Hugging Face's logo

Undi95
/
toxicqa-Llama2-7B-lora

PEFT
Safetensors
llama
Generated from Trainer
8-bit precision
bitsandbytes
Model card Files Community
Undi95 commited on
Commit
7460a81
•
1 Parent(s): c13dc16

Upload folder using huggingface_hub

Browse files
Files changed (23) hide show
  1. README.md +165 -0
  2. adapter_config.json +28 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-1408/README.md +220 -0
  5. checkpoint-1408/adapter_config.json +28 -0
  6. checkpoint-1408/adapter_model.safetensors +3 -0
  7. checkpoint-1408/optimizer.pt +3 -0
  8. checkpoint-1408/rng_state.pth +3 -0
  9. checkpoint-1408/scheduler.pt +3 -0
  10. checkpoint-1408/trainer_state.json +0 -0
  11. checkpoint-1408/training_args.bin +3 -0
  12. checkpoint-704/README.md +220 -0
  13. checkpoint-704/adapter_config.json +28 -0
  14. checkpoint-704/adapter_model.safetensors +3 -0
  15. checkpoint-704/optimizer.pt +3 -0
  16. checkpoint-704/rng_state.pth +3 -0
  17. checkpoint-704/scheduler.pt +3 -0
  18. checkpoint-704/trainer_state.json +4285 -0
  19. checkpoint-704/training_args.bin +3 -0
  20. config.json +41 -0
  21. special_tokens_map.json +30 -0
  22. tokenizer.model +3 -0
  23. tokenizer_config.json +43 -0
README.md ADDED
@@ -0,0 +1,165 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ tags:
4
+ - generated_from_trainer
5
+ base_model: NousResearch/Llama-2-7b-hf
6
+ model-index:
7
+ - name: lora-out
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
+ should probably proofread and complete it, then remove this comment. -->
13
+
14
+ [<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
15
+ <details><summary>See axolotl config</summary>
16
+
17
+ axolotl version: `0.3.0`
18
+ ```yaml
19
+ base_model: NousResearch/Llama-2-7b-hf
20
+ model_type: LlamaForCausalLM
21
+ tokenizer_type: LlamaTokenizer
22
+ is_llama_derived_model: true
23
+
24
+ load_in_8bit: true
25
+ load_in_4bit: false
26
+ strict: false
27
+
28
+ datasets:
29
+ - path: dataset
30
+ type: sharegpt
31
+ dataset_prepared_path:
32
+ val_set_size: 0.05
33
+ output_dir: ./lora-out
34
+
35
+ sequence_len: 4096
36
+ sample_packing: true
37
+ pad_to_sequence_len: true
38
+
39
+ adapter: lora
40
+ lora_model_dir:
41
+ lora_r: 128
42
+ lora_alpha: 64
43
+ lora_dropout: 0.05
44
+ lora_target_linear: true
45
+ lora_fan_in_fan_out:
46
+
47
+ wandb_project: toxicLlama-2-13B
48
+ wandb_entity:
49
+ wandb_watch:
50
+ wandb_name:
51
+ wandb_log_model:
52
+
53
+ gradient_accumulation_steps: 1
54
+ micro_batch_size: 2
55
+ num_epochs: 2
56
+ optimizer: adamw_bnb_8bit
57
+ lr_scheduler: cosine
58
+ learning_rate: 0.0002
59
+ eval_batch_size: 2
60
+
61
+ train_on_inputs: false
62
+ group_by_length: false
63
+ bf16: true
64
+ fp16: false
65
+ tf32: false
66
+
67
+ gradient_checkpointing: true
68
+ early_stopping_patience:
69
+ resume_from_checkpoint:
70
+ local_rank:
71
+ logging_steps: 1
72
+ xformers_attention:
73
+ flash_attention: true
74
+
75
+ warmup_steps: 10
76
+ evals_per_epoch: 4
77
+ eval_table_size:
78
+ eval_table_max_new_tokens: 128
79
+ saves_per_epoch: 1
80
+ debug:
81
+ deepspeed:
82
+ weight_decay: 0.0
83
+ fsdp:
84
+ fsdp_config:
85
+ special_tokens:
86
+ bos_token: "<s>"
87
+ eos_token: "</s>"
88
+ unk_token: "<unk>"
89
+
90
+ ```
91
+
92
+ </details><br>
93
+
94
+ # lora-out
95
+
96
+ This model is a fine-tuned version of [NousResearch/Llama-2-7b-hf](https://huggingface.co/NousResearch/Llama-2-7b-hf) on the None dataset.
97
+ It achieves the following results on the evaluation set:
98
+ - Loss: 0.8100
99
+
100
+ ## Model description
101
+
102
+ More information needed
103
+
104
+ ## Intended uses & limitations
105
+
106
+ More information needed
107
+
108
+ ## Training and evaluation data
109
+
110
+ More information needed
111
+
112
+ ## Training procedure
113
+
114
+ ### Training hyperparameters
115
+
116
+ The following hyperparameters were used during training:
117
+ - learning_rate: 0.0002
118
+ - train_batch_size: 2
119
+ - eval_batch_size: 2
120
+ - seed: 42
121
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
122
+ - lr_scheduler_type: cosine
123
+ - lr_scheduler_warmup_steps: 10
124
+ - num_epochs: 2
125
+
126
+ ### Training results
127
+
128
+ | Training Loss | Epoch | Step | Validation Loss |
129
+ |:-------------:|:-----:|:----:|:---------------:|
130
+ | 1.0748 | 0.0 | 1 | 1.1154 |
131
+ | 0.8635 | 0.25 | 176 | 0.8732 |
132
+ | 0.8284 | 0.5 | 352 | 0.8463 |
133
+ | 0.7928 | 0.75 | 528 | 0.8295 |
134
+ | 0.8313 | 1.0 | 704 | 0.8155 |
135
+ | 0.6694 | 1.23 | 880 | 0.8196 |
136
+ | 0.636 | 1.48 | 1056 | 0.8144 |
137
+ | 0.6842 | 1.73 | 1232 | 0.8105 |
138
+ | 0.6277 | 1.98 | 1408 | 0.8100 |
139
+
140
+
141
+ ### Framework versions
142
+
143
+ - Transformers 4.36.2
144
+ - Pytorch 2.0.1+cu118
145
+ - Datasets 2.16.1
146
+ - Tokenizers 0.15.0
147
+ ## Training procedure
148
+
149
+
150
+ The following `bitsandbytes` quantization config was used during training:
151
+ - quant_method: bitsandbytes
152
+ - load_in_8bit: True
153
+ - load_in_4bit: False
154
+ - llm_int8_threshold: 6.0
155
+ - llm_int8_skip_modules: None
156
+ - llm_int8_enable_fp32_cpu_offload: False
157
+ - llm_int8_has_fp16_weight: False
158
+ - bnb_4bit_quant_type: fp4
159
+ - bnb_4bit_use_double_quant: False
160
+ - bnb_4bit_compute_dtype: float32
161
+
162
+ ### Framework versions
163
+
164
+
165
+ - PEFT 0.6.0
adapter_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "NousResearch/Llama-2-7b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 64,
12
+ "lora_dropout": 0.05,
13
+ "modules_to_save": null,
14
+ "peft_type": "LORA",
15
+ "r": 128,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
+ "target_modules": [
19
+ "v_proj",
20
+ "down_proj",
21
+ "o_proj",
22
+ "k_proj",
23
+ "up_proj",
24
+ "q_proj",
25
+ "gate_proj"
26
+ ],
27
+ "task_type": "CAUSAL_LM"
28
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12305da53ae6196afdc92ca1b3775df38d729cff6284920ef8ddaddbe8d29ca9
3
+ size 1279424269
checkpoint-1408/README.md ADDED
@@ -0,0 +1,220 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: NousResearch/Llama-2-7b-hf
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+
201
+
202
+ ## Training procedure
203
+
204
+
205
+ The following `bitsandbytes` quantization config was used during training:
206
+ - quant_method: bitsandbytes
207
+ - load_in_8bit: True
208
+ - load_in_4bit: False
209
+ - llm_int8_threshold: 6.0
210
+ - llm_int8_skip_modules: None
211
+ - llm_int8_enable_fp32_cpu_offload: False
212
+ - llm_int8_has_fp16_weight: False
213
+ - bnb_4bit_quant_type: fp4
214
+ - bnb_4bit_use_double_quant: False
215
+ - bnb_4bit_compute_dtype: float32
216
+
217
+ ### Framework versions
218
+
219
+
220
+ - PEFT 0.6.0
checkpoint-1408/adapter_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "NousResearch/Llama-2-7b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 64,
12
+ "lora_dropout": 0.05,
13
+ "modules_to_save": null,
14
+ "peft_type": "LORA",
15
+ "r": 128,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
+ "target_modules": [
19
+ "v_proj",
20
+ "down_proj",
21
+ "o_proj",
22
+ "k_proj",
23
+ "up_proj",
24
+ "q_proj",
25
+ "gate_proj"
26
+ ],
27
+ "task_type": "CAUSAL_LM"
28
+ }
checkpoint-1408/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c7579d180085330879e781afc3f4f0154a180aebcd957065dd9378c0e44c123
3
+ size 1279323952
checkpoint-1408/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f448999b92527df1e93cb303365bc82057f51648ac97cbef4c1ba6357beaa99c
3
+ size 641407583
checkpoint-1408/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:de8ecce3856ee8a5a019f1bd95ec82d256e81cf54c2ecca1a587e5e1e730a0ae
3
+ size 14575
checkpoint-1408/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7a8648e8a6deb1ff1aeae517c966f755d4197ce2169c5f9eb756f7cdd2d5963
3
+ size 627
checkpoint-1408/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1408/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdfce1d53b772a89b2df019b8f67a3ae168e5ca0b13341505c6bdac951bcb1c7
3
+ size 4731
checkpoint-704/README.md ADDED
@@ -0,0 +1,220 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: NousResearch/Llama-2-7b-hf
4
+ ---
5
+
6
+ # Model Card for Model ID
7
+
8
+ <!-- Provide a quick summary of what the model is/does. -->
9
+
10
+
11
+
12
+ ## Model Details
13
+
14
+ ### Model Description
15
+
16
+ <!-- Provide a longer summary of what this model is. -->
17
+
18
+
19
+
20
+ - **Developed by:** [More Information Needed]
21
+ - **Funded by [optional]:** [More Information Needed]
22
+ - **Shared by [optional]:** [More Information Needed]
23
+ - **Model type:** [More Information Needed]
24
+ - **Language(s) (NLP):** [More Information Needed]
25
+ - **License:** [More Information Needed]
26
+ - **Finetuned from model [optional]:** [More Information Needed]
27
+
28
+ ### Model Sources [optional]
29
+
30
+ <!-- Provide the basic links for the model. -->
31
+
32
+ - **Repository:** [More Information Needed]
33
+ - **Paper [optional]:** [More Information Needed]
34
+ - **Demo [optional]:** [More Information Needed]
35
+
36
+ ## Uses
37
+
38
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
+
40
+ ### Direct Use
41
+
42
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
+
44
+ [More Information Needed]
45
+
46
+ ### Downstream Use [optional]
47
+
48
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Out-of-Scope Use
53
+
54
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
+
56
+ [More Information Needed]
57
+
58
+ ## Bias, Risks, and Limitations
59
+
60
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ### Recommendations
65
+
66
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
+
68
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
+
70
+ ## How to Get Started with the Model
71
+
72
+ Use the code below to get started with the model.
73
+
74
+ [More Information Needed]
75
+
76
+ ## Training Details
77
+
78
+ ### Training Data
79
+
80
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
81
+
82
+ [More Information Needed]
83
+
84
+ ### Training Procedure
85
+
86
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
+
88
+ #### Preprocessing [optional]
89
+
90
+ [More Information Needed]
91
+
92
+
93
+ #### Training Hyperparameters
94
+
95
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
+
97
+ #### Speeds, Sizes, Times [optional]
98
+
99
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
+
101
+ [More Information Needed]
102
+
103
+ ## Evaluation
104
+
105
+ <!-- This section describes the evaluation protocols and provides the results. -->
106
+
107
+ ### Testing Data, Factors & Metrics
108
+
109
+ #### Testing Data
110
+
111
+ <!-- This should link to a Dataset Card if possible. -->
112
+
113
+ [More Information Needed]
114
+
115
+ #### Factors
116
+
117
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Metrics
122
+
123
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
+
125
+ [More Information Needed]
126
+
127
+ ### Results
128
+
129
+ [More Information Needed]
130
+
131
+ #### Summary
132
+
133
+
134
+
135
+ ## Model Examination [optional]
136
+
137
+ <!-- Relevant interpretability work for the model goes here -->
138
+
139
+ [More Information Needed]
140
+
141
+ ## Environmental Impact
142
+
143
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
+
145
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
+
147
+ - **Hardware Type:** [More Information Needed]
148
+ - **Hours used:** [More Information Needed]
149
+ - **Cloud Provider:** [More Information Needed]
150
+ - **Compute Region:** [More Information Needed]
151
+ - **Carbon Emitted:** [More Information Needed]
152
+
153
+ ## Technical Specifications [optional]
154
+
155
+ ### Model Architecture and Objective
156
+
157
+ [More Information Needed]
158
+
159
+ ### Compute Infrastructure
160
+
161
+ [More Information Needed]
162
+
163
+ #### Hardware
164
+
165
+ [More Information Needed]
166
+
167
+ #### Software
168
+
169
+ [More Information Needed]
170
+
171
+ ## Citation [optional]
172
+
173
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
+
175
+ **BibTeX:**
176
+
177
+ [More Information Needed]
178
+
179
+ **APA:**
180
+
181
+ [More Information Needed]
182
+
183
+ ## Glossary [optional]
184
+
185
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
+
187
+ [More Information Needed]
188
+
189
+ ## More Information [optional]
190
+
191
+ [More Information Needed]
192
+
193
+ ## Model Card Authors [optional]
194
+
195
+ [More Information Needed]
196
+
197
+ ## Model Card Contact
198
+
199
+ [More Information Needed]
200
+
201
+
202
+ ## Training procedure
203
+
204
+
205
+ The following `bitsandbytes` quantization config was used during training:
206
+ - quant_method: bitsandbytes
207
+ - load_in_8bit: True
208
+ - load_in_4bit: False
209
+ - llm_int8_threshold: 6.0
210
+ - llm_int8_skip_modules: None
211
+ - llm_int8_enable_fp32_cpu_offload: False
212
+ - llm_int8_has_fp16_weight: False
213
+ - bnb_4bit_quant_type: fp4
214
+ - bnb_4bit_use_double_quant: False
215
+ - bnb_4bit_compute_dtype: float32
216
+
217
+ ### Framework versions
218
+
219
+
220
+ - PEFT 0.6.0
checkpoint-704/adapter_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "NousResearch/Llama-2-7b-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "lora_alpha": 64,
12
+ "lora_dropout": 0.05,
13
+ "modules_to_save": null,
14
+ "peft_type": "LORA",
15
+ "r": 128,
16
+ "rank_pattern": {},
17
+ "revision": null,
18
+ "target_modules": [
19
+ "v_proj",
20
+ "down_proj",
21
+ "o_proj",
22
+ "k_proj",
23
+ "up_proj",
24
+ "q_proj",
25
+ "gate_proj"
26
+ ],
27
+ "task_type": "CAUSAL_LM"
28
+ }
checkpoint-704/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dca2825b1f7efc41e1543a9022a554d0daa67a6e8a16e807f1e6f4be6691b994
3
+ size 1279323952
checkpoint-704/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d79c0bc6e189337cd40d41499eba5826454bcf56ece6cac901e5db118fad6fe9
3
+ size 641407583
checkpoint-704/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4994115d4b01f4329f284d1448614f69aa7936ea49352583f0ab43440b52252e
3
+ size 14575
checkpoint-704/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e6efc38b0d24183ce289e3eb97069f90ae3001c3354b59fa84ed9c34eef7e30
3
+ size 627
checkpoint-704/trainer_state.json ADDED
@@ -0,0 +1,4285 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.8154956698417664,
3
+ "best_model_checkpoint": "./lora-out/checkpoint-704",
4
+ "epoch": 1.0,
5
+ "eval_steps": 176,
6
+ "global_step": 704,
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
+ "learning_rate": 2e-05,
14
+ "loss": 1.0748,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.0,
19
+ "eval_loss": 1.1153677701950073,
20
+ "eval_runtime": 59.6311,
21
+ "eval_samples_per_second": 6.708,
22
+ "eval_steps_per_second": 3.354,
23
+ "step": 1
24
+ },
25
+ {
26
+ "epoch": 0.0,
27
+ "learning_rate": 4e-05,
28
+ "loss": 1.1247,
29
+ "step": 2
30
+ },
31
+ {
32
+ "epoch": 0.0,
33
+ "learning_rate": 6e-05,
34
+ "loss": 1.1111,
35
+ "step": 3
36
+ },
37
+ {
38
+ "epoch": 0.01,
39
+ "learning_rate": 8e-05,
40
+ "loss": 1.0777,
41
+ "step": 4
42
+ },
43
+ {
44
+ "epoch": 0.01,
45
+ "learning_rate": 0.0001,
46
+ "loss": 1.033,
47
+ "step": 5
48
+ },
49
+ {
50
+ "epoch": 0.01,
51
+ "learning_rate": 0.00012,
52
+ "loss": 1.0021,
53
+ "step": 6
54
+ },
55
+ {
56
+ "epoch": 0.01,
57
+ "learning_rate": 0.00014,
58
+ "loss": 1.0017,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.01,
63
+ "learning_rate": 0.00016,
64
+ "loss": 1.0205,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.01,
69
+ "learning_rate": 0.00018,
70
+ "loss": 0.95,
71
+ "step": 9
72
+ },
73
+ {
74
+ "epoch": 0.01,
75
+ "learning_rate": 0.0002,
76
+ "loss": 1.0236,
77
+ "step": 10
78
+ },
79
+ {
80
+ "epoch": 0.02,
81
+ "learning_rate": 0.00019999974750358046,
82
+ "loss": 0.9656,
83
+ "step": 11
84
+ },
85
+ {
86
+ "epoch": 0.02,
87
+ "learning_rate": 0.00019999899001559682,
88
+ "loss": 0.9769,
89
+ "step": 12
90
+ },
91
+ {
92
+ "epoch": 0.02,
93
+ "learning_rate": 0.00019999772753987444,
94
+ "loss": 0.984,
95
+ "step": 13
96
+ },
97
+ {
98
+ "epoch": 0.02,
99
+ "learning_rate": 0.0001999959600827887,
100
+ "loss": 0.9227,
101
+ "step": 14
102
+ },
103
+ {
104
+ "epoch": 0.02,
105
+ "learning_rate": 0.0001999936876532651,
106
+ "loss": 0.9342,
107
+ "step": 15
108
+ },
109
+ {
110
+ "epoch": 0.02,
111
+ "learning_rate": 0.00019999091026277928,
112
+ "loss": 0.991,
113
+ "step": 16
114
+ },
115
+ {
116
+ "epoch": 0.02,
117
+ "learning_rate": 0.00019998762792535683,
118
+ "loss": 0.9632,
119
+ "step": 17
120
+ },
121
+ {
122
+ "epoch": 0.03,
123
+ "learning_rate": 0.00019998384065757335,
124
+ "loss": 0.9574,
125
+ "step": 18
126
+ },
127
+ {
128
+ "epoch": 0.03,
129
+ "learning_rate": 0.00019997954847855427,
130
+ "loss": 0.9778,
131
+ "step": 19
132
+ },
133
+ {
134
+ "epoch": 0.03,
135
+ "learning_rate": 0.00019997475140997475,
136
+ "loss": 0.8948,
137
+ "step": 20
138
+ },
139
+ {
140
+ "epoch": 0.03,
141
+ "learning_rate": 0.00019996944947605968,
142
+ "loss": 0.9359,
143
+ "step": 21
144
+ },
145
+ {
146
+ "epoch": 0.03,
147
+ "learning_rate": 0.00019996364270358346,
148
+ "loss": 0.8931,
149
+ "step": 22
150
+ },
151
+ {
152
+ "epoch": 0.03,
153
+ "learning_rate": 0.00019995733112186982,
154
+ "loss": 0.9468,
155
+ "step": 23
156
+ },
157
+ {
158
+ "epoch": 0.03,
159
+ "learning_rate": 0.00019995051476279186,
160
+ "loss": 0.9246,
161
+ "step": 24
162
+ },
163
+ {
164
+ "epoch": 0.04,
165
+ "learning_rate": 0.00019994319366077167,
166
+ "loss": 0.9223,
167
+ "step": 25
168
+ },
169
+ {
170
+ "epoch": 0.04,
171
+ "learning_rate": 0.00019993536785278032,
172
+ "loss": 0.8973,
173
+ "step": 26
174
+ },
175
+ {
176
+ "epoch": 0.04,
177
+ "learning_rate": 0.00019992703737833748,
178
+ "loss": 0.9316,
179
+ "step": 27
180
+ },
181
+ {
182
+ "epoch": 0.04,
183
+ "learning_rate": 0.0001999182022795116,
184
+ "loss": 0.9499,
185
+ "step": 28
186
+ },
187
+ {
188
+ "epoch": 0.04,
189
+ "learning_rate": 0.00019990886260091916,
190
+ "loss": 0.9069,
191
+ "step": 29
192
+ },
193
+ {
194
+ "epoch": 0.04,
195
+ "learning_rate": 0.00019989901838972496,
196
+ "loss": 0.9034,
197
+ "step": 30
198
+ },
199
+ {
200
+ "epoch": 0.04,
201
+ "learning_rate": 0.0001998886696956415,
202
+ "loss": 0.9329,
203
+ "step": 31
204
+ },
205
+ {
206
+ "epoch": 0.05,
207
+ "learning_rate": 0.000199877816570929,
208
+ "loss": 0.944,
209
+ "step": 32
210
+ },
211
+ {
212
+ "epoch": 0.05,
213
+ "learning_rate": 0.00019986645907039497,
214
+ "loss": 0.899,
215
+ "step": 33
216
+ },
217
+ {
218
+ "epoch": 0.05,
219
+ "learning_rate": 0.0001998545972513939,
220
+ "loss": 0.8994,
221
+ "step": 34
222
+ },
223
+ {
224
+ "epoch": 0.05,
225
+ "learning_rate": 0.00019984223117382714,
226
+ "loss": 0.88,
227
+ "step": 35
228
+ },
229
+ {
230
+ "epoch": 0.05,
231
+ "learning_rate": 0.00019982936090014256,
232
+ "loss": 0.9321,
233
+ "step": 36
234
+ },
235
+ {
236
+ "epoch": 0.05,
237
+ "learning_rate": 0.0001998159864953341,
238
+ "loss": 0.9437,
239
+ "step": 37
240
+ },
241
+ {
242
+ "epoch": 0.05,
243
+ "learning_rate": 0.0001998021080269415,
244
+ "loss": 0.8879,
245
+ "step": 38
246
+ },
247
+ {
248
+ "epoch": 0.06,
249
+ "learning_rate": 0.0001997877255650501,
250
+ "loss": 0.8921,
251
+ "step": 39
252
+ },
253
+ {
254
+ "epoch": 0.06,
255
+ "learning_rate": 0.00019977283918229022,
256
+ "loss": 0.8943,
257
+ "step": 40
258
+ },
259
+ {
260
+ "epoch": 0.06,
261
+ "learning_rate": 0.00019975744895383706,
262
+ "loss": 0.8991,
263
+ "step": 41
264
+ },
265
+ {
266
+ "epoch": 0.06,
267
+ "learning_rate": 0.00019974155495741024,
268
+ "loss": 0.8881,
269
+ "step": 42
270
+ },
271
+ {
272
+ "epoch": 0.06,
273
+ "learning_rate": 0.0001997251572732732,
274
+ "loss": 0.9018,
275
+ "step": 43
276
+ },
277
+ {
278
+ "epoch": 0.06,
279
+ "learning_rate": 0.00019970825598423315,
280
+ "loss": 0.8722,
281
+ "step": 44
282
+ },
283
+ {
284
+ "epoch": 0.06,
285
+ "learning_rate": 0.00019969085117564034,
286
+ "loss": 0.9355,
287
+ "step": 45
288
+ },
289
+ {
290
+ "epoch": 0.07,
291
+ "learning_rate": 0.0001996729429353878,
292
+ "loss": 0.9524,
293
+ "step": 46
294
+ },
295
+ {
296
+ "epoch": 0.07,
297
+ "learning_rate": 0.0001996545313539109,
298
+ "loss": 0.9006,
299
+ "step": 47
300
+ },
301
+ {
302
+ "epoch": 0.07,
303
+ "learning_rate": 0.00019963561652418683,
304
+ "loss": 0.869,
305
+ "step": 48
306
+ },
307
+ {
308
+ "epoch": 0.07,
309
+ "learning_rate": 0.0001996161985417341,
310
+ "loss": 0.8397,
311
+ "step": 49
312
+ },
313
+ {
314
+ "epoch": 0.07,
315
+ "learning_rate": 0.00019959627750461208,
316
+ "loss": 0.8915,
317
+ "step": 50
318
+ },
319
+ {
320
+ "epoch": 0.07,
321
+ "learning_rate": 0.0001995758535134206,
322
+ "loss": 0.8423,
323
+ "step": 51
324
+ },
325
+ {
326
+ "epoch": 0.07,
327
+ "learning_rate": 0.0001995549266712994,
328
+ "loss": 0.8994,
329
+ "step": 52
330
+ },
331
+ {
332
+ "epoch": 0.08,
333
+ "learning_rate": 0.00019953349708392752,
334
+ "loss": 0.8939,
335
+ "step": 53
336
+ },
337
+ {
338
+ "epoch": 0.08,
339
+ "learning_rate": 0.0001995115648595228,
340
+ "loss": 0.93,
341
+ "step": 54
342
+ },
343
+ {
344
+ "epoch": 0.08,
345
+ "learning_rate": 0.00019948913010884147,
346
+ "loss": 0.8913,
347
+ "step": 55
348
+ },
349
+ {
350
+ "epoch": 0.08,
351
+ "learning_rate": 0.00019946619294517736,
352
+ "loss": 0.8927,
353
+ "step": 56
354
+ },
355
+ {
356
+ "epoch": 0.08,
357
+ "learning_rate": 0.00019944275348436153,
358
+ "loss": 0.875,
359
+ "step": 57
360
+ },
361
+ {
362
+ "epoch": 0.08,
363
+ "learning_rate": 0.00019941881184476154,
364
+ "loss": 0.8199,
365
+ "step": 58
366
+ },
367
+ {
368
+ "epoch": 0.08,
369
+ "learning_rate": 0.000199394368147281,
370
+ "loss": 0.9137,
371
+ "step": 59
372
+ },
373
+ {
374
+ "epoch": 0.09,
375
+ "learning_rate": 0.00019936942251535882,
376
+ "loss": 0.8572,
377
+ "step": 60
378
+ },
379
+ {
380
+ "epoch": 0.09,
381
+ "learning_rate": 0.00019934397507496865,
382
+ "loss": 0.9061,
383
+ "step": 61
384
+ },
385
+ {
386
+ "epoch": 0.09,
387
+ "learning_rate": 0.00019931802595461826,
388
+ "loss": 0.9258,
389
+ "step": 62
390
+ },
391
+ {
392
+ "epoch": 0.09,
393
+ "learning_rate": 0.0001992915752853488,
394
+ "loss": 0.945,
395
+ "step": 63
396
+ },
397
+ {
398
+ "epoch": 0.09,
399
+ "learning_rate": 0.00019926462320073429,
400
+ "loss": 0.8826,
401
+ "step": 64
402
+ },
403
+ {
404
+ "epoch": 0.09,
405
+ "learning_rate": 0.00019923716983688086,
406
+ "loss": 0.8929,
407
+ "step": 65
408
+ },
409
+ {
410
+ "epoch": 0.09,
411
+ "learning_rate": 0.00019920921533242596,
412
+ "loss": 0.8848,
413
+ "step": 66
414
+ },
415
+ {
416
+ "epoch": 0.1,
417
+ "learning_rate": 0.00019918075982853793,
418
+ "loss": 0.9109,
419
+ "step": 67
420
+ },
421
+ {
422
+ "epoch": 0.1,
423
+ "learning_rate": 0.0001991518034689149,
424
+ "loss": 0.9073,
425
+ "step": 68
426
+ },
427
+ {
428
+ "epoch": 0.1,
429
+ "learning_rate": 0.00019912234639978454,
430
+ "loss": 0.9215,
431
+ "step": 69
432
+ },
433
+ {
434
+ "epoch": 0.1,
435
+ "learning_rate": 0.00019909238876990285,
436
+ "loss": 0.8772,
437
+ "step": 70
438
+ },
439
+ {
440
+ "epoch": 0.1,
441
+ "learning_rate": 0.00019906193073055374,
442
+ "loss": 0.8677,
443
+ "step": 71
444
+ },
445
+ {
446
+ "epoch": 0.1,
447
+ "learning_rate": 0.00019903097243554815,
448
+ "loss": 0.8607,
449
+ "step": 72
450
+ },
451
+ {
452
+ "epoch": 0.1,
453
+ "learning_rate": 0.00019899951404122327,
454
+ "loss": 0.8926,
455
+ "step": 73
456
+ },
457
+ {
458
+ "epoch": 0.11,
459
+ "learning_rate": 0.00019896755570644165,
460
+ "loss": 0.8746,
461
+ "step": 74
462
+ },
463
+ {
464
+ "epoch": 0.11,
465
+ "learning_rate": 0.00019893509759259068,
466
+ "loss": 0.8866,
467
+ "step": 75
468
+ },
469
+ {
470
+ "epoch": 0.11,
471
+ "learning_rate": 0.00019890213986358148,
472
+ "loss": 0.853,
473
+ "step": 76
474
+ },
475
+ {
476
+ "epoch": 0.11,
477
+ "learning_rate": 0.00019886868268584822,
478
+ "loss": 0.8957,
479
+ "step": 77
480
+ },
481
+ {
482
+ "epoch": 0.11,
483
+ "learning_rate": 0.00019883472622834723,
484
+ "loss": 0.9247,
485
+ "step": 78
486
+ },
487
+ {
488
+ "epoch": 0.11,
489
+ "learning_rate": 0.00019880027066255623,
490
+ "loss": 0.8831,
491
+ "step": 79
492
+ },
493
+ {
494
+ "epoch": 0.11,
495
+ "learning_rate": 0.00019876531616247337,
496
+ "loss": 0.8998,
497
+ "step": 80
498
+ },
499
+ {
500
+ "epoch": 0.12,
501
+ "learning_rate": 0.00019872986290461633,
502
+ "loss": 0.8242,
503
+ "step": 81
504
+ },
505
+ {
506
+ "epoch": 0.12,
507
+ "learning_rate": 0.00019869391106802154,
508
+ "loss": 0.8815,
509
+ "step": 82
510
+ },
511
+ {
512
+ "epoch": 0.12,
513
+ "learning_rate": 0.00019865746083424317,
514
+ "loss": 0.8705,
515
+ "step": 83
516
+ },
517
+ {
518
+ "epoch": 0.12,
519
+ "learning_rate": 0.00019862051238735232,
520
+ "loss": 0.8767,
521
+ "step": 84
522
+ },
523
+ {
524
+ "epoch": 0.12,
525
+ "learning_rate": 0.00019858306591393602,
526
+ "loss": 0.9376,
527
+ "step": 85
528
+ },
529
+ {
530
+ "epoch": 0.12,
531
+ "learning_rate": 0.00019854512160309625,
532
+ "loss": 0.875,
533
+ "step": 86
534
+ },
535
+ {
536
+ "epoch": 0.12,
537
+ "learning_rate": 0.00019850667964644908,
538
+ "loss": 0.8512,
539
+ "step": 87
540
+ },
541
+ {
542
+ "epoch": 0.12,
543
+ "learning_rate": 0.00019846774023812364,
544
+ "loss": 0.8755,
545
+ "step": 88
546
+ },
547
+ {
548
+ "epoch": 0.13,
549
+ "learning_rate": 0.00019842830357476112,
550
+ "loss": 0.8862,
551
+ "step": 89
552
+ },
553
+ {
554
+ "epoch": 0.13,
555
+ "learning_rate": 0.0001983883698555139,
556
+ "loss": 0.8209,
557
+ "step": 90
558
+ },
559
+ {
560
+ "epoch": 0.13,
561
+ "learning_rate": 0.00019834793928204435,
562
+ "loss": 0.8633,
563
+ "step": 91
564
+ },
565
+ {
566
+ "epoch": 0.13,
567
+ "learning_rate": 0.000198307012058524,
568
+ "loss": 0.8421,
569
+ "step": 92
570
+ },
571
+ {
572
+ "epoch": 0.13,
573
+ "learning_rate": 0.00019826558839163236,
574
+ "loss": 0.8771,
575
+ "step": 93
576
+ },
577
+ {
578
+ "epoch": 0.13,
579
+ "learning_rate": 0.00019822366849055602,
580
+ "loss": 0.8392,
581
+ "step": 94
582
+ },
583
+ {
584
+ "epoch": 0.13,
585
+ "learning_rate": 0.0001981812525669875,
586
+ "loss": 0.8565,
587
+ "step": 95
588
+ },
589
+ {
590
+ "epoch": 0.14,
591
+ "learning_rate": 0.00019813834083512414,
592
+ "loss": 0.8965,
593
+ "step": 96
594
+ },
595
+ {
596
+ "epoch": 0.14,
597
+ "learning_rate": 0.00019809493351166711,
598
+ "loss": 0.8669,
599
+ "step": 97
600
+ },
601
+ {
602
+ "epoch": 0.14,
603
+ "learning_rate": 0.0001980510308158203,
604
+ "loss": 0.8476,
605
+ "step": 98
606
+ },
607
+ {
608
+ "epoch": 0.14,
609
+ "learning_rate": 0.00019800663296928918,
610
+ "loss": 0.8604,
611
+ "step": 99
612
+ },
613
+ {
614
+ "epoch": 0.14,
615
+ "learning_rate": 0.0001979617401962797,
616
+ "loss": 0.8687,
617
+ "step": 100
618
+ },
619
+ {
620
+ "epoch": 0.14,
621
+ "learning_rate": 0.0001979163527234971,
622
+ "loss": 0.8716,
623
+ "step": 101
624
+ },
625
+ {
626
+ "epoch": 0.14,
627
+ "learning_rate": 0.00019787047078014496,
628
+ "loss": 0.8727,
629
+ "step": 102
630
+ },
631
+ {
632
+ "epoch": 0.15,
633
+ "learning_rate": 0.00019782409459792371,
634
+ "loss": 0.8484,
635
+ "step": 103
636
+ },
637
+ {
638
+ "epoch": 0.15,
639
+ "learning_rate": 0.00019777722441102985,
640
+ "loss": 0.8811,
641
+ "step": 104
642
+ },
643
+ {
644
+ "epoch": 0.15,
645
+ "learning_rate": 0.00019772986045615438,
646
+ "loss": 0.9194,
647
+ "step": 105
648
+ },
649
+ {
650
+ "epoch": 0.15,
651
+ "learning_rate": 0.00019768200297248193,
652
+ "loss": 0.8592,
653
+ "step": 106
654
+ },
655
+ {
656
+ "epoch": 0.15,
657
+ "learning_rate": 0.0001976336522016893,
658
+ "loss": 0.856,
659
+ "step": 107
660
+ },
661
+ {
662
+ "epoch": 0.15,
663
+ "learning_rate": 0.00019758480838794453,
664
+ "loss": 0.8841,
665
+ "step": 108
666
+ },
667
+ {
668
+ "epoch": 0.15,
669
+ "learning_rate": 0.0001975354717779053,
670
+ "loss": 0.8962,
671
+ "step": 109
672
+ },
673
+ {
674
+ "epoch": 0.16,
675
+ "learning_rate": 0.000197485642620718,
676
+ "loss": 0.8415,
677
+ "step": 110
678
+ },
679
+ {
680
+ "epoch": 0.16,
681
+ "learning_rate": 0.00019743532116801624,
682
+ "loss": 0.8843,
683
+ "step": 111
684
+ },
685
+ {
686
+ "epoch": 0.16,
687
+ "learning_rate": 0.0001973845076739198,
688
+ "loss": 0.8918,
689
+ "step": 112
690
+ },
691
+ {
692
+ "epoch": 0.16,
693
+ "learning_rate": 0.00019733320239503322,
694
+ "loss": 0.836,
695
+ "step": 113
696
+ },
697
+ {
698
+ "epoch": 0.16,
699
+ "learning_rate": 0.00019728140559044445,
700
+ "loss": 0.8442,
701
+ "step": 114
702
+ },
703
+ {
704
+ "epoch": 0.16,
705
+ "learning_rate": 0.00019722911752172363,
706
+ "loss": 0.8471,
707
+ "step": 115
708
+ },
709
+ {
710
+ "epoch": 0.16,
711
+ "learning_rate": 0.00019717633845292175,
712
+ "loss": 0.8407,
713
+ "step": 116
714
+ },
715
+ {
716
+ "epoch": 0.17,
717
+ "learning_rate": 0.00019712306865056936,
718
+ "loss": 0.8982,
719
+ "step": 117
720
+ },
721
+ {
722
+ "epoch": 0.17,
723
+ "learning_rate": 0.00019706930838367517,
724
+ "loss": 0.8233,
725
+ "step": 118
726
+ },
727
+ {
728
+ "epoch": 0.17,
729
+ "learning_rate": 0.0001970150579237246,
730
+ "loss": 0.8739,
731
+ "step": 119
732
+ },
733
+ {
734
+ "epoch": 0.17,
735
+ "learning_rate": 0.0001969603175446787,
736
+ "loss": 0.8956,
737
+ "step": 120
738
+ },
739
+ {
740
+ "epoch": 0.17,
741
+ "learning_rate": 0.00019690508752297234,
742
+ "loss": 0.8368,
743
+ "step": 121
744
+ },
745
+ {
746
+ "epoch": 0.17,
747
+ "learning_rate": 0.00019684936813751326,
748
+ "loss": 0.8628,
749
+ "step": 122
750
+ },
751
+ {
752
+ "epoch": 0.17,
753
+ "learning_rate": 0.00019679315966968035,
754
+ "loss": 0.8485,
755
+ "step": 123
756
+ },
757
+ {
758
+ "epoch": 0.18,
759
+ "learning_rate": 0.00019673646240332232,
760
+ "loss": 0.8465,
761
+ "step": 124
762
+ },
763
+ {
764
+ "epoch": 0.18,
765
+ "learning_rate": 0.00019667927662475636,
766
+ "loss": 0.8258,
767
+ "step": 125
768
+ },
769
+ {
770
+ "epoch": 0.18,
771
+ "learning_rate": 0.0001966216026227665,
772
+ "loss": 0.8668,
773
+ "step": 126
774
+ },
775
+ {
776
+ "epoch": 0.18,
777
+ "learning_rate": 0.00019656344068860233,
778
+ "loss": 0.8556,
779
+ "step": 127
780
+ },
781
+ {
782
+ "epoch": 0.18,
783
+ "learning_rate": 0.00019650479111597748,
784
+ "loss": 0.8997,
785
+ "step": 128
786
+ },
787
+ {
788
+ "epoch": 0.18,
789
+ "learning_rate": 0.00019644565420106805,
790
+ "loss": 0.8958,
791
+ "step": 129
792
+ },
793
+ {
794
+ "epoch": 0.18,
795
+ "learning_rate": 0.0001963860302425113,
796
+ "loss": 0.8866,
797
+ "step": 130
798
+ },
799
+ {
800
+ "epoch": 0.19,
801
+ "learning_rate": 0.00019632591954140387,
802
+ "loss": 0.8723,
803
+ "step": 131
804
+ },
805
+ {
806
+ "epoch": 0.19,
807
+ "learning_rate": 0.00019626532240130055,
808
+ "loss": 0.8614,
809
+ "step": 132
810
+ },
811
+ {
812
+ "epoch": 0.19,
813
+ "learning_rate": 0.00019620423912821252,
814
+ "loss": 0.8564,
815
+ "step": 133
816
+ },
817
+ {
818
+ "epoch": 0.19,
819
+ "learning_rate": 0.00019614267003060593,
820
+ "loss": 0.8998,
821
+ "step": 134
822
+ },
823
+ {
824
+ "epoch": 0.19,
825
+ "learning_rate": 0.00019608061541940037,
826
+ "loss": 0.8743,
827
+ "step": 135
828
+ },
829
+ {
830
+ "epoch": 0.19,
831
+ "learning_rate": 0.00019601807560796713,
832
+ "loss": 0.8084,
833
+ "step": 136
834
+ },
835
+ {
836
+ "epoch": 0.19,
837
+ "learning_rate": 0.00019595505091212783,
838
+ "loss": 0.8503,
839
+ "step": 137
840
+ },
841
+ {
842
+ "epoch": 0.2,
843
+ "learning_rate": 0.0001958915416501526,
844
+ "loss": 0.8093,
845
+ "step": 138
846
+ },
847
+ {
848
+ "epoch": 0.2,
849
+ "learning_rate": 0.00019582754814275873,
850
+ "loss": 0.8413,
851
+ "step": 139
852
+ },
853
+ {
854
+ "epoch": 0.2,
855
+ "learning_rate": 0.00019576307071310882,
856
+ "loss": 0.9218,
857
+ "step": 140
858
+ },
859
+ {
860
+ "epoch": 0.2,
861
+ "learning_rate": 0.00019569810968680926,
862
+ "loss": 0.8248,
863
+ "step": 141
864
+ },
865
+ {
866
+ "epoch": 0.2,
867
+ "learning_rate": 0.00019563266539190862,
868
+ "loss": 0.8592,
869
+ "step": 142
870
+ },
871
+ {
872
+ "epoch": 0.2,
873
+ "learning_rate": 0.00019556673815889587,
874
+ "loss": 0.8835,
875
+ "step": 143
876
+ },
877
+ {
878
+ "epoch": 0.2,
879
+ "learning_rate": 0.00019550032832069882,
880
+ "loss": 0.8353,
881
+ "step": 144
882
+ },
883
+ {
884
+ "epoch": 0.21,
885
+ "learning_rate": 0.00019543343621268244,
886
+ "loss": 0.8212,
887
+ "step": 145
888
+ },
889
+ {
890
+ "epoch": 0.21,
891
+ "learning_rate": 0.000195366062172647,
892
+ "loss": 0.8606,
893
+ "step": 146
894
+ },
895
+ {
896
+ "epoch": 0.21,
897
+ "learning_rate": 0.00019529820654082665,
898
+ "loss": 0.8585,
899
+ "step": 147
900
+ },
901
+ {
902
+ "epoch": 0.21,
903
+ "learning_rate": 0.00019522986965988745,
904
+ "loss": 0.9144,
905
+ "step": 148
906
+ },
907
+ {
908
+ "epoch": 0.21,
909
+ "learning_rate": 0.00019516105187492575,
910
+ "loss": 0.9081,
911
+ "step": 149
912
+ },
913
+ {
914
+ "epoch": 0.21,
915
+ "learning_rate": 0.00019509175353346644,
916
+ "loss": 0.823,
917
+ "step": 150
918
+ },
919
+ {
920
+ "epoch": 0.21,
921
+ "learning_rate": 0.0001950219749854612,
922
+ "loss": 0.8732,
923
+ "step": 151
924
+ },
925
+ {
926
+ "epoch": 0.22,
927
+ "learning_rate": 0.00019495171658328664,
928
+ "loss": 0.8625,
929
+ "step": 152
930
+ },
931
+ {
932
+ "epoch": 0.22,
933
+ "learning_rate": 0.00019488097868174275,
934
+ "loss": 0.8403,
935
+ "step": 153
936
+ },
937
+ {
938
+ "epoch": 0.22,
939
+ "learning_rate": 0.00019480976163805078,
940
+ "loss": 0.8427,
941
+ "step": 154
942
+ },
943
+ {
944
+ "epoch": 0.22,
945
+ "learning_rate": 0.00019473806581185175,
946
+ "loss": 0.8417,
947
+ "step": 155
948
+ },
949
+ {
950
+ "epoch": 0.22,
951
+ "learning_rate": 0.00019466589156520448,
952
+ "loss": 0.8334,
953
+ "step": 156
954
+ },
955
+ {
956
+ "epoch": 0.22,
957
+ "learning_rate": 0.00019459323926258366,
958
+ "loss": 0.8514,
959
+ "step": 157
960
+ },
961
+ {
962
+ "epoch": 0.22,
963
+ "learning_rate": 0.00019452010927087826,
964
+ "loss": 0.7686,
965
+ "step": 158
966
+ },
967
+ {
968
+ "epoch": 0.23,
969
+ "learning_rate": 0.00019444650195938953,
970
+ "loss": 0.8638,
971
+ "step": 159
972
+ },
973
+ {
974
+ "epoch": 0.23,
975
+ "learning_rate": 0.00019437241769982907,
976
+ "loss": 0.8867,
977
+ "step": 160
978
+ },
979
+ {
980
+ "epoch": 0.23,
981
+ "learning_rate": 0.00019429785686631714,
982
+ "loss": 0.8116,
983
+ "step": 161
984
+ },
985
+ {
986
+ "epoch": 0.23,
987
+ "learning_rate": 0.00019422281983538054,
988
+ "loss": 0.8035,
989
+ "step": 162
990
+ },
991
+ {
992
+ "epoch": 0.23,
993
+ "learning_rate": 0.000194147306985951,
994
+ "loss": 0.8244,
995
+ "step": 163
996
+ },
997
+ {
998
+ "epoch": 0.23,
999
+ "learning_rate": 0.0001940713186993629,
1000
+ "loss": 0.8609,
1001
+ "step": 164
1002
+ },
1003
+ {
1004
+ "epoch": 0.23,
1005
+ "learning_rate": 0.00019399485535935172,
1006
+ "loss": 0.8357,
1007
+ "step": 165
1008
+ },
1009
+ {
1010
+ "epoch": 0.24,
1011
+ "learning_rate": 0.00019391791735205182,
1012
+ "loss": 0.8386,
1013
+ "step": 166
1014
+ },
1015
+ {
1016
+ "epoch": 0.24,
1017
+ "learning_rate": 0.00019384050506599462,
1018
+ "loss": 0.8402,
1019
+ "step": 167
1020
+ },
1021
+ {
1022
+ "epoch": 0.24,
1023
+ "learning_rate": 0.00019376261889210664,
1024
+ "loss": 0.8348,
1025
+ "step": 168
1026
+ },
1027
+ {
1028
+ "epoch": 0.24,
1029
+ "learning_rate": 0.00019368425922370748,
1030
+ "loss": 0.8332,
1031
+ "step": 169
1032
+ },
1033
+ {
1034
+ "epoch": 0.24,
1035
+ "learning_rate": 0.00019360542645650784,
1036
+ "loss": 0.8906,
1037
+ "step": 170
1038
+ },
1039
+ {
1040
+ "epoch": 0.24,
1041
+ "learning_rate": 0.00019352612098860755,
1042
+ "loss": 0.7958,
1043
+ "step": 171
1044
+ },
1045
+ {
1046
+ "epoch": 0.24,
1047
+ "learning_rate": 0.00019344634322049356,
1048
+ "loss": 0.8493,
1049
+ "step": 172
1050
+ },
1051
+ {
1052
+ "epoch": 0.25,
1053
+ "learning_rate": 0.00019336609355503787,
1054
+ "loss": 0.8244,
1055
+ "step": 173
1056
+ },
1057
+ {
1058
+ "epoch": 0.25,
1059
+ "learning_rate": 0.00019328537239749553,
1060
+ "loss": 0.8464,
1061
+ "step": 174
1062
+ },
1063
+ {
1064
+ "epoch": 0.25,
1065
+ "learning_rate": 0.00019320418015550265,
1066
+ "loss": 0.8518,
1067
+ "step": 175
1068
+ },
1069
+ {
1070
+ "epoch": 0.25,
1071
+ "learning_rate": 0.00019312251723907422,
1072
+ "loss": 0.8635,
1073
+ "step": 176
1074
+ },
1075
+ {
1076
+ "epoch": 0.25,
1077
+ "eval_loss": 0.873156726360321,
1078
+ "eval_runtime": 58.1152,
1079
+ "eval_samples_per_second": 6.883,
1080
+ "eval_steps_per_second": 3.441,
1081
+ "step": 176
1082
+ },
1083
+ {
1084
+ "epoch": 0.25,
1085
+ "learning_rate": 0.0001930403840606021,
1086
+ "loss": 0.9014,
1087
+ "step": 177
1088
+ },
1089
+ {
1090
+ "epoch": 0.25,
1091
+ "learning_rate": 0.00019295778103485298,
1092
+ "loss": 0.8353,
1093
+ "step": 178
1094
+ },
1095
+ {
1096
+ "epoch": 0.25,
1097
+ "learning_rate": 0.00019287470857896622,
1098
+ "loss": 0.8487,
1099
+ "step": 179
1100
+ },
1101
+ {
1102
+ "epoch": 0.26,
1103
+ "learning_rate": 0.00019279116711245177,
1104
+ "loss": 0.8616,
1105
+ "step": 180
1106
+ },
1107
+ {
1108
+ "epoch": 0.26,
1109
+ "learning_rate": 0.00019270715705718808,
1110
+ "loss": 0.8197,
1111
+ "step": 181
1112
+ },
1113
+ {
1114
+ "epoch": 0.26,
1115
+ "learning_rate": 0.00019262267883741986,
1116
+ "loss": 0.8525,
1117
+ "step": 182
1118
+ },
1119
+ {
1120
+ "epoch": 0.26,
1121
+ "learning_rate": 0.0001925377328797561,
1122
+ "loss": 0.8649,
1123
+ "step": 183
1124
+ },
1125
+ {
1126
+ "epoch": 0.26,
1127
+ "learning_rate": 0.00019245231961316782,
1128
+ "loss": 0.8416,
1129
+ "step": 184
1130
+ },
1131
+ {
1132
+ "epoch": 0.26,
1133
+ "learning_rate": 0.00019236643946898588,
1134
+ "loss": 0.8973,
1135
+ "step": 185
1136
+ },
1137
+ {
1138
+ "epoch": 0.26,
1139
+ "learning_rate": 0.00019228009288089885,
1140
+ "loss": 0.7733,
1141
+ "step": 186
1142
+ },
1143
+ {
1144
+ "epoch": 0.27,
1145
+ "learning_rate": 0.00019219328028495083,
1146
+ "loss": 0.8239,
1147
+ "step": 187
1148
+ },
1149
+ {
1150
+ "epoch": 0.27,
1151
+ "learning_rate": 0.00019210600211953918,
1152
+ "loss": 0.7993,
1153
+ "step": 188
1154
+ },
1155
+ {
1156
+ "epoch": 0.27,
1157
+ "learning_rate": 0.00019201825882541245,
1158
+ "loss": 0.8195,
1159
+ "step": 189
1160
+ },
1161
+ {
1162
+ "epoch": 0.27,
1163
+ "learning_rate": 0.00019193005084566797,
1164
+ "loss": 0.7977,
1165
+ "step": 190
1166
+ },
1167
+ {
1168
+ "epoch": 0.27,
1169
+ "learning_rate": 0.00019184137862574973,
1170
+ "loss": 0.85,
1171
+ "step": 191
1172
+ },
1173
+ {
1174
+ "epoch": 0.27,
1175
+ "learning_rate": 0.00019175224261344602,
1176
+ "loss": 0.8245,
1177
+ "step": 192
1178
+ },
1179
+ {
1180
+ "epoch": 0.27,
1181
+ "learning_rate": 0.00019166264325888742,
1182
+ "loss": 0.8929,
1183
+ "step": 193
1184
+ },
1185
+ {
1186
+ "epoch": 0.28,
1187
+ "learning_rate": 0.00019157258101454416,
1188
+ "loss": 0.8194,
1189
+ "step": 194
1190
+ },
1191
+ {
1192
+ "epoch": 0.28,
1193
+ "learning_rate": 0.00019148205633522414,
1194
+ "loss": 0.8493,
1195
+ "step": 195
1196
+ },
1197
+ {
1198
+ "epoch": 0.28,
1199
+ "learning_rate": 0.00019139106967807062,
1200
+ "loss": 0.8489,
1201
+ "step": 196
1202
+ },
1203
+ {
1204
+ "epoch": 0.28,
1205
+ "learning_rate": 0.00019129962150255957,
1206
+ "loss": 0.862,
1207
+ "step": 197
1208
+ },
1209
+ {
1210
+ "epoch": 0.28,
1211
+ "learning_rate": 0.00019120771227049778,
1212
+ "loss": 0.8723,
1213
+ "step": 198
1214
+ },
1215
+ {
1216
+ "epoch": 0.28,
1217
+ "learning_rate": 0.00019111534244602033,
1218
+ "loss": 0.8193,
1219
+ "step": 199
1220
+ },
1221
+ {
1222
+ "epoch": 0.28,
1223
+ "learning_rate": 0.00019102251249558812,
1224
+ "loss": 0.8619,
1225
+ "step": 200
1226
+ },
1227
+ {
1228
+ "epoch": 0.29,
1229
+ "learning_rate": 0.00019092922288798585,
1230
+ "loss": 0.8299,
1231
+ "step": 201
1232
+ },
1233
+ {
1234
+ "epoch": 0.29,
1235
+ "learning_rate": 0.0001908354740943193,
1236
+ "loss": 0.8403,
1237
+ "step": 202
1238
+ },
1239
+ {
1240
+ "epoch": 0.29,
1241
+ "learning_rate": 0.0001907412665880132,
1242
+ "loss": 0.8623,
1243
+ "step": 203
1244
+ },
1245
+ {
1246
+ "epoch": 0.29,
1247
+ "learning_rate": 0.00019064660084480868,
1248
+ "loss": 0.8998,
1249
+ "step": 204
1250
+ },
1251
+ {
1252
+ "epoch": 0.29,
1253
+ "learning_rate": 0.00019055147734276098,
1254
+ "loss": 0.8595,
1255
+ "step": 205
1256
+ },
1257
+ {
1258
+ "epoch": 0.29,
1259
+ "learning_rate": 0.00019045589656223696,
1260
+ "loss": 0.8264,
1261
+ "step": 206
1262
+ },
1263
+ {
1264
+ "epoch": 0.29,
1265
+ "learning_rate": 0.00019035985898591275,
1266
+ "loss": 0.8368,
1267
+ "step": 207
1268
+ },
1269
+ {
1270
+ "epoch": 0.3,
1271
+ "learning_rate": 0.0001902633650987712,
1272
+ "loss": 0.8436,
1273
+ "step": 208
1274
+ },
1275
+ {
1276
+ "epoch": 0.3,
1277
+ "learning_rate": 0.00019016641538809954,
1278
+ "loss": 0.8625,
1279
+ "step": 209
1280
+ },
1281
+ {
1282
+ "epoch": 0.3,
1283
+ "learning_rate": 0.00019006901034348688,
1284
+ "loss": 0.8775,
1285
+ "step": 210
1286
+ },
1287
+ {
1288
+ "epoch": 0.3,
1289
+ "learning_rate": 0.0001899711504568217,
1290
+ "loss": 0.7985,
1291
+ "step": 211
1292
+ },
1293
+ {
1294
+ "epoch": 0.3,
1295
+ "learning_rate": 0.00018987283622228944,
1296
+ "loss": 0.8656,
1297
+ "step": 212
1298
+ },
1299
+ {
1300
+ "epoch": 0.3,
1301
+ "learning_rate": 0.00018977406813636992,
1302
+ "loss": 0.8672,
1303
+ "step": 213
1304
+ },
1305
+ {
1306
+ "epoch": 0.3,
1307
+ "learning_rate": 0.00018967484669783492,
1308
+ "loss": 0.8057,
1309
+ "step": 214
1310
+ },
1311
+ {
1312
+ "epoch": 0.31,
1313
+ "learning_rate": 0.0001895751724077456,
1314
+ "loss": 0.8596,
1315
+ "step": 215
1316
+ },
1317
+ {
1318
+ "epoch": 0.31,
1319
+ "learning_rate": 0.00018947504576944998,
1320
+ "loss": 0.8373,
1321
+ "step": 216
1322
+ },
1323
+ {
1324
+ "epoch": 0.31,
1325
+ "learning_rate": 0.0001893744672885804,
1326
+ "loss": 0.8559,
1327
+ "step": 217
1328
+ },
1329
+ {
1330
+ "epoch": 0.31,
1331
+ "learning_rate": 0.000189273437473051,
1332
+ "loss": 0.7931,
1333
+ "step": 218
1334
+ },
1335
+ {
1336
+ "epoch": 0.31,
1337
+ "learning_rate": 0.00018917195683305517,
1338
+ "loss": 0.8594,
1339
+ "step": 219
1340
+ },
1341
+ {
1342
+ "epoch": 0.31,
1343
+ "learning_rate": 0.00018907002588106276,
1344
+ "loss": 0.877,
1345
+ "step": 220
1346
+ },
1347
+ {
1348
+ "epoch": 0.31,
1349
+ "learning_rate": 0.00018896764513181785,
1350
+ "loss": 0.8297,
1351
+ "step": 221
1352
+ },
1353
+ {
1354
+ "epoch": 0.32,
1355
+ "learning_rate": 0.0001888648151023359,
1356
+ "loss": 0.8407,
1357
+ "step": 222
1358
+ },
1359
+ {
1360
+ "epoch": 0.32,
1361
+ "learning_rate": 0.00018876153631190116,
1362
+ "loss": 0.8597,
1363
+ "step": 223
1364
+ },
1365
+ {
1366
+ "epoch": 0.32,
1367
+ "learning_rate": 0.00018865780928206417,
1368
+ "loss": 0.8647,
1369
+ "step": 224
1370
+ },
1371
+ {
1372
+ "epoch": 0.32,
1373
+ "learning_rate": 0.00018855363453663894,
1374
+ "loss": 0.8075,
1375
+ "step": 225
1376
+ },
1377
+ {
1378
+ "epoch": 0.32,
1379
+ "learning_rate": 0.0001884490126017005,
1380
+ "loss": 0.8438,
1381
+ "step": 226
1382
+ },
1383
+ {
1384
+ "epoch": 0.32,
1385
+ "learning_rate": 0.00018834394400558217,
1386
+ "loss": 0.8274,
1387
+ "step": 227
1388
+ },
1389
+ {
1390
+ "epoch": 0.32,
1391
+ "learning_rate": 0.0001882384292788728,
1392
+ "loss": 0.8721,
1393
+ "step": 228
1394
+ },
1395
+ {
1396
+ "epoch": 0.33,
1397
+ "learning_rate": 0.0001881324689544142,
1398
+ "loss": 0.8204,
1399
+ "step": 229
1400
+ },
1401
+ {
1402
+ "epoch": 0.33,
1403
+ "learning_rate": 0.00018802606356729844,
1404
+ "loss": 0.8424,
1405
+ "step": 230
1406
+ },
1407
+ {
1408
+ "epoch": 0.33,
1409
+ "learning_rate": 0.0001879192136548651,
1410
+ "loss": 0.8044,
1411
+ "step": 231
1412
+ },
1413
+ {
1414
+ "epoch": 0.33,
1415
+ "learning_rate": 0.00018781191975669859,
1416
+ "loss": 0.8082,
1417
+ "step": 232
1418
+ },
1419
+ {
1420
+ "epoch": 0.33,
1421
+ "learning_rate": 0.0001877041824146254,
1422
+ "loss": 0.8438,
1423
+ "step": 233
1424
+ },
1425
+ {
1426
+ "epoch": 0.33,
1427
+ "learning_rate": 0.0001875960021727114,
1428
+ "loss": 0.8748,
1429
+ "step": 234
1430
+ },
1431
+ {
1432
+ "epoch": 0.33,
1433
+ "learning_rate": 0.00018748737957725904,
1434
+ "loss": 0.854,
1435
+ "step": 235
1436
+ },
1437
+ {
1438
+ "epoch": 0.34,
1439
+ "learning_rate": 0.0001873783151768047,
1440
+ "loss": 0.851,
1441
+ "step": 236
1442
+ },
1443
+ {
1444
+ "epoch": 0.34,
1445
+ "learning_rate": 0.00018726880952211575,
1446
+ "loss": 0.8547,
1447
+ "step": 237
1448
+ },
1449
+ {
1450
+ "epoch": 0.34,
1451
+ "learning_rate": 0.0001871588631661879,
1452
+ "loss": 0.8399,
1453
+ "step": 238
1454
+ },
1455
+ {
1456
+ "epoch": 0.34,
1457
+ "learning_rate": 0.0001870484766642424,
1458
+ "loss": 0.8749,
1459
+ "step": 239
1460
+ },
1461
+ {
1462
+ "epoch": 0.34,
1463
+ "learning_rate": 0.00018693765057372318,
1464
+ "loss": 0.8127,
1465
+ "step": 240
1466
+ },
1467
+ {
1468
+ "epoch": 0.34,
1469
+ "learning_rate": 0.00018682638545429407,
1470
+ "loss": 0.8481,
1471
+ "step": 241
1472
+ },
1473
+ {
1474
+ "epoch": 0.34,
1475
+ "learning_rate": 0.00018671468186783592,
1476
+ "loss": 0.8164,
1477
+ "step": 242
1478
+ },
1479
+ {
1480
+ "epoch": 0.35,
1481
+ "learning_rate": 0.00018660254037844388,
1482
+ "loss": 0.8275,
1483
+ "step": 243
1484
+ },
1485
+ {
1486
+ "epoch": 0.35,
1487
+ "learning_rate": 0.0001864899615524244,
1488
+ "loss": 0.799,
1489
+ "step": 244
1490
+ },
1491
+ {
1492
+ "epoch": 0.35,
1493
+ "learning_rate": 0.00018637694595829252,
1494
+ "loss": 0.7889,
1495
+ "step": 245
1496
+ },
1497
+ {
1498
+ "epoch": 0.35,
1499
+ "learning_rate": 0.0001862634941667689,
1500
+ "loss": 0.8524,
1501
+ "step": 246
1502
+ },
1503
+ {
1504
+ "epoch": 0.35,
1505
+ "learning_rate": 0.00018614960675077696,
1506
+ "loss": 0.8409,
1507
+ "step": 247
1508
+ },
1509
+ {
1510
+ "epoch": 0.35,
1511
+ "learning_rate": 0.00018603528428544,
1512
+ "loss": 0.7696,
1513
+ "step": 248
1514
+ },
1515
+ {
1516
+ "epoch": 0.35,
1517
+ "learning_rate": 0.00018592052734807825,
1518
+ "loss": 0.8262,
1519
+ "step": 249
1520
+ },
1521
+ {
1522
+ "epoch": 0.36,
1523
+ "learning_rate": 0.00018580533651820603,
1524
+ "loss": 0.8208,
1525
+ "step": 250
1526
+ },
1527
+ {
1528
+ "epoch": 0.36,
1529
+ "learning_rate": 0.00018568971237752882,
1530
+ "loss": 0.8284,
1531
+ "step": 251
1532
+ },
1533
+ {
1534
+ "epoch": 0.36,
1535
+ "learning_rate": 0.00018557365550994024,
1536
+ "loss": 0.8471,
1537
+ "step": 252
1538
+ },
1539
+ {
1540
+ "epoch": 0.36,
1541
+ "learning_rate": 0.00018545716650151915,
1542
+ "loss": 0.8417,
1543
+ "step": 253
1544
+ },
1545
+ {
1546
+ "epoch": 0.36,
1547
+ "learning_rate": 0.0001853402459405267,
1548
+ "loss": 0.8051,
1549
+ "step": 254
1550
+ },
1551
+ {
1552
+ "epoch": 0.36,
1553
+ "learning_rate": 0.00018522289441740334,
1554
+ "loss": 0.8433,
1555
+ "step": 255
1556
+ },
1557
+ {
1558
+ "epoch": 0.36,
1559
+ "learning_rate": 0.00018510511252476587,
1560
+ "loss": 0.8526,
1561
+ "step": 256
1562
+ },
1563
+ {
1564
+ "epoch": 0.37,
1565
+ "learning_rate": 0.00018498690085740445,
1566
+ "loss": 0.805,
1567
+ "step": 257
1568
+ },
1569
+ {
1570
+ "epoch": 0.37,
1571
+ "learning_rate": 0.00018486826001227948,
1572
+ "loss": 0.8524,
1573
+ "step": 258
1574
+ },
1575
+ {
1576
+ "epoch": 0.37,
1577
+ "learning_rate": 0.00018474919058851877,
1578
+ "loss": 0.8487,
1579
+ "step": 259
1580
+ },
1581
+ {
1582
+ "epoch": 0.37,
1583
+ "learning_rate": 0.00018462969318741433,
1584
+ "loss": 0.8604,
1585
+ "step": 260
1586
+ },
1587
+ {
1588
+ "epoch": 0.37,
1589
+ "learning_rate": 0.00018450976841241951,
1590
+ "loss": 0.8518,
1591
+ "step": 261
1592
+ },
1593
+ {
1594
+ "epoch": 0.37,
1595
+ "learning_rate": 0.0001843894168691459,
1596
+ "loss": 0.769,
1597
+ "step": 262
1598
+ },
1599
+ {
1600
+ "epoch": 0.37,
1601
+ "learning_rate": 0.0001842686391653601,
1602
+ "loss": 0.8178,
1603
+ "step": 263
1604
+ },
1605
+ {
1606
+ "epoch": 0.38,
1607
+ "learning_rate": 0.00018414743591098085,
1608
+ "loss": 0.8224,
1609
+ "step": 264
1610
+ },
1611
+ {
1612
+ "epoch": 0.38,
1613
+ "learning_rate": 0.00018402580771807595,
1614
+ "loss": 0.8227,
1615
+ "step": 265
1616
+ },
1617
+ {
1618
+ "epoch": 0.38,
1619
+ "learning_rate": 0.00018390375520085903,
1620
+ "loss": 0.8333,
1621
+ "step": 266
1622
+ },
1623
+ {
1624
+ "epoch": 0.38,
1625
+ "learning_rate": 0.00018378127897568662,
1626
+ "loss": 0.8207,
1627
+ "step": 267
1628
+ },
1629
+ {
1630
+ "epoch": 0.38,
1631
+ "learning_rate": 0.00018365837966105483,
1632
+ "loss": 0.8144,
1633
+ "step": 268
1634
+ },
1635
+ {
1636
+ "epoch": 0.38,
1637
+ "learning_rate": 0.00018353505787759643,
1638
+ "loss": 0.8289,
1639
+ "step": 269
1640
+ },
1641
+ {
1642
+ "epoch": 0.38,
1643
+ "learning_rate": 0.0001834113142480776,
1644
+ "loss": 0.8387,
1645
+ "step": 270
1646
+ },
1647
+ {
1648
+ "epoch": 0.38,
1649
+ "learning_rate": 0.00018328714939739476,
1650
+ "loss": 0.8306,
1651
+ "step": 271
1652
+ },
1653
+ {
1654
+ "epoch": 0.39,
1655
+ "learning_rate": 0.00018316256395257155,
1656
+ "loss": 0.8485,
1657
+ "step": 272
1658
+ },
1659
+ {
1660
+ "epoch": 0.39,
1661
+ "learning_rate": 0.00018303755854275555,
1662
+ "loss": 0.7812,
1663
+ "step": 273
1664
+ },
1665
+ {
1666
+ "epoch": 0.39,
1667
+ "learning_rate": 0.00018291213379921512,
1668
+ "loss": 0.7986,
1669
+ "step": 274
1670
+ },
1671
+ {
1672
+ "epoch": 0.39,
1673
+ "learning_rate": 0.00018278629035533623,
1674
+ "loss": 0.8012,
1675
+ "step": 275
1676
+ },
1677
+ {
1678
+ "epoch": 0.39,
1679
+ "learning_rate": 0.00018266002884661928,
1680
+ "loss": 0.8854,
1681
+ "step": 276
1682
+ },
1683
+ {
1684
+ "epoch": 0.39,
1685
+ "learning_rate": 0.00018253334991067581,
1686
+ "loss": 0.8528,
1687
+ "step": 277
1688
+ },
1689
+ {
1690
+ "epoch": 0.39,
1691
+ "learning_rate": 0.00018240625418722541,
1692
+ "loss": 0.8012,
1693
+ "step": 278
1694
+ },
1695
+ {
1696
+ "epoch": 0.4,
1697
+ "learning_rate": 0.00018227874231809238,
1698
+ "loss": 0.8785,
1699
+ "step": 279
1700
+ },
1701
+ {
1702
+ "epoch": 0.4,
1703
+ "learning_rate": 0.00018215081494720248,
1704
+ "loss": 0.8013,
1705
+ "step": 280
1706
+ },
1707
+ {
1708
+ "epoch": 0.4,
1709
+ "learning_rate": 0.00018202247272057983,
1710
+ "loss": 0.7962,
1711
+ "step": 281
1712
+ },
1713
+ {
1714
+ "epoch": 0.4,
1715
+ "learning_rate": 0.00018189371628634347,
1716
+ "loss": 0.8746,
1717
+ "step": 282
1718
+ },
1719
+ {
1720
+ "epoch": 0.4,
1721
+ "learning_rate": 0.00018176454629470414,
1722
+ "loss": 0.8755,
1723
+ "step": 283
1724
+ },
1725
+ {
1726
+ "epoch": 0.4,
1727
+ "learning_rate": 0.0001816349633979611,
1728
+ "loss": 0.8665,
1729
+ "step": 284
1730
+ },
1731
+ {
1732
+ "epoch": 0.4,
1733
+ "learning_rate": 0.00018150496825049866,
1734
+ "loss": 0.8163,
1735
+ "step": 285
1736
+ },
1737
+ {
1738
+ "epoch": 0.41,
1739
+ "learning_rate": 0.00018137456150878303,
1740
+ "loss": 0.8478,
1741
+ "step": 286
1742
+ },
1743
+ {
1744
+ "epoch": 0.41,
1745
+ "learning_rate": 0.0001812437438313589,
1746
+ "loss": 0.7785,
1747
+ "step": 287
1748
+ },
1749
+ {
1750
+ "epoch": 0.41,
1751
+ "learning_rate": 0.00018111251587884616,
1752
+ "loss": 0.7899,
1753
+ "step": 288
1754
+ },
1755
+ {
1756
+ "epoch": 0.41,
1757
+ "learning_rate": 0.00018098087831393663,
1758
+ "loss": 0.8274,
1759
+ "step": 289
1760
+ },
1761
+ {
1762
+ "epoch": 0.41,
1763
+ "learning_rate": 0.00018084883180139052,
1764
+ "loss": 0.8124,
1765
+ "step": 290
1766
+ },
1767
+ {
1768
+ "epoch": 0.41,
1769
+ "learning_rate": 0.00018071637700803334,
1770
+ "loss": 0.8438,
1771
+ "step": 291
1772
+ },
1773
+ {
1774
+ "epoch": 0.41,
1775
+ "learning_rate": 0.0001805835146027522,
1776
+ "loss": 0.8542,
1777
+ "step": 292
1778
+ },
1779
+ {
1780
+ "epoch": 0.42,
1781
+ "learning_rate": 0.00018045024525649284,
1782
+ "loss": 0.8308,
1783
+ "step": 293
1784
+ },
1785
+ {
1786
+ "epoch": 0.42,
1787
+ "learning_rate": 0.00018031656964225584,
1788
+ "loss": 0.8677,
1789
+ "step": 294
1790
+ },
1791
+ {
1792
+ "epoch": 0.42,
1793
+ "learning_rate": 0.00018018248843509354,
1794
+ "loss": 0.8398,
1795
+ "step": 295
1796
+ },
1797
+ {
1798
+ "epoch": 0.42,
1799
+ "learning_rate": 0.00018004800231210638,
1800
+ "loss": 0.8078,
1801
+ "step": 296
1802
+ },
1803
+ {
1804
+ "epoch": 0.42,
1805
+ "learning_rate": 0.0001799131119524397,
1806
+ "loss": 0.7889,
1807
+ "step": 297
1808
+ },
1809
+ {
1810
+ "epoch": 0.42,
1811
+ "learning_rate": 0.0001797778180372801,
1812
+ "loss": 0.8287,
1813
+ "step": 298
1814
+ },
1815
+ {
1816
+ "epoch": 0.42,
1817
+ "learning_rate": 0.00017964212124985224,
1818
+ "loss": 0.8584,
1819
+ "step": 299
1820
+ },
1821
+ {
1822
+ "epoch": 0.43,
1823
+ "learning_rate": 0.0001795060222754151,
1824
+ "loss": 0.8178,
1825
+ "step": 300
1826
+ },
1827
+ {
1828
+ "epoch": 0.43,
1829
+ "learning_rate": 0.0001793695218012588,
1830
+ "loss": 0.8409,
1831
+ "step": 301
1832
+ },
1833
+ {
1834
+ "epoch": 0.43,
1835
+ "learning_rate": 0.000179232620516701,
1836
+ "loss": 0.8806,
1837
+ "step": 302
1838
+ },
1839
+ {
1840
+ "epoch": 0.43,
1841
+ "learning_rate": 0.00017909531911308325,
1842
+ "loss": 0.8563,
1843
+ "step": 303
1844
+ },
1845
+ {
1846
+ "epoch": 0.43,
1847
+ "learning_rate": 0.00017895761828376795,
1848
+ "loss": 0.8051,
1849
+ "step": 304
1850
+ },
1851
+ {
1852
+ "epoch": 0.43,
1853
+ "learning_rate": 0.00017881951872413435,
1854
+ "loss": 0.8475,
1855
+ "step": 305
1856
+ },
1857
+ {
1858
+ "epoch": 0.43,
1859
+ "learning_rate": 0.00017868102113157534,
1860
+ "loss": 0.804,
1861
+ "step": 306
1862
+ },
1863
+ {
1864
+ "epoch": 0.44,
1865
+ "learning_rate": 0.0001785421262054939,
1866
+ "loss": 0.8033,
1867
+ "step": 307
1868
+ },
1869
+ {
1870
+ "epoch": 0.44,
1871
+ "learning_rate": 0.0001784028346472994,
1872
+ "loss": 0.8219,
1873
+ "step": 308
1874
+ },
1875
+ {
1876
+ "epoch": 0.44,
1877
+ "learning_rate": 0.00017826314716040423,
1878
+ "loss": 0.8595,
1879
+ "step": 309
1880
+ },
1881
+ {
1882
+ "epoch": 0.44,
1883
+ "learning_rate": 0.00017812306445022025,
1884
+ "loss": 0.8405,
1885
+ "step": 310
1886
+ },
1887
+ {
1888
+ "epoch": 0.44,
1889
+ "learning_rate": 0.00017798258722415508,
1890
+ "loss": 0.7675,
1891
+ "step": 311
1892
+ },
1893
+ {
1894
+ "epoch": 0.44,
1895
+ "learning_rate": 0.0001778417161916087,
1896
+ "loss": 0.8026,
1897
+ "step": 312
1898
+ },
1899
+ {
1900
+ "epoch": 0.44,
1901
+ "learning_rate": 0.00017770045206396963,
1902
+ "loss": 0.8292,
1903
+ "step": 313
1904
+ },
1905
+ {
1906
+ "epoch": 0.45,
1907
+ "learning_rate": 0.0001775587955546117,
1908
+ "loss": 0.8287,
1909
+ "step": 314
1910
+ },
1911
+ {
1912
+ "epoch": 0.45,
1913
+ "learning_rate": 0.0001774167473788901,
1914
+ "loss": 0.8299,
1915
+ "step": 315
1916
+ },
1917
+ {
1918
+ "epoch": 0.45,
1919
+ "learning_rate": 0.00017727430825413792,
1920
+ "loss": 0.8101,
1921
+ "step": 316
1922
+ },
1923
+ {
1924
+ "epoch": 0.45,
1925
+ "learning_rate": 0.00017713147889966262,
1926
+ "loss": 0.7898,
1927
+ "step": 317
1928
+ },
1929
+ {
1930
+ "epoch": 0.45,
1931
+ "learning_rate": 0.0001769882600367421,
1932
+ "loss": 0.8084,
1933
+ "step": 318
1934
+ },
1935
+ {
1936
+ "epoch": 0.45,
1937
+ "learning_rate": 0.00017684465238862148,
1938
+ "loss": 0.8087,
1939
+ "step": 319
1940
+ },
1941
+ {
1942
+ "epoch": 0.45,
1943
+ "learning_rate": 0.000176700656680509,
1944
+ "loss": 0.8065,
1945
+ "step": 320
1946
+ },
1947
+ {
1948
+ "epoch": 0.46,
1949
+ "learning_rate": 0.00017655627363957277,
1950
+ "loss": 0.8368,
1951
+ "step": 321
1952
+ },
1953
+ {
1954
+ "epoch": 0.46,
1955
+ "learning_rate": 0.0001764115039949367,
1956
+ "loss": 0.8231,
1957
+ "step": 322
1958
+ },
1959
+ {
1960
+ "epoch": 0.46,
1961
+ "learning_rate": 0.0001762663484776772,
1962
+ "loss": 0.8333,
1963
+ "step": 323
1964
+ },
1965
+ {
1966
+ "epoch": 0.46,
1967
+ "learning_rate": 0.00017612080782081923,
1968
+ "loss": 0.8453,
1969
+ "step": 324
1970
+ },
1971
+ {
1972
+ "epoch": 0.46,
1973
+ "learning_rate": 0.00017597488275933267,
1974
+ "loss": 0.8669,
1975
+ "step": 325
1976
+ },
1977
+ {
1978
+ "epoch": 0.46,
1979
+ "learning_rate": 0.00017582857403012866,
1980
+ "loss": 0.8022,
1981
+ "step": 326
1982
+ },
1983
+ {
1984
+ "epoch": 0.46,
1985
+ "learning_rate": 0.00017568188237205575,
1986
+ "loss": 0.8116,
1987
+ "step": 327
1988
+ },
1989
+ {
1990
+ "epoch": 0.47,
1991
+ "learning_rate": 0.00017553480852589636,
1992
+ "loss": 0.8307,
1993
+ "step": 328
1994
+ },
1995
+ {
1996
+ "epoch": 0.47,
1997
+ "learning_rate": 0.00017538735323436284,
1998
+ "loss": 0.7986,
1999
+ "step": 329
2000
+ },
2001
+ {
2002
+ "epoch": 0.47,
2003
+ "learning_rate": 0.00017523951724209388,
2004
+ "loss": 0.8278,
2005
+ "step": 330
2006
+ },
2007
+ {
2008
+ "epoch": 0.47,
2009
+ "learning_rate": 0.00017509130129565067,
2010
+ "loss": 0.8397,
2011
+ "step": 331
2012
+ },
2013
+ {
2014
+ "epoch": 0.47,
2015
+ "learning_rate": 0.0001749427061435131,
2016
+ "loss": 0.7862,
2017
+ "step": 332
2018
+ },
2019
+ {
2020
+ "epoch": 0.47,
2021
+ "learning_rate": 0.00017479373253607606,
2022
+ "loss": 0.8574,
2023
+ "step": 333
2024
+ },
2025
+ {
2026
+ "epoch": 0.47,
2027
+ "learning_rate": 0.00017464438122564556,
2028
+ "loss": 0.8312,
2029
+ "step": 334
2030
+ },
2031
+ {
2032
+ "epoch": 0.48,
2033
+ "learning_rate": 0.0001744946529664351,
2034
+ "loss": 0.8666,
2035
+ "step": 335
2036
+ },
2037
+ {
2038
+ "epoch": 0.48,
2039
+ "learning_rate": 0.00017434454851456162,
2040
+ "loss": 0.8027,
2041
+ "step": 336
2042
+ },
2043
+ {
2044
+ "epoch": 0.48,
2045
+ "learning_rate": 0.00017419406862804183,
2046
+ "loss": 0.8962,
2047
+ "step": 337
2048
+ },
2049
+ {
2050
+ "epoch": 0.48,
2051
+ "learning_rate": 0.00017404321406678844,
2052
+ "loss": 0.8286,
2053
+ "step": 338
2054
+ },
2055
+ {
2056
+ "epoch": 0.48,
2057
+ "learning_rate": 0.00017389198559260614,
2058
+ "loss": 0.8167,
2059
+ "step": 339
2060
+ },
2061
+ {
2062
+ "epoch": 0.48,
2063
+ "learning_rate": 0.00017374038396918788,
2064
+ "loss": 0.831,
2065
+ "step": 340
2066
+ },
2067
+ {
2068
+ "epoch": 0.48,
2069
+ "learning_rate": 0.00017358840996211104,
2070
+ "loss": 0.8218,
2071
+ "step": 341
2072
+ },
2073
+ {
2074
+ "epoch": 0.49,
2075
+ "learning_rate": 0.00017343606433883348,
2076
+ "loss": 0.7902,
2077
+ "step": 342
2078
+ },
2079
+ {
2080
+ "epoch": 0.49,
2081
+ "learning_rate": 0.00017328334786868968,
2082
+ "loss": 0.8418,
2083
+ "step": 343
2084
+ },
2085
+ {
2086
+ "epoch": 0.49,
2087
+ "learning_rate": 0.00017313026132288682,
2088
+ "loss": 0.853,
2089
+ "step": 344
2090
+ },
2091
+ {
2092
+ "epoch": 0.49,
2093
+ "learning_rate": 0.00017297680547450107,
2094
+ "loss": 0.8153,
2095
+ "step": 345
2096
+ },
2097
+ {
2098
+ "epoch": 0.49,
2099
+ "learning_rate": 0.00017282298109847345,
2100
+ "loss": 0.8121,
2101
+ "step": 346
2102
+ },
2103
+ {
2104
+ "epoch": 0.49,
2105
+ "learning_rate": 0.00017266878897160605,
2106
+ "loss": 0.8085,
2107
+ "step": 347
2108
+ },
2109
+ {
2110
+ "epoch": 0.49,
2111
+ "learning_rate": 0.00017251422987255802,
2112
+ "loss": 0.8342,
2113
+ "step": 348
2114
+ },
2115
+ {
2116
+ "epoch": 0.5,
2117
+ "learning_rate": 0.0001723593045818418,
2118
+ "loss": 0.7986,
2119
+ "step": 349
2120
+ },
2121
+ {
2122
+ "epoch": 0.5,
2123
+ "learning_rate": 0.00017220401388181903,
2124
+ "loss": 0.7917,
2125
+ "step": 350
2126
+ },
2127
+ {
2128
+ "epoch": 0.5,
2129
+ "learning_rate": 0.00017204835855669657,
2130
+ "loss": 0.8111,
2131
+ "step": 351
2132
+ },
2133
+ {
2134
+ "epoch": 0.5,
2135
+ "learning_rate": 0.00017189233939252267,
2136
+ "loss": 0.8284,
2137
+ "step": 352
2138
+ },
2139
+ {
2140
+ "epoch": 0.5,
2141
+ "eval_loss": 0.846347987651825,
2142
+ "eval_runtime": 58.037,
2143
+ "eval_samples_per_second": 6.892,
2144
+ "eval_steps_per_second": 3.446,
2145
+ "step": 352
2146
+ },
2147
+ {
2148
+ "epoch": 0.5,
2149
+ "learning_rate": 0.00017173595717718295,
2150
+ "loss": 0.8536,
2151
+ "step": 353
2152
+ },
2153
+ {
2154
+ "epoch": 0.5,
2155
+ "learning_rate": 0.00017157921270039646,
2156
+ "loss": 0.8388,
2157
+ "step": 354
2158
+ },
2159
+ {
2160
+ "epoch": 0.5,
2161
+ "learning_rate": 0.0001714221067537115,
2162
+ "loss": 0.7963,
2163
+ "step": 355
2164
+ },
2165
+ {
2166
+ "epoch": 0.51,
2167
+ "learning_rate": 0.00017126464013050185,
2168
+ "loss": 0.7786,
2169
+ "step": 356
2170
+ },
2171
+ {
2172
+ "epoch": 0.51,
2173
+ "learning_rate": 0.00017110681362596272,
2174
+ "loss": 0.8085,
2175
+ "step": 357
2176
+ },
2177
+ {
2178
+ "epoch": 0.51,
2179
+ "learning_rate": 0.00017094862803710664,
2180
+ "loss": 0.8535,
2181
+ "step": 358
2182
+ },
2183
+ {
2184
+ "epoch": 0.51,
2185
+ "learning_rate": 0.00017079008416275954,
2186
+ "loss": 0.8492,
2187
+ "step": 359
2188
+ },
2189
+ {
2190
+ "epoch": 0.51,
2191
+ "learning_rate": 0.00017063118280355655,
2192
+ "loss": 0.884,
2193
+ "step": 360
2194
+ },
2195
+ {
2196
+ "epoch": 0.51,
2197
+ "learning_rate": 0.00017047192476193825,
2198
+ "loss": 0.8504,
2199
+ "step": 361
2200
+ },
2201
+ {
2202
+ "epoch": 0.51,
2203
+ "learning_rate": 0.0001703123108421463,
2204
+ "loss": 0.8377,
2205
+ "step": 362
2206
+ },
2207
+ {
2208
+ "epoch": 0.52,
2209
+ "learning_rate": 0.00017015234185021957,
2210
+ "loss": 0.8179,
2211
+ "step": 363
2212
+ },
2213
+ {
2214
+ "epoch": 0.52,
2215
+ "learning_rate": 0.00016999201859399,
2216
+ "loss": 0.8584,
2217
+ "step": 364
2218
+ },
2219
+ {
2220
+ "epoch": 0.52,
2221
+ "learning_rate": 0.00016983134188307858,
2222
+ "loss": 0.8151,
2223
+ "step": 365
2224
+ },
2225
+ {
2226
+ "epoch": 0.52,
2227
+ "learning_rate": 0.0001696703125288912,
2228
+ "loss": 0.7949,
2229
+ "step": 366
2230
+ },
2231
+ {
2232
+ "epoch": 0.52,
2233
+ "learning_rate": 0.0001695089313446145,
2234
+ "loss": 0.7939,
2235
+ "step": 367
2236
+ },
2237
+ {
2238
+ "epoch": 0.52,
2239
+ "learning_rate": 0.000169347199145212,
2240
+ "loss": 0.8024,
2241
+ "step": 368
2242
+ },
2243
+ {
2244
+ "epoch": 0.52,
2245
+ "learning_rate": 0.00016918511674741965,
2246
+ "loss": 0.8164,
2247
+ "step": 369
2248
+ },
2249
+ {
2250
+ "epoch": 0.53,
2251
+ "learning_rate": 0.00016902268496974201,
2252
+ "loss": 0.8424,
2253
+ "step": 370
2254
+ },
2255
+ {
2256
+ "epoch": 0.53,
2257
+ "learning_rate": 0.00016885990463244785,
2258
+ "loss": 0.8081,
2259
+ "step": 371
2260
+ },
2261
+ {
2262
+ "epoch": 0.53,
2263
+ "learning_rate": 0.0001686967765575663,
2264
+ "loss": 0.8728,
2265
+ "step": 372
2266
+ },
2267
+ {
2268
+ "epoch": 0.53,
2269
+ "learning_rate": 0.0001685333015688824,
2270
+ "loss": 0.8443,
2271
+ "step": 373
2272
+ },
2273
+ {
2274
+ "epoch": 0.53,
2275
+ "learning_rate": 0.00016836948049193316,
2276
+ "loss": 0.7933,
2277
+ "step": 374
2278
+ },
2279
+ {
2280
+ "epoch": 0.53,
2281
+ "learning_rate": 0.00016820531415400334,
2282
+ "loss": 0.8349,
2283
+ "step": 375
2284
+ },
2285
+ {
2286
+ "epoch": 0.53,
2287
+ "learning_rate": 0.00016804080338412108,
2288
+ "loss": 0.864,
2289
+ "step": 376
2290
+ },
2291
+ {
2292
+ "epoch": 0.54,
2293
+ "learning_rate": 0.00016787594901305404,
2294
+ "loss": 0.8422,
2295
+ "step": 377
2296
+ },
2297
+ {
2298
+ "epoch": 0.54,
2299
+ "learning_rate": 0.000167710751873305,
2300
+ "loss": 0.7993,
2301
+ "step": 378
2302
+ },
2303
+ {
2304
+ "epoch": 0.54,
2305
+ "learning_rate": 0.0001675452127991077,
2306
+ "loss": 0.8051,
2307
+ "step": 379
2308
+ },
2309
+ {
2310
+ "epoch": 0.54,
2311
+ "learning_rate": 0.00016737933262642258,
2312
+ "loss": 0.7417,
2313
+ "step": 380
2314
+ },
2315
+ {
2316
+ "epoch": 0.54,
2317
+ "learning_rate": 0.0001672131121929326,
2318
+ "loss": 0.7987,
2319
+ "step": 381
2320
+ },
2321
+ {
2322
+ "epoch": 0.54,
2323
+ "learning_rate": 0.00016704655233803912,
2324
+ "loss": 0.8387,
2325
+ "step": 382
2326
+ },
2327
+ {
2328
+ "epoch": 0.54,
2329
+ "learning_rate": 0.00016687965390285747,
2330
+ "loss": 0.815,
2331
+ "step": 383
2332
+ },
2333
+ {
2334
+ "epoch": 0.55,
2335
+ "learning_rate": 0.00016671241773021276,
2336
+ "loss": 0.8293,
2337
+ "step": 384
2338
+ },
2339
+ {
2340
+ "epoch": 0.55,
2341
+ "learning_rate": 0.0001665448446646357,
2342
+ "loss": 0.7945,
2343
+ "step": 385
2344
+ },
2345
+ {
2346
+ "epoch": 0.55,
2347
+ "learning_rate": 0.00016637693555235825,
2348
+ "loss": 0.801,
2349
+ "step": 386
2350
+ },
2351
+ {
2352
+ "epoch": 0.55,
2353
+ "learning_rate": 0.00016620869124130944,
2354
+ "loss": 0.8003,
2355
+ "step": 387
2356
+ },
2357
+ {
2358
+ "epoch": 0.55,
2359
+ "learning_rate": 0.000166040112581111,
2360
+ "loss": 0.8379,
2361
+ "step": 388
2362
+ },
2363
+ {
2364
+ "epoch": 0.55,
2365
+ "learning_rate": 0.00016587120042307305,
2366
+ "loss": 0.7929,
2367
+ "step": 389
2368
+ },
2369
+ {
2370
+ "epoch": 0.55,
2371
+ "learning_rate": 0.00016570195562018992,
2372
+ "loss": 0.7693,
2373
+ "step": 390
2374
+ },
2375
+ {
2376
+ "epoch": 0.56,
2377
+ "learning_rate": 0.00016553237902713574,
2378
+ "loss": 0.8191,
2379
+ "step": 391
2380
+ },
2381
+ {
2382
+ "epoch": 0.56,
2383
+ "learning_rate": 0.00016536247150026017,
2384
+ "loss": 0.826,
2385
+ "step": 392
2386
+ },
2387
+ {
2388
+ "epoch": 0.56,
2389
+ "learning_rate": 0.000165192233897584,
2390
+ "loss": 0.814,
2391
+ "step": 393
2392
+ },
2393
+ {
2394
+ "epoch": 0.56,
2395
+ "learning_rate": 0.00016502166707879504,
2396
+ "loss": 0.8159,
2397
+ "step": 394
2398
+ },
2399
+ {
2400
+ "epoch": 0.56,
2401
+ "learning_rate": 0.00016485077190524341,
2402
+ "loss": 0.8702,
2403
+ "step": 395
2404
+ },
2405
+ {
2406
+ "epoch": 0.56,
2407
+ "learning_rate": 0.00016467954923993756,
2408
+ "loss": 0.8188,
2409
+ "step": 396
2410
+ },
2411
+ {
2412
+ "epoch": 0.56,
2413
+ "learning_rate": 0.00016450799994753966,
2414
+ "loss": 0.8238,
2415
+ "step": 397
2416
+ },
2417
+ {
2418
+ "epoch": 0.57,
2419
+ "learning_rate": 0.00016433612489436135,
2420
+ "loss": 0.8168,
2421
+ "step": 398
2422
+ },
2423
+ {
2424
+ "epoch": 0.57,
2425
+ "learning_rate": 0.00016416392494835935,
2426
+ "loss": 0.7938,
2427
+ "step": 399
2428
+ },
2429
+ {
2430
+ "epoch": 0.57,
2431
+ "learning_rate": 0.00016399140097913105,
2432
+ "loss": 0.8115,
2433
+ "step": 400
2434
+ },
2435
+ {
2436
+ "epoch": 0.57,
2437
+ "learning_rate": 0.00016381855385791015,
2438
+ "loss": 0.8096,
2439
+ "step": 401
2440
+ },
2441
+ {
2442
+ "epoch": 0.57,
2443
+ "learning_rate": 0.00016364538445756224,
2444
+ "loss": 0.7986,
2445
+ "step": 402
2446
+ },
2447
+ {
2448
+ "epoch": 0.57,
2449
+ "learning_rate": 0.00016347189365258034,
2450
+ "loss": 0.7957,
2451
+ "step": 403
2452
+ },
2453
+ {
2454
+ "epoch": 0.57,
2455
+ "learning_rate": 0.0001632980823190807,
2456
+ "loss": 0.8502,
2457
+ "step": 404
2458
+ },
2459
+ {
2460
+ "epoch": 0.58,
2461
+ "learning_rate": 0.000163123951334798,
2462
+ "loss": 0.8214,
2463
+ "step": 405
2464
+ },
2465
+ {
2466
+ "epoch": 0.58,
2467
+ "learning_rate": 0.00016294950157908132,
2468
+ "loss": 0.8192,
2469
+ "step": 406
2470
+ },
2471
+ {
2472
+ "epoch": 0.58,
2473
+ "learning_rate": 0.00016277473393288937,
2474
+ "loss": 0.8425,
2475
+ "step": 407
2476
+ },
2477
+ {
2478
+ "epoch": 0.58,
2479
+ "learning_rate": 0.00016259964927878626,
2480
+ "loss": 0.7956,
2481
+ "step": 408
2482
+ },
2483
+ {
2484
+ "epoch": 0.58,
2485
+ "learning_rate": 0.000162424248500937,
2486
+ "loss": 0.7574,
2487
+ "step": 409
2488
+ },
2489
+ {
2490
+ "epoch": 0.58,
2491
+ "learning_rate": 0.0001622485324851029,
2492
+ "loss": 0.8459,
2493
+ "step": 410
2494
+ },
2495
+ {
2496
+ "epoch": 0.58,
2497
+ "learning_rate": 0.00016207250211863728,
2498
+ "loss": 0.7867,
2499
+ "step": 411
2500
+ },
2501
+ {
2502
+ "epoch": 0.59,
2503
+ "learning_rate": 0.00016189615829048095,
2504
+ "loss": 0.8068,
2505
+ "step": 412
2506
+ },
2507
+ {
2508
+ "epoch": 0.59,
2509
+ "learning_rate": 0.00016171950189115751,
2510
+ "loss": 0.8387,
2511
+ "step": 413
2512
+ },
2513
+ {
2514
+ "epoch": 0.59,
2515
+ "learning_rate": 0.0001615425338127692,
2516
+ "loss": 0.7998,
2517
+ "step": 414
2518
+ },
2519
+ {
2520
+ "epoch": 0.59,
2521
+ "learning_rate": 0.00016136525494899208,
2522
+ "loss": 0.819,
2523
+ "step": 415
2524
+ },
2525
+ {
2526
+ "epoch": 0.59,
2527
+ "learning_rate": 0.00016118766619507176,
2528
+ "loss": 0.8168,
2529
+ "step": 416
2530
+ },
2531
+ {
2532
+ "epoch": 0.59,
2533
+ "learning_rate": 0.00016100976844781877,
2534
+ "loss": 0.801,
2535
+ "step": 417
2536
+ },
2537
+ {
2538
+ "epoch": 0.59,
2539
+ "learning_rate": 0.00016083156260560387,
2540
+ "loss": 0.8251,
2541
+ "step": 418
2542
+ },
2543
+ {
2544
+ "epoch": 0.6,
2545
+ "learning_rate": 0.00016065304956835395,
2546
+ "loss": 0.8453,
2547
+ "step": 419
2548
+ },
2549
+ {
2550
+ "epoch": 0.6,
2551
+ "learning_rate": 0.00016047423023754696,
2552
+ "loss": 0.8419,
2553
+ "step": 420
2554
+ },
2555
+ {
2556
+ "epoch": 0.6,
2557
+ "learning_rate": 0.00016029510551620777,
2558
+ "loss": 0.7758,
2559
+ "step": 421
2560
+ },
2561
+ {
2562
+ "epoch": 0.6,
2563
+ "learning_rate": 0.00016011567630890336,
2564
+ "loss": 0.8178,
2565
+ "step": 422
2566
+ },
2567
+ {
2568
+ "epoch": 0.6,
2569
+ "learning_rate": 0.0001599359435217384,
2570
+ "loss": 0.8313,
2571
+ "step": 423
2572
+ },
2573
+ {
2574
+ "epoch": 0.6,
2575
+ "learning_rate": 0.00015975590806235058,
2576
+ "loss": 0.8286,
2577
+ "step": 424
2578
+ },
2579
+ {
2580
+ "epoch": 0.6,
2581
+ "learning_rate": 0.0001595755708399061,
2582
+ "loss": 0.8054,
2583
+ "step": 425
2584
+ },
2585
+ {
2586
+ "epoch": 0.61,
2587
+ "learning_rate": 0.000159394932765095,
2588
+ "loss": 0.834,
2589
+ "step": 426
2590
+ },
2591
+ {
2592
+ "epoch": 0.61,
2593
+ "learning_rate": 0.00015921399475012663,
2594
+ "loss": 0.8343,
2595
+ "step": 427
2596
+ },
2597
+ {
2598
+ "epoch": 0.61,
2599
+ "learning_rate": 0.000159032757708725,
2600
+ "loss": 0.813,
2601
+ "step": 428
2602
+ },
2603
+ {
2604
+ "epoch": 0.61,
2605
+ "learning_rate": 0.00015885122255612425,
2606
+ "loss": 0.7911,
2607
+ "step": 429
2608
+ },
2609
+ {
2610
+ "epoch": 0.61,
2611
+ "learning_rate": 0.00015866939020906377,
2612
+ "loss": 0.8037,
2613
+ "step": 430
2614
+ },
2615
+ {
2616
+ "epoch": 0.61,
2617
+ "learning_rate": 0.00015848726158578403,
2618
+ "loss": 0.838,
2619
+ "step": 431
2620
+ },
2621
+ {
2622
+ "epoch": 0.61,
2623
+ "learning_rate": 0.0001583048376060215,
2624
+ "loss": 0.8053,
2625
+ "step": 432
2626
+ },
2627
+ {
2628
+ "epoch": 0.62,
2629
+ "learning_rate": 0.00015812211919100411,
2630
+ "loss": 0.8014,
2631
+ "step": 433
2632
+ },
2633
+ {
2634
+ "epoch": 0.62,
2635
+ "learning_rate": 0.00015793910726344694,
2636
+ "loss": 0.7935,
2637
+ "step": 434
2638
+ },
2639
+ {
2640
+ "epoch": 0.62,
2641
+ "learning_rate": 0.00015775580274754697,
2642
+ "loss": 0.7803,
2643
+ "step": 435
2644
+ },
2645
+ {
2646
+ "epoch": 0.62,
2647
+ "learning_rate": 0.00015757220656897896,
2648
+ "loss": 0.8563,
2649
+ "step": 436
2650
+ },
2651
+ {
2652
+ "epoch": 0.62,
2653
+ "learning_rate": 0.00015738831965489048,
2654
+ "loss": 0.7825,
2655
+ "step": 437
2656
+ },
2657
+ {
2658
+ "epoch": 0.62,
2659
+ "learning_rate": 0.0001572041429338972,
2660
+ "loss": 0.8035,
2661
+ "step": 438
2662
+ },
2663
+ {
2664
+ "epoch": 0.62,
2665
+ "learning_rate": 0.00015701967733607844,
2666
+ "loss": 0.8199,
2667
+ "step": 439
2668
+ },
2669
+ {
2670
+ "epoch": 0.62,
2671
+ "learning_rate": 0.00015683492379297222,
2672
+ "loss": 0.8314,
2673
+ "step": 440
2674
+ },
2675
+ {
2676
+ "epoch": 0.63,
2677
+ "learning_rate": 0.00015664988323757072,
2678
+ "loss": 0.7746,
2679
+ "step": 441
2680
+ },
2681
+ {
2682
+ "epoch": 0.63,
2683
+ "learning_rate": 0.00015646455660431552,
2684
+ "loss": 0.7724,
2685
+ "step": 442
2686
+ },
2687
+ {
2688
+ "epoch": 0.63,
2689
+ "learning_rate": 0.0001562789448290928,
2690
+ "loss": 0.8263,
2691
+ "step": 443
2692
+ },
2693
+ {
2694
+ "epoch": 0.63,
2695
+ "learning_rate": 0.00015609304884922878,
2696
+ "loss": 0.7956,
2697
+ "step": 444
2698
+ },
2699
+ {
2700
+ "epoch": 0.63,
2701
+ "learning_rate": 0.0001559068696034848,
2702
+ "loss": 0.8388,
2703
+ "step": 445
2704
+ },
2705
+ {
2706
+ "epoch": 0.63,
2707
+ "learning_rate": 0.00015572040803205273,
2708
+ "loss": 0.8226,
2709
+ "step": 446
2710
+ },
2711
+ {
2712
+ "epoch": 0.63,
2713
+ "learning_rate": 0.0001555336650765502,
2714
+ "loss": 0.8014,
2715
+ "step": 447
2716
+ },
2717
+ {
2718
+ "epoch": 0.64,
2719
+ "learning_rate": 0.00015534664168001568,
2720
+ "loss": 0.8371,
2721
+ "step": 448
2722
+ },
2723
+ {
2724
+ "epoch": 0.64,
2725
+ "learning_rate": 0.000155159338786904,
2726
+ "loss": 0.8162,
2727
+ "step": 449
2728
+ },
2729
+ {
2730
+ "epoch": 0.64,
2731
+ "learning_rate": 0.00015497175734308135,
2732
+ "loss": 0.8334,
2733
+ "step": 450
2734
+ },
2735
+ {
2736
+ "epoch": 0.64,
2737
+ "learning_rate": 0.00015478389829582057,
2738
+ "loss": 0.802,
2739
+ "step": 451
2740
+ },
2741
+ {
2742
+ "epoch": 0.64,
2743
+ "learning_rate": 0.00015459576259379637,
2744
+ "loss": 0.8153,
2745
+ "step": 452
2746
+ },
2747
+ {
2748
+ "epoch": 0.64,
2749
+ "learning_rate": 0.00015440735118708062,
2750
+ "loss": 0.8418,
2751
+ "step": 453
2752
+ },
2753
+ {
2754
+ "epoch": 0.64,
2755
+ "learning_rate": 0.0001542186650271374,
2756
+ "loss": 0.7649,
2757
+ "step": 454
2758
+ },
2759
+ {
2760
+ "epoch": 0.65,
2761
+ "learning_rate": 0.00015402970506681832,
2762
+ "loss": 0.8178,
2763
+ "step": 455
2764
+ },
2765
+ {
2766
+ "epoch": 0.65,
2767
+ "learning_rate": 0.0001538404722603577,
2768
+ "loss": 0.7929,
2769
+ "step": 456
2770
+ },
2771
+ {
2772
+ "epoch": 0.65,
2773
+ "learning_rate": 0.00015365096756336756,
2774
+ "loss": 0.7714,
2775
+ "step": 457
2776
+ },
2777
+ {
2778
+ "epoch": 0.65,
2779
+ "learning_rate": 0.00015346119193283313,
2780
+ "loss": 0.7547,
2781
+ "step": 458
2782
+ },
2783
+ {
2784
+ "epoch": 0.65,
2785
+ "learning_rate": 0.0001532711463271077,
2786
+ "loss": 0.8495,
2787
+ "step": 459
2788
+ },
2789
+ {
2790
+ "epoch": 0.65,
2791
+ "learning_rate": 0.000153080831705908,
2792
+ "loss": 0.8482,
2793
+ "step": 460
2794
+ },
2795
+ {
2796
+ "epoch": 0.65,
2797
+ "learning_rate": 0.00015289024903030924,
2798
+ "loss": 0.7974,
2799
+ "step": 461
2800
+ },
2801
+ {
2802
+ "epoch": 0.66,
2803
+ "learning_rate": 0.00015269939926274028,
2804
+ "loss": 0.8091,
2805
+ "step": 462
2806
+ },
2807
+ {
2808
+ "epoch": 0.66,
2809
+ "learning_rate": 0.00015250828336697876,
2810
+ "loss": 0.771,
2811
+ "step": 463
2812
+ },
2813
+ {
2814
+ "epoch": 0.66,
2815
+ "learning_rate": 0.00015231690230814633,
2816
+ "loss": 0.8292,
2817
+ "step": 464
2818
+ },
2819
+ {
2820
+ "epoch": 0.66,
2821
+ "learning_rate": 0.00015212525705270356,
2822
+ "loss": 0.8259,
2823
+ "step": 465
2824
+ },
2825
+ {
2826
+ "epoch": 0.66,
2827
+ "learning_rate": 0.00015193334856844528,
2828
+ "loss": 0.8219,
2829
+ "step": 466
2830
+ },
2831
+ {
2832
+ "epoch": 0.66,
2833
+ "learning_rate": 0.00015174117782449563,
2834
+ "loss": 0.8284,
2835
+ "step": 467
2836
+ },
2837
+ {
2838
+ "epoch": 0.66,
2839
+ "learning_rate": 0.00015154874579130308,
2840
+ "loss": 0.8191,
2841
+ "step": 468
2842
+ },
2843
+ {
2844
+ "epoch": 0.67,
2845
+ "learning_rate": 0.0001513560534406356,
2846
+ "loss": 0.8275,
2847
+ "step": 469
2848
+ },
2849
+ {
2850
+ "epoch": 0.67,
2851
+ "learning_rate": 0.0001511631017455758,
2852
+ "loss": 0.7894,
2853
+ "step": 470
2854
+ },
2855
+ {
2856
+ "epoch": 0.67,
2857
+ "learning_rate": 0.00015096989168051595,
2858
+ "loss": 0.8161,
2859
+ "step": 471
2860
+ },
2861
+ {
2862
+ "epoch": 0.67,
2863
+ "learning_rate": 0.00015077642422115295,
2864
+ "loss": 0.8246,
2865
+ "step": 472
2866
+ },
2867
+ {
2868
+ "epoch": 0.67,
2869
+ "learning_rate": 0.0001505827003444837,
2870
+ "loss": 0.8306,
2871
+ "step": 473
2872
+ },
2873
+ {
2874
+ "epoch": 0.67,
2875
+ "learning_rate": 0.00015038872102879981,
2876
+ "loss": 0.7859,
2877
+ "step": 474
2878
+ },
2879
+ {
2880
+ "epoch": 0.67,
2881
+ "learning_rate": 0.00015019448725368305,
2882
+ "loss": 0.7873,
2883
+ "step": 475
2884
+ },
2885
+ {
2886
+ "epoch": 0.68,
2887
+ "learning_rate": 0.00015000000000000001,
2888
+ "loss": 0.812,
2889
+ "step": 476
2890
+ },
2891
+ {
2892
+ "epoch": 0.68,
2893
+ "learning_rate": 0.00014980526024989738,
2894
+ "loss": 0.8562,
2895
+ "step": 477
2896
+ },
2897
+ {
2898
+ "epoch": 0.68,
2899
+ "learning_rate": 0.00014961026898679703,
2900
+ "loss": 0.8009,
2901
+ "step": 478
2902
+ },
2903
+ {
2904
+ "epoch": 0.68,
2905
+ "learning_rate": 0.0001494150271953908,
2906
+ "loss": 0.8271,
2907
+ "step": 479
2908
+ },
2909
+ {
2910
+ "epoch": 0.68,
2911
+ "learning_rate": 0.00014921953586163577,
2912
+ "loss": 0.8172,
2913
+ "step": 480
2914
+ },
2915
+ {
2916
+ "epoch": 0.68,
2917
+ "learning_rate": 0.0001490237959727492,
2918
+ "loss": 0.7966,
2919
+ "step": 481
2920
+ },
2921
+ {
2922
+ "epoch": 0.68,
2923
+ "learning_rate": 0.00014882780851720344,
2924
+ "loss": 0.8242,
2925
+ "step": 482
2926
+ },
2927
+ {
2928
+ "epoch": 0.69,
2929
+ "learning_rate": 0.00014863157448472122,
2930
+ "loss": 0.7957,
2931
+ "step": 483
2932
+ },
2933
+ {
2934
+ "epoch": 0.69,
2935
+ "learning_rate": 0.0001484350948662703,
2936
+ "loss": 0.8224,
2937
+ "step": 484
2938
+ },
2939
+ {
2940
+ "epoch": 0.69,
2941
+ "learning_rate": 0.00014823837065405863,
2942
+ "loss": 0.784,
2943
+ "step": 485
2944
+ },
2945
+ {
2946
+ "epoch": 0.69,
2947
+ "learning_rate": 0.0001480414028415295,
2948
+ "loss": 0.8139,
2949
+ "step": 486
2950
+ },
2951
+ {
2952
+ "epoch": 0.69,
2953
+ "learning_rate": 0.00014784419242335614,
2954
+ "loss": 0.7678,
2955
+ "step": 487
2956
+ },
2957
+ {
2958
+ "epoch": 0.69,
2959
+ "learning_rate": 0.00014764674039543718,
2960
+ "loss": 0.772,
2961
+ "step": 488
2962
+ },
2963
+ {
2964
+ "epoch": 0.69,
2965
+ "learning_rate": 0.00014744904775489107,
2966
+ "loss": 0.8111,
2967
+ "step": 489
2968
+ },
2969
+ {
2970
+ "epoch": 0.7,
2971
+ "learning_rate": 0.0001472511155000516,
2972
+ "loss": 0.8256,
2973
+ "step": 490
2974
+ },
2975
+ {
2976
+ "epoch": 0.7,
2977
+ "learning_rate": 0.00014705294463046248,
2978
+ "loss": 0.8292,
2979
+ "step": 491
2980
+ },
2981
+ {
2982
+ "epoch": 0.7,
2983
+ "learning_rate": 0.00014685453614687231,
2984
+ "loss": 0.7988,
2985
+ "step": 492
2986
+ },
2987
+ {
2988
+ "epoch": 0.7,
2989
+ "learning_rate": 0.0001466558910512298,
2990
+ "loss": 0.7818,
2991
+ "step": 493
2992
+ },
2993
+ {
2994
+ "epoch": 0.7,
2995
+ "learning_rate": 0.00014645701034667847,
2996
+ "loss": 0.8344,
2997
+ "step": 494
2998
+ },
2999
+ {
3000
+ "epoch": 0.7,
3001
+ "learning_rate": 0.0001462578950375516,
3002
+ "loss": 0.8063,
3003
+ "step": 495
3004
+ },
3005
+ {
3006
+ "epoch": 0.7,
3007
+ "learning_rate": 0.00014605854612936728,
3008
+ "loss": 0.7822,
3009
+ "step": 496
3010
+ },
3011
+ {
3012
+ "epoch": 0.71,
3013
+ "learning_rate": 0.00014585896462882317,
3014
+ "loss": 0.8224,
3015
+ "step": 497
3016
+ },
3017
+ {
3018
+ "epoch": 0.71,
3019
+ "learning_rate": 0.00014565915154379162,
3020
+ "loss": 0.8074,
3021
+ "step": 498
3022
+ },
3023
+ {
3024
+ "epoch": 0.71,
3025
+ "learning_rate": 0.00014545910788331433,
3026
+ "loss": 0.7734,
3027
+ "step": 499
3028
+ },
3029
+ {
3030
+ "epoch": 0.71,
3031
+ "learning_rate": 0.0001452588346575975,
3032
+ "loss": 0.7801,
3033
+ "step": 500
3034
+ },
3035
+ {
3036
+ "epoch": 0.71,
3037
+ "learning_rate": 0.00014505833287800662,
3038
+ "loss": 0.8452,
3039
+ "step": 501
3040
+ },
3041
+ {
3042
+ "epoch": 0.71,
3043
+ "learning_rate": 0.00014485760355706123,
3044
+ "loss": 0.7961,
3045
+ "step": 502
3046
+ },
3047
+ {
3048
+ "epoch": 0.71,
3049
+ "learning_rate": 0.00014465664770843008,
3050
+ "loss": 0.8104,
3051
+ "step": 503
3052
+ },
3053
+ {
3054
+ "epoch": 0.72,
3055
+ "learning_rate": 0.00014445546634692582,
3056
+ "loss": 0.8137,
3057
+ "step": 504
3058
+ },
3059
+ {
3060
+ "epoch": 0.72,
3061
+ "learning_rate": 0.0001442540604884999,
3062
+ "loss": 0.7949,
3063
+ "step": 505
3064
+ },
3065
+ {
3066
+ "epoch": 0.72,
3067
+ "learning_rate": 0.00014405243115023748,
3068
+ "loss": 0.7968,
3069
+ "step": 506
3070
+ },
3071
+ {
3072
+ "epoch": 0.72,
3073
+ "learning_rate": 0.00014385057935035228,
3074
+ "loss": 0.7942,
3075
+ "step": 507
3076
+ },
3077
+ {
3078
+ "epoch": 0.72,
3079
+ "learning_rate": 0.00014364850610818145,
3080
+ "loss": 0.7847,
3081
+ "step": 508
3082
+ },
3083
+ {
3084
+ "epoch": 0.72,
3085
+ "learning_rate": 0.0001434462124441804,
3086
+ "loss": 0.7803,
3087
+ "step": 509
3088
+ },
3089
+ {
3090
+ "epoch": 0.72,
3091
+ "learning_rate": 0.00014324369937991765,
3092
+ "loss": 0.9007,
3093
+ "step": 510
3094
+ },
3095
+ {
3096
+ "epoch": 0.73,
3097
+ "learning_rate": 0.00014304096793806958,
3098
+ "loss": 0.8213,
3099
+ "step": 511
3100
+ },
3101
+ {
3102
+ "epoch": 0.73,
3103
+ "learning_rate": 0.0001428380191424156,
3104
+ "loss": 0.8121,
3105
+ "step": 512
3106
+ },
3107
+ {
3108
+ "epoch": 0.73,
3109
+ "learning_rate": 0.00014263485401783252,
3110
+ "loss": 0.7978,
3111
+ "step": 513
3112
+ },
3113
+ {
3114
+ "epoch": 0.73,
3115
+ "learning_rate": 0.00014243147359028967,
3116
+ "loss": 0.8319,
3117
+ "step": 514
3118
+ },
3119
+ {
3120
+ "epoch": 0.73,
3121
+ "learning_rate": 0.00014222787888684363,
3122
+ "loss": 0.7793,
3123
+ "step": 515
3124
+ },
3125
+ {
3126
+ "epoch": 0.73,
3127
+ "learning_rate": 0.0001420240709356331,
3128
+ "loss": 0.7744,
3129
+ "step": 516
3130
+ },
3131
+ {
3132
+ "epoch": 0.73,
3133
+ "learning_rate": 0.00014182005076587365,
3134
+ "loss": 0.769,
3135
+ "step": 517
3136
+ },
3137
+ {
3138
+ "epoch": 0.74,
3139
+ "learning_rate": 0.00014161581940785252,
3140
+ "loss": 0.8161,
3141
+ "step": 518
3142
+ },
3143
+ {
3144
+ "epoch": 0.74,
3145
+ "learning_rate": 0.0001414113778929234,
3146
+ "loss": 0.7841,
3147
+ "step": 519
3148
+ },
3149
+ {
3150
+ "epoch": 0.74,
3151
+ "learning_rate": 0.00014120672725350137,
3152
+ "loss": 0.7876,
3153
+ "step": 520
3154
+ },
3155
+ {
3156
+ "epoch": 0.74,
3157
+ "learning_rate": 0.00014100186852305743,
3158
+ "loss": 0.7723,
3159
+ "step": 521
3160
+ },
3161
+ {
3162
+ "epoch": 0.74,
3163
+ "learning_rate": 0.00014079680273611358,
3164
+ "loss": 0.8469,
3165
+ "step": 522
3166
+ },
3167
+ {
3168
+ "epoch": 0.74,
3169
+ "learning_rate": 0.0001405915309282373,
3170
+ "loss": 0.754,
3171
+ "step": 523
3172
+ },
3173
+ {
3174
+ "epoch": 0.74,
3175
+ "learning_rate": 0.00014038605413603652,
3176
+ "loss": 0.8394,
3177
+ "step": 524
3178
+ },
3179
+ {
3180
+ "epoch": 0.75,
3181
+ "learning_rate": 0.00014018037339715437,
3182
+ "loss": 0.813,
3183
+ "step": 525
3184
+ },
3185
+ {
3186
+ "epoch": 0.75,
3187
+ "learning_rate": 0.00013997448975026382,
3188
+ "loss": 0.7784,
3189
+ "step": 526
3190
+ },
3191
+ {
3192
+ "epoch": 0.75,
3193
+ "learning_rate": 0.00013976840423506257,
3194
+ "loss": 0.8189,
3195
+ "step": 527
3196
+ },
3197
+ {
3198
+ "epoch": 0.75,
3199
+ "learning_rate": 0.0001395621178922677,
3200
+ "loss": 0.7928,
3201
+ "step": 528
3202
+ },
3203
+ {
3204
+ "epoch": 0.75,
3205
+ "eval_loss": 0.8294777870178223,
3206
+ "eval_runtime": 58.0411,
3207
+ "eval_samples_per_second": 6.892,
3208
+ "eval_steps_per_second": 3.446,
3209
+ "step": 528
3210
+ },
3211
+ {
3212
+ "epoch": 0.75,
3213
+ "learning_rate": 0.00013935563176361042,
3214
+ "loss": 0.7904,
3215
+ "step": 529
3216
+ },
3217
+ {
3218
+ "epoch": 0.75,
3219
+ "learning_rate": 0.00013914894689183097,
3220
+ "loss": 0.7809,
3221
+ "step": 530
3222
+ },
3223
+ {
3224
+ "epoch": 0.75,
3225
+ "learning_rate": 0.00013894206432067308,
3226
+ "loss": 0.7594,
3227
+ "step": 531
3228
+ },
3229
+ {
3230
+ "epoch": 0.76,
3231
+ "learning_rate": 0.00013873498509487902,
3232
+ "loss": 0.8133,
3233
+ "step": 532
3234
+ },
3235
+ {
3236
+ "epoch": 0.76,
3237
+ "learning_rate": 0.000138527710260184,
3238
+ "loss": 0.7861,
3239
+ "step": 533
3240
+ },
3241
+ {
3242
+ "epoch": 0.76,
3243
+ "learning_rate": 0.00013832024086331103,
3244
+ "loss": 0.798,
3245
+ "step": 534
3246
+ },
3247
+ {
3248
+ "epoch": 0.76,
3249
+ "learning_rate": 0.0001381125779519658,
3250
+ "loss": 0.8223,
3251
+ "step": 535
3252
+ },
3253
+ {
3254
+ "epoch": 0.76,
3255
+ "learning_rate": 0.00013790472257483108,
3256
+ "loss": 0.7924,
3257
+ "step": 536
3258
+ },
3259
+ {
3260
+ "epoch": 0.76,
3261
+ "learning_rate": 0.00013769667578156165,
3262
+ "loss": 0.8164,
3263
+ "step": 537
3264
+ },
3265
+ {
3266
+ "epoch": 0.76,
3267
+ "learning_rate": 0.00013748843862277898,
3268
+ "loss": 0.8427,
3269
+ "step": 538
3270
+ },
3271
+ {
3272
+ "epoch": 0.77,
3273
+ "learning_rate": 0.00013728001215006574,
3274
+ "loss": 0.8305,
3275
+ "step": 539
3276
+ },
3277
+ {
3278
+ "epoch": 0.77,
3279
+ "learning_rate": 0.0001370713974159607,
3280
+ "loss": 0.805,
3281
+ "step": 540
3282
+ },
3283
+ {
3284
+ "epoch": 0.77,
3285
+ "learning_rate": 0.0001368625954739534,
3286
+ "loss": 0.7933,
3287
+ "step": 541
3288
+ },
3289
+ {
3290
+ "epoch": 0.77,
3291
+ "learning_rate": 0.00013665360737847857,
3292
+ "loss": 0.8358,
3293
+ "step": 542
3294
+ },
3295
+ {
3296
+ "epoch": 0.77,
3297
+ "learning_rate": 0.00013644443418491125,
3298
+ "loss": 0.813,
3299
+ "step": 543
3300
+ },
3301
+ {
3302
+ "epoch": 0.77,
3303
+ "learning_rate": 0.00013623507694956102,
3304
+ "loss": 0.7695,
3305
+ "step": 544
3306
+ },
3307
+ {
3308
+ "epoch": 0.77,
3309
+ "learning_rate": 0.0001360255367296669,
3310
+ "loss": 0.825,
3311
+ "step": 545
3312
+ },
3313
+ {
3314
+ "epoch": 0.78,
3315
+ "learning_rate": 0.00013581581458339207,
3316
+ "loss": 0.7928,
3317
+ "step": 546
3318
+ },
3319
+ {
3320
+ "epoch": 0.78,
3321
+ "learning_rate": 0.00013560591156981831,
3322
+ "loss": 0.8634,
3323
+ "step": 547
3324
+ },
3325
+ {
3326
+ "epoch": 0.78,
3327
+ "learning_rate": 0.00013539582874894083,
3328
+ "loss": 0.7986,
3329
+ "step": 548
3330
+ },
3331
+ {
3332
+ "epoch": 0.78,
3333
+ "learning_rate": 0.00013518556718166282,
3334
+ "loss": 0.8162,
3335
+ "step": 549
3336
+ },
3337
+ {
3338
+ "epoch": 0.78,
3339
+ "learning_rate": 0.00013497512792979012,
3340
+ "loss": 0.8048,
3341
+ "step": 550
3342
+ },
3343
+ {
3344
+ "epoch": 0.78,
3345
+ "learning_rate": 0.0001347645120560259,
3346
+ "loss": 0.8305,
3347
+ "step": 551
3348
+ },
3349
+ {
3350
+ "epoch": 0.78,
3351
+ "learning_rate": 0.00013455372062396524,
3352
+ "loss": 0.7757,
3353
+ "step": 552
3354
+ },
3355
+ {
3356
+ "epoch": 0.79,
3357
+ "learning_rate": 0.00013434275469808974,
3358
+ "loss": 0.8034,
3359
+ "step": 553
3360
+ },
3361
+ {
3362
+ "epoch": 0.79,
3363
+ "learning_rate": 0.0001341316153437623,
3364
+ "loss": 0.8023,
3365
+ "step": 554
3366
+ },
3367
+ {
3368
+ "epoch": 0.79,
3369
+ "learning_rate": 0.0001339203036272215,
3370
+ "loss": 0.8447,
3371
+ "step": 555
3372
+ },
3373
+ {
3374
+ "epoch": 0.79,
3375
+ "learning_rate": 0.00013370882061557635,
3376
+ "loss": 0.8172,
3377
+ "step": 556
3378
+ },
3379
+ {
3380
+ "epoch": 0.79,
3381
+ "learning_rate": 0.00013349716737680092,
3382
+ "loss": 0.8713,
3383
+ "step": 557
3384
+ },
3385
+ {
3386
+ "epoch": 0.79,
3387
+ "learning_rate": 0.00013328534497972894,
3388
+ "loss": 0.7882,
3389
+ "step": 558
3390
+ },
3391
+ {
3392
+ "epoch": 0.79,
3393
+ "learning_rate": 0.00013307335449404836,
3394
+ "loss": 0.812,
3395
+ "step": 559
3396
+ },
3397
+ {
3398
+ "epoch": 0.8,
3399
+ "learning_rate": 0.0001328611969902959,
3400
+ "loss": 0.8051,
3401
+ "step": 560
3402
+ },
3403
+ {
3404
+ "epoch": 0.8,
3405
+ "learning_rate": 0.00013264887353985175,
3406
+ "loss": 0.7853,
3407
+ "step": 561
3408
+ },
3409
+ {
3410
+ "epoch": 0.8,
3411
+ "learning_rate": 0.00013243638521493424,
3412
+ "loss": 0.8625,
3413
+ "step": 562
3414
+ },
3415
+ {
3416
+ "epoch": 0.8,
3417
+ "learning_rate": 0.00013222373308859406,
3418
+ "loss": 0.7404,
3419
+ "step": 563
3420
+ },
3421
+ {
3422
+ "epoch": 0.8,
3423
+ "learning_rate": 0.00013201091823470936,
3424
+ "loss": 0.7813,
3425
+ "step": 564
3426
+ },
3427
+ {
3428
+ "epoch": 0.8,
3429
+ "learning_rate": 0.00013179794172797976,
3430
+ "loss": 0.7803,
3431
+ "step": 565
3432
+ },
3433
+ {
3434
+ "epoch": 0.8,
3435
+ "learning_rate": 0.00013158480464392144,
3436
+ "loss": 0.8549,
3437
+ "step": 566
3438
+ },
3439
+ {
3440
+ "epoch": 0.81,
3441
+ "learning_rate": 0.00013137150805886147,
3442
+ "loss": 0.7503,
3443
+ "step": 567
3444
+ },
3445
+ {
3446
+ "epoch": 0.81,
3447
+ "learning_rate": 0.0001311580530499322,
3448
+ "loss": 0.8602,
3449
+ "step": 568
3450
+ },
3451
+ {
3452
+ "epoch": 0.81,
3453
+ "learning_rate": 0.0001309444406950663,
3454
+ "loss": 0.8061,
3455
+ "step": 569
3456
+ },
3457
+ {
3458
+ "epoch": 0.81,
3459
+ "learning_rate": 0.00013073067207299073,
3460
+ "loss": 0.8279,
3461
+ "step": 570
3462
+ },
3463
+ {
3464
+ "epoch": 0.81,
3465
+ "learning_rate": 0.00013051674826322176,
3466
+ "loss": 0.7735,
3467
+ "step": 571
3468
+ },
3469
+ {
3470
+ "epoch": 0.81,
3471
+ "learning_rate": 0.0001303026703460594,
3472
+ "loss": 0.7717,
3473
+ "step": 572
3474
+ },
3475
+ {
3476
+ "epoch": 0.81,
3477
+ "learning_rate": 0.00013008843940258164,
3478
+ "loss": 0.7514,
3479
+ "step": 573
3480
+ },
3481
+ {
3482
+ "epoch": 0.82,
3483
+ "learning_rate": 0.00012987405651463952,
3484
+ "loss": 0.7604,
3485
+ "step": 574
3486
+ },
3487
+ {
3488
+ "epoch": 0.82,
3489
+ "learning_rate": 0.00012965952276485128,
3490
+ "loss": 0.7884,
3491
+ "step": 575
3492
+ },
3493
+ {
3494
+ "epoch": 0.82,
3495
+ "learning_rate": 0.00012944483923659693,
3496
+ "loss": 0.8118,
3497
+ "step": 576
3498
+ },
3499
+ {
3500
+ "epoch": 0.82,
3501
+ "learning_rate": 0.00012923000701401297,
3502
+ "loss": 0.8253,
3503
+ "step": 577
3504
+ },
3505
+ {
3506
+ "epoch": 0.82,
3507
+ "learning_rate": 0.00012901502718198663,
3508
+ "loss": 0.7794,
3509
+ "step": 578
3510
+ },
3511
+ {
3512
+ "epoch": 0.82,
3513
+ "learning_rate": 0.0001287999008261508,
3514
+ "loss": 0.8049,
3515
+ "step": 579
3516
+ },
3517
+ {
3518
+ "epoch": 0.82,
3519
+ "learning_rate": 0.00012858462903287814,
3520
+ "loss": 0.8225,
3521
+ "step": 580
3522
+ },
3523
+ {
3524
+ "epoch": 0.83,
3525
+ "learning_rate": 0.00012836921288927574,
3526
+ "loss": 0.7848,
3527
+ "step": 581
3528
+ },
3529
+ {
3530
+ "epoch": 0.83,
3531
+ "learning_rate": 0.00012815365348317975,
3532
+ "loss": 0.8081,
3533
+ "step": 582
3534
+ },
3535
+ {
3536
+ "epoch": 0.83,
3537
+ "learning_rate": 0.00012793795190314973,
3538
+ "loss": 0.7644,
3539
+ "step": 583
3540
+ },
3541
+ {
3542
+ "epoch": 0.83,
3543
+ "learning_rate": 0.00012772210923846317,
3544
+ "loss": 0.8084,
3545
+ "step": 584
3546
+ },
3547
+ {
3548
+ "epoch": 0.83,
3549
+ "learning_rate": 0.00012750612657911012,
3550
+ "loss": 0.7859,
3551
+ "step": 585
3552
+ },
3553
+ {
3554
+ "epoch": 0.83,
3555
+ "learning_rate": 0.0001272900050157875,
3556
+ "loss": 0.8209,
3557
+ "step": 586
3558
+ },
3559
+ {
3560
+ "epoch": 0.83,
3561
+ "learning_rate": 0.00012707374563989375,
3562
+ "loss": 0.8092,
3563
+ "step": 587
3564
+ },
3565
+ {
3566
+ "epoch": 0.84,
3567
+ "learning_rate": 0.00012685734954352327,
3568
+ "loss": 0.8402,
3569
+ "step": 588
3570
+ },
3571
+ {
3572
+ "epoch": 0.84,
3573
+ "learning_rate": 0.0001266408178194608,
3574
+ "loss": 0.7294,
3575
+ "step": 589
3576
+ },
3577
+ {
3578
+ "epoch": 0.84,
3579
+ "learning_rate": 0.00012642415156117605,
3580
+ "loss": 0.8243,
3581
+ "step": 590
3582
+ },
3583
+ {
3584
+ "epoch": 0.84,
3585
+ "learning_rate": 0.0001262073518628181,
3586
+ "loss": 0.8151,
3587
+ "step": 591
3588
+ },
3589
+ {
3590
+ "epoch": 0.84,
3591
+ "learning_rate": 0.00012599041981920995,
3592
+ "loss": 0.7833,
3593
+ "step": 592
3594
+ },
3595
+ {
3596
+ "epoch": 0.84,
3597
+ "learning_rate": 0.00012577335652584284,
3598
+ "loss": 0.8321,
3599
+ "step": 593
3600
+ },
3601
+ {
3602
+ "epoch": 0.84,
3603
+ "learning_rate": 0.00012555616307887086,
3604
+ "loss": 0.7716,
3605
+ "step": 594
3606
+ },
3607
+ {
3608
+ "epoch": 0.85,
3609
+ "learning_rate": 0.00012533884057510538,
3610
+ "loss": 0.7746,
3611
+ "step": 595
3612
+ },
3613
+ {
3614
+ "epoch": 0.85,
3615
+ "learning_rate": 0.00012512139011200947,
3616
+ "loss": 0.8502,
3617
+ "step": 596
3618
+ },
3619
+ {
3620
+ "epoch": 0.85,
3621
+ "learning_rate": 0.00012490381278769242,
3622
+ "loss": 0.8417,
3623
+ "step": 597
3624
+ },
3625
+ {
3626
+ "epoch": 0.85,
3627
+ "learning_rate": 0.00012468610970090411,
3628
+ "loss": 0.8002,
3629
+ "step": 598
3630
+ },
3631
+ {
3632
+ "epoch": 0.85,
3633
+ "learning_rate": 0.00012446828195102956,
3634
+ "loss": 0.762,
3635
+ "step": 599
3636
+ },
3637
+ {
3638
+ "epoch": 0.85,
3639
+ "learning_rate": 0.00012425033063808328,
3640
+ "loss": 0.8302,
3641
+ "step": 600
3642
+ },
3643
+ {
3644
+ "epoch": 0.85,
3645
+ "learning_rate": 0.00012403225686270384,
3646
+ "loss": 0.8135,
3647
+ "step": 601
3648
+ },
3649
+ {
3650
+ "epoch": 0.86,
3651
+ "learning_rate": 0.00012381406172614812,
3652
+ "loss": 0.7791,
3653
+ "step": 602
3654
+ },
3655
+ {
3656
+ "epoch": 0.86,
3657
+ "learning_rate": 0.000123595746330286,
3658
+ "loss": 0.7753,
3659
+ "step": 603
3660
+ },
3661
+ {
3662
+ "epoch": 0.86,
3663
+ "learning_rate": 0.0001233773117775946,
3664
+ "loss": 0.7365,
3665
+ "step": 604
3666
+ },
3667
+ {
3668
+ "epoch": 0.86,
3669
+ "learning_rate": 0.0001231587591711527,
3670
+ "loss": 0.8144,
3671
+ "step": 605
3672
+ },
3673
+ {
3674
+ "epoch": 0.86,
3675
+ "learning_rate": 0.00012294008961463539,
3676
+ "loss": 0.7906,
3677
+ "step": 606
3678
+ },
3679
+ {
3680
+ "epoch": 0.86,
3681
+ "learning_rate": 0.00012272130421230818,
3682
+ "loss": 0.8327,
3683
+ "step": 607
3684
+ },
3685
+ {
3686
+ "epoch": 0.86,
3687
+ "learning_rate": 0.0001225024040690218,
3688
+ "loss": 0.7626,
3689
+ "step": 608
3690
+ },
3691
+ {
3692
+ "epoch": 0.87,
3693
+ "learning_rate": 0.00012228339029020624,
3694
+ "loss": 0.8493,
3695
+ "step": 609
3696
+ },
3697
+ {
3698
+ "epoch": 0.87,
3699
+ "learning_rate": 0.00012206426398186534,
3700
+ "loss": 0.7778,
3701
+ "step": 610
3702
+ },
3703
+ {
3704
+ "epoch": 0.87,
3705
+ "learning_rate": 0.00012184502625057139,
3706
+ "loss": 0.805,
3707
+ "step": 611
3708
+ },
3709
+ {
3710
+ "epoch": 0.87,
3711
+ "learning_rate": 0.00012162567820345912,
3712
+ "loss": 0.8099,
3713
+ "step": 612
3714
+ },
3715
+ {
3716
+ "epoch": 0.87,
3717
+ "learning_rate": 0.00012140622094822054,
3718
+ "loss": 0.7626,
3719
+ "step": 613
3720
+ },
3721
+ {
3722
+ "epoch": 0.87,
3723
+ "learning_rate": 0.00012118665559309906,
3724
+ "loss": 0.8807,
3725
+ "step": 614
3726
+ },
3727
+ {
3728
+ "epoch": 0.87,
3729
+ "learning_rate": 0.00012096698324688392,
3730
+ "loss": 0.8007,
3731
+ "step": 615
3732
+ },
3733
+ {
3734
+ "epoch": 0.88,
3735
+ "learning_rate": 0.00012074720501890484,
3736
+ "loss": 0.7872,
3737
+ "step": 616
3738
+ },
3739
+ {
3740
+ "epoch": 0.88,
3741
+ "learning_rate": 0.00012052732201902608,
3742
+ "loss": 0.7814,
3743
+ "step": 617
3744
+ },
3745
+ {
3746
+ "epoch": 0.88,
3747
+ "learning_rate": 0.00012030733535764107,
3748
+ "loss": 0.8192,
3749
+ "step": 618
3750
+ },
3751
+ {
3752
+ "epoch": 0.88,
3753
+ "learning_rate": 0.0001200872461456667,
3754
+ "loss": 0.8097,
3755
+ "step": 619
3756
+ },
3757
+ {
3758
+ "epoch": 0.88,
3759
+ "learning_rate": 0.0001198670554945377,
3760
+ "loss": 0.7605,
3761
+ "step": 620
3762
+ },
3763
+ {
3764
+ "epoch": 0.88,
3765
+ "learning_rate": 0.00011964676451620112,
3766
+ "loss": 0.817,
3767
+ "step": 621
3768
+ },
3769
+ {
3770
+ "epoch": 0.88,
3771
+ "learning_rate": 0.00011942637432311059,
3772
+ "loss": 0.8013,
3773
+ "step": 622
3774
+ },
3775
+ {
3776
+ "epoch": 0.88,
3777
+ "learning_rate": 0.00011920588602822083,
3778
+ "loss": 0.7394,
3779
+ "step": 623
3780
+ },
3781
+ {
3782
+ "epoch": 0.89,
3783
+ "learning_rate": 0.00011898530074498194,
3784
+ "loss": 0.8615,
3785
+ "step": 624
3786
+ },
3787
+ {
3788
+ "epoch": 0.89,
3789
+ "learning_rate": 0.00011876461958733381,
3790
+ "loss": 0.745,
3791
+ "step": 625
3792
+ },
3793
+ {
3794
+ "epoch": 0.89,
3795
+ "learning_rate": 0.00011854384366970046,
3796
+ "loss": 0.8218,
3797
+ "step": 626
3798
+ },
3799
+ {
3800
+ "epoch": 0.89,
3801
+ "learning_rate": 0.00011832297410698447,
3802
+ "loss": 0.776,
3803
+ "step": 627
3804
+ },
3805
+ {
3806
+ "epoch": 0.89,
3807
+ "learning_rate": 0.00011810201201456134,
3808
+ "loss": 0.781,
3809
+ "step": 628
3810
+ },
3811
+ {
3812
+ "epoch": 0.89,
3813
+ "learning_rate": 0.00011788095850827381,
3814
+ "loss": 0.7886,
3815
+ "step": 629
3816
+ },
3817
+ {
3818
+ "epoch": 0.89,
3819
+ "learning_rate": 0.00011765981470442624,
3820
+ "loss": 0.7804,
3821
+ "step": 630
3822
+ },
3823
+ {
3824
+ "epoch": 0.9,
3825
+ "learning_rate": 0.00011743858171977899,
3826
+ "loss": 0.7484,
3827
+ "step": 631
3828
+ },
3829
+ {
3830
+ "epoch": 0.9,
3831
+ "learning_rate": 0.00011721726067154282,
3832
+ "loss": 0.7553,
3833
+ "step": 632
3834
+ },
3835
+ {
3836
+ "epoch": 0.9,
3837
+ "learning_rate": 0.00011699585267737317,
3838
+ "loss": 0.8006,
3839
+ "step": 633
3840
+ },
3841
+ {
3842
+ "epoch": 0.9,
3843
+ "learning_rate": 0.00011677435885536452,
3844
+ "loss": 0.7649,
3845
+ "step": 634
3846
+ },
3847
+ {
3848
+ "epoch": 0.9,
3849
+ "learning_rate": 0.00011655278032404489,
3850
+ "loss": 0.7723,
3851
+ "step": 635
3852
+ },
3853
+ {
3854
+ "epoch": 0.9,
3855
+ "learning_rate": 0.00011633111820236991,
3856
+ "loss": 0.8024,
3857
+ "step": 636
3858
+ },
3859
+ {
3860
+ "epoch": 0.9,
3861
+ "learning_rate": 0.00011610937360971747,
3862
+ "loss": 0.7931,
3863
+ "step": 637
3864
+ },
3865
+ {
3866
+ "epoch": 0.91,
3867
+ "learning_rate": 0.00011588754766588188,
3868
+ "loss": 0.847,
3869
+ "step": 638
3870
+ },
3871
+ {
3872
+ "epoch": 0.91,
3873
+ "learning_rate": 0.00011566564149106822,
3874
+ "loss": 0.7786,
3875
+ "step": 639
3876
+ },
3877
+ {
3878
+ "epoch": 0.91,
3879
+ "learning_rate": 0.00011544365620588688,
3880
+ "loss": 0.8177,
3881
+ "step": 640
3882
+ },
3883
+ {
3884
+ "epoch": 0.91,
3885
+ "learning_rate": 0.00011522159293134758,
3886
+ "loss": 0.7655,
3887
+ "step": 641
3888
+ },
3889
+ {
3890
+ "epoch": 0.91,
3891
+ "learning_rate": 0.00011499945278885395,
3892
+ "loss": 0.7785,
3893
+ "step": 642
3894
+ },
3895
+ {
3896
+ "epoch": 0.91,
3897
+ "learning_rate": 0.00011477723690019788,
3898
+ "loss": 0.8031,
3899
+ "step": 643
3900
+ },
3901
+ {
3902
+ "epoch": 0.91,
3903
+ "learning_rate": 0.0001145549463875536,
3904
+ "loss": 0.844,
3905
+ "step": 644
3906
+ },
3907
+ {
3908
+ "epoch": 0.92,
3909
+ "learning_rate": 0.00011433258237347235,
3910
+ "loss": 0.7841,
3911
+ "step": 645
3912
+ },
3913
+ {
3914
+ "epoch": 0.92,
3915
+ "learning_rate": 0.00011411014598087644,
3916
+ "loss": 0.8228,
3917
+ "step": 646
3918
+ },
3919
+ {
3920
+ "epoch": 0.92,
3921
+ "learning_rate": 0.00011388763833305371,
3922
+ "loss": 0.8026,
3923
+ "step": 647
3924
+ },
3925
+ {
3926
+ "epoch": 0.92,
3927
+ "learning_rate": 0.00011366506055365194,
3928
+ "loss": 0.8072,
3929
+ "step": 648
3930
+ },
3931
+ {
3932
+ "epoch": 0.92,
3933
+ "learning_rate": 0.00011344241376667284,
3934
+ "loss": 0.8113,
3935
+ "step": 649
3936
+ },
3937
+ {
3938
+ "epoch": 0.92,
3939
+ "learning_rate": 0.00011321969909646683,
3940
+ "loss": 0.8151,
3941
+ "step": 650
3942
+ },
3943
+ {
3944
+ "epoch": 0.92,
3945
+ "learning_rate": 0.00011299691766772709,
3946
+ "loss": 0.8332,
3947
+ "step": 651
3948
+ },
3949
+ {
3950
+ "epoch": 0.93,
3951
+ "learning_rate": 0.00011277407060548373,
3952
+ "loss": 0.7731,
3953
+ "step": 652
3954
+ },
3955
+ {
3956
+ "epoch": 0.93,
3957
+ "learning_rate": 0.00011255115903509861,
3958
+ "loss": 0.8198,
3959
+ "step": 653
3960
+ },
3961
+ {
3962
+ "epoch": 0.93,
3963
+ "learning_rate": 0.00011232818408225909,
3964
+ "loss": 0.7763,
3965
+ "step": 654
3966
+ },
3967
+ {
3968
+ "epoch": 0.93,
3969
+ "learning_rate": 0.0001121051468729728,
3970
+ "loss": 0.8101,
3971
+ "step": 655
3972
+ },
3973
+ {
3974
+ "epoch": 0.93,
3975
+ "learning_rate": 0.00011188204853356163,
3976
+ "loss": 0.7844,
3977
+ "step": 656
3978
+ },
3979
+ {
3980
+ "epoch": 0.93,
3981
+ "learning_rate": 0.00011165889019065618,
3982
+ "loss": 0.8057,
3983
+ "step": 657
3984
+ },
3985
+ {
3986
+ "epoch": 0.93,
3987
+ "learning_rate": 0.0001114356729711902,
3988
+ "loss": 0.7925,
3989
+ "step": 658
3990
+ },
3991
+ {
3992
+ "epoch": 0.94,
3993
+ "learning_rate": 0.00011121239800239458,
3994
+ "loss": 0.7849,
3995
+ "step": 659
3996
+ },
3997
+ {
3998
+ "epoch": 0.94,
3999
+ "learning_rate": 0.00011098906641179194,
4000
+ "loss": 0.811,
4001
+ "step": 660
4002
+ },
4003
+ {
4004
+ "epoch": 0.94,
4005
+ "learning_rate": 0.00011076567932719088,
4006
+ "loss": 0.7957,
4007
+ "step": 661
4008
+ },
4009
+ {
4010
+ "epoch": 0.94,
4011
+ "learning_rate": 0.00011054223787668008,
4012
+ "loss": 0.8227,
4013
+ "step": 662
4014
+ },
4015
+ {
4016
+ "epoch": 0.94,
4017
+ "learning_rate": 0.00011031874318862294,
4018
+ "loss": 0.8388,
4019
+ "step": 663
4020
+ },
4021
+ {
4022
+ "epoch": 0.94,
4023
+ "learning_rate": 0.00011009519639165162,
4024
+ "loss": 0.7837,
4025
+ "step": 664
4026
+ },
4027
+ {
4028
+ "epoch": 0.94,
4029
+ "learning_rate": 0.00010987159861466143,
4030
+ "loss": 0.8168,
4031
+ "step": 665
4032
+ },
4033
+ {
4034
+ "epoch": 0.95,
4035
+ "learning_rate": 0.00010964795098680512,
4036
+ "loss": 0.8015,
4037
+ "step": 666
4038
+ },
4039
+ {
4040
+ "epoch": 0.95,
4041
+ "learning_rate": 0.0001094242546374872,
4042
+ "loss": 0.8834,
4043
+ "step": 667
4044
+ },
4045
+ {
4046
+ "epoch": 0.95,
4047
+ "learning_rate": 0.00010920051069635822,
4048
+ "loss": 0.7445,
4049
+ "step": 668
4050
+ },
4051
+ {
4052
+ "epoch": 0.95,
4053
+ "learning_rate": 0.00010897672029330906,
4054
+ "loss": 0.84,
4055
+ "step": 669
4056
+ },
4057
+ {
4058
+ "epoch": 0.95,
4059
+ "learning_rate": 0.00010875288455846522,
4060
+ "loss": 0.8329,
4061
+ "step": 670
4062
+ },
4063
+ {
4064
+ "epoch": 0.95,
4065
+ "learning_rate": 0.00010852900462218117,
4066
+ "loss": 0.8016,
4067
+ "step": 671
4068
+ },
4069
+ {
4070
+ "epoch": 0.95,
4071
+ "learning_rate": 0.0001083050816150345,
4072
+ "loss": 0.7757,
4073
+ "step": 672
4074
+ },
4075
+ {
4076
+ "epoch": 0.96,
4077
+ "learning_rate": 0.0001080811166678204,
4078
+ "loss": 0.7858,
4079
+ "step": 673
4080
+ },
4081
+ {
4082
+ "epoch": 0.96,
4083
+ "learning_rate": 0.0001078571109115458,
4084
+ "loss": 0.7934,
4085
+ "step": 674
4086
+ },
4087
+ {
4088
+ "epoch": 0.96,
4089
+ "learning_rate": 0.00010763306547742375,
4090
+ "loss": 0.8384,
4091
+ "step": 675
4092
+ },
4093
+ {
4094
+ "epoch": 0.96,
4095
+ "learning_rate": 0.0001074089814968676,
4096
+ "loss": 0.7513,
4097
+ "step": 676
4098
+ },
4099
+ {
4100
+ "epoch": 0.96,
4101
+ "learning_rate": 0.00010718486010148547,
4102
+ "loss": 0.8118,
4103
+ "step": 677
4104
+ },
4105
+ {
4106
+ "epoch": 0.96,
4107
+ "learning_rate": 0.00010696070242307432,
4108
+ "loss": 0.7644,
4109
+ "step": 678
4110
+ },
4111
+ {
4112
+ "epoch": 0.96,
4113
+ "learning_rate": 0.00010673650959361439,
4114
+ "loss": 0.7744,
4115
+ "step": 679
4116
+ },
4117
+ {
4118
+ "epoch": 0.97,
4119
+ "learning_rate": 0.00010651228274526339,
4120
+ "loss": 0.8201,
4121
+ "step": 680
4122
+ },
4123
+ {
4124
+ "epoch": 0.97,
4125
+ "learning_rate": 0.00010628802301035085,
4126
+ "loss": 0.8065,
4127
+ "step": 681
4128
+ },
4129
+ {
4130
+ "epoch": 0.97,
4131
+ "learning_rate": 0.00010606373152137241,
4132
+ "loss": 0.851,
4133
+ "step": 682
4134
+ },
4135
+ {
4136
+ "epoch": 0.97,
4137
+ "learning_rate": 0.000105839409410984,
4138
+ "loss": 0.8175,
4139
+ "step": 683
4140
+ },
4141
+ {
4142
+ "epoch": 0.97,
4143
+ "learning_rate": 0.00010561505781199618,
4144
+ "loss": 0.7985,
4145
+ "step": 684
4146
+ },
4147
+ {
4148
+ "epoch": 0.97,
4149
+ "learning_rate": 0.00010539067785736856,
4150
+ "loss": 0.801,
4151
+ "step": 685
4152
+ },
4153
+ {
4154
+ "epoch": 0.97,
4155
+ "learning_rate": 0.00010516627068020373,
4156
+ "loss": 0.8111,
4157
+ "step": 686
4158
+ },
4159
+ {
4160
+ "epoch": 0.98,
4161
+ "learning_rate": 0.00010494183741374194,
4162
+ "loss": 0.8076,
4163
+ "step": 687
4164
+ },
4165
+ {
4166
+ "epoch": 0.98,
4167
+ "learning_rate": 0.0001047173791913551,
4168
+ "loss": 0.7516,
4169
+ "step": 688
4170
+ },
4171
+ {
4172
+ "epoch": 0.98,
4173
+ "learning_rate": 0.00010449289714654109,
4174
+ "loss": 0.7644,
4175
+ "step": 689
4176
+ },
4177
+ {
4178
+ "epoch": 0.98,
4179
+ "learning_rate": 0.00010426839241291828,
4180
+ "loss": 0.8133,
4181
+ "step": 690
4182
+ },
4183
+ {
4184
+ "epoch": 0.98,
4185
+ "learning_rate": 0.00010404386612421942,
4186
+ "loss": 0.8554,
4187
+ "step": 691
4188
+ },
4189
+ {
4190
+ "epoch": 0.98,
4191
+ "learning_rate": 0.0001038193194142862,
4192
+ "loss": 0.8334,
4193
+ "step": 692
4194
+ },
4195
+ {
4196
+ "epoch": 0.98,
4197
+ "learning_rate": 0.00010359475341706346,
4198
+ "loss": 0.8241,
4199
+ "step": 693
4200
+ },
4201
+ {
4202
+ "epoch": 0.99,
4203
+ "learning_rate": 0.00010337016926659333,
4204
+ "loss": 0.8197,
4205
+ "step": 694
4206
+ },
4207
+ {
4208
+ "epoch": 0.99,
4209
+ "learning_rate": 0.0001031455680970098,
4210
+ "loss": 0.8107,
4211
+ "step": 695
4212
+ },
4213
+ {
4214
+ "epoch": 0.99,
4215
+ "learning_rate": 0.00010292095104253259,
4216
+ "loss": 0.798,
4217
+ "step": 696
4218
+ },
4219
+ {
4220
+ "epoch": 0.99,
4221
+ "learning_rate": 0.00010269631923746176,
4222
+ "loss": 0.8339,
4223
+ "step": 697
4224
+ },
4225
+ {
4226
+ "epoch": 0.99,
4227
+ "learning_rate": 0.00010247167381617191,
4228
+ "loss": 0.7761,
4229
+ "step": 698
4230
+ },
4231
+ {
4232
+ "epoch": 0.99,
4233
+ "learning_rate": 0.00010224701591310625,
4234
+ "loss": 0.8617,
4235
+ "step": 699
4236
+ },
4237
+ {
4238
+ "epoch": 0.99,
4239
+ "learning_rate": 0.00010202234666277115,
4240
+ "loss": 0.7663,
4241
+ "step": 700
4242
+ },
4243
+ {
4244
+ "epoch": 1.0,
4245
+ "learning_rate": 0.00010179766719973023,
4246
+ "loss": 0.7603,
4247
+ "step": 701
4248
+ },
4249
+ {
4250
+ "epoch": 1.0,
4251
+ "learning_rate": 0.00010157297865859865,
4252
+ "loss": 0.7736,
4253
+ "step": 702
4254
+ },
4255
+ {
4256
+ "epoch": 1.0,
4257
+ "learning_rate": 0.0001013482821740375,
4258
+ "loss": 0.8078,
4259
+ "step": 703
4260
+ },
4261
+ {
4262
+ "epoch": 1.0,
4263
+ "learning_rate": 0.00010112357888074793,
4264
+ "loss": 0.8313,
4265
+ "step": 704
4266
+ },
4267
+ {
4268
+ "epoch": 1.0,
4269
+ "eval_loss": 0.8154956698417664,
4270
+ "eval_runtime": 58.0041,
4271
+ "eval_samples_per_second": 6.896,
4272
+ "eval_steps_per_second": 3.448,
4273
+ "step": 704
4274
+ }
4275
+ ],
4276
+ "logging_steps": 1,
4277
+ "max_steps": 1408,
4278
+ "num_input_tokens_seen": 0,
4279
+ "num_train_epochs": 2,
4280
+ "save_steps": 704,
4281
+ "total_flos": 2.3970054853676237e+17,
4282
+ "train_batch_size": 2,
4283
+ "trial_name": null,
4284
+ "trial_params": null
4285
+ }
checkpoint-704/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fdfce1d53b772a89b2df019b8f67a3ae168e5ca0b13341505c6bdac951bcb1c7
3
+ size 4731
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "NousResearch/Llama-2-7b-hf",
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": 4096,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 11008,
14
+ "max_position_embeddings": 4096,
15
+ "model_type": "llama",
16
+ "num_attention_heads": 32,
17
+ "num_hidden_layers": 32,
18
+ "num_key_value_heads": 32,
19
+ "pad_token_id": 0,
20
+ "pretraining_tp": 1,
21
+ "quantization_config": {
22
+ "bnb_4bit_compute_dtype": "float32",
23
+ "bnb_4bit_quant_type": "fp4",
24
+ "bnb_4bit_use_double_quant": false,
25
+ "llm_int8_enable_fp32_cpu_offload": false,
26
+ "llm_int8_has_fp16_weight": false,
27
+ "llm_int8_skip_modules": null,
28
+ "llm_int8_threshold": 6.0,
29
+ "load_in_4bit": false,
30
+ "load_in_8bit": true,
31
+ "quant_method": "bitsandbytes"
32
+ },
33
+ "rms_norm_eps": 1e-05,
34
+ "rope_scaling": null,
35
+ "rope_theta": 10000.0,
36
+ "tie_word_embeddings": false,
37
+ "torch_dtype": "float16",
38
+ "transformers_version": "4.36.2",
39
+ "use_cache": false,
40
+ "vocab_size": 32000
41
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": {
17
+ "content": "<unk>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "bos_token": "<s>",
31
+ "clean_up_tokenization_spaces": false,
32
+ "eos_token": "</s>",
33
+ "legacy": false,
34
+ "model_max_length": 1000000000000000019884624838656,
35
+ "pad_token": "<unk>",
36
+ "sp_model_kwargs": {},
37
+ "spaces_between_special_tokens": false,
38
+ "tokenizer_class": "LlamaTokenizer",
39
+ "trust_remote_code": false,
40
+ "unk_token": "<unk>",
41
+ "use_default_system_prompt": false,
42
+ "use_fast": true
43
+ }