rate": 7e-05,
65
+ "loss": 1.394,
66
+ "step": 7
67
+ },
68
+ {
69
+ "epoch": 0.02,
70
+ "grad_norm": 1.0408940315246582,
71
+ "learning_rate": 8e-05,
72
+ "loss": 1.1983,
73
+ "step": 8
74
+ },
75
+ {
76
+ "epoch": 0.02,
77
+ "grad_norm": 1.217496633529663,
78
+ "learning_rate": 9e-05,
79
+ "loss": 1.4083,
80
+ "step": 9
81
+ },
82
+ {
83
+ "epoch": 0.02,
84
+ "grad_norm": 1.0259888172149658,
85
+ "learning_rate": 0.0001,
86
+ "loss": 1.527,
87
+ "step": 10
88
+ },
89
+ {
90
+ "epoch": 0.03,
91
+ "grad_norm": 0.8085443377494812,
92
+ "learning_rate": 0.00011000000000000002,
93
+ "loss": 1.1892,
94
+ "step": 11
95
+ },
96
+ {
97
+ "epoch": 0.03,
98
+ "grad_norm": 1.468948245048523,
99
+ "learning_rate": 0.00012,
100
+ "loss": 1.6414,
101
+ "step": 12
102
+ },
103
+ {
104
+ "epoch": 0.03,
105
+ "grad_norm": 1.1099722385406494,
106
+ "learning_rate": 0.00013000000000000002,
107
+ "loss": 1.8412,
108
+ "step": 13
109
+ },
110
+ {
111
+ "epoch": 0.03,
112
+ "grad_norm": 1.6442979574203491,
113
+ "learning_rate": 0.00014,
114
+ "loss": 1.3497,
115
+ "step": 14
116
+ },
117
+ {
118
+ "epoch": 0.04,
119
+ "grad_norm": 0.8905379176139832,
120
+ "learning_rate": 0.00015000000000000001,
121
+ "loss": 0.7155,
122
+ "step": 15
123
+ },
124
+ {
125
+ "epoch": 0.04,
126
+ "grad_norm": 1.1754573583602905,
127
+ "learning_rate": 0.00016,
128
+ "loss": 2.5543,
129
+ "step": 16
130
+ },
131
+ {
132
+ "epoch": 0.04,
133
+ "grad_norm": 1.4055070877075195,
134
+ "learning_rate": 0.00017,
135
+ "loss": 1.3637,
136
+ "step": 17
137
+ },
138
+ {
139
+ "epoch": 0.04,
140
+ "grad_norm": 1.1854274272918701,
141
+ "learning_rate": 0.00018,
142
+ "loss": 0.9771,
143
+ "step": 18
144
+ },
145
+ {
146
+ "epoch": 0.04,
147
+ "grad_norm": 1.0462994575500488,
148
+ "learning_rate": 0.00019,
149
+ "loss": 1.3884,
150
+ "step": 19
151
+ },
152
+ {
153
+ "epoch": 0.05,
154
+ "grad_norm": 1.3554202318191528,
155
+ "learning_rate": 0.0002,
156
+ "loss": 1.3547,
157
+ "step": 20
158
+ },
159
+ {
160
+ "epoch": 0.05,
161
+ "grad_norm": 1.1008217334747314,
162
+ "learning_rate": 0.00019999700625010443,
163
+ "loss": 0.7969,
164
+ "step": 21
165
+ },
166
+ {
167
+ "epoch": 0.05,
168
+ "grad_norm": 1.003470540046692,
169
+ "learning_rate": 0.0001999880251796685,
170
+ "loss": 1.3606,
171
+ "step": 22
172
+ },
173
+ {
174
+ "epoch": 0.05,
175
+ "grad_norm": 0.8562943935394287,
176
+ "learning_rate": 0.00019997305732643374,
177
+ "loss": 1.1796,
178
+ "step": 23
179
+ },
180
+ {
181
+ "epoch": 0.06,
182
+ "grad_norm": 1.0724220275878906,
183
+ "learning_rate": 0.00019995210358660038,
184
+ "loss": 1.2886,
185
+ "step": 24
186
+ },
187
+ {
188
+ "epoch": 0.06,
189
+ "grad_norm": 1.0882127285003662,
190
+ "learning_rate": 0.00019992516521477352,
191
+ "loss": 1.2085,
192
+ "step": 25
193
+ },
194
+ {
195
+ "epoch": 0.06,
196
+ "grad_norm": 0.8728991150856018,
197
+ "learning_rate": 0.00019989224382388813,
198
+ "loss": 1.2168,
199
+ "step": 26
200
+ },
201
+ {
202
+ "epoch": 0.06,
203
+ "grad_norm": 1.0858105421066284,
204
+ "learning_rate": 0.00019985334138511237,
205
+ "loss": 1.1404,
206
+ "step": 27
207
+ },
208
+ {
209
+ "epoch": 0.07,
210
+ "grad_norm": 1.2043813467025757,
211
+ "learning_rate": 0.00019980846022772978,
212
+ "loss": 1.4167,
213
+ "step": 28
214
+ },
215
+ {
216
+ "epoch": 0.07,
217
+ "grad_norm": 0.8915519714355469,
218
+ "learning_rate": 0.00019975760303899952,
219
+ "loss": 1.3995,
220
+ "step": 29
221
+ },
222
+ {
223
+ "epoch": 0.07,
224
+ "grad_norm": 0.8792808055877686,
225
+ "learning_rate": 0.0001997007728639956,
226
+ "loss": 1.2375,
227
+ "step": 30
228
+ },
229
+ {
230
+ "epoch": 0.07,
231
+ "grad_norm": 0.8958542346954346,
232
+ "learning_rate": 0.0001996379731054247,
233
+ "loss": 1.0913,
234
+ "step": 31
235
+ },
236
+ {
237
+ "epoch": 0.08,
238
+ "grad_norm": 1.0441163778305054,
239
+ "learning_rate": 0.00019956920752342225,
240
+ "loss": 1.2435,
241
+ "step": 32
242
+ },
243
+ {
244
+ "epoch": 0.08,
245
+ "grad_norm": 1.2748677730560303,
246
+ "learning_rate": 0.00019949448023532726,
247
+ "loss": 1.5485,
248
+ "step": 33
249
+ },
250
+ {
251
+ "epoch": 0.08,
252
+ "grad_norm": 1.4271036386489868,
253
+ "learning_rate": 0.00019941379571543596,
254
+ "loss": 1.2604,
255
+ "step": 34
256
+ },
257
+ {
258
+ "epoch": 0.08,
259
+ "grad_norm": 1.1877238750457764,
260
+ "learning_rate": 0.00019932715879473386,
261
+ "loss": 1.391,
262
+ "step": 35
263
+ },
264
+ {
265
+ "epoch": 0.08,
266
+ "grad_norm": 0.8864910006523132,
267
+ "learning_rate": 0.00019923457466060636,
268
+ "loss": 0.8182,
269
+ "step": 36
270
+ },
271
+ {
272
+ "epoch": 0.09,
273
+ "grad_norm": 0.9275113344192505,
274
+ "learning_rate": 0.00019913604885652832,
275
+ "loss": 1.1828,
276
+ "step": 37
277
+ },
278
+ {
279
+ "epoch": 0.09,
280
+ "grad_norm": 0.8808721899986267,
281
+ "learning_rate": 0.00019903158728173205,
282
+ "loss": 0.5187,
283
+ "step": 38
284
+ },
285
+ {
286
+ "epoch": 0.09,
287
+ "grad_norm": 1.3636075258255005,
288
+ "learning_rate": 0.00019892119619085413,
289
+ "loss": 1.2401,
290
+ "step": 39
291
+ },
292
+ {
293
+ "epoch": 0.09,
294
+ "grad_norm": 0.9511018991470337,
295
+ "learning_rate": 0.00019880488219356087,
296
+ "loss": 1.3375,
297
+ "step": 40
298
+ },
299
+ {
300
+ "epoch": 0.1,
301
+ "grad_norm": 0.8974116444587708,
302
+ "learning_rate": 0.00019868265225415265,
303
+ "loss": 1.0299,
304
+ "step": 41
305
+ },
306
+ {
307
+ "epoch": 0.1,
308
+ "grad_norm": 1.2469041347503662,
309
+ "learning_rate": 0.00019855451369114676,
310
+ "loss": 1.3814,
311
+ "step": 42
312
+ },
313
+ {
314
+ "epoch": 0.1,
315
+ "grad_norm": 0.8123692870140076,
316
+ "learning_rate": 0.0001984204741768395,
317
+ "loss": 1.1434,
318
+ "step": 43
319
+ },
320
+ {
321
+ "epoch": 0.1,
322
+ "grad_norm": 0.8598824143409729,
323
+ "learning_rate": 0.00019828054173684644,
324
+ "loss": 0.9911,
325
+ "step": 44
326
+ },
327
+ {
328
+ "epoch": 0.11,
329
+ "grad_norm": 1.1425665616989136,
330
+ "learning_rate": 0.00019813472474962217,
331
+ "loss": 1.4346,
332
+ "step": 45
333
+ },
334
+ {
335
+ "epoch": 0.11,
336
+ "grad_norm": 0.8056216835975647,
337
+ "learning_rate": 0.00019798303194595846,
338
+ "loss": 1.221,
339
+ "step": 46
340
+ },
341
+ {
342
+ "epoch": 0.11,
343
+ "grad_norm": 1.738286018371582,
344
+ "learning_rate": 0.00019782547240846166,
345
+ "loss": 1.5189,
346
+ "step": 47
347
+ },
348
+ {
349
+ "epoch": 0.11,
350
+ "grad_norm": 1.21600341796875,
351
+ "learning_rate": 0.00019766205557100868,
352
+ "loss": 1.0516,
353
+ "step": 48
354
+ },
355
+ {
356
+ "epoch": 0.12,
357
+ "grad_norm": 0.9683429598808289,
358
+ "learning_rate": 0.00019749279121818235,
359
+ "loss": 1.2459,
360
+ "step": 49
361
+ },
362
+ {
363
+ "epoch": 0.12,
364
+ "grad_norm": 1.1180105209350586,
365
+ "learning_rate": 0.00019731768948468549,
366
+ "loss": 1.5256,
367
+ "step": 50
368
+ },
369
+ {
370
+ "epoch": 0.12,
371
+ "grad_norm": 1.0404270887374878,
372
+ "learning_rate": 0.00019713676085473397,
373
+ "loss": 1.125,
374
+ "step": 51
375
+ },
376
+ {
377
+ "epoch": 0.12,
378
+ "grad_norm": 1.1551333665847778,
379
+ "learning_rate": 0.00019695001616142915,
380
+ "loss": 1.4839,
381
+ "step": 52
382
+ },
383
+ {
384
+ "epoch": 0.12,
385
+ "grad_norm": 1.0433716773986816,
386
+ "learning_rate": 0.00019675746658610917,
387
+ "loss": 1.3104,
388
+ "step": 53
389
+ },
390
+ {
391
+ "epoch": 0.13,
392
+ "grad_norm": 1.487308144569397,
393
+ "learning_rate": 0.0001965591236576794,
394
+ "loss": 1.1968,
395
+ "step": 54
396
+ },
397
+ {
398
+ "epoch": 0.13,
399
+ "grad_norm": 0.8388357758522034,
400
+ "learning_rate": 0.0001963549992519223,
401
+ "loss": 1.3537,
402
+ "step": 55
403
+ },
404
+ {
405
+ "epoch": 0.13,
406
+ "grad_norm": 1.3177778720855713,
407
+ "learning_rate": 0.00019614510559078625,
408
+ "loss": 1.3413,
409
+ "step": 56
410
+ },
411
+ {
412
+ "epoch": 0.13,
413
+ "grad_norm": 1.2819414138793945,
414
+ "learning_rate": 0.00019592945524165374,
415
+ "loss": 1.0727,
416
+ "step": 57
417
+ },
418
+ {
419
+ "epoch": 0.14,
420
+ "grad_norm": 1.0701121091842651,
421
+ "learning_rate": 0.00019570806111658898,
422
+ "loss": 1.5754,
423
+ "step": 58
424
+ },
425
+ {
426
+ "epoch": 0.14,
427
+ "grad_norm": 0.8484037518501282,
428
+ "learning_rate": 0.0001954809364715648,
429
+ "loss": 1.2085,
430
+ "step": 59
431
+ },
432
+ {
433
+ "epoch": 0.14,
434
+ "grad_norm": 1.3628385066986084,
435
+ "learning_rate": 0.00019524809490566877,
436
+ "loss": 1.2707,
437
+ "step": 60
438
+ },
439
+ {
440
+ "epoch": 0.14,
441
+ "grad_norm": 0.9246305823326111,
442
+ "learning_rate": 0.00019500955036028922,
443
+ "loss": 1.1952,
444
+ "step": 61
445
+ },
446
+ {
447
+ "epoch": 0.15,
448
+ "grad_norm": 0.927686870098114,
449
+ "learning_rate": 0.00019476531711828027,
450
+ "loss": 1.1894,
451
+ "step": 62
452
+ },
453
+ {
454
+ "epoch": 0.15,
455
+ "grad_norm": 0.9189894795417786,
456
+ "learning_rate": 0.00019451540980310676,
457
+ "loss": 1.0086,
458
+ "step": 63
459
+ },
460
+ {
461
+ "epoch": 0.15,
462
+ "grad_norm": 0.8920311331748962,
463
+ "learning_rate": 0.0001942598433779687,
464
+ "loss": 1.355,
465
+ "step": 64
466
+ },
467
+ {
468
+ "epoch": 0.15,
469
+ "grad_norm": 1.1602280139923096,
470
+ "learning_rate": 0.00019399863314490526,
471
+ "loss": 1.1758,
472
+ "step": 65
473
+ },
474
+ {
475
+ "epoch": 0.15,
476
+ "grad_norm": 0.9801963567733765,
477
+ "learning_rate": 0.00019373179474387858,
478
+ "loss": 1.211,
479
+ "step": 66
480
+ },
481
+ {
482
+ "epoch": 0.16,
483
+ "grad_norm": 0.9547926187515259,
484
+ "learning_rate": 0.0001934593441518374,
485
+ "loss": 1.0282,
486
+ "step": 67
487
+ },
488
+ {
489
+ "epoch": 0.16,
490
+ "grad_norm": 0.9999262690544128,
491
+ "learning_rate": 0.00019318129768176032,
492
+ "loss": 1.1899,
493
+ "step": 68
494
+ },
495
+ {
496
+ "epoch": 0.16,
497
+ "grad_norm": 0.9414704442024231,
498
+ "learning_rate": 0.00019289767198167916,
499
+ "loss": 1.1515,
500
+ "step": 69
501
+ },
502
+ {
503
+ "epoch": 0.16,
504
+ "grad_norm": 0.9921344518661499,
505
+ "learning_rate": 0.0001926084840336821,
506
+ "loss": 1.0423,
507
+ "step": 70
508
+ },
509
+ {
510
+ "epoch": 0.17,
511
+ "grad_norm": 0.8750163912773132,
512
+ "learning_rate": 0.00019231375115289696,
513
+ "loss": 0.6788,
514
+ "step": 71
515
+ },
516
+ {
517
+ "epoch": 0.17,
518
+ "grad_norm": 1.3444360494613647,
519
+ "learning_rate": 0.00019201349098645434,
520
+ "loss": 1.241,
521
+ "step": 72
522
+ },
523
+ {
524
+ "epoch": 0.17,
525
+ "grad_norm": 0.6930075883865356,
526
+ "learning_rate": 0.00019170772151243107,
527
+ "loss": 1.0857,
528
+ "step": 73
529
+ },
530
+ {
531
+ "epoch": 0.17,
532
+ "grad_norm": 0.9557262659072876,
533
+ "learning_rate": 0.0001913964610387738,
534
+ "loss": 1.4148,
535
+ "step": 74
536
+ },
537
+ {
538
+ "epoch": 0.18,
539
+ "grad_norm": 0.9436715245246887,
540
+ "learning_rate": 0.00019107972820220267,
541
+ "loss": 1.1334,
542
+ "step": 75
543
+ },
544
+ {
545
+ "epoch": 0.18,
546
+ "grad_norm": 1.321735143661499,
547
+ "learning_rate": 0.00019075754196709572,
548
+ "loss": 1.4699,
549
+ "step": 76
550
+ },
551
+ {
552
+ "epoch": 0.18,
553
+ "grad_norm": 0.824515163898468,
554
+ "learning_rate": 0.00019042992162435302,
555
+ "loss": 0.9975,
556
+ "step": 77
557
+ },
558
+ {
559
+ "epoch": 0.18,
560
+ "grad_norm": 1.0362204313278198,
561
+ "learning_rate": 0.0001900968867902419,
562
+ "loss": 1.146,
563
+ "step": 78
564
+ },
565
+ {
566
+ "epoch": 0.19,
567
+ "grad_norm": 0.9640101194381714,
568
+ "learning_rate": 0.00018975845740522244,
569
+ "loss": 1.3615,
570
+ "step": 79
571
+ },
572
+ {
573
+ "epoch": 0.19,
574
+ "grad_norm": 0.9828828573226929,
575
+ "learning_rate": 0.0001894146537327533,
576
+ "loss": 1.0824,
577
+ "step": 80
578
+ },
579
+ {
580
+ "epoch": 0.19,
581
+ "grad_norm": 1.0409202575683594,
582
+ "learning_rate": 0.00018906549635807861,
583
+ "loss": 1.4383,
584
+ "step": 81
585
+ },
586
+ {
587
+ "epoch": 0.19,
588
+ "grad_norm": 1.0056612491607666,
589
+ "learning_rate": 0.00018871100618699554,
590
+ "loss": 1.4139,
591
+ "step": 82
592
+ },
593
+ {
594
+ "epoch": 0.19,
595
+ "grad_norm": 0.9796703457832336,
596
+ "learning_rate": 0.0001883512044446023,
597
+ "loss": 1.6273,
598
+ "step": 83
599
+ },
600
+ {
601
+ "epoch": 0.2,
602
+ "grad_norm": 0.887485682964325,
603
+ "learning_rate": 0.00018798611267402746,
604
+ "loss": 1.3315,
605
+ "step": 84
606
+ },
607
+ {
608
+ "epoch": 0.2,
609
+ "grad_norm": 0.959770679473877,
610
+ "learning_rate": 0.00018761575273514003,
611
+ "loss": 1.3016,
612
+ "step": 85
613
+ },
614
+ {
615
+ "epoch": 0.2,
616
+ "grad_norm": 0.9511293172836304,
617
+ "learning_rate": 0.00018724014680324057,
618
+ "loss": 1.2996,
619
+ "step": 86
620
+ },
621
+ {
622
+ "epoch": 0.2,
623
+ "grad_norm": 0.9222784638404846,
624
+ "learning_rate": 0.0001868593173677335,
625
+ "loss": 1.1622,
626
+ "step": 87
627
+ },
628
+ {
629
+ "epoch": 0.21,
630
+ "grad_norm": 0.7448104619979858,
631
+ "learning_rate": 0.00018647328723078038,
632
+ "loss": 0.8966,
633
+ "step": 88
634
+ },
635
+ {
636
+ "epoch": 0.21,
637
+ "grad_norm": 1.2796077728271484,
638
+ "learning_rate": 0.000186082079505935,
639
+ "loss": 1.0091,
640
+ "step": 89
641
+ },
642
+ {
643
+ "epoch": 0.21,
644
+ "grad_norm": 0.784875750541687,
645
+ "learning_rate": 0.00018568571761675893,
646
+ "loss": 1.0961,
647
+ "step": 90
648
+ },
649
+ {
650
+ "epoch": 0.21,
651
+ "grad_norm": 0.8300296068191528,
652
+ "learning_rate": 0.00018528422529541952,
653
+ "loss": 0.7698,
654
+ "step": 91
655
+ },
656
+ {
657
+ "epoch": 0.22,
658
+ "grad_norm": 0.8383649587631226,
659
+ "learning_rate": 0.0001848776265812687,
660
+ "loss": 1.0525,
661
+ "step": 92
662
+ },
663
+ {
664
+ "epoch": 0.22,
665
+ "grad_norm": 0.9814484119415283,
666
+ "learning_rate": 0.0001844659458194036,
667
+ "loss": 1.2679,
668
+ "step": 93
669
+ },
670
+ {
671
+ "epoch": 0.22,
672
+ "grad_norm": 0.8606963753700256,
673
+ "learning_rate": 0.00018404920765920896,
674
+ "loss": 1.4763,
675
+ "step": 94
676
+ },
677
+ {
678
+ "epoch": 0.22,
679
+ "grad_norm": 1.2019658088684082,
680
+ "learning_rate": 0.00018362743705288125,
681
+ "loss": 1.376,
682
+ "step": 95
683
+ },
684
+ {
685
+ "epoch": 0.23,
686
+ "grad_norm": 0.8579021692276001,
687
+ "learning_rate": 0.00018320065925393468,
688
+ "loss": 1.3184,
689
+ "step": 96
690
+ },
691
+ {
692
+ "epoch": 0.23,
693
+ "grad_norm": 1.0116324424743652,
694
+ "learning_rate": 0.00018276889981568906,
695
+ "loss": 0.939,
696
+ "step": 97
697
+ },
698
+ {
699
+ "epoch": 0.23,
700
+ "grad_norm": 0.9020506143569946,
701
+ "learning_rate": 0.00018233218458973984,
702
+ "loss": 1.3675,
703
+ "step": 98
704
+ },
705
+ {
706
+ "epoch": 0.23,
707
+ "grad_norm": 0.7909868955612183,
708
+ "learning_rate": 0.00018189053972441025,
709
+ "loss": 1.0777,
710
+ "step": 99
711
+ },
712
+ {
713
+ "epoch": 0.23,
714
+ "grad_norm": 0.9654402136802673,
715
+ "learning_rate": 0.00018144399166318572,
716
+ "loss": 1.2697,
717
+ "step": 100
718
+ },
719
+ {
720
+ "epoch": 0.24,
721
+ "grad_norm": 0.9301828742027283,
722
+ "learning_rate": 0.0001809925671431304,
723
+ "loss": 1.3651,
724
+ "step": 101
725
+ },
726
+ {
727
+ "epoch": 0.24,
728
+ "grad_norm": 0.8000981211662292,
729
+ "learning_rate": 0.00018053629319328662,
730
+ "loss": 0.9806,
731
+ "step": 102
732
+ },
733
+ {
734
+ "epoch": 0.24,
735
+ "grad_norm": 0.7500132918357849,
736
+ "learning_rate": 0.00018007519713305605,
737
+ "loss": 1.0699,
738
+ "step": 103
739
+ },
740
+ {
741
+ "epoch": 0.24,
742
+ "grad_norm": 1.0019265413284302,
743
+ "learning_rate": 0.00017960930657056438,
744
+ "loss": 1.2713,
745
+ "step": 104
746
+ },
747
+ {
748
+ "epoch": 0.25,
749
+ "grad_norm": 0.8942422270774841,
750
+ "learning_rate": 0.00017913864940100808,
751
+ "loss": 0.9149,
752
+ "step": 105
753
+ },
754
+ {
755
+ "epoch": 0.25,
756
+ "grad_norm": 0.7721758484840393,
757
+ "learning_rate": 0.00017866325380498416,
758
+ "loss": 1.0378,
759
+ "step": 106
760
+ },
761
+ {
762
+ "epoch": 0.25,
763
+ "grad_norm": 0.9199188947677612,
764
+ "learning_rate": 0.000178183148246803,
765
+ "loss": 1.3192,
766
+ "step": 107
767
+ },
768
+ {
769
+ "epoch": 0.25,
770
+ "eval_loss": 1.1375699043273926,
771
+ "eval_runtime": 5.9115,
772
+ "eval_samples_per_second": 16.916,
773
+ "eval_steps_per_second": 16.916,
774
+ "step": 107
775
+ },
776
+ {
777
+ "epoch": 0.25,
778
+ "grad_norm": 0.7156293988227844,
779
+ "learning_rate": 0.0001776983614727838,
780
+ "loss": 0.854,
781
+ "step": 108
782
+ },
783
+ {
784
+ "epoch": 0.26,
785
+ "grad_norm": 1.0115437507629395,
786
+ "learning_rate": 0.00017720892250953373,
787
+ "loss": 0.9842,
788
+ "step": 109
789
+ },
790
+ {
791
+ "epoch": 0.26,
792
+ "grad_norm": 1.112038493156433,
793
+ "learning_rate": 0.00017671486066220965,
794
+ "loss": 1.0456,
795
+ "step": 110
796
+ },
797
+ {
798
+ "epoch": 0.26,
799
+ "grad_norm": 0.9239675402641296,
800
+ "learning_rate": 0.00017621620551276366,
801
+ "loss": 1.1104,
802
+ "step": 111
803
+ },
804
+ {
805
+ "epoch": 0.26,
806
+ "grad_norm": 0.8196420073509216,
807
+ "learning_rate": 0.00017571298691817177,
808
+ "loss": 0.7432,
809
+ "step": 112
810
+ },
811
+ {
812
+ "epoch": 0.27,
813
+ "grad_norm": 0.7901629209518433,
814
+ "learning_rate": 0.00017520523500864625,
815
+ "loss": 0.8345,
816
+ "step": 113
817
+ },
818
+ {
819
+ "epoch": 0.27,
820
+ "grad_norm": 1.0044571161270142,
821
+ "learning_rate": 0.0001746929801858317,
822
+ "loss": 1.1328,
823
+ "step": 114
824
+ },
825
+ {
826
+ "epoch": 0.27,
827
+ "grad_norm": 1.10731041431427,
828
+ "learning_rate": 0.00017417625312098452,
829
+ "loss": 1.1772,
830
+ "step": 115
831
+ },
832
+ {
833
+ "epoch": 0.27,
834
+ "grad_norm": 1.0169037580490112,
835
+ "learning_rate": 0.0001736550847531366,
836
+ "loss": 1.0579,
837
+ "step": 116
838
+ },
839
+ {
840
+ "epoch": 0.27,
841
+ "grad_norm": 0.8641659021377563,
842
+ "learning_rate": 0.00017312950628724295,
843
+ "loss": 0.8655,
844
+ "step": 117
845
+ },
846
+ {
847
+ "epoch": 0.28,
848
+ "grad_norm": 1.0836398601531982,
849
+ "learning_rate": 0.0001725995491923131,
850
+ "loss": 1.1701,
851
+ "step": 118
852
+ },
853
+ {
854
+ "epoch": 0.28,
855
+ "grad_norm": 1.0795177221298218,
856
+ "learning_rate": 0.00017206524519952697,
857
+ "loss": 1.3704,
858
+ "step": 119
859
+ },
860
+ {
861
+ "epoch": 0.28,
862
+ "grad_norm": 0.7752137184143066,
863
+ "learning_rate": 0.00017152662630033505,
864
+ "loss": 0.9499,
865
+ "step": 120
866
+ },
867
+ {
868
+ "epoch": 0.28,
869
+ "grad_norm": 0.8980644345283508,
870
+ "learning_rate": 0.00017098372474454277,
871
+ "loss": 0.915,
872
+ "step": 121
873
+ },
874
+ {
875
+ "epoch": 0.29,
876
+ "grad_norm": 0.7363729476928711,
877
+ "learning_rate": 0.00017043657303837963,
878
+ "loss": 0.9518,
879
+ "step": 122
880
+ },
881
+ {
882
+ "epoch": 0.29,
883
+ "grad_norm": 0.9289439916610718,
884
+ "learning_rate": 0.000169885203942553,
885
+ "loss": 1.1618,
886
+ "step": 123
887
+ },
888
+ {
889
+ "epoch": 0.29,
890
+ "grad_norm": 0.785468578338623,
891
+ "learning_rate": 0.0001693296504702862,
892
+ "loss": 1.1148,
893
+ "step": 124
894
+ },
895
+ {
896
+ "epoch": 0.29,
897
+ "grad_norm": 0.9278959035873413,
898
+ "learning_rate": 0.00016876994588534234,
899
+ "loss": 1.1782,
900
+ "step": 125
901
+ },
902
+ {
903
+ "epoch": 0.3,
904
+ "grad_norm": 0.8510153889656067,
905
+ "learning_rate": 0.00016820612370003221,
906
+ "loss": 1.1296,
907
+ "step": 126
908
+ },
909
+ {
910
+ "epoch": 0.3,
911
+ "grad_norm": 0.9500170350074768,
912
+ "learning_rate": 0.000167638217673208,
913
+ "loss": 0.8094,
914
+ "step": 127
915
+ },
916
+ {
917
+ "epoch": 0.3,
918
+ "grad_norm": 0.8451842665672302,
919
+ "learning_rate": 0.00016706626180824186,
920
+ "loss": 1.0722,
921
+ "step": 128
922
+ },
923
+ {
924
+ "epoch": 0.3,
925
+ "grad_norm": 0.6593143939971924,
926
+ "learning_rate": 0.00016649029035099,
927
+ "loss": 0.5532,
928
+ "step": 129
929
+ },
930
+ {
931
+ "epoch": 0.31,
932
+ "grad_norm": 1.0398962497711182,
933
+ "learning_rate": 0.0001659103377877423,
934
+ "loss": 0.8373,
935
+ "step": 130
936
+ },
937
+ {
938
+ "epoch": 0.31,
939
+ "grad_norm": 0.8942617774009705,
940
+ "learning_rate": 0.0001653264388431572,
941
+ "loss": 1.0992,
942
+ "step": 131
943
+ },
944
+ {
945
+ "epoch": 0.31,
946
+ "grad_norm": 0.7003504633903503,
947
+ "learning_rate": 0.00016473862847818277,
948
+ "loss": 0.699,
949
+ "step": 132
950
+ },
951
+ {
952
+ "epoch": 0.31,
953
+ "grad_norm": 0.7734900116920471,
954
+ "learning_rate": 0.00016414694188796345,
955
+ "loss": 0.9312,
956
+ "step": 133
957
+ },
958
+ {
959
+ "epoch": 0.31,
960
+ "grad_norm": 1.0958092212677002,
961
+ "learning_rate": 0.00016355141449973256,
962
+ "loss": 1.1076,
963
+ "step": 134
964
+ },
965
+ {
966
+ "epoch": 0.32,
967
+ "grad_norm": 0.8197342157363892,
968
+ "learning_rate": 0.0001629520819706912,
969
+ "loss": 1.1782,
970
+ "step": 135
971
+ },
972
+ {
973
+ "epoch": 0.32,
974
+ "grad_norm": 0.7238759994506836,
975
+ "learning_rate": 0.00016234898018587337,
976
+ "loss": 1.106,
977
+ "step": 136
978
+ },
979
+ {
980
+ "epoch": 0.32,
981
+ "grad_norm": 0.9790141582489014,
982
+ "learning_rate": 0.0001617421452559971,
983
+ "loss": 1.0013,
984
+ "step": 137
985
+ },
986
+ {
987
+ "epoch": 0.32,
988
+ "grad_norm": 0.8993759155273438,
989
+ "learning_rate": 0.0001611316135153026,
990
+ "loss": 1.0726,
991
+ "step": 138
992
+ },
993
+ {
994
+ "epoch": 0.33,
995
+ "grad_norm": 0.7299768328666687,
996
+ "learning_rate": 0.00016051742151937655,
997
+ "loss": 0.9232,
998
+ "step": 139
999
+ },
1000
+ {
1001
+ "epoch": 0.33,
1002
+ "grad_norm": 1.0451937913894653,
1003
+ "learning_rate": 0.0001598996060429634,
1004
+ "loss": 1.2411,
1005
+ "step": 140
1006
+ },
1007
+ {
1008
+ "epoch": 0.33,
1009
+ "grad_norm": 0.9933896064758301,
1010
+ "learning_rate": 0.00015927820407776353,
1011
+ "loss": 0.6061,
1012
+ "step": 141
1013
+ },
1014
+ {
1015
+ "epoch": 0.33,
1016
+ "grad_norm": 0.6830273866653442,
1017
+ "learning_rate": 0.0001586532528302183,
1018
+ "loss": 0.9581,
1019
+ "step": 142
1020
+ },
1021
+ {
1022
+ "epoch": 0.34,
1023
+ "grad_norm": 0.7997300624847412,
1024
+ "learning_rate": 0.00015802478971928242,
1025
+ "loss": 0.7054,
1026
+ "step": 143
1027
+ },
1028
+ {
1029
+ "epoch": 0.34,
1030
+ "grad_norm": 0.8350067138671875,
1031
+ "learning_rate": 0.0001573928523741832,
1032
+ "loss": 1.1718,
1033
+ "step": 144
1034
+ },
1035
+ {
1036
+ "epoch": 0.34,
1037
+ "grad_norm": 0.9091382026672363,
1038
+ "learning_rate": 0.00015675747863216801,
1039
+ "loss": 1.0653,
1040
+ "step": 145
1041
+ },
1042
+ {
1043
+ "epoch": 0.34,
1044
+ "grad_norm": 0.9306054711341858,
1045
+ "learning_rate": 0.00015611870653623825,
1046
+ "loss": 1.3816,
1047
+ "step": 146
1048
+ },
1049
+ {
1050
+ "epoch": 0.35,
1051
+ "grad_norm": 1.1074793338775635,
1052
+ "learning_rate": 0.00015547657433287183,
1053
+ "loss": 1.4123,
1054
+ "step": 147
1055
+ },
1056
+ {
1057
+ "epoch": 0.35,
1058
+ "grad_norm": 0.8237017393112183,
1059
+ "learning_rate": 0.0001548311204697331,
1060
+ "loss": 1.1059,
1061
+ "step": 148
1062
+ },
1063
+ {
1064
+ "epoch": 0.35,
1065
+ "grad_norm": 1.4395543336868286,
1066
+ "learning_rate": 0.00015418238359337077,
1067
+ "loss": 1.2508,
1068
+ "step": 149
1069
+ },
1070
+ {
1071
+ "epoch": 0.35,
1072
+ "grad_norm": 1.2550331354141235,
1073
+ "learning_rate": 0.00015353040254690393,
1074
+ "loss": 1.8478,
1075
+ "step": 150
1076
+ },
1077
+ {
1078
+ "epoch": 0.35,
1079
+ "grad_norm": 0.7537391781806946,
1080
+ "learning_rate": 0.0001528752163676964,
1081
+ "loss": 0.8779,
1082
+ "step": 151
1083
+ },
1084
+ {
1085
+ "epoch": 0.36,
1086
+ "grad_norm": 1.1749041080474854,
1087
+ "learning_rate": 0.00015221686428501928,
1088
+ "loss": 1.1649,
1089
+ "step": 152
1090
+ },
1091
+ {
1092
+ "epoch": 0.36,
1093
+ "grad_norm": 0.9532057046890259,
1094
+ "learning_rate": 0.00015155538571770218,
1095
+ "loss": 1.1493,
1096
+ "step": 153
1097
+ },
1098
+ {
1099
+ "epoch": 0.36,
1100
+ "grad_norm": 1.0025173425674438,
1101
+ "learning_rate": 0.0001508908202717729,
1102
+ "loss": 1.2429,
1103
+ "step": 154
1104
+ },
1105
+ {
1106
+ "epoch": 0.36,
1107
+ "grad_norm": 0.9108004570007324,
1108
+ "learning_rate": 0.00015022320773808612,
1109
+ "loss": 1.5981,
1110
+ "step": 155
1111
+ },
1112
+ {
1113
+ "epoch": 0.37,
1114
+ "grad_norm": 0.9072840809822083,
1115
+ "learning_rate": 0.00014955258808994096,
1116
+ "loss": 1.2374,
1117
+ "step": 156
1118
+ },
1119
+ {
1120
+ "epoch": 0.37,
1121
+ "grad_norm": 0.8806712031364441,
1122
+ "learning_rate": 0.00014887900148068735,
1123
+ "loss": 1.1942,
1124
+ "step": 157
1125
+ },
1126
+ {
1127
+ "epoch": 0.37,
1128
+ "grad_norm": 1.1253175735473633,
1129
+ "learning_rate": 0.0001482024882413222,
1130
+ "loss": 1.1235,
1131
+ "step": 158
1132
+ },
1133
+ {
1134
+ "epoch": 0.37,
1135
+ "grad_norm": 0.8281373977661133,
1136
+ "learning_rate": 0.00014752308887807427,
1137
+ "loss": 1.1125,
1138
+ "step": 159
1139
+ },
1140
+ {
1141
+ "epoch": 0.38,
1142
+ "grad_norm": 0.9860203862190247,
1143
+ "learning_rate": 0.00014684084406997903,
1144
+ "loss": 1.12,
1145
+ "step": 160
1146
+ },
1147
+ {
1148
+ "epoch": 0.38,
1149
+ "grad_norm": 0.9580567479133606,
1150
+ "learning_rate": 0.00014615579466644292,
1151
+ "loss": 1.1726,
1152
+ "step": 161
1153
+ },
1154
+ {
1155
+ "epoch": 0.38,
1156
+ "grad_norm": 0.704282820224762,
1157
+ "learning_rate": 0.00014546798168479756,
1158
+ "loss": 1.1812,
1159
+ "step": 162
1160
+ },
1161
+ {
1162
+ "epoch": 0.38,
1163
+ "grad_norm": 0.8092489838600159,
1164
+ "learning_rate": 0.00014477744630784378,
1165
+ "loss": 0.9743,
1166
+ "step": 163
1167
+ },
1168
+ {
1169
+ "epoch": 0.38,
1170
+ "grad_norm": 0.875648021697998,
1171
+ "learning_rate": 0.00014408422988138584,
1172
+ "loss": 1.019,
1173
+ "step": 164
1174
+ },
1175
+ {
1176
+ "epoch": 0.39,
1177
+ "grad_norm": 0.8234091401100159,
1178
+ "learning_rate": 0.00014338837391175582,
1179
+ "loss": 1.2794,
1180
+ "step": 165
1181
+ },
1182
+ {
1183
+ "epoch": 0.39,
1184
+ "grad_norm": 0.8568207025527954,
1185
+ "learning_rate": 0.00014268992006332846,
1186
+ "loss": 1.1104,
1187
+ "step": 166
1188
+ },
1189
+ {
1190
+ "epoch": 0.39,
1191
+ "grad_norm": 0.9901588559150696,
1192
+ "learning_rate": 0.00014198891015602646,
1193
+ "loss": 1.1187,
1194
+ "step": 167
1195
+ },
1196
+ {
1197
+ "epoch": 0.39,
1198
+ "grad_norm": 1.0761953592300415,
1199
+ "learning_rate": 0.0001412853861628166,
1200
+ "loss": 1.5204,
1201
+ "step": 168
1202
+ },
1203
+ {
1204
+ "epoch": 0.4,
1205
+ "grad_norm": 0.8662331104278564,
1206
+ "learning_rate": 0.0001405793902071964,
1207
+ "loss": 1.2189,
1208
+ "step": 169
1209
+ },
1210
+ {
1211
+ "epoch": 0.4,
1212
+ "grad_norm": 0.8049296736717224,
1213
+ "learning_rate": 0.00013987096456067236,
1214
+ "loss": 1.3537,
1215
+ "step": 170
1216
+ },
1217
+ {
1218
+ "epoch": 0.4,
1219
+ "grad_norm": 0.8470759391784668,
1220
+ "learning_rate": 0.00013916015164022852,
1221
+ "loss": 1.0526,
1222
+ "step": 171
1223
+ },
1224
+ {
1225
+ "epoch": 0.4,
1226
+ "grad_norm": 0.9302281737327576,
1227
+ "learning_rate": 0.00013844699400578696,
1228
+ "loss": 0.9932,
1229
+ "step": 172
1230
+ },
1231
+ {
1232
+ "epoch": 0.41,
1233
+ "grad_norm": 1.1691091060638428,
1234
+ "learning_rate": 0.00013773153435765964,
1235
+ "loss": 1.1426,
1236
+ "step": 173
1237
+ },
1238
+ {
1239
+ "epoch": 0.41,
1240
+ "grad_norm": 0.9937382340431213,
1241
+ "learning_rate": 0.00013701381553399145,
1242
+ "loss": 1.326,
1243
+ "step": 174
1244
+ },
1245
+ {
1246
+ "epoch": 0.41,
1247
+ "grad_norm": 0.8379032015800476,
1248
+ "learning_rate": 0.00013629388050819547,
1249
+ "loss": 1.0955,
1250
+ "step": 175
1251
+ },
1252
+ {
1253
+ "epoch": 0.41,
1254
+ "grad_norm": 0.9588186144828796,
1255
+ "learning_rate": 0.00013557177238637986,
1256
+ "loss": 1.2328,
1257
+ "step": 176
1258
+ },
1259
+ {
1260
+ "epoch": 0.42,
1261
+ "grad_norm": 0.7939326167106628,
1262
+ "learning_rate": 0.00013484753440476692,
1263
+ "loss": 1.1649,
1264
+ "step": 177
1265
+ },
1266
+ {
1267
+ "epoch": 0.42,
1268
+ "grad_norm": 0.8485954403877258,
1269
+ "learning_rate": 0.00013412120992710425,
1270
+ "loss": 1.034,
1271
+ "step": 178
1272
+ },
1273
+ {
1274
+ "epoch": 0.42,
1275
+ "grad_norm": 0.923659086227417,
1276
+ "learning_rate": 0.00013339284244206847,
1277
+ "loss": 1.245,
1278
+ "step": 179
1279
+ },
1280
+ {
1281
+ "epoch": 0.42,
1282
+ "grad_norm": 0.9998313784599304,
1283
+ "learning_rate": 0.00013266247556066122,
1284
+ "loss": 1.4282,
1285
+ "step": 180
1286
+ },
1287
+ {
1288
+ "epoch": 0.42,
1289
+ "grad_norm": 1.1997437477111816,
1290
+ "learning_rate": 0.000131930153013598,
1291
+ "loss": 1.3608,
1292
+ "step": 181
1293
+ },
1294
+ {
1295
+ "epoch": 0.43,
1296
+ "grad_norm": 1.101218581199646,
1297
+ "learning_rate": 0.0001311959186486898,
1298
+ "loss": 1.3061,
1299
+ "step": 182
1300
+ },
1301
+ {
1302
+ "epoch": 0.43,
1303
+ "grad_norm": 1.1238192319869995,
1304
+ "learning_rate": 0.0001304598164282176,
1305
+ "loss": 1.1255,
1306
+ "step": 183
1307
+ },
1308
+ {
1309
+ "epoch": 0.43,
1310
+ "grad_norm": 1.3340548276901245,
1311
+ "learning_rate": 0.00012972189042630044,
1312
+ "loss": 1.5872,
1313
+ "step": 184
1314
+ },
1315
+ {
1316
+ "epoch": 0.43,
1317
+ "grad_norm": 0.8796036839485168,
1318
+ "learning_rate": 0.00012898218482625606,
1319
+ "loss": 1.1176,
1320
+ "step": 185
1321
+ },
1322
+ {
1323
+ "epoch": 0.44,
1324
+ "grad_norm": 0.9776093363761902,
1325
+ "learning_rate": 0.0001282407439179557,
1326
+ "loss": 1.1615,
1327
+ "step": 186
1328
+ },
1329
+ {
1330
+ "epoch": 0.44,
1331
+ "grad_norm": 1.3886358737945557,
1332
+ "learning_rate": 0.0001274976120951723,
1333
+ "loss": 1.0288,
1334
+ "step": 187
1335
+ },
1336
+ {
1337
+ "epoch": 0.44,
1338
+ "grad_norm": 0.9199085235595703,
1339
+ "learning_rate": 0.00012675283385292212,
1340
+ "loss": 0.9302,
1341
+ "step": 188
1342
+ },
1343
+ {
1344
+ "epoch": 0.44,
1345
+ "grad_norm": 1.1090227365493774,
1346
+ "learning_rate": 0.00012600645378480082,
1347
+ "loss": 1.214,
1348
+ "step": 189
1349
+ },
1350
+ {
1351
+ "epoch": 0.45,
1352
+ "grad_norm": 0.8784577250480652,
1353
+ "learning_rate": 0.00012525851658031352,
1354
+ "loss": 0.8802,
1355
+ "step": 190
1356
+ },
1357
+ {
1358
+ "epoch": 0.45,
1359
+ "grad_norm": 0.860397219657898,
1360
+ "learning_rate": 0.0001245090670221987,
1361
+ "loss": 1.0587,
1362
+ "step": 191
1363
+ },
1364
+ {
1365
+ "epoch": 0.45,
1366
+ "grad_norm": 1.014009952545166,
1367
+ "learning_rate": 0.00012375814998374712,
1368
+ "loss": 1.2103,
1369
+ "step": 192
1370
+ },
1371
+ {
1372
+ "epoch": 0.45,
1373
+ "grad_norm": 0.6379606127738953,
1374
+ "learning_rate": 0.00012300581042611492,
1375
+ "loss": 0.4779,
1376
+ "step": 193
1377
+ },
1378
+ {
1379
+ "epoch": 0.46,
1380
+ "grad_norm": 0.8548885583877563,
1381
+ "learning_rate": 0.00012225209339563145,
1382
+ "loss": 1.2684,
1383
+ "step": 194
1384
+ },
1385
+ {
1386
+ "epoch": 0.46,
1387
+ "grad_norm": 0.5329421162605286,
1388
+ "learning_rate": 0.00012149704402110243,
1389
+ "loss": 0.5498,
1390
+ "step": 195
1391
+ },
1392
+ {
1393
+ "epoch": 0.46,
1394
+ "grad_norm": 0.8345950245857239,
1395
+ "learning_rate": 0.00012074070751110751,
1396
+ "loss": 1.2684,
1397
+ "step": 196
1398
+ },
1399
+ {
1400
+ "epoch": 0.46,
1401
+ "grad_norm": 0.8862521052360535,
1402
+ "learning_rate": 0.00011998312915129371,
1403
+ "loss": 1.1864,
1404
+ "step": 197
1405
+ },
1406
+ {
1407
+ "epoch": 0.46,
1408
+ "grad_norm": 0.9100441336631775,
1409
+ "learning_rate": 0.0001192243543016637,
1410
+ "loss": 1.2897,
1411
+ "step": 198
1412
+ },
1413
+ {
1414
+ "epoch": 0.47,
1415
+ "grad_norm": 1.1576776504516602,
1416
+ "learning_rate": 0.00011846442839386003,
1417
+ "loss": 1.3155,
1418
+ "step": 199
1419
+ },
1420
+ {
1421
+ "epoch": 0.47,
1422
+ "grad_norm": 1.105965495109558,
1423
+ "learning_rate": 0.00011770339692844483,
1424
+ "loss": 0.9913,
1425
+ "step": 200
1426
+ },
1427
+ {
1428
+ "epoch": 0.47,
1429
+ "grad_norm": 0.696582019329071,
1430
+ "learning_rate": 0.00011694130547217554,
1431
+ "loss": 0.9719,
1432
+ "step": 201
1433
+ },
1434
+ {
1435
+ "epoch": 0.47,
1436
+ "grad_norm": 0.9100035429000854,
1437
+ "learning_rate": 0.0001161781996552765,
1438
+ "loss": 1.0476,
1439
+ "step": 202
1440
+ },
1441
+ {
1442
+ "epoch": 0.48,
1443
+ "grad_norm": 0.9169908165931702,
1444
+ "learning_rate": 0.00011541412516870684,
1445
+ "loss": 0.909,
1446
+ "step": 203
1447
+ },
1448
+ {
1449
+ "epoch": 0.48,
1450
+ "grad_norm": 0.8421806693077087,
1451
+ "learning_rate": 0.00011464912776142494,
1452
+ "loss": 1.1626,
1453
+ "step": 204
1454
+ },
1455
+ {
1456
+ "epoch": 0.48,
1457
+ "grad_norm": 0.8510706424713135,
1458
+ "learning_rate": 0.00011388325323764888,
1459
+ "loss": 1.4659,
1460
+ "step": 205
1461
+ },
1462
+ {
1463
+ "epoch": 0.48,
1464
+ "grad_norm": 1.1053885221481323,
1465
+ "learning_rate": 0.00011311654745411425,
1466
+ "loss": 1.4878,
1467
+ "step": 206
1468
+ },
1469
+ {
1470
+ "epoch": 0.49,
1471
+ "grad_norm": 0.933678388595581,
1472
+ "learning_rate": 0.00011234905631732819,
1473
+ "loss": 0.9186,
1474
+ "step": 207
1475
+ },
1476
+ {
1477
+ "epoch": 0.49,
1478
+ "grad_norm": 0.8687578439712524,
1479
+ "learning_rate": 0.00011158082578082089,
1480
+ "loss": 0.7517,
1481
+ "step": 208
1482
+ },
1483
+ {
1484
+ "epoch": 0.49,
1485
+ "grad_norm": 0.8849249482154846,
1486
+ "learning_rate": 0.00011081190184239419,
1487
+ "loss": 1.0061,
1488
+ "step": 209
1489
+ },
1490
+ {
1491
+ "epoch": 0.49,
1492
+ "grad_norm": 0.8970105648040771,
1493
+ "learning_rate": 0.00011004233054136725,
1494
+ "loss": 1.0486,
1495
+ "step": 210
1496
+ },
1497
+ {
1498
+ "epoch": 0.5,
1499
+ "grad_norm": 1.0767931938171387,
1500
+ "learning_rate": 0.00010927215795582012,
1501
+ "loss": 1.2699,
1502
+ "step": 211
1503
+ },
1504
+ {
1505
+ "epoch": 0.5,
1506
+ "grad_norm": 0.9513278007507324,
1507
+ "learning_rate": 0.00010850143019983474,
1508
+ "loss": 1.1946,
1509
+ "step": 212
1510
+ },
1511
+ {
1512
+ "epoch": 0.5,
1513
+ "grad_norm": 1.0254892110824585,
1514
+ "learning_rate": 0.0001077301934207339,
1515
+ "loss": 1.4022,
1516
+ "step": 213
1517
+ },
1518
+ {
1519
+ "epoch": 0.5,
1520
+ "grad_norm": 0.7836260199546814,
1521
+ "learning_rate": 0.00010695849379631813,
1522
+ "loss": 1.3095,
1523
+ "step": 214
1524
+ },
1525
+ {
1526
+ "epoch": 0.5,
1527
+ "eval_loss": 1.1224589347839355,
1528
+ "eval_runtime": 5.9458,
1529
+ "eval_samples_per_second": 16.819,
1530
+ "eval_steps_per_second": 16.819,
1531
+ "step": 214
1532
+ },
1533
+ {
1534
+ "epoch": 0.5,
1535
+ "grad_norm": 0.9416055679321289,
1536
+ "learning_rate": 0.00010618637753210085,
1537
+ "loss": 1.0996,
1538
+ "step": 215
1539
+ },
1540
+ {
1541
+ "epoch": 0.51,
1542
+ "grad_norm": 0.8331229090690613,
1543
+ "learning_rate": 0.00010541389085854176,
1544
+ "loss": 0.9657,
1545
+ "step": 216
1546
+ },
1547
+ {
1548
+ "epoch": 0.51,
1549
+ "grad_norm": 0.7178126573562622,
1550
+ "learning_rate": 0.00010464108002827882,
1551
+ "loss": 0.7778,
1552
+ "step": 217
1553
+ },
1554
+ {
1555
+ "epoch": 0.51,
1556
+ "grad_norm": 1.8502916097640991,
1557
+ "learning_rate": 0.00010386799131335889,
1558
+ "loss": 1.4472,
1559
+ "step": 218
1560
+ },
1561
+ {
1562
+ "epoch": 0.51,
1563
+ "grad_norm": 0.8758464455604553,
1564
+ "learning_rate": 0.00010309467100246713,
1565
+ "loss": 1.1188,
1566
+ "step": 219
1567
+ },
1568
+ {
1569
+ "epoch": 0.52,
1570
+ "grad_norm": 0.8960602879524231,
1571
+ "learning_rate": 0.00010232116539815558,
1572
+ "loss": 1.1453,
1573
+ "step": 220
1574
+ },
1575
+ {
1576
+ "epoch": 0.52,
1577
+ "grad_norm": 0.8716662526130676,
1578
+ "learning_rate": 0.00010154752081407066,
1579
+ "loss": 1.1906,
1580
+ "step": 221
1581
+ },
1582
+ {
1583
+ "epoch": 0.52,
1584
+ "grad_norm": 1.0372726917266846,
1585
+ "learning_rate": 0.00010077378357218021,
1586
+ "loss": 1.2253,
1587
+ "step": 222
1588
+ },
1589
+ {
1590
+ "epoch": 0.52,
1591
+ "grad_norm": 1.2147200107574463,
1592
+ "learning_rate": 0.0001,
1593
+ "loss": 1.1959,
1594
+ "step": 223
1595
+ },
1596
+ {
1597
+ "epoch": 0.53,
1598
+ "grad_norm": 1.0374308824539185,
1599
+ "learning_rate": 9.92262164278198e-05,
1600
+ "loss": 1.5055,
1601
+ "step": 224
1602
+ },
1603
+ {
1604
+ "epoch": 0.53,
1605
+ "grad_norm": 1.026852011680603,
1606
+ "learning_rate": 9.845247918592937e-05,
1607
+ "loss": 1.3998,
1608
+ "step": 225
1609
+ },
1610
+ {
1611
+ "epoch": 0.53,
1612
+ "grad_norm": 1.0135536193847656,
1613
+ "learning_rate": 9.767883460184443e-05,
1614
+ "loss": 0.8342,
1615
+ "step": 226
1616
+ },
1617
+ {
1618
+ "epoch": 0.53,
1619
+ "grad_norm": 0.6277703642845154,
1620
+ "learning_rate": 9.69053289975329e-05,
1621
+ "loss": 0.6104,
1622
+ "step": 227
1623
+ },
1624
+ {
1625
+ "epoch": 0.54,
1626
+ "grad_norm": 0.8149964213371277,
1627
+ "learning_rate": 9.613200868664112e-05,
1628
+ "loss": 1.1625,
1629
+ "step": 228
1630
+ },
1631
+ {
1632
+ "epoch": 0.54,
1633
+ "grad_norm": 1.0715911388397217,
1634
+ "learning_rate": 9.53589199717212e-05,
1635
+ "loss": 1.0454,
1636
+ "step": 229
1637
+ },
1638
+ {
1639
+ "epoch": 0.54,
1640
+ "grad_norm": 0.9631131887435913,
1641
+ "learning_rate": 9.458610914145826e-05,
1642
+ "loss": 1.0737,
1643
+ "step": 230
1644
+ },
1645
+ {
1646
+ "epoch": 0.54,
1647
+ "grad_norm": 0.8216869235038757,
1648
+ "learning_rate": 9.381362246789917e-05,
1649
+ "loss": 1.0467,
1650
+ "step": 231
1651
+ },
1652
+ {
1653
+ "epoch": 0.54,
1654
+ "grad_norm": 1.1427477598190308,
1655
+ "learning_rate": 9.304150620368188e-05,
1656
+ "loss": 1.2994,
1657
+ "step": 232
1658
+ },
1659
+ {
1660
+ "epoch": 0.55,
1661
+ "grad_norm": 0.8490440845489502,
1662
+ "learning_rate": 9.226980657926614e-05,
1663
+ "loss": 1.0697,
1664
+ "step": 233
1665
+ },
1666
+ {
1667
+ "epoch": 0.55,
1668
+ "grad_norm": 1.339479684829712,
1669
+ "learning_rate": 9.149856980016529e-05,
1670
+ "loss": 1.6431,
1671
+ "step": 234
1672
+ },
1673
+ {
1674
+ "epoch": 0.55,
1675
+ "grad_norm": 0.7970044016838074,
1676
+ "learning_rate": 9.072784204417995e-05,
1677
+ "loss": 1.0936,
1678
+ "step": 235
1679
+ },
1680
+ {
1681
+ "epoch": 0.55,
1682
+ "grad_norm": 1.135980486869812,
1683
+ "learning_rate": 8.995766945863277e-05,
1684
+ "loss": 1.2341,
1685
+ "step": 236
1686
+ },
1687
+ {
1688
+ "epoch": 0.56,
1689
+ "grad_norm": 0.663905680179596,
1690
+ "learning_rate": 8.918809815760585e-05,
1691
+ "loss": 0.8369,
1692
+ "step": 237
1693
+ },
1694
+ {
1695
+ "epoch": 0.56,
1696
+ "grad_norm": 0.7740184664726257,
1697
+ "learning_rate": 8.841917421917912e-05,
1698
+ "loss": 1.0605,
1699
+ "step": 238
1700
+ },
1701
+ {
1702
+ "epoch": 0.56,
1703
+ "grad_norm": 0.9601121544837952,
1704
+ "learning_rate": 8.765094368267186e-05,
1705
+ "loss": 1.1868,
1706
+ "step": 239
1707
+ },
1708
+ {
1709
+ "epoch": 0.56,
1710
+ "grad_norm": 0.9471805095672607,
1711
+ "learning_rate": 8.688345254588578e-05,
1712
+ "loss": 1.1898,
1713
+ "step": 240
1714
+ },
1715
+ {
1716
+ "epoch": 0.57,
1717
+ "grad_norm": 0.8799854516983032,
1718
+ "learning_rate": 8.611674676235115e-05,
1719
+ "loss": 1.0781,
1720
+ "step": 241
1721
+ },
1722
+ {
1723
+ "epoch": 0.57,
1724
+ "grad_norm": 1.255386233329773,
1725
+ "learning_rate": 8.535087223857508e-05,
1726
+ "loss": 1.0936,
1727
+ "step": 242
1728
+ },
1729
+ {
1730
+ "epoch": 0.57,
1731
+ "grad_norm": 0.8445800542831421,
1732
+ "learning_rate": 8.458587483129316e-05,
1733
+ "loss": 1.2126,
1734
+ "step": 243
1735
+ },
1736
+ {
1737
+ "epoch": 0.57,
1738
+ "grad_norm": 0.7667735815048218,
1739
+ "learning_rate": 8.382180034472353e-05,
1740
+ "loss": 1.1693,
1741
+ "step": 244
1742
+ },
1743
+ {
1744
+ "epoch": 0.58,
1745
+ "grad_norm": 0.8405752182006836,
1746
+ "learning_rate": 8.305869452782446e-05,
1747
+ "loss": 1.2401,
1748
+ "step": 245
1749
+ },
1750
+ {
1751
+ "epoch": 0.58,
1752
+ "grad_norm": 1.1948215961456299,
1753
+ "learning_rate": 8.229660307155518e-05,
1754
+ "loss": 1.4077,
1755
+ "step": 246
1756
+ },
1757
+ {
1758
+ "epoch": 0.58,
1759
+ "grad_norm": 1.0930763483047485,
1760
+ "learning_rate": 8.153557160613998e-05,
1761
+ "loss": 1.3625,
1762
+ "step": 247
1763
+ },
1764
+ {
1765
+ "epoch": 0.58,
1766
+ "grad_norm": 0.8063607215881348,
1767
+ "learning_rate": 8.077564569833632e-05,
1768
+ "loss": 1.0265,
1769
+ "step": 248
1770
+ },
1771
+ {
1772
+ "epoch": 0.58,
1773
+ "grad_norm": 0.8202766180038452,
1774
+ "learning_rate": 8.00168708487063e-05,
1775
+ "loss": 1.154,
1776
+ "step": 249
1777
+ },
1778
+ {
1779
+ "epoch": 0.59,
1780
+ "grad_norm": 1.57820463180542,
1781
+ "learning_rate": 7.92592924888925e-05,
1782
+ "loss": 1.5653,
1783
+ "step": 250
1784
+ },
1785
+ {
1786
+ "epoch": 0.59,
1787
+ "grad_norm": 0.9507820010185242,
1788
+ "learning_rate": 7.85029559788976e-05,
1789
+ "loss": 1.0407,
1790
+ "step": 251
1791
+ },
1792
+ {
1793
+ "epoch": 0.59,
1794
+ "grad_norm": 0.890489935874939,
1795
+ "learning_rate": 7.774790660436858e-05,
1796
+ "loss": 1.2915,
1797
+ "step": 252
1798
+ },
1799
+ {
1800
+ "epoch": 0.59,
1801
+ "grad_norm": 0.9816429018974304,
1802
+ "learning_rate": 7.699418957388512e-05,
1803
+ "loss": 1.3749,
1804
+ "step": 253
1805
+ },
1806
+ {
1807
+ "epoch": 0.6,
1808
+ "grad_norm": 0.8766152262687683,
1809
+ "learning_rate": 7.624185001625292e-05,
1810
+ "loss": 1.0009,
1811
+ "step": 254
1812
+ },
1813
+ {
1814
+ "epoch": 0.6,
1815
+ "grad_norm": 1.0196894407272339,
1816
+ "learning_rate": 7.549093297780132e-05,
1817
+ "loss": 1.2238,
1818
+ "step": 255
1819
+ },
1820
+ {
1821
+ "epoch": 0.6,
1822
+ "grad_norm": 1.0334336757659912,
1823
+ "learning_rate": 7.474148341968652e-05,
1824
+ "loss": 0.8767,
1825
+ "step": 256
1826
+ },
1827
+ {
1828
+ "epoch": 0.6,
1829
+ "grad_norm": 0.8634482026100159,
1830
+ "learning_rate": 7.39935462151992e-05,
1831
+ "loss": 1.4253,
1832
+ "step": 257
1833
+ },
1834
+ {
1835
+ "epoch": 0.61,
1836
+ "grad_norm": 0.9882122278213501,
1837
+ "learning_rate": 7.324716614707793e-05,
1838
+ "loss": 1.3695,
1839
+ "step": 258
1840
+ },
1841
+ {
1842
+ "epoch": 0.61,
1843
+ "grad_norm": 0.7892170548439026,
1844
+ "learning_rate": 7.250238790482773e-05,
1845
+ "loss": 1.1509,
1846
+ "step": 259
1847
+ },
1848
+ {
1849
+ "epoch": 0.61,
1850
+ "grad_norm": 0.858100414276123,
1851
+ "learning_rate": 7.175925608204428e-05,
1852
+ "loss": 1.1307,
1853
+ "step": 260
1854
+ },
1855
+ {
1856
+ "epoch": 0.61,
1857
+ "grad_norm": 1.1348087787628174,
1858
+ "learning_rate": 7.101781517374398e-05,
1859
+ "loss": 1.0939,
1860
+ "step": 261
1861
+ },
1862
+ {
1863
+ "epoch": 0.62,
1864
+ "grad_norm": 1.0761414766311646,
1865
+ "learning_rate": 7.027810957369957e-05,
1866
+ "loss": 1.4693,
1867
+ "step": 262
1868
+ },
1869
+ {
1870
+ "epoch": 0.62,
1871
+ "grad_norm": 0.971714198589325,
1872
+ "learning_rate": 6.954018357178241e-05,
1873
+ "loss": 1.333,
1874
+ "step": 263
1875
+ },
1876
+ {
1877
+ "epoch": 0.62,
1878
+ "grad_norm": 0.748940646648407,
1879
+ "learning_rate": 6.880408135131022e-05,
1880
+ "loss": 0.9636,
1881
+ "step": 264
1882
+ },
1883
+ {
1884
+ "epoch": 0.62,
1885
+ "grad_norm": 0.8127878904342651,
1886
+ "learning_rate": 6.806984698640202e-05,
1887
+ "loss": 0.8037,
1888
+ "step": 265
1889
+ },
1890
+ {
1891
+ "epoch": 0.62,
1892
+ "grad_norm": 0.8631081581115723,
1893
+ "learning_rate": 6.733752443933878e-05,
1894
+ "loss": 0.977,
1895
+ "step": 266
1896
+ },
1897
+ {
1898
+ "epoch": 0.63,
1899
+ "grad_norm": 0.8358796238899231,
1900
+ "learning_rate": 6.660715755793154e-05,
1901
+ "loss": 1.0804,
1902
+ "step": 267
1903
+ },
1904
+ {
1905
+ "epoch": 0.63,
1906
+ "grad_norm": 0.9398666620254517,
1907
+ "learning_rate": 6.587879007289576e-05,
1908
+ "loss": 1.357,
1909
+ "step": 268
1910
+ },
1911
+ {
1912
+ "epoch": 0.63,
1913
+ "grad_norm": 1.1066324710845947,
1914
+ "learning_rate": 6.515246559523312e-05,
1915
+ "loss": 1.2569,
1916
+ "step": 269
1917
+ },
1918
+ {
1919
+ "epoch": 0.63,
1920
+ "grad_norm": 1.116248607635498,
1921
+ "learning_rate": 6.442822761362015e-05,
1922
+ "loss": 1.2156,
1923
+ "step": 270
1924
+ },
1925
+ {
1926
+ "epoch": 0.64,
1927
+ "grad_norm": 1.0011835098266602,
1928
+ "learning_rate": 6.370611949180457e-05,
1929
+ "loss": 1.2736,
1930
+ "step": 271
1931
+ },
1932
+ {
1933
+ "epoch": 0.64,
1934
+ "grad_norm": 0.8156257271766663,
1935
+ "learning_rate": 6.298618446600856e-05,
1936
+ "loss": 0.9655,
1937
+ "step": 272
1938
+ },
1939
+ {
1940
+ "epoch": 0.64,
1941
+ "grad_norm": 0.8273128271102905,
1942
+ "learning_rate": 6.22684656423404e-05,
1943
+ "loss": 0.7496,
1944
+ "step": 273
1945
+ },
1946
+ {
1947
+ "epoch": 0.64,
1948
+ "grad_norm": 0.855414628982544,
1949
+ "learning_rate": 6.155300599421306e-05,
1950
+ "loss": 1.1053,
1951
+ "step": 274
1952
+ },
1953
+ {
1954
+ "epoch": 0.65,
1955
+ "grad_norm": 0.8847690224647522,
1956
+ "learning_rate": 6.0839848359771536e-05,
1957
+ "loss": 1.2419,
1958
+ "step": 275
1959
+ },
1960
+ {
1961
+ "epoch": 0.65,
1962
+ "grad_norm": 0.7055469751358032,
1963
+ "learning_rate": 6.012903543932766e-05,
1964
+ "loss": 0.9169,
1965
+ "step": 276
1966
+ },
1967
+ {
1968
+ "epoch": 0.65,
1969
+ "grad_norm": 1.0745937824249268,
1970
+ "learning_rate": 5.9420609792803604e-05,
1971
+ "loss": 1.2988,
1972
+ "step": 277
1973
+ },
1974
+ {
1975
+ "epoch": 0.65,
1976
+ "grad_norm": 1.488973617553711,
1977
+ "learning_rate": 5.871461383718344e-05,
1978
+ "loss": 1.3937,
1979
+ "step": 278
1980
+ },
1981
+ {
1982
+ "epoch": 0.65,
1983
+ "grad_norm": 0.8680241703987122,
1984
+ "learning_rate": 5.801108984397354e-05,
1985
+ "loss": 0.7092,
1986
+ "step": 279
1987
+ },
1988
+ {
1989
+ "epoch": 0.66,
1990
+ "grad_norm": 0.7179112434387207,
1991
+ "learning_rate": 5.7310079936671545e-05,
1992
+ "loss": 0.9667,
1993
+ "step": 280
1994
+ },
1995
+ {
1996
+ "epoch": 0.66,
1997
+ "grad_norm": 0.944363534450531,
1998
+ "learning_rate": 5.6611626088244194e-05,
1999
+ "loss": 1.2307,
2000
+ "step": 281
2001
+ },
2002
+ {
2003
+ "epoch": 0.66,
2004
+ "grad_norm": 0.797068178653717,
2005
+ "learning_rate": 5.59157701186142e-05,
2006
+ "loss": 1.1412,
2007
+ "step": 282
2008
+ },
2009
+ {
2010
+ "epoch": 0.66,
2011
+ "grad_norm": 0.9618858695030212,
2012
+ "learning_rate": 5.522255369215622e-05,
2013
+ "loss": 1.1069,
2014
+ "step": 283
2015
+ },
2016
+ {
2017
+ "epoch": 0.67,
2018
+ "grad_norm": 1.047471284866333,
2019
+ "learning_rate": 5.453201831520245e-05,
2020
+ "loss": 1.0406,
2021
+ "step": 284
2022
+ },
2023
+ {
2024
+ "epoch": 0.67,
2025
+ "grad_norm": 0.8067429065704346,
2026
+ "learning_rate": 5.38442053335571e-05,
2027
+ "loss": 1.1467,
2028
+ "step": 285
2029
+ },
2030
+ {
2031
+ "epoch": 0.67,
2032
+ "grad_norm": 0.8198029398918152,
2033
+ "learning_rate": 5.3159155930021e-05,
2034
+ "loss": 1.1139,
2035
+ "step": 286
2036
+ },
2037
+ {
2038
+ "epoch": 0.67,
2039
+ "grad_norm": 0.7728237509727478,
2040
+ "learning_rate": 5.247691112192577e-05,
2041
+ "loss": 0.8623,
2042
+ "step": 287
2043
+ },
2044
+ {
2045
+ "epoch": 0.68,
2046
+ "grad_norm": 0.8685657978057861,
2047
+ "learning_rate": 5.179751175867784e-05,
2048
+ "loss": 1.0307,
2049
+ "step": 288
2050
+ },
2051
+ {
2052
+ "epoch": 0.68,
2053
+ "grad_norm": 0.7406818866729736,
2054
+ "learning_rate": 5.112099851931265e-05,
2055
+ "loss": 0.9348,
2056
+ "step": 289
2057
+ },
2058
+ {
2059
+ "epoch": 0.68,
2060
+ "grad_norm": 1.017289638519287,
2061
+ "learning_rate": 5.044741191005908e-05,
2062
+ "loss": 1.131,
2063
+ "step": 290
2064
+ },
2065
+ {
2066
+ "epoch": 0.68,
2067
+ "grad_norm": 0.7423017024993896,
2068
+ "learning_rate": 4.9776792261913896e-05,
2069
+ "loss": 1.0436,
2070
+ "step": 291
2071
+ },
2072
+ {
2073
+ "epoch": 0.69,
2074
+ "grad_norm": 0.7347959876060486,
2075
+ "learning_rate": 4.910917972822713e-05,
2076
+ "loss": 1.0401,
2077
+ "step": 292
2078
+ },
2079
+ {
2080
+ "epoch": 0.69,
2081
+ "grad_norm": 0.9067183136940002,
2082
+ "learning_rate": 4.844461428229782e-05,
2083
+ "loss": 1.3648,
2084
+ "step": 293
2085
+ },
2086
+ {
2087
+ "epoch": 0.69,
2088
+ "grad_norm": 0.8086903691291809,
2089
+ "learning_rate": 4.7783135714980744e-05,
2090
+ "loss": 1.1079,
2091
+ "step": 294
2092
+ },
2093
+ {
2094
+ "epoch": 0.69,
2095
+ "grad_norm": 0.8241545557975769,
2096
+ "learning_rate": 4.712478363230362e-05,
2097
+ "loss": 1.3574,
2098
+ "step": 295
2099
+ },
2100
+ {
2101
+ "epoch": 0.69,
2102
+ "grad_norm": 0.8495978713035583,
2103
+ "learning_rate": 4.646959745309609e-05,
2104
+ "loss": 1.152,
2105
+ "step": 296
2106
+ },
2107
+ {
2108
+ "epoch": 0.7,
2109
+ "grad_norm": 0.8596299290657043,
2110
+ "learning_rate": 4.581761640662927e-05,
2111
+ "loss": 1.2126,
2112
+ "step": 297
2113
+ },
2114
+ {
2115
+ "epoch": 0.7,
2116
+ "grad_norm": 0.9820694327354431,
2117
+ "learning_rate": 4.516887953026691e-05,
2118
+ "loss": 1.0377,
2119
+ "step": 298
2120
+ },
2121
+ {
2122
+ "epoch": 0.7,
2123
+ "grad_norm": 0.8206368088722229,
2124
+ "learning_rate": 4.452342566712818e-05,
2125
+ "loss": 1.0879,
2126
+ "step": 299
2127
+ },
2128
+ {
2129
+ "epoch": 0.7,
2130
+ "grad_norm": 0.9816787242889404,
2131
+ "learning_rate": 4.388129346376178e-05,
2132
+ "loss": 1.1499,
2133
+ "step": 300
2134
+ },
2135
+ {
2136
+ "epoch": 0.71,
2137
+ "grad_norm": 0.8422172665596008,
2138
+ "learning_rate": 4.3242521367832015e-05,
2139
+ "loss": 1.0807,
2140
+ "step": 301
2141
+ },
2142
+ {
2143
+ "epoch": 0.71,
2144
+ "grad_norm": 0.9412662386894226,
2145
+ "learning_rate": 4.260714762581677e-05,
2146
+ "loss": 1.067,
2147
+ "step": 302
2148
+ },
2149
+ {
2150
+ "epoch": 0.71,
2151
+ "grad_norm": 0.8067646026611328,
2152
+ "learning_rate": 4.197521028071765e-05,
2153
+ "loss": 1.2295,
2154
+ "step": 303
2155
+ },
2156
+ {
2157
+ "epoch": 0.71,
2158
+ "grad_norm": 1.0259385108947754,
2159
+ "learning_rate": 4.13467471697817e-05,
2160
+ "loss": 0.9919,
2161
+ "step": 304
2162
+ },
2163
+ {
2164
+ "epoch": 0.72,
2165
+ "grad_norm": 0.8335386514663696,
2166
+ "learning_rate": 4.0721795922236496e-05,
2167
+ "loss": 1.0636,
2168
+ "step": 305
2169
+ },
2170
+ {
2171
+ "epoch": 0.72,
2172
+ "grad_norm": 1.1450220346450806,
2173
+ "learning_rate": 4.010039395703664e-05,
2174
+ "loss": 1.1161,
2175
+ "step": 306
2176
+ },
2177
+ {
2178
+ "epoch": 0.72,
2179
+ "grad_norm": 0.9241200089454651,
2180
+ "learning_rate": 3.948257848062351e-05,
2181
+ "loss": 1.2009,
2182
+ "step": 307
2183
+ },
2184
+ {
2185
+ "epoch": 0.72,
2186
+ "grad_norm": 1.1173282861709595,
2187
+ "learning_rate": 3.8868386484697417e-05,
2188
+ "loss": 1.195,
2189
+ "step": 308
2190
+ },
2191
+ {
2192
+ "epoch": 0.73,
2193
+ "grad_norm": 0.9357693791389465,
2194
+ "learning_rate": 3.825785474400291e-05,
2195
+ "loss": 1.0678,
2196
+ "step": 309
2197
+ },
2198
+ {
2199
+ "epoch": 0.73,
2200
+ "grad_norm": 1.0329748392105103,
2201
+ "learning_rate": 3.7651019814126654e-05,
2202
+ "loss": 1.4416,
2203
+ "step": 310
2204
+ },
2205
+ {
2206
+ "epoch": 0.73,
2207
+ "grad_norm": 0.9039399027824402,
2208
+ "learning_rate": 3.7047918029308815e-05,
2209
+ "loss": 1.0469,
2210
+ "step": 311
2211
+ },
2212
+ {
2213
+ "epoch": 0.73,
2214
+ "grad_norm": 1.1105232238769531,
2215
+ "learning_rate": 3.6448585500267485e-05,
2216
+ "loss": 1.2242,
2217
+ "step": 312
2218
+ },
2219
+ {
2220
+ "epoch": 0.73,
2221
+ "grad_norm": 0.9321856498718262,
2222
+ "learning_rate": 3.5853058112036596e-05,
2223
+ "loss": 1.2051,
2224
+ "step": 313
2225
+ },
2226
+ {
2227
+ "epoch": 0.74,
2228
+ "grad_norm": 0.8780171871185303,
2229
+ "learning_rate": 3.5261371521817244e-05,
2230
+ "loss": 1.0514,
2231
+ "step": 314
2232
+ },
2233
+ {
2234
+ "epoch": 0.74,
2235
+ "grad_norm": 1.3793842792510986,
2236
+ "learning_rate": 3.467356115684284e-05,
2237
+ "loss": 1.2994,
2238
+ "step": 315
2239
+ },
2240
+ {
2241
+ "epoch": 0.74,
2242
+ "grad_norm": 0.9189188480377197,
2243
+ "learning_rate": 3.408966221225773e-05,
2244
+ "loss": 0.9158,
2245
+ "step": 316
2246
+ },
2247
+ {
2248
+ "epoch": 0.74,
2249
+ "grad_norm": 0.979479193687439,
2250
+ "learning_rate": 3.350970964900998e-05,
2251
+ "loss": 1.0156,
2252
+ "step": 317
2253
+ },
2254
+ {
2255
+ "epoch": 0.75,
2256
+ "grad_norm": 1.0084733963012695,
2257
+ "learning_rate": 3.293373819175816e-05,
2258
+ "loss": 1.1564,
2259
+ "step": 318
2260
+ },
2261
+ {
2262
+ "epoch": 0.75,
2263
+ "grad_norm": 0.708268404006958,
2264
+ "learning_rate": 3.236178232679202e-05,
2265
+ "loss": 0.6927,
2266
+ "step": 319
2267
+ },
2268
+ {
2269
+ "epoch": 0.75,
2270
+ "grad_norm": 0.8528488278388977,
2271
+ "learning_rate": 3.1793876299967816e-05,
2272
+ "loss": 1.2442,
2273
+ "step": 320
2274
+ },
2275
+ {
2276
+ "epoch": 0.75,
2277
+ "grad_norm": 0.9393438100814819,
2278
+ "learning_rate": 3.123005411465766e-05,
2279
+ "loss": 1.3178,
2280
+ "step": 321
2281
+ },
2282
+ {
2283
+ "epoch": 0.75,
2284
+ "eval_loss": 1.1110302209854126,
2285
+ "eval_runtime": 6.0903,
2286
+ "eval_samples_per_second": 16.42,
2287
+ "eval_steps_per_second": 16.42,
2288
+ "step": 321
2289
+ },
2290
+ {
2291
+ "epoch": 0.76,
2292
+ "grad_norm": 0.8227856159210205,
2293
+ "learning_rate": 3.0670349529713816e-05,
2294
+ "loss": 1.0518,
2295
+ "step": 322
2296
+ },
2297
+ {
2298
+ "epoch": 0.76,
2299
+ "grad_norm": 0.8559508323669434,
2300
+ "learning_rate": 3.0114796057447026e-05,
2301
+ "loss": 1.0809,
2302
+ "step": 323
2303
+ },
2304
+ {
2305
+ "epoch": 0.76,
2306
+ "grad_norm": 0.8127685189247131,
2307
+ "learning_rate": 2.9563426961620367e-05,
2308
+ "loss": 1.2037,
2309
+ "step": 324
2310
+ },
2311
+ {
2312
+ "epoch": 0.76,
2313
+ "grad_norm": 0.8787775039672852,
2314
+ "learning_rate": 2.901627525545726e-05,
2315
+ "loss": 1.0329,
2316
+ "step": 325
2317
+ },
2318
+ {
2319
+ "epoch": 0.77,
2320
+ "grad_norm": 0.9373772740364075,
2321
+ "learning_rate": 2.8473373699664997e-05,
2322
+ "loss": 1.2817,
2323
+ "step": 326
2324
+ },
2325
+ {
2326
+ "epoch": 0.77,
2327
+ "grad_norm": 0.8289972543716431,
2328
+ "learning_rate": 2.793475480047303e-05,
2329
+ "loss": 0.9784,
2330
+ "step": 327
2331
+ },
2332
+ {
2333
+ "epoch": 0.77,
2334
+ "grad_norm": 1.0169614553451538,
2335
+ "learning_rate": 2.7400450807686938e-05,
2336
+ "loss": 1.2231,
2337
+ "step": 328
2338
+ },
2339
+ {
2340
+ "epoch": 0.77,
2341
+ "grad_norm": 0.8996503949165344,
2342
+ "learning_rate": 2.687049371275705e-05,
2343
+ "loss": 1.0073,
2344
+ "step": 329
2345
+ },
2346
+ {
2347
+ "epoch": 0.77,
2348
+ "grad_norm": 1.0333868265151978,
2349
+ "learning_rate": 2.6344915246863412e-05,
2350
+ "loss": 1.2796,
2351
+ "step": 330
2352
+ },
2353
+ {
2354
+ "epoch": 0.78,
2355
+ "grad_norm": 0.8676667213439941,
2356
+ "learning_rate": 2.582374687901553e-05,
2357
+ "loss": 1.3943,
2358
+ "step": 331
2359
+ },
2360
+ {
2361
+ "epoch": 0.78,
2362
+ "grad_norm": 0.9073312878608704,
2363
+ "learning_rate": 2.5307019814168342e-05,
2364
+ "loss": 1.4125,
2365
+ "step": 332
2366
+ },
2367
+ {
2368
+ "epoch": 0.78,
2369
+ "grad_norm": 0.7397946119308472,
2370
+ "learning_rate": 2.4794764991353748e-05,
2371
+ "loss": 0.9884,
2372
+ "step": 333
2373
+ },
2374
+ {
2375
+ "epoch": 0.78,
2376
+ "grad_norm": 0.8001339435577393,
2377
+ "learning_rate": 2.4287013081828257e-05,
2378
+ "loss": 1.0607,
2379
+ "step": 334
2380
+ },
2381
+ {
2382
+ "epoch": 0.79,
2383
+ "grad_norm": 0.7958093881607056,
2384
+ "learning_rate": 2.3783794487236365e-05,
2385
+ "loss": 1.0917,
2386
+ "step": 335
2387
+ },
2388
+ {
2389
+ "epoch": 0.79,
2390
+ "grad_norm": 0.7405944466590881,
2391
+ "learning_rate": 2.328513933779034e-05,
2392
+ "loss": 1.1819,
2393
+ "step": 336
2394
+ },
2395
+ {
2396
+ "epoch": 0.79,
2397
+ "grad_norm": 0.8886005878448486,
2398
+ "learning_rate": 2.2791077490466262e-05,
2399
+ "loss": 1.1103,
2400
+ "step": 337
2401
+ },
2402
+ {
2403
+ "epoch": 0.79,
2404
+ "grad_norm": 0.9879472255706787,
2405
+ "learning_rate": 2.2301638527216194e-05,
2406
+ "loss": 1.5232,
2407
+ "step": 338
2408
+ },
2409
+ {
2410
+ "epoch": 0.8,
2411
+ "grad_norm": 0.6994455456733704,
2412
+ "learning_rate": 2.181685175319702e-05,
2413
+ "loss": 1.0496,
2414
+ "step": 339
2415
+ },
2416
+ {
2417
+ "epoch": 0.8,
2418
+ "grad_norm": 1.373903512954712,
2419
+ "learning_rate": 2.1336746195015846e-05,
2420
+ "loss": 1.2936,
2421
+ "step": 340
2422
+ },
2423
+ {
2424
+ "epoch": 0.8,
2425
+ "grad_norm": 1.3520807027816772,
2426
+ "learning_rate": 2.0861350598991945e-05,
2427
+ "loss": 1.5331,
2428
+ "step": 341
2429
+ },
2430
+ {
2431
+ "epoch": 0.8,
2432
+ "grad_norm": 0.7715508341789246,
2433
+ "learning_rate": 2.0390693429435627e-05,
2434
+ "loss": 1.0536,
2435
+ "step": 342
2436
+ },
2437
+ {
2438
+ "epoch": 0.81,
2439
+ "grad_norm": 0.9478192925453186,
2440
+ "learning_rate": 1.992480286694397e-05,
2441
+ "loss": 1.3032,
2442
+ "step": 343
2443
+ },
2444
+ {
2445
+ "epoch": 0.81,
2446
+ "grad_norm": 0.8158330917358398,
2447
+ "learning_rate": 1.946370680671341e-05,
2448
+ "loss": 1.1646,
2449
+ "step": 344
2450
+ },
2451
+ {
2452
+ "epoch": 0.81,
2453
+ "grad_norm": 1.5716131925582886,
2454
+ "learning_rate": 1.90074328568696e-05,
2455
+ "loss": 1.8497,
2456
+ "step": 345
2457
+ },
2458
+ {
2459
+ "epoch": 0.81,
2460
+ "grad_norm": 0.9847422242164612,
2461
+ "learning_rate": 1.85560083368143e-05,
2462
+ "loss": 1.181,
2463
+ "step": 346
2464
+ },
2465
+ {
2466
+ "epoch": 0.81,
2467
+ "grad_norm": 1.0614407062530518,
2468
+ "learning_rate": 1.8109460275589773e-05,
2469
+ "loss": 1.3074,
2470
+ "step": 347
2471
+ },
2472
+ {
2473
+ "epoch": 0.82,
2474
+ "grad_norm": 0.9004321098327637,
2475
+ "learning_rate": 1.766781541026018e-05,
2476
+ "loss": 1.0441,
2477
+ "step": 348
2478
+ },
2479
+ {
2480
+ "epoch": 0.82,
2481
+ "grad_norm": 1.0964152812957764,
2482
+ "learning_rate": 1.7231100184310956e-05,
2483
+ "loss": 1.1583,
2484
+ "step": 349
2485
+ },
2486
+ {
2487
+ "epoch": 0.82,
2488
+ "grad_norm": 0.939444363117218,
2489
+ "learning_rate": 1.679934074606533e-05,
2490
+ "loss": 1.2871,
2491
+ "step": 350
2492
+ },
2493
+ {
2494
+ "epoch": 0.82,
2495
+ "grad_norm": 0.8655087351799011,
2496
+ "learning_rate": 1.6372562947118763e-05,
2497
+ "loss": 1.1802,
2498
+ "step": 351
2499
+ },
2500
+ {
2501
+ "epoch": 0.83,
2502
+ "grad_norm": 0.8392437696456909,
2503
+ "learning_rate": 1.5950792340791043e-05,
2504
+ "loss": 1.0773,
2505
+ "step": 352
2506
+ },
2507
+ {
2508
+ "epoch": 0.83,
2509
+ "grad_norm": 0.856883704662323,
2510
+ "learning_rate": 1.5534054180596415e-05,
2511
+ "loss": 1.111,
2512
+ "step": 353
2513
+ },
2514
+ {
2515
+ "epoch": 0.83,
2516
+ "grad_norm": 0.9138004183769226,
2517
+ "learning_rate": 1.5122373418731306e-05,
2518
+ "loss": 0.9742,
2519
+ "step": 354
2520
+ },
2521
+ {
2522
+ "epoch": 0.83,
2523
+ "grad_norm": 0.8892737627029419,
2524
+ "learning_rate": 1.4715774704580453e-05,
2525
+ "loss": 1.3606,
2526
+ "step": 355
2527
+ },
2528
+ {
2529
+ "epoch": 0.84,
2530
+ "grad_norm": 0.8643128275871277,
2531
+ "learning_rate": 1.4314282383241096e-05,
2532
+ "loss": 1.2462,
2533
+ "step": 356
2534
+ },
2535
+ {
2536
+ "epoch": 0.84,
2537
+ "grad_norm": 0.8623273372650146,
2538
+ "learning_rate": 1.3917920494065029e-05,
2539
+ "loss": 0.9758,
2540
+ "step": 357
2541
+ },
2542
+ {
2543
+ "epoch": 0.84,
2544
+ "grad_norm": 0.9717140793800354,
2545
+ "learning_rate": 1.3526712769219618e-05,
2546
+ "loss": 1.1752,
2547
+ "step": 358
2548
+ },
2549
+ {
2550
+ "epoch": 0.84,
2551
+ "grad_norm": 1.449154257774353,
2552
+ "learning_rate": 1.3140682632266543e-05,
2553
+ "loss": 0.974,
2554
+ "step": 359
2555
+ },
2556
+ {
2557
+ "epoch": 0.85,
2558
+ "grad_norm": 0.8342477083206177,
2559
+ "learning_rate": 1.2759853196759453e-05,
2560
+ "loss": 1.0768,
2561
+ "step": 360
2562
+ },
2563
+ {
2564
+ "epoch": 0.85,
2565
+ "grad_norm": 0.699179470539093,
2566
+ "learning_rate": 1.2384247264859972e-05,
2567
+ "loss": 1.0079,
2568
+ "step": 361
2569
+ },
2570
+ {
2571
+ "epoch": 0.85,
2572
+ "grad_norm": 0.9927258491516113,
2573
+ "learning_rate": 1.201388732597255e-05,
2574
+ "loss": 1.16,
2575
+ "step": 362
2576
+ },
2577
+ {
2578
+ "epoch": 0.85,
2579
+ "grad_norm": 0.8921307325363159,
2580
+ "learning_rate": 1.1648795555397719e-05,
2581
+ "loss": 1.1092,
2582
+ "step": 363
2583
+ },
2584
+ {
2585
+ "epoch": 0.85,
2586
+ "grad_norm": 0.8358816504478455,
2587
+ "learning_rate": 1.1288993813004467e-05,
2588
+ "loss": 0.9703,
2589
+ "step": 364
2590
+ },
2591
+ {
2592
+ "epoch": 0.86,
2593
+ "grad_norm": 1.0156934261322021,
2594
+ "learning_rate": 1.0934503641921402e-05,
2595
+ "loss": 1.3109,
2596
+ "step": 365
2597
+ },
2598
+ {
2599
+ "epoch": 0.86,
2600
+ "grad_norm": 0.7621612548828125,
2601
+ "learning_rate": 1.0585346267246743e-05,
2602
+ "loss": 0.5513,
2603
+ "step": 366
2604
+ },
2605
+ {
2606
+ "epoch": 0.86,
2607
+ "grad_norm": 1.0127166509628296,
2608
+ "learning_rate": 1.0241542594777576e-05,
2609
+ "loss": 0.9964,
2610
+ "step": 367
2611
+ },
2612
+ {
2613
+ "epoch": 0.86,
2614
+ "grad_norm": 0.8973998427391052,
2615
+ "learning_rate": 9.903113209758096e-06,
2616
+ "loss": 1.1234,
2617
+ "step": 368
2618
+ },
2619
+ {
2620
+ "epoch": 0.87,
2621
+ "grad_norm": 1.2160240411758423,
2622
+ "learning_rate": 9.570078375647006e-06,
2623
+ "loss": 1.8291,
2624
+ "step": 369
2625
+ },
2626
+ {
2627
+ "epoch": 0.87,
2628
+ "grad_norm": 0.9975070357322693,
2629
+ "learning_rate": 9.242458032904311e-06,
2630
+ "loss": 1.0111,
2631
+ "step": 370
2632
+ },
2633
+ {
2634
+ "epoch": 0.87,
2635
+ "grad_norm": 1.1726033687591553,
2636
+ "learning_rate": 8.92027179779732e-06,
2637
+ "loss": 1.3626,
2638
+ "step": 371
2639
+ },
2640
+ {
2641
+ "epoch": 0.87,
2642
+ "grad_norm": 0.8554450869560242,
2643
+ "learning_rate": 8.603538961226232e-06,
2644
+ "loss": 1.2957,
2645
+ "step": 372
2646
+ },
2647
+ {
2648
+ "epoch": 0.88,
2649
+ "grad_norm": 0.8581518530845642,
2650
+ "learning_rate": 8.29227848756895e-06,
2651
+ "loss": 1.1524,
2652
+ "step": 373
2653
+ },
2654
+ {
2655
+ "epoch": 0.88,
2656
+ "grad_norm": 0.834023118019104,
2657
+ "learning_rate": 7.986509013545673e-06,
2658
+ "loss": 1.0957,
2659
+ "step": 374
2660
+ },
2661
+ {
2662
+ "epoch": 0.88,
2663
+ "grad_norm": 0.9786900281906128,
2664
+ "learning_rate": 7.686248847103072e-06,
2665
+ "loss": 0.9576,
2666
+ "step": 375
2667
+ },
2668
+ {
2669
+ "epoch": 0.88,
2670
+ "grad_norm": 1.0205256938934326,
2671
+ "learning_rate": 7.3915159663179075e-06,
2672
+ "loss": 1.4658,
2673
+ "step": 376
2674
+ },
2675
+ {
2676
+ "epoch": 0.88,
2677
+ "grad_norm": 1.0132267475128174,
2678
+ "learning_rate": 7.102328018320858e-06,
2679
+ "loss": 1.1489,
2680
+ "step": 377
2681
+ },
2682
+ {
2683
+ "epoch": 0.89,
2684
+ "grad_norm": 0.8851252198219299,
2685
+ "learning_rate": 6.818702318239689e-06,
2686
+ "loss": 1.1875,
2687
+ "step": 378
2688
+ },
2689
+ {
2690
+ "epoch": 0.89,
2691
+ "grad_norm": 0.658548891544342,
2692
+ "learning_rate": 6.540655848162602e-06,
2693
+ "loss": 0.7974,
2694
+ "step": 379
2695
+ },
2696
+ {
2697
+ "epoch": 0.89,
2698
+ "grad_norm": 0.877100944519043,
2699
+ "learning_rate": 6.268205256121396e-06,
2700
+ "loss": 1.2753,
2701
+ "step": 380
2702
+ },
2703
+ {
2704
+ "epoch": 0.89,
2705
+ "grad_norm": 0.9094898104667664,
2706
+ "learning_rate": 6.001366855094748e-06,
2707
+ "loss": 1.2567,
2708
+ "step": 381
2709
+ },
2710
+ {
2711
+ "epoch": 0.9,
2712
+ "grad_norm": 1.538279414176941,
2713
+ "learning_rate": 5.7401566220313005e-06,
2714
+ "loss": 0.9588,
2715
+ "step": 382
2716
+ },
2717
+ {
2718
+ "epoch": 0.9,
2719
+ "grad_norm": 0.7195213437080383,
2720
+ "learning_rate": 5.484590196893247e-06,
2721
+ "loss": 0.7562,
2722
+ "step": 383
2723
+ },
2724
+ {
2725
+ "epoch": 0.9,
2726
+ "grad_norm": 0.9021754860877991,
2727
+ "learning_rate": 5.2346828817197655e-06,
2728
+ "loss": 1.3057,
2729
+ "step": 384
2730
+ },
2731
+ {
2732
+ "epoch": 0.9,
2733
+ "grad_norm": 0.8472947478294373,
2734
+ "learning_rate": 4.990449639710815e-06,
2735
+ "loss": 1.308,
2736
+ "step": 385
2737
+ },
2738
+ {
2739
+ "epoch": 0.91,
2740
+ "grad_norm": 0.9867368936538696,
2741
+ "learning_rate": 4.7519050943312325e-06,
2742
+ "loss": 1.2082,
2743
+ "step": 386
2744
+ },
2745
+ {
2746
+ "epoch": 0.91,
2747
+ "grad_norm": 0.7588489055633545,
2748
+ "learning_rate": 4.5190635284352075e-06,
2749
+ "loss": 1.1311,
2750
+ "step": 387
2751
+ },
2752
+ {
2753
+ "epoch": 0.91,
2754
+ "grad_norm": 0.763214111328125,
2755
+ "learning_rate": 4.291938883411007e-06,
2756
+ "loss": 1.2086,
2757
+ "step": 388
2758
+ },
2759
+ {
2760
+ "epoch": 0.91,
2761
+ "grad_norm": 0.8648312091827393,
2762
+ "learning_rate": 4.070544758346273e-06,
2763
+ "loss": 1.005,
2764
+ "step": 389
2765
+ },
2766
+ {
2767
+ "epoch": 0.92,
2768
+ "grad_norm": 1.0920186042785645,
2769
+ "learning_rate": 3.85489440921376e-06,
2770
+ "loss": 1.0133,
2771
+ "step": 390
2772
+ },
2773
+ {
2774
+ "epoch": 0.92,
2775
+ "grad_norm": 0.7375323176383972,
2776
+ "learning_rate": 3.6450007480777093e-06,
2777
+ "loss": 0.9391,
2778
+ "step": 391
2779
+ },
2780
+ {
2781
+ "epoch": 0.92,
2782
+ "grad_norm": 0.8165771961212158,
2783
+ "learning_rate": 3.440876342320609e-06,
2784
+ "loss": 1.0479,
2785
+ "step": 392
2786
+ },
2787
+ {
2788
+ "epoch": 0.92,
2789
+ "grad_norm": 0.8869404792785645,
2790
+ "learning_rate": 3.2425334138908583e-06,
2791
+ "loss": 1.08,
2792
+ "step": 393
2793
+ },
2794
+ {
2795
+ "epoch": 0.92,
2796
+ "grad_norm": 0.7623941898345947,
2797
+ "learning_rate": 3.049983838570858e-06,
2798
+ "loss": 1.1899,
2799
+ "step": 394
2800
+ },
2801
+ {
2802
+ "epoch": 0.93,
2803
+ "grad_norm": 0.9894457459449768,
2804
+ "learning_rate": 2.863239145266028e-06,
2805
+ "loss": 1.3285,
2806
+ "step": 395
2807
+ },
2808
+ {
2809
+ "epoch": 0.93,
2810
+ "grad_norm": 0.9151179194450378,
2811
+ "learning_rate": 2.682310515314512e-06,
2812
+ "loss": 1.0843,
2813
+ "step": 396
2814
+ },
2815
+ {
2816
+ "epoch": 0.93,
2817
+ "grad_norm": 0.9756556749343872,
2818
+ "learning_rate": 2.5072087818176382e-06,
2819
+ "loss": 1.3407,
2820
+ "step": 397
2821
+ },
2822
+ {
2823
+ "epoch": 0.93,
2824
+ "grad_norm": 0.8172594308853149,
2825
+ "learning_rate": 2.3379444289913342e-06,
2826
+ "loss": 1.0445,
2827
+ "step": 398
2828
+ },
2829
+ {
2830
+ "epoch": 0.94,
2831
+ "grad_norm": 1.0837291479110718,
2832
+ "learning_rate": 2.174527591538367e-06,
2833
+ "loss": 1.2118,
2834
+ "step": 399
2835
+ },
2836
+ {
2837
+ "epoch": 0.94,
2838
+ "grad_norm": 0.9625768661499023,
2839
+ "learning_rate": 2.016968054041546e-06,
2840
+ "loss": 1.1475,
2841
+ "step": 400
2842
+ },
2843
+ {
2844
+ "epoch": 0.94,
2845
+ "grad_norm": 0.9723854660987854,
2846
+ "learning_rate": 1.8652752503778404e-06,
2847
+ "loss": 1.1693,
2848
+ "step": 401
2849
+ },
2850
+ {
2851
+ "epoch": 0.94,
2852
+ "grad_norm": 1.2548187971115112,
2853
+ "learning_rate": 1.7194582631535617e-06,
2854
+ "loss": 1.3045,
2855
+ "step": 402
2856
+ },
2857
+ {
2858
+ "epoch": 0.95,
2859
+ "grad_norm": 0.6179590821266174,
2860
+ "learning_rate": 1.5795258231605103e-06,
2861
+ "loss": 0.4655,
2862
+ "step": 403
2863
+ },
2864
+ {
2865
+ "epoch": 0.95,
2866
+ "grad_norm": 0.8926462531089783,
2867
+ "learning_rate": 1.4454863088532388e-06,
2868
+ "loss": 0.8854,
2869
+ "step": 404
2870
+ },
2871
+ {
2872
+ "epoch": 0.95,
2873
+ "grad_norm": 1.0097955465316772,
2874
+ "learning_rate": 1.317347745847386e-06,
2875
+ "loss": 1.3568,
2876
+ "step": 405
2877
+ },
2878
+ {
2879
+ "epoch": 0.95,
2880
+ "grad_norm": 0.8205281496047974,
2881
+ "learning_rate": 1.19511780643915e-06,
2882
+ "loss": 0.9675,
2883
+ "step": 406
2884
+ },
2885
+ {
2886
+ "epoch": 0.96,
2887
+ "grad_norm": 0.9589681029319763,
2888
+ "learning_rate": 1.0788038091458897e-06,
2889
+ "loss": 1.1202,
2890
+ "step": 407
2891
+ },
2892
+ {
2893
+ "epoch": 0.96,
2894
+ "grad_norm": 0.8816035985946655,
2895
+ "learning_rate": 9.684127182679526e-07,
2896
+ "loss": 1.1746,
2897
+ "step": 408
2898
+ },
2899
+ {
2900
+ "epoch": 0.96,
2901
+ "grad_norm": 0.8051217198371887,
2902
+ "learning_rate": 8.639511434716863e-07,
2903
+ "loss": 0.975,
2904
+ "step": 409
2905
+ },
2906
+ {
2907
+ "epoch": 0.96,
2908
+ "grad_norm": 0.8666868209838867,
2909
+ "learning_rate": 7.654253393936439e-07,
2910
+ "loss": 1.2602,
2911
+ "step": 410
2912
+ },
2913
+ {
2914
+ "epoch": 0.96,
2915
+ "grad_norm": 0.9826870560646057,
2916
+ "learning_rate": 6.728412052661504e-07,
2917
+ "loss": 1.1426,
2918
+ "step": 411
2919
+ },
2920
+ {
2921
+ "epoch": 0.97,
2922
+ "grad_norm": 1.006511926651001,
2923
+ "learning_rate": 5.862042845640403e-07,
2924
+ "loss": 1.1518,
2925
+ "step": 412
2926
+ },
2927
+ {
2928
+ "epoch": 0.97,
2929
+ "grad_norm": 0.7082429528236389,
2930
+ "learning_rate": 5.055197646727572e-07,
2931
+ "loss": 0.9126,
2932
+ "step": 413
2933
+ },
2934
+ {
2935
+ "epoch": 0.97,
2936
+ "grad_norm": 1.163749098777771,
2937
+ "learning_rate": 4.307924765777682e-07,
2938
+ "loss": 1.3254,
2939
+ "step": 414
2940
+ },
2941
+ {
2942
+ "epoch": 0.97,
2943
+ "grad_norm": 0.9457451105117798,
2944
+ "learning_rate": 3.620268945752847e-07,
2945
+ "loss": 1.605,
2946
+ "step": 415
2947
+ },
2948
+ {
2949
+ "epoch": 0.98,
2950
+ "grad_norm": 0.7310549020767212,
2951
+ "learning_rate": 2.9922713600439854e-07,
2952
+ "loss": 0.9009,
2953
+ "step": 416
2954
+ },
2955
+ {
2956
+ "epoch": 0.98,
2957
+ "grad_norm": 1.1131972074508667,
2958
+ "learning_rate": 2.423969610005017e-07,
2959
+ "loss": 1.1344,
2960
+ "step": 417
2961
+ },
2962
+ {
2963
+ "epoch": 0.98,
2964
+ "grad_norm": 0.9373881816864014,
2965
+ "learning_rate": 1.915397722702217e-07,
2966
+ "loss": 1.0736,
2967
+ "step": 418
2968
+ },
2969
+ {
2970
+ "epoch": 0.98,
2971
+ "grad_norm": 0.819059431552887,
2972
+ "learning_rate": 1.4665861488761813e-07,
2973
+ "loss": 0.8256,
2974
+ "step": 419
2975
+ },
2976
+ {
2977
+ "epoch": 0.99,
2978
+ "grad_norm": 0.8947961926460266,
2979
+ "learning_rate": 1.0775617611189503e-07,
2980
+ "loss": 1.3029,
2981
+ "step": 420
2982
+ },
2983
+ {
2984
+ "epoch": 0.99,
2985
+ "grad_norm": 0.7374153137207031,
2986
+ "learning_rate": 7.483478522649634e-08,
2987
+ "loss": 1.0694,
2988
+ "step": 421
2989
+ },
2990
+ {
2991
+ "epoch": 0.99,
2992
+ "grad_norm": 0.8008151650428772,
2993
+ "learning_rate": 4.789641339963957e-08,
2994
+ "loss": 1.0821,
2995
+ "step": 422
2996
+ },
2997
+ {
2998
+ "epoch": 0.99,
2999
+ "grad_norm": 1.0294992923736572,
3000
+ "learning_rate": 2.6942673566265897e-08,
3001
+ "loss": 1.3806,
3002
+ "step": 423
3003
+ },
3004
+ {
3005
+ "epoch": 1.0,
3006
+ "grad_norm": 0.9338717460632324,
3007
+ "learning_rate": 1.1974820331517312e-08,
3008
+ "loss": 1.3182,
3009
+ "step": 424
3010
+ },
3011
+ {
3012
+ "epoch": 1.0,
3013
+ "grad_norm": 0.9727728366851807,
3014
+ "learning_rate": 2.9937498955745493e-09,
3015
+ "loss": 1.3143,
3016
+ "step": 425
3017
+ },
3018
+ {
3019
+ "epoch": 1.0,
3020
+ "grad_norm": 0.943352997303009,
3021
+ "learning_rate": 0.0,
3022
+ "loss": 0.9565,
3023
+ "step": 426
3024
+ }
3025
+ ],
3026
+ "logging_steps": 1,
3027
+ "max_steps": 426,
3028
+ "num_input_tokens_seen": 0,
3029
+ "num_train_epochs": 1,
3030
+ "save_steps": 500,
3031
+ "total_flos": 7758429166387200.0,
3032
+ "train_batch_size": 1,
3033
+ "trial_name": null,
3034
+ "trial_params": null
3035
+ }
checkpoint-426/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a35ee37b6bc55089085e74bcdcb56850bc492bb11f261c2c3c93fb7268fe3c72
3
+ size 5752
config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": true,
23
+ "_load_in_8bit": false,
24
+ "bnb_4bit_compute_dtype": "float16",
25
+ "bnb_4bit_quant_storage": "float32",
26
+ "bnb_4bit_quant_type": "nf4",
27
+ "bnb_4bit_use_double_quant": true,
28
+ "llm_int8_enable_fp32_cpu_offload": false,
29
+ "llm_int8_has_fp16_weight": false,
30
+ "llm_int8_skip_modules": null,
31
+ "llm_int8_threshold": 6.0,
32
+ "load_in_4bit": true,
33
+ "load_in_8bit": false,
34
+ "quant_method": "bitsandbytes"
35
+ },
36
+ "rms_norm_eps": 1e-06,
37
+ "rope_scaling": null,
38
+ "rope_theta": 10000.0,
39
+ "tie_word_embeddings": false,
40
+ "torch_dtype": "float16",
41
+ "transformers_version": "4.40.0.dev0",
42
+ "use_cache": false,
43
+ "vocab_size": 32000
44
+ }
runs/Apr10_21-24-13_89272c674ade/events.out.tfevents.1712784253.89272c674ade.474.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:393343d1dd6e7eb4beef31671e52c92c348a3b20cbf9166832f5796c154b3d7f
3
+ size 96859
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
+ }