Hugging Face's logo

myownip
/
axolotl-openllama-1k-qlora-v03

PEFT
TensorBoard
Safetensors
llama
Generated from Trainer
4-bit precision
bitsandbytes
Model card Files Metrics Community
myownip commited on
Commit
7562595
•
1 Parent(s): 6831457

Upload folder using huggingface_hub

Browse files
Files changed (41) hide show
  1. README.md +147 -0
  2. adapter_config.json +33 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-398/README.md +202 -0
  5. checkpoint-398/adapter_config.json +33 -0
  6. checkpoint-398/adapter_model.safetensors +3 -0
  7. checkpoint-398/optimizer.pt +3 -0
  8. checkpoint-398/rng_state.pth +3 -0
  9. checkpoint-398/scheduler.pt +3 -0
  10. checkpoint-398/trainer_state.json +2871 -0
  11. checkpoint-398/training_args.bin +3 -0
  12. checkpoint-431/README.md +202 -0
  13. checkpoint-431/adapter_config.json +33 -0
  14. checkpoint-431/adapter_model.safetensors +3 -0
  15. checkpoint-431/optimizer.pt +3 -0
  16. checkpoint-431/rng_state.pth +3 -0
  17. checkpoint-431/scheduler.pt +3 -0
  18. checkpoint-431/trainer_state.json +3070 -0
  19. checkpoint-431/training_args.bin +3 -0
  20. checkpoint-597/README.md +202 -0
  21. checkpoint-597/adapter_config.json +33 -0
  22. checkpoint-597/adapter_model.safetensors +3 -0
  23. checkpoint-597/optimizer.pt +3 -0
  24. checkpoint-597/rng_state.pth +3 -0
  25. checkpoint-597/scheduler.pt +3 -0
  26. checkpoint-597/trainer_state.json +0 -0
  27. checkpoint-597/training_args.bin +3 -0
  28. checkpoint-796/README.md +202 -0
  29. checkpoint-796/adapter_config.json +33 -0
  30. checkpoint-796/adapter_model.safetensors +3 -0
  31. checkpoint-796/optimizer.pt +3 -0
  32. checkpoint-796/rng_state.pth +3 -0
  33. checkpoint-796/scheduler.pt +3 -0
  34. checkpoint-796/trainer_state.json +0 -0
  35. checkpoint-796/training_args.bin +3 -0
  36. config.json +43 -0
  37. runs/Mar13_21-58-24_8711e78fac20/events.out.tfevents.1710367104.8711e78fac20.40.0 +3 -0
  38. runs/Mar13_22-06-09_8711e78fac20/events.out.tfevents.1710367570.8711e78fac20.172.0 +3 -0
  39. special_tokens_map.json +24 -0
  40. tokenizer.model +3 -0
  41. tokenizer_config.json +43 -0
README.md ADDED
@@ -0,0 +1,147 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - generated_from_trainer
6
+ base_model: openlm-research/open_llama_3b_v2
7
+ model-index:
8
+ - name: qlora-out
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ [<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
16
+ <details><summary>See axolotl config</summary>
17
+
18
+ axolotl version: `0.4.0`
19
+ ```yaml
20
+ base_model: openlm-research/open_llama_3b_v2
21
+ model_type: LlamaForCausalLM
22
+ tokenizer_type: LlamaTokenizer
23
+ load_in_8bit: false
24
+ load_in_4bit: true
25
+ strict: false
26
+ push_dataset_to_hub:
27
+ datasets:
28
+ - path: mhenrichsen/alpaca_2k_test
29
+ type: alpaca
30
+ dataset_prepared_path:
31
+ val_set_size: 0.05
32
+ adapter: qlora
33
+ lora_model_dir:
34
+ sequence_len: 1024
35
+ sample_packing: true
36
+ lora_r: 8
37
+ lora_alpha: 32
38
+ lora_dropout: 0.05
39
+ lora_target_modules:
40
+ lora_target_linear: true
41
+ lora_fan_in_fan_out:
42
+ wandb_project:
43
+ wandb_entity:
44
+ wandb_watch:
45
+ wandb_name:
46
+ wandb_log_model:
47
+ output_dir: ./qlora-out
48
+ gradient_accumulation_steps: 1
49
+ micro_batch_size: 2
50
+ num_epochs: 4
51
+ optimizer: paged_adamw_32bit
52
+ torchdistx_path:
53
+ lr_scheduler: cosine
54
+ learning_rate: 0.0002
55
+ train_on_inputs: false
56
+ group_by_length: false
57
+ bf16: false
58
+ fp16: true
59
+ tf32: false
60
+ gradient_checkpointing: true
61
+ early_stopping_patience:
62
+ resume_from_checkpoint:
63
+ local_rank:
64
+ logging_steps: 1
65
+ xformers_attention:
66
+ flash_attention: true
67
+ gptq_groupsize:
68
+ gptq_model_v1:
69
+ warmup_steps: 20
70
+ evals_per_epoch: 4
71
+ saves_per_epoch: 1
72
+ debug:
73
+ deepspeed:
74
+ weight_decay: 0.1
75
+ fsdp:
76
+ fsdp_config:
77
+ special_tokens:
78
+ bos_token: "<s>"
79
+ eos_token: "</s>"
80
+ unk_token: "<unk>"
81
+
82
+ ```
83
+
84
+ </details><br>
85
+
86
+ # qlora-out
87
+
88
+ This model is a fine-tuned version of [openlm-research/open_llama_3b_v2](https://huggingface.co/openlm-research/open_llama_3b_v2) on the None dataset.
89
+ It achieves the following results on the evaluation set:
90
+ - Loss: 1.4177
91
+
92
+ ## Model description
93
+
94
+ More information needed
95
+
96
+ ## Intended uses & limitations
97
+
98
+ More information needed
99
+
100
+ ## Training and evaluation data
101
+
102
+ More information needed
103
+
104
+ ## Training procedure
105
+
106
+ ### Training hyperparameters
107
+
108
+ The following hyperparameters were used during training:
109
+ - learning_rate: 0.0002
110
+ - train_batch_size: 2
111
+ - eval_batch_size: 2
112
+ - seed: 42
113
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
114
+ - lr_scheduler_type: cosine
115
+ - lr_scheduler_warmup_steps: 20
116
+ - num_epochs: 4
117
+ - mixed_precision_training: Native AMP
118
+
119
+ ### Training results
120
+
121
+ | Training Loss | Epoch | Step | Validation Loss |
122
+ |:-------------:|:-----:|:----:|:---------------:|
123
+ | 1.3026 | 0.01 | 1 | 1.3435 |
124
+ | 1.1146 | 0.25 | 50 | 1.1476 |
125
+ | 1.2387 | 0.5 | 100 | 1.1319 |
126
+ | 1.4159 | 0.75 | 150 | 1.1192 |
127
+ | 1.2807 | 1.01 | 200 | 1.1153 |
128
+ | 1.0465 | 1.24 | 250 | 1.1569 |
129
+ | 0.9577 | 1.49 | 300 | 1.1493 |
130
+ | 1.1257 | 1.74 | 350 | 1.1462 |
131
+ | 0.9404 | 1.99 | 400 | 1.1520 |
132
+ | 0.7161 | 2.22 | 450 | 1.2603 |
133
+ | 0.5897 | 2.47 | 500 | 1.2661 |
134
+ | 0.5271 | 2.72 | 550 | 1.2814 |
135
+ | 0.6239 | 2.97 | 600 | 1.2705 |
136
+ | 0.3486 | 3.21 | 650 | 1.3848 |
137
+ | 0.5591 | 3.46 | 700 | 1.4171 |
138
+ | 0.3804 | 3.71 | 750 | 1.4177 |
139
+
140
+
141
+ ### Framework versions
142
+
143
+ - PEFT 0.9.0
144
+ - Transformers 4.38.2
145
+ - Pytorch 2.1.2+cu118
146
+ - Datasets 2.18.0
147
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openlm-research/open_llama_3b_v2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 32,
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": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "v_proj",
24
+ "o_proj",
25
+ "q_proj",
26
+ "down_proj",
27
+ "up_proj",
28
+ "gate_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:ccf1d2eb3fe8099e23e6c42a359689426d253e71d33c399490b017b152b6d6e7
3
+ size 50982842
checkpoint-398/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: openlm-research/open_llama_3b_v2
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.9.0
checkpoint-398/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openlm-research/open_llama_3b_v2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 32,
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": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "v_proj",
24
+ "o_proj",
25
+ "q_proj",
26
+ "down_proj",
27
+ "up_proj",
28
+ "gate_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
checkpoint-398/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b6d7016774d54258741e87e7a11113ba7bfdad918c9560a3585544926ce0ee7
3
+ size 50899792
checkpoint-398/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58eb8a13f40a6bfa712cb3c9c7e1633673cb857b530d9a99f7237b6c46c703f9
3
+ size 101919290
checkpoint-398/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b89fc1d394f02fc6ad0be146152ad044e5c16ee7b8e840eed13f3689d19910c5
3
+ size 14244
checkpoint-398/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07f80e9e26e4de9744278d23fb86ee2e353bf61ec1fe7a592c6b05e590e547be
3
+ size 1064
checkpoint-398/trainer_state.json ADDED
@@ -0,0 +1,2871 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.979899497487437,
5
+ "eval_steps": 50,
6
+ "global_step": 398,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "grad_norm": 0.6762334704399109,
14
+ "learning_rate": 1e-05,
15
+ "loss": 1.3026,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.01,
20
+ "eval_loss": 1.343465805053711,
21
+ "eval_runtime": 2.9584,
22
+ "eval_samples_per_second": 33.802,
23
+ "eval_steps_per_second": 16.901,
24
+ "step": 1
25
+ },
26
+ {
27
+ "epoch": 0.01,
28
+ "grad_norm": 0.7722721695899963,
29
+ "learning_rate": 2e-05,
30
+ "loss": 1.5419,
31
+ "step": 2
32
+ },
33
+ {
34
+ "epoch": 0.02,
35
+ "grad_norm": 0.6532348394393921,
36
+ "learning_rate": 3e-05,
37
+ "loss": 1.4429,
38
+ "step": 3
39
+ },
40
+ {
41
+ "epoch": 0.02,
42
+ "grad_norm": 0.8427589535713196,
43
+ "learning_rate": 4e-05,
44
+ "loss": 1.4,
45
+ "step": 4
46
+ },
47
+ {
48
+ "epoch": 0.03,
49
+ "grad_norm": 0.9355791807174683,
50
+ "learning_rate": 5e-05,
51
+ "loss": 1.2583,
52
+ "step": 5
53
+ },
54
+ {
55
+ "epoch": 0.03,
56
+ "grad_norm": 0.6357808113098145,
57
+ "learning_rate": 6e-05,
58
+ "loss": 1.2655,
59
+ "step": 6
60
+ },
61
+ {
62
+ "epoch": 0.04,
63
+ "grad_norm": 0.963829517364502,
64
+ "learning_rate": 7e-05,
65
+ "loss": 1.42,
66
+ "step": 7
67
+ },
68
+ {
69
+ "epoch": 0.04,
70
+ "grad_norm": 0.6698102951049805,
71
+ "learning_rate": 8e-05,
72
+ "loss": 1.3938,
73
+ "step": 8
74
+ },
75
+ {
76
+ "epoch": 0.05,
77
+ "grad_norm": 0.5394894480705261,
78
+ "learning_rate": 9e-05,
79
+ "loss": 1.2234,
80
+ "step": 9
81
+ },
82
+ {
83
+ "epoch": 0.05,
84
+ "grad_norm": 0.8773290514945984,
85
+ "learning_rate": 0.0001,
86
+ "loss": 1.4257,
87
+ "step": 10
88
+ },
89
+ {
90
+ "epoch": 0.06,
91
+ "grad_norm": 0.7960235476493835,
92
+ "learning_rate": 0.00011000000000000002,
93
+ "loss": 1.4272,
94
+ "step": 11
95
+ },
96
+ {
97
+ "epoch": 0.06,
98
+ "grad_norm": 0.7909610867500305,
99
+ "learning_rate": 0.00012,
100
+ "loss": 1.352,
101
+ "step": 12
102
+ },
103
+ {
104
+ "epoch": 0.07,
105
+ "grad_norm": 0.8417578339576721,
106
+ "learning_rate": 0.00013000000000000002,
107
+ "loss": 1.2048,
108
+ "step": 13
109
+ },
110
+ {
111
+ "epoch": 0.07,
112
+ "grad_norm": 0.8076886534690857,
113
+ "learning_rate": 0.00014,
114
+ "loss": 1.4186,
115
+ "step": 14
116
+ },
117
+ {
118
+ "epoch": 0.08,
119
+ "grad_norm": 0.7543106079101562,
120
+ "learning_rate": 0.00015000000000000001,
121
+ "loss": 1.0873,
122
+ "step": 15
123
+ },
124
+ {
125
+ "epoch": 0.08,
126
+ "grad_norm": 0.9430835247039795,
127
+ "learning_rate": 0.00016,
128
+ "loss": 1.4061,
129
+ "step": 16
130
+ },
131
+ {
132
+ "epoch": 0.09,
133
+ "grad_norm": 0.7473496794700623,
134
+ "learning_rate": 0.00017,
135
+ "loss": 1.1407,
136
+ "step": 17
137
+ },
138
+ {
139
+ "epoch": 0.09,
140
+ "grad_norm": 0.8123806715011597,
141
+ "learning_rate": 0.00018,
142
+ "loss": 1.4394,
143
+ "step": 18
144
+ },
145
+ {
146
+ "epoch": 0.1,
147
+ "grad_norm": 0.7778059244155884,
148
+ "learning_rate": 0.00019,
149
+ "loss": 1.2752,
150
+ "step": 19
151
+ },
152
+ {
153
+ "epoch": 0.1,
154
+ "grad_norm": 0.7027471661567688,
155
+ "learning_rate": 0.0002,
156
+ "loss": 1.3107,
157
+ "step": 20
158
+ },
159
+ {
160
+ "epoch": 0.11,
161
+ "grad_norm": 0.8443830609321594,
162
+ "learning_rate": 0.00019999918050612108,
163
+ "loss": 1.2204,
164
+ "step": 21
165
+ },
166
+ {
167
+ "epoch": 0.11,
168
+ "grad_norm": 0.6853266358375549,
169
+ "learning_rate": 0.00019999672203791565,
170
+ "loss": 1.2231,
171
+ "step": 22
172
+ },
173
+ {
174
+ "epoch": 0.12,
175
+ "grad_norm": 0.8757483959197998,
176
+ "learning_rate": 0.00019999262463567773,
177
+ "loss": 1.2069,
178
+ "step": 23
179
+ },
180
+ {
181
+ "epoch": 0.12,
182
+ "grad_norm": 0.7184014320373535,
183
+ "learning_rate": 0.00019998688836656323,
184
+ "loss": 1.2124,
185
+ "step": 24
186
+ },
187
+ {
188
+ "epoch": 0.13,
189
+ "grad_norm": 0.6530072093009949,
190
+ "learning_rate": 0.0001999795133245889,
191
+ "loss": 1.1672,
192
+ "step": 25
193
+ },
194
+ {
195
+ "epoch": 0.13,
196
+ "grad_norm": 0.7211533188819885,
197
+ "learning_rate": 0.0001999704996306308,
198
+ "loss": 1.3207,
199
+ "step": 26
200
+ },
201
+ {
202
+ "epoch": 0.14,
203
+ "grad_norm": 0.7048207521438599,
204
+ "learning_rate": 0.00019995984743242226,
205
+ "loss": 1.2003,
206
+ "step": 27
207
+ },
208
+ {
209
+ "epoch": 0.14,
210
+ "grad_norm": 0.6881248354911804,
211
+ "learning_rate": 0.00019994755690455152,
212
+ "loss": 1.117,
213
+ "step": 28
214
+ },
215
+ {
216
+ "epoch": 0.15,
217
+ "grad_norm": 0.7877801656723022,
218
+ "learning_rate": 0.00019993362824845875,
219
+ "loss": 1.0531,
220
+ "step": 29
221
+ },
222
+ {
223
+ "epoch": 0.15,
224
+ "grad_norm": 0.749905526638031,
225
+ "learning_rate": 0.000199918061692433,
226
+ "loss": 1.1462,
227
+ "step": 30
228
+ },
229
+ {
230
+ "epoch": 0.16,
231
+ "grad_norm": 0.67184978723526,
232
+ "learning_rate": 0.00019990085749160822,
233
+ "loss": 1.0939,
234
+ "step": 31
235
+ },
236
+ {
237
+ "epoch": 0.16,
238
+ "grad_norm": 0.6622844934463501,
239
+ "learning_rate": 0.0001998820159279591,
240
+ "loss": 1.1369,
241
+ "step": 32
242
+ },
243
+ {
244
+ "epoch": 0.17,
245
+ "grad_norm": 0.763306736946106,
246
+ "learning_rate": 0.00019986153731029656,
247
+ "loss": 1.3525,
248
+ "step": 33
249
+ },
250
+ {
251
+ "epoch": 0.17,
252
+ "grad_norm": 0.6171010136604309,
253
+ "learning_rate": 0.0001998394219742627,
254
+ "loss": 0.8807,
255
+ "step": 34
256
+ },
257
+ {
258
+ "epoch": 0.18,
259
+ "grad_norm": 0.7575845718383789,
260
+ "learning_rate": 0.00019981567028232514,
261
+ "loss": 1.206,
262
+ "step": 35
263
+ },
264
+ {
265
+ "epoch": 0.18,
266
+ "grad_norm": 0.5694592595100403,
267
+ "learning_rate": 0.00019979028262377118,
268
+ "loss": 0.9079,
269
+ "step": 36
270
+ },
271
+ {
272
+ "epoch": 0.19,
273
+ "grad_norm": 0.7056426405906677,
274
+ "learning_rate": 0.00019976325941470146,
275
+ "loss": 1.1133,
276
+ "step": 37
277
+ },
278
+ {
279
+ "epoch": 0.19,
280
+ "grad_norm": 0.6812122464179993,
281
+ "learning_rate": 0.00019973460109802305,
282
+ "loss": 1.2707,
283
+ "step": 38
284
+ },
285
+ {
286
+ "epoch": 0.2,
287
+ "grad_norm": 0.5790569186210632,
288
+ "learning_rate": 0.0001997043081434423,
289
+ "loss": 1.0047,
290
+ "step": 39
291
+ },
292
+ {
293
+ "epoch": 0.2,
294
+ "grad_norm": 0.6529936790466309,
295
+ "learning_rate": 0.00019967238104745696,
296
+ "loss": 1.0917,
297
+ "step": 40
298
+ },
299
+ {
300
+ "epoch": 0.21,
301
+ "grad_norm": 0.6274911165237427,
302
+ "learning_rate": 0.00019963882033334826,
303
+ "loss": 1.2586,
304
+ "step": 41
305
+ },
306
+ {
307
+ "epoch": 0.21,
308
+ "grad_norm": 0.6666668653488159,
309
+ "learning_rate": 0.00019960362655117218,
310
+ "loss": 1.1187,
311
+ "step": 42
312
+ },
313
+ {
314
+ "epoch": 0.22,
315
+ "grad_norm": 0.6239954233169556,
316
+ "learning_rate": 0.00019956680027775051,
317
+ "loss": 1.0343,
318
+ "step": 43
319
+ },
320
+ {
321
+ "epoch": 0.22,
322
+ "grad_norm": 0.6892250180244446,
323
+ "learning_rate": 0.0001995283421166614,
324
+ "loss": 1.0254,
325
+ "step": 44
326
+ },
327
+ {
328
+ "epoch": 0.23,
329
+ "grad_norm": 0.7392664551734924,
330
+ "learning_rate": 0.00019948825269822934,
331
+ "loss": 1.0592,
332
+ "step": 45
333
+ },
334
+ {
335
+ "epoch": 0.23,
336
+ "grad_norm": 0.7541553378105164,
337
+ "learning_rate": 0.00019944653267951504,
338
+ "loss": 1.2297,
339
+ "step": 46
340
+ },
341
+ {
342
+ "epoch": 0.24,
343
+ "grad_norm": 0.685874342918396,
344
+ "learning_rate": 0.00019940318274430449,
345
+ "loss": 1.321,
346
+ "step": 47
347
+ },
348
+ {
349
+ "epoch": 0.24,
350
+ "grad_norm": 0.7901135087013245,
351
+ "learning_rate": 0.00019935820360309777,
352
+ "loss": 1.2583,
353
+ "step": 48
354
+ },
355
+ {
356
+ "epoch": 0.25,
357
+ "grad_norm": 0.6619594693183899,
358
+ "learning_rate": 0.00019931159599309757,
359
+ "loss": 0.9762,
360
+ "step": 49
361
+ },
362
+ {
363
+ "epoch": 0.25,
364
+ "grad_norm": 0.6059371829032898,
365
+ "learning_rate": 0.00019926336067819684,
366
+ "loss": 1.1146,
367
+ "step": 50
368
+ },
369
+ {
370
+ "epoch": 0.25,
371
+ "eval_loss": 1.1476221084594727,
372
+ "eval_runtime": 2.9589,
373
+ "eval_samples_per_second": 33.796,
374
+ "eval_steps_per_second": 16.898,
375
+ "step": 50
376
+ },
377
+ {
378
+ "epoch": 0.26,
379
+ "grad_norm": 0.6533025503158569,
380
+ "learning_rate": 0.00019921349844896654,
381
+ "loss": 1.2439,
382
+ "step": 51
383
+ },
384
+ {
385
+ "epoch": 0.26,
386
+ "grad_norm": 0.5473713278770447,
387
+ "learning_rate": 0.00019916201012264254,
388
+ "loss": 0.8464,
389
+ "step": 52
390
+ },
391
+ {
392
+ "epoch": 0.27,
393
+ "grad_norm": 0.6035101413726807,
394
+ "learning_rate": 0.00019910889654311208,
395
+ "loss": 1.1297,
396
+ "step": 53
397
+ },
398
+ {
399
+ "epoch": 0.27,
400
+ "grad_norm": 0.7092946767807007,
401
+ "learning_rate": 0.00019905415858090036,
402
+ "loss": 1.0365,
403
+ "step": 54
404
+ },
405
+ {
406
+ "epoch": 0.28,
407
+ "grad_norm": 0.602556049823761,
408
+ "learning_rate": 0.00019899779713315575,
409
+ "loss": 1.1238,
410
+ "step": 55
411
+ },
412
+ {
413
+ "epoch": 0.28,
414
+ "grad_norm": 0.6566863059997559,
415
+ "learning_rate": 0.00019893981312363562,
416
+ "loss": 1.1097,
417
+ "step": 56
418
+ },
419
+ {
420
+ "epoch": 0.29,
421
+ "grad_norm": 0.6582695245742798,
422
+ "learning_rate": 0.00019888020750269067,
423
+ "loss": 1.3681,
424
+ "step": 57
425
+ },
426
+ {
427
+ "epoch": 0.29,
428
+ "grad_norm": 0.509901225566864,
429
+ "learning_rate": 0.00019881898124724981,
430
+ "loss": 0.7163,
431
+ "step": 58
432
+ },
433
+ {
434
+ "epoch": 0.3,
435
+ "grad_norm": 0.6406445503234863,
436
+ "learning_rate": 0.0001987561353608038,
437
+ "loss": 1.1309,
438
+ "step": 59
439
+ },
440
+ {
441
+ "epoch": 0.3,
442
+ "grad_norm": 0.5770175457000732,
443
+ "learning_rate": 0.00019869167087338907,
444
+ "loss": 1.1706,
445
+ "step": 60
446
+ },
447
+ {
448
+ "epoch": 0.31,
449
+ "grad_norm": 0.6582055687904358,
450
+ "learning_rate": 0.00019862558884157068,
451
+ "loss": 1.1121,
452
+ "step": 61
453
+ },
454
+ {
455
+ "epoch": 0.31,
456
+ "grad_norm": 0.7646100521087646,
457
+ "learning_rate": 0.00019855789034842504,
458
+ "loss": 1.1313,
459
+ "step": 62
460
+ },
461
+ {
462
+ "epoch": 0.32,
463
+ "grad_norm": 0.7127470970153809,
464
+ "learning_rate": 0.00019848857650352214,
465
+ "loss": 1.258,
466
+ "step": 63
467
+ },
468
+ {
469
+ "epoch": 0.32,
470
+ "grad_norm": 0.5721624493598938,
471
+ "learning_rate": 0.00019841764844290744,
472
+ "loss": 1.0163,
473
+ "step": 64
474
+ },
475
+ {
476
+ "epoch": 0.33,
477
+ "grad_norm": 0.6494898796081543,
478
+ "learning_rate": 0.00019834510732908315,
479
+ "loss": 1.1974,
480
+ "step": 65
481
+ },
482
+ {
483
+ "epoch": 0.33,
484
+ "grad_norm": 0.6703062057495117,
485
+ "learning_rate": 0.00019827095435098925,
486
+ "loss": 1.1376,
487
+ "step": 66
488
+ },
489
+ {
490
+ "epoch": 0.34,
491
+ "grad_norm": 0.696711003780365,
492
+ "learning_rate": 0.000198195190723984,
493
+ "loss": 0.9931,
494
+ "step": 67
495
+ },
496
+ {
497
+ "epoch": 0.34,
498
+ "grad_norm": 0.6563432216644287,
499
+ "learning_rate": 0.0001981178176898239,
500
+ "loss": 1.2047,
501
+ "step": 68
502
+ },
503
+ {
504
+ "epoch": 0.35,
505
+ "grad_norm": 0.7269361019134521,
506
+ "learning_rate": 0.0001980388365166436,
507
+ "loss": 1.6113,
508
+ "step": 69
509
+ },
510
+ {
511
+ "epoch": 0.35,
512
+ "grad_norm": 0.6356198191642761,
513
+ "learning_rate": 0.0001979582484989348,
514
+ "loss": 1.3778,
515
+ "step": 70
516
+ },
517
+ {
518
+ "epoch": 0.36,
519
+ "grad_norm": 0.6009278893470764,
520
+ "learning_rate": 0.00019787605495752528,
521
+ "loss": 1.2131,
522
+ "step": 71
523
+ },
524
+ {
525
+ "epoch": 0.36,
526
+ "grad_norm": 0.49109163880348206,
527
+ "learning_rate": 0.00019779225723955707,
528
+ "loss": 0.8246,
529
+ "step": 72
530
+ },
531
+ {
532
+ "epoch": 0.37,
533
+ "grad_norm": 0.5709823966026306,
534
+ "learning_rate": 0.00019770685671846456,
535
+ "loss": 1.0578,
536
+ "step": 73
537
+ },
538
+ {
539
+ "epoch": 0.37,
540
+ "grad_norm": 0.5613502860069275,
541
+ "learning_rate": 0.0001976198547939518,
542
+ "loss": 0.8883,
543
+ "step": 74
544
+ },
545
+ {
546
+ "epoch": 0.38,
547
+ "grad_norm": 0.750335156917572,
548
+ "learning_rate": 0.0001975312528919697,
549
+ "loss": 1.1836,
550
+ "step": 75
551
+ },
552
+ {
553
+ "epoch": 0.38,
554
+ "grad_norm": 0.6157568693161011,
555
+ "learning_rate": 0.00019744105246469263,
556
+ "loss": 1.0637,
557
+ "step": 76
558
+ },
559
+ {
560
+ "epoch": 0.39,
561
+ "grad_norm": 0.6417941451072693,
562
+ "learning_rate": 0.00019734925499049447,
563
+ "loss": 1.2824,
564
+ "step": 77
565
+ },
566
+ {
567
+ "epoch": 0.39,
568
+ "grad_norm": 0.8214441537857056,
569
+ "learning_rate": 0.0001972558619739246,
570
+ "loss": 1.1942,
571
+ "step": 78
572
+ },
573
+ {
574
+ "epoch": 0.4,
575
+ "grad_norm": 0.6943228244781494,
576
+ "learning_rate": 0.00019716087494568317,
577
+ "loss": 1.3261,
578
+ "step": 79
579
+ },
580
+ {
581
+ "epoch": 0.4,
582
+ "grad_norm": 0.739622950553894,
583
+ "learning_rate": 0.00019706429546259593,
584
+ "loss": 1.2639,
585
+ "step": 80
586
+ },
587
+ {
588
+ "epoch": 0.41,
589
+ "grad_norm": 0.6374944448471069,
590
+ "learning_rate": 0.00019696612510758876,
591
+ "loss": 0.9929,
592
+ "step": 81
593
+ },
594
+ {
595
+ "epoch": 0.41,
596
+ "grad_norm": 0.7595279812812805,
597
+ "learning_rate": 0.00019686636548966178,
598
+ "loss": 1.2859,
599
+ "step": 82
600
+ },
601
+ {
602
+ "epoch": 0.42,
603
+ "grad_norm": 0.6465960144996643,
604
+ "learning_rate": 0.00019676501824386294,
605
+ "loss": 1.0333,
606
+ "step": 83
607
+ },
608
+ {
609
+ "epoch": 0.42,
610
+ "grad_norm": 0.7063401341438293,
611
+ "learning_rate": 0.00019666208503126112,
612
+ "loss": 1.2189,
613
+ "step": 84
614
+ },
615
+ {
616
+ "epoch": 0.43,
617
+ "grad_norm": 0.631826639175415,
618
+ "learning_rate": 0.00019655756753891916,
619
+ "loss": 1.2583,
620
+ "step": 85
621
+ },
622
+ {
623
+ "epoch": 0.43,
624
+ "grad_norm": 0.6506052017211914,
625
+ "learning_rate": 0.0001964514674798659,
626
+ "loss": 1.2019,
627
+ "step": 86
628
+ },
629
+ {
630
+ "epoch": 0.44,
631
+ "grad_norm": 0.7421661615371704,
632
+ "learning_rate": 0.00019634378659306832,
633
+ "loss": 1.2122,
634
+ "step": 87
635
+ },
636
+ {
637
+ "epoch": 0.44,
638
+ "grad_norm": 0.5749310851097107,
639
+ "learning_rate": 0.00019623452664340306,
640
+ "loss": 1.0522,
641
+ "step": 88
642
+ },
643
+ {
644
+ "epoch": 0.45,
645
+ "grad_norm": 0.6523499488830566,
646
+ "learning_rate": 0.0001961236894216272,
647
+ "loss": 1.2135,
648
+ "step": 89
649
+ },
650
+ {
651
+ "epoch": 0.45,
652
+ "grad_norm": 0.5970554947853088,
653
+ "learning_rate": 0.00019601127674434928,
654
+ "loss": 1.0297,
655
+ "step": 90
656
+ },
657
+ {
658
+ "epoch": 0.46,
659
+ "grad_norm": 0.587348461151123,
660
+ "learning_rate": 0.00019589729045399934,
661
+ "loss": 1.0214,
662
+ "step": 91
663
+ },
664
+ {
665
+ "epoch": 0.46,
666
+ "grad_norm": 0.6518609523773193,
667
+ "learning_rate": 0.00019578173241879872,
668
+ "loss": 0.9928,
669
+ "step": 92
670
+ },
671
+ {
672
+ "epoch": 0.47,
673
+ "grad_norm": 0.7513082027435303,
674
+ "learning_rate": 0.00019566460453272945,
675
+ "loss": 1.1204,
676
+ "step": 93
677
+ },
678
+ {
679
+ "epoch": 0.47,
680
+ "grad_norm": 0.8648024201393127,
681
+ "learning_rate": 0.0001955459087155033,
682
+ "loss": 1.3671,
683
+ "step": 94
684
+ },
685
+ {
686
+ "epoch": 0.48,
687
+ "grad_norm": 0.6207080483436584,
688
+ "learning_rate": 0.0001954256469125301,
689
+ "loss": 1.1286,
690
+ "step": 95
691
+ },
692
+ {
693
+ "epoch": 0.48,
694
+ "grad_norm": 0.6174007058143616,
695
+ "learning_rate": 0.0001953038210948861,
696
+ "loss": 1.145,
697
+ "step": 96
698
+ },
699
+ {
700
+ "epoch": 0.49,
701
+ "grad_norm": 0.6160337328910828,
702
+ "learning_rate": 0.00019518043325928157,
703
+ "loss": 1.2688,
704
+ "step": 97
705
+ },
706
+ {
707
+ "epoch": 0.49,
708
+ "grad_norm": 0.662702202796936,
709
+ "learning_rate": 0.00019505548542802804,
710
+ "loss": 1.1212,
711
+ "step": 98
712
+ },
713
+ {
714
+ "epoch": 0.5,
715
+ "grad_norm": 0.7133952379226685,
716
+ "learning_rate": 0.00019492897964900512,
717
+ "loss": 1.0514,
718
+ "step": 99
719
+ },
720
+ {
721
+ "epoch": 0.5,
722
+ "grad_norm": 0.7767614126205444,
723
+ "learning_rate": 0.00019480091799562704,
724
+ "loss": 1.2387,
725
+ "step": 100
726
+ },
727
+ {
728
+ "epoch": 0.5,
729
+ "eval_loss": 1.1319388151168823,
730
+ "eval_runtime": 2.9089,
731
+ "eval_samples_per_second": 34.377,
732
+ "eval_steps_per_second": 17.189,
733
+ "step": 100
734
+ },
735
+ {
736
+ "epoch": 0.51,
737
+ "grad_norm": 0.6398429870605469,
738
+ "learning_rate": 0.00019467130256680868,
739
+ "loss": 1.0076,
740
+ "step": 101
741
+ },
742
+ {
743
+ "epoch": 0.51,
744
+ "grad_norm": 0.6510715484619141,
745
+ "learning_rate": 0.00019454013548693102,
746
+ "loss": 1.2372,
747
+ "step": 102
748
+ },
749
+ {
750
+ "epoch": 0.52,
751
+ "grad_norm": 0.7204650044441223,
752
+ "learning_rate": 0.00019440741890580643,
753
+ "loss": 1.0999,
754
+ "step": 103
755
+ },
756
+ {
757
+ "epoch": 0.52,
758
+ "grad_norm": 0.6531095504760742,
759
+ "learning_rate": 0.00019427315499864344,
760
+ "loss": 1.1123,
761
+ "step": 104
762
+ },
763
+ {
764
+ "epoch": 0.53,
765
+ "grad_norm": 0.5871708989143372,
766
+ "learning_rate": 0.00019413734596601104,
767
+ "loss": 1.2162,
768
+ "step": 105
769
+ },
770
+ {
771
+ "epoch": 0.53,
772
+ "grad_norm": 0.6323477625846863,
773
+ "learning_rate": 0.00019399999403380266,
774
+ "loss": 1.1369,
775
+ "step": 106
776
+ },
777
+ {
778
+ "epoch": 0.54,
779
+ "grad_norm": 0.6977123618125916,
780
+ "learning_rate": 0.00019386110145319963,
781
+ "loss": 1.0952,
782
+ "step": 107
783
+ },
784
+ {
785
+ "epoch": 0.54,
786
+ "grad_norm": 0.6638639569282532,
787
+ "learning_rate": 0.00019372067050063438,
788
+ "loss": 1.1125,
789
+ "step": 108
790
+ },
791
+ {
792
+ "epoch": 0.55,
793
+ "grad_norm": 0.6010698676109314,
794
+ "learning_rate": 0.000193578703477753,
795
+ "loss": 1.1715,
796
+ "step": 109
797
+ },
798
+ {
799
+ "epoch": 0.55,
800
+ "grad_norm": 0.5837023258209229,
801
+ "learning_rate": 0.00019343520271137763,
802
+ "loss": 0.8489,
803
+ "step": 110
804
+ },
805
+ {
806
+ "epoch": 0.56,
807
+ "grad_norm": 0.6870157718658447,
808
+ "learning_rate": 0.0001932901705534683,
809
+ "loss": 1.0953,
810
+ "step": 111
811
+ },
812
+ {
813
+ "epoch": 0.56,
814
+ "grad_norm": 0.5713046789169312,
815
+ "learning_rate": 0.00019314360938108425,
816
+ "loss": 1.1113,
817
+ "step": 112
818
+ },
819
+ {
820
+ "epoch": 0.57,
821
+ "grad_norm": 0.5966447591781616,
822
+ "learning_rate": 0.00019299552159634517,
823
+ "loss": 1.2646,
824
+ "step": 113
825
+ },
826
+ {
827
+ "epoch": 0.57,
828
+ "grad_norm": 0.6116918921470642,
829
+ "learning_rate": 0.00019284590962639176,
830
+ "loss": 1.0807,
831
+ "step": 114
832
+ },
833
+ {
834
+ "epoch": 0.58,
835
+ "grad_norm": 0.5885886549949646,
836
+ "learning_rate": 0.0001926947759233459,
837
+ "loss": 0.9551,
838
+ "step": 115
839
+ },
840
+ {
841
+ "epoch": 0.58,
842
+ "grad_norm": 0.5844876766204834,
843
+ "learning_rate": 0.00019254212296427044,
844
+ "loss": 1.0009,
845
+ "step": 116
846
+ },
847
+ {
848
+ "epoch": 0.59,
849
+ "grad_norm": 0.5967299342155457,
850
+ "learning_rate": 0.0001923879532511287,
851
+ "loss": 0.863,
852
+ "step": 117
853
+ },
854
+ {
855
+ "epoch": 0.59,
856
+ "grad_norm": 0.543732762336731,
857
+ "learning_rate": 0.0001922322693107434,
858
+ "loss": 0.8331,
859
+ "step": 118
860
+ },
861
+ {
862
+ "epoch": 0.6,
863
+ "grad_norm": 0.6925728917121887,
864
+ "learning_rate": 0.0001920750736947553,
865
+ "loss": 1.1044,
866
+ "step": 119
867
+ },
868
+ {
869
+ "epoch": 0.6,
870
+ "grad_norm": 0.5720507502555847,
871
+ "learning_rate": 0.00019191636897958122,
872
+ "loss": 1.2173,
873
+ "step": 120
874
+ },
875
+ {
876
+ "epoch": 0.61,
877
+ "grad_norm": 0.6664772033691406,
878
+ "learning_rate": 0.0001917561577663721,
879
+ "loss": 0.9849,
880
+ "step": 121
881
+ },
882
+ {
883
+ "epoch": 0.61,
884
+ "grad_norm": 0.6026978492736816,
885
+ "learning_rate": 0.00019159444268097012,
886
+ "loss": 1.2952,
887
+ "step": 122
888
+ },
889
+ {
890
+ "epoch": 0.62,
891
+ "grad_norm": 0.6648169755935669,
892
+ "learning_rate": 0.00019143122637386566,
893
+ "loss": 0.8417,
894
+ "step": 123
895
+ },
896
+ {
897
+ "epoch": 0.62,
898
+ "grad_norm": 0.7643215656280518,
899
+ "learning_rate": 0.00019126651152015403,
900
+ "loss": 1.1142,
901
+ "step": 124
902
+ },
903
+ {
904
+ "epoch": 0.63,
905
+ "grad_norm": 0.6389123797416687,
906
+ "learning_rate": 0.00019110030081949156,
907
+ "loss": 1.2387,
908
+ "step": 125
909
+ },
910
+ {
911
+ "epoch": 0.63,
912
+ "grad_norm": 0.7826026678085327,
913
+ "learning_rate": 0.00019093259699605125,
914
+ "loss": 1.1407,
915
+ "step": 126
916
+ },
917
+ {
918
+ "epoch": 0.64,
919
+ "grad_norm": 0.6801394820213318,
920
+ "learning_rate": 0.0001907634027984782,
921
+ "loss": 0.932,
922
+ "step": 127
923
+ },
924
+ {
925
+ "epoch": 0.64,
926
+ "grad_norm": 0.6450052857398987,
927
+ "learning_rate": 0.0001905927209998447,
928
+ "loss": 1.3197,
929
+ "step": 128
930
+ },
931
+ {
932
+ "epoch": 0.65,
933
+ "grad_norm": 0.6216878890991211,
934
+ "learning_rate": 0.00019042055439760444,
935
+ "loss": 1.2593,
936
+ "step": 129
937
+ },
938
+ {
939
+ "epoch": 0.65,
940
+ "grad_norm": 0.6000977158546448,
941
+ "learning_rate": 0.000190246905813547,
942
+ "loss": 0.9974,
943
+ "step": 130
944
+ },
945
+ {
946
+ "epoch": 0.66,
947
+ "grad_norm": 0.5806196928024292,
948
+ "learning_rate": 0.0001900717780937514,
949
+ "loss": 1.1792,
950
+ "step": 131
951
+ },
952
+ {
953
+ "epoch": 0.66,
954
+ "grad_norm": 0.6986164450645447,
955
+ "learning_rate": 0.00018989517410853955,
956
+ "loss": 1.252,
957
+ "step": 132
958
+ },
959
+ {
960
+ "epoch": 0.67,
961
+ "grad_norm": 0.6852320432662964,
962
+ "learning_rate": 0.0001897170967524291,
963
+ "loss": 1.098,
964
+ "step": 133
965
+ },
966
+ {
967
+ "epoch": 0.67,
968
+ "grad_norm": 0.6186272501945496,
969
+ "learning_rate": 0.00018953754894408616,
970
+ "loss": 1.1099,
971
+ "step": 134
972
+ },
973
+ {
974
+ "epoch": 0.68,
975
+ "grad_norm": 0.7196840643882751,
976
+ "learning_rate": 0.0001893565336262773,
977
+ "loss": 1.1809,
978
+ "step": 135
979
+ },
980
+ {
981
+ "epoch": 0.68,
982
+ "grad_norm": 0.6523413062095642,
983
+ "learning_rate": 0.00018917405376582145,
984
+ "loss": 1.2383,
985
+ "step": 136
986
+ },
987
+ {
988
+ "epoch": 0.69,
989
+ "grad_norm": 0.7788291573524475,
990
+ "learning_rate": 0.00018899011235354115,
991
+ "loss": 1.023,
992
+ "step": 137
993
+ },
994
+ {
995
+ "epoch": 0.69,
996
+ "grad_norm": 0.5616946220397949,
997
+ "learning_rate": 0.00018880471240421365,
998
+ "loss": 0.8242,
999
+ "step": 138
1000
+ },
1001
+ {
1002
+ "epoch": 0.7,
1003
+ "grad_norm": 0.6670994758605957,
1004
+ "learning_rate": 0.00018861785695652142,
1005
+ "loss": 1.2797,
1006
+ "step": 139
1007
+ },
1008
+ {
1009
+ "epoch": 0.7,
1010
+ "grad_norm": 0.6285648345947266,
1011
+ "learning_rate": 0.00018842954907300236,
1012
+ "loss": 1.0959,
1013
+ "step": 140
1014
+ },
1015
+ {
1016
+ "epoch": 0.71,
1017
+ "grad_norm": 0.6495100855827332,
1018
+ "learning_rate": 0.00018823979183999964,
1019
+ "loss": 1.1426,
1020
+ "step": 141
1021
+ },
1022
+ {
1023
+ "epoch": 0.71,
1024
+ "grad_norm": 0.7513198256492615,
1025
+ "learning_rate": 0.00018804858836761107,
1026
+ "loss": 1.2578,
1027
+ "step": 142
1028
+ },
1029
+ {
1030
+ "epoch": 0.72,
1031
+ "grad_norm": 0.5422288775444031,
1032
+ "learning_rate": 0.0001878559417896382,
1033
+ "loss": 0.9833,
1034
+ "step": 143
1035
+ },
1036
+ {
1037
+ "epoch": 0.72,
1038
+ "grad_norm": 0.605277419090271,
1039
+ "learning_rate": 0.0001876618552635348,
1040
+ "loss": 1.2323,
1041
+ "step": 144
1042
+ },
1043
+ {
1044
+ "epoch": 0.73,
1045
+ "grad_norm": 0.7177323698997498,
1046
+ "learning_rate": 0.00018746633197035527,
1047
+ "loss": 1.2153,
1048
+ "step": 145
1049
+ },
1050
+ {
1051
+ "epoch": 0.73,
1052
+ "grad_norm": 0.5417729020118713,
1053
+ "learning_rate": 0.00018726937511470246,
1054
+ "loss": 0.9367,
1055
+ "step": 146
1056
+ },
1057
+ {
1058
+ "epoch": 0.74,
1059
+ "grad_norm": 0.6895157098770142,
1060
+ "learning_rate": 0.00018707098792467515,
1061
+ "loss": 1.3363,
1062
+ "step": 147
1063
+ },
1064
+ {
1065
+ "epoch": 0.74,
1066
+ "grad_norm": 0.5565975308418274,
1067
+ "learning_rate": 0.00018687117365181512,
1068
+ "loss": 1.0385,
1069
+ "step": 148
1070
+ },
1071
+ {
1072
+ "epoch": 0.75,
1073
+ "grad_norm": 0.7168130278587341,
1074
+ "learning_rate": 0.00018666993557105377,
1075
+ "loss": 1.2281,
1076
+ "step": 149
1077
+ },
1078
+ {
1079
+ "epoch": 0.75,
1080
+ "grad_norm": 0.839598536491394,
1081
+ "learning_rate": 0.00018646727698065865,
1082
+ "loss": 1.4159,
1083
+ "step": 150
1084
+ },
1085
+ {
1086
+ "epoch": 0.75,
1087
+ "eval_loss": 1.119249939918518,
1088
+ "eval_runtime": 2.9417,
1089
+ "eval_samples_per_second": 33.994,
1090
+ "eval_steps_per_second": 16.997,
1091
+ "step": 150
1092
+ },
1093
+ {
1094
+ "epoch": 0.76,
1095
+ "grad_norm": 0.5981218814849854,
1096
+ "learning_rate": 0.00018626320120217923,
1097
+ "loss": 1.0671,
1098
+ "step": 151
1099
+ },
1100
+ {
1101
+ "epoch": 0.76,
1102
+ "grad_norm": 0.6944805383682251,
1103
+ "learning_rate": 0.00018605771158039253,
1104
+ "loss": 1.3229,
1105
+ "step": 152
1106
+ },
1107
+ {
1108
+ "epoch": 0.77,
1109
+ "grad_norm": 0.6238952875137329,
1110
+ "learning_rate": 0.00018585081148324832,
1111
+ "loss": 1.1578,
1112
+ "step": 153
1113
+ },
1114
+ {
1115
+ "epoch": 0.77,
1116
+ "grad_norm": 0.6363958120346069,
1117
+ "learning_rate": 0.00018564250430181387,
1118
+ "loss": 1.3265,
1119
+ "step": 154
1120
+ },
1121
+ {
1122
+ "epoch": 0.78,
1123
+ "grad_norm": 0.5761409401893616,
1124
+ "learning_rate": 0.00018543279345021834,
1125
+ "loss": 1.1844,
1126
+ "step": 155
1127
+ },
1128
+ {
1129
+ "epoch": 0.78,
1130
+ "grad_norm": 0.810093104839325,
1131
+ "learning_rate": 0.00018522168236559695,
1132
+ "loss": 1.2033,
1133
+ "step": 156
1134
+ },
1135
+ {
1136
+ "epoch": 0.79,
1137
+ "grad_norm": 0.7487497329711914,
1138
+ "learning_rate": 0.0001850091745080345,
1139
+ "loss": 1.1043,
1140
+ "step": 157
1141
+ },
1142
+ {
1143
+ "epoch": 0.79,
1144
+ "grad_norm": 0.6162795424461365,
1145
+ "learning_rate": 0.00018479527336050878,
1146
+ "loss": 1.2486,
1147
+ "step": 158
1148
+ },
1149
+ {
1150
+ "epoch": 0.8,
1151
+ "grad_norm": 0.5720970034599304,
1152
+ "learning_rate": 0.00018457998242883344,
1153
+ "loss": 1.0381,
1154
+ "step": 159
1155
+ },
1156
+ {
1157
+ "epoch": 0.8,
1158
+ "grad_norm": 0.6686292886734009,
1159
+ "learning_rate": 0.00018436330524160047,
1160
+ "loss": 1.502,
1161
+ "step": 160
1162
+ },
1163
+ {
1164
+ "epoch": 0.81,
1165
+ "grad_norm": 0.5931655764579773,
1166
+ "learning_rate": 0.00018414524535012244,
1167
+ "loss": 1.0813,
1168
+ "step": 161
1169
+ },
1170
+ {
1171
+ "epoch": 0.81,
1172
+ "grad_norm": 0.6548634171485901,
1173
+ "learning_rate": 0.00018392580632837423,
1174
+ "loss": 1.3147,
1175
+ "step": 162
1176
+ },
1177
+ {
1178
+ "epoch": 0.82,
1179
+ "grad_norm": 0.559681236743927,
1180
+ "learning_rate": 0.00018370499177293464,
1181
+ "loss": 1.1096,
1182
+ "step": 163
1183
+ },
1184
+ {
1185
+ "epoch": 0.82,
1186
+ "grad_norm": 0.6365666389465332,
1187
+ "learning_rate": 0.00018348280530292713,
1188
+ "loss": 1.2062,
1189
+ "step": 164
1190
+ },
1191
+ {
1192
+ "epoch": 0.83,
1193
+ "grad_norm": 0.616242527961731,
1194
+ "learning_rate": 0.00018325925055996076,
1195
+ "loss": 1.1219,
1196
+ "step": 165
1197
+ },
1198
+ {
1199
+ "epoch": 0.83,
1200
+ "grad_norm": 0.6588903069496155,
1201
+ "learning_rate": 0.0001830343312080704,
1202
+ "loss": 1.2697,
1203
+ "step": 166
1204
+ },
1205
+ {
1206
+ "epoch": 0.84,
1207
+ "grad_norm": 0.5880855321884155,
1208
+ "learning_rate": 0.00018280805093365672,
1209
+ "loss": 1.1511,
1210
+ "step": 167
1211
+ },
1212
+ {
1213
+ "epoch": 0.84,
1214
+ "grad_norm": 0.7549880743026733,
1215
+ "learning_rate": 0.00018258041344542566,
1216
+ "loss": 1.2181,
1217
+ "step": 168
1218
+ },
1219
+ {
1220
+ "epoch": 0.85,
1221
+ "grad_norm": 0.6862443089485168,
1222
+ "learning_rate": 0.00018235142247432782,
1223
+ "loss": 1.8496,
1224
+ "step": 169
1225
+ },
1226
+ {
1227
+ "epoch": 0.85,
1228
+ "grad_norm": 0.5903118848800659,
1229
+ "learning_rate": 0.0001821210817734972,
1230
+ "loss": 1.2092,
1231
+ "step": 170
1232
+ },
1233
+ {
1234
+ "epoch": 0.86,
1235
+ "grad_norm": 0.6936279535293579,
1236
+ "learning_rate": 0.00018188939511818965,
1237
+ "loss": 1.0635,
1238
+ "step": 171
1239
+ },
1240
+ {
1241
+ "epoch": 0.86,
1242
+ "grad_norm": 0.6887457370758057,
1243
+ "learning_rate": 0.0001816563663057211,
1244
+ "loss": 0.9387,
1245
+ "step": 172
1246
+ },
1247
+ {
1248
+ "epoch": 0.87,
1249
+ "grad_norm": 0.6930254101753235,
1250
+ "learning_rate": 0.00018142199915540527,
1251
+ "loss": 1.1651,
1252
+ "step": 173
1253
+ },
1254
+ {
1255
+ "epoch": 0.87,
1256
+ "grad_norm": 0.6529977321624756,
1257
+ "learning_rate": 0.00018118629750849105,
1258
+ "loss": 1.2512,
1259
+ "step": 174
1260
+ },
1261
+ {
1262
+ "epoch": 0.88,
1263
+ "grad_norm": 0.705954372882843,
1264
+ "learning_rate": 0.0001809492652280996,
1265
+ "loss": 1.2601,
1266
+ "step": 175
1267
+ },
1268
+ {
1269
+ "epoch": 0.88,
1270
+ "grad_norm": 0.6263706088066101,
1271
+ "learning_rate": 0.00018071090619916093,
1272
+ "loss": 1.0446,
1273
+ "step": 176
1274
+ },
1275
+ {
1276
+ "epoch": 0.89,
1277
+ "grad_norm": 0.7754440307617188,
1278
+ "learning_rate": 0.00018047122432835038,
1279
+ "loss": 1.2517,
1280
+ "step": 177
1281
+ },
1282
+ {
1283
+ "epoch": 0.89,
1284
+ "grad_norm": 0.6904909610748291,
1285
+ "learning_rate": 0.0001802302235440245,
1286
+ "loss": 1.3028,
1287
+ "step": 178
1288
+ },
1289
+ {
1290
+ "epoch": 0.9,
1291
+ "grad_norm": 0.6373815536499023,
1292
+ "learning_rate": 0.0001799879077961566,
1293
+ "loss": 0.7538,
1294
+ "step": 179
1295
+ },
1296
+ {
1297
+ "epoch": 0.9,
1298
+ "grad_norm": 0.6192349791526794,
1299
+ "learning_rate": 0.00017974428105627208,
1300
+ "loss": 1.1583,
1301
+ "step": 180
1302
+ },
1303
+ {
1304
+ "epoch": 0.91,
1305
+ "grad_norm": 0.6500440239906311,
1306
+ "learning_rate": 0.00017949934731738347,
1307
+ "loss": 1.189,
1308
+ "step": 181
1309
+ },
1310
+ {
1311
+ "epoch": 0.91,
1312
+ "grad_norm": 0.5701293349266052,
1313
+ "learning_rate": 0.0001792531105939247,
1314
+ "loss": 0.9937,
1315
+ "step": 182
1316
+ },
1317
+ {
1318
+ "epoch": 0.92,
1319
+ "grad_norm": 0.6383854150772095,
1320
+ "learning_rate": 0.0001790055749216856,
1321
+ "loss": 1.0381,
1322
+ "step": 183
1323
+ },
1324
+ {
1325
+ "epoch": 0.92,
1326
+ "grad_norm": 0.7212352156639099,
1327
+ "learning_rate": 0.00017875674435774547,
1328
+ "loss": 1.2023,
1329
+ "step": 184
1330
+ },
1331
+ {
1332
+ "epoch": 0.93,
1333
+ "grad_norm": 0.7195665836334229,
1334
+ "learning_rate": 0.00017850662298040678,
1335
+ "loss": 1.4138,
1336
+ "step": 185
1337
+ },
1338
+ {
1339
+ "epoch": 0.93,
1340
+ "grad_norm": 0.6174137592315674,
1341
+ "learning_rate": 0.0001782552148891283,
1342
+ "loss": 0.8007,
1343
+ "step": 186
1344
+ },
1345
+ {
1346
+ "epoch": 0.94,
1347
+ "grad_norm": 0.672179102897644,
1348
+ "learning_rate": 0.00017800252420445788,
1349
+ "loss": 1.1403,
1350
+ "step": 187
1351
+ },
1352
+ {
1353
+ "epoch": 0.94,
1354
+ "grad_norm": 0.6487817168235779,
1355
+ "learning_rate": 0.00017774855506796496,
1356
+ "loss": 1.169,
1357
+ "step": 188
1358
+ },
1359
+ {
1360
+ "epoch": 0.95,
1361
+ "grad_norm": 0.7027740478515625,
1362
+ "learning_rate": 0.0001774933116421725,
1363
+ "loss": 1.2268,
1364
+ "step": 189
1365
+ },
1366
+ {
1367
+ "epoch": 0.95,
1368
+ "grad_norm": 0.7178415060043335,
1369
+ "learning_rate": 0.00017723679811048904,
1370
+ "loss": 1.2785,
1371
+ "step": 190
1372
+ },
1373
+ {
1374
+ "epoch": 0.96,
1375
+ "grad_norm": 0.682354748249054,
1376
+ "learning_rate": 0.00017697901867713995,
1377
+ "loss": 1.2195,
1378
+ "step": 191
1379
+ },
1380
+ {
1381
+ "epoch": 0.96,
1382
+ "grad_norm": 0.7199010252952576,
1383
+ "learning_rate": 0.00017671997756709863,
1384
+ "loss": 1.4132,
1385
+ "step": 192
1386
+ },
1387
+ {
1388
+ "epoch": 0.97,
1389
+ "grad_norm": 0.7743118405342102,
1390
+ "learning_rate": 0.0001764596790260171,
1391
+ "loss": 0.9824,
1392
+ "step": 193
1393
+ },
1394
+ {
1395
+ "epoch": 0.97,
1396
+ "grad_norm": 0.7540227174758911,
1397
+ "learning_rate": 0.00017619812732015664,
1398
+ "loss": 1.0527,
1399
+ "step": 194
1400
+ },
1401
+ {
1402
+ "epoch": 0.98,
1403
+ "grad_norm": 0.6113067269325256,
1404
+ "learning_rate": 0.00017593532673631766,
1405
+ "loss": 1.2446,
1406
+ "step": 195
1407
+ },
1408
+ {
1409
+ "epoch": 0.98,
1410
+ "grad_norm": 0.6951828598976135,
1411
+ "learning_rate": 0.00017567128158176953,
1412
+ "loss": 1.3333,
1413
+ "step": 196
1414
+ },
1415
+ {
1416
+ "epoch": 0.99,
1417
+ "grad_norm": 0.570866584777832,
1418
+ "learning_rate": 0.00017540599618418007,
1419
+ "loss": 1.0012,
1420
+ "step": 197
1421
+ },
1422
+ {
1423
+ "epoch": 0.99,
1424
+ "grad_norm": 0.5432811379432678,
1425
+ "learning_rate": 0.00017513947489154443,
1426
+ "loss": 1.1343,
1427
+ "step": 198
1428
+ },
1429
+ {
1430
+ "epoch": 1.0,
1431
+ "grad_norm": 0.6711558103561401,
1432
+ "learning_rate": 0.00017487172207211396,
1433
+ "loss": 1.0945,
1434
+ "step": 199
1435
+ },
1436
+ {
1437
+ "epoch": 1.01,
1438
+ "grad_norm": 0.675626814365387,
1439
+ "learning_rate": 0.0001746027421143246,
1440
+ "loss": 1.2807,
1441
+ "step": 200
1442
+ },
1443
+ {
1444
+ "epoch": 1.01,
1445
+ "eval_loss": 1.1153115034103394,
1446
+ "eval_runtime": 3.0007,
1447
+ "eval_samples_per_second": 33.326,
1448
+ "eval_steps_per_second": 16.663,
1449
+ "step": 200
1450
+ },
1451
+ {
1452
+ "epoch": 1.01,
1453
+ "grad_norm": 0.6204088926315308,
1454
+ "learning_rate": 0.00017433253942672496,
1455
+ "loss": 1.2167,
1456
+ "step": 201
1457
+ },
1458
+ {
1459
+ "epoch": 1.02,
1460
+ "grad_norm": 0.6080848574638367,
1461
+ "learning_rate": 0.000174061118437904,
1462
+ "loss": 0.979,
1463
+ "step": 202
1464
+ },
1465
+ {
1466
+ "epoch": 1.02,
1467
+ "grad_norm": 0.8325397372245789,
1468
+ "learning_rate": 0.00017378848359641847,
1469
+ "loss": 0.9095,
1470
+ "step": 203
1471
+ },
1472
+ {
1473
+ "epoch": 1.01,
1474
+ "grad_norm": 0.6108893752098083,
1475
+ "learning_rate": 0.00017351463937072004,
1476
+ "loss": 1.0784,
1477
+ "step": 204
1478
+ },
1479
+ {
1480
+ "epoch": 1.01,
1481
+ "grad_norm": 0.6140009164810181,
1482
+ "learning_rate": 0.00017323959024908209,
1483
+ "loss": 1.131,
1484
+ "step": 205
1485
+ },
1486
+ {
1487
+ "epoch": 1.02,
1488
+ "grad_norm": 0.7503536343574524,
1489
+ "learning_rate": 0.00017296334073952605,
1490
+ "loss": 1.0152,
1491
+ "step": 206
1492
+ },
1493
+ {
1494
+ "epoch": 1.02,
1495
+ "grad_norm": 0.6903036236763,
1496
+ "learning_rate": 0.0001726858953697475,
1497
+ "loss": 1.1751,
1498
+ "step": 207
1499
+ },
1500
+ {
1501
+ "epoch": 1.03,
1502
+ "grad_norm": 0.6842136979103088,
1503
+ "learning_rate": 0.00017240725868704218,
1504
+ "loss": 0.9362,
1505
+ "step": 208
1506
+ },
1507
+ {
1508
+ "epoch": 1.03,
1509
+ "grad_norm": 0.6317443251609802,
1510
+ "learning_rate": 0.00017212743525823112,
1511
+ "loss": 1.0199,
1512
+ "step": 209
1513
+ },
1514
+ {
1515
+ "epoch": 1.04,
1516
+ "grad_norm": 0.6331597566604614,
1517
+ "learning_rate": 0.0001718464296695861,
1518
+ "loss": 0.8634,
1519
+ "step": 210
1520
+ },
1521
+ {
1522
+ "epoch": 1.04,
1523
+ "grad_norm": 0.7953663468360901,
1524
+ "learning_rate": 0.0001715642465267543,
1525
+ "loss": 1.0635,
1526
+ "step": 211
1527
+ },
1528
+ {
1529
+ "epoch": 1.05,
1530
+ "grad_norm": 0.6130046248435974,
1531
+ "learning_rate": 0.00017128089045468294,
1532
+ "loss": 0.8426,
1533
+ "step": 212
1534
+ },
1535
+ {
1536
+ "epoch": 1.05,
1537
+ "grad_norm": 0.5984789729118347,
1538
+ "learning_rate": 0.00017099636609754329,
1539
+ "loss": 0.7435,
1540
+ "step": 213
1541
+ },
1542
+ {
1543
+ "epoch": 1.06,
1544
+ "grad_norm": 0.8032707571983337,
1545
+ "learning_rate": 0.00017071067811865476,
1546
+ "loss": 0.9271,
1547
+ "step": 214
1548
+ },
1549
+ {
1550
+ "epoch": 1.06,
1551
+ "grad_norm": 0.78273606300354,
1552
+ "learning_rate": 0.00017042383120040834,
1553
+ "loss": 0.8695,
1554
+ "step": 215
1555
+ },
1556
+ {
1557
+ "epoch": 1.07,
1558
+ "grad_norm": 0.7779294848442078,
1559
+ "learning_rate": 0.00017013583004418993,
1560
+ "loss": 1.085,
1561
+ "step": 216
1562
+ },
1563
+ {
1564
+ "epoch": 1.07,
1565
+ "grad_norm": 0.7201984524726868,
1566
+ "learning_rate": 0.00016984667937030318,
1567
+ "loss": 0.8079,
1568
+ "step": 217
1569
+ },
1570
+ {
1571
+ "epoch": 1.08,
1572
+ "grad_norm": 0.6246169805526733,
1573
+ "learning_rate": 0.00016955638391789228,
1574
+ "loss": 0.7941,
1575
+ "step": 218
1576
+ },
1577
+ {
1578
+ "epoch": 1.08,
1579
+ "grad_norm": 0.7627923488616943,
1580
+ "learning_rate": 0.00016926494844486412,
1581
+ "loss": 0.9281,
1582
+ "step": 219
1583
+ },
1584
+ {
1585
+ "epoch": 1.09,
1586
+ "grad_norm": 0.6979169249534607,
1587
+ "learning_rate": 0.00016897237772781044,
1588
+ "loss": 0.8461,
1589
+ "step": 220
1590
+ },
1591
+ {
1592
+ "epoch": 1.09,
1593
+ "grad_norm": 0.7872811555862427,
1594
+ "learning_rate": 0.00016867867656192946,
1595
+ "loss": 0.9413,
1596
+ "step": 221
1597
+ },
1598
+ {
1599
+ "epoch": 1.1,
1600
+ "grad_norm": 0.7482172846794128,
1601
+ "learning_rate": 0.00016838384976094738,
1602
+ "loss": 0.9107,
1603
+ "step": 222
1604
+ },
1605
+ {
1606
+ "epoch": 1.1,
1607
+ "grad_norm": 0.8587368130683899,
1608
+ "learning_rate": 0.00016808790215703935,
1609
+ "loss": 0.9886,
1610
+ "step": 223
1611
+ },
1612
+ {
1613
+ "epoch": 1.11,
1614
+ "grad_norm": 0.732606828212738,
1615
+ "learning_rate": 0.00016779083860075033,
1616
+ "loss": 0.6831,
1617
+ "step": 224
1618
+ },
1619
+ {
1620
+ "epoch": 1.11,
1621
+ "grad_norm": 0.9272279143333435,
1622
+ "learning_rate": 0.0001674926639609157,
1623
+ "loss": 1.1396,
1624
+ "step": 225
1625
+ },
1626
+ {
1627
+ "epoch": 1.12,
1628
+ "grad_norm": 0.6473172307014465,
1629
+ "learning_rate": 0.00016719338312458124,
1630
+ "loss": 0.8299,
1631
+ "step": 226
1632
+ },
1633
+ {
1634
+ "epoch": 1.12,
1635
+ "grad_norm": 0.8427954316139221,
1636
+ "learning_rate": 0.00016689300099692332,
1637
+ "loss": 0.9203,
1638
+ "step": 227
1639
+ },
1640
+ {
1641
+ "epoch": 1.13,
1642
+ "grad_norm": 0.8205825090408325,
1643
+ "learning_rate": 0.00016659152250116812,
1644
+ "loss": 0.8532,
1645
+ "step": 228
1646
+ },
1647
+ {
1648
+ "epoch": 1.13,
1649
+ "grad_norm": 0.7522780299186707,
1650
+ "learning_rate": 0.00016628895257851135,
1651
+ "loss": 0.7687,
1652
+ "step": 229
1653
+ },
1654
+ {
1655
+ "epoch": 1.14,
1656
+ "grad_norm": 0.8582683205604553,
1657
+ "learning_rate": 0.000165985296188037,
1658
+ "loss": 0.9217,
1659
+ "step": 230
1660
+ },
1661
+ {
1662
+ "epoch": 1.14,
1663
+ "grad_norm": 0.8408709168434143,
1664
+ "learning_rate": 0.0001656805583066361,
1665
+ "loss": 1.0371,
1666
+ "step": 231
1667
+ },
1668
+ {
1669
+ "epoch": 1.15,
1670
+ "grad_norm": 0.9851942658424377,
1671
+ "learning_rate": 0.00016537474392892528,
1672
+ "loss": 1.044,
1673
+ "step": 232
1674
+ },
1675
+ {
1676
+ "epoch": 1.15,
1677
+ "grad_norm": 0.8463842868804932,
1678
+ "learning_rate": 0.00016506785806716465,
1679
+ "loss": 0.9521,
1680
+ "step": 233
1681
+ },
1682
+ {
1683
+ "epoch": 1.16,
1684
+ "grad_norm": 0.825255811214447,
1685
+ "learning_rate": 0.00016475990575117605,
1686
+ "loss": 0.8524,
1687
+ "step": 234
1688
+ },
1689
+ {
1690
+ "epoch": 1.16,
1691
+ "grad_norm": 1.1519947052001953,
1692
+ "learning_rate": 0.0001644508920282601,
1693
+ "loss": 0.9906,
1694
+ "step": 235
1695
+ },
1696
+ {
1697
+ "epoch": 1.17,
1698
+ "grad_norm": 0.8102966547012329,
1699
+ "learning_rate": 0.000164140821963114,
1700
+ "loss": 0.9192,
1701
+ "step": 236
1702
+ },
1703
+ {
1704
+ "epoch": 1.17,
1705
+ "grad_norm": 1.0159798860549927,
1706
+ "learning_rate": 0.0001638297006377481,
1707
+ "loss": 1.0234,
1708
+ "step": 237
1709
+ },
1710
+ {
1711
+ "epoch": 1.18,
1712
+ "grad_norm": 1.0157923698425293,
1713
+ "learning_rate": 0.00016351753315140287,
1714
+ "loss": 0.8921,
1715
+ "step": 238
1716
+ },
1717
+ {
1718
+ "epoch": 1.18,
1719
+ "grad_norm": 0.8466264009475708,
1720
+ "learning_rate": 0.00016320432462046516,
1721
+ "loss": 0.7098,
1722
+ "step": 239
1723
+ },
1724
+ {
1725
+ "epoch": 1.19,
1726
+ "grad_norm": 0.8298121690750122,
1727
+ "learning_rate": 0.00016289008017838445,
1728
+ "loss": 0.8517,
1729
+ "step": 240
1730
+ },
1731
+ {
1732
+ "epoch": 1.19,
1733
+ "grad_norm": 1.2163349390029907,
1734
+ "learning_rate": 0.00016257480497558873,
1735
+ "loss": 1.1172,
1736
+ "step": 241
1737
+ },
1738
+ {
1739
+ "epoch": 1.2,
1740
+ "grad_norm": 0.9839556217193604,
1741
+ "learning_rate": 0.0001622585041793999,
1742
+ "loss": 1.1022,
1743
+ "step": 242
1744
+ },
1745
+ {
1746
+ "epoch": 1.2,
1747
+ "grad_norm": 0.7986888289451599,
1748
+ "learning_rate": 0.00016194118297394936,
1749
+ "loss": 0.7826,
1750
+ "step": 243
1751
+ },
1752
+ {
1753
+ "epoch": 1.21,
1754
+ "grad_norm": 0.9318971037864685,
1755
+ "learning_rate": 0.00016162284656009274,
1756
+ "loss": 0.8899,
1757
+ "step": 244
1758
+ },
1759
+ {
1760
+ "epoch": 1.21,
1761
+ "grad_norm": 1.0234252214431763,
1762
+ "learning_rate": 0.00016130350015532496,
1763
+ "loss": 0.8831,
1764
+ "step": 245
1765
+ },
1766
+ {
1767
+ "epoch": 1.22,
1768
+ "grad_norm": 0.8264230489730835,
1769
+ "learning_rate": 0.00016098314899369446,
1770
+ "loss": 1.1389,
1771
+ "step": 246
1772
+ },
1773
+ {
1774
+ "epoch": 1.22,
1775
+ "grad_norm": 0.8845193982124329,
1776
+ "learning_rate": 0.0001606617983257176,
1777
+ "loss": 1.0822,
1778
+ "step": 247
1779
+ },
1780
+ {
1781
+ "epoch": 1.23,
1782
+ "grad_norm": 0.9044338464736938,
1783
+ "learning_rate": 0.00016033945341829248,
1784
+ "loss": 1.0556,
1785
+ "step": 248
1786
+ },
1787
+ {
1788
+ "epoch": 1.23,
1789
+ "grad_norm": 0.9660309553146362,
1790
+ "learning_rate": 0.00016001611955461265,
1791
+ "loss": 1.0331,
1792
+ "step": 249
1793
+ },
1794
+ {
1795
+ "epoch": 1.24,
1796
+ "grad_norm": 1.0728851556777954,
1797
+ "learning_rate": 0.0001596918020340805,
1798
+ "loss": 1.0465,
1799
+ "step": 250
1800
+ },
1801
+ {
1802
+ "epoch": 1.24,
1803
+ "eval_loss": 1.1568788290023804,
1804
+ "eval_runtime": 2.9063,
1805
+ "eval_samples_per_second": 34.408,
1806
+ "eval_steps_per_second": 17.204,
1807
+ "step": 250
1808
+ },
1809
+ {
1810
+ "epoch": 1.24,
1811
+ "grad_norm": 0.9447798728942871,
1812
+ "learning_rate": 0.00015936650617222063,
1813
+ "loss": 0.9487,
1814
+ "step": 251
1815
+ },
1816
+ {
1817
+ "epoch": 1.25,
1818
+ "grad_norm": 1.0429767370224,
1819
+ "learning_rate": 0.00015904023730059228,
1820
+ "loss": 1.006,
1821
+ "step": 252
1822
+ },
1823
+ {
1824
+ "epoch": 1.25,
1825
+ "grad_norm": 0.9871753454208374,
1826
+ "learning_rate": 0.00015871300076670234,
1827
+ "loss": 0.9494,
1828
+ "step": 253
1829
+ },
1830
+ {
1831
+ "epoch": 1.26,
1832
+ "grad_norm": 0.7644299268722534,
1833
+ "learning_rate": 0.00015838480193391754,
1834
+ "loss": 0.6077,
1835
+ "step": 254
1836
+ },
1837
+ {
1838
+ "epoch": 1.26,
1839
+ "grad_norm": 1.1654846668243408,
1840
+ "learning_rate": 0.0001580556461813766,
1841
+ "loss": 1.0632,
1842
+ "step": 255
1843
+ },
1844
+ {
1845
+ "epoch": 1.27,
1846
+ "grad_norm": 1.0508393049240112,
1847
+ "learning_rate": 0.00015772553890390197,
1848
+ "loss": 0.8754,
1849
+ "step": 256
1850
+ },
1851
+ {
1852
+ "epoch": 1.27,
1853
+ "grad_norm": 0.8676743507385254,
1854
+ "learning_rate": 0.0001573944855119115,
1855
+ "loss": 1.007,
1856
+ "step": 257
1857
+ },
1858
+ {
1859
+ "epoch": 1.28,
1860
+ "grad_norm": 1.178464412689209,
1861
+ "learning_rate": 0.00015706249143132982,
1862
+ "loss": 1.041,
1863
+ "step": 258
1864
+ },
1865
+ {
1866
+ "epoch": 1.28,
1867
+ "grad_norm": 1.0226370096206665,
1868
+ "learning_rate": 0.00015672956210349923,
1869
+ "loss": 1.1114,
1870
+ "step": 259
1871
+ },
1872
+ {
1873
+ "epoch": 1.29,
1874
+ "grad_norm": 0.9840787649154663,
1875
+ "learning_rate": 0.00015639570298509064,
1876
+ "loss": 0.9043,
1877
+ "step": 260
1878
+ },
1879
+ {
1880
+ "epoch": 1.29,
1881
+ "grad_norm": 1.0564519166946411,
1882
+ "learning_rate": 0.0001560609195480142,
1883
+ "loss": 0.9696,
1884
+ "step": 261
1885
+ },
1886
+ {
1887
+ "epoch": 1.3,
1888
+ "grad_norm": 0.9174713492393494,
1889
+ "learning_rate": 0.00015572521727932935,
1890
+ "loss": 0.9849,
1891
+ "step": 262
1892
+ },
1893
+ {
1894
+ "epoch": 1.3,
1895
+ "grad_norm": 0.7333153486251831,
1896
+ "learning_rate": 0.00015538860168115527,
1897
+ "loss": 0.7286,
1898
+ "step": 263
1899
+ },
1900
+ {
1901
+ "epoch": 1.31,
1902
+ "grad_norm": 0.9282216429710388,
1903
+ "learning_rate": 0.00015505107827058036,
1904
+ "loss": 0.8975,
1905
+ "step": 264
1906
+ },
1907
+ {
1908
+ "epoch": 1.31,
1909
+ "grad_norm": 1.003192663192749,
1910
+ "learning_rate": 0.00015471265257957202,
1911
+ "loss": 1.1836,
1912
+ "step": 265
1913
+ },
1914
+ {
1915
+ "epoch": 1.32,
1916
+ "grad_norm": 0.8726491928100586,
1917
+ "learning_rate": 0.00015437333015488587,
1918
+ "loss": 0.9313,
1919
+ "step": 266
1920
+ },
1921
+ {
1922
+ "epoch": 1.32,
1923
+ "grad_norm": 0.9721888899803162,
1924
+ "learning_rate": 0.00015403311655797492,
1925
+ "loss": 0.8935,
1926
+ "step": 267
1927
+ },
1928
+ {
1929
+ "epoch": 1.33,
1930
+ "grad_norm": 1.0440247058868408,
1931
+ "learning_rate": 0.0001536920173648984,
1932
+ "loss": 0.9741,
1933
+ "step": 268
1934
+ },
1935
+ {
1936
+ "epoch": 1.33,
1937
+ "grad_norm": 0.9814698100090027,
1938
+ "learning_rate": 0.00015335003816623028,
1939
+ "loss": 0.8982,
1940
+ "step": 269
1941
+ },
1942
+ {
1943
+ "epoch": 1.34,
1944
+ "grad_norm": 0.904926598072052,
1945
+ "learning_rate": 0.00015300718456696778,
1946
+ "loss": 0.8579,
1947
+ "step": 270
1948
+ },
1949
+ {
1950
+ "epoch": 1.34,
1951
+ "grad_norm": 1.0483490228652954,
1952
+ "learning_rate": 0.00015266346218643947,
1953
+ "loss": 0.8108,
1954
+ "step": 271
1955
+ },
1956
+ {
1957
+ "epoch": 1.35,
1958
+ "grad_norm": 0.9156501293182373,
1959
+ "learning_rate": 0.000152318876658213,
1960
+ "loss": 0.9442,
1961
+ "step": 272
1962
+ },
1963
+ {
1964
+ "epoch": 1.35,
1965
+ "grad_norm": 0.9268532395362854,
1966
+ "learning_rate": 0.00015197343363000307,
1967
+ "loss": 0.8243,
1968
+ "step": 273
1969
+ },
1970
+ {
1971
+ "epoch": 1.36,
1972
+ "grad_norm": 0.8675321340560913,
1973
+ "learning_rate": 0.00015162713876357858,
1974
+ "loss": 0.7758,
1975
+ "step": 274
1976
+ },
1977
+ {
1978
+ "epoch": 1.36,
1979
+ "grad_norm": 0.9131675362586975,
1980
+ "learning_rate": 0.00015127999773467002,
1981
+ "loss": 0.8845,
1982
+ "step": 275
1983
+ },
1984
+ {
1985
+ "epoch": 1.37,
1986
+ "grad_norm": 1.0260008573532104,
1987
+ "learning_rate": 0.00015093201623287631,
1988
+ "loss": 0.9032,
1989
+ "step": 276
1990
+ },
1991
+ {
1992
+ "epoch": 1.37,
1993
+ "grad_norm": 1.044528841972351,
1994
+ "learning_rate": 0.00015058319996157172,
1995
+ "loss": 1.0489,
1996
+ "step": 277
1997
+ },
1998
+ {
1999
+ "epoch": 1.38,
2000
+ "grad_norm": 0.9959388375282288,
2001
+ "learning_rate": 0.0001502335546378122,
2002
+ "loss": 0.858,
2003
+ "step": 278
2004
+ },
2005
+ {
2006
+ "epoch": 1.38,
2007
+ "grad_norm": 0.8414021730422974,
2008
+ "learning_rate": 0.00014988308599224183,
2009
+ "loss": 0.782,
2010
+ "step": 279
2011
+ },
2012
+ {
2013
+ "epoch": 1.39,
2014
+ "grad_norm": 0.9205671548843384,
2015
+ "learning_rate": 0.00014953179976899878,
2016
+ "loss": 0.8376,
2017
+ "step": 280
2018
+ },
2019
+ {
2020
+ "epoch": 1.39,
2021
+ "grad_norm": 0.9481040239334106,
2022
+ "learning_rate": 0.0001491797017256212,
2023
+ "loss": 0.851,
2024
+ "step": 281
2025
+ },
2026
+ {
2027
+ "epoch": 1.4,
2028
+ "grad_norm": 0.8266577124595642,
2029
+ "learning_rate": 0.00014882679763295306,
2030
+ "loss": 0.7228,
2031
+ "step": 282
2032
+ },
2033
+ {
2034
+ "epoch": 1.4,
2035
+ "grad_norm": 1.0222742557525635,
2036
+ "learning_rate": 0.0001484730932750491,
2037
+ "loss": 0.7955,
2038
+ "step": 283
2039
+ },
2040
+ {
2041
+ "epoch": 1.41,
2042
+ "grad_norm": 1.0014468431472778,
2043
+ "learning_rate": 0.00014811859444908052,
2044
+ "loss": 0.9107,
2045
+ "step": 284
2046
+ },
2047
+ {
2048
+ "epoch": 1.41,
2049
+ "grad_norm": 0.9157910346984863,
2050
+ "learning_rate": 0.00014776330696523963,
2051
+ "loss": 1.0208,
2052
+ "step": 285
2053
+ },
2054
+ {
2055
+ "epoch": 1.42,
2056
+ "grad_norm": 1.0565227270126343,
2057
+ "learning_rate": 0.00014740723664664483,
2058
+ "loss": 0.6496,
2059
+ "step": 286
2060
+ },
2061
+ {
2062
+ "epoch": 1.42,
2063
+ "grad_norm": 1.0323175191879272,
2064
+ "learning_rate": 0.00014705038932924503,
2065
+ "loss": 1.0043,
2066
+ "step": 287
2067
+ },
2068
+ {
2069
+ "epoch": 1.43,
2070
+ "grad_norm": 1.0063213109970093,
2071
+ "learning_rate": 0.00014669277086172406,
2072
+ "loss": 1.1286,
2073
+ "step": 288
2074
+ },
2075
+ {
2076
+ "epoch": 1.43,
2077
+ "grad_norm": 0.8602890968322754,
2078
+ "learning_rate": 0.00014633438710540489,
2079
+ "loss": 0.7254,
2080
+ "step": 289
2081
+ },
2082
+ {
2083
+ "epoch": 1.44,
2084
+ "grad_norm": 0.9782769083976746,
2085
+ "learning_rate": 0.00014597524393415335,
2086
+ "loss": 0.7086,
2087
+ "step": 290
2088
+ },
2089
+ {
2090
+ "epoch": 1.44,
2091
+ "grad_norm": 0.9836515784263611,
2092
+ "learning_rate": 0.00014561534723428205,
2093
+ "loss": 0.8405,
2094
+ "step": 291
2095
+ },
2096
+ {
2097
+ "epoch": 1.45,
2098
+ "grad_norm": 1.0674123764038086,
2099
+ "learning_rate": 0.00014525470290445392,
2100
+ "loss": 1.0317,
2101
+ "step": 292
2102
+ },
2103
+ {
2104
+ "epoch": 1.45,
2105
+ "grad_norm": 0.9632031917572021,
2106
+ "learning_rate": 0.00014489331685558525,
2107
+ "loss": 0.9473,
2108
+ "step": 293
2109
+ },
2110
+ {
2111
+ "epoch": 1.46,
2112
+ "grad_norm": 1.0105828046798706,
2113
+ "learning_rate": 0.00014453119501074924,
2114
+ "loss": 0.8199,
2115
+ "step": 294
2116
+ },
2117
+ {
2118
+ "epoch": 1.46,
2119
+ "grad_norm": 1.0012938976287842,
2120
+ "learning_rate": 0.00014416834330507856,
2121
+ "loss": 0.9099,
2122
+ "step": 295
2123
+ },
2124
+ {
2125
+ "epoch": 1.47,
2126
+ "grad_norm": 1.0367400646209717,
2127
+ "learning_rate": 0.00014380476768566824,
2128
+ "loss": 1.0958,
2129
+ "step": 296
2130
+ },
2131
+ {
2132
+ "epoch": 1.47,
2133
+ "grad_norm": 0.7329337000846863,
2134
+ "learning_rate": 0.00014344047411147818,
2135
+ "loss": 0.6189,
2136
+ "step": 297
2137
+ },
2138
+ {
2139
+ "epoch": 1.48,
2140
+ "grad_norm": 0.9014643430709839,
2141
+ "learning_rate": 0.00014307546855323549,
2142
+ "loss": 0.8168,
2143
+ "step": 298
2144
+ },
2145
+ {
2146
+ "epoch": 1.48,
2147
+ "grad_norm": 0.7568360567092896,
2148
+ "learning_rate": 0.00014270975699333654,
2149
+ "loss": 0.7857,
2150
+ "step": 299
2151
+ },
2152
+ {
2153
+ "epoch": 1.49,
2154
+ "grad_norm": 0.9573479890823364,
2155
+ "learning_rate": 0.00014234334542574906,
2156
+ "loss": 0.9577,
2157
+ "step": 300
2158
+ },
2159
+ {
2160
+ "epoch": 1.49,
2161
+ "eval_loss": 1.149274230003357,
2162
+ "eval_runtime": 2.9222,
2163
+ "eval_samples_per_second": 34.22,
2164
+ "eval_steps_per_second": 17.11,
2165
+ "step": 300
2166
+ },
2167
+ {
2168
+ "epoch": 1.49,
2169
+ "grad_norm": 0.8575048446655273,
2170
+ "learning_rate": 0.00014197623985591373,
2171
+ "loss": 0.8521,
2172
+ "step": 301
2173
+ },
2174
+ {
2175
+ "epoch": 1.5,
2176
+ "grad_norm": 0.990980863571167,
2177
+ "learning_rate": 0.00014160844630064595,
2178
+ "loss": 1.0642,
2179
+ "step": 302
2180
+ },
2181
+ {
2182
+ "epoch": 1.5,
2183
+ "grad_norm": 1.1145374774932861,
2184
+ "learning_rate": 0.00014123997078803707,
2185
+ "loss": 0.8963,
2186
+ "step": 303
2187
+ },
2188
+ {
2189
+ "epoch": 1.51,
2190
+ "grad_norm": 0.9661507606506348,
2191
+ "learning_rate": 0.00014087081935735564,
2192
+ "loss": 0.9473,
2193
+ "step": 304
2194
+ },
2195
+ {
2196
+ "epoch": 1.51,
2197
+ "grad_norm": 1.019618272781372,
2198
+ "learning_rate": 0.00014050099805894837,
2199
+ "loss": 0.9048,
2200
+ "step": 305
2201
+ },
2202
+ {
2203
+ "epoch": 1.52,
2204
+ "grad_norm": 0.871661365032196,
2205
+ "learning_rate": 0.00014013051295414108,
2206
+ "loss": 0.6644,
2207
+ "step": 306
2208
+ },
2209
+ {
2210
+ "epoch": 1.52,
2211
+ "grad_norm": 0.9834782481193542,
2212
+ "learning_rate": 0.00013975937011513932,
2213
+ "loss": 0.9226,
2214
+ "step": 307
2215
+ },
2216
+ {
2217
+ "epoch": 1.53,
2218
+ "grad_norm": 0.9938518404960632,
2219
+ "learning_rate": 0.00013938757562492873,
2220
+ "loss": 0.9608,
2221
+ "step": 308
2222
+ },
2223
+ {
2224
+ "epoch": 1.53,
2225
+ "grad_norm": 1.0692541599273682,
2226
+ "learning_rate": 0.00013901513557717553,
2227
+ "loss": 0.9646,
2228
+ "step": 309
2229
+ },
2230
+ {
2231
+ "epoch": 1.54,
2232
+ "grad_norm": 1.039904236793518,
2233
+ "learning_rate": 0.00013864205607612648,
2234
+ "loss": 0.7799,
2235
+ "step": 310
2236
+ },
2237
+ {
2238
+ "epoch": 1.54,
2239
+ "grad_norm": 0.9138852953910828,
2240
+ "learning_rate": 0.000138268343236509,
2241
+ "loss": 0.8297,
2242
+ "step": 311
2243
+ },
2244
+ {
2245
+ "epoch": 1.55,
2246
+ "grad_norm": 1.01775324344635,
2247
+ "learning_rate": 0.00013789400318343068,
2248
+ "loss": 0.8992,
2249
+ "step": 312
2250
+ },
2251
+ {
2252
+ "epoch": 1.55,
2253
+ "grad_norm": 1.0052934885025024,
2254
+ "learning_rate": 0.0001375190420522792,
2255
+ "loss": 0.8212,
2256
+ "step": 313
2257
+ },
2258
+ {
2259
+ "epoch": 1.56,
2260
+ "grad_norm": 1.0567269325256348,
2261
+ "learning_rate": 0.00013714346598862166,
2262
+ "loss": 1.0402,
2263
+ "step": 314
2264
+ },
2265
+ {
2266
+ "epoch": 1.56,
2267
+ "grad_norm": 0.8707680702209473,
2268
+ "learning_rate": 0.00013676728114810367,
2269
+ "loss": 0.8864,
2270
+ "step": 315
2271
+ },
2272
+ {
2273
+ "epoch": 1.57,
2274
+ "grad_norm": 0.959578812122345,
2275
+ "learning_rate": 0.00013639049369634876,
2276
+ "loss": 0.7048,
2277
+ "step": 316
2278
+ },
2279
+ {
2280
+ "epoch": 1.57,
2281
+ "grad_norm": 1.0675721168518066,
2282
+ "learning_rate": 0.00013601310980885714,
2283
+ "loss": 1.0025,
2284
+ "step": 317
2285
+ },
2286
+ {
2287
+ "epoch": 1.58,
2288
+ "grad_norm": 0.8831722736358643,
2289
+ "learning_rate": 0.0001356351356709045,
2290
+ "loss": 0.8058,
2291
+ "step": 318
2292
+ },
2293
+ {
2294
+ "epoch": 1.58,
2295
+ "grad_norm": 1.0400885343551636,
2296
+ "learning_rate": 0.00013525657747744072,
2297
+ "loss": 1.0273,
2298
+ "step": 319
2299
+ },
2300
+ {
2301
+ "epoch": 1.59,
2302
+ "grad_norm": 1.0046364068984985,
2303
+ "learning_rate": 0.00013487744143298822,
2304
+ "loss": 0.8441,
2305
+ "step": 320
2306
+ },
2307
+ {
2308
+ "epoch": 1.59,
2309
+ "grad_norm": 1.029714822769165,
2310
+ "learning_rate": 0.0001344977337515404,
2311
+ "loss": 0.7771,
2312
+ "step": 321
2313
+ },
2314
+ {
2315
+ "epoch": 1.6,
2316
+ "grad_norm": 0.8168841004371643,
2317
+ "learning_rate": 0.0001341174606564596,
2318
+ "loss": 0.8024,
2319
+ "step": 322
2320
+ },
2321
+ {
2322
+ "epoch": 1.6,
2323
+ "grad_norm": 0.9833108186721802,
2324
+ "learning_rate": 0.00013373662838037537,
2325
+ "loss": 0.9065,
2326
+ "step": 323
2327
+ },
2328
+ {
2329
+ "epoch": 1.61,
2330
+ "grad_norm": 0.9366996884346008,
2331
+ "learning_rate": 0.00013335524316508208,
2332
+ "loss": 0.9436,
2333
+ "step": 324
2334
+ },
2335
+ {
2336
+ "epoch": 1.61,
2337
+ "grad_norm": 0.8757138848304749,
2338
+ "learning_rate": 0.00013297331126143667,
2339
+ "loss": 0.8399,
2340
+ "step": 325
2341
+ },
2342
+ {
2343
+ "epoch": 1.62,
2344
+ "grad_norm": 1.1467972993850708,
2345
+ "learning_rate": 0.00013259083892925633,
2346
+ "loss": 1.1416,
2347
+ "step": 326
2348
+ },
2349
+ {
2350
+ "epoch": 1.62,
2351
+ "grad_norm": 0.9916189312934875,
2352
+ "learning_rate": 0.00013220783243721572,
2353
+ "loss": 0.9531,
2354
+ "step": 327
2355
+ },
2356
+ {
2357
+ "epoch": 1.63,
2358
+ "grad_norm": 0.9911974668502808,
2359
+ "learning_rate": 0.0001318242980627444,
2360
+ "loss": 0.9476,
2361
+ "step": 328
2362
+ },
2363
+ {
2364
+ "epoch": 1.63,
2365
+ "grad_norm": 1.0219913721084595,
2366
+ "learning_rate": 0.0001314402420919238,
2367
+ "loss": 0.9288,
2368
+ "step": 329
2369
+ },
2370
+ {
2371
+ "epoch": 1.64,
2372
+ "grad_norm": 1.0889464616775513,
2373
+ "learning_rate": 0.00013105567081938424,
2374
+ "loss": 0.8025,
2375
+ "step": 330
2376
+ },
2377
+ {
2378
+ "epoch": 1.64,
2379
+ "grad_norm": 0.8797928690910339,
2380
+ "learning_rate": 0.00013067059054820183,
2381
+ "loss": 0.9002,
2382
+ "step": 331
2383
+ },
2384
+ {
2385
+ "epoch": 1.65,
2386
+ "grad_norm": 1.0043346881866455,
2387
+ "learning_rate": 0.00013028500758979506,
2388
+ "loss": 0.8971,
2389
+ "step": 332
2390
+ },
2391
+ {
2392
+ "epoch": 1.65,
2393
+ "grad_norm": 0.9221352934837341,
2394
+ "learning_rate": 0.00012989892826382145,
2395
+ "loss": 0.8181,
2396
+ "step": 333
2397
+ },
2398
+ {
2399
+ "epoch": 1.66,
2400
+ "grad_norm": 1.2053778171539307,
2401
+ "learning_rate": 0.00012951235889807386,
2402
+ "loss": 0.9374,
2403
+ "step": 334
2404
+ },
2405
+ {
2406
+ "epoch": 1.66,
2407
+ "grad_norm": 1.2230528593063354,
2408
+ "learning_rate": 0.00012912530582837682,
2409
+ "loss": 0.9123,
2410
+ "step": 335
2411
+ },
2412
+ {
2413
+ "epoch": 1.67,
2414
+ "grad_norm": 0.8403642773628235,
2415
+ "learning_rate": 0.00012873777539848283,
2416
+ "loss": 0.9323,
2417
+ "step": 336
2418
+ },
2419
+ {
2420
+ "epoch": 1.67,
2421
+ "grad_norm": 1.1632657051086426,
2422
+ "learning_rate": 0.00012834977395996818,
2423
+ "loss": 1.1916,
2424
+ "step": 337
2425
+ },
2426
+ {
2427
+ "epoch": 1.68,
2428
+ "grad_norm": 0.9937611222267151,
2429
+ "learning_rate": 0.0001279613078721289,
2430
+ "loss": 1.141,
2431
+ "step": 338
2432
+ },
2433
+ {
2434
+ "epoch": 1.68,
2435
+ "grad_norm": 0.8973978161811829,
2436
+ "learning_rate": 0.0001275723835018767,
2437
+ "loss": 0.8399,
2438
+ "step": 339
2439
+ },
2440
+ {
2441
+ "epoch": 1.69,
2442
+ "grad_norm": 1.0466402769088745,
2443
+ "learning_rate": 0.0001271830072236343,
2444
+ "loss": 0.8127,
2445
+ "step": 340
2446
+ },
2447
+ {
2448
+ "epoch": 1.69,
2449
+ "grad_norm": 0.9691051244735718,
2450
+ "learning_rate": 0.0001267931854192313,
2451
+ "loss": 0.9794,
2452
+ "step": 341
2453
+ },
2454
+ {
2455
+ "epoch": 1.7,
2456
+ "grad_norm": 0.925682544708252,
2457
+ "learning_rate": 0.0001264029244777993,
2458
+ "loss": 0.8233,
2459
+ "step": 342
2460
+ },
2461
+ {
2462
+ "epoch": 1.7,
2463
+ "grad_norm": 0.9783278703689575,
2464
+ "learning_rate": 0.00012601223079566743,
2465
+ "loss": 0.9542,
2466
+ "step": 343
2467
+ },
2468
+ {
2469
+ "epoch": 1.71,
2470
+ "grad_norm": 0.9945007562637329,
2471
+ "learning_rate": 0.00012562111077625722,
2472
+ "loss": 1.0757,
2473
+ "step": 344
2474
+ },
2475
+ {
2476
+ "epoch": 1.71,
2477
+ "grad_norm": 1.1597148180007935,
2478
+ "learning_rate": 0.000125229570829978,
2479
+ "loss": 1.1052,
2480
+ "step": 345
2481
+ },
2482
+ {
2483
+ "epoch": 1.72,
2484
+ "grad_norm": 0.7987023591995239,
2485
+ "learning_rate": 0.0001248376173741215,
2486
+ "loss": 0.8602,
2487
+ "step": 346
2488
+ },
2489
+ {
2490
+ "epoch": 1.72,
2491
+ "grad_norm": 0.8969370126724243,
2492
+ "learning_rate": 0.00012444525683275688,
2493
+ "loss": 1.6019,
2494
+ "step": 347
2495
+ },
2496
+ {
2497
+ "epoch": 1.73,
2498
+ "grad_norm": 1.0622583627700806,
2499
+ "learning_rate": 0.00012405249563662537,
2500
+ "loss": 1.0735,
2501
+ "step": 348
2502
+ },
2503
+ {
2504
+ "epoch": 1.73,
2505
+ "grad_norm": 1.0987950563430786,
2506
+ "learning_rate": 0.00012365934022303491,
2507
+ "loss": 0.9973,
2508
+ "step": 349
2509
+ },
2510
+ {
2511
+ "epoch": 1.74,
2512
+ "grad_norm": 0.9930221438407898,
2513
+ "learning_rate": 0.00012326579703575462,
2514
+ "loss": 1.1257,
2515
+ "step": 350
2516
+ },
2517
+ {
2518
+ "epoch": 1.74,
2519
+ "eval_loss": 1.1461950540542603,
2520
+ "eval_runtime": 2.9343,
2521
+ "eval_samples_per_second": 34.08,
2522
+ "eval_steps_per_second": 17.04,
2523
+ "step": 350
2524
+ },
2525
+ {
2526
+ "epoch": 1.74,
2527
+ "grad_norm": 1.0799540281295776,
2528
+ "learning_rate": 0.00012287187252490913,
2529
+ "loss": 0.8758,
2530
+ "step": 351
2531
+ },
2532
+ {
2533
+ "epoch": 1.75,
2534
+ "grad_norm": 1.0633143186569214,
2535
+ "learning_rate": 0.00012247757314687297,
2536
+ "loss": 1.0396,
2537
+ "step": 352
2538
+ },
2539
+ {
2540
+ "epoch": 1.75,
2541
+ "grad_norm": 0.9504884481430054,
2542
+ "learning_rate": 0.00012208290536416463,
2543
+ "loss": 0.8192,
2544
+ "step": 353
2545
+ },
2546
+ {
2547
+ "epoch": 1.76,
2548
+ "grad_norm": 0.8587303161621094,
2549
+ "learning_rate": 0.00012168787564534078,
2550
+ "loss": 0.748,
2551
+ "step": 354
2552
+ },
2553
+ {
2554
+ "epoch": 1.76,
2555
+ "grad_norm": 1.3652898073196411,
2556
+ "learning_rate": 0.0001212924904648902,
2557
+ "loss": 1.0768,
2558
+ "step": 355
2559
+ },
2560
+ {
2561
+ "epoch": 1.77,
2562
+ "grad_norm": 1.0679266452789307,
2563
+ "learning_rate": 0.00012089675630312754,
2564
+ "loss": 0.9099,
2565
+ "step": 356
2566
+ },
2567
+ {
2568
+ "epoch": 1.77,
2569
+ "grad_norm": 1.2426522970199585,
2570
+ "learning_rate": 0.00012050067964608724,
2571
+ "loss": 0.9869,
2572
+ "step": 357
2573
+ },
2574
+ {
2575
+ "epoch": 1.78,
2576
+ "grad_norm": 0.9639490246772766,
2577
+ "learning_rate": 0.00012010426698541728,
2578
+ "loss": 0.6993,
2579
+ "step": 358
2580
+ },
2581
+ {
2582
+ "epoch": 1.78,
2583
+ "grad_norm": 1.1884175539016724,
2584
+ "learning_rate": 0.0001197075248182726,
2585
+ "loss": 0.9868,
2586
+ "step": 359
2587
+ },
2588
+ {
2589
+ "epoch": 1.79,
2590
+ "grad_norm": 0.9860052466392517,
2591
+ "learning_rate": 0.00011931045964720881,
2592
+ "loss": 0.7148,
2593
+ "step": 360
2594
+ },
2595
+ {
2596
+ "epoch": 1.79,
2597
+ "grad_norm": 0.8812693357467651,
2598
+ "learning_rate": 0.00011891307798007536,
2599
+ "loss": 0.9295,
2600
+ "step": 361
2601
+ },
2602
+ {
2603
+ "epoch": 1.8,
2604
+ "grad_norm": 1.032242774963379,
2605
+ "learning_rate": 0.00011851538632990921,
2606
+ "loss": 1.2292,
2607
+ "step": 362
2608
+ },
2609
+ {
2610
+ "epoch": 1.8,
2611
+ "grad_norm": 0.9777809381484985,
2612
+ "learning_rate": 0.00011811739121482777,
2613
+ "loss": 1.0646,
2614
+ "step": 363
2615
+ },
2616
+ {
2617
+ "epoch": 1.81,
2618
+ "grad_norm": 1.0464228391647339,
2619
+ "learning_rate": 0.0001177190991579223,
2620
+ "loss": 0.9703,
2621
+ "step": 364
2622
+ },
2623
+ {
2624
+ "epoch": 1.81,
2625
+ "grad_norm": 0.9763212203979492,
2626
+ "learning_rate": 0.00011732051668715081,
2627
+ "loss": 0.7753,
2628
+ "step": 365
2629
+ },
2630
+ {
2631
+ "epoch": 1.82,
2632
+ "grad_norm": 1.114912748336792,
2633
+ "learning_rate": 0.00011692165033523117,
2634
+ "loss": 0.9979,
2635
+ "step": 366
2636
+ },
2637
+ {
2638
+ "epoch": 1.82,
2639
+ "grad_norm": 0.8752657771110535,
2640
+ "learning_rate": 0.00011652250663953415,
2641
+ "loss": 0.9964,
2642
+ "step": 367
2643
+ },
2644
+ {
2645
+ "epoch": 1.83,
2646
+ "grad_norm": 0.9158682823181152,
2647
+ "learning_rate": 0.00011612309214197599,
2648
+ "loss": 0.7576,
2649
+ "step": 368
2650
+ },
2651
+ {
2652
+ "epoch": 1.83,
2653
+ "grad_norm": 0.8457457423210144,
2654
+ "learning_rate": 0.00011572341338891144,
2655
+ "loss": 0.9144,
2656
+ "step": 369
2657
+ },
2658
+ {
2659
+ "epoch": 1.84,
2660
+ "grad_norm": 1.0021049976348877,
2661
+ "learning_rate": 0.00011532347693102632,
2662
+ "loss": 0.9226,
2663
+ "step": 370
2664
+ },
2665
+ {
2666
+ "epoch": 1.84,
2667
+ "grad_norm": 0.9614117741584778,
2668
+ "learning_rate": 0.00011492328932323022,
2669
+ "loss": 1.0214,
2670
+ "step": 371
2671
+ },
2672
+ {
2673
+ "epoch": 1.85,
2674
+ "grad_norm": 0.9289172291755676,
2675
+ "learning_rate": 0.00011452285712454904,
2676
+ "loss": 0.8793,
2677
+ "step": 372
2678
+ },
2679
+ {
2680
+ "epoch": 1.85,
2681
+ "grad_norm": 1.0654929876327515,
2682
+ "learning_rate": 0.00011412218689801748,
2683
+ "loss": 1.1519,
2684
+ "step": 373
2685
+ },
2686
+ {
2687
+ "epoch": 1.86,
2688
+ "grad_norm": 1.0563515424728394,
2689
+ "learning_rate": 0.00011372128521057155,
2690
+ "loss": 0.9859,
2691
+ "step": 374
2692
+ },
2693
+ {
2694
+ "epoch": 1.86,
2695
+ "grad_norm": 1.011228322982788,
2696
+ "learning_rate": 0.00011332015863294076,
2697
+ "loss": 0.9138,
2698
+ "step": 375
2699
+ },
2700
+ {
2701
+ "epoch": 1.87,
2702
+ "grad_norm": 0.942287802696228,
2703
+ "learning_rate": 0.00011291881373954065,
2704
+ "loss": 0.8865,
2705
+ "step": 376
2706
+ },
2707
+ {
2708
+ "epoch": 1.87,
2709
+ "grad_norm": 0.9734610319137573,
2710
+ "learning_rate": 0.00011251725710836489,
2711
+ "loss": 0.8578,
2712
+ "step": 377
2713
+ },
2714
+ {
2715
+ "epoch": 1.88,
2716
+ "grad_norm": 1.184990406036377,
2717
+ "learning_rate": 0.00011211549532087749,
2718
+ "loss": 1.0107,
2719
+ "step": 378
2720
+ },
2721
+ {
2722
+ "epoch": 1.88,
2723
+ "grad_norm": 1.033831238746643,
2724
+ "learning_rate": 0.00011171353496190498,
2725
+ "loss": 1.0496,
2726
+ "step": 379
2727
+ },
2728
+ {
2729
+ "epoch": 1.89,
2730
+ "grad_norm": 1.018054485321045,
2731
+ "learning_rate": 0.00011131138261952845,
2732
+ "loss": 0.8782,
2733
+ "step": 380
2734
+ },
2735
+ {
2736
+ "epoch": 1.89,
2737
+ "grad_norm": 0.9694205522537231,
2738
+ "learning_rate": 0.00011090904488497549,
2739
+ "loss": 0.9928,
2740
+ "step": 381
2741
+ },
2742
+ {
2743
+ "epoch": 1.9,
2744
+ "grad_norm": 0.9095280170440674,
2745
+ "learning_rate": 0.0001105065283525124,
2746
+ "loss": 0.9821,
2747
+ "step": 382
2748
+ },
2749
+ {
2750
+ "epoch": 1.9,
2751
+ "grad_norm": 0.8029172420501709,
2752
+ "learning_rate": 0.00011010383961933581,
2753
+ "loss": 0.6811,
2754
+ "step": 383
2755
+ },
2756
+ {
2757
+ "epoch": 1.91,
2758
+ "grad_norm": 0.9388089776039124,
2759
+ "learning_rate": 0.00010970098528546481,
2760
+ "loss": 0.9703,
2761
+ "step": 384
2762
+ },
2763
+ {
2764
+ "epoch": 1.91,
2765
+ "grad_norm": 0.8639506697654724,
2766
+ "learning_rate": 0.00010929797195363259,
2767
+ "loss": 0.8579,
2768
+ "step": 385
2769
+ },
2770
+ {
2771
+ "epoch": 1.92,
2772
+ "grad_norm": 1.001845121383667,
2773
+ "learning_rate": 0.0001088948062291783,
2774
+ "loss": 1.038,
2775
+ "step": 386
2776
+ },
2777
+ {
2778
+ "epoch": 1.92,
2779
+ "grad_norm": 0.9668776392936707,
2780
+ "learning_rate": 0.00010849149471993882,
2781
+ "loss": 0.9457,
2782
+ "step": 387
2783
+ },
2784
+ {
2785
+ "epoch": 1.93,
2786
+ "grad_norm": 0.8607358932495117,
2787
+ "learning_rate": 0.00010808804403614043,
2788
+ "loss": 0.8795,
2789
+ "step": 388
2790
+ },
2791
+ {
2792
+ "epoch": 1.93,
2793
+ "grad_norm": 1.0189685821533203,
2794
+ "learning_rate": 0.00010768446079029044,
2795
+ "loss": 0.9203,
2796
+ "step": 389
2797
+ },
2798
+ {
2799
+ "epoch": 1.94,
2800
+ "grad_norm": 0.9952776432037354,
2801
+ "learning_rate": 0.0001072807515970688,
2802
+ "loss": 1.0368,
2803
+ "step": 390
2804
+ },
2805
+ {
2806
+ "epoch": 1.94,
2807
+ "grad_norm": 1.057427167892456,
2808
+ "learning_rate": 0.00010687692307321984,
2809
+ "loss": 1.0568,
2810
+ "step": 391
2811
+ },
2812
+ {
2813
+ "epoch": 1.95,
2814
+ "grad_norm": 0.822589099407196,
2815
+ "learning_rate": 0.00010647298183744359,
2816
+ "loss": 0.9598,
2817
+ "step": 392
2818
+ },
2819
+ {
2820
+ "epoch": 1.95,
2821
+ "grad_norm": 0.9903733730316162,
2822
+ "learning_rate": 0.00010606893451028743,
2823
+ "loss": 1.0595,
2824
+ "step": 393
2825
+ },
2826
+ {
2827
+ "epoch": 1.96,
2828
+ "grad_norm": 1.0125857591629028,
2829
+ "learning_rate": 0.00010566478771403763,
2830
+ "loss": 0.9646,
2831
+ "step": 394
2832
+ },
2833
+ {
2834
+ "epoch": 1.96,
2835
+ "grad_norm": 0.899347722530365,
2836
+ "learning_rate": 0.00010526054807261067,
2837
+ "loss": 1.0054,
2838
+ "step": 395
2839
+ },
2840
+ {
2841
+ "epoch": 1.97,
2842
+ "grad_norm": 1.0629827976226807,
2843
+ "learning_rate": 0.00010485622221144484,
2844
+ "loss": 0.9319,
2845
+ "step": 396
2846
+ },
2847
+ {
2848
+ "epoch": 1.97,
2849
+ "grad_norm": 0.9910023212432861,
2850
+ "learning_rate": 0.00010445181675739144,
2851
+ "loss": 0.9388,
2852
+ "step": 397
2853
+ },
2854
+ {
2855
+ "epoch": 1.98,
2856
+ "grad_norm": 0.8644474744796753,
2857
+ "learning_rate": 0.00010404733833860639,
2858
+ "loss": 0.8007,
2859
+ "step": 398
2860
+ }
2861
+ ],
2862
+ "logging_steps": 1,
2863
+ "max_steps": 796,
2864
+ "num_input_tokens_seen": 0,
2865
+ "num_train_epochs": 4,
2866
+ "save_steps": 199,
2867
+ "total_flos": 1.531569014464512e+16,
2868
+ "train_batch_size": 2,
2869
+ "trial_name": null,
2870
+ "trial_params": null
2871
+ }
checkpoint-398/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d84e26ce27315b618f94e914bb6b67f0bb5aa37c3903b14adcd26c9fca9f3f82
3
+ size 5624
checkpoint-431/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: openlm-research/open_llama_3b_v2
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.9.0
checkpoint-431/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openlm-research/open_llama_3b_v2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 32,
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": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "q_proj",
23
+ "v_proj",
24
+ "o_proj",
25
+ "up_proj",
26
+ "down_proj",
27
+ "gate_proj",
28
+ "k_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
checkpoint-431/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33c327d1f8c2e6bfc45c4c172b410eddab59af41c4d7f45ed14020b5bd6d9c0f
3
+ size 50899792
checkpoint-431/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76dbbb915b03d456a516c3d532d7833adc4c05e4b9f29e9a2d1c46ff2a6222b7
3
+ size 101919290
checkpoint-431/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e766413e4f3b39c1d0ac620807b3bd3fd4dac79e0a0eed6a4a60c5746642e0a6
3
+ size 14244
checkpoint-431/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23f73a16ac262980457d80b6b9e4834ebbf9e3ee06ac2280222318fcdf9e15a8
3
+ size 1064
checkpoint-431/trainer_state.json ADDED
@@ -0,0 +1,3070 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 108,
6
+ "global_step": 431,
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": 0.8690947890281677,
14
+ "learning_rate": 1e-05,
15
+ "loss": 1.2567,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.0,
20
+ "eval_loss": 1.3469510078430176,
21
+ "eval_runtime": 4.9827,
22
+ "eval_samples_per_second": 20.069,
23
+ "eval_steps_per_second": 20.069,
24
+ "step": 1
25
+ },
26
+ {
27
+ "epoch": 0.0,
28
+ "grad_norm": 0.781764566898346,
29
+ "learning_rate": 2e-05,
30
+ "loss": 1.3328,
31
+ "step": 2
32
+ },
33
+ {
34
+ "epoch": 0.01,
35
+ "grad_norm": 1.0551249980926514,
36
+ "learning_rate": 3e-05,
37
+ "loss": 1.6572,
38
+ "step": 3
39
+ },
40
+ {
41
+ "epoch": 0.01,
42
+ "grad_norm": 1.39896821975708,
43
+ "learning_rate": 4e-05,
44
+ "loss": 1.5495,
45
+ "step": 4
46
+ },
47
+ {
48
+ "epoch": 0.01,
49
+ "grad_norm": 0.6837481260299683,
50
+ "learning_rate": 5e-05,
51
+ "loss": 1.4339,
52
+ "step": 5
53
+ },
54
+ {
55
+ "epoch": 0.01,
56
+ "grad_norm": 1.1228013038635254,
57
+ "learning_rate": 6e-05,
58
+ "loss": 1.2666,
59
+ "step": 6
60
+ },
61
+ {
62
+ "epoch": 0.02,
63
+ "grad_norm": 0.9632679224014282,
64
+ "learning_rate": 7e-05,
65
+ "loss": 1.4717,
66
+ "step": 7
67
+ },
68
+ {
69
+ "epoch": 0.02,
70
+ "grad_norm": 0.8218845725059509,
71
+ "learning_rate": 8e-05,
72
+ "loss": 1.4052,
73
+ "step": 8
74
+ },
75
+ {
76
+ "epoch": 0.02,
77
+ "grad_norm": 0.8262380361557007,
78
+ "learning_rate": 9e-05,
79
+ "loss": 1.2844,
80
+ "step": 9
81
+ },
82
+ {
83
+ "epoch": 0.02,
84
+ "grad_norm": 0.7251008749008179,
85
+ "learning_rate": 0.0001,
86
+ "loss": 1.1436,
87
+ "step": 10
88
+ },
89
+ {
90
+ "epoch": 0.03,
91
+ "grad_norm": 0.8107339143753052,
92
+ "learning_rate": 0.00011000000000000002,
93
+ "loss": 1.4666,
94
+ "step": 11
95
+ },
96
+ {
97
+ "epoch": 0.03,
98
+ "grad_norm": 0.8182777166366577,
99
+ "learning_rate": 0.00012,
100
+ "loss": 0.9755,
101
+ "step": 12
102
+ },
103
+ {
104
+ "epoch": 0.03,
105
+ "grad_norm": 1.3238070011138916,
106
+ "learning_rate": 0.00013000000000000002,
107
+ "loss": 1.5148,
108
+ "step": 13
109
+ },
110
+ {
111
+ "epoch": 0.03,
112
+ "grad_norm": 1.0324606895446777,
113
+ "learning_rate": 0.00014,
114
+ "loss": 1.1514,
115
+ "step": 14
116
+ },
117
+ {
118
+ "epoch": 0.03,
119
+ "grad_norm": 2.189394950866699,
120
+ "learning_rate": 0.00015000000000000001,
121
+ "loss": 1.2768,
122
+ "step": 15
123
+ },
124
+ {
125
+ "epoch": 0.04,
126
+ "grad_norm": 1.4134596586227417,
127
+ "learning_rate": 0.00016,
128
+ "loss": 1.3283,
129
+ "step": 16
130
+ },
131
+ {
132
+ "epoch": 0.04,
133
+ "grad_norm": 1.0440051555633545,
134
+ "learning_rate": 0.00017,
135
+ "loss": 1.297,
136
+ "step": 17
137
+ },
138
+ {
139
+ "epoch": 0.04,
140
+ "grad_norm": 1.0499480962753296,
141
+ "learning_rate": 0.00018,
142
+ "loss": 1.321,
143
+ "step": 18
144
+ },
145
+ {
146
+ "epoch": 0.04,
147
+ "grad_norm": 0.8342623710632324,
148
+ "learning_rate": 0.00019,
149
+ "loss": 1.2034,
150
+ "step": 19
151
+ },
152
+ {
153
+ "epoch": 0.05,
154
+ "grad_norm": 1.6799418926239014,
155
+ "learning_rate": 0.0002,
156
+ "loss": 1.1958,
157
+ "step": 20
158
+ },
159
+ {
160
+ "epoch": 0.05,
161
+ "grad_norm": 1.3825242519378662,
162
+ "learning_rate": 0.00019999707864731247,
163
+ "loss": 1.2299,
164
+ "step": 21
165
+ },
166
+ {
167
+ "epoch": 0.05,
168
+ "grad_norm": 1.3574899435043335,
169
+ "learning_rate": 0.00019998831475993593,
170
+ "loss": 1.3222,
171
+ "step": 22
172
+ },
173
+ {
174
+ "epoch": 0.05,
175
+ "grad_norm": 1.4586331844329834,
176
+ "learning_rate": 0.00019997370884991842,
177
+ "loss": 1.5933,
178
+ "step": 23
179
+ },
180
+ {
181
+ "epoch": 0.06,
182
+ "grad_norm": 1.102039098739624,
183
+ "learning_rate": 0.0001999532617706403,
184
+ "loss": 1.3109,
185
+ "step": 24
186
+ },
187
+ {
188
+ "epoch": 0.06,
189
+ "grad_norm": 0.9768686890602112,
190
+ "learning_rate": 0.00019992697471676413,
191
+ "loss": 1.3154,
192
+ "step": 25
193
+ },
194
+ {
195
+ "epoch": 0.06,
196
+ "grad_norm": 1.053282380104065,
197
+ "learning_rate": 0.00019989484922416502,
198
+ "loss": 1.2525,
199
+ "step": 26
200
+ },
201
+ {
202
+ "epoch": 0.06,
203
+ "grad_norm": 0.9527302980422974,
204
+ "learning_rate": 0.0001998568871698409,
205
+ "loss": 1.1981,
206
+ "step": 27
207
+ },
208
+ {
209
+ "epoch": 0.06,
210
+ "grad_norm": 0.9766901135444641,
211
+ "learning_rate": 0.00019981309077180272,
212
+ "loss": 1.0464,
213
+ "step": 28
214
+ },
215
+ {
216
+ "epoch": 0.07,
217
+ "grad_norm": 0.9839689135551453,
218
+ "learning_rate": 0.00019976346258894503,
219
+ "loss": 1.3556,
220
+ "step": 29
221
+ },
222
+ {
223
+ "epoch": 0.07,
224
+ "grad_norm": 1.3148291110992432,
225
+ "learning_rate": 0.00019970800552089623,
226
+ "loss": 1.2592,
227
+ "step": 30
228
+ },
229
+ {
230
+ "epoch": 0.07,
231
+ "grad_norm": 0.9205895066261292,
232
+ "learning_rate": 0.00019964672280784954,
233
+ "loss": 1.1584,
234
+ "step": 31
235
+ },
236
+ {
237
+ "epoch": 0.07,
238
+ "grad_norm": 1.2833977937698364,
239
+ "learning_rate": 0.00019957961803037326,
240
+ "loss": 1.1818,
241
+ "step": 32
242
+ },
243
+ {
244
+ "epoch": 0.08,
245
+ "grad_norm": 1.0097846984863281,
246
+ "learning_rate": 0.00019950669510920184,
247
+ "loss": 1.006,
248
+ "step": 33
249
+ },
250
+ {
251
+ "epoch": 0.08,
252
+ "grad_norm": 0.9942007660865784,
253
+ "learning_rate": 0.0001994279583050067,
254
+ "loss": 1.1939,
255
+ "step": 34
256
+ },
257
+ {
258
+ "epoch": 0.08,
259
+ "grad_norm": 1.3307287693023682,
260
+ "learning_rate": 0.00019934341221814739,
261
+ "loss": 1.5808,
262
+ "step": 35
263
+ },
264
+ {
265
+ "epoch": 0.08,
266
+ "grad_norm": 1.069790005683899,
267
+ "learning_rate": 0.0001992530617884026,
268
+ "loss": 0.9687,
269
+ "step": 36
270
+ },
271
+ {
272
+ "epoch": 0.09,
273
+ "grad_norm": 1.0054887533187866,
274
+ "learning_rate": 0.00019915691229468178,
275
+ "loss": 1.1882,
276
+ "step": 37
277
+ },
278
+ {
279
+ "epoch": 0.09,
280
+ "grad_norm": 1.6451635360717773,
281
+ "learning_rate": 0.00019905496935471658,
282
+ "loss": 1.7297,
283
+ "step": 38
284
+ },
285
+ {
286
+ "epoch": 0.09,
287
+ "grad_norm": 0.9728072881698608,
288
+ "learning_rate": 0.0001989472389247326,
289
+ "loss": 1.5091,
290
+ "step": 39
291
+ },
292
+ {
293
+ "epoch": 0.09,
294
+ "grad_norm": 1.2531479597091675,
295
+ "learning_rate": 0.00019883372729910152,
296
+ "loss": 1.1329,
297
+ "step": 40
298
+ },
299
+ {
300
+ "epoch": 0.1,
301
+ "grad_norm": 0.9028749465942383,
302
+ "learning_rate": 0.0001987144411099731,
303
+ "loss": 1.2428,
304
+ "step": 41
305
+ },
306
+ {
307
+ "epoch": 0.1,
308
+ "grad_norm": 0.9710382223129272,
309
+ "learning_rate": 0.000198589387326888,
310
+ "loss": 1.2192,
311
+ "step": 42
312
+ },
313
+ {
314
+ "epoch": 0.1,
315
+ "grad_norm": 0.8959193825721741,
316
+ "learning_rate": 0.00019845857325637031,
317
+ "loss": 1.5278,
318
+ "step": 43
319
+ },
320
+ {
321
+ "epoch": 0.1,
322
+ "grad_norm": 0.9432021379470825,
323
+ "learning_rate": 0.00019832200654150076,
324
+ "loss": 1.1128,
325
+ "step": 44
326
+ },
327
+ {
328
+ "epoch": 0.1,
329
+ "grad_norm": 0.8419432640075684,
330
+ "learning_rate": 0.0001981796951614701,
331
+ "loss": 1.0904,
332
+ "step": 45
333
+ },
334
+ {
335
+ "epoch": 0.11,
336
+ "grad_norm": 1.1038836240768433,
337
+ "learning_rate": 0.00019803164743111302,
338
+ "loss": 1.2347,
339
+ "step": 46
340
+ },
341
+ {
342
+ "epoch": 0.11,
343
+ "grad_norm": 1.2094361782073975,
344
+ "learning_rate": 0.00019787787200042223,
345
+ "loss": 1.3438,
346
+ "step": 47
347
+ },
348
+ {
349
+ "epoch": 0.11,
350
+ "grad_norm": 0.8724287748336792,
351
+ "learning_rate": 0.00019771837785404305,
352
+ "loss": 0.9631,
353
+ "step": 48
354
+ },
355
+ {
356
+ "epoch": 0.11,
357
+ "grad_norm": 1.0441787242889404,
358
+ "learning_rate": 0.00019755317431074859,
359
+ "loss": 1.3894,
360
+ "step": 49
361
+ },
362
+ {
363
+ "epoch": 0.12,
364
+ "grad_norm": 1.0916489362716675,
365
+ "learning_rate": 0.0001973822710228951,
366
+ "loss": 1.019,
367
+ "step": 50
368
+ },
369
+ {
370
+ "epoch": 0.12,
371
+ "grad_norm": 1.1637184619903564,
372
+ "learning_rate": 0.00019720567797585817,
373
+ "loss": 1.3044,
374
+ "step": 51
375
+ },
376
+ {
377
+ "epoch": 0.12,
378
+ "grad_norm": 1.023861289024353,
379
+ "learning_rate": 0.0001970234054874493,
380
+ "loss": 1.1711,
381
+ "step": 52
382
+ },
383
+ {
384
+ "epoch": 0.12,
385
+ "grad_norm": 1.2109915018081665,
386
+ "learning_rate": 0.0001968354642073129,
387
+ "loss": 1.2524,
388
+ "step": 53
389
+ },
390
+ {
391
+ "epoch": 0.13,
392
+ "grad_norm": 1.138853907585144,
393
+ "learning_rate": 0.00019664186511630433,
394
+ "loss": 0.8772,
395
+ "step": 54
396
+ },
397
+ {
398
+ "epoch": 0.13,
399
+ "grad_norm": 0.777930498123169,
400
+ "learning_rate": 0.000196442619525848,
401
+ "loss": 1.2304,
402
+ "step": 55
403
+ },
404
+ {
405
+ "epoch": 0.13,
406
+ "grad_norm": 1.1599185466766357,
407
+ "learning_rate": 0.00019623773907727682,
408
+ "loss": 1.4214,
409
+ "step": 56
410
+ },
411
+ {
412
+ "epoch": 0.13,
413
+ "grad_norm": 1.1862808465957642,
414
+ "learning_rate": 0.0001960272357411517,
415
+ "loss": 1.3264,
416
+ "step": 57
417
+ },
418
+ {
419
+ "epoch": 0.13,
420
+ "grad_norm": 0.8586333394050598,
421
+ "learning_rate": 0.0001958111218165624,
422
+ "loss": 1.1423,
423
+ "step": 58
424
+ },
425
+ {
426
+ "epoch": 0.14,
427
+ "grad_norm": 0.9254307746887207,
428
+ "learning_rate": 0.00019558940993040885,
429
+ "loss": 1.0662,
430
+ "step": 59
431
+ },
432
+ {
433
+ "epoch": 0.14,
434
+ "grad_norm": 0.8446744084358215,
435
+ "learning_rate": 0.00019536211303666323,
436
+ "loss": 1.3621,
437
+ "step": 60
438
+ },
439
+ {
440
+ "epoch": 0.14,
441
+ "grad_norm": 0.9066444635391235,
442
+ "learning_rate": 0.00019512924441561348,
443
+ "loss": 0.5954,
444
+ "step": 61
445
+ },
446
+ {
447
+ "epoch": 0.14,
448
+ "grad_norm": 0.8354930281639099,
449
+ "learning_rate": 0.00019489081767308698,
450
+ "loss": 1.2582,
451
+ "step": 62
452
+ },
453
+ {
454
+ "epoch": 0.15,
455
+ "grad_norm": 1.3045473098754883,
456
+ "learning_rate": 0.00019464684673965583,
457
+ "loss": 1.3778,
458
+ "step": 63
459
+ },
460
+ {
461
+ "epoch": 0.15,
462
+ "grad_norm": 0.7895586490631104,
463
+ "learning_rate": 0.0001943973458698229,
464
+ "loss": 0.9473,
465
+ "step": 64
466
+ },
467
+ {
468
+ "epoch": 0.15,
469
+ "grad_norm": 0.9904553294181824,
470
+ "learning_rate": 0.00019414232964118892,
471
+ "loss": 1.114,
472
+ "step": 65
473
+ },
474
+ {
475
+ "epoch": 0.15,
476
+ "grad_norm": 1.0157588720321655,
477
+ "learning_rate": 0.00019388181295360078,
478
+ "loss": 1.0409,
479
+ "step": 66
480
+ },
481
+ {
482
+ "epoch": 0.16,
483
+ "grad_norm": 1.1616543531417847,
484
+ "learning_rate": 0.00019361581102828095,
485
+ "loss": 1.0549,
486
+ "step": 67
487
+ },
488
+ {
489
+ "epoch": 0.16,
490
+ "grad_norm": 0.8383068442344666,
491
+ "learning_rate": 0.0001933443394069383,
492
+ "loss": 1.1849,
493
+ "step": 68
494
+ },
495
+ {
496
+ "epoch": 0.16,
497
+ "grad_norm": 0.9804612994194031,
498
+ "learning_rate": 0.00019306741395085976,
499
+ "loss": 1.1631,
500
+ "step": 69
501
+ },
502
+ {
503
+ "epoch": 0.16,
504
+ "grad_norm": 0.7936605215072632,
505
+ "learning_rate": 0.0001927850508399839,
506
+ "loss": 1.1084,
507
+ "step": 70
508
+ },
509
+ {
510
+ "epoch": 0.16,
511
+ "grad_norm": 0.963039755821228,
512
+ "learning_rate": 0.00019249726657195532,
513
+ "loss": 1.3065,
514
+ "step": 71
515
+ },
516
+ {
517
+ "epoch": 0.17,
518
+ "grad_norm": 0.9112350940704346,
519
+ "learning_rate": 0.00019220407796116098,
520
+ "loss": 1.3096,
521
+ "step": 72
522
+ },
523
+ {
524
+ "epoch": 0.17,
525
+ "grad_norm": 1.536594033241272,
526
+ "learning_rate": 0.00019190550213774756,
527
+ "loss": 1.1588,
528
+ "step": 73
529
+ },
530
+ {
531
+ "epoch": 0.17,
532
+ "grad_norm": 0.8441547751426697,
533
+ "learning_rate": 0.00019160155654662076,
534
+ "loss": 0.7042,
535
+ "step": 74
536
+ },
537
+ {
538
+ "epoch": 0.17,
539
+ "grad_norm": 1.0272456407546997,
540
+ "learning_rate": 0.00019129225894642593,
541
+ "loss": 1.0812,
542
+ "step": 75
543
+ },
544
+ {
545
+ "epoch": 0.18,
546
+ "grad_norm": 0.8152709603309631,
547
+ "learning_rate": 0.00019097762740851061,
548
+ "loss": 1.0435,
549
+ "step": 76
550
+ },
551
+ {
552
+ "epoch": 0.18,
553
+ "grad_norm": 1.0662580728530884,
554
+ "learning_rate": 0.0001906576803158686,
555
+ "loss": 1.1471,
556
+ "step": 77
557
+ },
558
+ {
559
+ "epoch": 0.18,
560
+ "grad_norm": 1.001435399055481,
561
+ "learning_rate": 0.0001903324363620659,
562
+ "loss": 1.0687,
563
+ "step": 78
564
+ },
565
+ {
566
+ "epoch": 0.18,
567
+ "grad_norm": 0.5956882834434509,
568
+ "learning_rate": 0.0001900019145501484,
569
+ "loss": 0.6305,
570
+ "step": 79
571
+ },
572
+ {
573
+ "epoch": 0.19,
574
+ "grad_norm": 0.8889302015304565,
575
+ "learning_rate": 0.0001896661341915318,
576
+ "loss": 1.4107,
577
+ "step": 80
578
+ },
579
+ {
580
+ "epoch": 0.19,
581
+ "grad_norm": 0.8901386260986328,
582
+ "learning_rate": 0.0001893251149048732,
583
+ "loss": 1.0325,
584
+ "step": 81
585
+ },
586
+ {
587
+ "epoch": 0.19,
588
+ "grad_norm": 0.9966225624084473,
589
+ "learning_rate": 0.00018897887661492474,
590
+ "loss": 1.121,
591
+ "step": 82
592
+ },
593
+ {
594
+ "epoch": 0.19,
595
+ "grad_norm": 1.066897988319397,
596
+ "learning_rate": 0.00018862743955136966,
597
+ "loss": 1.4332,
598
+ "step": 83
599
+ },
600
+ {
601
+ "epoch": 0.19,
602
+ "grad_norm": 0.7830447554588318,
603
+ "learning_rate": 0.0001882708242476401,
604
+ "loss": 1.0578,
605
+ "step": 84
606
+ },
607
+ {
608
+ "epoch": 0.2,
609
+ "grad_norm": 0.8779483437538147,
610
+ "learning_rate": 0.00018790905153971758,
611
+ "loss": 0.8687,
612
+ "step": 85
613
+ },
614
+ {
615
+ "epoch": 0.2,
616
+ "grad_norm": 0.8768894076347351,
617
+ "learning_rate": 0.00018754214256491562,
618
+ "loss": 1.0531,
619
+ "step": 86
620
+ },
621
+ {
622
+ "epoch": 0.2,
623
+ "grad_norm": 0.729383647441864,
624
+ "learning_rate": 0.00018717011876064453,
625
+ "loss": 1.0437,
626
+ "step": 87
627
+ },
628
+ {
629
+ "epoch": 0.2,
630
+ "grad_norm": 0.8910384178161621,
631
+ "learning_rate": 0.0001867930018631592,
632
+ "loss": 1.0108,
633
+ "step": 88
634
+ },
635
+ {
636
+ "epoch": 0.21,
637
+ "grad_norm": 0.8145522475242615,
638
+ "learning_rate": 0.00018641081390628877,
639
+ "loss": 1.1992,
640
+ "step": 89
641
+ },
642
+ {
643
+ "epoch": 0.21,
644
+ "grad_norm": 0.8675424456596375,
645
+ "learning_rate": 0.00018602357722014964,
646
+ "loss": 1.2933,
647
+ "step": 90
648
+ },
649
+ {
650
+ "epoch": 0.21,
651
+ "grad_norm": 0.7332017421722412,
652
+ "learning_rate": 0.00018563131442984044,
653
+ "loss": 1.0419,
654
+ "step": 91
655
+ },
656
+ {
657
+ "epoch": 0.21,
658
+ "grad_norm": 1.0502578020095825,
659
+ "learning_rate": 0.00018523404845412027,
660
+ "loss": 1.1473,
661
+ "step": 92
662
+ },
663
+ {
664
+ "epoch": 0.22,
665
+ "grad_norm": 1.497140645980835,
666
+ "learning_rate": 0.0001848318025040697,
667
+ "loss": 1.3815,
668
+ "step": 93
669
+ },
670
+ {
671
+ "epoch": 0.22,
672
+ "grad_norm": 0.7718933820724487,
673
+ "learning_rate": 0.00018442460008173445,
674
+ "loss": 1.0141,
675
+ "step": 94
676
+ },
677
+ {
678
+ "epoch": 0.22,
679
+ "grad_norm": 0.8856745958328247,
680
+ "learning_rate": 0.0001840124649787524,
681
+ "loss": 1.0661,
682
+ "step": 95
683
+ },
684
+ {
685
+ "epoch": 0.22,
686
+ "grad_norm": 1.0631150007247925,
687
+ "learning_rate": 0.0001835954212749632,
688
+ "loss": 1.1029,
689
+ "step": 96
690
+ },
691
+ {
692
+ "epoch": 0.23,
693
+ "grad_norm": 0.7661423683166504,
694
+ "learning_rate": 0.0001831734933370019,
695
+ "loss": 0.8733,
696
+ "step": 97
697
+ },
698
+ {
699
+ "epoch": 0.23,
700
+ "grad_norm": 0.8283933997154236,
701
+ "learning_rate": 0.0001827467058168748,
702
+ "loss": 0.885,
703
+ "step": 98
704
+ },
705
+ {
706
+ "epoch": 0.23,
707
+ "grad_norm": 1.1238280534744263,
708
+ "learning_rate": 0.00018231508365051922,
709
+ "loss": 1.315,
710
+ "step": 99
711
+ },
712
+ {
713
+ "epoch": 0.23,
714
+ "grad_norm": 0.8286343812942505,
715
+ "learning_rate": 0.0001818786520563467,
716
+ "loss": 1.3218,
717
+ "step": 100
718
+ },
719
+ {
720
+ "epoch": 0.23,
721
+ "grad_norm": 1.028365969657898,
722
+ "learning_rate": 0.00018143743653376942,
723
+ "loss": 1.0926,
724
+ "step": 101
725
+ },
726
+ {
727
+ "epoch": 0.24,
728
+ "grad_norm": 1.0841856002807617,
729
+ "learning_rate": 0.0001809914628617105,
730
+ "loss": 1.5602,
731
+ "step": 102
732
+ },
733
+ {
734
+ "epoch": 0.24,
735
+ "grad_norm": 0.7237169146537781,
736
+ "learning_rate": 0.00018054075709709756,
737
+ "loss": 1.1558,
738
+ "step": 103
739
+ },
740
+ {
741
+ "epoch": 0.24,
742
+ "grad_norm": 1.2722384929656982,
743
+ "learning_rate": 0.00018008534557334064,
744
+ "loss": 1.289,
745
+ "step": 104
746
+ },
747
+ {
748
+ "epoch": 0.24,
749
+ "grad_norm": 0.8757687211036682,
750
+ "learning_rate": 0.00017962525489879325,
751
+ "loss": 1.2181,
752
+ "step": 105
753
+ },
754
+ {
755
+ "epoch": 0.25,
756
+ "grad_norm": 0.9850521683692932,
757
+ "learning_rate": 0.00017916051195519797,
758
+ "loss": 0.9446,
759
+ "step": 106
760
+ },
761
+ {
762
+ "epoch": 0.25,
763
+ "grad_norm": 0.8803001642227173,
764
+ "learning_rate": 0.00017869114389611575,
765
+ "loss": 0.9984,
766
+ "step": 107
767
+ },
768
+ {
769
+ "epoch": 0.25,
770
+ "grad_norm": 0.8752118945121765,
771
+ "learning_rate": 0.0001782171781453394,
772
+ "loss": 1.1776,
773
+ "step": 108
774
+ },
775
+ {
776
+ "epoch": 0.25,
777
+ "eval_loss": 1.1369682550430298,
778
+ "eval_runtime": 5.0569,
779
+ "eval_samples_per_second": 19.775,
780
+ "eval_steps_per_second": 19.775,
781
+ "step": 108
782
+ },
783
+ {
784
+ "epoch": 0.25,
785
+ "grad_norm": 0.8919423222541809,
786
+ "learning_rate": 0.00017773864239529132,
787
+ "loss": 1.0642,
788
+ "step": 109
789
+ },
790
+ {
791
+ "epoch": 0.26,
792
+ "grad_norm": 0.8584320545196533,
793
+ "learning_rate": 0.0001772555646054055,
794
+ "loss": 1.0944,
795
+ "step": 110
796
+ },
797
+ {
798
+ "epoch": 0.26,
799
+ "grad_norm": 0.8071363568305969,
800
+ "learning_rate": 0.00017676797300049393,
801
+ "loss": 1.231,
802
+ "step": 111
803
+ },
804
+ {
805
+ "epoch": 0.26,
806
+ "grad_norm": 1.2383997440338135,
807
+ "learning_rate": 0.00017627589606909755,
808
+ "loss": 1.2194,
809
+ "step": 112
810
+ },
811
+ {
812
+ "epoch": 0.26,
813
+ "grad_norm": 0.557596743106842,
814
+ "learning_rate": 0.00017577936256182167,
815
+ "loss": 0.4546,
816
+ "step": 113
817
+ },
818
+ {
819
+ "epoch": 0.26,
820
+ "grad_norm": 0.9943146109580994,
821
+ "learning_rate": 0.0001752784014896562,
822
+ "loss": 1.3126,
823
+ "step": 114
824
+ },
825
+ {
826
+ "epoch": 0.27,
827
+ "grad_norm": 0.7767719626426697,
828
+ "learning_rate": 0.00017477304212228057,
829
+ "loss": 1.0385,
830
+ "step": 115
831
+ },
832
+ {
833
+ "epoch": 0.27,
834
+ "grad_norm": 0.849905788898468,
835
+ "learning_rate": 0.0001742633139863538,
836
+ "loss": 1.1383,
837
+ "step": 116
838
+ },
839
+ {
840
+ "epoch": 0.27,
841
+ "grad_norm": 1.038552165031433,
842
+ "learning_rate": 0.00017374924686378905,
843
+ "loss": 1.2284,
844
+ "step": 117
845
+ },
846
+ {
847
+ "epoch": 0.27,
848
+ "grad_norm": 1.1603586673736572,
849
+ "learning_rate": 0.0001732308707900137,
850
+ "loss": 1.1579,
851
+ "step": 118
852
+ },
853
+ {
854
+ "epoch": 0.28,
855
+ "grad_norm": 0.8285987973213196,
856
+ "learning_rate": 0.0001727082160522145,
857
+ "loss": 1.1173,
858
+ "step": 119
859
+ },
860
+ {
861
+ "epoch": 0.28,
862
+ "grad_norm": 0.8592681288719177,
863
+ "learning_rate": 0.0001721813131875679,
864
+ "loss": 1.0445,
865
+ "step": 120
866
+ },
867
+ {
868
+ "epoch": 0.28,
869
+ "grad_norm": 0.9642465710639954,
870
+ "learning_rate": 0.00017165019298145585,
871
+ "loss": 0.7456,
872
+ "step": 121
873
+ },
874
+ {
875
+ "epoch": 0.28,
876
+ "grad_norm": 0.8067343235015869,
877
+ "learning_rate": 0.00017111488646566727,
878
+ "loss": 1.1104,
879
+ "step": 122
880
+ },
881
+ {
882
+ "epoch": 0.29,
883
+ "grad_norm": 0.9156467914581299,
884
+ "learning_rate": 0.00017057542491658468,
885
+ "loss": 1.321,
886
+ "step": 123
887
+ },
888
+ {
889
+ "epoch": 0.29,
890
+ "grad_norm": 0.9722065329551697,
891
+ "learning_rate": 0.000170031839853357,
892
+ "loss": 1.3756,
893
+ "step": 124
894
+ },
895
+ {
896
+ "epoch": 0.29,
897
+ "grad_norm": 0.9338364005088806,
898
+ "learning_rate": 0.00016948416303605795,
899
+ "loss": 1.4055,
900
+ "step": 125
901
+ },
902
+ {
903
+ "epoch": 0.29,
904
+ "grad_norm": 0.7224316000938416,
905
+ "learning_rate": 0.0001689324264638304,
906
+ "loss": 0.8181,
907
+ "step": 126
908
+ },
909
+ {
910
+ "epoch": 0.29,
911
+ "grad_norm": 0.6058412194252014,
912
+ "learning_rate": 0.00016837666237301663,
913
+ "loss": 0.5958,
914
+ "step": 127
915
+ },
916
+ {
917
+ "epoch": 0.3,
918
+ "grad_norm": 0.9860162138938904,
919
+ "learning_rate": 0.00016781690323527511,
920
+ "loss": 1.1325,
921
+ "step": 128
922
+ },
923
+ {
924
+ "epoch": 0.3,
925
+ "grad_norm": 0.8655012845993042,
926
+ "learning_rate": 0.00016725318175568306,
927
+ "loss": 1.0151,
928
+ "step": 129
929
+ },
930
+ {
931
+ "epoch": 0.3,
932
+ "grad_norm": 0.7968864440917969,
933
+ "learning_rate": 0.00016668553087082567,
934
+ "loss": 1.288,
935
+ "step": 130
936
+ },
937
+ {
938
+ "epoch": 0.3,
939
+ "grad_norm": 0.8813712000846863,
940
+ "learning_rate": 0.0001661139837468717,
941
+ "loss": 1.3115,
942
+ "step": 131
943
+ },
944
+ {
945
+ "epoch": 0.31,
946
+ "grad_norm": 0.7319151163101196,
947
+ "learning_rate": 0.00016553857377763566,
948
+ "loss": 1.0332,
949
+ "step": 132
950
+ },
951
+ {
952
+ "epoch": 0.31,
953
+ "grad_norm": 1.1282501220703125,
954
+ "learning_rate": 0.0001649593345826268,
955
+ "loss": 1.2969,
956
+ "step": 133
957
+ },
958
+ {
959
+ "epoch": 0.31,
960
+ "grad_norm": 0.799923837184906,
961
+ "learning_rate": 0.00016437630000508464,
962
+ "loss": 0.9233,
963
+ "step": 134
964
+ },
965
+ {
966
+ "epoch": 0.31,
967
+ "grad_norm": 1.0394463539123535,
968
+ "learning_rate": 0.00016378950411000183,
969
+ "loss": 1.2451,
970
+ "step": 135
971
+ },
972
+ {
973
+ "epoch": 0.32,
974
+ "grad_norm": 1.032335877418518,
975
+ "learning_rate": 0.00016319898118213365,
976
+ "loss": 1.0193,
977
+ "step": 136
978
+ },
979
+ {
980
+ "epoch": 0.32,
981
+ "grad_norm": 1.0510966777801514,
982
+ "learning_rate": 0.00016260476572399496,
983
+ "loss": 1.4264,
984
+ "step": 137
985
+ },
986
+ {
987
+ "epoch": 0.32,
988
+ "grad_norm": 0.8531592488288879,
989
+ "learning_rate": 0.00016200689245384424,
990
+ "loss": 1.0608,
991
+ "step": 138
992
+ },
993
+ {
994
+ "epoch": 0.32,
995
+ "grad_norm": 1.2909444570541382,
996
+ "learning_rate": 0.00016140539630365522,
997
+ "loss": 1.116,
998
+ "step": 139
999
+ },
1000
+ {
1001
+ "epoch": 0.32,
1002
+ "grad_norm": 0.6848735213279724,
1003
+ "learning_rate": 0.00016080031241707578,
1004
+ "loss": 0.9243,
1005
+ "step": 140
1006
+ },
1007
+ {
1008
+ "epoch": 0.33,
1009
+ "grad_norm": 0.8676778674125671,
1010
+ "learning_rate": 0.0001601916761473747,
1011
+ "loss": 1.1039,
1012
+ "step": 141
1013
+ },
1014
+ {
1015
+ "epoch": 0.33,
1016
+ "grad_norm": 1.1020722389221191,
1017
+ "learning_rate": 0.00015957952305537597,
1018
+ "loss": 1.1554,
1019
+ "step": 142
1020
+ },
1021
+ {
1022
+ "epoch": 0.33,
1023
+ "grad_norm": 0.883198082447052,
1024
+ "learning_rate": 0.00015896388890738127,
1025
+ "loss": 1.2192,
1026
+ "step": 143
1027
+ },
1028
+ {
1029
+ "epoch": 0.33,
1030
+ "grad_norm": 0.8749895691871643,
1031
+ "learning_rate": 0.00015834480967308003,
1032
+ "loss": 1.0946,
1033
+ "step": 144
1034
+ },
1035
+ {
1036
+ "epoch": 0.34,
1037
+ "grad_norm": 1.0936459302902222,
1038
+ "learning_rate": 0.00015772232152344795,
1039
+ "loss": 1.2226,
1040
+ "step": 145
1041
+ },
1042
+ {
1043
+ "epoch": 0.34,
1044
+ "grad_norm": 1.003266453742981,
1045
+ "learning_rate": 0.0001570964608286336,
1046
+ "loss": 1.0759,
1047
+ "step": 146
1048
+ },
1049
+ {
1050
+ "epoch": 0.34,
1051
+ "grad_norm": 0.9020757675170898,
1052
+ "learning_rate": 0.00015646726415583344,
1053
+ "loss": 0.8462,
1054
+ "step": 147
1055
+ },
1056
+ {
1057
+ "epoch": 0.34,
1058
+ "grad_norm": 0.9358006119728088,
1059
+ "learning_rate": 0.0001558347682671553,
1060
+ "loss": 1.4268,
1061
+ "step": 148
1062
+ },
1063
+ {
1064
+ "epoch": 0.35,
1065
+ "grad_norm": 1.215381383895874,
1066
+ "learning_rate": 0.00015519901011747044,
1067
+ "loss": 1.1037,
1068
+ "step": 149
1069
+ },
1070
+ {
1071
+ "epoch": 0.35,
1072
+ "grad_norm": 1.1152362823486328,
1073
+ "learning_rate": 0.00015456002685225448,
1074
+ "loss": 1.3482,
1075
+ "step": 150
1076
+ },
1077
+ {
1078
+ "epoch": 0.35,
1079
+ "grad_norm": 1.1836152076721191,
1080
+ "learning_rate": 0.00015391785580541698,
1081
+ "loss": 1.3423,
1082
+ "step": 151
1083
+ },
1084
+ {
1085
+ "epoch": 0.35,
1086
+ "grad_norm": 0.8727456331253052,
1087
+ "learning_rate": 0.0001532725344971202,
1088
+ "loss": 1.6155,
1089
+ "step": 152
1090
+ },
1091
+ {
1092
+ "epoch": 0.35,
1093
+ "grad_norm": 1.0771983861923218,
1094
+ "learning_rate": 0.0001526241006315869,
1095
+ "loss": 1.4055,
1096
+ "step": 153
1097
+ },
1098
+ {
1099
+ "epoch": 0.36,
1100
+ "grad_norm": 0.7963141798973083,
1101
+ "learning_rate": 0.00015197259209489747,
1102
+ "loss": 1.407,
1103
+ "step": 154
1104
+ },
1105
+ {
1106
+ "epoch": 0.36,
1107
+ "grad_norm": 0.9456453323364258,
1108
+ "learning_rate": 0.00015131804695277612,
1109
+ "loss": 1.1586,
1110
+ "step": 155
1111
+ },
1112
+ {
1113
+ "epoch": 0.36,
1114
+ "grad_norm": 0.9810377359390259,
1115
+ "learning_rate": 0.00015066050344836706,
1116
+ "loss": 1.1881,
1117
+ "step": 156
1118
+ },
1119
+ {
1120
+ "epoch": 0.36,
1121
+ "grad_norm": 0.8879066109657288,
1122
+ "learning_rate": 0.00015000000000000001,
1123
+ "loss": 1.2283,
1124
+ "step": 157
1125
+ },
1126
+ {
1127
+ "epoch": 0.37,
1128
+ "grad_norm": 0.6425898671150208,
1129
+ "learning_rate": 0.0001493365751989454,
1130
+ "loss": 0.8447,
1131
+ "step": 158
1132
+ },
1133
+ {
1134
+ "epoch": 0.37,
1135
+ "grad_norm": 0.7046767473220825,
1136
+ "learning_rate": 0.0001486702678071598,
1137
+ "loss": 0.8164,
1138
+ "step": 159
1139
+ },
1140
+ {
1141
+ "epoch": 0.37,
1142
+ "grad_norm": 0.7500093579292297,
1143
+ "learning_rate": 0.00014800111675502094,
1144
+ "loss": 0.9141,
1145
+ "step": 160
1146
+ },
1147
+ {
1148
+ "epoch": 0.37,
1149
+ "grad_norm": 0.8106879591941833,
1150
+ "learning_rate": 0.00014732916113905335,
1151
+ "loss": 1.1849,
1152
+ "step": 161
1153
+ },
1154
+ {
1155
+ "epoch": 0.38,
1156
+ "grad_norm": 1.1407498121261597,
1157
+ "learning_rate": 0.0001466544402196439,
1158
+ "loss": 1.0616,
1159
+ "step": 162
1160
+ },
1161
+ {
1162
+ "epoch": 0.38,
1163
+ "grad_norm": 0.8494256138801575,
1164
+ "learning_rate": 0.00014597699341874806,
1165
+ "loss": 0.7441,
1166
+ "step": 163
1167
+ },
1168
+ {
1169
+ "epoch": 0.38,
1170
+ "grad_norm": 0.8430727124214172,
1171
+ "learning_rate": 0.00014529686031758643,
1172
+ "loss": 0.8714,
1173
+ "step": 164
1174
+ },
1175
+ {
1176
+ "epoch": 0.38,
1177
+ "grad_norm": 1.0597652196884155,
1178
+ "learning_rate": 0.00014461408065433227,
1179
+ "loss": 1.372,
1180
+ "step": 165
1181
+ },
1182
+ {
1183
+ "epoch": 0.39,
1184
+ "grad_norm": 0.7789810299873352,
1185
+ "learning_rate": 0.00014392869432178971,
1186
+ "loss": 1.0039,
1187
+ "step": 166
1188
+ },
1189
+ {
1190
+ "epoch": 0.39,
1191
+ "grad_norm": 0.7892749905586243,
1192
+ "learning_rate": 0.00014324074136506284,
1193
+ "loss": 1.1196,
1194
+ "step": 167
1195
+ },
1196
+ {
1197
+ "epoch": 0.39,
1198
+ "grad_norm": 1.3845864534378052,
1199
+ "learning_rate": 0.00014255026197921596,
1200
+ "loss": 1.0634,
1201
+ "step": 168
1202
+ },
1203
+ {
1204
+ "epoch": 0.39,
1205
+ "grad_norm": 0.9081052541732788,
1206
+ "learning_rate": 0.00014185729650692533,
1207
+ "loss": 1.2649,
1208
+ "step": 169
1209
+ },
1210
+ {
1211
+ "epoch": 0.39,
1212
+ "grad_norm": 0.9540420174598694,
1213
+ "learning_rate": 0.0001411618854361218,
1214
+ "loss": 1.3414,
1215
+ "step": 170
1216
+ },
1217
+ {
1218
+ "epoch": 0.4,
1219
+ "grad_norm": 1.3803868293762207,
1220
+ "learning_rate": 0.00014046406939762545,
1221
+ "loss": 1.1241,
1222
+ "step": 171
1223
+ },
1224
+ {
1225
+ "epoch": 0.4,
1226
+ "grad_norm": 1.0107030868530273,
1227
+ "learning_rate": 0.0001397638891627714,
1228
+ "loss": 1.2018,
1229
+ "step": 172
1230
+ },
1231
+ {
1232
+ "epoch": 0.4,
1233
+ "grad_norm": 1.2234842777252197,
1234
+ "learning_rate": 0.00013906138564102793,
1235
+ "loss": 1.2857,
1236
+ "step": 173
1237
+ },
1238
+ {
1239
+ "epoch": 0.4,
1240
+ "grad_norm": 1.0201656818389893,
1241
+ "learning_rate": 0.00013835659987760605,
1242
+ "loss": 1.306,
1243
+ "step": 174
1244
+ },
1245
+ {
1246
+ "epoch": 0.41,
1247
+ "grad_norm": 0.9351068735122681,
1248
+ "learning_rate": 0.0001376495730510614,
1249
+ "loss": 1.3483,
1250
+ "step": 175
1251
+ },
1252
+ {
1253
+ "epoch": 0.41,
1254
+ "grad_norm": 1.046683430671692,
1255
+ "learning_rate": 0.0001369403464708884,
1256
+ "loss": 1.2294,
1257
+ "step": 176
1258
+ },
1259
+ {
1260
+ "epoch": 0.41,
1261
+ "grad_norm": 1.1101959943771362,
1262
+ "learning_rate": 0.00013622896157510658,
1263
+ "loss": 1.3065,
1264
+ "step": 177
1265
+ },
1266
+ {
1267
+ "epoch": 0.41,
1268
+ "grad_norm": 0.7936177849769592,
1269
+ "learning_rate": 0.00013551545992783947,
1270
+ "loss": 0.8732,
1271
+ "step": 178
1272
+ },
1273
+ {
1274
+ "epoch": 0.42,
1275
+ "grad_norm": 0.8901675939559937,
1276
+ "learning_rate": 0.0001347998832168862,
1277
+ "loss": 1.0109,
1278
+ "step": 179
1279
+ },
1280
+ {
1281
+ "epoch": 0.42,
1282
+ "grad_norm": 1.0607932806015015,
1283
+ "learning_rate": 0.0001340822732512857,
1284
+ "loss": 1.459,
1285
+ "step": 180
1286
+ },
1287
+ {
1288
+ "epoch": 0.42,
1289
+ "grad_norm": 0.9324591159820557,
1290
+ "learning_rate": 0.00013336267195887398,
1291
+ "loss": 1.1962,
1292
+ "step": 181
1293
+ },
1294
+ {
1295
+ "epoch": 0.42,
1296
+ "grad_norm": 0.8911722898483276,
1297
+ "learning_rate": 0.00013264112138383445,
1298
+ "loss": 0.9749,
1299
+ "step": 182
1300
+ },
1301
+ {
1302
+ "epoch": 0.42,
1303
+ "grad_norm": 0.7679628133773804,
1304
+ "learning_rate": 0.00013191766368424133,
1305
+ "loss": 1.1086,
1306
+ "step": 183
1307
+ },
1308
+ {
1309
+ "epoch": 0.43,
1310
+ "grad_norm": 1.0897005796432495,
1311
+ "learning_rate": 0.00013119234112959655,
1312
+ "loss": 0.9565,
1313
+ "step": 184
1314
+ },
1315
+ {
1316
+ "epoch": 0.43,
1317
+ "grad_norm": 0.969784677028656,
1318
+ "learning_rate": 0.00013046519609836,
1319
+ "loss": 1.3534,
1320
+ "step": 185
1321
+ },
1322
+ {
1323
+ "epoch": 0.43,
1324
+ "grad_norm": 0.7279093265533447,
1325
+ "learning_rate": 0.00012973627107547346,
1326
+ "loss": 1.261,
1327
+ "step": 186
1328
+ },
1329
+ {
1330
+ "epoch": 0.43,
1331
+ "grad_norm": 0.9844076037406921,
1332
+ "learning_rate": 0.0001290056086498785,
1333
+ "loss": 1.2582,
1334
+ "step": 187
1335
+ },
1336
+ {
1337
+ "epoch": 0.44,
1338
+ "grad_norm": 0.8580243587493896,
1339
+ "learning_rate": 0.00012827325151202782,
1340
+ "loss": 1.2099,
1341
+ "step": 188
1342
+ },
1343
+ {
1344
+ "epoch": 0.44,
1345
+ "grad_norm": 0.9180042147636414,
1346
+ "learning_rate": 0.00012753924245139135,
1347
+ "loss": 1.1773,
1348
+ "step": 189
1349
+ },
1350
+ {
1351
+ "epoch": 0.44,
1352
+ "grad_norm": 0.8368516564369202,
1353
+ "learning_rate": 0.00012680362435395595,
1354
+ "loss": 1.1123,
1355
+ "step": 190
1356
+ },
1357
+ {
1358
+ "epoch": 0.44,
1359
+ "grad_norm": 1.815647006034851,
1360
+ "learning_rate": 0.00012606644019971968,
1361
+ "loss": 1.4643,
1362
+ "step": 191
1363
+ },
1364
+ {
1365
+ "epoch": 0.45,
1366
+ "grad_norm": 0.8074012398719788,
1367
+ "learning_rate": 0.00012532773306018076,
1368
+ "loss": 1.1728,
1369
+ "step": 192
1370
+ },
1371
+ {
1372
+ "epoch": 0.45,
1373
+ "grad_norm": 1.0798611640930176,
1374
+ "learning_rate": 0.00012458754609582097,
1375
+ "loss": 1.003,
1376
+ "step": 193
1377
+ },
1378
+ {
1379
+ "epoch": 0.45,
1380
+ "grad_norm": 0.7999426126480103,
1381
+ "learning_rate": 0.00012384592255358385,
1382
+ "loss": 1.0413,
1383
+ "step": 194
1384
+ },
1385
+ {
1386
+ "epoch": 0.45,
1387
+ "grad_norm": 0.9997307062149048,
1388
+ "learning_rate": 0.00012310290576434795,
1389
+ "loss": 1.4449,
1390
+ "step": 195
1391
+ },
1392
+ {
1393
+ "epoch": 0.45,
1394
+ "grad_norm": 0.7481138706207275,
1395
+ "learning_rate": 0.00012235853914039515,
1396
+ "loss": 0.9215,
1397
+ "step": 196
1398
+ },
1399
+ {
1400
+ "epoch": 0.46,
1401
+ "grad_norm": 0.967931866645813,
1402
+ "learning_rate": 0.00012161286617287419,
1403
+ "loss": 0.944,
1404
+ "step": 197
1405
+ },
1406
+ {
1407
+ "epoch": 0.46,
1408
+ "grad_norm": 0.8793982863426208,
1409
+ "learning_rate": 0.00012086593042925964,
1410
+ "loss": 1.2297,
1411
+ "step": 198
1412
+ },
1413
+ {
1414
+ "epoch": 0.46,
1415
+ "grad_norm": 0.6650402545928955,
1416
+ "learning_rate": 0.00012011777555080638,
1417
+ "loss": 0.6958,
1418
+ "step": 199
1419
+ },
1420
+ {
1421
+ "epoch": 0.46,
1422
+ "grad_norm": 1.2048183679580688,
1423
+ "learning_rate": 0.00011936844524999966,
1424
+ "loss": 1.3445,
1425
+ "step": 200
1426
+ },
1427
+ {
1428
+ "epoch": 0.47,
1429
+ "grad_norm": 0.845542848110199,
1430
+ "learning_rate": 0.00011861798330800125,
1431
+ "loss": 1.0368,
1432
+ "step": 201
1433
+ },
1434
+ {
1435
+ "epoch": 0.47,
1436
+ "grad_norm": 1.0240477323532104,
1437
+ "learning_rate": 0.00011786643357209136,
1438
+ "loss": 0.9428,
1439
+ "step": 202
1440
+ },
1441
+ {
1442
+ "epoch": 0.47,
1443
+ "grad_norm": 1.2319940328598022,
1444
+ "learning_rate": 0.00011711383995310681,
1445
+ "loss": 1.117,
1446
+ "step": 203
1447
+ },
1448
+ {
1449
+ "epoch": 0.47,
1450
+ "grad_norm": 0.8866223692893982,
1451
+ "learning_rate": 0.00011636024642287546,
1452
+ "loss": 1.1306,
1453
+ "step": 204
1454
+ },
1455
+ {
1456
+ "epoch": 0.48,
1457
+ "grad_norm": 0.8940457701683044,
1458
+ "learning_rate": 0.00011560569701164697,
1459
+ "loss": 1.3186,
1460
+ "step": 205
1461
+ },
1462
+ {
1463
+ "epoch": 0.48,
1464
+ "grad_norm": 2.126744508743286,
1465
+ "learning_rate": 0.00011485023580552039,
1466
+ "loss": 1.4878,
1467
+ "step": 206
1468
+ },
1469
+ {
1470
+ "epoch": 0.48,
1471
+ "grad_norm": 0.8328797817230225,
1472
+ "learning_rate": 0.00011409390694386817,
1473
+ "loss": 1.179,
1474
+ "step": 207
1475
+ },
1476
+ {
1477
+ "epoch": 0.48,
1478
+ "grad_norm": 0.8258645534515381,
1479
+ "learning_rate": 0.00011333675461675739,
1480
+ "loss": 1.0724,
1481
+ "step": 208
1482
+ },
1483
+ {
1484
+ "epoch": 0.48,
1485
+ "grad_norm": 0.9328184723854065,
1486
+ "learning_rate": 0.00011257882306236775,
1487
+ "loss": 1.2991,
1488
+ "step": 209
1489
+ },
1490
+ {
1491
+ "epoch": 0.49,
1492
+ "grad_norm": 0.7689169645309448,
1493
+ "learning_rate": 0.00011182015656440692,
1494
+ "loss": 1.0181,
1495
+ "step": 210
1496
+ },
1497
+ {
1498
+ "epoch": 0.49,
1499
+ "grad_norm": 0.9105891585350037,
1500
+ "learning_rate": 0.00011106079944952317,
1501
+ "loss": 1.2888,
1502
+ "step": 211
1503
+ },
1504
+ {
1505
+ "epoch": 0.49,
1506
+ "grad_norm": 1.0148977041244507,
1507
+ "learning_rate": 0.00011030079608471544,
1508
+ "loss": 0.897,
1509
+ "step": 212
1510
+ },
1511
+ {
1512
+ "epoch": 0.49,
1513
+ "grad_norm": 1.0647752285003662,
1514
+ "learning_rate": 0.00010954019087474124,
1515
+ "loss": 1.4844,
1516
+ "step": 213
1517
+ },
1518
+ {
1519
+ "epoch": 0.5,
1520
+ "grad_norm": 0.8448371887207031,
1521
+ "learning_rate": 0.00010877902825952197,
1522
+ "loss": 1.0933,
1523
+ "step": 214
1524
+ },
1525
+ {
1526
+ "epoch": 0.5,
1527
+ "grad_norm": 0.9057207107543945,
1528
+ "learning_rate": 0.00010801735271154669,
1529
+ "loss": 1.1517,
1530
+ "step": 215
1531
+ },
1532
+ {
1533
+ "epoch": 0.5,
1534
+ "grad_norm": 0.9787666201591492,
1535
+ "learning_rate": 0.00010725520873327361,
1536
+ "loss": 1.1082,
1537
+ "step": 216
1538
+ },
1539
+ {
1540
+ "epoch": 0.5,
1541
+ "eval_loss": 1.1230801343917847,
1542
+ "eval_runtime": 5.7682,
1543
+ "eval_samples_per_second": 17.336,
1544
+ "eval_steps_per_second": 17.336,
1545
+ "step": 216
1546
+ },
1547
+ {
1548
+ "epoch": 0.5,
1549
+ "grad_norm": 0.8504732251167297,
1550
+ "learning_rate": 0.00010649264085452988,
1551
+ "loss": 0.9788,
1552
+ "step": 217
1553
+ },
1554
+ {
1555
+ "epoch": 0.51,
1556
+ "grad_norm": 1.1127643585205078,
1557
+ "learning_rate": 0.00010572969362990998,
1558
+ "loss": 1.2786,
1559
+ "step": 218
1560
+ },
1561
+ {
1562
+ "epoch": 0.51,
1563
+ "grad_norm": 0.9895543456077576,
1564
+ "learning_rate": 0.0001049664116361724,
1565
+ "loss": 1.1889,
1566
+ "step": 219
1567
+ },
1568
+ {
1569
+ "epoch": 0.51,
1570
+ "grad_norm": 1.016451358795166,
1571
+ "learning_rate": 0.0001042028394696352,
1572
+ "loss": 1.0312,
1573
+ "step": 220
1574
+ },
1575
+ {
1576
+ "epoch": 0.51,
1577
+ "grad_norm": 1.0872440338134766,
1578
+ "learning_rate": 0.00010343902174357039,
1579
+ "loss": 1.1398,
1580
+ "step": 221
1581
+ },
1582
+ {
1583
+ "epoch": 0.52,
1584
+ "grad_norm": 0.7803768515586853,
1585
+ "learning_rate": 0.00010267500308559732,
1586
+ "loss": 1.1324,
1587
+ "step": 222
1588
+ },
1589
+ {
1590
+ "epoch": 0.52,
1591
+ "grad_norm": 1.0301629304885864,
1592
+ "learning_rate": 0.0001019108281350752,
1593
+ "loss": 1.0781,
1594
+ "step": 223
1595
+ },
1596
+ {
1597
+ "epoch": 0.52,
1598
+ "grad_norm": 0.812462568283081,
1599
+ "learning_rate": 0.0001011465415404949,
1600
+ "loss": 1.2578,
1601
+ "step": 224
1602
+ },
1603
+ {
1604
+ "epoch": 0.52,
1605
+ "grad_norm": 0.8909915089607239,
1606
+ "learning_rate": 0.0001003821879568704,
1607
+ "loss": 0.9458,
1608
+ "step": 225
1609
+ },
1610
+ {
1611
+ "epoch": 0.52,
1612
+ "grad_norm": 0.9123939275741577,
1613
+ "learning_rate": 9.96178120431296e-05,
1614
+ "loss": 1.1767,
1615
+ "step": 226
1616
+ },
1617
+ {
1618
+ "epoch": 0.53,
1619
+ "grad_norm": 0.8767213821411133,
1620
+ "learning_rate": 9.88534584595051e-05,
1621
+ "loss": 1.0316,
1622
+ "step": 227
1623
+ },
1624
+ {
1625
+ "epoch": 0.53,
1626
+ "grad_norm": 0.7890278100967407,
1627
+ "learning_rate": 9.80891718649248e-05,
1628
+ "loss": 0.9678,
1629
+ "step": 228
1630
+ },
1631
+ {
1632
+ "epoch": 0.53,
1633
+ "grad_norm": 0.8505096435546875,
1634
+ "learning_rate": 9.732499691440266e-05,
1635
+ "loss": 1.3792,
1636
+ "step": 229
1637
+ },
1638
+ {
1639
+ "epoch": 0.53,
1640
+ "grad_norm": 0.7998389601707458,
1641
+ "learning_rate": 9.656097825642961e-05,
1642
+ "loss": 1.2242,
1643
+ "step": 230
1644
+ },
1645
+ {
1646
+ "epoch": 0.54,
1647
+ "grad_norm": 0.7925727367401123,
1648
+ "learning_rate": 9.579716053036479e-05,
1649
+ "loss": 0.9716,
1650
+ "step": 231
1651
+ },
1652
+ {
1653
+ "epoch": 0.54,
1654
+ "grad_norm": 0.8605330586433411,
1655
+ "learning_rate": 9.503358836382761e-05,
1656
+ "loss": 1.2368,
1657
+ "step": 232
1658
+ },
1659
+ {
1660
+ "epoch": 0.54,
1661
+ "grad_norm": 0.971171498298645,
1662
+ "learning_rate": 9.427030637009003e-05,
1663
+ "loss": 1.1605,
1664
+ "step": 233
1665
+ },
1666
+ {
1667
+ "epoch": 0.54,
1668
+ "grad_norm": 1.1337116956710815,
1669
+ "learning_rate": 9.35073591454701e-05,
1670
+ "loss": 0.8344,
1671
+ "step": 234
1672
+ },
1673
+ {
1674
+ "epoch": 0.55,
1675
+ "grad_norm": 0.8743841648101807,
1676
+ "learning_rate": 9.274479126672641e-05,
1677
+ "loss": 1.109,
1678
+ "step": 235
1679
+ },
1680
+ {
1681
+ "epoch": 0.55,
1682
+ "grad_norm": 0.9219100475311279,
1683
+ "learning_rate": 9.198264728845332e-05,
1684
+ "loss": 1.3502,
1685
+ "step": 236
1686
+ },
1687
+ {
1688
+ "epoch": 0.55,
1689
+ "grad_norm": 0.800612211227417,
1690
+ "learning_rate": 9.122097174047805e-05,
1691
+ "loss": 1.1531,
1692
+ "step": 237
1693
+ },
1694
+ {
1695
+ "epoch": 0.55,
1696
+ "grad_norm": 0.7432851791381836,
1697
+ "learning_rate": 9.045980912525879e-05,
1698
+ "loss": 0.9259,
1699
+ "step": 238
1700
+ },
1701
+ {
1702
+ "epoch": 0.55,
1703
+ "grad_norm": 0.7886478304862976,
1704
+ "learning_rate": 8.969920391528458e-05,
1705
+ "loss": 0.8643,
1706
+ "step": 239
1707
+ },
1708
+ {
1709
+ "epoch": 0.56,
1710
+ "grad_norm": 0.7922253012657166,
1711
+ "learning_rate": 8.893920055047686e-05,
1712
+ "loss": 1.0508,
1713
+ "step": 240
1714
+ },
1715
+ {
1716
+ "epoch": 0.56,
1717
+ "grad_norm": 1.112563967704773,
1718
+ "learning_rate": 8.81798434355931e-05,
1719
+ "loss": 1.1824,
1720
+ "step": 241
1721
+ },
1722
+ {
1723
+ "epoch": 0.56,
1724
+ "grad_norm": 0.7471961379051208,
1725
+ "learning_rate": 8.742117693763227e-05,
1726
+ "loss": 1.0301,
1727
+ "step": 242
1728
+ },
1729
+ {
1730
+ "epoch": 0.56,
1731
+ "grad_norm": 0.8147581815719604,
1732
+ "learning_rate": 8.666324538324264e-05,
1733
+ "loss": 1.0864,
1734
+ "step": 243
1735
+ },
1736
+ {
1737
+ "epoch": 0.57,
1738
+ "grad_norm": 0.8042265772819519,
1739
+ "learning_rate": 8.590609305613184e-05,
1740
+ "loss": 1.2239,
1741
+ "step": 244
1742
+ },
1743
+ {
1744
+ "epoch": 0.57,
1745
+ "grad_norm": 0.7160305976867676,
1746
+ "learning_rate": 8.514976419447964e-05,
1747
+ "loss": 1.1087,
1748
+ "step": 245
1749
+ },
1750
+ {
1751
+ "epoch": 0.57,
1752
+ "grad_norm": 0.7818716168403625,
1753
+ "learning_rate": 8.439430298835304e-05,
1754
+ "loss": 1.2434,
1755
+ "step": 246
1756
+ },
1757
+ {
1758
+ "epoch": 0.57,
1759
+ "grad_norm": 0.8190276622772217,
1760
+ "learning_rate": 8.363975357712457e-05,
1761
+ "loss": 1.2271,
1762
+ "step": 247
1763
+ },
1764
+ {
1765
+ "epoch": 0.58,
1766
+ "grad_norm": 0.8305711150169373,
1767
+ "learning_rate": 8.28861600468932e-05,
1768
+ "loss": 0.8105,
1769
+ "step": 248
1770
+ },
1771
+ {
1772
+ "epoch": 0.58,
1773
+ "grad_norm": 0.8715688586235046,
1774
+ "learning_rate": 8.213356642790867e-05,
1775
+ "loss": 0.8814,
1776
+ "step": 249
1777
+ },
1778
+ {
1779
+ "epoch": 0.58,
1780
+ "grad_norm": 0.8992697596549988,
1781
+ "learning_rate": 8.138201669199879e-05,
1782
+ "loss": 1.1118,
1783
+ "step": 250
1784
+ },
1785
+ {
1786
+ "epoch": 0.58,
1787
+ "grad_norm": 0.9572819471359253,
1788
+ "learning_rate": 8.063155475000037e-05,
1789
+ "loss": 1.0346,
1790
+ "step": 251
1791
+ },
1792
+ {
1793
+ "epoch": 0.58,
1794
+ "grad_norm": 0.7713218927383423,
1795
+ "learning_rate": 7.988222444919364e-05,
1796
+ "loss": 0.9758,
1797
+ "step": 252
1798
+ },
1799
+ {
1800
+ "epoch": 0.59,
1801
+ "grad_norm": 1.0017491579055786,
1802
+ "learning_rate": 7.913406957074037e-05,
1803
+ "loss": 1.2187,
1804
+ "step": 253
1805
+ },
1806
+ {
1807
+ "epoch": 0.59,
1808
+ "grad_norm": 0.6401370763778687,
1809
+ "learning_rate": 7.838713382712583e-05,
1810
+ "loss": 0.6044,
1811
+ "step": 254
1812
+ },
1813
+ {
1814
+ "epoch": 0.59,
1815
+ "grad_norm": 0.7729225158691406,
1816
+ "learning_rate": 7.76414608596049e-05,
1817
+ "loss": 1.0083,
1818
+ "step": 255
1819
+ },
1820
+ {
1821
+ "epoch": 0.59,
1822
+ "grad_norm": 0.7392654418945312,
1823
+ "learning_rate": 7.68970942356521e-05,
1824
+ "loss": 0.6903,
1825
+ "step": 256
1826
+ },
1827
+ {
1828
+ "epoch": 0.6,
1829
+ "grad_norm": 1.2603764533996582,
1830
+ "learning_rate": 7.615407744641619e-05,
1831
+ "loss": 1.0388,
1832
+ "step": 257
1833
+ },
1834
+ {
1835
+ "epoch": 0.6,
1836
+ "grad_norm": 1.094689130783081,
1837
+ "learning_rate": 7.541245390417906e-05,
1838
+ "loss": 1.1617,
1839
+ "step": 258
1840
+ },
1841
+ {
1842
+ "epoch": 0.6,
1843
+ "grad_norm": 0.7427874803543091,
1844
+ "learning_rate": 7.467226693981925e-05,
1845
+ "loss": 0.977,
1846
+ "step": 259
1847
+ },
1848
+ {
1849
+ "epoch": 0.6,
1850
+ "grad_norm": 1.2259175777435303,
1851
+ "learning_rate": 7.393355980028039e-05,
1852
+ "loss": 1.3001,
1853
+ "step": 260
1854
+ },
1855
+ {
1856
+ "epoch": 0.61,
1857
+ "grad_norm": 0.7839226722717285,
1858
+ "learning_rate": 7.319637564604412e-05,
1859
+ "loss": 1.3195,
1860
+ "step": 261
1861
+ },
1862
+ {
1863
+ "epoch": 0.61,
1864
+ "grad_norm": 0.9151708483695984,
1865
+ "learning_rate": 7.246075754860868e-05,
1866
+ "loss": 0.8617,
1867
+ "step": 262
1868
+ },
1869
+ {
1870
+ "epoch": 0.61,
1871
+ "grad_norm": 0.9389089345932007,
1872
+ "learning_rate": 7.172674848797219e-05,
1873
+ "loss": 1.1299,
1874
+ "step": 263
1875
+ },
1876
+ {
1877
+ "epoch": 0.61,
1878
+ "grad_norm": 0.9104580879211426,
1879
+ "learning_rate": 7.099439135012153e-05,
1880
+ "loss": 1.4089,
1881
+ "step": 264
1882
+ },
1883
+ {
1884
+ "epoch": 0.61,
1885
+ "grad_norm": 1.0850249528884888,
1886
+ "learning_rate": 7.026372892452653e-05,
1887
+ "loss": 1.1907,
1888
+ "step": 265
1889
+ },
1890
+ {
1891
+ "epoch": 0.62,
1892
+ "grad_norm": 0.6725065112113953,
1893
+ "learning_rate": 6.953480390164e-05,
1894
+ "loss": 0.7875,
1895
+ "step": 266
1896
+ },
1897
+ {
1898
+ "epoch": 0.62,
1899
+ "grad_norm": 1.6374472379684448,
1900
+ "learning_rate": 6.880765887040343e-05,
1901
+ "loss": 1.4271,
1902
+ "step": 267
1903
+ },
1904
+ {
1905
+ "epoch": 0.62,
1906
+ "grad_norm": 0.691931426525116,
1907
+ "learning_rate": 6.808233631575867e-05,
1908
+ "loss": 0.7463,
1909
+ "step": 268
1910
+ },
1911
+ {
1912
+ "epoch": 0.62,
1913
+ "grad_norm": 0.9267165660858154,
1914
+ "learning_rate": 6.735887861616556e-05,
1915
+ "loss": 1.2322,
1916
+ "step": 269
1917
+ },
1918
+ {
1919
+ "epoch": 0.63,
1920
+ "grad_norm": 0.8684428930282593,
1921
+ "learning_rate": 6.663732804112603e-05,
1922
+ "loss": 1.1335,
1923
+ "step": 270
1924
+ },
1925
+ {
1926
+ "epoch": 0.63,
1927
+ "grad_norm": 0.9493190050125122,
1928
+ "learning_rate": 6.591772674871434e-05,
1929
+ "loss": 1.25,
1930
+ "step": 271
1931
+ },
1932
+ {
1933
+ "epoch": 0.63,
1934
+ "grad_norm": 0.8068637847900391,
1935
+ "learning_rate": 6.520011678311382e-05,
1936
+ "loss": 1.2451,
1937
+ "step": 272
1938
+ },
1939
+ {
1940
+ "epoch": 0.63,
1941
+ "grad_norm": 1.043681263923645,
1942
+ "learning_rate": 6.448454007216054e-05,
1943
+ "loss": 1.1108,
1944
+ "step": 273
1945
+ },
1946
+ {
1947
+ "epoch": 0.64,
1948
+ "grad_norm": 1.0855973958969116,
1949
+ "learning_rate": 6.377103842489343e-05,
1950
+ "loss": 0.8609,
1951
+ "step": 274
1952
+ },
1953
+ {
1954
+ "epoch": 0.64,
1955
+ "grad_norm": 0.7895878553390503,
1956
+ "learning_rate": 6.305965352911161e-05,
1957
+ "loss": 0.9859,
1958
+ "step": 275
1959
+ },
1960
+ {
1961
+ "epoch": 0.64,
1962
+ "grad_norm": 0.89215087890625,
1963
+ "learning_rate": 6.235042694893862e-05,
1964
+ "loss": 1.3013,
1965
+ "step": 276
1966
+ },
1967
+ {
1968
+ "epoch": 0.64,
1969
+ "grad_norm": 1.1240825653076172,
1970
+ "learning_rate": 6.164340012239396e-05,
1971
+ "loss": 1.2481,
1972
+ "step": 277
1973
+ },
1974
+ {
1975
+ "epoch": 0.65,
1976
+ "grad_norm": 0.8804110288619995,
1977
+ "learning_rate": 6.093861435897208e-05,
1978
+ "loss": 1.354,
1979
+ "step": 278
1980
+ },
1981
+ {
1982
+ "epoch": 0.65,
1983
+ "grad_norm": 0.8492618799209595,
1984
+ "learning_rate": 6.02361108372286e-05,
1985
+ "loss": 1.196,
1986
+ "step": 279
1987
+ },
1988
+ {
1989
+ "epoch": 0.65,
1990
+ "grad_norm": 0.7941075563430786,
1991
+ "learning_rate": 5.953593060237457e-05,
1992
+ "loss": 1.1934,
1993
+ "step": 280
1994
+ },
1995
+ {
1996
+ "epoch": 0.65,
1997
+ "grad_norm": 0.831951379776001,
1998
+ "learning_rate": 5.883811456387821e-05,
1999
+ "loss": 0.9866,
2000
+ "step": 281
2001
+ },
2002
+ {
2003
+ "epoch": 0.65,
2004
+ "grad_norm": 0.8883886337280273,
2005
+ "learning_rate": 5.8142703493074714e-05,
2006
+ "loss": 0.9827,
2007
+ "step": 282
2008
+ },
2009
+ {
2010
+ "epoch": 0.66,
2011
+ "grad_norm": 0.8326197862625122,
2012
+ "learning_rate": 5.7449738020784085e-05,
2013
+ "loss": 1.2837,
2014
+ "step": 283
2015
+ },
2016
+ {
2017
+ "epoch": 0.66,
2018
+ "grad_norm": 0.8599093556404114,
2019
+ "learning_rate": 5.675925863493721e-05,
2020
+ "loss": 1.1051,
2021
+ "step": 284
2022
+ },
2023
+ {
2024
+ "epoch": 0.66,
2025
+ "grad_norm": 0.9380517601966858,
2026
+ "learning_rate": 5.607130567821031e-05,
2027
+ "loss": 1.1518,
2028
+ "step": 285
2029
+ },
2030
+ {
2031
+ "epoch": 0.66,
2032
+ "grad_norm": 0.9544126391410828,
2033
+ "learning_rate": 5.5385919345667715e-05,
2034
+ "loss": 1.2313,
2035
+ "step": 286
2036
+ },
2037
+ {
2038
+ "epoch": 0.67,
2039
+ "grad_norm": 1.0354762077331543,
2040
+ "learning_rate": 5.4703139682413586e-05,
2041
+ "loss": 1.2945,
2042
+ "step": 287
2043
+ },
2044
+ {
2045
+ "epoch": 0.67,
2046
+ "grad_norm": 0.9215995669364929,
2047
+ "learning_rate": 5.402300658125197e-05,
2048
+ "loss": 0.9978,
2049
+ "step": 288
2050
+ },
2051
+ {
2052
+ "epoch": 0.67,
2053
+ "grad_norm": 0.7332642674446106,
2054
+ "learning_rate": 5.334555978035609e-05,
2055
+ "loss": 0.9027,
2056
+ "step": 289
2057
+ },
2058
+ {
2059
+ "epoch": 0.67,
2060
+ "grad_norm": 1.0011807680130005,
2061
+ "learning_rate": 5.267083886094668e-05,
2062
+ "loss": 1.4776,
2063
+ "step": 290
2064
+ },
2065
+ {
2066
+ "epoch": 0.68,
2067
+ "grad_norm": 0.9119071364402771,
2068
+ "learning_rate": 5.199888324497907e-05,
2069
+ "loss": 1.0865,
2070
+ "step": 291
2071
+ },
2072
+ {
2073
+ "epoch": 0.68,
2074
+ "grad_norm": 0.905940055847168,
2075
+ "learning_rate": 5.132973219284023e-05,
2076
+ "loss": 1.1807,
2077
+ "step": 292
2078
+ },
2079
+ {
2080
+ "epoch": 0.68,
2081
+ "grad_norm": 0.8075955510139465,
2082
+ "learning_rate": 5.0663424801054595e-05,
2083
+ "loss": 1.1457,
2084
+ "step": 293
2085
+ },
2086
+ {
2087
+ "epoch": 0.68,
2088
+ "grad_norm": 0.9124309420585632,
2089
+ "learning_rate": 5.000000000000002e-05,
2090
+ "loss": 1.0714,
2091
+ "step": 294
2092
+ },
2093
+ {
2094
+ "epoch": 0.68,
2095
+ "grad_norm": 1.0044463872909546,
2096
+ "learning_rate": 4.9339496551632944e-05,
2097
+ "loss": 1.4506,
2098
+ "step": 295
2099
+ },
2100
+ {
2101
+ "epoch": 0.69,
2102
+ "grad_norm": 0.9177619218826294,
2103
+ "learning_rate": 4.8681953047223914e-05,
2104
+ "loss": 0.8655,
2105
+ "step": 296
2106
+ },
2107
+ {
2108
+ "epoch": 0.69,
2109
+ "grad_norm": 1.0393673181533813,
2110
+ "learning_rate": 4.8027407905102585e-05,
2111
+ "loss": 1.1033,
2112
+ "step": 297
2113
+ },
2114
+ {
2115
+ "epoch": 0.69,
2116
+ "grad_norm": 0.7895400524139404,
2117
+ "learning_rate": 4.73758993684131e-05,
2118
+ "loss": 0.8526,
2119
+ "step": 298
2120
+ },
2121
+ {
2122
+ "epoch": 0.69,
2123
+ "grad_norm": 0.6922112107276917,
2124
+ "learning_rate": 4.672746550287985e-05,
2125
+ "loss": 0.9227,
2126
+ "step": 299
2127
+ },
2128
+ {
2129
+ "epoch": 0.7,
2130
+ "grad_norm": 1.066853404045105,
2131
+ "learning_rate": 4.6082144194583056e-05,
2132
+ "loss": 1.4358,
2133
+ "step": 300
2134
+ },
2135
+ {
2136
+ "epoch": 0.7,
2137
+ "grad_norm": 0.911865770816803,
2138
+ "learning_rate": 4.543997314774553e-05,
2139
+ "loss": 1.0442,
2140
+ "step": 301
2141
+ },
2142
+ {
2143
+ "epoch": 0.7,
2144
+ "grad_norm": 0.8865716457366943,
2145
+ "learning_rate": 4.4800989882529574e-05,
2146
+ "loss": 1.139,
2147
+ "step": 302
2148
+ },
2149
+ {
2150
+ "epoch": 0.7,
2151
+ "grad_norm": 0.9678537249565125,
2152
+ "learning_rate": 4.41652317328447e-05,
2153
+ "loss": 1.2122,
2154
+ "step": 303
2155
+ },
2156
+ {
2157
+ "epoch": 0.71,
2158
+ "grad_norm": 0.9031918048858643,
2159
+ "learning_rate": 4.3532735844166574e-05,
2160
+ "loss": 1.0144,
2161
+ "step": 304
2162
+ },
2163
+ {
2164
+ "epoch": 0.71,
2165
+ "grad_norm": 0.8521971702575684,
2166
+ "learning_rate": 4.2903539171366393e-05,
2167
+ "loss": 1.1825,
2168
+ "step": 305
2169
+ },
2170
+ {
2171
+ "epoch": 0.71,
2172
+ "grad_norm": 1.575620174407959,
2173
+ "learning_rate": 4.227767847655205e-05,
2174
+ "loss": 1.2486,
2175
+ "step": 306
2176
+ },
2177
+ {
2178
+ "epoch": 0.71,
2179
+ "grad_norm": 0.8732744455337524,
2180
+ "learning_rate": 4.165519032691998e-05,
2181
+ "loss": 1.2195,
2182
+ "step": 307
2183
+ },
2184
+ {
2185
+ "epoch": 0.71,
2186
+ "grad_norm": 0.6721953749656677,
2187
+ "learning_rate": 4.1036111092618725e-05,
2188
+ "loss": 0.9281,
2189
+ "step": 308
2190
+ },
2191
+ {
2192
+ "epoch": 0.72,
2193
+ "grad_norm": 0.8912865519523621,
2194
+ "learning_rate": 4.042047694462404e-05,
2195
+ "loss": 1.059,
2196
+ "step": 309
2197
+ },
2198
+ {
2199
+ "epoch": 0.72,
2200
+ "grad_norm": 0.9858503341674805,
2201
+ "learning_rate": 3.9808323852625316e-05,
2202
+ "loss": 1.1077,
2203
+ "step": 310
2204
+ },
2205
+ {
2206
+ "epoch": 0.72,
2207
+ "grad_norm": 1.1012351512908936,
2208
+ "learning_rate": 3.919968758292425e-05,
2209
+ "loss": 1.3365,
2210
+ "step": 311
2211
+ },
2212
+ {
2213
+ "epoch": 0.72,
2214
+ "grad_norm": 0.9721686244010925,
2215
+ "learning_rate": 3.859460369634479e-05,
2216
+ "loss": 1.1154,
2217
+ "step": 312
2218
+ },
2219
+ {
2220
+ "epoch": 0.73,
2221
+ "grad_norm": 0.9255664348602295,
2222
+ "learning_rate": 3.799310754615578e-05,
2223
+ "loss": 1.3577,
2224
+ "step": 313
2225
+ },
2226
+ {
2227
+ "epoch": 0.73,
2228
+ "grad_norm": 0.6955008506774902,
2229
+ "learning_rate": 3.7395234276005087e-05,
2230
+ "loss": 0.8553,
2231
+ "step": 314
2232
+ },
2233
+ {
2234
+ "epoch": 0.73,
2235
+ "grad_norm": 1.000977635383606,
2236
+ "learning_rate": 3.6801018817866375e-05,
2237
+ "loss": 1.0546,
2238
+ "step": 315
2239
+ },
2240
+ {
2241
+ "epoch": 0.73,
2242
+ "grad_norm": 0.9564784169197083,
2243
+ "learning_rate": 3.62104958899982e-05,
2244
+ "loss": 1.1865,
2245
+ "step": 316
2246
+ },
2247
+ {
2248
+ "epoch": 0.74,
2249
+ "grad_norm": 1.024553894996643,
2250
+ "learning_rate": 3.562369999491536e-05,
2251
+ "loss": 1.4169,
2252
+ "step": 317
2253
+ },
2254
+ {
2255
+ "epoch": 0.74,
2256
+ "grad_norm": 0.7676483392715454,
2257
+ "learning_rate": 3.504066541737323e-05,
2258
+ "loss": 1.0726,
2259
+ "step": 318
2260
+ },
2261
+ {
2262
+ "epoch": 0.74,
2263
+ "grad_norm": 0.7936806082725525,
2264
+ "learning_rate": 3.4461426222364336e-05,
2265
+ "loss": 1.0008,
2266
+ "step": 319
2267
+ },
2268
+ {
2269
+ "epoch": 0.74,
2270
+ "grad_norm": 0.8642453551292419,
2271
+ "learning_rate": 3.3886016253128326e-05,
2272
+ "loss": 1.1144,
2273
+ "step": 320
2274
+ },
2275
+ {
2276
+ "epoch": 0.74,
2277
+ "grad_norm": 0.9978905320167542,
2278
+ "learning_rate": 3.3314469129174364e-05,
2279
+ "loss": 1.3217,
2280
+ "step": 321
2281
+ },
2282
+ {
2283
+ "epoch": 0.75,
2284
+ "grad_norm": 0.955431342124939,
2285
+ "learning_rate": 3.2746818244316956e-05,
2286
+ "loss": 1.0409,
2287
+ "step": 322
2288
+ },
2289
+ {
2290
+ "epoch": 0.75,
2291
+ "grad_norm": 1.0944534540176392,
2292
+ "learning_rate": 3.2183096764724915e-05,
2293
+ "loss": 1.4031,
2294
+ "step": 323
2295
+ },
2296
+ {
2297
+ "epoch": 0.75,
2298
+ "grad_norm": 0.9666315913200378,
2299
+ "learning_rate": 3.16233376269834e-05,
2300
+ "loss": 1.4093,
2301
+ "step": 324
2302
+ },
2303
+ {
2304
+ "epoch": 0.75,
2305
+ "eval_loss": 1.1115437746047974,
2306
+ "eval_runtime": 5.4226,
2307
+ "eval_samples_per_second": 18.441,
2308
+ "eval_steps_per_second": 18.441,
2309
+ "step": 324
2310
+ },
2311
+ {
2312
+ "epoch": 0.75,
2313
+ "grad_norm": 0.7884992957115173,
2314
+ "learning_rate": 3.106757353616966e-05,
2315
+ "loss": 0.8406,
2316
+ "step": 325
2317
+ },
2318
+ {
2319
+ "epoch": 0.76,
2320
+ "grad_norm": 0.8696621656417847,
2321
+ "learning_rate": 3.0515836963942056e-05,
2322
+ "loss": 1.2945,
2323
+ "step": 326
2324
+ },
2325
+ {
2326
+ "epoch": 0.76,
2327
+ "grad_norm": 0.8080437183380127,
2328
+ "learning_rate": 2.9968160146643022e-05,
2329
+ "loss": 1.1088,
2330
+ "step": 327
2331
+ },
2332
+ {
2333
+ "epoch": 0.76,
2334
+ "grad_norm": 1.0544288158416748,
2335
+ "learning_rate": 2.9424575083415362e-05,
2336
+ "loss": 1.5478,
2337
+ "step": 328
2338
+ },
2339
+ {
2340
+ "epoch": 0.76,
2341
+ "grad_norm": 0.9855386018753052,
2342
+ "learning_rate": 2.888511353433274e-05,
2343
+ "loss": 1.1224,
2344
+ "step": 329
2345
+ },
2346
+ {
2347
+ "epoch": 0.77,
2348
+ "grad_norm": 0.8086793422698975,
2349
+ "learning_rate": 2.8349807018544174e-05,
2350
+ "loss": 1.195,
2351
+ "step": 330
2352
+ },
2353
+ {
2354
+ "epoch": 0.77,
2355
+ "grad_norm": 0.7815480828285217,
2356
+ "learning_rate": 2.7818686812432136e-05,
2357
+ "loss": 1.1851,
2358
+ "step": 331
2359
+ },
2360
+ {
2361
+ "epoch": 0.77,
2362
+ "grad_norm": 1.0230026245117188,
2363
+ "learning_rate": 2.7291783947785543e-05,
2364
+ "loss": 1.506,
2365
+ "step": 332
2366
+ },
2367
+ {
2368
+ "epoch": 0.77,
2369
+ "grad_norm": 0.9805968403816223,
2370
+ "learning_rate": 2.6769129209986322e-05,
2371
+ "loss": 1.4654,
2372
+ "step": 333
2373
+ },
2374
+ {
2375
+ "epoch": 0.77,
2376
+ "grad_norm": 0.9589524865150452,
2377
+ "learning_rate": 2.6250753136210983e-05,
2378
+ "loss": 0.8588,
2379
+ "step": 334
2380
+ },
2381
+ {
2382
+ "epoch": 0.78,
2383
+ "grad_norm": 0.792204737663269,
2384
+ "learning_rate": 2.5736686013646228e-05,
2385
+ "loss": 1.1536,
2386
+ "step": 335
2387
+ },
2388
+ {
2389
+ "epoch": 0.78,
2390
+ "grad_norm": 1.411292314529419,
2391
+ "learning_rate": 2.5226957877719436e-05,
2392
+ "loss": 1.1398,
2393
+ "step": 336
2394
+ },
2395
+ {
2396
+ "epoch": 0.78,
2397
+ "grad_norm": 1.2491581439971924,
2398
+ "learning_rate": 2.4721598510343858e-05,
2399
+ "loss": 1.3447,
2400
+ "step": 337
2401
+ },
2402
+ {
2403
+ "epoch": 0.78,
2404
+ "grad_norm": 0.9221548438072205,
2405
+ "learning_rate": 2.4220637438178317e-05,
2406
+ "loss": 1.0514,
2407
+ "step": 338
2408
+ },
2409
+ {
2410
+ "epoch": 0.79,
2411
+ "grad_norm": 1.1859768629074097,
2412
+ "learning_rate": 2.372410393090243e-05,
2413
+ "loss": 1.1367,
2414
+ "step": 339
2415
+ },
2416
+ {
2417
+ "epoch": 0.79,
2418
+ "grad_norm": 0.8900750279426575,
2419
+ "learning_rate": 2.3232026999506062e-05,
2420
+ "loss": 1.2437,
2421
+ "step": 340
2422
+ },
2423
+ {
2424
+ "epoch": 0.79,
2425
+ "grad_norm": 0.8685091137886047,
2426
+ "learning_rate": 2.2744435394594497e-05,
2427
+ "loss": 1.2518,
2428
+ "step": 341
2429
+ },
2430
+ {
2431
+ "epoch": 0.79,
2432
+ "grad_norm": 0.7511561512947083,
2433
+ "learning_rate": 2.22613576047087e-05,
2434
+ "loss": 0.9799,
2435
+ "step": 342
2436
+ },
2437
+ {
2438
+ "epoch": 0.8,
2439
+ "grad_norm": 0.9015387296676636,
2440
+ "learning_rate": 2.1782821854660606e-05,
2441
+ "loss": 1.096,
2442
+ "step": 343
2443
+ },
2444
+ {
2445
+ "epoch": 0.8,
2446
+ "grad_norm": 1.0497876405715942,
2447
+ "learning_rate": 2.130885610388428e-05,
2448
+ "loss": 1.335,
2449
+ "step": 344
2450
+ },
2451
+ {
2452
+ "epoch": 0.8,
2453
+ "grad_norm": 1.0829570293426514,
2454
+ "learning_rate": 2.0839488044802036e-05,
2455
+ "loss": 1.7663,
2456
+ "step": 345
2457
+ },
2458
+ {
2459
+ "epoch": 0.8,
2460
+ "grad_norm": 0.7861917614936829,
2461
+ "learning_rate": 2.037474510120676e-05,
2462
+ "loss": 1.1382,
2463
+ "step": 346
2464
+ },
2465
+ {
2466
+ "epoch": 0.81,
2467
+ "grad_norm": 0.9880712628364563,
2468
+ "learning_rate": 1.9914654426659374e-05,
2469
+ "loss": 1.0684,
2470
+ "step": 347
2471
+ },
2472
+ {
2473
+ "epoch": 0.81,
2474
+ "grad_norm": 0.882596492767334,
2475
+ "learning_rate": 1.945924290290242e-05,
2476
+ "loss": 1.1136,
2477
+ "step": 348
2478
+ },
2479
+ {
2480
+ "epoch": 0.81,
2481
+ "grad_norm": 1.4436990022659302,
2482
+ "learning_rate": 1.9008537138289527e-05,
2483
+ "loss": 1.5471,
2484
+ "step": 349
2485
+ },
2486
+ {
2487
+ "epoch": 0.81,
2488
+ "grad_norm": 0.8502001166343689,
2489
+ "learning_rate": 1.8562563466230576e-05,
2490
+ "loss": 1.2033,
2491
+ "step": 350
2492
+ },
2493
+ {
2494
+ "epoch": 0.81,
2495
+ "grad_norm": 0.8720386624336243,
2496
+ "learning_rate": 1.8121347943653332e-05,
2497
+ "loss": 1.3097,
2498
+ "step": 351
2499
+ },
2500
+ {
2501
+ "epoch": 0.82,
2502
+ "grad_norm": 0.6880800724029541,
2503
+ "learning_rate": 1.7684916349480794e-05,
2504
+ "loss": 0.943,
2505
+ "step": 352
2506
+ },
2507
+ {
2508
+ "epoch": 0.82,
2509
+ "grad_norm": 1.0741546154022217,
2510
+ "learning_rate": 1.7253294183125223e-05,
2511
+ "loss": 1.1616,
2512
+ "step": 353
2513
+ },
2514
+ {
2515
+ "epoch": 0.82,
2516
+ "grad_norm": 0.966346263885498,
2517
+ "learning_rate": 1.6826506662998097e-05,
2518
+ "loss": 1.353,
2519
+ "step": 354
2520
+ },
2521
+ {
2522
+ "epoch": 0.82,
2523
+ "grad_norm": 0.7042524814605713,
2524
+ "learning_rate": 1.64045787250368e-05,
2525
+ "loss": 1.102,
2526
+ "step": 355
2527
+ },
2528
+ {
2529
+ "epoch": 0.83,
2530
+ "grad_norm": 0.8945311903953552,
2531
+ "learning_rate": 1.5987535021247667e-05,
2532
+ "loss": 1.0206,
2533
+ "step": 356
2534
+ },
2535
+ {
2536
+ "epoch": 0.83,
2537
+ "grad_norm": 0.7530681490898132,
2538
+ "learning_rate": 1.5575399918265542e-05,
2539
+ "loss": 1.1307,
2540
+ "step": 357
2541
+ },
2542
+ {
2543
+ "epoch": 0.83,
2544
+ "grad_norm": 0.9247698783874512,
2545
+ "learning_rate": 1.5168197495930315e-05,
2546
+ "loss": 1.4552,
2547
+ "step": 358
2548
+ },
2549
+ {
2550
+ "epoch": 0.83,
2551
+ "grad_norm": 0.8581532835960388,
2552
+ "learning_rate": 1.476595154587973e-05,
2553
+ "loss": 0.9923,
2554
+ "step": 359
2555
+ },
2556
+ {
2557
+ "epoch": 0.84,
2558
+ "grad_norm": 0.9739102125167847,
2559
+ "learning_rate": 1.436868557015959e-05,
2560
+ "loss": 1.3055,
2561
+ "step": 360
2562
+ },
2563
+ {
2564
+ "epoch": 0.84,
2565
+ "grad_norm": 1.1127710342407227,
2566
+ "learning_rate": 1.3976422779850384e-05,
2567
+ "loss": 1.3258,
2568
+ "step": 361
2569
+ },
2570
+ {
2571
+ "epoch": 0.84,
2572
+ "grad_norm": 0.7771569490432739,
2573
+ "learning_rate": 1.3589186093711226e-05,
2574
+ "loss": 0.9989,
2575
+ "step": 362
2576
+ },
2577
+ {
2578
+ "epoch": 0.84,
2579
+ "grad_norm": 0.9304590821266174,
2580
+ "learning_rate": 1.3206998136840831e-05,
2581
+ "loss": 1.2083,
2582
+ "step": 363
2583
+ },
2584
+ {
2585
+ "epoch": 0.84,
2586
+ "grad_norm": 1.4773048162460327,
2587
+ "learning_rate": 1.2829881239355468e-05,
2588
+ "loss": 1.3064,
2589
+ "step": 364
2590
+ },
2591
+ {
2592
+ "epoch": 0.85,
2593
+ "grad_norm": 1.212895393371582,
2594
+ "learning_rate": 1.2457857435084408e-05,
2595
+ "loss": 2.7761,
2596
+ "step": 365
2597
+ },
2598
+ {
2599
+ "epoch": 0.85,
2600
+ "grad_norm": 0.7761988639831543,
2601
+ "learning_rate": 1.2090948460282414e-05,
2602
+ "loss": 1.1132,
2603
+ "step": 366
2604
+ },
2605
+ {
2606
+ "epoch": 0.85,
2607
+ "grad_norm": 1.2847517728805542,
2608
+ "learning_rate": 1.1729175752359922e-05,
2609
+ "loss": 1.3685,
2610
+ "step": 367
2611
+ },
2612
+ {
2613
+ "epoch": 0.85,
2614
+ "grad_norm": 0.8102625608444214,
2615
+ "learning_rate": 1.1372560448630376e-05,
2616
+ "loss": 1.1863,
2617
+ "step": 368
2618
+ },
2619
+ {
2620
+ "epoch": 0.86,
2621
+ "grad_norm": 0.7838321328163147,
2622
+ "learning_rate": 1.102112338507526e-05,
2623
+ "loss": 1.153,
2624
+ "step": 369
2625
+ },
2626
+ {
2627
+ "epoch": 0.86,
2628
+ "grad_norm": 1.1303842067718506,
2629
+ "learning_rate": 1.067488509512683e-05,
2630
+ "loss": 0.9941,
2631
+ "step": 370
2632
+ },
2633
+ {
2634
+ "epoch": 0.86,
2635
+ "grad_norm": 0.917951226234436,
2636
+ "learning_rate": 1.0333865808468202e-05,
2637
+ "loss": 1.0509,
2638
+ "step": 371
2639
+ },
2640
+ {
2641
+ "epoch": 0.86,
2642
+ "grad_norm": 1.0915648937225342,
2643
+ "learning_rate": 9.998085449851635e-06,
2644
+ "loss": 0.999,
2645
+ "step": 372
2646
+ },
2647
+ {
2648
+ "epoch": 0.87,
2649
+ "grad_norm": 0.8339885473251343,
2650
+ "learning_rate": 9.667563637934129e-06,
2651
+ "loss": 0.8986,
2652
+ "step": 373
2653
+ },
2654
+ {
2655
+ "epoch": 0.87,
2656
+ "grad_norm": 0.8891353607177734,
2657
+ "learning_rate": 9.342319684131395e-06,
2658
+ "loss": 1.1157,
2659
+ "step": 374
2660
+ },
2661
+ {
2662
+ "epoch": 0.87,
2663
+ "grad_norm": 1.030124545097351,
2664
+ "learning_rate": 9.02237259148938e-06,
2665
+ "loss": 1.1778,
2666
+ "step": 375
2667
+ },
2668
+ {
2669
+ "epoch": 0.87,
2670
+ "grad_norm": 0.877785325050354,
2671
+ "learning_rate": 8.70774105357407e-06,
2672
+ "loss": 1.2047,
2673
+ "step": 376
2674
+ },
2675
+ {
2676
+ "epoch": 0.87,
2677
+ "grad_norm": 0.9380621910095215,
2678
+ "learning_rate": 8.398443453379267e-06,
2679
+ "loss": 1.1252,
2680
+ "step": 377
2681
+ },
2682
+ {
2683
+ "epoch": 0.88,
2684
+ "grad_norm": 0.879805326461792,
2685
+ "learning_rate": 8.094497862252471e-06,
2686
+ "loss": 1.3558,
2687
+ "step": 378
2688
+ },
2689
+ {
2690
+ "epoch": 0.88,
2691
+ "grad_norm": 1.081350564956665,
2692
+ "learning_rate": 7.795922038839032e-06,
2693
+ "loss": 1.3124,
2694
+ "step": 379
2695
+ },
2696
+ {
2697
+ "epoch": 0.88,
2698
+ "grad_norm": 0.8164477348327637,
2699
+ "learning_rate": 7.502733428044683e-06,
2700
+ "loss": 1.0358,
2701
+ "step": 380
2702
+ },
2703
+ {
2704
+ "epoch": 0.88,
2705
+ "grad_norm": 0.8706633448600769,
2706
+ "learning_rate": 7.214949160016115e-06,
2707
+ "loss": 1.1296,
2708
+ "step": 381
2709
+ },
2710
+ {
2711
+ "epoch": 0.89,
2712
+ "grad_norm": 0.9576455354690552,
2713
+ "learning_rate": 6.932586049140255e-06,
2714
+ "loss": 1.126,
2715
+ "step": 382
2716
+ },
2717
+ {
2718
+ "epoch": 0.89,
2719
+ "grad_norm": 0.9922929406166077,
2720
+ "learning_rate": 6.655660593061719e-06,
2721
+ "loss": 1.2583,
2722
+ "step": 383
2723
+ },
2724
+ {
2725
+ "epoch": 0.89,
2726
+ "grad_norm": 0.8414103984832764,
2727
+ "learning_rate": 6.384188971719052e-06,
2728
+ "loss": 1.1785,
2729
+ "step": 384
2730
+ },
2731
+ {
2732
+ "epoch": 0.89,
2733
+ "grad_norm": 1.0709247589111328,
2734
+ "learning_rate": 6.11818704639926e-06,
2735
+ "loss": 1.4718,
2736
+ "step": 385
2737
+ },
2738
+ {
2739
+ "epoch": 0.9,
2740
+ "grad_norm": 1.3863894939422607,
2741
+ "learning_rate": 5.857670358811096e-06,
2742
+ "loss": 1.2066,
2743
+ "step": 386
2744
+ },
2745
+ {
2746
+ "epoch": 0.9,
2747
+ "grad_norm": 1.3715444803237915,
2748
+ "learning_rate": 5.6026541301771095e-06,
2749
+ "loss": 1.0306,
2750
+ "step": 387
2751
+ },
2752
+ {
2753
+ "epoch": 0.9,
2754
+ "grad_norm": 0.759107768535614,
2755
+ "learning_rate": 5.353153260344179e-06,
2756
+ "loss": 0.4256,
2757
+ "step": 388
2758
+ },
2759
+ {
2760
+ "epoch": 0.9,
2761
+ "grad_norm": 0.7659134864807129,
2762
+ "learning_rate": 5.109182326913054e-06,
2763
+ "loss": 1.1735,
2764
+ "step": 389
2765
+ },
2766
+ {
2767
+ "epoch": 0.9,
2768
+ "grad_norm": 0.8766858577728271,
2769
+ "learning_rate": 4.870755584386544e-06,
2770
+ "loss": 1.135,
2771
+ "step": 390
2772
+ },
2773
+ {
2774
+ "epoch": 0.91,
2775
+ "grad_norm": 1.0143039226531982,
2776
+ "learning_rate": 4.63788696333678e-06,
2777
+ "loss": 1.3126,
2778
+ "step": 391
2779
+ },
2780
+ {
2781
+ "epoch": 0.91,
2782
+ "grad_norm": 0.8704203963279724,
2783
+ "learning_rate": 4.410590069591192e-06,
2784
+ "loss": 1.0583,
2785
+ "step": 392
2786
+ },
2787
+ {
2788
+ "epoch": 0.91,
2789
+ "grad_norm": 1.024915337562561,
2790
+ "learning_rate": 4.188878183437594e-06,
2791
+ "loss": 0.9828,
2792
+ "step": 393
2793
+ },
2794
+ {
2795
+ "epoch": 0.91,
2796
+ "grad_norm": 0.9721056222915649,
2797
+ "learning_rate": 3.972764258848305e-06,
2798
+ "loss": 0.8001,
2799
+ "step": 394
2800
+ },
2801
+ {
2802
+ "epoch": 0.92,
2803
+ "grad_norm": 0.8740870356559753,
2804
+ "learning_rate": 3.7622609227231818e-06,
2805
+ "loss": 1.2904,
2806
+ "step": 395
2807
+ },
2808
+ {
2809
+ "epoch": 0.92,
2810
+ "grad_norm": 1.0467931032180786,
2811
+ "learning_rate": 3.5573804741519833e-06,
2812
+ "loss": 1.0275,
2813
+ "step": 396
2814
+ },
2815
+ {
2816
+ "epoch": 0.92,
2817
+ "grad_norm": 0.8117631673812866,
2818
+ "learning_rate": 3.3581348836956738e-06,
2819
+ "loss": 0.8395,
2820
+ "step": 397
2821
+ },
2822
+ {
2823
+ "epoch": 0.92,
2824
+ "grad_norm": 0.8695462942123413,
2825
+ "learning_rate": 3.1645357926870955e-06,
2826
+ "loss": 1.0884,
2827
+ "step": 398
2828
+ },
2829
+ {
2830
+ "epoch": 0.93,
2831
+ "grad_norm": 1.1519237756729126,
2832
+ "learning_rate": 2.9765945125507235e-06,
2833
+ "loss": 1.3518,
2834
+ "step": 399
2835
+ },
2836
+ {
2837
+ "epoch": 0.93,
2838
+ "grad_norm": 1.028548240661621,
2839
+ "learning_rate": 2.7943220241418377e-06,
2840
+ "loss": 1.6295,
2841
+ "step": 400
2842
+ },
2843
+ {
2844
+ "epoch": 0.93,
2845
+ "grad_norm": 1.0098282098770142,
2846
+ "learning_rate": 2.6177289771049274e-06,
2847
+ "loss": 1.2196,
2848
+ "step": 401
2849
+ },
2850
+ {
2851
+ "epoch": 0.93,
2852
+ "grad_norm": 0.8729552030563354,
2853
+ "learning_rate": 2.4468256892514417e-06,
2854
+ "loss": 0.8682,
2855
+ "step": 402
2856
+ },
2857
+ {
2858
+ "epoch": 0.94,
2859
+ "grad_norm": 0.7072241306304932,
2860
+ "learning_rate": 2.281622145956952e-06,
2861
+ "loss": 0.7565,
2862
+ "step": 403
2863
+ },
2864
+ {
2865
+ "epoch": 0.94,
2866
+ "grad_norm": 0.8818898797035217,
2867
+ "learning_rate": 2.122127999577783e-06,
2868
+ "loss": 1.0552,
2869
+ "step": 404
2870
+ },
2871
+ {
2872
+ "epoch": 0.94,
2873
+ "grad_norm": 1.243230938911438,
2874
+ "learning_rate": 1.9683525688869773e-06,
2875
+ "loss": 1.2233,
2876
+ "step": 405
2877
+ },
2878
+ {
2879
+ "epoch": 0.94,
2880
+ "grad_norm": 0.8495836853981018,
2881
+ "learning_rate": 1.8203048385299181e-06,
2882
+ "loss": 1.1537,
2883
+ "step": 406
2884
+ },
2885
+ {
2886
+ "epoch": 0.94,
2887
+ "grad_norm": 1.1537418365478516,
2888
+ "learning_rate": 1.6779934584992718e-06,
2889
+ "loss": 1.2535,
2890
+ "step": 407
2891
+ },
2892
+ {
2893
+ "epoch": 0.95,
2894
+ "grad_norm": 0.8221082091331482,
2895
+ "learning_rate": 1.5414267436297037e-06,
2896
+ "loss": 1.1052,
2897
+ "step": 408
2898
+ },
2899
+ {
2900
+ "epoch": 0.95,
2901
+ "grad_norm": 0.7960305213928223,
2902
+ "learning_rate": 1.4106126731119996e-06,
2903
+ "loss": 1.2468,
2904
+ "step": 409
2905
+ },
2906
+ {
2907
+ "epoch": 0.95,
2908
+ "grad_norm": 0.9470089077949524,
2909
+ "learning_rate": 1.2855588900269056e-06,
2910
+ "loss": 1.193,
2911
+ "step": 410
2912
+ },
2913
+ {
2914
+ "epoch": 0.95,
2915
+ "grad_norm": 0.9635655283927917,
2916
+ "learning_rate": 1.1662727008984964e-06,
2917
+ "loss": 1.474,
2918
+ "step": 411
2919
+ },
2920
+ {
2921
+ "epoch": 0.96,
2922
+ "grad_norm": 0.8623586297035217,
2923
+ "learning_rate": 1.0527610752673944e-06,
2924
+ "loss": 1.138,
2925
+ "step": 412
2926
+ },
2927
+ {
2928
+ "epoch": 0.96,
2929
+ "grad_norm": 0.8060218095779419,
2930
+ "learning_rate": 9.450306452834179e-07,
2931
+ "loss": 1.137,
2932
+ "step": 413
2933
+ },
2934
+ {
2935
+ "epoch": 0.96,
2936
+ "grad_norm": 1.009254813194275,
2937
+ "learning_rate": 8.430877053182129e-07,
2938
+ "loss": 1.1293,
2939
+ "step": 414
2940
+ },
2941
+ {
2942
+ "epoch": 0.96,
2943
+ "grad_norm": 0.8745806813240051,
2944
+ "learning_rate": 7.469382115974032e-07,
2945
+ "loss": 1.2717,
2946
+ "step": 415
2947
+ },
2948
+ {
2949
+ "epoch": 0.97,
2950
+ "grad_norm": 0.8753180503845215,
2951
+ "learning_rate": 6.565877818526245e-07,
2952
+ "loss": 1.6308,
2953
+ "step": 416
2954
+ },
2955
+ {
2956
+ "epoch": 0.97,
2957
+ "grad_norm": 0.7818780541419983,
2958
+ "learning_rate": 5.72041694993286e-07,
2959
+ "loss": 0.7243,
2960
+ "step": 417
2961
+ },
2962
+ {
2963
+ "epoch": 0.97,
2964
+ "grad_norm": 1.1222724914550781,
2965
+ "learning_rate": 4.933048907981741e-07,
2966
+ "loss": 1.2855,
2967
+ "step": 418
2968
+ },
2969
+ {
2970
+ "epoch": 0.97,
2971
+ "grad_norm": 0.7851914763450623,
2972
+ "learning_rate": 4.203819696267486e-07,
2973
+ "loss": 1.1144,
2974
+ "step": 419
2975
+ },
2976
+ {
2977
+ "epoch": 0.97,
2978
+ "grad_norm": 1.2824020385742188,
2979
+ "learning_rate": 3.532771921504696e-07,
2980
+ "loss": 0.9624,
2981
+ "step": 420
2982
+ },
2983
+ {
2984
+ "epoch": 0.98,
2985
+ "grad_norm": 0.9270046353340149,
2986
+ "learning_rate": 2.919944791037632e-07,
2987
+ "loss": 1.1227,
2988
+ "step": 421
2989
+ },
2990
+ {
2991
+ "epoch": 0.98,
2992
+ "grad_norm": 0.9121943712234497,
2993
+ "learning_rate": 2.3653741105499338e-07,
2994
+ "loss": 1.3002,
2995
+ "step": 422
2996
+ },
2997
+ {
2998
+ "epoch": 0.98,
2999
+ "grad_norm": 0.8677274584770203,
3000
+ "learning_rate": 1.8690922819727398e-07,
3001
+ "loss": 1.2818,
3002
+ "step": 423
3003
+ },
3004
+ {
3005
+ "epoch": 0.98,
3006
+ "grad_norm": 0.9009543657302856,
3007
+ "learning_rate": 1.4311283015910893e-07,
3008
+ "loss": 1.4132,
3009
+ "step": 424
3010
+ },
3011
+ {
3012
+ "epoch": 0.99,
3013
+ "grad_norm": 0.7963302135467529,
3014
+ "learning_rate": 1.0515077583498344e-07,
3015
+ "loss": 1.0084,
3016
+ "step": 425
3017
+ },
3018
+ {
3019
+ "epoch": 0.99,
3020
+ "grad_norm": 0.7949801087379456,
3021
+ "learning_rate": 7.302528323589464e-08,
3022
+ "loss": 0.8838,
3023
+ "step": 426
3024
+ },
3025
+ {
3026
+ "epoch": 0.99,
3027
+ "grad_norm": 0.845206081867218,
3028
+ "learning_rate": 4.6738229359732935e-08,
3029
+ "loss": 0.9944,
3030
+ "step": 427
3031
+ },
3032
+ {
3033
+ "epoch": 0.99,
3034
+ "grad_norm": 0.7643120288848877,
3035
+ "learning_rate": 2.6291150081603212e-08,
3036
+ "loss": 1.1573,
3037
+ "step": 428
3038
+ },
3039
+ {
3040
+ "epoch": 1.0,
3041
+ "grad_norm": 0.9906445145606995,
3042
+ "learning_rate": 1.168524006410765e-08,
3043
+ "loss": 1.3015,
3044
+ "step": 429
3045
+ },
3046
+ {
3047
+ "epoch": 1.0,
3048
+ "grad_norm": 1.216629147529602,
3049
+ "learning_rate": 2.921352687534906e-09,
3050
+ "loss": 0.9889,
3051
+ "step": 430
3052
+ },
3053
+ {
3054
+ "epoch": 1.0,
3055
+ "grad_norm": 0.8904082179069519,
3056
+ "learning_rate": 0.0,
3057
+ "loss": 1.1844,
3058
+ "step": 431
3059
+ }
3060
+ ],
3061
+ "logging_steps": 1,
3062
+ "max_steps": 431,
3063
+ "num_input_tokens_seen": 0,
3064
+ "num_train_epochs": 1,
3065
+ "save_steps": 500,
3066
+ "total_flos": 7839152706846720.0,
3067
+ "train_batch_size": 1,
3068
+ "trial_name": null,
3069
+ "trial_params": null
3070
+ }
checkpoint-431/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20ae85fcf3e62c3900e9e6048e9302b4db3cc1e6fbd50b029bc0739c47ef99b5
3
+ size 5624
checkpoint-597/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: openlm-research/open_llama_3b_v2
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.9.0
checkpoint-597/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openlm-research/open_llama_3b_v2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 32,
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": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "v_proj",
24
+ "o_proj",
25
+ "q_proj",
26
+ "down_proj",
27
+ "up_proj",
28
+ "gate_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
checkpoint-597/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0f1f41f014b771db793fe3989264831c3ef8df78bac770780ea2ed6a5da31fe
3
+ size 50899792
checkpoint-597/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d8f2bb2165989a836390f9898996deca00393348e405a4d33a1e049a58a4a6e
3
+ size 101919290
checkpoint-597/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc2f1b74cde4a4420e9e40dc897108fc36ea3a012b622f2e0123ed439bcd1152
3
+ size 14244
checkpoint-597/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2462c7ed82882387b0c2e4a03b3e21bb60ef12f7c00dbb318874547acc66bc62
3
+ size 1064
checkpoint-597/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-597/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d84e26ce27315b618f94e914bb6b67f0bb5aa37c3903b14adcd26c9fca9f3f82
3
+ size 5624
checkpoint-796/README.md ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: openlm-research/open_llama_3b_v2
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+ ### Framework versions
201
+
202
+ - PEFT 0.9.0
checkpoint-796/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openlm-research/open_llama_3b_v2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 32,
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": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "v_proj",
24
+ "o_proj",
25
+ "q_proj",
26
+ "down_proj",
27
+ "up_proj",
28
+ "gate_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
checkpoint-796/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:842013bff659a1aeece0bd337debda2899c1b7f9c9fce662fc2a102cd6d462ed
3
+ size 50899792
checkpoint-796/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c789c934200ad012c91419cf47b9e24f7c2811b9c98f262f5c606103d4d1b13
3
+ size 101919290
checkpoint-796/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d963dde01b390675a90cd7d31c489e6ee115fa4d8a03b6f3619b526cbbfa719
3
+ size 14244
checkpoint-796/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61fd7e6c894e16612a66cbd972048c812ca7c4e2a3781a69ab7e52fea2402ea2
3
+ size 1064
checkpoint-796/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-796/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d84e26ce27315b618f94e914bb6b67f0bb5aa37c3903b14adcd26c9fca9f3f82
3
+ size 5624
config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openlm-research/open_llama_3b_v2",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 3200,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8640,
14
+ "max_position_embeddings": 2048,
15
+ "model_type": "llama",
16
+ "num_attention_heads": 32,
17
+ "num_hidden_layers": 26,
18
+ "num_key_value_heads": 32,
19
+ "pad_token_id": 0,
20
+ "pretraining_tp": 1,
21
+ "quantization_config": {
22
+ "_load_in_4bit": true,
23
+ "_load_in_8bit": false,
24
+ "bnb_4bit_compute_dtype": "float16",
25
+ "bnb_4bit_quant_type": "nf4",
26
+ "bnb_4bit_use_double_quant": true,
27
+ "llm_int8_enable_fp32_cpu_offload": false,
28
+ "llm_int8_has_fp16_weight": false,
29
+ "llm_int8_skip_modules": null,
30
+ "llm_int8_threshold": 6.0,
31
+ "load_in_4bit": true,
32
+ "load_in_8bit": false,
33
+ "quant_method": "bitsandbytes"
34
+ },
35
+ "rms_norm_eps": 1e-06,
36
+ "rope_scaling": null,
37
+ "rope_theta": 10000.0,
38
+ "tie_word_embeddings": false,
39
+ "torch_dtype": "float16",
40
+ "transformers_version": "4.38.2",
41
+ "use_cache": false,
42
+ "vocab_size": 32000
43
+ }
runs/Mar13_21-58-24_8711e78fac20/events.out.tfevents.1710367104.8711e78fac20.40.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7631a30f73d64fe416f3d56cc639c398b6f148c0744264cb60d4cc07ce49a72e
3
+ size 97683
runs/Mar13_22-06-09_8711e78fac20/events.out.tfevents.1710367570.8711e78fac20.172.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac5f6f47b09946d9d1ec181e9edd7640acb54188e8ce1f21a9384bd741b8b978
3
+ size 177950
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91b289e85fa20fd375d8b33dc12f77616f18abc6359804471d1fafcb425fecb8
3
+ size 511574
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "legacy": true,
35
+ "model_max_length": 2048,
36
+ "pad_token": "</s>",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "unk_token": "<unk>",
41
+ "use_default_system_prompt": false,
42
+ "use_fast": true
43
+ }