Hugging Face's logo

CognitiveLab
/
Ambari-One-Thrid-PEFT_final

PEFT
Safetensors
llama
Generated from Trainer
8-bit precision
bitsandbytes
Model card Files Community
AdithyaSK commited on
Commit
d32e812
•
1 Parent(s): fea3b38

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +144 -0
  2. adapter_config.json +33 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-534/README.md +202 -0
  5. checkpoint-534/adapter_config.json +33 -0
  6. checkpoint-534/adapter_model.safetensors +3 -0
  7. checkpoint-534/optimizer.pt +3 -0
  8. checkpoint-534/rng_state_0.pth +3 -0
  9. checkpoint-534/rng_state_1.pth +3 -0
  10. checkpoint-534/scheduler.pt +3 -0
  11. checkpoint-534/special_tokens_map.json +24 -0
  12. checkpoint-534/tokenizer.model +3 -0
  13. checkpoint-534/tokenizer_config.json +44 -0
  14. checkpoint-534/trainer_state.json +3759 -0
  15. checkpoint-534/training_args.bin +3 -0
  16. checkpoint-623/README.md +202 -0
  17. checkpoint-623/adapter_config.json +33 -0
  18. checkpoint-623/adapter_model.safetensors +3 -0
  19. checkpoint-623/optimizer.pt +3 -0
  20. checkpoint-623/rng_state_0.pth +3 -0
  21. checkpoint-623/rng_state_1.pth +3 -0
  22. checkpoint-623/scheduler.pt +3 -0
  23. checkpoint-623/special_tokens_map.json +24 -0
  24. checkpoint-623/tokenizer.model +3 -0
  25. checkpoint-623/tokenizer_config.json +44 -0
  26. checkpoint-623/trainer_state.json +0 -0
  27. checkpoint-623/training_args.bin +3 -0
  28. checkpoint-712/README.md +202 -0
  29. checkpoint-712/adapter_config.json +33 -0
  30. checkpoint-712/adapter_model.safetensors +3 -0
  31. checkpoint-712/optimizer.pt +3 -0
  32. checkpoint-712/rng_state_0.pth +3 -0
  33. checkpoint-712/rng_state_1.pth +3 -0
  34. checkpoint-712/scheduler.pt +3 -0
  35. checkpoint-712/special_tokens_map.json +24 -0
  36. checkpoint-712/tokenizer.model +3 -0
  37. checkpoint-712/tokenizer_config.json +44 -0
  38. checkpoint-712/trainer_state.json +0 -0
  39. checkpoint-712/training_args.bin +3 -0
  40. checkpoint-801/README.md +202 -0
  41. checkpoint-801/adapter_config.json +33 -0
  42. checkpoint-801/adapter_model.safetensors +3 -0
  43. checkpoint-801/optimizer.pt +3 -0
  44. checkpoint-801/rng_state_0.pth +3 -0
  45. checkpoint-801/rng_state_1.pth +3 -0
  46. checkpoint-801/scheduler.pt +3 -0
  47. checkpoint-801/special_tokens_map.json +24 -0
  48. checkpoint-801/tokenizer.model +3 -0
  49. checkpoint-801/tokenizer_config.json +44 -0
  50. checkpoint-801/trainer_state.json +0 -0
README.md ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ tags:
4
+ - generated_from_trainer
5
+ base_model: meta-llama/Llama-2-7b-hf
6
+ model-index:
7
+ - name: kannada_out_peft_instruct_final_no_ext
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
+ should probably proofread and complete it, then remove this comment. -->
13
+
14
+ [<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)
15
+ <details><summary>See axolotl config</summary>
16
+
17
+ axolotl version: `0.4.0`
18
+ ```yaml
19
+ base_model: meta-llama/Llama-2-7b-hf
20
+ model_type: LlamaForCausalLM
21
+ tokenizer_type: LlamaTokenizer
22
+ tokenizer_config: meta-llama/Llama-2-7b-hf
23
+ is_llama_derived_model: true
24
+
25
+ load_in_8bit: true
26
+ load_in_4bit: false
27
+ strict: false
28
+
29
+ datasets:
30
+ - path: Cognitive-Lab/Kannada_Bilingual_Instruct
31
+ type: completion
32
+ dataset_prepared_path:
33
+ val_set_size: 0.05
34
+ output_dir: ./kannada_out_peft_instruct_final_no_ext
35
+
36
+ sequence_len: 4096
37
+ sample_packing: true
38
+ pad_to_sequence_len: true
39
+
40
+ adapter: lora
41
+ lora_model_dir:
42
+ lora_r: 256
43
+ lora_alpha: 128
44
+ lora_dropout: 0.05
45
+ lora_target_linear: true
46
+ lora_fan_in_fan_out:
47
+
48
+ wandb_project: Ambari-Instruct-No-Extension
49
+ wandb_entity:
50
+ wandb_watch:
51
+ wandb_name:
52
+ wandb_log_model:
53
+
54
+ gradient_accumulation_steps: 4
55
+ micro_batch_size: 14
56
+ num_epochs: 1
57
+ optimizer: adamw_bnb_8bit
58
+ lr_scheduler: cosine
59
+ learning_rate: 0.0002
60
+
61
+ train_on_inputs: false
62
+ group_by_length: false
63
+ bf16: true
64
+ fp16: false
65
+ tf32: false
66
+
67
+ gradient_checkpointing: true
68
+ early_stopping_patience:
69
+ resume_from_checkpoint:
70
+ local_rank:
71
+ logging_steps: 1
72
+ xformers_attention:
73
+ flash_attention: true
74
+
75
+ warmup_steps: 10
76
+ evals_per_epoch: 1
77
+ eval_table_size:
78
+ eval_table_max_new_tokens: 128
79
+ saves_per_epoch: 10
80
+ debug:
81
+ # deepspeed: deepspeed/zero2.json
82
+ weight_decay: 0.0
83
+ fsdp:
84
+ fsdp_config:
85
+ special_tokens:
86
+ bos_token: "<s>"
87
+ eos_token: "</s>"
88
+ unk_token: "<unk>"
89
+
90
+ ```
91
+
92
+ </details><br>
93
+
94
+ # kannada_out_peft_instruct_final_no_ext
95
+
96
+ This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the None dataset.
97
+ It achieves the following results on the evaluation set:
98
+ - Loss: 0.4334
99
+
100
+ ## Model description
101
+
102
+ More information needed
103
+
104
+ ## Intended uses & limitations
105
+
106
+ More information needed
107
+
108
+ ## Training and evaluation data
109
+
110
+ More information needed
111
+
112
+ ## Training procedure
113
+
114
+ ### Training hyperparameters
115
+
116
+ The following hyperparameters were used during training:
117
+ - learning_rate: 0.0002
118
+ - train_batch_size: 14
119
+ - eval_batch_size: 14
120
+ - seed: 42
121
+ - distributed_type: multi-GPU
122
+ - num_devices: 2
123
+ - gradient_accumulation_steps: 4
124
+ - total_train_batch_size: 112
125
+ - total_eval_batch_size: 28
126
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
127
+ - lr_scheduler_type: cosine
128
+ - lr_scheduler_warmup_steps: 10
129
+ - num_epochs: 1
130
+
131
+ ### Training results
132
+
133
+ | Training Loss | Epoch | Step | Validation Loss |
134
+ |:-------------:|:-----:|:----:|:---------------:|
135
+ | 0.3109 | 1.0 | 888 | 0.4334 |
136
+
137
+
138
+ ### Framework versions
139
+
140
+ - PEFT 0.9.0
141
+ - Transformers 4.40.0.dev0
142
+ - Pytorch 2.1.1+cu121
143
+ - Datasets 2.18.0
144
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 128,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 256,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "q_proj",
24
+ "o_proj",
25
+ "up_proj",
26
+ "v_proj",
27
+ "gate_proj",
28
+ "down_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a7c59a78c24b6a39425411326714edd9c4e6c1eac4e85cf0030d9bb774e8957
3
+ size 2558688074
checkpoint-534/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: meta-llama/Llama-2-7b-hf
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.9.0
checkpoint-534/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 128,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 256,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "q_proj",
24
+ "o_proj",
25
+ "up_proj",
26
+ "v_proj",
27
+ "gate_proj",
28
+ "down_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
checkpoint-534/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0bd312f5de575ec68662b3da82a89c318d7f5f2501b2de401d1ff9856cff94ae
3
+ size 2558587064
checkpoint-534/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34bf0232283b939f747bc4567d1dd0c6aa4be2e7dcc8af0ffcd9772e621bbb2c
3
+ size 1282290004
checkpoint-534/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb74977da5b3ca87a7a46b746c002bd459658ee940d8bbab4f55ce7b2a48504d
3
+ size 14512
checkpoint-534/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b483961abd4c4ee6b4eda8bb421ad8e6fa5d5d34ed139fb1b8d68df050a900f0
3
+ size 14512
checkpoint-534/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d81533da857ca6d796428267f511bb800d5c318aa7be539d44ce5728a76bd07
3
+ size 1064
checkpoint-534/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-534/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-534/tokenizer_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": false,
35
+ "model_max_length": 1000000000000000019884624838656,
36
+ "pad_token": "</s>",
37
+ "padding_side": "right",
38
+ "sp_model_kwargs": {},
39
+ "spaces_between_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false,
43
+ "use_fast": true
44
+ }
checkpoint-534/trainer_state.json ADDED
@@ -0,0 +1,3759 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.60084388185654,
5
+ "eval_steps": 500,
6
+ "global_step": 534,
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": 0.33328112959861755,
14
+ "learning_rate": 2e-05,
15
+ "loss": 0.9029,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0,
20
+ "grad_norm": 0.341234415769577,
21
+ "learning_rate": 4e-05,
22
+ "loss": 0.9276,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.0,
27
+ "grad_norm": 0.3182106614112854,
28
+ "learning_rate": 6e-05,
29
+ "loss": 0.9148,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.0,
34
+ "grad_norm": 0.22865141928195953,
35
+ "learning_rate": 8e-05,
36
+ "loss": 0.8421,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.01,
41
+ "grad_norm": 0.29670122265815735,
42
+ "learning_rate": 0.0001,
43
+ "loss": 0.7984,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.01,
48
+ "grad_norm": 0.28762516379356384,
49
+ "learning_rate": 0.00012,
50
+ "loss": 0.7625,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.01,
55
+ "grad_norm": 0.23907965421676636,
56
+ "learning_rate": 0.00014,
57
+ "loss": 0.7226,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.01,
62
+ "grad_norm": 0.19532263278961182,
63
+ "learning_rate": 0.00016,
64
+ "loss": 0.7043,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.01,
69
+ "grad_norm": 0.1425202488899231,
70
+ "learning_rate": 0.00018,
71
+ "loss": 0.6784,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.01,
76
+ "grad_norm": 0.10882167518138885,
77
+ "learning_rate": 0.0002,
78
+ "loss": 0.6579,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.01,
83
+ "grad_norm": 0.11275648325681686,
84
+ "learning_rate": 0.00019999935985220405,
85
+ "loss": 0.6592,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.01,
90
+ "grad_norm": 0.10155748575925827,
91
+ "learning_rate": 0.00019999743941701188,
92
+ "loss": 0.6554,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.01,
97
+ "grad_norm": 0.0845816433429718,
98
+ "learning_rate": 0.0001999942387190108,
99
+ "loss": 0.6513,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.02,
104
+ "grad_norm": 0.09046202898025513,
105
+ "learning_rate": 0.0001999897577991792,
106
+ "loss": 0.6267,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.02,
111
+ "grad_norm": 0.08296829462051392,
112
+ "learning_rate": 0.00019998399671488612,
113
+ "loss": 0.6434,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.02,
118
+ "grad_norm": 0.07594181597232819,
119
+ "learning_rate": 0.00019997695553989042,
120
+ "loss": 0.6096,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.02,
125
+ "grad_norm": 0.0683172270655632,
126
+ "learning_rate": 0.00019996863436433997,
127
+ "loss": 0.6143,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.02,
132
+ "grad_norm": 0.05627186596393585,
133
+ "learning_rate": 0.0001999590332947704,
134
+ "loss": 0.6024,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.02,
139
+ "grad_norm": 0.05644279345870018,
140
+ "learning_rate": 0.00019994815245410384,
141
+ "loss": 0.595,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.02,
146
+ "grad_norm": 0.05661479011178017,
147
+ "learning_rate": 0.00019993599198164715,
148
+ "loss": 0.5759,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.02,
153
+ "grad_norm": 0.04077177122235298,
154
+ "learning_rate": 0.00019992255203309033,
155
+ "loss": 0.582,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.02,
160
+ "grad_norm": 0.04510512948036194,
161
+ "learning_rate": 0.00019990783278050448,
162
+ "loss": 0.5751,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.03,
167
+ "grad_norm": 0.0470162108540535,
168
+ "learning_rate": 0.00019989183441233952,
169
+ "loss": 0.5716,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.03,
174
+ "grad_norm": 0.04402562975883484,
175
+ "learning_rate": 0.00019987455713342187,
176
+ "loss": 0.564,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.03,
181
+ "grad_norm": 0.045594893395900726,
182
+ "learning_rate": 0.00019985600116495173,
183
+ "loss": 0.5657,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.03,
188
+ "grad_norm": 0.037670280784368515,
189
+ "learning_rate": 0.0001998361667445004,
190
+ "loss": 0.5619,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.03,
195
+ "grad_norm": 0.034366946667432785,
196
+ "learning_rate": 0.00019981505412600706,
197
+ "loss": 0.554,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 0.03,
202
+ "grad_norm": 0.044084370136260986,
203
+ "learning_rate": 0.00019979266357977564,
204
+ "loss": 0.5527,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 0.03,
209
+ "grad_norm": 0.04839107394218445,
210
+ "learning_rate": 0.0001997689953924713,
211
+ "loss": 0.5534,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 0.03,
216
+ "grad_norm": 0.04943666234612465,
217
+ "learning_rate": 0.0001997440498671168,
218
+ "loss": 0.5357,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 0.03,
223
+ "grad_norm": 0.05644814297556877,
224
+ "learning_rate": 0.00019971782732308867,
225
+ "loss": 0.5388,
226
+ "step": 31
227
+ },
228
+ {
229
+ "epoch": 0.04,
230
+ "grad_norm": 0.05796538665890694,
231
+ "learning_rate": 0.00019969032809611287,
232
+ "loss": 0.5327,
233
+ "step": 32
234
+ },
235
+ {
236
+ "epoch": 0.04,
237
+ "grad_norm": 0.05399211496114731,
238
+ "learning_rate": 0.0001996615525382609,
239
+ "loss": 0.5447,
240
+ "step": 33
241
+ },
242
+ {
243
+ "epoch": 0.04,
244
+ "grad_norm": 0.03574652597308159,
245
+ "learning_rate": 0.0001996315010179449,
246
+ "loss": 0.5213,
247
+ "step": 34
248
+ },
249
+ {
250
+ "epoch": 0.04,
251
+ "grad_norm": 0.04394914582371712,
252
+ "learning_rate": 0.00019960017391991314,
253
+ "loss": 0.5247,
254
+ "step": 35
255
+ },
256
+ {
257
+ "epoch": 0.04,
258
+ "grad_norm": 0.05197073519229889,
259
+ "learning_rate": 0.00019956757164524516,
260
+ "loss": 0.5253,
261
+ "step": 36
262
+ },
263
+ {
264
+ "epoch": 0.04,
265
+ "grad_norm": 0.05605654790997505,
266
+ "learning_rate": 0.00019953369461134634,
267
+ "loss": 0.5289,
268
+ "step": 37
269
+ },
270
+ {
271
+ "epoch": 0.04,
272
+ "grad_norm": 0.05167644843459129,
273
+ "learning_rate": 0.00019949854325194294,
274
+ "loss": 0.5223,
275
+ "step": 38
276
+ },
277
+ {
278
+ "epoch": 0.04,
279
+ "grad_norm": 0.04118728265166283,
280
+ "learning_rate": 0.0001994621180170762,
281
+ "loss": 0.52,
282
+ "step": 39
283
+ },
284
+ {
285
+ "epoch": 0.05,
286
+ "grad_norm": 0.0373263917863369,
287
+ "learning_rate": 0.00019942441937309684,
288
+ "loss": 0.5076,
289
+ "step": 40
290
+ },
291
+ {
292
+ "epoch": 0.05,
293
+ "grad_norm": 0.04870600998401642,
294
+ "learning_rate": 0.0001993854478026589,
295
+ "loss": 0.5287,
296
+ "step": 41
297
+ },
298
+ {
299
+ "epoch": 0.05,
300
+ "grad_norm": 0.05014103651046753,
301
+ "learning_rate": 0.00019934520380471372,
302
+ "loss": 0.5048,
303
+ "step": 42
304
+ },
305
+ {
306
+ "epoch": 0.05,
307
+ "grad_norm": 0.04807833209633827,
308
+ "learning_rate": 0.0001993036878945034,
309
+ "loss": 0.5075,
310
+ "step": 43
311
+ },
312
+ {
313
+ "epoch": 0.05,
314
+ "grad_norm": 0.040045421570539474,
315
+ "learning_rate": 0.0001992609006035543,
316
+ "loss": 0.4954,
317
+ "step": 44
318
+ },
319
+ {
320
+ "epoch": 0.05,
321
+ "grad_norm": 0.04202349856495857,
322
+ "learning_rate": 0.00019921684247967028,
323
+ "loss": 0.4953,
324
+ "step": 45
325
+ },
326
+ {
327
+ "epoch": 0.05,
328
+ "grad_norm": 0.041629109531641006,
329
+ "learning_rate": 0.0001991715140869255,
330
+ "loss": 0.501,
331
+ "step": 46
332
+ },
333
+ {
334
+ "epoch": 0.05,
335
+ "grad_norm": 0.04767894372344017,
336
+ "learning_rate": 0.0001991249160056574,
337
+ "loss": 0.4878,
338
+ "step": 47
339
+ },
340
+ {
341
+ "epoch": 0.05,
342
+ "grad_norm": 0.05395263060927391,
343
+ "learning_rate": 0.00019907704883245916,
344
+ "loss": 0.5014,
345
+ "step": 48
346
+ },
347
+ {
348
+ "epoch": 0.06,
349
+ "grad_norm": 0.06040235981345177,
350
+ "learning_rate": 0.00019902791318017205,
351
+ "loss": 0.5043,
352
+ "step": 49
353
+ },
354
+ {
355
+ "epoch": 0.06,
356
+ "grad_norm": 0.06749273091554642,
357
+ "learning_rate": 0.0001989775096778777,
358
+ "loss": 0.4931,
359
+ "step": 50
360
+ },
361
+ {
362
+ "epoch": 0.06,
363
+ "grad_norm": 0.06482889503240585,
364
+ "learning_rate": 0.00019892583897088994,
365
+ "loss": 0.4869,
366
+ "step": 51
367
+ },
368
+ {
369
+ "epoch": 0.06,
370
+ "grad_norm": 0.045358914881944656,
371
+ "learning_rate": 0.0001988729017207465,
372
+ "loss": 0.479,
373
+ "step": 52
374
+ },
375
+ {
376
+ "epoch": 0.06,
377
+ "grad_norm": 0.043360061943531036,
378
+ "learning_rate": 0.00019881869860520073,
379
+ "loss": 0.4953,
380
+ "step": 53
381
+ },
382
+ {
383
+ "epoch": 0.06,
384
+ "grad_norm": 0.060205183923244476,
385
+ "learning_rate": 0.00019876323031821266,
386
+ "loss": 0.4705,
387
+ "step": 54
388
+ },
389
+ {
390
+ "epoch": 0.06,
391
+ "grad_norm": 0.05729120969772339,
392
+ "learning_rate": 0.00019870649756994037,
393
+ "loss": 0.4887,
394
+ "step": 55
395
+ },
396
+ {
397
+ "epoch": 0.06,
398
+ "grad_norm": 0.03843148052692413,
399
+ "learning_rate": 0.00019864850108673073,
400
+ "loss": 0.4737,
401
+ "step": 56
402
+ },
403
+ {
404
+ "epoch": 0.06,
405
+ "grad_norm": 0.053673889487981796,
406
+ "learning_rate": 0.00019858924161111015,
407
+ "loss": 0.4817,
408
+ "step": 57
409
+ },
410
+ {
411
+ "epoch": 0.07,
412
+ "grad_norm": 0.05148368701338768,
413
+ "learning_rate": 0.00019852871990177503,
414
+ "loss": 0.4763,
415
+ "step": 58
416
+ },
417
+ {
418
+ "epoch": 0.07,
419
+ "grad_norm": 0.05371672287583351,
420
+ "learning_rate": 0.00019846693673358226,
421
+ "loss": 0.4751,
422
+ "step": 59
423
+ },
424
+ {
425
+ "epoch": 0.07,
426
+ "grad_norm": 0.05490916967391968,
427
+ "learning_rate": 0.00019840389289753896,
428
+ "loss": 0.457,
429
+ "step": 60
430
+ },
431
+ {
432
+ "epoch": 0.07,
433
+ "grad_norm": 0.04629400372505188,
434
+ "learning_rate": 0.00019833958920079255,
435
+ "loss": 0.4692,
436
+ "step": 61
437
+ },
438
+ {
439
+ "epoch": 0.07,
440
+ "grad_norm": 0.051137253642082214,
441
+ "learning_rate": 0.00019827402646662047,
442
+ "loss": 0.4614,
443
+ "step": 62
444
+ },
445
+ {
446
+ "epoch": 0.07,
447
+ "grad_norm": 0.051790811121463776,
448
+ "learning_rate": 0.0001982072055344195,
449
+ "loss": 0.4594,
450
+ "step": 63
451
+ },
452
+ {
453
+ "epoch": 0.07,
454
+ "grad_norm": 0.0445956289768219,
455
+ "learning_rate": 0.00019813912725969509,
456
+ "loss": 0.4601,
457
+ "step": 64
458
+ },
459
+ {
460
+ "epoch": 0.07,
461
+ "grad_norm": 0.04766576737165451,
462
+ "learning_rate": 0.0001980697925140504,
463
+ "loss": 0.4631,
464
+ "step": 65
465
+ },
466
+ {
467
+ "epoch": 0.07,
468
+ "grad_norm": 0.04839074984192848,
469
+ "learning_rate": 0.0001979992021851751,
470
+ "loss": 0.4605,
471
+ "step": 66
472
+ },
473
+ {
474
+ "epoch": 0.08,
475
+ "grad_norm": 0.04736727103590965,
476
+ "learning_rate": 0.0001979273571768341,
477
+ "loss": 0.4617,
478
+ "step": 67
479
+ },
480
+ {
481
+ "epoch": 0.08,
482
+ "grad_norm": 0.057293377816677094,
483
+ "learning_rate": 0.0001978542584088558,
484
+ "loss": 0.4621,
485
+ "step": 68
486
+ },
487
+ {
488
+ "epoch": 0.08,
489
+ "grad_norm": 0.05025665834546089,
490
+ "learning_rate": 0.0001977799068171206,
491
+ "loss": 0.4671,
492
+ "step": 69
493
+ },
494
+ {
495
+ "epoch": 0.08,
496
+ "grad_norm": 0.057366687804460526,
497
+ "learning_rate": 0.0001977043033535486,
498
+ "loss": 0.4521,
499
+ "step": 70
500
+ },
501
+ {
502
+ "epoch": 0.08,
503
+ "grad_norm": 0.07595837116241455,
504
+ "learning_rate": 0.00019762744898608762,
505
+ "loss": 0.4671,
506
+ "step": 71
507
+ },
508
+ {
509
+ "epoch": 0.08,
510
+ "grad_norm": 0.07574213296175003,
511
+ "learning_rate": 0.0001975493446987007,
512
+ "loss": 0.4664,
513
+ "step": 72
514
+ },
515
+ {
516
+ "epoch": 0.08,
517
+ "grad_norm": 0.06472938507795334,
518
+ "learning_rate": 0.00019746999149135362,
519
+ "loss": 0.456,
520
+ "step": 73
521
+ },
522
+ {
523
+ "epoch": 0.08,
524
+ "grad_norm": 0.05983012542128563,
525
+ "learning_rate": 0.00019738939038000205,
526
+ "loss": 0.4459,
527
+ "step": 74
528
+ },
529
+ {
530
+ "epoch": 0.08,
531
+ "grad_norm": 0.05136057734489441,
532
+ "learning_rate": 0.00019730754239657842,
533
+ "loss": 0.4486,
534
+ "step": 75
535
+ },
536
+ {
537
+ "epoch": 0.09,
538
+ "grad_norm": 0.06191498041152954,
539
+ "learning_rate": 0.00019722444858897878,
540
+ "loss": 0.4424,
541
+ "step": 76
542
+ },
543
+ {
544
+ "epoch": 0.09,
545
+ "grad_norm": 0.06742191314697266,
546
+ "learning_rate": 0.0001971401100210496,
547
+ "loss": 0.458,
548
+ "step": 77
549
+ },
550
+ {
551
+ "epoch": 0.09,
552
+ "grad_norm": 0.06019548326730728,
553
+ "learning_rate": 0.00019705452777257377,
554
+ "loss": 0.4423,
555
+ "step": 78
556
+ },
557
+ {
558
+ "epoch": 0.09,
559
+ "grad_norm": 0.05012982338666916,
560
+ "learning_rate": 0.0001969677029392571,
561
+ "loss": 0.4466,
562
+ "step": 79
563
+ },
564
+ {
565
+ "epoch": 0.09,
566
+ "grad_norm": 0.0552060566842556,
567
+ "learning_rate": 0.00019687963663271409,
568
+ "loss": 0.4534,
569
+ "step": 80
570
+ },
571
+ {
572
+ "epoch": 0.09,
573
+ "grad_norm": 0.05883748456835747,
574
+ "learning_rate": 0.00019679032998045376,
575
+ "loss": 0.4409,
576
+ "step": 81
577
+ },
578
+ {
579
+ "epoch": 0.09,
580
+ "grad_norm": 0.07146705687046051,
581
+ "learning_rate": 0.00019669978412586528,
582
+ "loss": 0.4582,
583
+ "step": 82
584
+ },
585
+ {
586
+ "epoch": 0.09,
587
+ "grad_norm": 0.054095230996608734,
588
+ "learning_rate": 0.00019660800022820317,
589
+ "loss": 0.4487,
590
+ "step": 83
591
+ },
592
+ {
593
+ "epoch": 0.09,
594
+ "grad_norm": 0.04927053675055504,
595
+ "learning_rate": 0.00019651497946257266,
596
+ "loss": 0.4429,
597
+ "step": 84
598
+ },
599
+ {
600
+ "epoch": 0.1,
601
+ "grad_norm": 0.06037526577711105,
602
+ "learning_rate": 0.00019642072301991455,
603
+ "loss": 0.4456,
604
+ "step": 85
605
+ },
606
+ {
607
+ "epoch": 0.1,
608
+ "grad_norm": 0.05555957555770874,
609
+ "learning_rate": 0.00019632523210698987,
610
+ "loss": 0.4382,
611
+ "step": 86
612
+ },
613
+ {
614
+ "epoch": 0.1,
615
+ "grad_norm": 0.04606284573674202,
616
+ "learning_rate": 0.00019622850794636455,
617
+ "loss": 0.4411,
618
+ "step": 87
619
+ },
620
+ {
621
+ "epoch": 0.1,
622
+ "grad_norm": 0.04605920985341072,
623
+ "learning_rate": 0.00019613055177639384,
624
+ "loss": 0.4326,
625
+ "step": 88
626
+ },
627
+ {
628
+ "epoch": 0.1,
629
+ "grad_norm": 0.050325632095336914,
630
+ "learning_rate": 0.0001960313648512062,
631
+ "loss": 0.4338,
632
+ "step": 89
633
+ },
634
+ {
635
+ "epoch": 0.1,
636
+ "grad_norm": 0.04921424016356468,
637
+ "learning_rate": 0.00019593094844068748,
638
+ "loss": 0.4316,
639
+ "step": 90
640
+ },
641
+ {
642
+ "epoch": 0.1,
643
+ "grad_norm": 0.04333706200122833,
644
+ "learning_rate": 0.00019582930383046457,
645
+ "loss": 0.4441,
646
+ "step": 91
647
+ },
648
+ {
649
+ "epoch": 0.1,
650
+ "grad_norm": 0.048454612493515015,
651
+ "learning_rate": 0.0001957264323218889,
652
+ "loss": 0.4382,
653
+ "step": 92
654
+ },
655
+ {
656
+ "epoch": 0.1,
657
+ "grad_norm": 0.0541059784591198,
658
+ "learning_rate": 0.00019562233523201986,
659
+ "loss": 0.4328,
660
+ "step": 93
661
+ },
662
+ {
663
+ "epoch": 0.11,
664
+ "grad_norm": 0.043696511536836624,
665
+ "learning_rate": 0.00019551701389360795,
666
+ "loss": 0.4335,
667
+ "step": 94
668
+ },
669
+ {
670
+ "epoch": 0.11,
671
+ "grad_norm": 0.04407835751771927,
672
+ "learning_rate": 0.00019541046965507758,
673
+ "loss": 0.4327,
674
+ "step": 95
675
+ },
676
+ {
677
+ "epoch": 0.11,
678
+ "grad_norm": 0.05477238819003105,
679
+ "learning_rate": 0.00019530270388050998,
680
+ "loss": 0.4294,
681
+ "step": 96
682
+ },
683
+ {
684
+ "epoch": 0.11,
685
+ "grad_norm": 0.05609311908483505,
686
+ "learning_rate": 0.00019519371794962556,
687
+ "loss": 0.4305,
688
+ "step": 97
689
+ },
690
+ {
691
+ "epoch": 0.11,
692
+ "grad_norm": 0.045145273208618164,
693
+ "learning_rate": 0.00019508351325776642,
694
+ "loss": 0.4395,
695
+ "step": 98
696
+ },
697
+ {
698
+ "epoch": 0.11,
699
+ "grad_norm": 0.04475285857915878,
700
+ "learning_rate": 0.00019497209121587837,
701
+ "loss": 0.4284,
702
+ "step": 99
703
+ },
704
+ {
705
+ "epoch": 0.11,
706
+ "grad_norm": 0.04405711591243744,
707
+ "learning_rate": 0.00019485945325049288,
708
+ "loss": 0.4214,
709
+ "step": 100
710
+ },
711
+ {
712
+ "epoch": 0.11,
713
+ "grad_norm": 0.04461454227566719,
714
+ "learning_rate": 0.0001947456008037089,
715
+ "loss": 0.4154,
716
+ "step": 101
717
+ },
718
+ {
719
+ "epoch": 0.11,
720
+ "grad_norm": 0.04791221395134926,
721
+ "learning_rate": 0.00019463053533317425,
722
+ "loss": 0.4248,
723
+ "step": 102
724
+ },
725
+ {
726
+ "epoch": 0.12,
727
+ "grad_norm": 0.05543987452983856,
728
+ "learning_rate": 0.00019451425831206706,
729
+ "loss": 0.4303,
730
+ "step": 103
731
+ },
732
+ {
733
+ "epoch": 0.12,
734
+ "grad_norm": 0.06330578774213791,
735
+ "learning_rate": 0.00019439677122907697,
736
+ "loss": 0.4274,
737
+ "step": 104
738
+ },
739
+ {
740
+ "epoch": 0.12,
741
+ "grad_norm": 0.05569112300872803,
742
+ "learning_rate": 0.00019427807558838588,
743
+ "loss": 0.4234,
744
+ "step": 105
745
+ },
746
+ {
747
+ "epoch": 0.12,
748
+ "grad_norm": 0.047680530697107315,
749
+ "learning_rate": 0.00019415817290964883,
750
+ "loss": 0.4155,
751
+ "step": 106
752
+ },
753
+ {
754
+ "epoch": 0.12,
755
+ "grad_norm": 0.05214262008666992,
756
+ "learning_rate": 0.0001940370647279746,
757
+ "loss": 0.4224,
758
+ "step": 107
759
+ },
760
+ {
761
+ "epoch": 0.12,
762
+ "grad_norm": 0.06332990527153015,
763
+ "learning_rate": 0.00019391475259390584,
764
+ "loss": 0.4233,
765
+ "step": 108
766
+ },
767
+ {
768
+ "epoch": 0.12,
769
+ "grad_norm": 0.05726313218474388,
770
+ "learning_rate": 0.00019379123807339942,
771
+ "loss": 0.4118,
772
+ "step": 109
773
+ },
774
+ {
775
+ "epoch": 0.12,
776
+ "grad_norm": 0.044936031103134155,
777
+ "learning_rate": 0.00019366652274780628,
778
+ "loss": 0.4296,
779
+ "step": 110
780
+ },
781
+ {
782
+ "epoch": 0.12,
783
+ "grad_norm": 0.05117325484752655,
784
+ "learning_rate": 0.0001935406082138513,
785
+ "loss": 0.4287,
786
+ "step": 111
787
+ },
788
+ {
789
+ "epoch": 0.13,
790
+ "grad_norm": 0.058542776852846146,
791
+ "learning_rate": 0.00019341349608361267,
792
+ "loss": 0.4213,
793
+ "step": 112
794
+ },
795
+ {
796
+ "epoch": 0.13,
797
+ "grad_norm": 0.056066304445266724,
798
+ "learning_rate": 0.00019328518798450138,
799
+ "loss": 0.4174,
800
+ "step": 113
801
+ },
802
+ {
803
+ "epoch": 0.13,
804
+ "grad_norm": 0.049762677401304245,
805
+ "learning_rate": 0.00019315568555924035,
806
+ "loss": 0.418,
807
+ "step": 114
808
+ },
809
+ {
810
+ "epoch": 0.13,
811
+ "grad_norm": 0.043821126222610474,
812
+ "learning_rate": 0.00019302499046584348,
813
+ "loss": 0.4012,
814
+ "step": 115
815
+ },
816
+ {
817
+ "epoch": 0.13,
818
+ "grad_norm": 0.05036221817135811,
819
+ "learning_rate": 0.00019289310437759427,
820
+ "loss": 0.4237,
821
+ "step": 116
822
+ },
823
+ {
824
+ "epoch": 0.13,
825
+ "grad_norm": 0.050889529287815094,
826
+ "learning_rate": 0.00019276002898302447,
827
+ "loss": 0.4144,
828
+ "step": 117
829
+ },
830
+ {
831
+ "epoch": 0.13,
832
+ "grad_norm": 0.04269757494330406,
833
+ "learning_rate": 0.0001926257659858925,
834
+ "loss": 0.4078,
835
+ "step": 118
836
+ },
837
+ {
838
+ "epoch": 0.13,
839
+ "grad_norm": 0.04927165433764458,
840
+ "learning_rate": 0.00019249031710516162,
841
+ "loss": 0.4155,
842
+ "step": 119
843
+ },
844
+ {
845
+ "epoch": 0.14,
846
+ "grad_norm": 0.05124311521649361,
847
+ "learning_rate": 0.00019235368407497788,
848
+ "loss": 0.3966,
849
+ "step": 120
850
+ },
851
+ {
852
+ "epoch": 0.14,
853
+ "grad_norm": 0.04073040187358856,
854
+ "learning_rate": 0.00019221586864464786,
855
+ "loss": 0.4064,
856
+ "step": 121
857
+ },
858
+ {
859
+ "epoch": 0.14,
860
+ "grad_norm": 0.04988453537225723,
861
+ "learning_rate": 0.00019207687257861655,
862
+ "loss": 0.4197,
863
+ "step": 122
864
+ },
865
+ {
866
+ "epoch": 0.14,
867
+ "grad_norm": 0.05227258801460266,
868
+ "learning_rate": 0.0001919366976564444,
869
+ "loss": 0.414,
870
+ "step": 123
871
+ },
872
+ {
873
+ "epoch": 0.14,
874
+ "grad_norm": 0.0466819666326046,
875
+ "learning_rate": 0.00019179534567278475,
876
+ "loss": 0.4173,
877
+ "step": 124
878
+ },
879
+ {
880
+ "epoch": 0.14,
881
+ "grad_norm": 0.047633491456508636,
882
+ "learning_rate": 0.00019165281843736085,
883
+ "loss": 0.4085,
884
+ "step": 125
885
+ },
886
+ {
887
+ "epoch": 0.14,
888
+ "grad_norm": 0.05280464142560959,
889
+ "learning_rate": 0.00019150911777494258,
890
+ "loss": 0.4051,
891
+ "step": 126
892
+ },
893
+ {
894
+ "epoch": 0.14,
895
+ "grad_norm": 0.052302148193120956,
896
+ "learning_rate": 0.00019136424552532318,
897
+ "loss": 0.42,
898
+ "step": 127
899
+ },
900
+ {
901
+ "epoch": 0.14,
902
+ "grad_norm": 0.04875241965055466,
903
+ "learning_rate": 0.00019121820354329577,
904
+ "loss": 0.4258,
905
+ "step": 128
906
+ },
907
+ {
908
+ "epoch": 0.15,
909
+ "grad_norm": 0.04654408246278763,
910
+ "learning_rate": 0.0001910709936986293,
911
+ "loss": 0.409,
912
+ "step": 129
913
+ },
914
+ {
915
+ "epoch": 0.15,
916
+ "grad_norm": 0.05745020881295204,
917
+ "learning_rate": 0.00019092261787604492,
918
+ "loss": 0.4059,
919
+ "step": 130
920
+ },
921
+ {
922
+ "epoch": 0.15,
923
+ "grad_norm": 0.06945241987705231,
924
+ "learning_rate": 0.00019077307797519183,
925
+ "loss": 0.4038,
926
+ "step": 131
927
+ },
928
+ {
929
+ "epoch": 0.15,
930
+ "grad_norm": 0.06346461176872253,
931
+ "learning_rate": 0.00019062237591062272,
932
+ "loss": 0.4031,
933
+ "step": 132
934
+ },
935
+ {
936
+ "epoch": 0.15,
937
+ "grad_norm": 0.058026187121868134,
938
+ "learning_rate": 0.00019047051361176953,
939
+ "loss": 0.4126,
940
+ "step": 133
941
+ },
942
+ {
943
+ "epoch": 0.15,
944
+ "grad_norm": 0.04755179584026337,
945
+ "learning_rate": 0.0001903174930229185,
946
+ "loss": 0.4209,
947
+ "step": 134
948
+ },
949
+ {
950
+ "epoch": 0.15,
951
+ "grad_norm": 0.05765068158507347,
952
+ "learning_rate": 0.0001901633161031856,
953
+ "loss": 0.4067,
954
+ "step": 135
955
+ },
956
+ {
957
+ "epoch": 0.15,
958
+ "grad_norm": 0.05687811225652695,
959
+ "learning_rate": 0.000190007984826491,
960
+ "loss": 0.3975,
961
+ "step": 136
962
+ },
963
+ {
964
+ "epoch": 0.15,
965
+ "grad_norm": 0.04930473491549492,
966
+ "learning_rate": 0.0001898515011815343,
967
+ "loss": 0.4146,
968
+ "step": 137
969
+ },
970
+ {
971
+ "epoch": 0.16,
972
+ "grad_norm": 0.05147051811218262,
973
+ "learning_rate": 0.0001896938671717687,
974
+ "loss": 0.4035,
975
+ "step": 138
976
+ },
977
+ {
978
+ "epoch": 0.16,
979
+ "grad_norm": 0.05680418014526367,
980
+ "learning_rate": 0.0001895350848153754,
981
+ "loss": 0.4049,
982
+ "step": 139
983
+ },
984
+ {
985
+ "epoch": 0.16,
986
+ "grad_norm": 0.0444297268986702,
987
+ "learning_rate": 0.00018937515614523797,
988
+ "loss": 0.4085,
989
+ "step": 140
990
+ },
991
+ {
992
+ "epoch": 0.16,
993
+ "grad_norm": 0.05083802342414856,
994
+ "learning_rate": 0.00018921408320891612,
995
+ "loss": 0.4036,
996
+ "step": 141
997
+ },
998
+ {
999
+ "epoch": 0.16,
1000
+ "grad_norm": 0.04978756606578827,
1001
+ "learning_rate": 0.00018905186806861957,
1002
+ "loss": 0.4058,
1003
+ "step": 142
1004
+ },
1005
+ {
1006
+ "epoch": 0.16,
1007
+ "grad_norm": 0.04963681101799011,
1008
+ "learning_rate": 0.00018888851280118155,
1009
+ "loss": 0.3977,
1010
+ "step": 143
1011
+ },
1012
+ {
1013
+ "epoch": 0.16,
1014
+ "grad_norm": 0.0466095507144928,
1015
+ "learning_rate": 0.00018872401949803237,
1016
+ "loss": 0.3945,
1017
+ "step": 144
1018
+ },
1019
+ {
1020
+ "epoch": 0.16,
1021
+ "grad_norm": 0.04972768574953079,
1022
+ "learning_rate": 0.00018855839026517257,
1023
+ "loss": 0.4151,
1024
+ "step": 145
1025
+ },
1026
+ {
1027
+ "epoch": 0.16,
1028
+ "grad_norm": 0.054370637983083725,
1029
+ "learning_rate": 0.0001883916272231459,
1030
+ "loss": 0.3944,
1031
+ "step": 146
1032
+ },
1033
+ {
1034
+ "epoch": 0.17,
1035
+ "grad_norm": 0.054699357599020004,
1036
+ "learning_rate": 0.00018822373250701224,
1037
+ "loss": 0.3989,
1038
+ "step": 147
1039
+ },
1040
+ {
1041
+ "epoch": 0.17,
1042
+ "grad_norm": 0.054452769458293915,
1043
+ "learning_rate": 0.00018805470826632024,
1044
+ "loss": 0.3984,
1045
+ "step": 148
1046
+ },
1047
+ {
1048
+ "epoch": 0.17,
1049
+ "grad_norm": 0.04596908017992973,
1050
+ "learning_rate": 0.00018788455666507981,
1051
+ "loss": 0.4018,
1052
+ "step": 149
1053
+ },
1054
+ {
1055
+ "epoch": 0.17,
1056
+ "grad_norm": 0.054354868829250336,
1057
+ "learning_rate": 0.00018771327988173435,
1058
+ "loss": 0.3985,
1059
+ "step": 150
1060
+ },
1061
+ {
1062
+ "epoch": 0.17,
1063
+ "grad_norm": 0.05570242181420326,
1064
+ "learning_rate": 0.00018754088010913304,
1065
+ "loss": 0.3818,
1066
+ "step": 151
1067
+ },
1068
+ {
1069
+ "epoch": 0.17,
1070
+ "grad_norm": 0.054722048342227936,
1071
+ "learning_rate": 0.00018736735955450251,
1072
+ "loss": 0.4111,
1073
+ "step": 152
1074
+ },
1075
+ {
1076
+ "epoch": 0.17,
1077
+ "grad_norm": 0.04620000347495079,
1078
+ "learning_rate": 0.00018719272043941882,
1079
+ "loss": 0.3949,
1080
+ "step": 153
1081
+ },
1082
+ {
1083
+ "epoch": 0.17,
1084
+ "grad_norm": 0.048443205654621124,
1085
+ "learning_rate": 0.00018701696499977884,
1086
+ "loss": 0.3856,
1087
+ "step": 154
1088
+ },
1089
+ {
1090
+ "epoch": 0.17,
1091
+ "grad_norm": 0.06628945469856262,
1092
+ "learning_rate": 0.00018684009548577168,
1093
+ "loss": 0.4048,
1094
+ "step": 155
1095
+ },
1096
+ {
1097
+ "epoch": 0.18,
1098
+ "grad_norm": 0.05339967459440231,
1099
+ "learning_rate": 0.00018666211416184999,
1100
+ "loss": 0.3894,
1101
+ "step": 156
1102
+ },
1103
+ {
1104
+ "epoch": 0.18,
1105
+ "grad_norm": 0.04650304839015007,
1106
+ "learning_rate": 0.00018648302330670082,
1107
+ "loss": 0.4004,
1108
+ "step": 157
1109
+ },
1110
+ {
1111
+ "epoch": 0.18,
1112
+ "grad_norm": 0.05634591728448868,
1113
+ "learning_rate": 0.00018630282521321645,
1114
+ "loss": 0.4033,
1115
+ "step": 158
1116
+ },
1117
+ {
1118
+ "epoch": 0.18,
1119
+ "grad_norm": 0.048666685819625854,
1120
+ "learning_rate": 0.00018612152218846513,
1121
+ "loss": 0.399,
1122
+ "step": 159
1123
+ },
1124
+ {
1125
+ "epoch": 0.18,
1126
+ "grad_norm": 0.04597772657871246,
1127
+ "learning_rate": 0.0001859391165536615,
1128
+ "loss": 0.3931,
1129
+ "step": 160
1130
+ },
1131
+ {
1132
+ "epoch": 0.18,
1133
+ "grad_norm": 0.0526028610765934,
1134
+ "learning_rate": 0.00018575561064413689,
1135
+ "loss": 0.3879,
1136
+ "step": 161
1137
+ },
1138
+ {
1139
+ "epoch": 0.18,
1140
+ "grad_norm": 0.05867009237408638,
1141
+ "learning_rate": 0.00018557100680930937,
1142
+ "loss": 0.3905,
1143
+ "step": 162
1144
+ },
1145
+ {
1146
+ "epoch": 0.18,
1147
+ "grad_norm": 0.05077454075217247,
1148
+ "learning_rate": 0.00018538530741265364,
1149
+ "loss": 0.395,
1150
+ "step": 163
1151
+ },
1152
+ {
1153
+ "epoch": 0.18,
1154
+ "grad_norm": 0.0461389385163784,
1155
+ "learning_rate": 0.00018519851483167097,
1156
+ "loss": 0.4016,
1157
+ "step": 164
1158
+ },
1159
+ {
1160
+ "epoch": 0.19,
1161
+ "grad_norm": 0.059010252356529236,
1162
+ "learning_rate": 0.00018501063145785846,
1163
+ "loss": 0.3823,
1164
+ "step": 165
1165
+ },
1166
+ {
1167
+ "epoch": 0.19,
1168
+ "grad_norm": 0.06437338888645172,
1169
+ "learning_rate": 0.00018482165969667874,
1170
+ "loss": 0.3918,
1171
+ "step": 166
1172
+ },
1173
+ {
1174
+ "epoch": 0.19,
1175
+ "grad_norm": 0.04585932940244675,
1176
+ "learning_rate": 0.00018463160196752887,
1177
+ "loss": 0.3808,
1178
+ "step": 167
1179
+ },
1180
+ {
1181
+ "epoch": 0.19,
1182
+ "grad_norm": 0.05361521616578102,
1183
+ "learning_rate": 0.00018444046070370963,
1184
+ "loss": 0.3858,
1185
+ "step": 168
1186
+ },
1187
+ {
1188
+ "epoch": 0.19,
1189
+ "grad_norm": 0.05653822794556618,
1190
+ "learning_rate": 0.00018424823835239417,
1191
+ "loss": 0.3785,
1192
+ "step": 169
1193
+ },
1194
+ {
1195
+ "epoch": 0.19,
1196
+ "grad_norm": 0.04439689964056015,
1197
+ "learning_rate": 0.0001840549373745968,
1198
+ "loss": 0.3894,
1199
+ "step": 170
1200
+ },
1201
+ {
1202
+ "epoch": 0.19,
1203
+ "grad_norm": 0.05564529448747635,
1204
+ "learning_rate": 0.00018386056024514137,
1205
+ "loss": 0.3883,
1206
+ "step": 171
1207
+ },
1208
+ {
1209
+ "epoch": 0.19,
1210
+ "grad_norm": 0.06035888195037842,
1211
+ "learning_rate": 0.00018366510945262972,
1212
+ "loss": 0.3855,
1213
+ "step": 172
1214
+ },
1215
+ {
1216
+ "epoch": 0.19,
1217
+ "grad_norm": 0.044238511472940445,
1218
+ "learning_rate": 0.0001834685874994098,
1219
+ "loss": 0.3934,
1220
+ "step": 173
1221
+ },
1222
+ {
1223
+ "epoch": 0.2,
1224
+ "grad_norm": 0.050235260277986526,
1225
+ "learning_rate": 0.00018327099690154344,
1226
+ "loss": 0.3819,
1227
+ "step": 174
1228
+ },
1229
+ {
1230
+ "epoch": 0.2,
1231
+ "grad_norm": 0.051336683332920074,
1232
+ "learning_rate": 0.00018307234018877434,
1233
+ "loss": 0.3897,
1234
+ "step": 175
1235
+ },
1236
+ {
1237
+ "epoch": 0.2,
1238
+ "grad_norm": 0.045052576810121536,
1239
+ "learning_rate": 0.0001828726199044957,
1240
+ "loss": 0.3822,
1241
+ "step": 176
1242
+ },
1243
+ {
1244
+ "epoch": 0.2,
1245
+ "grad_norm": 0.05162283405661583,
1246
+ "learning_rate": 0.00018267183860571753,
1247
+ "loss": 0.4047,
1248
+ "step": 177
1249
+ },
1250
+ {
1251
+ "epoch": 0.2,
1252
+ "grad_norm": 0.0488157793879509,
1253
+ "learning_rate": 0.00018246999886303383,
1254
+ "loss": 0.3947,
1255
+ "step": 178
1256
+ },
1257
+ {
1258
+ "epoch": 0.2,
1259
+ "grad_norm": 0.04454487934708595,
1260
+ "learning_rate": 0.00018226710326059006,
1261
+ "loss": 0.3942,
1262
+ "step": 179
1263
+ },
1264
+ {
1265
+ "epoch": 0.2,
1266
+ "grad_norm": 0.0500001423060894,
1267
+ "learning_rate": 0.0001820631543960496,
1268
+ "loss": 0.3826,
1269
+ "step": 180
1270
+ },
1271
+ {
1272
+ "epoch": 0.2,
1273
+ "grad_norm": 0.04919865354895592,
1274
+ "learning_rate": 0.00018185815488056076,
1275
+ "loss": 0.3791,
1276
+ "step": 181
1277
+ },
1278
+ {
1279
+ "epoch": 0.2,
1280
+ "grad_norm": 0.04547140747308731,
1281
+ "learning_rate": 0.00018165210733872336,
1282
+ "loss": 0.3879,
1283
+ "step": 182
1284
+ },
1285
+ {
1286
+ "epoch": 0.21,
1287
+ "grad_norm": 0.044622063636779785,
1288
+ "learning_rate": 0.00018144501440855496,
1289
+ "loss": 0.3778,
1290
+ "step": 183
1291
+ },
1292
+ {
1293
+ "epoch": 0.21,
1294
+ "grad_norm": 0.04467932507395744,
1295
+ "learning_rate": 0.00018123687874145721,
1296
+ "loss": 0.3994,
1297
+ "step": 184
1298
+ },
1299
+ {
1300
+ "epoch": 0.21,
1301
+ "grad_norm": 0.04281982406973839,
1302
+ "learning_rate": 0.0001810277030021819,
1303
+ "loss": 0.3817,
1304
+ "step": 185
1305
+ },
1306
+ {
1307
+ "epoch": 0.21,
1308
+ "grad_norm": 0.05303504317998886,
1309
+ "learning_rate": 0.00018081748986879679,
1310
+ "loss": 0.3749,
1311
+ "step": 186
1312
+ },
1313
+ {
1314
+ "epoch": 0.21,
1315
+ "grad_norm": 0.046573616564273834,
1316
+ "learning_rate": 0.00018060624203265134,
1317
+ "loss": 0.3866,
1318
+ "step": 187
1319
+ },
1320
+ {
1321
+ "epoch": 0.21,
1322
+ "grad_norm": 0.044320229440927505,
1323
+ "learning_rate": 0.00018039396219834237,
1324
+ "loss": 0.3732,
1325
+ "step": 188
1326
+ },
1327
+ {
1328
+ "epoch": 0.21,
1329
+ "grad_norm": 0.05708359181880951,
1330
+ "learning_rate": 0.00018018065308367912,
1331
+ "loss": 0.3863,
1332
+ "step": 189
1333
+ },
1334
+ {
1335
+ "epoch": 0.21,
1336
+ "grad_norm": 0.045029208064079285,
1337
+ "learning_rate": 0.00017996631741964888,
1338
+ "loss": 0.3862,
1339
+ "step": 190
1340
+ },
1341
+ {
1342
+ "epoch": 0.21,
1343
+ "grad_norm": 0.055195923894643784,
1344
+ "learning_rate": 0.00017975095795038165,
1345
+ "loss": 0.3835,
1346
+ "step": 191
1347
+ },
1348
+ {
1349
+ "epoch": 0.22,
1350
+ "grad_norm": 0.048293352127075195,
1351
+ "learning_rate": 0.00017953457743311523,
1352
+ "loss": 0.374,
1353
+ "step": 192
1354
+ },
1355
+ {
1356
+ "epoch": 0.22,
1357
+ "grad_norm": 0.04677055403590202,
1358
+ "learning_rate": 0.00017931717863815987,
1359
+ "loss": 0.377,
1360
+ "step": 193
1361
+ },
1362
+ {
1363
+ "epoch": 0.22,
1364
+ "grad_norm": 0.04955766722559929,
1365
+ "learning_rate": 0.00017909876434886273,
1366
+ "loss": 0.3808,
1367
+ "step": 194
1368
+ },
1369
+ {
1370
+ "epoch": 0.22,
1371
+ "grad_norm": 0.04526973515748978,
1372
+ "learning_rate": 0.00017887933736157233,
1373
+ "loss": 0.3796,
1374
+ "step": 195
1375
+ },
1376
+ {
1377
+ "epoch": 0.22,
1378
+ "grad_norm": 0.043622203171253204,
1379
+ "learning_rate": 0.00017865890048560277,
1380
+ "loss": 0.376,
1381
+ "step": 196
1382
+ },
1383
+ {
1384
+ "epoch": 0.22,
1385
+ "grad_norm": 0.046581387519836426,
1386
+ "learning_rate": 0.0001784374565431976,
1387
+ "loss": 0.3716,
1388
+ "step": 197
1389
+ },
1390
+ {
1391
+ "epoch": 0.22,
1392
+ "grad_norm": 0.04433497413992882,
1393
+ "learning_rate": 0.00017821500836949386,
1394
+ "loss": 0.3715,
1395
+ "step": 198
1396
+ },
1397
+ {
1398
+ "epoch": 0.22,
1399
+ "grad_norm": 0.04146367311477661,
1400
+ "learning_rate": 0.00017799155881248572,
1401
+ "loss": 0.3809,
1402
+ "step": 199
1403
+ },
1404
+ {
1405
+ "epoch": 0.23,
1406
+ "grad_norm": 0.045288585126399994,
1407
+ "learning_rate": 0.000177767110732988,
1408
+ "loss": 0.3885,
1409
+ "step": 200
1410
+ },
1411
+ {
1412
+ "epoch": 0.23,
1413
+ "grad_norm": 0.04070120304822922,
1414
+ "learning_rate": 0.00017754166700459958,
1415
+ "loss": 0.3713,
1416
+ "step": 201
1417
+ },
1418
+ {
1419
+ "epoch": 0.23,
1420
+ "grad_norm": 0.042820919305086136,
1421
+ "learning_rate": 0.00017731523051366658,
1422
+ "loss": 0.3839,
1423
+ "step": 202
1424
+ },
1425
+ {
1426
+ "epoch": 0.23,
1427
+ "grad_norm": 0.04416365176439285,
1428
+ "learning_rate": 0.00017708780415924539,
1429
+ "loss": 0.3728,
1430
+ "step": 203
1431
+ },
1432
+ {
1433
+ "epoch": 0.23,
1434
+ "grad_norm": 0.04461952671408653,
1435
+ "learning_rate": 0.00017685939085306562,
1436
+ "loss": 0.373,
1437
+ "step": 204
1438
+ },
1439
+ {
1440
+ "epoch": 0.23,
1441
+ "grad_norm": 0.04675828292965889,
1442
+ "learning_rate": 0.00017662999351949278,
1443
+ "loss": 0.3711,
1444
+ "step": 205
1445
+ },
1446
+ {
1447
+ "epoch": 0.23,
1448
+ "grad_norm": 0.04258272796869278,
1449
+ "learning_rate": 0.00017639961509549078,
1450
+ "loss": 0.3782,
1451
+ "step": 206
1452
+ },
1453
+ {
1454
+ "epoch": 0.23,
1455
+ "grad_norm": 0.04638506844639778,
1456
+ "learning_rate": 0.00017616825853058443,
1457
+ "loss": 0.3592,
1458
+ "step": 207
1459
+ },
1460
+ {
1461
+ "epoch": 0.23,
1462
+ "grad_norm": 0.04781416058540344,
1463
+ "learning_rate": 0.00017593592678682166,
1464
+ "loss": 0.383,
1465
+ "step": 208
1466
+ },
1467
+ {
1468
+ "epoch": 0.24,
1469
+ "grad_norm": 0.04813629388809204,
1470
+ "learning_rate": 0.00017570262283873552,
1471
+ "loss": 0.3775,
1472
+ "step": 209
1473
+ },
1474
+ {
1475
+ "epoch": 0.24,
1476
+ "grad_norm": 0.046996332705020905,
1477
+ "learning_rate": 0.00017546834967330617,
1478
+ "loss": 0.3815,
1479
+ "step": 210
1480
+ },
1481
+ {
1482
+ "epoch": 0.24,
1483
+ "grad_norm": 0.04889595881104469,
1484
+ "learning_rate": 0.00017523311028992268,
1485
+ "loss": 0.3636,
1486
+ "step": 211
1487
+ },
1488
+ {
1489
+ "epoch": 0.24,
1490
+ "grad_norm": 0.04298345744609833,
1491
+ "learning_rate": 0.00017499690770034443,
1492
+ "loss": 0.3672,
1493
+ "step": 212
1494
+ },
1495
+ {
1496
+ "epoch": 0.24,
1497
+ "grad_norm": 0.04219110682606697,
1498
+ "learning_rate": 0.00017475974492866278,
1499
+ "loss": 0.3801,
1500
+ "step": 213
1501
+ },
1502
+ {
1503
+ "epoch": 0.24,
1504
+ "grad_norm": 0.051573265343904495,
1505
+ "learning_rate": 0.00017452162501126227,
1506
+ "loss": 0.3778,
1507
+ "step": 214
1508
+ },
1509
+ {
1510
+ "epoch": 0.24,
1511
+ "grad_norm": 0.048954349011182785,
1512
+ "learning_rate": 0.00017428255099678167,
1513
+ "loss": 0.3849,
1514
+ "step": 215
1515
+ },
1516
+ {
1517
+ "epoch": 0.24,
1518
+ "grad_norm": 0.042610183358192444,
1519
+ "learning_rate": 0.0001740425259460751,
1520
+ "loss": 0.3682,
1521
+ "step": 216
1522
+ },
1523
+ {
1524
+ "epoch": 0.24,
1525
+ "grad_norm": 0.04517417773604393,
1526
+ "learning_rate": 0.00017380155293217264,
1527
+ "loss": 0.3827,
1528
+ "step": 217
1529
+ },
1530
+ {
1531
+ "epoch": 0.25,
1532
+ "grad_norm": 0.04968888312578201,
1533
+ "learning_rate": 0.00017355963504024123,
1534
+ "loss": 0.3821,
1535
+ "step": 218
1536
+ },
1537
+ {
1538
+ "epoch": 0.25,
1539
+ "grad_norm": 0.051313381642103195,
1540
+ "learning_rate": 0.0001733167753675449,
1541
+ "loss": 0.381,
1542
+ "step": 219
1543
+ },
1544
+ {
1545
+ "epoch": 0.25,
1546
+ "grad_norm": 0.044351302087306976,
1547
+ "learning_rate": 0.0001730729770234054,
1548
+ "loss": 0.381,
1549
+ "step": 220
1550
+ },
1551
+ {
1552
+ "epoch": 0.25,
1553
+ "grad_norm": 0.03970547392964363,
1554
+ "learning_rate": 0.00017282824312916218,
1555
+ "loss": 0.3698,
1556
+ "step": 221
1557
+ },
1558
+ {
1559
+ "epoch": 0.25,
1560
+ "grad_norm": 0.04822370782494545,
1561
+ "learning_rate": 0.00017258257681813244,
1562
+ "loss": 0.3838,
1563
+ "step": 222
1564
+ },
1565
+ {
1566
+ "epoch": 0.25,
1567
+ "grad_norm": 0.045927174389362335,
1568
+ "learning_rate": 0.0001723359812355712,
1569
+ "loss": 0.3662,
1570
+ "step": 223
1571
+ },
1572
+ {
1573
+ "epoch": 0.25,
1574
+ "grad_norm": 0.042983219027519226,
1575
+ "learning_rate": 0.00017208845953863076,
1576
+ "loss": 0.3574,
1577
+ "step": 224
1578
+ },
1579
+ {
1580
+ "epoch": 0.25,
1581
+ "grad_norm": 0.0422198586165905,
1582
+ "learning_rate": 0.0001718400148963206,
1583
+ "loss": 0.3559,
1584
+ "step": 225
1585
+ },
1586
+ {
1587
+ "epoch": 0.25,
1588
+ "grad_norm": 0.042307570576667786,
1589
+ "learning_rate": 0.00017159065048946644,
1590
+ "loss": 0.3834,
1591
+ "step": 226
1592
+ },
1593
+ {
1594
+ "epoch": 0.26,
1595
+ "grad_norm": 0.04701109230518341,
1596
+ "learning_rate": 0.0001713403695106698,
1597
+ "loss": 0.3718,
1598
+ "step": 227
1599
+ },
1600
+ {
1601
+ "epoch": 0.26,
1602
+ "grad_norm": 0.04007503017783165,
1603
+ "learning_rate": 0.00017108917516426704,
1604
+ "loss": 0.3785,
1605
+ "step": 228
1606
+ },
1607
+ {
1608
+ "epoch": 0.26,
1609
+ "grad_norm": 0.04560061916708946,
1610
+ "learning_rate": 0.00017083707066628832,
1611
+ "loss": 0.3713,
1612
+ "step": 229
1613
+ },
1614
+ {
1615
+ "epoch": 0.26,
1616
+ "grad_norm": 0.04315731301903725,
1617
+ "learning_rate": 0.00017058405924441636,
1618
+ "loss": 0.3702,
1619
+ "step": 230
1620
+ },
1621
+ {
1622
+ "epoch": 0.26,
1623
+ "grad_norm": 0.040260497480630875,
1624
+ "learning_rate": 0.0001703301441379453,
1625
+ "loss": 0.367,
1626
+ "step": 231
1627
+ },
1628
+ {
1629
+ "epoch": 0.26,
1630
+ "grad_norm": 0.04882992431521416,
1631
+ "learning_rate": 0.000170075328597739,
1632
+ "loss": 0.3737,
1633
+ "step": 232
1634
+ },
1635
+ {
1636
+ "epoch": 0.26,
1637
+ "grad_norm": 0.04410382732748985,
1638
+ "learning_rate": 0.0001698196158861896,
1639
+ "loss": 0.3625,
1640
+ "step": 233
1641
+ },
1642
+ {
1643
+ "epoch": 0.26,
1644
+ "grad_norm": 0.04889338091015816,
1645
+ "learning_rate": 0.00016956300927717575,
1646
+ "loss": 0.3697,
1647
+ "step": 234
1648
+ },
1649
+ {
1650
+ "epoch": 0.26,
1651
+ "grad_norm": 0.044603537768125534,
1652
+ "learning_rate": 0.00016930551205602043,
1653
+ "loss": 0.3729,
1654
+ "step": 235
1655
+ },
1656
+ {
1657
+ "epoch": 0.27,
1658
+ "grad_norm": 0.0539550743997097,
1659
+ "learning_rate": 0.00016904712751944931,
1660
+ "loss": 0.3625,
1661
+ "step": 236
1662
+ },
1663
+ {
1664
+ "epoch": 0.27,
1665
+ "grad_norm": 0.04753349721431732,
1666
+ "learning_rate": 0.00016878785897554818,
1667
+ "loss": 0.3662,
1668
+ "step": 237
1669
+ },
1670
+ {
1671
+ "epoch": 0.27,
1672
+ "grad_norm": 0.04425463080406189,
1673
+ "learning_rate": 0.0001685277097437208,
1674
+ "loss": 0.3595,
1675
+ "step": 238
1676
+ },
1677
+ {
1678
+ "epoch": 0.27,
1679
+ "grad_norm": 0.04160892590880394,
1680
+ "learning_rate": 0.0001682666831546463,
1681
+ "loss": 0.3679,
1682
+ "step": 239
1683
+ },
1684
+ {
1685
+ "epoch": 0.27,
1686
+ "grad_norm": 0.04600003361701965,
1687
+ "learning_rate": 0.0001680047825502366,
1688
+ "loss": 0.3702,
1689
+ "step": 240
1690
+ },
1691
+ {
1692
+ "epoch": 0.27,
1693
+ "grad_norm": 0.03887678310275078,
1694
+ "learning_rate": 0.00016774201128359357,
1695
+ "loss": 0.3633,
1696
+ "step": 241
1697
+ },
1698
+ {
1699
+ "epoch": 0.27,
1700
+ "grad_norm": 0.0477156862616539,
1701
+ "learning_rate": 0.00016747837271896622,
1702
+ "loss": 0.3702,
1703
+ "step": 242
1704
+ },
1705
+ {
1706
+ "epoch": 0.27,
1707
+ "grad_norm": 0.04244072362780571,
1708
+ "learning_rate": 0.00016721387023170737,
1709
+ "loss": 0.3668,
1710
+ "step": 243
1711
+ },
1712
+ {
1713
+ "epoch": 0.27,
1714
+ "grad_norm": 0.04049496725201607,
1715
+ "learning_rate": 0.0001669485072082308,
1716
+ "loss": 0.3785,
1717
+ "step": 244
1718
+ },
1719
+ {
1720
+ "epoch": 0.28,
1721
+ "grad_norm": 0.04432998597621918,
1722
+ "learning_rate": 0.00016668228704596756,
1723
+ "loss": 0.3703,
1724
+ "step": 245
1725
+ },
1726
+ {
1727
+ "epoch": 0.28,
1728
+ "grad_norm": 0.0432085320353508,
1729
+ "learning_rate": 0.00016641521315332265,
1730
+ "loss": 0.3615,
1731
+ "step": 246
1732
+ },
1733
+ {
1734
+ "epoch": 0.28,
1735
+ "grad_norm": 0.03820549696683884,
1736
+ "learning_rate": 0.00016614728894963135,
1737
+ "loss": 0.3483,
1738
+ "step": 247
1739
+ },
1740
+ {
1741
+ "epoch": 0.28,
1742
+ "grad_norm": 0.04436295107007027,
1743
+ "learning_rate": 0.00016587851786511543,
1744
+ "loss": 0.3661,
1745
+ "step": 248
1746
+ },
1747
+ {
1748
+ "epoch": 0.28,
1749
+ "grad_norm": 0.04371733218431473,
1750
+ "learning_rate": 0.00016560890334083926,
1751
+ "loss": 0.3503,
1752
+ "step": 249
1753
+ },
1754
+ {
1755
+ "epoch": 0.28,
1756
+ "grad_norm": 0.039205193519592285,
1757
+ "learning_rate": 0.00016533844882866568,
1758
+ "loss": 0.3482,
1759
+ "step": 250
1760
+ },
1761
+ {
1762
+ "epoch": 0.28,
1763
+ "grad_norm": 0.04308384284377098,
1764
+ "learning_rate": 0.00016506715779121187,
1765
+ "loss": 0.373,
1766
+ "step": 251
1767
+ },
1768
+ {
1769
+ "epoch": 0.28,
1770
+ "grad_norm": 0.040143441408872604,
1771
+ "learning_rate": 0.00016479503370180507,
1772
+ "loss": 0.3609,
1773
+ "step": 252
1774
+ },
1775
+ {
1776
+ "epoch": 0.28,
1777
+ "grad_norm": 0.03845199570059776,
1778
+ "learning_rate": 0.000164522080044438,
1779
+ "loss": 0.3644,
1780
+ "step": 253
1781
+ },
1782
+ {
1783
+ "epoch": 0.29,
1784
+ "grad_norm": 0.039730221033096313,
1785
+ "learning_rate": 0.00016424830031372425,
1786
+ "loss": 0.3514,
1787
+ "step": 254
1788
+ },
1789
+ {
1790
+ "epoch": 0.29,
1791
+ "grad_norm": 0.04021477699279785,
1792
+ "learning_rate": 0.00016397369801485366,
1793
+ "loss": 0.3566,
1794
+ "step": 255
1795
+ },
1796
+ {
1797
+ "epoch": 0.29,
1798
+ "grad_norm": 0.03929613530635834,
1799
+ "learning_rate": 0.00016369827666354745,
1800
+ "loss": 0.3649,
1801
+ "step": 256
1802
+ },
1803
+ {
1804
+ "epoch": 0.29,
1805
+ "grad_norm": 0.040966227650642395,
1806
+ "learning_rate": 0.0001634220397860129,
1807
+ "loss": 0.3661,
1808
+ "step": 257
1809
+ },
1810
+ {
1811
+ "epoch": 0.29,
1812
+ "grad_norm": 0.036787159740924835,
1813
+ "learning_rate": 0.0001631449909188987,
1814
+ "loss": 0.3572,
1815
+ "step": 258
1816
+ },
1817
+ {
1818
+ "epoch": 0.29,
1819
+ "grad_norm": 0.039864350110292435,
1820
+ "learning_rate": 0.00016286713360924918,
1821
+ "loss": 0.3593,
1822
+ "step": 259
1823
+ },
1824
+ {
1825
+ "epoch": 0.29,
1826
+ "grad_norm": 0.039622630923986435,
1827
+ "learning_rate": 0.00016258847141445928,
1828
+ "loss": 0.3711,
1829
+ "step": 260
1830
+ },
1831
+ {
1832
+ "epoch": 0.29,
1833
+ "grad_norm": 0.03840857744216919,
1834
+ "learning_rate": 0.00016230900790222878,
1835
+ "loss": 0.3537,
1836
+ "step": 261
1837
+ },
1838
+ {
1839
+ "epoch": 0.29,
1840
+ "grad_norm": 0.03800921142101288,
1841
+ "learning_rate": 0.00016202874665051674,
1842
+ "loss": 0.3662,
1843
+ "step": 262
1844
+ },
1845
+ {
1846
+ "epoch": 0.3,
1847
+ "grad_norm": 0.03894530236721039,
1848
+ "learning_rate": 0.0001617476912474956,
1849
+ "loss": 0.3633,
1850
+ "step": 263
1851
+ },
1852
+ {
1853
+ "epoch": 0.3,
1854
+ "grad_norm": 0.04486812278628349,
1855
+ "learning_rate": 0.00016146584529150526,
1856
+ "loss": 0.3594,
1857
+ "step": 264
1858
+ },
1859
+ {
1860
+ "epoch": 0.3,
1861
+ "grad_norm": 0.0394977331161499,
1862
+ "learning_rate": 0.00016118321239100712,
1863
+ "loss": 0.3473,
1864
+ "step": 265
1865
+ },
1866
+ {
1867
+ "epoch": 0.3,
1868
+ "grad_norm": 0.03960977867245674,
1869
+ "learning_rate": 0.0001608997961645377,
1870
+ "loss": 0.363,
1871
+ "step": 266
1872
+ },
1873
+ {
1874
+ "epoch": 0.3,
1875
+ "grad_norm": 0.04547852650284767,
1876
+ "learning_rate": 0.00016061560024066248,
1877
+ "loss": 0.3698,
1878
+ "step": 267
1879
+ },
1880
+ {
1881
+ "epoch": 0.3,
1882
+ "grad_norm": 0.045081403106451035,
1883
+ "learning_rate": 0.00016033062825792935,
1884
+ "loss": 0.3684,
1885
+ "step": 268
1886
+ },
1887
+ {
1888
+ "epoch": 0.3,
1889
+ "grad_norm": 0.03854916989803314,
1890
+ "learning_rate": 0.00016004488386482205,
1891
+ "loss": 0.3467,
1892
+ "step": 269
1893
+ },
1894
+ {
1895
+ "epoch": 0.3,
1896
+ "grad_norm": 0.042972736060619354,
1897
+ "learning_rate": 0.0001597583707197134,
1898
+ "loss": 0.3589,
1899
+ "step": 270
1900
+ },
1901
+ {
1902
+ "epoch": 0.3,
1903
+ "grad_norm": 0.0463341549038887,
1904
+ "learning_rate": 0.0001594710924908186,
1905
+ "loss": 0.3517,
1906
+ "step": 271
1907
+ },
1908
+ {
1909
+ "epoch": 0.31,
1910
+ "grad_norm": 0.04469626024365425,
1911
+ "learning_rate": 0.00015918305285614822,
1912
+ "loss": 0.3509,
1913
+ "step": 272
1914
+ },
1915
+ {
1916
+ "epoch": 0.31,
1917
+ "grad_norm": 0.037887342274188995,
1918
+ "learning_rate": 0.0001588942555034609,
1919
+ "loss": 0.3622,
1920
+ "step": 273
1921
+ },
1922
+ {
1923
+ "epoch": 0.31,
1924
+ "grad_norm": 0.046203695237636566,
1925
+ "learning_rate": 0.00015860470413021642,
1926
+ "loss": 0.3587,
1927
+ "step": 274
1928
+ },
1929
+ {
1930
+ "epoch": 0.31,
1931
+ "grad_norm": 0.043443821370601654,
1932
+ "learning_rate": 0.00015831440244352832,
1933
+ "loss": 0.3742,
1934
+ "step": 275
1935
+ },
1936
+ {
1937
+ "epoch": 0.31,
1938
+ "grad_norm": 0.037403471767902374,
1939
+ "learning_rate": 0.00015802335416011625,
1940
+ "loss": 0.357,
1941
+ "step": 276
1942
+ },
1943
+ {
1944
+ "epoch": 0.31,
1945
+ "grad_norm": 0.041037216782569885,
1946
+ "learning_rate": 0.00015773156300625857,
1947
+ "loss": 0.347,
1948
+ "step": 277
1949
+ },
1950
+ {
1951
+ "epoch": 0.31,
1952
+ "grad_norm": 0.04024377465248108,
1953
+ "learning_rate": 0.00015743903271774455,
1954
+ "loss": 0.3584,
1955
+ "step": 278
1956
+ },
1957
+ {
1958
+ "epoch": 0.31,
1959
+ "grad_norm": 0.04302544146776199,
1960
+ "learning_rate": 0.0001571457670398266,
1961
+ "loss": 0.3754,
1962
+ "step": 279
1963
+ },
1964
+ {
1965
+ "epoch": 0.32,
1966
+ "grad_norm": 0.04049861803650856,
1967
+ "learning_rate": 0.00015685176972717223,
1968
+ "loss": 0.3617,
1969
+ "step": 280
1970
+ },
1971
+ {
1972
+ "epoch": 0.32,
1973
+ "grad_norm": 0.04286986216902733,
1974
+ "learning_rate": 0.0001565570445438161,
1975
+ "loss": 0.3459,
1976
+ "step": 281
1977
+ },
1978
+ {
1979
+ "epoch": 0.32,
1980
+ "grad_norm": 0.04489251226186752,
1981
+ "learning_rate": 0.00015626159526311174,
1982
+ "loss": 0.342,
1983
+ "step": 282
1984
+ },
1985
+ {
1986
+ "epoch": 0.32,
1987
+ "grad_norm": 0.039040662348270416,
1988
+ "learning_rate": 0.00015596542566768327,
1989
+ "loss": 0.3579,
1990
+ "step": 283
1991
+ },
1992
+ {
1993
+ "epoch": 0.32,
1994
+ "grad_norm": 0.041434600949287415,
1995
+ "learning_rate": 0.00015566853954937694,
1996
+ "loss": 0.3607,
1997
+ "step": 284
1998
+ },
1999
+ {
2000
+ "epoch": 0.32,
2001
+ "grad_norm": 0.040113747119903564,
2002
+ "learning_rate": 0.00015537094070921267,
2003
+ "loss": 0.3581,
2004
+ "step": 285
2005
+ },
2006
+ {
2007
+ "epoch": 0.32,
2008
+ "grad_norm": 0.047135982662439346,
2009
+ "learning_rate": 0.00015507263295733528,
2010
+ "loss": 0.355,
2011
+ "step": 286
2012
+ },
2013
+ {
2014
+ "epoch": 0.32,
2015
+ "grad_norm": 0.04679419472813606,
2016
+ "learning_rate": 0.00015477362011296575,
2017
+ "loss": 0.3506,
2018
+ "step": 287
2019
+ },
2020
+ {
2021
+ "epoch": 0.32,
2022
+ "grad_norm": 0.04256124794483185,
2023
+ "learning_rate": 0.00015447390600435238,
2024
+ "loss": 0.3547,
2025
+ "step": 288
2026
+ },
2027
+ {
2028
+ "epoch": 0.33,
2029
+ "grad_norm": 0.041164420545101166,
2030
+ "learning_rate": 0.00015417349446872168,
2031
+ "loss": 0.3533,
2032
+ "step": 289
2033
+ },
2034
+ {
2035
+ "epoch": 0.33,
2036
+ "grad_norm": 0.04434530436992645,
2037
+ "learning_rate": 0.00015387238935222927,
2038
+ "loss": 0.3655,
2039
+ "step": 290
2040
+ },
2041
+ {
2042
+ "epoch": 0.33,
2043
+ "grad_norm": 0.042684029787778854,
2044
+ "learning_rate": 0.00015357059450991068,
2045
+ "loss": 0.3388,
2046
+ "step": 291
2047
+ },
2048
+ {
2049
+ "epoch": 0.33,
2050
+ "grad_norm": 0.03664594888687134,
2051
+ "learning_rate": 0.00015326811380563204,
2052
+ "loss": 0.3415,
2053
+ "step": 292
2054
+ },
2055
+ {
2056
+ "epoch": 0.33,
2057
+ "grad_norm": 0.044670287519693375,
2058
+ "learning_rate": 0.0001529649511120404,
2059
+ "loss": 0.3639,
2060
+ "step": 293
2061
+ },
2062
+ {
2063
+ "epoch": 0.33,
2064
+ "grad_norm": 0.043318092823028564,
2065
+ "learning_rate": 0.00015266111031051442,
2066
+ "loss": 0.3705,
2067
+ "step": 294
2068
+ },
2069
+ {
2070
+ "epoch": 0.33,
2071
+ "grad_norm": 0.03925277665257454,
2072
+ "learning_rate": 0.00015235659529111445,
2073
+ "loss": 0.3452,
2074
+ "step": 295
2075
+ },
2076
+ {
2077
+ "epoch": 0.33,
2078
+ "grad_norm": 0.048936985433101654,
2079
+ "learning_rate": 0.00015205140995253283,
2080
+ "loss": 0.3511,
2081
+ "step": 296
2082
+ },
2083
+ {
2084
+ "epoch": 0.33,
2085
+ "grad_norm": 0.04244200140237808,
2086
+ "learning_rate": 0.00015174555820204408,
2087
+ "loss": 0.3573,
2088
+ "step": 297
2089
+ },
2090
+ {
2091
+ "epoch": 0.34,
2092
+ "grad_norm": 0.041375573724508286,
2093
+ "learning_rate": 0.00015143904395545466,
2094
+ "loss": 0.3474,
2095
+ "step": 298
2096
+ },
2097
+ {
2098
+ "epoch": 0.34,
2099
+ "grad_norm": 0.04903846234083176,
2100
+ "learning_rate": 0.0001511318711370529,
2101
+ "loss": 0.3677,
2102
+ "step": 299
2103
+ },
2104
+ {
2105
+ "epoch": 0.34,
2106
+ "grad_norm": 0.03968316316604614,
2107
+ "learning_rate": 0.0001508240436795589,
2108
+ "loss": 0.3504,
2109
+ "step": 300
2110
+ },
2111
+ {
2112
+ "epoch": 0.34,
2113
+ "grad_norm": 0.038755469024181366,
2114
+ "learning_rate": 0.00015051556552407399,
2115
+ "loss": 0.3594,
2116
+ "step": 301
2117
+ },
2118
+ {
2119
+ "epoch": 0.34,
2120
+ "grad_norm": 0.03975263237953186,
2121
+ "learning_rate": 0.00015020644062003046,
2122
+ "loss": 0.3536,
2123
+ "step": 302
2124
+ },
2125
+ {
2126
+ "epoch": 0.34,
2127
+ "grad_norm": 0.03559111803770065,
2128
+ "learning_rate": 0.0001498966729251408,
2129
+ "loss": 0.3489,
2130
+ "step": 303
2131
+ },
2132
+ {
2133
+ "epoch": 0.34,
2134
+ "grad_norm": 0.0421811044216156,
2135
+ "learning_rate": 0.0001495862664053471,
2136
+ "loss": 0.3558,
2137
+ "step": 304
2138
+ },
2139
+ {
2140
+ "epoch": 0.34,
2141
+ "grad_norm": 0.04090610146522522,
2142
+ "learning_rate": 0.00014927522503477048,
2143
+ "loss": 0.3532,
2144
+ "step": 305
2145
+ },
2146
+ {
2147
+ "epoch": 0.34,
2148
+ "grad_norm": 0.04141204059123993,
2149
+ "learning_rate": 0.00014896355279565976,
2150
+ "loss": 0.361,
2151
+ "step": 306
2152
+ },
2153
+ {
2154
+ "epoch": 0.35,
2155
+ "grad_norm": 0.039628610014915466,
2156
+ "learning_rate": 0.00014865125367834092,
2157
+ "loss": 0.353,
2158
+ "step": 307
2159
+ },
2160
+ {
2161
+ "epoch": 0.35,
2162
+ "grad_norm": 0.040805645287036896,
2163
+ "learning_rate": 0.00014833833168116582,
2164
+ "loss": 0.3442,
2165
+ "step": 308
2166
+ },
2167
+ {
2168
+ "epoch": 0.35,
2169
+ "grad_norm": 0.03766808658838272,
2170
+ "learning_rate": 0.00014802479081046102,
2171
+ "loss": 0.3429,
2172
+ "step": 309
2173
+ },
2174
+ {
2175
+ "epoch": 0.35,
2176
+ "grad_norm": 0.03960324451327324,
2177
+ "learning_rate": 0.00014771063508047636,
2178
+ "loss": 0.3641,
2179
+ "step": 310
2180
+ },
2181
+ {
2182
+ "epoch": 0.35,
2183
+ "grad_norm": 0.03804844617843628,
2184
+ "learning_rate": 0.0001473958685133339,
2185
+ "loss": 0.3462,
2186
+ "step": 311
2187
+ },
2188
+ {
2189
+ "epoch": 0.35,
2190
+ "grad_norm": 0.03872362896800041,
2191
+ "learning_rate": 0.0001470804951389761,
2192
+ "loss": 0.3529,
2193
+ "step": 312
2194
+ },
2195
+ {
2196
+ "epoch": 0.35,
2197
+ "grad_norm": 0.038420502096414566,
2198
+ "learning_rate": 0.00014676451899511437,
2199
+ "loss": 0.3594,
2200
+ "step": 313
2201
+ },
2202
+ {
2203
+ "epoch": 0.35,
2204
+ "grad_norm": 0.038893669843673706,
2205
+ "learning_rate": 0.00014644794412717736,
2206
+ "loss": 0.3461,
2207
+ "step": 314
2208
+ },
2209
+ {
2210
+ "epoch": 0.35,
2211
+ "grad_norm": 0.039783939719200134,
2212
+ "learning_rate": 0.00014613077458825913,
2213
+ "loss": 0.3641,
2214
+ "step": 315
2215
+ },
2216
+ {
2217
+ "epoch": 0.36,
2218
+ "grad_norm": 0.040646251291036606,
2219
+ "learning_rate": 0.0001458130144390673,
2220
+ "loss": 0.3408,
2221
+ "step": 316
2222
+ },
2223
+ {
2224
+ "epoch": 0.36,
2225
+ "grad_norm": 0.03997185081243515,
2226
+ "learning_rate": 0.00014549466774787108,
2227
+ "loss": 0.3476,
2228
+ "step": 317
2229
+ },
2230
+ {
2231
+ "epoch": 0.36,
2232
+ "grad_norm": 0.03733928129076958,
2233
+ "learning_rate": 0.00014517573859044907,
2234
+ "loss": 0.348,
2235
+ "step": 318
2236
+ },
2237
+ {
2238
+ "epoch": 0.36,
2239
+ "grad_norm": 0.04240603372454643,
2240
+ "learning_rate": 0.00014485623105003732,
2241
+ "loss": 0.3526,
2242
+ "step": 319
2243
+ },
2244
+ {
2245
+ "epoch": 0.36,
2246
+ "grad_norm": 0.04083820432424545,
2247
+ "learning_rate": 0.00014453614921727668,
2248
+ "loss": 0.358,
2249
+ "step": 320
2250
+ },
2251
+ {
2252
+ "epoch": 0.36,
2253
+ "grad_norm": 0.038078900426626205,
2254
+ "learning_rate": 0.00014421549719016081,
2255
+ "loss": 0.3383,
2256
+ "step": 321
2257
+ },
2258
+ {
2259
+ "epoch": 0.36,
2260
+ "grad_norm": 0.043162036687135696,
2261
+ "learning_rate": 0.00014389427907398342,
2262
+ "loss": 0.3502,
2263
+ "step": 322
2264
+ },
2265
+ {
2266
+ "epoch": 0.36,
2267
+ "grad_norm": 0.04186828061938286,
2268
+ "learning_rate": 0.00014357249898128594,
2269
+ "loss": 0.355,
2270
+ "step": 323
2271
+ },
2272
+ {
2273
+ "epoch": 0.36,
2274
+ "grad_norm": 0.041309986263513565,
2275
+ "learning_rate": 0.0001432501610318047,
2276
+ "loss": 0.3634,
2277
+ "step": 324
2278
+ },
2279
+ {
2280
+ "epoch": 0.37,
2281
+ "grad_norm": 0.03907867148518562,
2282
+ "learning_rate": 0.00014292726935241827,
2283
+ "loss": 0.3431,
2284
+ "step": 325
2285
+ },
2286
+ {
2287
+ "epoch": 0.37,
2288
+ "grad_norm": 0.04641426354646683,
2289
+ "learning_rate": 0.00014260382807709457,
2290
+ "loss": 0.3423,
2291
+ "step": 326
2292
+ },
2293
+ {
2294
+ "epoch": 0.37,
2295
+ "grad_norm": 0.042708221822977066,
2296
+ "learning_rate": 0.000142279841346838,
2297
+ "loss": 0.3459,
2298
+ "step": 327
2299
+ },
2300
+ {
2301
+ "epoch": 0.37,
2302
+ "grad_norm": 0.03900787606835365,
2303
+ "learning_rate": 0.00014195531330963635,
2304
+ "loss": 0.349,
2305
+ "step": 328
2306
+ },
2307
+ {
2308
+ "epoch": 0.37,
2309
+ "grad_norm": 0.04202708601951599,
2310
+ "learning_rate": 0.0001416302481204078,
2311
+ "loss": 0.3478,
2312
+ "step": 329
2313
+ },
2314
+ {
2315
+ "epoch": 0.37,
2316
+ "grad_norm": 0.04631033539772034,
2317
+ "learning_rate": 0.0001413046499409477,
2318
+ "loss": 0.3495,
2319
+ "step": 330
2320
+ },
2321
+ {
2322
+ "epoch": 0.37,
2323
+ "grad_norm": 0.04463302344083786,
2324
+ "learning_rate": 0.00014097852293987507,
2325
+ "loss": 0.3422,
2326
+ "step": 331
2327
+ },
2328
+ {
2329
+ "epoch": 0.37,
2330
+ "grad_norm": 0.03706689178943634,
2331
+ "learning_rate": 0.00014065187129257964,
2332
+ "loss": 0.3558,
2333
+ "step": 332
2334
+ },
2335
+ {
2336
+ "epoch": 0.37,
2337
+ "grad_norm": 0.04193353280425072,
2338
+ "learning_rate": 0.000140324699181168,
2339
+ "loss": 0.3539,
2340
+ "step": 333
2341
+ },
2342
+ {
2343
+ "epoch": 0.38,
2344
+ "grad_norm": 0.03598970174789429,
2345
+ "learning_rate": 0.00013999701079441028,
2346
+ "loss": 0.3412,
2347
+ "step": 334
2348
+ },
2349
+ {
2350
+ "epoch": 0.38,
2351
+ "grad_norm": 0.04364067688584328,
2352
+ "learning_rate": 0.00013966881032768643,
2353
+ "loss": 0.3496,
2354
+ "step": 335
2355
+ },
2356
+ {
2357
+ "epoch": 0.38,
2358
+ "grad_norm": 0.034331586211919785,
2359
+ "learning_rate": 0.00013934010198293257,
2360
+ "loss": 0.3495,
2361
+ "step": 336
2362
+ },
2363
+ {
2364
+ "epoch": 0.38,
2365
+ "grad_norm": 0.04461033269762993,
2366
+ "learning_rate": 0.00013901088996858717,
2367
+ "loss": 0.3555,
2368
+ "step": 337
2369
+ },
2370
+ {
2371
+ "epoch": 0.38,
2372
+ "grad_norm": 0.037957970052957535,
2373
+ "learning_rate": 0.0001386811784995371,
2374
+ "loss": 0.3535,
2375
+ "step": 338
2376
+ },
2377
+ {
2378
+ "epoch": 0.38,
2379
+ "grad_norm": 0.04143986850976944,
2380
+ "learning_rate": 0.0001383509717970638,
2381
+ "loss": 0.3552,
2382
+ "step": 339
2383
+ },
2384
+ {
2385
+ "epoch": 0.38,
2386
+ "grad_norm": 0.04176801070570946,
2387
+ "learning_rate": 0.0001380202740887891,
2388
+ "loss": 0.3462,
2389
+ "step": 340
2390
+ },
2391
+ {
2392
+ "epoch": 0.38,
2393
+ "grad_norm": 0.038572393357753754,
2394
+ "learning_rate": 0.00013768908960862123,
2395
+ "loss": 0.3582,
2396
+ "step": 341
2397
+ },
2398
+ {
2399
+ "epoch": 0.38,
2400
+ "grad_norm": 0.047118738293647766,
2401
+ "learning_rate": 0.0001373574225967004,
2402
+ "loss": 0.3451,
2403
+ "step": 342
2404
+ },
2405
+ {
2406
+ "epoch": 0.39,
2407
+ "grad_norm": 0.042480722069740295,
2408
+ "learning_rate": 0.00013702527729934482,
2409
+ "loss": 0.3355,
2410
+ "step": 343
2411
+ },
2412
+ {
2413
+ "epoch": 0.39,
2414
+ "grad_norm": 0.04277138039469719,
2415
+ "learning_rate": 0.00013669265796899607,
2416
+ "loss": 0.3395,
2417
+ "step": 344
2418
+ },
2419
+ {
2420
+ "epoch": 0.39,
2421
+ "grad_norm": 0.04838969558477402,
2422
+ "learning_rate": 0.0001363595688641648,
2423
+ "loss": 0.3419,
2424
+ "step": 345
2425
+ },
2426
+ {
2427
+ "epoch": 0.39,
2428
+ "grad_norm": 0.04388457536697388,
2429
+ "learning_rate": 0.00013602601424937604,
2430
+ "loss": 0.3412,
2431
+ "step": 346
2432
+ },
2433
+ {
2434
+ "epoch": 0.39,
2435
+ "grad_norm": 0.041595809161663055,
2436
+ "learning_rate": 0.00013569199839511494,
2437
+ "loss": 0.3488,
2438
+ "step": 347
2439
+ },
2440
+ {
2441
+ "epoch": 0.39,
2442
+ "grad_norm": 0.045120954513549805,
2443
+ "learning_rate": 0.0001353575255777717,
2444
+ "loss": 0.3486,
2445
+ "step": 348
2446
+ },
2447
+ {
2448
+ "epoch": 0.39,
2449
+ "grad_norm": 0.04108583182096481,
2450
+ "learning_rate": 0.00013502260007958706,
2451
+ "loss": 0.3522,
2452
+ "step": 349
2453
+ },
2454
+ {
2455
+ "epoch": 0.39,
2456
+ "grad_norm": 0.046396516263484955,
2457
+ "learning_rate": 0.00013468722618859743,
2458
+ "loss": 0.3474,
2459
+ "step": 350
2460
+ },
2461
+ {
2462
+ "epoch": 0.39,
2463
+ "grad_norm": 0.04111646115779877,
2464
+ "learning_rate": 0.0001343514081985799,
2465
+ "loss": 0.3522,
2466
+ "step": 351
2467
+ },
2468
+ {
2469
+ "epoch": 0.4,
2470
+ "grad_norm": 0.03970329090952873,
2471
+ "learning_rate": 0.00013401515040899746,
2472
+ "loss": 0.3504,
2473
+ "step": 352
2474
+ },
2475
+ {
2476
+ "epoch": 0.4,
2477
+ "grad_norm": 0.041780147701501846,
2478
+ "learning_rate": 0.00013367845712494372,
2479
+ "loss": 0.3464,
2480
+ "step": 353
2481
+ },
2482
+ {
2483
+ "epoch": 0.4,
2484
+ "grad_norm": 0.03978356719017029,
2485
+ "learning_rate": 0.000133341332657088,
2486
+ "loss": 0.3473,
2487
+ "step": 354
2488
+ },
2489
+ {
2490
+ "epoch": 0.4,
2491
+ "grad_norm": 0.04411393776535988,
2492
+ "learning_rate": 0.00013300378132161992,
2493
+ "loss": 0.3451,
2494
+ "step": 355
2495
+ },
2496
+ {
2497
+ "epoch": 0.4,
2498
+ "grad_norm": 0.04426994547247887,
2499
+ "learning_rate": 0.00013266580744019445,
2500
+ "loss": 0.3366,
2501
+ "step": 356
2502
+ },
2503
+ {
2504
+ "epoch": 0.4,
2505
+ "grad_norm": 0.037810515612363815,
2506
+ "learning_rate": 0.00013232741533987622,
2507
+ "loss": 0.342,
2508
+ "step": 357
2509
+ },
2510
+ {
2511
+ "epoch": 0.4,
2512
+ "grad_norm": 0.04545320197939873,
2513
+ "learning_rate": 0.00013198860935308444,
2514
+ "loss": 0.3553,
2515
+ "step": 358
2516
+ },
2517
+ {
2518
+ "epoch": 0.4,
2519
+ "grad_norm": 0.04098277539014816,
2520
+ "learning_rate": 0.00013164939381753713,
2521
+ "loss": 0.3458,
2522
+ "step": 359
2523
+ },
2524
+ {
2525
+ "epoch": 0.41,
2526
+ "grad_norm": 0.04471420869231224,
2527
+ "learning_rate": 0.00013130977307619594,
2528
+ "loss": 0.3318,
2529
+ "step": 360
2530
+ },
2531
+ {
2532
+ "epoch": 0.41,
2533
+ "grad_norm": 0.038031481206417084,
2534
+ "learning_rate": 0.00013096975147721017,
2535
+ "loss": 0.3533,
2536
+ "step": 361
2537
+ },
2538
+ {
2539
+ "epoch": 0.41,
2540
+ "grad_norm": 0.0440196692943573,
2541
+ "learning_rate": 0.00013062933337386142,
2542
+ "loss": 0.3299,
2543
+ "step": 362
2544
+ },
2545
+ {
2546
+ "epoch": 0.41,
2547
+ "grad_norm": 0.034114640206098557,
2548
+ "learning_rate": 0.00013028852312450763,
2549
+ "loss": 0.3401,
2550
+ "step": 363
2551
+ },
2552
+ {
2553
+ "epoch": 0.41,
2554
+ "grad_norm": 0.042403046041727066,
2555
+ "learning_rate": 0.00012994732509252744,
2556
+ "loss": 0.3458,
2557
+ "step": 364
2558
+ },
2559
+ {
2560
+ "epoch": 0.41,
2561
+ "grad_norm": 0.0356055311858654,
2562
+ "learning_rate": 0.00012960574364626412,
2563
+ "loss": 0.3498,
2564
+ "step": 365
2565
+ },
2566
+ {
2567
+ "epoch": 0.41,
2568
+ "grad_norm": 0.040617913007736206,
2569
+ "learning_rate": 0.00012926378315896998,
2570
+ "loss": 0.3454,
2571
+ "step": 366
2572
+ },
2573
+ {
2574
+ "epoch": 0.41,
2575
+ "grad_norm": 0.03738779574632645,
2576
+ "learning_rate": 0.00012892144800875,
2577
+ "loss": 0.342,
2578
+ "step": 367
2579
+ },
2580
+ {
2581
+ "epoch": 0.41,
2582
+ "grad_norm": 0.039146389812231064,
2583
+ "learning_rate": 0.00012857874257850605,
2584
+ "loss": 0.35,
2585
+ "step": 368
2586
+ },
2587
+ {
2588
+ "epoch": 0.42,
2589
+ "grad_norm": 0.03738423436880112,
2590
+ "learning_rate": 0.0001282356712558806,
2591
+ "loss": 0.3503,
2592
+ "step": 369
2593
+ },
2594
+ {
2595
+ "epoch": 0.42,
2596
+ "grad_norm": 0.03822367265820503,
2597
+ "learning_rate": 0.00012789223843320073,
2598
+ "loss": 0.3489,
2599
+ "step": 370
2600
+ },
2601
+ {
2602
+ "epoch": 0.42,
2603
+ "grad_norm": 0.03619171679019928,
2604
+ "learning_rate": 0.00012754844850742172,
2605
+ "loss": 0.3392,
2606
+ "step": 371
2607
+ },
2608
+ {
2609
+ "epoch": 0.42,
2610
+ "grad_norm": 0.03672062233090401,
2611
+ "learning_rate": 0.00012720430588007077,
2612
+ "loss": 0.3503,
2613
+ "step": 372
2614
+ },
2615
+ {
2616
+ "epoch": 0.42,
2617
+ "grad_norm": 0.038667939603328705,
2618
+ "learning_rate": 0.00012685981495719087,
2619
+ "loss": 0.3582,
2620
+ "step": 373
2621
+ },
2622
+ {
2623
+ "epoch": 0.42,
2624
+ "grad_norm": 0.033767540007829666,
2625
+ "learning_rate": 0.00012651498014928402,
2626
+ "loss": 0.3465,
2627
+ "step": 374
2628
+ },
2629
+ {
2630
+ "epoch": 0.42,
2631
+ "grad_norm": 0.0382758267223835,
2632
+ "learning_rate": 0.00012616980587125512,
2633
+ "loss": 0.3435,
2634
+ "step": 375
2635
+ },
2636
+ {
2637
+ "epoch": 0.42,
2638
+ "grad_norm": 0.03528289496898651,
2639
+ "learning_rate": 0.00012582429654235523,
2640
+ "loss": 0.3427,
2641
+ "step": 376
2642
+ },
2643
+ {
2644
+ "epoch": 0.42,
2645
+ "grad_norm": 0.03565202280879021,
2646
+ "learning_rate": 0.00012547845658612508,
2647
+ "loss": 0.3453,
2648
+ "step": 377
2649
+ },
2650
+ {
2651
+ "epoch": 0.43,
2652
+ "grad_norm": 0.03661198914051056,
2653
+ "learning_rate": 0.0001251322904303383,
2654
+ "loss": 0.3341,
2655
+ "step": 378
2656
+ },
2657
+ {
2658
+ "epoch": 0.43,
2659
+ "grad_norm": 0.03750570863485336,
2660
+ "learning_rate": 0.00012478580250694504,
2661
+ "loss": 0.3375,
2662
+ "step": 379
2663
+ },
2664
+ {
2665
+ "epoch": 0.43,
2666
+ "grad_norm": 0.03714427724480629,
2667
+ "learning_rate": 0.00012443899725201482,
2668
+ "loss": 0.3447,
2669
+ "step": 380
2670
+ },
2671
+ {
2672
+ "epoch": 0.43,
2673
+ "grad_norm": 0.03938335180282593,
2674
+ "learning_rate": 0.0001240918791056801,
2675
+ "loss": 0.3551,
2676
+ "step": 381
2677
+ },
2678
+ {
2679
+ "epoch": 0.43,
2680
+ "grad_norm": 0.037106383591890335,
2681
+ "learning_rate": 0.00012374445251207914,
2682
+ "loss": 0.3365,
2683
+ "step": 382
2684
+ },
2685
+ {
2686
+ "epoch": 0.43,
2687
+ "grad_norm": 0.039744965732097626,
2688
+ "learning_rate": 0.00012339672191929936,
2689
+ "loss": 0.3333,
2690
+ "step": 383
2691
+ },
2692
+ {
2693
+ "epoch": 0.43,
2694
+ "grad_norm": 0.04372095316648483,
2695
+ "learning_rate": 0.0001230486917793202,
2696
+ "loss": 0.344,
2697
+ "step": 384
2698
+ },
2699
+ {
2700
+ "epoch": 0.43,
2701
+ "grad_norm": 0.03674423322081566,
2702
+ "learning_rate": 0.00012270036654795613,
2703
+ "loss": 0.3269,
2704
+ "step": 385
2705
+ },
2706
+ {
2707
+ "epoch": 0.43,
2708
+ "grad_norm": 0.03725577890872955,
2709
+ "learning_rate": 0.00012235175068479984,
2710
+ "loss": 0.3394,
2711
+ "step": 386
2712
+ },
2713
+ {
2714
+ "epoch": 0.44,
2715
+ "grad_norm": 0.037114035338163376,
2716
+ "learning_rate": 0.00012200284865316475,
2717
+ "loss": 0.3411,
2718
+ "step": 387
2719
+ },
2720
+ {
2721
+ "epoch": 0.44,
2722
+ "grad_norm": 0.037100065499544144,
2723
+ "learning_rate": 0.00012165366492002832,
2724
+ "loss": 0.3351,
2725
+ "step": 388
2726
+ },
2727
+ {
2728
+ "epoch": 0.44,
2729
+ "grad_norm": 0.042357347905635834,
2730
+ "learning_rate": 0.00012130420395597437,
2731
+ "loss": 0.3458,
2732
+ "step": 389
2733
+ },
2734
+ {
2735
+ "epoch": 0.44,
2736
+ "grad_norm": 0.03712886944413185,
2737
+ "learning_rate": 0.0001209544702351363,
2738
+ "loss": 0.3385,
2739
+ "step": 390
2740
+ },
2741
+ {
2742
+ "epoch": 0.44,
2743
+ "grad_norm": 0.03989335149526596,
2744
+ "learning_rate": 0.00012060446823513949,
2745
+ "loss": 0.3431,
2746
+ "step": 391
2747
+ },
2748
+ {
2749
+ "epoch": 0.44,
2750
+ "grad_norm": 0.03781385347247124,
2751
+ "learning_rate": 0.0001202542024370441,
2752
+ "loss": 0.3444,
2753
+ "step": 392
2754
+ },
2755
+ {
2756
+ "epoch": 0.44,
2757
+ "grad_norm": 0.03825078904628754,
2758
+ "learning_rate": 0.00011990367732528773,
2759
+ "loss": 0.3394,
2760
+ "step": 393
2761
+ },
2762
+ {
2763
+ "epoch": 0.44,
2764
+ "grad_norm": 0.03969776630401611,
2765
+ "learning_rate": 0.00011955289738762796,
2766
+ "loss": 0.3444,
2767
+ "step": 394
2768
+ },
2769
+ {
2770
+ "epoch": 0.44,
2771
+ "grad_norm": 0.04011238366365433,
2772
+ "learning_rate": 0.00011920186711508479,
2773
+ "loss": 0.3466,
2774
+ "step": 395
2775
+ },
2776
+ {
2777
+ "epoch": 0.45,
2778
+ "grad_norm": 0.03450911492109299,
2779
+ "learning_rate": 0.00011885059100188341,
2780
+ "loss": 0.337,
2781
+ "step": 396
2782
+ },
2783
+ {
2784
+ "epoch": 0.45,
2785
+ "grad_norm": 0.038461893796920776,
2786
+ "learning_rate": 0.00011849907354539633,
2787
+ "loss": 0.3347,
2788
+ "step": 397
2789
+ },
2790
+ {
2791
+ "epoch": 0.45,
2792
+ "grad_norm": 0.037344567477703094,
2793
+ "learning_rate": 0.00011814731924608616,
2794
+ "loss": 0.3452,
2795
+ "step": 398
2796
+ },
2797
+ {
2798
+ "epoch": 0.45,
2799
+ "grad_norm": 0.03558855876326561,
2800
+ "learning_rate": 0.00011779533260744757,
2801
+ "loss": 0.3401,
2802
+ "step": 399
2803
+ },
2804
+ {
2805
+ "epoch": 0.45,
2806
+ "grad_norm": 0.04183749854564667,
2807
+ "learning_rate": 0.00011744311813595006,
2808
+ "loss": 0.3421,
2809
+ "step": 400
2810
+ },
2811
+ {
2812
+ "epoch": 0.45,
2813
+ "grad_norm": 0.03574857488274574,
2814
+ "learning_rate": 0.00011709068034097997,
2815
+ "loss": 0.3271,
2816
+ "step": 401
2817
+ },
2818
+ {
2819
+ "epoch": 0.45,
2820
+ "grad_norm": 0.03993195295333862,
2821
+ "learning_rate": 0.0001167380237347828,
2822
+ "loss": 0.3267,
2823
+ "step": 402
2824
+ },
2825
+ {
2826
+ "epoch": 0.45,
2827
+ "grad_norm": 0.03541812300682068,
2828
+ "learning_rate": 0.0001163851528324056,
2829
+ "loss": 0.3347,
2830
+ "step": 403
2831
+ },
2832
+ {
2833
+ "epoch": 0.45,
2834
+ "grad_norm": 0.037207312881946564,
2835
+ "learning_rate": 0.00011603207215163894,
2836
+ "loss": 0.3418,
2837
+ "step": 404
2838
+ },
2839
+ {
2840
+ "epoch": 0.46,
2841
+ "grad_norm": 0.03919894993305206,
2842
+ "learning_rate": 0.00011567878621295922,
2843
+ "loss": 0.3516,
2844
+ "step": 405
2845
+ },
2846
+ {
2847
+ "epoch": 0.46,
2848
+ "grad_norm": 0.0382562056183815,
2849
+ "learning_rate": 0.00011532529953947075,
2850
+ "loss": 0.3499,
2851
+ "step": 406
2852
+ },
2853
+ {
2854
+ "epoch": 0.46,
2855
+ "grad_norm": 0.03956792503595352,
2856
+ "learning_rate": 0.00011497161665684784,
2857
+ "loss": 0.347,
2858
+ "step": 407
2859
+ },
2860
+ {
2861
+ "epoch": 0.46,
2862
+ "grad_norm": 0.03666992112994194,
2863
+ "learning_rate": 0.0001146177420932768,
2864
+ "loss": 0.3292,
2865
+ "step": 408
2866
+ },
2867
+ {
2868
+ "epoch": 0.46,
2869
+ "grad_norm": 0.03818726912140846,
2870
+ "learning_rate": 0.00011426368037939813,
2871
+ "loss": 0.346,
2872
+ "step": 409
2873
+ },
2874
+ {
2875
+ "epoch": 0.46,
2876
+ "grad_norm": 0.0352170504629612,
2877
+ "learning_rate": 0.00011390943604824826,
2878
+ "loss": 0.3253,
2879
+ "step": 410
2880
+ },
2881
+ {
2882
+ "epoch": 0.46,
2883
+ "grad_norm": 0.04157968610525131,
2884
+ "learning_rate": 0.00011355501363520185,
2885
+ "loss": 0.3411,
2886
+ "step": 411
2887
+ },
2888
+ {
2889
+ "epoch": 0.46,
2890
+ "grad_norm": 0.037365157157182693,
2891
+ "learning_rate": 0.00011320041767791336,
2892
+ "loss": 0.329,
2893
+ "step": 412
2894
+ },
2895
+ {
2896
+ "epoch": 0.46,
2897
+ "grad_norm": 0.03966519236564636,
2898
+ "learning_rate": 0.00011284565271625922,
2899
+ "loss": 0.3412,
2900
+ "step": 413
2901
+ },
2902
+ {
2903
+ "epoch": 0.47,
2904
+ "grad_norm": 0.037126705050468445,
2905
+ "learning_rate": 0.00011249072329227959,
2906
+ "loss": 0.3387,
2907
+ "step": 414
2908
+ },
2909
+ {
2910
+ "epoch": 0.47,
2911
+ "grad_norm": 0.03643345460295677,
2912
+ "learning_rate": 0.00011213563395012026,
2913
+ "loss": 0.3392,
2914
+ "step": 415
2915
+ },
2916
+ {
2917
+ "epoch": 0.47,
2918
+ "grad_norm": 0.03495652228593826,
2919
+ "learning_rate": 0.0001117803892359744,
2920
+ "loss": 0.3327,
2921
+ "step": 416
2922
+ },
2923
+ {
2924
+ "epoch": 0.47,
2925
+ "grad_norm": 0.03979011997580528,
2926
+ "learning_rate": 0.00011142499369802444,
2927
+ "loss": 0.3298,
2928
+ "step": 417
2929
+ },
2930
+ {
2931
+ "epoch": 0.47,
2932
+ "grad_norm": 0.03582910820841789,
2933
+ "learning_rate": 0.00011106945188638378,
2934
+ "loss": 0.3424,
2935
+ "step": 418
2936
+ },
2937
+ {
2938
+ "epoch": 0.47,
2939
+ "grad_norm": 0.03996811434626579,
2940
+ "learning_rate": 0.00011071376835303858,
2941
+ "loss": 0.3335,
2942
+ "step": 419
2943
+ },
2944
+ {
2945
+ "epoch": 0.47,
2946
+ "grad_norm": 0.039901167154312134,
2947
+ "learning_rate": 0.00011035794765178941,
2948
+ "loss": 0.3333,
2949
+ "step": 420
2950
+ },
2951
+ {
2952
+ "epoch": 0.47,
2953
+ "grad_norm": 0.03835071623325348,
2954
+ "learning_rate": 0.00011000199433819305,
2955
+ "loss": 0.328,
2956
+ "step": 421
2957
+ },
2958
+ {
2959
+ "epoch": 0.47,
2960
+ "grad_norm": 0.03873496130108833,
2961
+ "learning_rate": 0.00010964591296950406,
2962
+ "loss": 0.3414,
2963
+ "step": 422
2964
+ },
2965
+ {
2966
+ "epoch": 0.48,
2967
+ "grad_norm": 0.03686109930276871,
2968
+ "learning_rate": 0.00010928970810461652,
2969
+ "loss": 0.3393,
2970
+ "step": 423
2971
+ },
2972
+ {
2973
+ "epoch": 0.48,
2974
+ "grad_norm": 0.038315027952194214,
2975
+ "learning_rate": 0.00010893338430400562,
2976
+ "loss": 0.3349,
2977
+ "step": 424
2978
+ },
2979
+ {
2980
+ "epoch": 0.48,
2981
+ "grad_norm": 0.036203257739543915,
2982
+ "learning_rate": 0.0001085769461296692,
2983
+ "loss": 0.332,
2984
+ "step": 425
2985
+ },
2986
+ {
2987
+ "epoch": 0.48,
2988
+ "grad_norm": 0.037686560302972794,
2989
+ "learning_rate": 0.00010822039814506964,
2990
+ "loss": 0.3437,
2991
+ "step": 426
2992
+ },
2993
+ {
2994
+ "epoch": 0.48,
2995
+ "grad_norm": 0.0374578982591629,
2996
+ "learning_rate": 0.00010786374491507494,
2997
+ "loss": 0.3424,
2998
+ "step": 427
2999
+ },
3000
+ {
3001
+ "epoch": 0.48,
3002
+ "grad_norm": 0.03514384478330612,
3003
+ "learning_rate": 0.00010750699100590076,
3004
+ "loss": 0.3317,
3005
+ "step": 428
3006
+ },
3007
+ {
3008
+ "epoch": 0.48,
3009
+ "grad_norm": 0.040791869163513184,
3010
+ "learning_rate": 0.00010715014098505162,
3011
+ "loss": 0.3389,
3012
+ "step": 429
3013
+ },
3014
+ {
3015
+ "epoch": 0.48,
3016
+ "grad_norm": 0.03556138649582863,
3017
+ "learning_rate": 0.00010679319942126264,
3018
+ "loss": 0.3424,
3019
+ "step": 430
3020
+ },
3021
+ {
3022
+ "epoch": 0.48,
3023
+ "grad_norm": 0.040588121861219406,
3024
+ "learning_rate": 0.00010643617088444094,
3025
+ "loss": 0.3418,
3026
+ "step": 431
3027
+ },
3028
+ {
3029
+ "epoch": 0.49,
3030
+ "grad_norm": 0.03443857282400131,
3031
+ "learning_rate": 0.0001060790599456071,
3032
+ "loss": 0.3395,
3033
+ "step": 432
3034
+ },
3035
+ {
3036
+ "epoch": 0.49,
3037
+ "grad_norm": 0.038418177515268326,
3038
+ "learning_rate": 0.00010572187117683674,
3039
+ "loss": 0.3325,
3040
+ "step": 433
3041
+ },
3042
+ {
3043
+ "epoch": 0.49,
3044
+ "grad_norm": 0.03434626758098602,
3045
+ "learning_rate": 0.0001053646091512019,
3046
+ "loss": 0.3378,
3047
+ "step": 434
3048
+ },
3049
+ {
3050
+ "epoch": 0.49,
3051
+ "grad_norm": 0.03498569130897522,
3052
+ "learning_rate": 0.0001050072784427126,
3053
+ "loss": 0.3375,
3054
+ "step": 435
3055
+ },
3056
+ {
3057
+ "epoch": 0.49,
3058
+ "grad_norm": 0.03690873831510544,
3059
+ "learning_rate": 0.00010464988362625812,
3060
+ "loss": 0.331,
3061
+ "step": 436
3062
+ },
3063
+ {
3064
+ "epoch": 0.49,
3065
+ "grad_norm": 0.040180571377277374,
3066
+ "learning_rate": 0.00010429242927754854,
3067
+ "loss": 0.3394,
3068
+ "step": 437
3069
+ },
3070
+ {
3071
+ "epoch": 0.49,
3072
+ "grad_norm": 0.03546585142612457,
3073
+ "learning_rate": 0.00010393491997305613,
3074
+ "loss": 0.3368,
3075
+ "step": 438
3076
+ },
3077
+ {
3078
+ "epoch": 0.49,
3079
+ "grad_norm": 0.03656655550003052,
3080
+ "learning_rate": 0.00010357736028995677,
3081
+ "loss": 0.332,
3082
+ "step": 439
3083
+ },
3084
+ {
3085
+ "epoch": 0.5,
3086
+ "grad_norm": 0.03666246309876442,
3087
+ "learning_rate": 0.00010321975480607129,
3088
+ "loss": 0.3342,
3089
+ "step": 440
3090
+ },
3091
+ {
3092
+ "epoch": 0.5,
3093
+ "grad_norm": 0.03500663861632347,
3094
+ "learning_rate": 0.00010286210809980697,
3095
+ "loss": 0.3389,
3096
+ "step": 441
3097
+ },
3098
+ {
3099
+ "epoch": 0.5,
3100
+ "grad_norm": 0.03510987013578415,
3101
+ "learning_rate": 0.0001025044247500988,
3102
+ "loss": 0.3338,
3103
+ "step": 442
3104
+ },
3105
+ {
3106
+ "epoch": 0.5,
3107
+ "grad_norm": 0.037753257900476456,
3108
+ "learning_rate": 0.00010214670933635095,
3109
+ "loss": 0.333,
3110
+ "step": 443
3111
+ },
3112
+ {
3113
+ "epoch": 0.5,
3114
+ "grad_norm": 0.03440658003091812,
3115
+ "learning_rate": 0.00010178896643837809,
3116
+ "loss": 0.3456,
3117
+ "step": 444
3118
+ },
3119
+ {
3120
+ "epoch": 0.5,
3121
+ "grad_norm": 0.0362187922000885,
3122
+ "learning_rate": 0.00010143120063634681,
3123
+ "loss": 0.3484,
3124
+ "step": 445
3125
+ },
3126
+ {
3127
+ "epoch": 0.5,
3128
+ "grad_norm": 0.034865692257881165,
3129
+ "learning_rate": 0.00010107341651071684,
3130
+ "loss": 0.3392,
3131
+ "step": 446
3132
+ },
3133
+ {
3134
+ "epoch": 0.5,
3135
+ "grad_norm": 0.035264838486909866,
3136
+ "learning_rate": 0.00010071561864218262,
3137
+ "loss": 0.3267,
3138
+ "step": 447
3139
+ },
3140
+ {
3141
+ "epoch": 0.5,
3142
+ "grad_norm": 0.035725150257349014,
3143
+ "learning_rate": 0.00010035781161161446,
3144
+ "loss": 0.3454,
3145
+ "step": 448
3146
+ },
3147
+ {
3148
+ "epoch": 0.51,
3149
+ "grad_norm": 0.03839932009577751,
3150
+ "learning_rate": 0.0001,
3151
+ "loss": 0.3304,
3152
+ "step": 449
3153
+ },
3154
+ {
3155
+ "epoch": 0.51,
3156
+ "grad_norm": 0.03366202488541603,
3157
+ "learning_rate": 9.964218838838554e-05,
3158
+ "loss": 0.3288,
3159
+ "step": 450
3160
+ },
3161
+ {
3162
+ "epoch": 0.51,
3163
+ "grad_norm": 0.03588353097438812,
3164
+ "learning_rate": 9.92843813578174e-05,
3165
+ "loss": 0.3406,
3166
+ "step": 451
3167
+ },
3168
+ {
3169
+ "epoch": 0.51,
3170
+ "grad_norm": 0.034711651504039764,
3171
+ "learning_rate": 9.892658348928316e-05,
3172
+ "loss": 0.3289,
3173
+ "step": 452
3174
+ },
3175
+ {
3176
+ "epoch": 0.51,
3177
+ "grad_norm": 0.03611205145716667,
3178
+ "learning_rate": 9.856879936365321e-05,
3179
+ "loss": 0.3294,
3180
+ "step": 453
3181
+ },
3182
+ {
3183
+ "epoch": 0.51,
3184
+ "grad_norm": 0.03515039384365082,
3185
+ "learning_rate": 9.821103356162189e-05,
3186
+ "loss": 0.3324,
3187
+ "step": 454
3188
+ },
3189
+ {
3190
+ "epoch": 0.51,
3191
+ "grad_norm": 0.03375555947422981,
3192
+ "learning_rate": 9.785329066364906e-05,
3193
+ "loss": 0.3214,
3194
+ "step": 455
3195
+ },
3196
+ {
3197
+ "epoch": 0.51,
3198
+ "grad_norm": 0.03730206564068794,
3199
+ "learning_rate": 9.749557524990121e-05,
3200
+ "loss": 0.3258,
3201
+ "step": 456
3202
+ },
3203
+ {
3204
+ "epoch": 0.51,
3205
+ "grad_norm": 0.033822208642959595,
3206
+ "learning_rate": 9.713789190019304e-05,
3207
+ "loss": 0.3353,
3208
+ "step": 457
3209
+ },
3210
+ {
3211
+ "epoch": 0.52,
3212
+ "grad_norm": 0.03683510422706604,
3213
+ "learning_rate": 9.678024519392871e-05,
3214
+ "loss": 0.3264,
3215
+ "step": 458
3216
+ },
3217
+ {
3218
+ "epoch": 0.52,
3219
+ "grad_norm": 0.037592969834804535,
3220
+ "learning_rate": 9.642263971004324e-05,
3221
+ "loss": 0.329,
3222
+ "step": 459
3223
+ },
3224
+ {
3225
+ "epoch": 0.52,
3226
+ "grad_norm": 0.03537925332784653,
3227
+ "learning_rate": 9.606508002694386e-05,
3228
+ "loss": 0.3448,
3229
+ "step": 460
3230
+ },
3231
+ {
3232
+ "epoch": 0.52,
3233
+ "grad_norm": 0.033904388546943665,
3234
+ "learning_rate": 9.570757072245147e-05,
3235
+ "loss": 0.3472,
3236
+ "step": 461
3237
+ },
3238
+ {
3239
+ "epoch": 0.52,
3240
+ "grad_norm": 0.036394666880369186,
3241
+ "learning_rate": 9.535011637374189e-05,
3242
+ "loss": 0.3288,
3243
+ "step": 462
3244
+ },
3245
+ {
3246
+ "epoch": 0.52,
3247
+ "grad_norm": 0.03391895070672035,
3248
+ "learning_rate": 9.499272155728742e-05,
3249
+ "loss": 0.3368,
3250
+ "step": 463
3251
+ },
3252
+ {
3253
+ "epoch": 0.52,
3254
+ "grad_norm": 0.037903424352407455,
3255
+ "learning_rate": 9.463539084879809e-05,
3256
+ "loss": 0.3443,
3257
+ "step": 464
3258
+ },
3259
+ {
3260
+ "epoch": 0.52,
3261
+ "grad_norm": 0.033752232789993286,
3262
+ "learning_rate": 9.427812882316329e-05,
3263
+ "loss": 0.3316,
3264
+ "step": 465
3265
+ },
3266
+ {
3267
+ "epoch": 0.52,
3268
+ "grad_norm": 0.03620493784546852,
3269
+ "learning_rate": 9.392094005439291e-05,
3270
+ "loss": 0.3318,
3271
+ "step": 466
3272
+ },
3273
+ {
3274
+ "epoch": 0.53,
3275
+ "grad_norm": 0.0346236452460289,
3276
+ "learning_rate": 9.356382911555907e-05,
3277
+ "loss": 0.3438,
3278
+ "step": 467
3279
+ },
3280
+ {
3281
+ "epoch": 0.53,
3282
+ "grad_norm": 0.03393540903925896,
3283
+ "learning_rate": 9.320680057873735e-05,
3284
+ "loss": 0.3459,
3285
+ "step": 468
3286
+ },
3287
+ {
3288
+ "epoch": 0.53,
3289
+ "grad_norm": 0.03790227323770523,
3290
+ "learning_rate": 9.28498590149484e-05,
3291
+ "loss": 0.3405,
3292
+ "step": 469
3293
+ },
3294
+ {
3295
+ "epoch": 0.53,
3296
+ "grad_norm": 0.03373325988650322,
3297
+ "learning_rate": 9.249300899409924e-05,
3298
+ "loss": 0.3329,
3299
+ "step": 470
3300
+ },
3301
+ {
3302
+ "epoch": 0.53,
3303
+ "grad_norm": 0.03829934075474739,
3304
+ "learning_rate": 9.213625508492508e-05,
3305
+ "loss": 0.3321,
3306
+ "step": 471
3307
+ },
3308
+ {
3309
+ "epoch": 0.53,
3310
+ "grad_norm": 0.03592018410563469,
3311
+ "learning_rate": 9.177960185493036e-05,
3312
+ "loss": 0.3186,
3313
+ "step": 472
3314
+ },
3315
+ {
3316
+ "epoch": 0.53,
3317
+ "grad_norm": 0.03398677706718445,
3318
+ "learning_rate": 9.142305387033081e-05,
3319
+ "loss": 0.3465,
3320
+ "step": 473
3321
+ },
3322
+ {
3323
+ "epoch": 0.53,
3324
+ "grad_norm": 0.03493371605873108,
3325
+ "learning_rate": 9.106661569599442e-05,
3326
+ "loss": 0.3358,
3327
+ "step": 474
3328
+ },
3329
+ {
3330
+ "epoch": 0.53,
3331
+ "grad_norm": 0.03722028434276581,
3332
+ "learning_rate": 9.071029189538353e-05,
3333
+ "loss": 0.3278,
3334
+ "step": 475
3335
+ },
3336
+ {
3337
+ "epoch": 0.54,
3338
+ "grad_norm": 0.033628251403570175,
3339
+ "learning_rate": 9.035408703049596e-05,
3340
+ "loss": 0.3315,
3341
+ "step": 476
3342
+ },
3343
+ {
3344
+ "epoch": 0.54,
3345
+ "grad_norm": 0.036100320518016815,
3346
+ "learning_rate": 8.9998005661807e-05,
3347
+ "loss": 0.322,
3348
+ "step": 477
3349
+ },
3350
+ {
3351
+ "epoch": 0.54,
3352
+ "grad_norm": 0.03419538959860802,
3353
+ "learning_rate": 8.96420523482106e-05,
3354
+ "loss": 0.3241,
3355
+ "step": 478
3356
+ },
3357
+ {
3358
+ "epoch": 0.54,
3359
+ "grad_norm": 0.03728823736310005,
3360
+ "learning_rate": 8.928623164696146e-05,
3361
+ "loss": 0.3312,
3362
+ "step": 479
3363
+ },
3364
+ {
3365
+ "epoch": 0.54,
3366
+ "grad_norm": 0.03538035228848457,
3367
+ "learning_rate": 8.893054811361624e-05,
3368
+ "loss": 0.3244,
3369
+ "step": 480
3370
+ },
3371
+ {
3372
+ "epoch": 0.54,
3373
+ "grad_norm": 0.036142975091934204,
3374
+ "learning_rate": 8.85750063019756e-05,
3375
+ "loss": 0.3401,
3376
+ "step": 481
3377
+ },
3378
+ {
3379
+ "epoch": 0.54,
3380
+ "grad_norm": 0.0373205728828907,
3381
+ "learning_rate": 8.821961076402563e-05,
3382
+ "loss": 0.3334,
3383
+ "step": 482
3384
+ },
3385
+ {
3386
+ "epoch": 0.54,
3387
+ "grad_norm": 0.035391587764024734,
3388
+ "learning_rate": 8.786436604987978e-05,
3389
+ "loss": 0.341,
3390
+ "step": 483
3391
+ },
3392
+ {
3393
+ "epoch": 0.54,
3394
+ "grad_norm": 0.03749712184071541,
3395
+ "learning_rate": 8.750927670772044e-05,
3396
+ "loss": 0.3179,
3397
+ "step": 484
3398
+ },
3399
+ {
3400
+ "epoch": 0.55,
3401
+ "grad_norm": 0.037055715918540955,
3402
+ "learning_rate": 8.715434728374083e-05,
3403
+ "loss": 0.3226,
3404
+ "step": 485
3405
+ },
3406
+ {
3407
+ "epoch": 0.55,
3408
+ "grad_norm": 0.03762289509177208,
3409
+ "learning_rate": 8.679958232208668e-05,
3410
+ "loss": 0.337,
3411
+ "step": 486
3412
+ },
3413
+ {
3414
+ "epoch": 0.55,
3415
+ "grad_norm": 0.03537070378661156,
3416
+ "learning_rate": 8.644498636479819e-05,
3417
+ "loss": 0.3306,
3418
+ "step": 487
3419
+ },
3420
+ {
3421
+ "epoch": 0.55,
3422
+ "grad_norm": 0.03710364177823067,
3423
+ "learning_rate": 8.609056395175175e-05,
3424
+ "loss": 0.3256,
3425
+ "step": 488
3426
+ },
3427
+ {
3428
+ "epoch": 0.55,
3429
+ "grad_norm": 0.03595058619976044,
3430
+ "learning_rate": 8.573631962060192e-05,
3431
+ "loss": 0.3337,
3432
+ "step": 489
3433
+ },
3434
+ {
3435
+ "epoch": 0.55,
3436
+ "grad_norm": 0.03377842903137207,
3437
+ "learning_rate": 8.538225790672322e-05,
3438
+ "loss": 0.3411,
3439
+ "step": 490
3440
+ },
3441
+ {
3442
+ "epoch": 0.55,
3443
+ "grad_norm": 0.036737628281116486,
3444
+ "learning_rate": 8.50283833431522e-05,
3445
+ "loss": 0.3286,
3446
+ "step": 491
3447
+ },
3448
+ {
3449
+ "epoch": 0.55,
3450
+ "grad_norm": 0.034495435655117035,
3451
+ "learning_rate": 8.467470046052927e-05,
3452
+ "loss": 0.3345,
3453
+ "step": 492
3454
+ },
3455
+ {
3456
+ "epoch": 0.55,
3457
+ "grad_norm": 0.036502912640571594,
3458
+ "learning_rate": 8.432121378704081e-05,
3459
+ "loss": 0.3237,
3460
+ "step": 493
3461
+ },
3462
+ {
3463
+ "epoch": 0.56,
3464
+ "grad_norm": 0.03458232432603836,
3465
+ "learning_rate": 8.396792784836108e-05,
3466
+ "loss": 0.3215,
3467
+ "step": 494
3468
+ },
3469
+ {
3470
+ "epoch": 0.56,
3471
+ "grad_norm": 0.036796458065509796,
3472
+ "learning_rate": 8.361484716759445e-05,
3473
+ "loss": 0.3375,
3474
+ "step": 495
3475
+ },
3476
+ {
3477
+ "epoch": 0.56,
3478
+ "grad_norm": 0.03392602503299713,
3479
+ "learning_rate": 8.326197626521723e-05,
3480
+ "loss": 0.335,
3481
+ "step": 496
3482
+ },
3483
+ {
3484
+ "epoch": 0.56,
3485
+ "grad_norm": 0.03623703867197037,
3486
+ "learning_rate": 8.290931965902008e-05,
3487
+ "loss": 0.3388,
3488
+ "step": 497
3489
+ },
3490
+ {
3491
+ "epoch": 0.56,
3492
+ "grad_norm": 0.03508217632770538,
3493
+ "learning_rate": 8.255688186404996e-05,
3494
+ "loss": 0.3267,
3495
+ "step": 498
3496
+ },
3497
+ {
3498
+ "epoch": 0.56,
3499
+ "grad_norm": 0.03563014790415764,
3500
+ "learning_rate": 8.220466739255244e-05,
3501
+ "loss": 0.3277,
3502
+ "step": 499
3503
+ },
3504
+ {
3505
+ "epoch": 0.56,
3506
+ "grad_norm": 0.035729020833969116,
3507
+ "learning_rate": 8.185268075391388e-05,
3508
+ "loss": 0.3348,
3509
+ "step": 500
3510
+ },
3511
+ {
3512
+ "epoch": 0.56,
3513
+ "grad_norm": 0.034020714461803436,
3514
+ "learning_rate": 8.150092645460366e-05,
3515
+ "loss": 0.32,
3516
+ "step": 501
3517
+ },
3518
+ {
3519
+ "epoch": 0.56,
3520
+ "grad_norm": 0.033443402498960495,
3521
+ "learning_rate": 8.114940899811662e-05,
3522
+ "loss": 0.3402,
3523
+ "step": 502
3524
+ },
3525
+ {
3526
+ "epoch": 0.57,
3527
+ "grad_norm": 0.03387276828289032,
3528
+ "learning_rate": 8.079813288491521e-05,
3529
+ "loss": 0.3319,
3530
+ "step": 503
3531
+ },
3532
+ {
3533
+ "epoch": 0.57,
3534
+ "grad_norm": 0.033886682242155075,
3535
+ "learning_rate": 8.044710261237207e-05,
3536
+ "loss": 0.3307,
3537
+ "step": 504
3538
+ },
3539
+ {
3540
+ "epoch": 0.57,
3541
+ "grad_norm": 0.03618338331580162,
3542
+ "learning_rate": 8.009632267471227e-05,
3543
+ "loss": 0.3286,
3544
+ "step": 505
3545
+ },
3546
+ {
3547
+ "epoch": 0.57,
3548
+ "grad_norm": 0.03527137264609337,
3549
+ "learning_rate": 7.974579756295591e-05,
3550
+ "loss": 0.3171,
3551
+ "step": 506
3552
+ },
3553
+ {
3554
+ "epoch": 0.57,
3555
+ "grad_norm": 0.036220818758010864,
3556
+ "learning_rate": 7.939553176486052e-05,
3557
+ "loss": 0.3396,
3558
+ "step": 507
3559
+ },
3560
+ {
3561
+ "epoch": 0.57,
3562
+ "grad_norm": 0.035215239971876144,
3563
+ "learning_rate": 7.904552976486372e-05,
3564
+ "loss": 0.3226,
3565
+ "step": 508
3566
+ },
3567
+ {
3568
+ "epoch": 0.57,
3569
+ "grad_norm": 0.0399019829928875,
3570
+ "learning_rate": 7.869579604402562e-05,
3571
+ "loss": 0.3245,
3572
+ "step": 509
3573
+ },
3574
+ {
3575
+ "epoch": 0.57,
3576
+ "grad_norm": 0.03600882738828659,
3577
+ "learning_rate": 7.83463350799717e-05,
3578
+ "loss": 0.3217,
3579
+ "step": 510
3580
+ },
3581
+ {
3582
+ "epoch": 0.57,
3583
+ "grad_norm": 0.04228314757347107,
3584
+ "learning_rate": 7.799715134683523e-05,
3585
+ "loss": 0.3282,
3586
+ "step": 511
3587
+ },
3588
+ {
3589
+ "epoch": 0.58,
3590
+ "grad_norm": 0.03412134200334549,
3591
+ "learning_rate": 7.764824931520018e-05,
3592
+ "loss": 0.3297,
3593
+ "step": 512
3594
+ },
3595
+ {
3596
+ "epoch": 0.58,
3597
+ "grad_norm": 0.03876680135726929,
3598
+ "learning_rate": 7.729963345204386e-05,
3599
+ "loss": 0.3232,
3600
+ "step": 513
3601
+ },
3602
+ {
3603
+ "epoch": 0.58,
3604
+ "grad_norm": 0.034336164593696594,
3605
+ "learning_rate": 7.695130822067984e-05,
3606
+ "loss": 0.32,
3607
+ "step": 514
3608
+ },
3609
+ {
3610
+ "epoch": 0.58,
3611
+ "grad_norm": 0.0393151193857193,
3612
+ "learning_rate": 7.660327808070064e-05,
3613
+ "loss": 0.3321,
3614
+ "step": 515
3615
+ },
3616
+ {
3617
+ "epoch": 0.58,
3618
+ "grad_norm": 0.03484569489955902,
3619
+ "learning_rate": 7.625554748792085e-05,
3620
+ "loss": 0.3315,
3621
+ "step": 516
3622
+ },
3623
+ {
3624
+ "epoch": 0.58,
3625
+ "grad_norm": 0.042867034673690796,
3626
+ "learning_rate": 7.59081208943199e-05,
3627
+ "loss": 0.3294,
3628
+ "step": 517
3629
+ },
3630
+ {
3631
+ "epoch": 0.58,
3632
+ "grad_norm": 0.03290384262800217,
3633
+ "learning_rate": 7.556100274798519e-05,
3634
+ "loss": 0.3284,
3635
+ "step": 518
3636
+ },
3637
+ {
3638
+ "epoch": 0.58,
3639
+ "grad_norm": 0.042282383888959885,
3640
+ "learning_rate": 7.521419749305497e-05,
3641
+ "loss": 0.3304,
3642
+ "step": 519
3643
+ },
3644
+ {
3645
+ "epoch": 0.59,
3646
+ "grad_norm": 0.03494780510663986,
3647
+ "learning_rate": 7.486770956966171e-05,
3648
+ "loss": 0.3378,
3649
+ "step": 520
3650
+ },
3651
+ {
3652
+ "epoch": 0.59,
3653
+ "grad_norm": 0.04240773990750313,
3654
+ "learning_rate": 7.452154341387493e-05,
3655
+ "loss": 0.3325,
3656
+ "step": 521
3657
+ },
3658
+ {
3659
+ "epoch": 0.59,
3660
+ "grad_norm": 0.037106744945049286,
3661
+ "learning_rate": 7.417570345764481e-05,
3662
+ "loss": 0.3267,
3663
+ "step": 522
3664
+ },
3665
+ {
3666
+ "epoch": 0.59,
3667
+ "grad_norm": 0.0382254458963871,
3668
+ "learning_rate": 7.383019412874489e-05,
3669
+ "loss": 0.3225,
3670
+ "step": 523
3671
+ },
3672
+ {
3673
+ "epoch": 0.59,
3674
+ "grad_norm": 0.037587832659482956,
3675
+ "learning_rate": 7.348501985071603e-05,
3676
+ "loss": 0.3197,
3677
+ "step": 524
3678
+ },
3679
+ {
3680
+ "epoch": 0.59,
3681
+ "grad_norm": 0.03936656191945076,
3682
+ "learning_rate": 7.314018504280917e-05,
3683
+ "loss": 0.3268,
3684
+ "step": 525
3685
+ },
3686
+ {
3687
+ "epoch": 0.59,
3688
+ "grad_norm": 0.037244342267513275,
3689
+ "learning_rate": 7.279569411992926e-05,
3690
+ "loss": 0.3254,
3691
+ "step": 526
3692
+ },
3693
+ {
3694
+ "epoch": 0.59,
3695
+ "grad_norm": 0.03609207645058632,
3696
+ "learning_rate": 7.24515514925783e-05,
3697
+ "loss": 0.3388,
3698
+ "step": 527
3699
+ },
3700
+ {
3701
+ "epoch": 0.59,
3702
+ "grad_norm": 0.03564932942390442,
3703
+ "learning_rate": 7.210776156679931e-05,
3704
+ "loss": 0.3307,
3705
+ "step": 528
3706
+ },
3707
+ {
3708
+ "epoch": 0.6,
3709
+ "grad_norm": 0.03292916342616081,
3710
+ "learning_rate": 7.176432874411941e-05,
3711
+ "loss": 0.3257,
3712
+ "step": 529
3713
+ },
3714
+ {
3715
+ "epoch": 0.6,
3716
+ "grad_norm": 0.0374549925327301,
3717
+ "learning_rate": 7.1421257421494e-05,
3718
+ "loss": 0.3417,
3719
+ "step": 530
3720
+ },
3721
+ {
3722
+ "epoch": 0.6,
3723
+ "grad_norm": 0.03253106027841568,
3724
+ "learning_rate": 7.107855199125002e-05,
3725
+ "loss": 0.3397,
3726
+ "step": 531
3727
+ },
3728
+ {
3729
+ "epoch": 0.6,
3730
+ "grad_norm": 0.035386502742767334,
3731
+ "learning_rate": 7.073621684103007e-05,
3732
+ "loss": 0.3329,
3733
+ "step": 532
3734
+ },
3735
+ {
3736
+ "epoch": 0.6,
3737
+ "grad_norm": 0.032128430902957916,
3738
+ "learning_rate": 7.039425635373589e-05,
3739
+ "loss": 0.3314,
3740
+ "step": 533
3741
+ },
3742
+ {
3743
+ "epoch": 0.6,
3744
+ "grad_norm": 0.03548587113618851,
3745
+ "learning_rate": 7.005267490747263e-05,
3746
+ "loss": 0.3267,
3747
+ "step": 534
3748
+ }
3749
+ ],
3750
+ "logging_steps": 1,
3751
+ "max_steps": 888,
3752
+ "num_input_tokens_seen": 0,
3753
+ "num_train_epochs": 1,
3754
+ "save_steps": 89,
3755
+ "total_flos": 1.065190361577554e+19,
3756
+ "train_batch_size": 14,
3757
+ "trial_name": null,
3758
+ "trial_params": null
3759
+ }
checkpoint-534/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0441fddfec498f33a86d551b67802f8f8fbf45a9463d322464e8ee1ed737775
3
+ size 5688
checkpoint-623/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: meta-llama/Llama-2-7b-hf
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.9.0
checkpoint-623/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 128,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 256,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "q_proj",
24
+ "o_proj",
25
+ "up_proj",
26
+ "v_proj",
27
+ "gate_proj",
28
+ "down_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
checkpoint-623/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf4a5e606db7773e8e82599c06bb4fa4948aeabebd5a956786ef2dfa62d4885a
3
+ size 2558587064
checkpoint-623/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e73db1184b0c4113bd8f959b0c8f5a854682219d19a27e5e6c71463b4288abc3
3
+ size 1282290004
checkpoint-623/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa2a89a7ed9e61c843edf8209e09605549f1d810132870ad02c9ab775929e75b
3
+ size 14512
checkpoint-623/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7009b1a2c5014bad2ad34d1b799e2f887a5be1420babc9788ad26e34cda61ed7
3
+ size 14512
checkpoint-623/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2839f2856d98883d5b86461a5c90ace9c9a4b49cc0b13bdb1c175254259099ea
3
+ size 1064
checkpoint-623/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-623/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-623/tokenizer_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": false,
35
+ "model_max_length": 1000000000000000019884624838656,
36
+ "pad_token": "</s>",
37
+ "padding_side": "right",
38
+ "sp_model_kwargs": {},
39
+ "spaces_between_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false,
43
+ "use_fast": true
44
+ }
checkpoint-623/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-623/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0441fddfec498f33a86d551b67802f8f8fbf45a9463d322464e8ee1ed737775
3
+ size 5688
checkpoint-712/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: meta-llama/Llama-2-7b-hf
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.9.0
checkpoint-712/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 128,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 256,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "q_proj",
24
+ "o_proj",
25
+ "up_proj",
26
+ "v_proj",
27
+ "gate_proj",
28
+ "down_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
checkpoint-712/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae9bde5b6e84011d900b84eb3a1236a375458481d970f14705fb5e96e7ae1866
3
+ size 2558587064
checkpoint-712/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9577628b036b9664397157482a1ef4cabaa66e7af815b50694cbf86f00a4010
3
+ size 1282290004
checkpoint-712/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55f3bf3630a619cfbcf68422929655380a25b658a0849b69700f222278e42414
3
+ size 14512
checkpoint-712/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6164a6b237f262e8746e2c2ad7ddb78854645a2c88596ebdf13ed76443c1e445
3
+ size 14512
checkpoint-712/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e482a6fb883c3fe32af8b8a791ff0944b0f976604825691001f5564002009129
3
+ size 1064
checkpoint-712/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-712/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-712/tokenizer_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": false,
35
+ "model_max_length": 1000000000000000019884624838656,
36
+ "pad_token": "</s>",
37
+ "padding_side": "right",
38
+ "sp_model_kwargs": {},
39
+ "spaces_between_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false,
43
+ "use_fast": true
44
+ }
checkpoint-712/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-712/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0441fddfec498f33a86d551b67802f8f8fbf45a9463d322464e8ee1ed737775
3
+ size 5688
checkpoint-801/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: meta-llama/Llama-2-7b-hf
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.9.0
checkpoint-801/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 128,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 256,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "q_proj",
24
+ "o_proj",
25
+ "up_proj",
26
+ "v_proj",
27
+ "gate_proj",
28
+ "down_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
checkpoint-801/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66f46df362b92d3c396a5e0310393f49a3373f3015516cbe7efb395ad8b62245
3
+ size 2558587064
checkpoint-801/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76a893b516d63ab1c44310fc645180bac0883578861562dfbcd2d9585124a392
3
+ size 1282290004
checkpoint-801/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92e6779d9edc0484f276f9adfc132a2908cff2dbd4a9b49a336d66b16e1cf310
3
+ size 14512
checkpoint-801/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4888e000ab16b4bd93988aa6c74ce5cf4c107b56eb9aae0abace27eb5de54b33
3
+ size 14512
checkpoint-801/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebc49d548ba20069d9e5435e61a8f67a4b6e68ec3500b949db85d9aac43a2807
3
+ size 1064
checkpoint-801/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-801/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-801/tokenizer_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": false,
35
+ "model_max_length": 1000000000000000019884624838656,
36
+ "pad_token": "</s>",
37
+ "padding_side": "right",
38
+ "sp_model_kwargs": {},
39
+ "spaces_between_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false,
43
+ "use_fast": true
44
+ }
checkpoint-801/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff