Hugging Face's logo

dante123
/
AutoRE

TensorBoard
Safetensors
Model card Files
xet
 Metrics Community
1
dante123 commited on
Commit
e237099
·
verified ·
1 Parent(s): ab5df65

Upload 28 files

Browse files
Files changed (28) hide show
  1. subject/README.md +129 -0
  2. subject/adapter_config.json +26 -0
  3. subject/adapter_model.safetensors +3 -0
  4. subject/all_results.json +11 -0
  5. subject/checkpoint-5390/README.md +218 -0
  6. subject/checkpoint-5390/adapter_config.json +26 -0
  7. subject/checkpoint-5390/adapter_model.safetensors +3 -0
  8. subject/checkpoint-5390/optimizer.pt +3 -0
  9. subject/checkpoint-5390/rng_state.pth +3 -0
  10. subject/checkpoint-5390/scheduler.pt +3 -0
  11. subject/checkpoint-5390/special_tokens_map.json +30 -0
  12. subject/checkpoint-5390/tokenizer.model +3 -0
  13. subject/checkpoint-5390/tokenizer_config.json +43 -0
  14. subject/checkpoint-5390/trainer_state.json +3679 -0
  15. subject/checkpoint-5390/training_args.bin +3 -0
  16. subject/eval_results.json +7 -0
  17. subject/runs/Jan26_17-21-12_platform/events.out.tfevents.1706289812.platform +3 -0
  18. subject/runs/Jan29_08-21-25_platform/events.out.tfevents.1706516497.platform +3 -0
  19. subject/runs/Jan29_08-21-25_platform/events.out.tfevents.1706516849.platform +3 -0
  20. subject/special_tokens_map.json +30 -0
  21. subject/tokenizer.model +3 -0
  22. subject/tokenizer_config.json +43 -0
  23. subject/train_results.json +7 -0
  24. subject/trainer_log.jsonl +0 -0
  25. subject/trainer_state.json +3688 -0
  26. subject/training_args.bin +3 -0
  27. subject/training_eval_loss.png +0 -0
  28. subject/training_loss.png +0 -0
subject/README.md ADDED
@@ -0,0 +1,129 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ library_name: peft
4
+ tags:
5
+ - llama-factory
6
+ - lora
7
+ - generated_from_trainer
8
+ base_model: /workspace/xll/checkpoints/lmsys/vicuna-7b-v1.5
9
+ model-index:
10
+ - name: subject
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # subject
18
+
19
+ This model is a fine-tuned version of [/workspace/xll/checkpoints/lmsys/vicuna-7b-v1.5](https://huggingface.co//workspace/xll/checkpoints/lmsys/vicuna-7b-v1.5) on the vicuna_subject_test dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.4113
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+
38
+ The following `bitsandbytes` quantization config was used during training:
39
+ - quant_method: bitsandbytes
40
+ - load_in_8bit: False
41
+ - load_in_4bit: True
42
+ - llm_int8_threshold: 6.0
43
+ - llm_int8_skip_modules: None
44
+ - llm_int8_enable_fp32_cpu_offload: False
45
+ - llm_int8_has_fp16_weight: False
46
+ - bnb_4bit_quant_type: nf4
47
+ - bnb_4bit_use_double_quant: True
48
+ - bnb_4bit_compute_dtype: float16
49
+
50
+ ### Training hyperparameters
51
+
52
+ The following hyperparameters were used during training:
53
+ - learning_rate: 5e-05
54
+ - train_batch_size: 8
55
+ - eval_batch_size: 8
56
+ - seed: 42
57
+ - gradient_accumulation_steps: 4
58
+ - total_train_batch_size: 32
59
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
+ - lr_scheduler_type: cosine
61
+ - training_steps: 5300
62
+ - mixed_precision_training: Native AMP
63
+
64
+ ### Training results
65
+
66
+ | Training Loss | Epoch | Step | Validation Loss |
67
+ |:-------------:|:-----:|:----:|:---------------:|
68
+ | 0.5382 | 0.04 | 100 | 0.5621 |
69
+ | 0.4788 | 0.07 | 200 | 0.5201 |
70
+ | 0.4682 | 0.11 | 300 | 0.4971 |
71
+ | 0.4778 | 0.15 | 400 | 0.4731 |
72
+ | 0.4541 | 0.18 | 500 | 0.4687 |
73
+ | 0.4786 | 0.22 | 600 | 0.4500 |
74
+ | 0.3974 | 0.26 | 700 | 0.4457 |
75
+ | 0.4142 | 0.29 | 800 | 0.4460 |
76
+ | 0.4374 | 0.33 | 900 | 0.4420 |
77
+ | 0.4008 | 0.37 | 1000 | 0.4419 |
78
+ | 0.3979 | 0.4 | 1100 | 0.4333 |
79
+ | 0.4108 | 0.44 | 1200 | 0.4304 |
80
+ | 0.3578 | 0.48 | 1300 | 0.4255 |
81
+ | 0.3895 | 0.51 | 1400 | 0.4196 |
82
+ | 0.3725 | 0.55 | 1500 | 0.4203 |
83
+ | 0.3836 | 0.59 | 1600 | 0.4204 |
84
+ | 0.3784 | 0.62 | 1700 | 0.4183 |
85
+ | 0.369 | 0.66 | 1800 | 0.4111 |
86
+ | 0.3409 | 0.7 | 1900 | 0.4120 |
87
+ | 0.388 | 0.73 | 2000 | 0.4151 |
88
+ | 0.3608 | 0.77 | 2100 | 0.4093 |
89
+ | 0.3171 | 0.81 | 2200 | 0.4079 |
90
+ | 0.3581 | 0.84 | 2300 | 0.4119 |
91
+ | 0.3389 | 0.88 | 2400 | 0.4119 |
92
+ | 0.3302 | 0.92 | 2500 | 0.4022 |
93
+ | 0.3553 | 0.95 | 2600 | 0.4055 |
94
+ | 0.3586 | 0.99 | 2700 | 0.4049 |
95
+ | 0.3014 | 1.03 | 2800 | 0.4126 |
96
+ | 0.3354 | 1.06 | 2900 | 0.4082 |
97
+ | 0.2954 | 1.1 | 3000 | 0.4158 |
98
+ | 0.3023 | 1.14 | 3100 | 0.4050 |
99
+ | 0.2896 | 1.17 | 3200 | 0.4053 |
100
+ | 0.3339 | 1.21 | 3300 | 0.4054 |
101
+ | 0.3118 | 1.25 | 3400 | 0.3964 |
102
+ | 0.3289 | 1.28 | 3500 | 0.3991 |
103
+ | 0.2984 | 1.32 | 3600 | 0.4059 |
104
+ | 0.3277 | 1.36 | 3700 | 0.3980 |
105
+ | 0.3011 | 1.39 | 3800 | 0.4045 |
106
+ | 0.3194 | 1.43 | 3900 | 0.4015 |
107
+ | 0.2921 | 1.47 | 4000 | 0.4009 |
108
+ | 0.2917 | 1.5 | 4100 | 0.4019 |
109
+ | 0.2792 | 1.54 | 4200 | 0.4046 |
110
+ | 0.2886 | 1.58 | 4300 | 0.4055 |
111
+ | 0.2947 | 1.61 | 4400 | 0.4051 |
112
+ | 0.2975 | 1.65 | 4500 | 0.4067 |
113
+ | 0.3091 | 1.69 | 4600 | 0.3947 |
114
+ | 0.2908 | 1.72 | 4700 | 0.4033 |
115
+ | 0.2864 | 1.76 | 4800 | 0.4096 |
116
+ | 0.2788 | 1.8 | 4900 | 0.4069 |
117
+ | 0.2942 | 1.83 | 5000 | 0.3997 |
118
+ | 0.2736 | 1.87 | 5100 | 0.4099 |
119
+ | 0.2905 | 1.91 | 5200 | 0.3983 |
120
+ | 0.2848 | 1.94 | 5300 | 0.4000 |
121
+
122
+
123
+ ### Framework versions
124
+
125
+ - PEFT 0.7.0
126
+ - Transformers 4.37.1
127
+ - Pytorch 2.1.2+cu121
128
+ - Datasets 2.16.1
129
+ - Tokenizers 0.15.1
subject/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/workspace/xll/checkpoints/lmsys/vicuna-7b-v1.5",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 16,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 100,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "v_proj",
23
+ "q_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
subject/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35f7e7a0fcf39f4b0c2bef24ad378bbc90bb86d37c47387c3061ce240a1b2bd9
3
+ size 23199744
subject/all_results.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.98,
3
+ "eval_loss": 0.41133925318717957,
4
+ "eval_runtime": 345.8577,
5
+ "eval_samples_per_second": 13.829,
6
+ "eval_steps_per_second": 1.729,
7
+ "train_loss": 4.6051067095080594e-05,
8
+ "train_runtime": 5.408,
9
+ "train_samples_per_second": 15680.552,
10
+ "train_steps_per_second": 980.035
11
+ }
subject/checkpoint-5390/README.md ADDED
@@ -0,0 +1,218 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ base_model: /workspace/xll/checkpoints/lmsys/vicuna-7b-v1.5
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
+ The following `bitsandbytes` quantization config was used during training:
205
+ - quant_method: bitsandbytes
206
+ - load_in_8bit: False
207
+ - load_in_4bit: True
208
+ - llm_int8_threshold: 6.0
209
+ - llm_int8_skip_modules: None
210
+ - llm_int8_enable_fp32_cpu_offload: False
211
+ - llm_int8_has_fp16_weight: False
212
+ - bnb_4bit_quant_type: nf4
213
+ - bnb_4bit_use_double_quant: True
214
+ - bnb_4bit_compute_dtype: float16
215
+
216
+ ### Framework versions
217
+
218
+ - PEFT 0.7.0
subject/checkpoint-5390/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/workspace/xll/checkpoints/lmsys/vicuna-7b-v1.5",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 16,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 100,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "q_proj",
23
+ "v_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
subject/checkpoint-5390/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df0ab9036907f9015aa75d06bbd7d0bf166ad324ca9e5fae05c187a25eb46b0e
3
+ size 10027008
subject/checkpoint-5390/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea28621c81fe15e6eb1593bdd325a06bf854696b2b06bdca450471a4d57f7394
3
+ size 10788864
subject/checkpoint-5390/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebdfb55abddaaf38a1bfaabb2c0b362c13d92c96777005104cef718be67a3df5
3
+ size 14244
subject/checkpoint-5390/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb4125f67bff6149440a68863a525799fb3d7c602cceb540a815d63102bf5581
3
+ size 1064
subject/checkpoint-5390/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": false,
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
+ }
subject/checkpoint-5390/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
subject/checkpoint-5390/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": 4096,
35
+ "pad_token": "<unk>",
36
+ "padding_side": "right",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "split_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
subject/checkpoint-5390/trainer_state.json ADDED
@@ -0,0 +1,3679 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.9747206448067411,
5
+ "eval_steps": 100,
6
+ "global_step": 5390,
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": 0.0001999998159397344,
14
+ "loss": 1.4113,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 0.00019999926375961516,
20
+ "loss": 0.8357,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 0.00019999834346167496,
26
+ "loss": 0.7122,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.01,
31
+ "learning_rate": 0.0001999970550493016,
32
+ "loss": 0.6653,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.02,
37
+ "learning_rate": 0.000199995398527238,
38
+ "loss": 0.6152,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.02,
43
+ "learning_rate": 0.00019999337390158218,
44
+ "loss": 0.5963,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.03,
49
+ "learning_rate": 0.00019999098117978715,
50
+ "loss": 0.5923,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.03,
55
+ "learning_rate": 0.00019998822037066105,
56
+ "loss": 0.5783,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.03,
61
+ "learning_rate": 0.00019998509148436697,
62
+ "loss": 0.5529,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.04,
67
+ "learning_rate": 0.000199981594532423,
68
+ "loss": 0.5382,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.04,
73
+ "eval_loss": 0.5621132850646973,
74
+ "eval_runtime": 398.8595,
75
+ "eval_samples_per_second": 11.992,
76
+ "eval_steps_per_second": 2.999,
77
+ "step": 100
78
+ },
79
+ {
80
+ "epoch": 0.04,
81
+ "learning_rate": 0.00019997772952770216,
82
+ "loss": 0.5489,
83
+ "step": 110
84
+ },
85
+ {
86
+ "epoch": 0.04,
87
+ "learning_rate": 0.00019997349648443225,
88
+ "loss": 0.5268,
89
+ "step": 120
90
+ },
91
+ {
92
+ "epoch": 0.05,
93
+ "learning_rate": 0.00019996889541819602,
94
+ "loss": 0.525,
95
+ "step": 130
96
+ },
97
+ {
98
+ "epoch": 0.05,
99
+ "learning_rate": 0.00019996392634593092,
100
+ "loss": 0.4926,
101
+ "step": 140
102
+ },
103
+ {
104
+ "epoch": 0.05,
105
+ "learning_rate": 0.00019995858928592916,
106
+ "loss": 0.5181,
107
+ "step": 150
108
+ },
109
+ {
110
+ "epoch": 0.06,
111
+ "learning_rate": 0.00019995288425783754,
112
+ "loss": 0.5225,
113
+ "step": 160
114
+ },
115
+ {
116
+ "epoch": 0.06,
117
+ "learning_rate": 0.00019994681128265743,
118
+ "loss": 0.4993,
119
+ "step": 170
120
+ },
121
+ {
122
+ "epoch": 0.07,
123
+ "learning_rate": 0.00019994037038274467,
124
+ "loss": 0.5387,
125
+ "step": 180
126
+ },
127
+ {
128
+ "epoch": 0.07,
129
+ "learning_rate": 0.0001999335615818096,
130
+ "loss": 0.465,
131
+ "step": 190
132
+ },
133
+ {
134
+ "epoch": 0.07,
135
+ "learning_rate": 0.00019992638490491676,
136
+ "loss": 0.4788,
137
+ "step": 200
138
+ },
139
+ {
140
+ "epoch": 0.07,
141
+ "eval_loss": 0.5201168060302734,
142
+ "eval_runtime": 398.3332,
143
+ "eval_samples_per_second": 12.008,
144
+ "eval_steps_per_second": 3.003,
145
+ "step": 200
146
+ },
147
+ {
148
+ "epoch": 0.08,
149
+ "learning_rate": 0.00019991884037848497,
150
+ "loss": 0.451,
151
+ "step": 210
152
+ },
153
+ {
154
+ "epoch": 0.08,
155
+ "learning_rate": 0.00019991092803028725,
156
+ "loss": 0.4586,
157
+ "step": 220
158
+ },
159
+ {
160
+ "epoch": 0.08,
161
+ "learning_rate": 0.00019990264788945052,
162
+ "loss": 0.4894,
163
+ "step": 230
164
+ },
165
+ {
166
+ "epoch": 0.09,
167
+ "learning_rate": 0.00019989399998645568,
168
+ "loss": 0.4666,
169
+ "step": 240
170
+ },
171
+ {
172
+ "epoch": 0.09,
173
+ "learning_rate": 0.00019988498435313744,
174
+ "loss": 0.4763,
175
+ "step": 250
176
+ },
177
+ {
178
+ "epoch": 0.1,
179
+ "learning_rate": 0.00019987560102268422,
180
+ "loss": 0.4539,
181
+ "step": 260
182
+ },
183
+ {
184
+ "epoch": 0.1,
185
+ "learning_rate": 0.00019986585002963793,
186
+ "loss": 0.5133,
187
+ "step": 270
188
+ },
189
+ {
190
+ "epoch": 0.1,
191
+ "learning_rate": 0.00019985573140989405,
192
+ "loss": 0.4704,
193
+ "step": 280
194
+ },
195
+ {
196
+ "epoch": 0.11,
197
+ "learning_rate": 0.00019984524520070125,
198
+ "loss": 0.4548,
199
+ "step": 290
200
+ },
201
+ {
202
+ "epoch": 0.11,
203
+ "learning_rate": 0.00019983439144066143,
204
+ "loss": 0.4682,
205
+ "step": 300
206
+ },
207
+ {
208
+ "epoch": 0.11,
209
+ "eval_loss": 0.497054785490036,
210
+ "eval_runtime": 399.4057,
211
+ "eval_samples_per_second": 11.975,
212
+ "eval_steps_per_second": 2.994,
213
+ "step": 300
214
+ },
215
+ {
216
+ "epoch": 0.11,
217
+ "learning_rate": 0.0001998231701697295,
218
+ "loss": 0.4867,
219
+ "step": 310
220
+ },
221
+ {
222
+ "epoch": 0.12,
223
+ "learning_rate": 0.0001998115814292133,
224
+ "loss": 0.4486,
225
+ "step": 320
226
+ },
227
+ {
228
+ "epoch": 0.12,
229
+ "learning_rate": 0.0001997996252617733,
230
+ "loss": 0.4472,
231
+ "step": 330
232
+ },
233
+ {
234
+ "epoch": 0.12,
235
+ "learning_rate": 0.00019978730171142268,
236
+ "loss": 0.4821,
237
+ "step": 340
238
+ },
239
+ {
240
+ "epoch": 0.13,
241
+ "learning_rate": 0.0001997746108235269,
242
+ "loss": 0.4475,
243
+ "step": 350
244
+ },
245
+ {
246
+ "epoch": 0.13,
247
+ "learning_rate": 0.00019976155264480377,
248
+ "loss": 0.4485,
249
+ "step": 360
250
+ },
251
+ {
252
+ "epoch": 0.14,
253
+ "learning_rate": 0.00019974812722332308,
254
+ "loss": 0.4547,
255
+ "step": 370
256
+ },
257
+ {
258
+ "epoch": 0.14,
259
+ "learning_rate": 0.0001997343346085066,
260
+ "loss": 0.4279,
261
+ "step": 380
262
+ },
263
+ {
264
+ "epoch": 0.14,
265
+ "learning_rate": 0.00019972017485112774,
266
+ "loss": 0.4748,
267
+ "step": 390
268
+ },
269
+ {
270
+ "epoch": 0.15,
271
+ "learning_rate": 0.0001997056480033115,
272
+ "loss": 0.4778,
273
+ "step": 400
274
+ },
275
+ {
276
+ "epoch": 0.15,
277
+ "eval_loss": 0.4730743169784546,
278
+ "eval_runtime": 403.2392,
279
+ "eval_samples_per_second": 11.861,
280
+ "eval_steps_per_second": 2.966,
281
+ "step": 400
282
+ },
283
+ {
284
+ "epoch": 0.15,
285
+ "learning_rate": 0.0001996907541185342,
286
+ "loss": 0.4393,
287
+ "step": 410
288
+ },
289
+ {
290
+ "epoch": 0.15,
291
+ "learning_rate": 0.00019967549325162324,
292
+ "loss": 0.4562,
293
+ "step": 420
294
+ },
295
+ {
296
+ "epoch": 0.16,
297
+ "learning_rate": 0.00019965986545875708,
298
+ "loss": 0.4389,
299
+ "step": 430
300
+ },
301
+ {
302
+ "epoch": 0.16,
303
+ "learning_rate": 0.0001996438707974648,
304
+ "loss": 0.4282,
305
+ "step": 440
306
+ },
307
+ {
308
+ "epoch": 0.16,
309
+ "learning_rate": 0.000199627509326626,
310
+ "loss": 0.4254,
311
+ "step": 450
312
+ },
313
+ {
314
+ "epoch": 0.17,
315
+ "learning_rate": 0.0001996107811064706,
316
+ "loss": 0.419,
317
+ "step": 460
318
+ },
319
+ {
320
+ "epoch": 0.17,
321
+ "learning_rate": 0.00019959368619857872,
322
+ "loss": 0.4261,
323
+ "step": 470
324
+ },
325
+ {
326
+ "epoch": 0.18,
327
+ "learning_rate": 0.0001995762246658801,
328
+ "loss": 0.4581,
329
+ "step": 480
330
+ },
331
+ {
332
+ "epoch": 0.18,
333
+ "learning_rate": 0.00019955839657265432,
334
+ "loss": 0.4333,
335
+ "step": 490
336
+ },
337
+ {
338
+ "epoch": 0.18,
339
+ "learning_rate": 0.00019954020198453018,
340
+ "loss": 0.4541,
341
+ "step": 500
342
+ },
343
+ {
344
+ "epoch": 0.18,
345
+ "eval_loss": 0.46874529123306274,
346
+ "eval_runtime": 403.7251,
347
+ "eval_samples_per_second": 11.847,
348
+ "eval_steps_per_second": 2.962,
349
+ "step": 500
350
+ },
351
+ {
352
+ "epoch": 0.19,
353
+ "learning_rate": 0.00019952164096848578,
354
+ "loss": 0.4482,
355
+ "step": 510
356
+ },
357
+ {
358
+ "epoch": 0.19,
359
+ "learning_rate": 0.00019950271359284795,
360
+ "loss": 0.4475,
361
+ "step": 520
362
+ },
363
+ {
364
+ "epoch": 0.19,
365
+ "learning_rate": 0.00019948341992729227,
366
+ "loss": 0.4339,
367
+ "step": 530
368
+ },
369
+ {
370
+ "epoch": 0.2,
371
+ "learning_rate": 0.00019946376004284272,
372
+ "loss": 0.4527,
373
+ "step": 540
374
+ },
375
+ {
376
+ "epoch": 0.2,
377
+ "learning_rate": 0.0001994437340118713,
378
+ "loss": 0.4009,
379
+ "step": 550
380
+ },
381
+ {
382
+ "epoch": 0.21,
383
+ "learning_rate": 0.000199423341908098,
384
+ "loss": 0.4496,
385
+ "step": 560
386
+ },
387
+ {
388
+ "epoch": 0.21,
389
+ "learning_rate": 0.0001994025838065903,
390
+ "loss": 0.4314,
391
+ "step": 570
392
+ },
393
+ {
394
+ "epoch": 0.21,
395
+ "learning_rate": 0.0001993814597837631,
396
+ "loss": 0.4454,
397
+ "step": 580
398
+ },
399
+ {
400
+ "epoch": 0.22,
401
+ "learning_rate": 0.00019935996991737818,
402
+ "loss": 0.4076,
403
+ "step": 590
404
+ },
405
+ {
406
+ "epoch": 0.22,
407
+ "learning_rate": 0.0001993381142865442,
408
+ "loss": 0.4786,
409
+ "step": 600
410
+ },
411
+ {
412
+ "epoch": 0.22,
413
+ "eval_loss": 0.4500006139278412,
414
+ "eval_runtime": 399.7496,
415
+ "eval_samples_per_second": 11.965,
416
+ "eval_steps_per_second": 2.992,
417
+ "step": 600
418
+ },
419
+ {
420
+ "epoch": 0.22,
421
+ "learning_rate": 0.00019931589297171628,
422
+ "loss": 0.4386,
423
+ "step": 610
424
+ },
425
+ {
426
+ "epoch": 0.23,
427
+ "learning_rate": 0.0001992933060546955,
428
+ "loss": 0.4805,
429
+ "step": 620
430
+ },
431
+ {
432
+ "epoch": 0.23,
433
+ "learning_rate": 0.00019927035361862904,
434
+ "loss": 0.4135,
435
+ "step": 630
436
+ },
437
+ {
438
+ "epoch": 0.23,
439
+ "learning_rate": 0.0001992470357480095,
440
+ "loss": 0.421,
441
+ "step": 640
442
+ },
443
+ {
444
+ "epoch": 0.24,
445
+ "learning_rate": 0.00019922335252867476,
446
+ "loss": 0.4224,
447
+ "step": 650
448
+ },
449
+ {
450
+ "epoch": 0.24,
451
+ "learning_rate": 0.00019919930404780766,
452
+ "loss": 0.4324,
453
+ "step": 660
454
+ },
455
+ {
456
+ "epoch": 0.25,
457
+ "learning_rate": 0.0001991748903939355,
458
+ "loss": 0.4062,
459
+ "step": 670
460
+ },
461
+ {
462
+ "epoch": 0.25,
463
+ "learning_rate": 0.00019915011165692997,
464
+ "loss": 0.4107,
465
+ "step": 680
466
+ },
467
+ {
468
+ "epoch": 0.25,
469
+ "learning_rate": 0.00019912496792800677,
470
+ "loss": 0.3953,
471
+ "step": 690
472
+ },
473
+ {
474
+ "epoch": 0.26,
475
+ "learning_rate": 0.00019909945929972502,
476
+ "loss": 0.3974,
477
+ "step": 700
478
+ },
479
+ {
480
+ "epoch": 0.26,
481
+ "eval_loss": 0.4456511437892914,
482
+ "eval_runtime": 403.5923,
483
+ "eval_samples_per_second": 11.851,
484
+ "eval_steps_per_second": 2.963,
485
+ "step": 700
486
+ },
487
+ {
488
+ "epoch": 0.26,
489
+ "learning_rate": 0.0001990735858659873,
490
+ "loss": 0.4006,
491
+ "step": 710
492
+ },
493
+ {
494
+ "epoch": 0.26,
495
+ "learning_rate": 0.000199047347722039,
496
+ "loss": 0.4258,
497
+ "step": 720
498
+ },
499
+ {
500
+ "epoch": 0.27,
501
+ "learning_rate": 0.00019902074496446815,
502
+ "loss": 0.4083,
503
+ "step": 730
504
+ },
505
+ {
506
+ "epoch": 0.27,
507
+ "learning_rate": 0.00019899377769120487,
508
+ "loss": 0.4097,
509
+ "step": 740
510
+ },
511
+ {
512
+ "epoch": 0.27,
513
+ "learning_rate": 0.00019896644600152135,
514
+ "loss": 0.4262,
515
+ "step": 750
516
+ },
517
+ {
518
+ "epoch": 0.28,
519
+ "learning_rate": 0.00019893874999603103,
520
+ "loss": 0.4204,
521
+ "step": 760
522
+ },
523
+ {
524
+ "epoch": 0.28,
525
+ "learning_rate": 0.0001989106897766887,
526
+ "loss": 0.4233,
527
+ "step": 770
528
+ },
529
+ {
530
+ "epoch": 0.29,
531
+ "learning_rate": 0.0001988822654467897,
532
+ "loss": 0.4107,
533
+ "step": 780
534
+ },
535
+ {
536
+ "epoch": 0.29,
537
+ "learning_rate": 0.00019885347711096993,
538
+ "loss": 0.4013,
539
+ "step": 790
540
+ },
541
+ {
542
+ "epoch": 0.29,
543
+ "learning_rate": 0.00019882432487520506,
544
+ "loss": 0.4142,
545
+ "step": 800
546
+ },
547
+ {
548
+ "epoch": 0.29,
549
+ "eval_loss": 0.4460389018058777,
550
+ "eval_runtime": 400.2256,
551
+ "eval_samples_per_second": 11.951,
552
+ "eval_steps_per_second": 2.988,
553
+ "step": 800
554
+ },
555
+ {
556
+ "epoch": 0.3,
557
+ "learning_rate": 0.0001987948088468105,
558
+ "loss": 0.4324,
559
+ "step": 810
560
+ },
561
+ {
562
+ "epoch": 0.3,
563
+ "learning_rate": 0.0001987649291344408,
564
+ "loss": 0.3975,
565
+ "step": 820
566
+ },
567
+ {
568
+ "epoch": 0.3,
569
+ "learning_rate": 0.00019873468584808934,
570
+ "loss": 0.4109,
571
+ "step": 830
572
+ },
573
+ {
574
+ "epoch": 0.31,
575
+ "learning_rate": 0.00019870407909908786,
576
+ "loss": 0.3987,
577
+ "step": 840
578
+ },
579
+ {
580
+ "epoch": 0.31,
581
+ "learning_rate": 0.00019867310900010605,
582
+ "loss": 0.3885,
583
+ "step": 850
584
+ },
585
+ {
586
+ "epoch": 0.32,
587
+ "learning_rate": 0.00019864177566515122,
588
+ "loss": 0.4122,
589
+ "step": 860
590
+ },
591
+ {
592
+ "epoch": 0.32,
593
+ "learning_rate": 0.00019861007920956786,
594
+ "loss": 0.405,
595
+ "step": 870
596
+ },
597
+ {
598
+ "epoch": 0.32,
599
+ "learning_rate": 0.00019857801975003704,
600
+ "loss": 0.404,
601
+ "step": 880
602
+ },
603
+ {
604
+ "epoch": 0.33,
605
+ "learning_rate": 0.0001985455974045763,
606
+ "loss": 0.382,
607
+ "step": 890
608
+ },
609
+ {
610
+ "epoch": 0.33,
611
+ "learning_rate": 0.0001985128122925389,
612
+ "loss": 0.4374,
613
+ "step": 900
614
+ },
615
+ {
616
+ "epoch": 0.33,
617
+ "eval_loss": 0.44203105568885803,
618
+ "eval_runtime": 401.2967,
619
+ "eval_samples_per_second": 11.919,
620
+ "eval_steps_per_second": 2.98,
621
+ "step": 900
622
+ },
623
+ {
624
+ "epoch": 0.33,
625
+ "learning_rate": 0.00019847966453461358,
626
+ "loss": 0.3859,
627
+ "step": 910
628
+ },
629
+ {
630
+ "epoch": 0.34,
631
+ "learning_rate": 0.00019844615425282405,
632
+ "loss": 0.4187,
633
+ "step": 920
634
+ },
635
+ {
636
+ "epoch": 0.34,
637
+ "learning_rate": 0.00019841228157052853,
638
+ "loss": 0.4078,
639
+ "step": 930
640
+ },
641
+ {
642
+ "epoch": 0.34,
643
+ "learning_rate": 0.0001983780466124193,
644
+ "loss": 0.4218,
645
+ "step": 940
646
+ },
647
+ {
648
+ "epoch": 0.35,
649
+ "learning_rate": 0.0001983434495045223,
650
+ "loss": 0.4109,
651
+ "step": 950
652
+ },
653
+ {
654
+ "epoch": 0.35,
655
+ "learning_rate": 0.00019830849037419656,
656
+ "loss": 0.4249,
657
+ "step": 960
658
+ },
659
+ {
660
+ "epoch": 0.36,
661
+ "learning_rate": 0.00019827316935013388,
662
+ "loss": 0.4073,
663
+ "step": 970
664
+ },
665
+ {
666
+ "epoch": 0.36,
667
+ "learning_rate": 0.0001982374865623581,
668
+ "loss": 0.4336,
669
+ "step": 980
670
+ },
671
+ {
672
+ "epoch": 0.36,
673
+ "learning_rate": 0.00019820144214222497,
674
+ "loss": 0.4056,
675
+ "step": 990
676
+ },
677
+ {
678
+ "epoch": 0.37,
679
+ "learning_rate": 0.00019816503622242137,
680
+ "loss": 0.4008,
681
+ "step": 1000
682
+ },
683
+ {
684
+ "epoch": 0.37,
685
+ "eval_loss": 0.4418930411338806,
686
+ "eval_runtime": 404.9805,
687
+ "eval_samples_per_second": 11.81,
688
+ "eval_steps_per_second": 2.953,
689
+ "step": 1000
690
+ },
691
+ {
692
+ "epoch": 0.37,
693
+ "learning_rate": 0.00019812826893696495,
694
+ "loss": 0.365,
695
+ "step": 1010
696
+ },
697
+ {
698
+ "epoch": 0.37,
699
+ "learning_rate": 0.00019809114042120367,
700
+ "loss": 0.4006,
701
+ "step": 1020
702
+ },
703
+ {
704
+ "epoch": 0.38,
705
+ "learning_rate": 0.0001980536508118152,
706
+ "loss": 0.3714,
707
+ "step": 1030
708
+ },
709
+ {
710
+ "epoch": 0.38,
711
+ "learning_rate": 0.00019801580024680652,
712
+ "loss": 0.3945,
713
+ "step": 1040
714
+ },
715
+ {
716
+ "epoch": 0.38,
717
+ "learning_rate": 0.00019797758886551324,
718
+ "loss": 0.4316,
719
+ "step": 1050
720
+ },
721
+ {
722
+ "epoch": 0.39,
723
+ "learning_rate": 0.0001979390168085994,
724
+ "loss": 0.3958,
725
+ "step": 1060
726
+ },
727
+ {
728
+ "epoch": 0.39,
729
+ "learning_rate": 0.00019790008421805664,
730
+ "loss": 0.419,
731
+ "step": 1070
732
+ },
733
+ {
734
+ "epoch": 0.4,
735
+ "learning_rate": 0.00019786079123720377,
736
+ "loss": 0.4048,
737
+ "step": 1080
738
+ },
739
+ {
740
+ "epoch": 0.4,
741
+ "learning_rate": 0.0001978211380106864,
742
+ "loss": 0.4002,
743
+ "step": 1090
744
+ },
745
+ {
746
+ "epoch": 0.4,
747
+ "learning_rate": 0.0001977811246844761,
748
+ "loss": 0.3979,
749
+ "step": 1100
750
+ },
751
+ {
752
+ "epoch": 0.4,
753
+ "eval_loss": 0.43328720331192017,
754
+ "eval_runtime": 398.734,
755
+ "eval_samples_per_second": 11.995,
756
+ "eval_steps_per_second": 2.999,
757
+ "step": 1100
758
+ },
759
+ {
760
+ "epoch": 0.41,
761
+ "learning_rate": 0.00019774075140587024,
762
+ "loss": 0.4047,
763
+ "step": 1110
764
+ },
765
+ {
766
+ "epoch": 0.41,
767
+ "learning_rate": 0.00019770001832349106,
768
+ "loss": 0.389,
769
+ "step": 1120
770
+ },
771
+ {
772
+ "epoch": 0.41,
773
+ "learning_rate": 0.00019765892558728542,
774
+ "loss": 0.3628,
775
+ "step": 1130
776
+ },
777
+ {
778
+ "epoch": 0.42,
779
+ "learning_rate": 0.0001976174733485242,
780
+ "loss": 0.3914,
781
+ "step": 1140
782
+ },
783
+ {
784
+ "epoch": 0.42,
785
+ "learning_rate": 0.00019757566175980146,
786
+ "loss": 0.3885,
787
+ "step": 1150
788
+ },
789
+ {
790
+ "epoch": 0.42,
791
+ "learning_rate": 0.00019753349097503437,
792
+ "loss": 0.4023,
793
+ "step": 1160
794
+ },
795
+ {
796
+ "epoch": 0.43,
797
+ "learning_rate": 0.0001974909611494622,
798
+ "loss": 0.4358,
799
+ "step": 1170
800
+ },
801
+ {
802
+ "epoch": 0.43,
803
+ "learning_rate": 0.00019744807243964597,
804
+ "loss": 0.3965,
805
+ "step": 1180
806
+ },
807
+ {
808
+ "epoch": 0.44,
809
+ "learning_rate": 0.00019740482500346779,
810
+ "loss": 0.3762,
811
+ "step": 1190
812
+ },
813
+ {
814
+ "epoch": 0.44,
815
+ "learning_rate": 0.0001973612190001304,
816
+ "loss": 0.4108,
817
+ "step": 1200
818
+ },
819
+ {
820
+ "epoch": 0.44,
821
+ "eval_loss": 0.4304308295249939,
822
+ "eval_runtime": 397.6793,
823
+ "eval_samples_per_second": 12.027,
824
+ "eval_steps_per_second": 3.007,
825
+ "step": 1200
826
+ },
827
+ {
828
+ "epoch": 0.44,
829
+ "learning_rate": 0.00019731725459015643,
830
+ "loss": 0.3838,
831
+ "step": 1210
832
+ },
833
+ {
834
+ "epoch": 0.45,
835
+ "learning_rate": 0.00019727293193538793,
836
+ "loss": 0.3782,
837
+ "step": 1220
838
+ },
839
+ {
840
+ "epoch": 0.45,
841
+ "learning_rate": 0.00019722825119898566,
842
+ "loss": 0.4034,
843
+ "step": 1230
844
+ },
845
+ {
846
+ "epoch": 0.45,
847
+ "learning_rate": 0.00019718321254542858,
848
+ "loss": 0.391,
849
+ "step": 1240
850
+ },
851
+ {
852
+ "epoch": 0.46,
853
+ "learning_rate": 0.0001971378161405132,
854
+ "loss": 0.4082,
855
+ "step": 1250
856
+ },
857
+ {
858
+ "epoch": 0.46,
859
+ "learning_rate": 0.0001970920621513531,
860
+ "loss": 0.3772,
861
+ "step": 1260
862
+ },
863
+ {
864
+ "epoch": 0.47,
865
+ "learning_rate": 0.00019704595074637805,
866
+ "loss": 0.3894,
867
+ "step": 1270
868
+ },
869
+ {
870
+ "epoch": 0.47,
871
+ "learning_rate": 0.00019699948209533355,
872
+ "loss": 0.3882,
873
+ "step": 1280
874
+ },
875
+ {
876
+ "epoch": 0.47,
877
+ "learning_rate": 0.00019695265636928032,
878
+ "loss": 0.3782,
879
+ "step": 1290
880
+ },
881
+ {
882
+ "epoch": 0.48,
883
+ "learning_rate": 0.0001969054737405934,
884
+ "loss": 0.3578,
885
+ "step": 1300
886
+ },
887
+ {
888
+ "epoch": 0.48,
889
+ "eval_loss": 0.42550337314605713,
890
+ "eval_runtime": 402.9262,
891
+ "eval_samples_per_second": 11.871,
892
+ "eval_steps_per_second": 2.968,
893
+ "step": 1300
894
+ },
895
+ {
896
+ "epoch": 0.48,
897
+ "learning_rate": 0.00019685793438296183,
898
+ "loss": 0.3644,
899
+ "step": 1310
900
+ },
901
+ {
902
+ "epoch": 0.48,
903
+ "learning_rate": 0.00019681003847138765,
904
+ "loss": 0.3702,
905
+ "step": 1320
906
+ },
907
+ {
908
+ "epoch": 0.49,
909
+ "learning_rate": 0.00019676178618218565,
910
+ "loss": 0.4213,
911
+ "step": 1330
912
+ },
913
+ {
914
+ "epoch": 0.49,
915
+ "learning_rate": 0.0001967131776929823,
916
+ "loss": 0.4126,
917
+ "step": 1340
918
+ },
919
+ {
920
+ "epoch": 0.49,
921
+ "learning_rate": 0.00019666421318271547,
922
+ "loss": 0.3643,
923
+ "step": 1350
924
+ },
925
+ {
926
+ "epoch": 0.5,
927
+ "learning_rate": 0.00019661489283163362,
928
+ "loss": 0.3854,
929
+ "step": 1360
930
+ },
931
+ {
932
+ "epoch": 0.5,
933
+ "learning_rate": 0.00019656521682129502,
934
+ "loss": 0.3933,
935
+ "step": 1370
936
+ },
937
+ {
938
+ "epoch": 0.51,
939
+ "learning_rate": 0.00019651518533456733,
940
+ "loss": 0.3841,
941
+ "step": 1380
942
+ },
943
+ {
944
+ "epoch": 0.51,
945
+ "learning_rate": 0.00019646479855562666,
946
+ "loss": 0.3677,
947
+ "step": 1390
948
+ },
949
+ {
950
+ "epoch": 0.51,
951
+ "learning_rate": 0.00019641405666995715,
952
+ "loss": 0.3895,
953
+ "step": 1400
954
+ },
955
+ {
956
+ "epoch": 0.51,
957
+ "eval_loss": 0.41961902379989624,
958
+ "eval_runtime": 402.2386,
959
+ "eval_samples_per_second": 11.891,
960
+ "eval_steps_per_second": 2.973,
961
+ "step": 1400
962
+ },
963
+ {
964
+ "epoch": 0.52,
965
+ "learning_rate": 0.00019636295986435003,
966
+ "loss": 0.3578,
967
+ "step": 1410
968
+ },
969
+ {
970
+ "epoch": 0.52,
971
+ "learning_rate": 0.00019631150832690318,
972
+ "loss": 0.3878,
973
+ "step": 1420
974
+ },
975
+ {
976
+ "epoch": 0.52,
977
+ "learning_rate": 0.00019625970224702025,
978
+ "loss": 0.3784,
979
+ "step": 1430
980
+ },
981
+ {
982
+ "epoch": 0.53,
983
+ "learning_rate": 0.00019620754181541008,
984
+ "loss": 0.4042,
985
+ "step": 1440
986
+ },
987
+ {
988
+ "epoch": 0.53,
989
+ "learning_rate": 0.0001961550272240859,
990
+ "loss": 0.361,
991
+ "step": 1450
992
+ },
993
+ {
994
+ "epoch": 0.53,
995
+ "learning_rate": 0.00019610215866636477,
996
+ "loss": 0.3877,
997
+ "step": 1460
998
+ },
999
+ {
1000
+ "epoch": 0.54,
1001
+ "learning_rate": 0.00019604893633686662,
1002
+ "loss": 0.3822,
1003
+ "step": 1470
1004
+ },
1005
+ {
1006
+ "epoch": 0.54,
1007
+ "learning_rate": 0.00019599536043151384,
1008
+ "loss": 0.3726,
1009
+ "step": 1480
1010
+ },
1011
+ {
1012
+ "epoch": 0.55,
1013
+ "learning_rate": 0.00019594143114753026,
1014
+ "loss": 0.3552,
1015
+ "step": 1490
1016
+ },
1017
+ {
1018
+ "epoch": 0.55,
1019
+ "learning_rate": 0.00019588714868344073,
1020
+ "loss": 0.3725,
1021
+ "step": 1500
1022
+ },
1023
+ {
1024
+ "epoch": 0.55,
1025
+ "eval_loss": 0.4203263521194458,
1026
+ "eval_runtime": 396.7561,
1027
+ "eval_samples_per_second": 12.055,
1028
+ "eval_steps_per_second": 3.014,
1029
+ "step": 1500
1030
+ },
1031
+ {
1032
+ "epoch": 0.55,
1033
+ "learning_rate": 0.00019583251323907006,
1034
+ "loss": 0.3993,
1035
+ "step": 1510
1036
+ },
1037
+ {
1038
+ "epoch": 0.56,
1039
+ "learning_rate": 0.0001957775250155426,
1040
+ "loss": 0.3557,
1041
+ "step": 1520
1042
+ },
1043
+ {
1044
+ "epoch": 0.56,
1045
+ "learning_rate": 0.0001957221842152813,
1046
+ "loss": 0.4027,
1047
+ "step": 1530
1048
+ },
1049
+ {
1050
+ "epoch": 0.56,
1051
+ "learning_rate": 0.00019566649104200696,
1052
+ "loss": 0.3547,
1053
+ "step": 1540
1054
+ },
1055
+ {
1056
+ "epoch": 0.57,
1057
+ "learning_rate": 0.00019561044570073763,
1058
+ "loss": 0.3978,
1059
+ "step": 1550
1060
+ },
1061
+ {
1062
+ "epoch": 0.57,
1063
+ "learning_rate": 0.00019555404839778767,
1064
+ "loss": 0.3919,
1065
+ "step": 1560
1066
+ },
1067
+ {
1068
+ "epoch": 0.58,
1069
+ "learning_rate": 0.00019549729934076717,
1070
+ "loss": 0.3713,
1071
+ "step": 1570
1072
+ },
1073
+ {
1074
+ "epoch": 0.58,
1075
+ "learning_rate": 0.00019544019873858102,
1076
+ "loss": 0.408,
1077
+ "step": 1580
1078
+ },
1079
+ {
1080
+ "epoch": 0.58,
1081
+ "learning_rate": 0.00019538274680142834,
1082
+ "loss": 0.3792,
1083
+ "step": 1590
1084
+ },
1085
+ {
1086
+ "epoch": 0.59,
1087
+ "learning_rate": 0.00019532494374080144,
1088
+ "loss": 0.3836,
1089
+ "step": 1600
1090
+ },
1091
+ {
1092
+ "epoch": 0.59,
1093
+ "eval_loss": 0.42036038637161255,
1094
+ "eval_runtime": 402.4927,
1095
+ "eval_samples_per_second": 11.883,
1096
+ "eval_steps_per_second": 2.971,
1097
+ "step": 1600
1098
+ },
1099
+ {
1100
+ "epoch": 0.59,
1101
+ "learning_rate": 0.00019526678976948525,
1102
+ "loss": 0.3411,
1103
+ "step": 1610
1104
+ },
1105
+ {
1106
+ "epoch": 0.59,
1107
+ "learning_rate": 0.0001952082851015565,
1108
+ "loss": 0.3863,
1109
+ "step": 1620
1110
+ },
1111
+ {
1112
+ "epoch": 0.6,
1113
+ "learning_rate": 0.00019514942995238287,
1114
+ "loss": 0.375,
1115
+ "step": 1630
1116
+ },
1117
+ {
1118
+ "epoch": 0.6,
1119
+ "learning_rate": 0.00019509022453862226,
1120
+ "loss": 0.3747,
1121
+ "step": 1640
1122
+ },
1123
+ {
1124
+ "epoch": 0.6,
1125
+ "learning_rate": 0.00019503066907822198,
1126
+ "loss": 0.36,
1127
+ "step": 1650
1128
+ },
1129
+ {
1130
+ "epoch": 0.61,
1131
+ "learning_rate": 0.00019497076379041786,
1132
+ "loss": 0.3919,
1133
+ "step": 1660
1134
+ },
1135
+ {
1136
+ "epoch": 0.61,
1137
+ "learning_rate": 0.00019491050889573357,
1138
+ "loss": 0.3699,
1139
+ "step": 1670
1140
+ },
1141
+ {
1142
+ "epoch": 0.62,
1143
+ "learning_rate": 0.00019484990461597978,
1144
+ "loss": 0.3829,
1145
+ "step": 1680
1146
+ },
1147
+ {
1148
+ "epoch": 0.62,
1149
+ "learning_rate": 0.00019478895117425323,
1150
+ "loss": 0.3589,
1151
+ "step": 1690
1152
+ },
1153
+ {
1154
+ "epoch": 0.62,
1155
+ "learning_rate": 0.00019472764879493616,
1156
+ "loss": 0.3784,
1157
+ "step": 1700
1158
+ },
1159
+ {
1160
+ "epoch": 0.62,
1161
+ "eval_loss": 0.4183010458946228,
1162
+ "eval_runtime": 404.1371,
1163
+ "eval_samples_per_second": 11.835,
1164
+ "eval_steps_per_second": 2.959,
1165
+ "step": 1700
1166
+ },
1167
+ {
1168
+ "epoch": 0.63,
1169
+ "learning_rate": 0.00019466599770369509,
1170
+ "loss": 0.3912,
1171
+ "step": 1710
1172
+ },
1173
+ {
1174
+ "epoch": 0.63,
1175
+ "learning_rate": 0.00019460399812748041,
1176
+ "loss": 0.374,
1177
+ "step": 1720
1178
+ },
1179
+ {
1180
+ "epoch": 0.63,
1181
+ "learning_rate": 0.0001945416502945253,
1182
+ "loss": 0.384,
1183
+ "step": 1730
1184
+ },
1185
+ {
1186
+ "epoch": 0.64,
1187
+ "learning_rate": 0.00019447895443434486,
1188
+ "loss": 0.3519,
1189
+ "step": 1740
1190
+ },
1191
+ {
1192
+ "epoch": 0.64,
1193
+ "learning_rate": 0.00019441591077773554,
1194
+ "loss": 0.3598,
1195
+ "step": 1750
1196
+ },
1197
+ {
1198
+ "epoch": 0.64,
1199
+ "learning_rate": 0.0001943525195567739,
1200
+ "loss": 0.3145,
1201
+ "step": 1760
1202
+ },
1203
+ {
1204
+ "epoch": 0.65,
1205
+ "learning_rate": 0.00019428878100481606,
1206
+ "loss": 0.3842,
1207
+ "step": 1770
1208
+ },
1209
+ {
1210
+ "epoch": 0.65,
1211
+ "learning_rate": 0.0001942246953564967,
1212
+ "loss": 0.3674,
1213
+ "step": 1780
1214
+ },
1215
+ {
1216
+ "epoch": 0.66,
1217
+ "learning_rate": 0.00019416026284772825,
1218
+ "loss": 0.3603,
1219
+ "step": 1790
1220
+ },
1221
+ {
1222
+ "epoch": 0.66,
1223
+ "learning_rate": 0.00019409548371570007,
1224
+ "loss": 0.369,
1225
+ "step": 1800
1226
+ },
1227
+ {
1228
+ "epoch": 0.66,
1229
+ "eval_loss": 0.4111176133155823,
1230
+ "eval_runtime": 402.9829,
1231
+ "eval_samples_per_second": 11.869,
1232
+ "eval_steps_per_second": 2.968,
1233
+ "step": 1800
1234
+ },
1235
+ {
1236
+ "epoch": 0.66,
1237
+ "learning_rate": 0.00019403035819887734,
1238
+ "loss": 0.3713,
1239
+ "step": 1810
1240
+ },
1241
+ {
1242
+ "epoch": 0.67,
1243
+ "learning_rate": 0.00019396488653700055,
1244
+ "loss": 0.3539,
1245
+ "step": 1820
1246
+ },
1247
+ {
1248
+ "epoch": 0.67,
1249
+ "learning_rate": 0.00019389906897108428,
1250
+ "loss": 0.3932,
1251
+ "step": 1830
1252
+ },
1253
+ {
1254
+ "epoch": 0.67,
1255
+ "learning_rate": 0.0001938329057434165,
1256
+ "loss": 0.3371,
1257
+ "step": 1840
1258
+ },
1259
+ {
1260
+ "epoch": 0.68,
1261
+ "learning_rate": 0.00019376639709755766,
1262
+ "loss": 0.3832,
1263
+ "step": 1850
1264
+ },
1265
+ {
1266
+ "epoch": 0.68,
1267
+ "learning_rate": 0.00019369954327833972,
1268
+ "loss": 0.3658,
1269
+ "step": 1860
1270
+ },
1271
+ {
1272
+ "epoch": 0.69,
1273
+ "learning_rate": 0.00019363234453186534,
1274
+ "loss": 0.3452,
1275
+ "step": 1870
1276
+ },
1277
+ {
1278
+ "epoch": 0.69,
1279
+ "learning_rate": 0.00019356480110550687,
1280
+ "loss": 0.3721,
1281
+ "step": 1880
1282
+ },
1283
+ {
1284
+ "epoch": 0.69,
1285
+ "learning_rate": 0.00019349691324790555,
1286
+ "loss": 0.3637,
1287
+ "step": 1890
1288
+ },
1289
+ {
1290
+ "epoch": 0.7,
1291
+ "learning_rate": 0.00019342868120897054,
1292
+ "loss": 0.3409,
1293
+ "step": 1900
1294
+ },
1295
+ {
1296
+ "epoch": 0.7,
1297
+ "eval_loss": 0.41198766231536865,
1298
+ "eval_runtime": 402.9688,
1299
+ "eval_samples_per_second": 11.869,
1300
+ "eval_steps_per_second": 2.968,
1301
+ "step": 1900
1302
+ },
1303
+ {
1304
+ "epoch": 0.7,
1305
+ "learning_rate": 0.00019336010523987796,
1306
+ "loss": 0.3424,
1307
+ "step": 1910
1308
+ },
1309
+ {
1310
+ "epoch": 0.7,
1311
+ "learning_rate": 0.00019329118559307,
1312
+ "loss": 0.3532,
1313
+ "step": 1920
1314
+ },
1315
+ {
1316
+ "epoch": 0.71,
1317
+ "learning_rate": 0.0001932219225222541,
1318
+ "loss": 0.3591,
1319
+ "step": 1930
1320
+ },
1321
+ {
1322
+ "epoch": 0.71,
1323
+ "learning_rate": 0.00019315231628240178,
1324
+ "loss": 0.3794,
1325
+ "step": 1940
1326
+ },
1327
+ {
1328
+ "epoch": 0.71,
1329
+ "learning_rate": 0.00019308236712974795,
1330
+ "loss": 0.3825,
1331
+ "step": 1950
1332
+ },
1333
+ {
1334
+ "epoch": 0.72,
1335
+ "learning_rate": 0.0001930120753217898,
1336
+ "loss": 0.3482,
1337
+ "step": 1960
1338
+ },
1339
+ {
1340
+ "epoch": 0.72,
1341
+ "learning_rate": 0.00019294144111728584,
1342
+ "loss": 0.3643,
1343
+ "step": 1970
1344
+ },
1345
+ {
1346
+ "epoch": 0.73,
1347
+ "learning_rate": 0.00019287046477625515,
1348
+ "loss": 0.3522,
1349
+ "step": 1980
1350
+ },
1351
+ {
1352
+ "epoch": 0.73,
1353
+ "learning_rate": 0.00019279914655997619,
1354
+ "loss": 0.3723,
1355
+ "step": 1990
1356
+ },
1357
+ {
1358
+ "epoch": 0.73,
1359
+ "learning_rate": 0.00019272748673098596,
1360
+ "loss": 0.388,
1361
+ "step": 2000
1362
+ },
1363
+ {
1364
+ "epoch": 0.73,
1365
+ "eval_loss": 0.41505494713783264,
1366
+ "eval_runtime": 404.7184,
1367
+ "eval_samples_per_second": 11.818,
1368
+ "eval_steps_per_second": 2.955,
1369
+ "step": 2000
1370
+ },
1371
+ {
1372
+ "epoch": 0.74,
1373
+ "learning_rate": 0.000192655485553079,
1374
+ "loss": 0.3739,
1375
+ "step": 2010
1376
+ },
1377
+ {
1378
+ "epoch": 0.74,
1379
+ "learning_rate": 0.00019258314329130641,
1380
+ "loss": 0.3625,
1381
+ "step": 2020
1382
+ },
1383
+ {
1384
+ "epoch": 0.74,
1385
+ "learning_rate": 0.00019251046021197496,
1386
+ "loss": 0.3174,
1387
+ "step": 2030
1388
+ },
1389
+ {
1390
+ "epoch": 0.75,
1391
+ "learning_rate": 0.00019243743658264593,
1392
+ "loss": 0.347,
1393
+ "step": 2040
1394
+ },
1395
+ {
1396
+ "epoch": 0.75,
1397
+ "learning_rate": 0.00019236407267213433,
1398
+ "loss": 0.3837,
1399
+ "step": 2050
1400
+ },
1401
+ {
1402
+ "epoch": 0.75,
1403
+ "learning_rate": 0.00019229036875050777,
1404
+ "loss": 0.3854,
1405
+ "step": 2060
1406
+ },
1407
+ {
1408
+ "epoch": 0.76,
1409
+ "learning_rate": 0.0001922163250890855,
1410
+ "loss": 0.3449,
1411
+ "step": 2070
1412
+ },
1413
+ {
1414
+ "epoch": 0.76,
1415
+ "learning_rate": 0.00019214194196043741,
1416
+ "loss": 0.3507,
1417
+ "step": 2080
1418
+ },
1419
+ {
1420
+ "epoch": 0.77,
1421
+ "learning_rate": 0.00019206721963838317,
1422
+ "loss": 0.3786,
1423
+ "step": 2090
1424
+ },
1425
+ {
1426
+ "epoch": 0.77,
1427
+ "learning_rate": 0.00019199215839799092,
1428
+ "loss": 0.3608,
1429
+ "step": 2100
1430
+ },
1431
+ {
1432
+ "epoch": 0.77,
1433
+ "eval_loss": 0.4092780351638794,
1434
+ "eval_runtime": 403.6583,
1435
+ "eval_samples_per_second": 11.849,
1436
+ "eval_steps_per_second": 2.963,
1437
+ "step": 2100
1438
+ },
1439
+ {
1440
+ "epoch": 0.77,
1441
+ "learning_rate": 0.0001919167585155765,
1442
+ "loss": 0.3233,
1443
+ "step": 2110
1444
+ },
1445
+ {
1446
+ "epoch": 0.78,
1447
+ "learning_rate": 0.00019184102026870235,
1448
+ "loss": 0.3679,
1449
+ "step": 2120
1450
+ },
1451
+ {
1452
+ "epoch": 0.78,
1453
+ "learning_rate": 0.0001917649439361765,
1454
+ "loss": 0.3426,
1455
+ "step": 2130
1456
+ },
1457
+ {
1458
+ "epoch": 0.78,
1459
+ "learning_rate": 0.00019168852979805162,
1460
+ "loss": 0.3272,
1461
+ "step": 2140
1462
+ },
1463
+ {
1464
+ "epoch": 0.79,
1465
+ "learning_rate": 0.00019161177813562379,
1466
+ "loss": 0.3572,
1467
+ "step": 2150
1468
+ },
1469
+ {
1470
+ "epoch": 0.79,
1471
+ "learning_rate": 0.0001915346892314316,
1472
+ "loss": 0.3429,
1473
+ "step": 2160
1474
+ },
1475
+ {
1476
+ "epoch": 0.8,
1477
+ "learning_rate": 0.0001914572633692552,
1478
+ "loss": 0.3325,
1479
+ "step": 2170
1480
+ },
1481
+ {
1482
+ "epoch": 0.8,
1483
+ "learning_rate": 0.00019137950083411505,
1484
+ "loss": 0.3444,
1485
+ "step": 2180
1486
+ },
1487
+ {
1488
+ "epoch": 0.8,
1489
+ "learning_rate": 0.000191301401912271,
1490
+ "loss": 0.351,
1491
+ "step": 2190
1492
+ },
1493
+ {
1494
+ "epoch": 0.81,
1495
+ "learning_rate": 0.00019122296689122123,
1496
+ "loss": 0.3171,
1497
+ "step": 2200
1498
+ },
1499
+ {
1500
+ "epoch": 0.81,
1501
+ "eval_loss": 0.4078885316848755,
1502
+ "eval_runtime": 403.041,
1503
+ "eval_samples_per_second": 11.867,
1504
+ "eval_steps_per_second": 2.967,
1505
+ "step": 2200
1506
+ },
1507
+ {
1508
+ "epoch": 0.81,
1509
+ "learning_rate": 0.0001911441960597012,
1510
+ "loss": 0.3626,
1511
+ "step": 2210
1512
+ },
1513
+ {
1514
+ "epoch": 0.81,
1515
+ "learning_rate": 0.0001910650897076824,
1516
+ "loss": 0.3529,
1517
+ "step": 2220
1518
+ },
1519
+ {
1520
+ "epoch": 0.82,
1521
+ "learning_rate": 0.00019098564812637165,
1522
+ "loss": 0.3727,
1523
+ "step": 2230
1524
+ },
1525
+ {
1526
+ "epoch": 0.82,
1527
+ "learning_rate": 0.0001909058716082097,
1528
+ "loss": 0.3111,
1529
+ "step": 2240
1530
+ },
1531
+ {
1532
+ "epoch": 0.82,
1533
+ "learning_rate": 0.0001908257604468703,
1534
+ "loss": 0.3846,
1535
+ "step": 2250
1536
+ },
1537
+ {
1538
+ "epoch": 0.83,
1539
+ "learning_rate": 0.00019074531493725906,
1540
+ "loss": 0.3612,
1541
+ "step": 2260
1542
+ },
1543
+ {
1544
+ "epoch": 0.83,
1545
+ "learning_rate": 0.0001906645353755124,
1546
+ "loss": 0.3573,
1547
+ "step": 2270
1548
+ },
1549
+ {
1550
+ "epoch": 0.84,
1551
+ "learning_rate": 0.00019058342205899656,
1552
+ "loss": 0.3902,
1553
+ "step": 2280
1554
+ },
1555
+ {
1556
+ "epoch": 0.84,
1557
+ "learning_rate": 0.0001905019752863062,
1558
+ "loss": 0.345,
1559
+ "step": 2290
1560
+ },
1561
+ {
1562
+ "epoch": 0.84,
1563
+ "learning_rate": 0.0001904201953572637,
1564
+ "loss": 0.3581,
1565
+ "step": 2300
1566
+ },
1567
+ {
1568
+ "epoch": 0.84,
1569
+ "eval_loss": 0.4118511974811554,
1570
+ "eval_runtime": 405.9796,
1571
+ "eval_samples_per_second": 11.781,
1572
+ "eval_steps_per_second": 2.946,
1573
+ "step": 2300
1574
+ },
1575
+ {
1576
+ "epoch": 0.85,
1577
+ "learning_rate": 0.00019033808257291768,
1578
+ "loss": 0.3303,
1579
+ "step": 2310
1580
+ },
1581
+ {
1582
+ "epoch": 0.85,
1583
+ "learning_rate": 0.00019025563723554223,
1584
+ "loss": 0.3432,
1585
+ "step": 2320
1586
+ },
1587
+ {
1588
+ "epoch": 0.85,
1589
+ "learning_rate": 0.00019017285964863554,
1590
+ "loss": 0.3436,
1591
+ "step": 2330
1592
+ },
1593
+ {
1594
+ "epoch": 0.86,
1595
+ "learning_rate": 0.0001900897501169189,
1596
+ "loss": 0.338,
1597
+ "step": 2340
1598
+ },
1599
+ {
1600
+ "epoch": 0.86,
1601
+ "learning_rate": 0.0001900063089463356,
1602
+ "loss": 0.3611,
1603
+ "step": 2350
1604
+ },
1605
+ {
1606
+ "epoch": 0.86,
1607
+ "learning_rate": 0.00018992253644404967,
1608
+ "loss": 0.3311,
1609
+ "step": 2360
1610
+ },
1611
+ {
1612
+ "epoch": 0.87,
1613
+ "learning_rate": 0.00018983843291844492,
1614
+ "loss": 0.3337,
1615
+ "step": 2370
1616
+ },
1617
+ {
1618
+ "epoch": 0.87,
1619
+ "learning_rate": 0.00018975399867912364,
1620
+ "loss": 0.358,
1621
+ "step": 2380
1622
+ },
1623
+ {
1624
+ "epoch": 0.88,
1625
+ "learning_rate": 0.0001896692340369057,
1626
+ "loss": 0.3802,
1627
+ "step": 2390
1628
+ },
1629
+ {
1630
+ "epoch": 0.88,
1631
+ "learning_rate": 0.00018958413930382704,
1632
+ "loss": 0.3389,
1633
+ "step": 2400
1634
+ },
1635
+ {
1636
+ "epoch": 0.88,
1637
+ "eval_loss": 0.41193756461143494,
1638
+ "eval_runtime": 402.9836,
1639
+ "eval_samples_per_second": 11.869,
1640
+ "eval_steps_per_second": 2.968,
1641
+ "step": 2400
1642
+ },
1643
+ {
1644
+ "epoch": 0.88,
1645
+ "learning_rate": 0.0001894987147931389,
1646
+ "loss": 0.3398,
1647
+ "step": 2410
1648
+ },
1649
+ {
1650
+ "epoch": 0.89,
1651
+ "learning_rate": 0.00018941296081930646,
1652
+ "loss": 0.3459,
1653
+ "step": 2420
1654
+ },
1655
+ {
1656
+ "epoch": 0.89,
1657
+ "learning_rate": 0.00018932687769800767,
1658
+ "loss": 0.3546,
1659
+ "step": 2430
1660
+ },
1661
+ {
1662
+ "epoch": 0.89,
1663
+ "learning_rate": 0.00018924046574613222,
1664
+ "loss": 0.3158,
1665
+ "step": 2440
1666
+ },
1667
+ {
1668
+ "epoch": 0.9,
1669
+ "learning_rate": 0.0001891537252817802,
1670
+ "loss": 0.3484,
1671
+ "step": 2450
1672
+ },
1673
+ {
1674
+ "epoch": 0.9,
1675
+ "learning_rate": 0.00018906665662426104,
1676
+ "loss": 0.3437,
1677
+ "step": 2460
1678
+ },
1679
+ {
1680
+ "epoch": 0.9,
1681
+ "learning_rate": 0.0001889792600940924,
1682
+ "loss": 0.3637,
1683
+ "step": 2470
1684
+ },
1685
+ {
1686
+ "epoch": 0.91,
1687
+ "learning_rate": 0.00018889153601299888,
1688
+ "loss": 0.3658,
1689
+ "step": 2480
1690
+ },
1691
+ {
1692
+ "epoch": 0.91,
1693
+ "learning_rate": 0.00018880348470391077,
1694
+ "loss": 0.3452,
1695
+ "step": 2490
1696
+ },
1697
+ {
1698
+ "epoch": 0.92,
1699
+ "learning_rate": 0.000188715106490963,
1700
+ "loss": 0.3302,
1701
+ "step": 2500
1702
+ },
1703
+ {
1704
+ "epoch": 0.92,
1705
+ "eval_loss": 0.4021553695201874,
1706
+ "eval_runtime": 399.2274,
1707
+ "eval_samples_per_second": 11.981,
1708
+ "eval_steps_per_second": 2.996,
1709
+ "step": 2500
1710
+ },
1711
+ {
1712
+ "epoch": 0.92,
1713
+ "learning_rate": 0.000188626401699494,
1714
+ "loss": 0.3511,
1715
+ "step": 2510
1716
+ },
1717
+ {
1718
+ "epoch": 0.92,
1719
+ "learning_rate": 0.00018853737065604426,
1720
+ "loss": 0.3232,
1721
+ "step": 2520
1722
+ },
1723
+ {
1724
+ "epoch": 0.93,
1725
+ "learning_rate": 0.00018844801368835532,
1726
+ "loss": 0.366,
1727
+ "step": 2530
1728
+ },
1729
+ {
1730
+ "epoch": 0.93,
1731
+ "learning_rate": 0.00018835833112536857,
1732
+ "loss": 0.3151,
1733
+ "step": 2540
1734
+ },
1735
+ {
1736
+ "epoch": 0.93,
1737
+ "learning_rate": 0.0001882683232972239,
1738
+ "loss": 0.3534,
1739
+ "step": 2550
1740
+ },
1741
+ {
1742
+ "epoch": 0.94,
1743
+ "learning_rate": 0.00018817799053525862,
1744
+ "loss": 0.3779,
1745
+ "step": 2560
1746
+ },
1747
+ {
1748
+ "epoch": 0.94,
1749
+ "learning_rate": 0.0001880873331720062,
1750
+ "loss": 0.3338,
1751
+ "step": 2570
1752
+ },
1753
+ {
1754
+ "epoch": 0.95,
1755
+ "learning_rate": 0.00018799635154119495,
1756
+ "loss": 0.3243,
1757
+ "step": 2580
1758
+ },
1759
+ {
1760
+ "epoch": 0.95,
1761
+ "learning_rate": 0.00018790504597774698,
1762
+ "loss": 0.3729,
1763
+ "step": 2590
1764
+ },
1765
+ {
1766
+ "epoch": 0.95,
1767
+ "learning_rate": 0.0001878134168177768,
1768
+ "loss": 0.3553,
1769
+ "step": 2600
1770
+ },
1771
+ {
1772
+ "epoch": 0.95,
1773
+ "eval_loss": 0.4054717421531677,
1774
+ "eval_runtime": 400.3737,
1775
+ "eval_samples_per_second": 11.946,
1776
+ "eval_steps_per_second": 2.987,
1777
+ "step": 2600
1778
+ },
1779
+ {
1780
+ "epoch": 0.96,
1781
+ "learning_rate": 0.00018772146439859015,
1782
+ "loss": 0.3268,
1783
+ "step": 2610
1784
+ },
1785
+ {
1786
+ "epoch": 0.96,
1787
+ "learning_rate": 0.00018762918905868277,
1788
+ "loss": 0.3304,
1789
+ "step": 2620
1790
+ },
1791
+ {
1792
+ "epoch": 0.96,
1793
+ "learning_rate": 0.00018753659113773913,
1794
+ "loss": 0.3606,
1795
+ "step": 2630
1796
+ },
1797
+ {
1798
+ "epoch": 0.97,
1799
+ "learning_rate": 0.0001874436709766312,
1800
+ "loss": 0.3421,
1801
+ "step": 2640
1802
+ },
1803
+ {
1804
+ "epoch": 0.97,
1805
+ "learning_rate": 0.00018735042891741718,
1806
+ "loss": 0.3629,
1807
+ "step": 2650
1808
+ },
1809
+ {
1810
+ "epoch": 0.97,
1811
+ "learning_rate": 0.0001872568653033402,
1812
+ "loss": 0.3636,
1813
+ "step": 2660
1814
+ },
1815
+ {
1816
+ "epoch": 0.98,
1817
+ "learning_rate": 0.00018716298047882714,
1818
+ "loss": 0.3485,
1819
+ "step": 2670
1820
+ },
1821
+ {
1822
+ "epoch": 0.98,
1823
+ "learning_rate": 0.00018706877478948735,
1824
+ "loss": 0.3169,
1825
+ "step": 2680
1826
+ },
1827
+ {
1828
+ "epoch": 0.99,
1829
+ "learning_rate": 0.00018697424858211126,
1830
+ "loss": 0.306,
1831
+ "step": 2690
1832
+ },
1833
+ {
1834
+ "epoch": 0.99,
1835
+ "learning_rate": 0.0001868794022046693,
1836
+ "loss": 0.3586,
1837
+ "step": 2700
1838
+ },
1839
+ {
1840
+ "epoch": 0.99,
1841
+ "eval_loss": 0.4048784375190735,
1842
+ "eval_runtime": 400.9367,
1843
+ "eval_samples_per_second": 11.93,
1844
+ "eval_steps_per_second": 2.983,
1845
+ "step": 2700
1846
+ },
1847
+ {
1848
+ "epoch": 0.99,
1849
+ "learning_rate": 0.00018678423600631042,
1850
+ "loss": 0.3311,
1851
+ "step": 2710
1852
+ },
1853
+ {
1854
+ "epoch": 1.0,
1855
+ "learning_rate": 0.00018668875033736094,
1856
+ "loss": 0.3066,
1857
+ "step": 2720
1858
+ },
1859
+ {
1860
+ "epoch": 1.0,
1861
+ "learning_rate": 0.00018659294554932324,
1862
+ "loss": 0.338,
1863
+ "step": 2730
1864
+ },
1865
+ {
1866
+ "epoch": 1.0,
1867
+ "learning_rate": 0.00018649682199487437,
1868
+ "loss": 0.3205,
1869
+ "step": 2740
1870
+ },
1871
+ {
1872
+ "epoch": 1.01,
1873
+ "learning_rate": 0.0001864003800278649,
1874
+ "loss": 0.3176,
1875
+ "step": 2750
1876
+ },
1877
+ {
1878
+ "epoch": 1.01,
1879
+ "learning_rate": 0.00018630362000331753,
1880
+ "loss": 0.2986,
1881
+ "step": 2760
1882
+ },
1883
+ {
1884
+ "epoch": 1.01,
1885
+ "learning_rate": 0.00018620654227742572,
1886
+ "loss": 0.3175,
1887
+ "step": 2770
1888
+ },
1889
+ {
1890
+ "epoch": 1.02,
1891
+ "learning_rate": 0.0001861091472075526,
1892
+ "loss": 0.3174,
1893
+ "step": 2780
1894
+ },
1895
+ {
1896
+ "epoch": 1.02,
1897
+ "learning_rate": 0.0001860114351522293,
1898
+ "loss": 0.2752,
1899
+ "step": 2790
1900
+ },
1901
+ {
1902
+ "epoch": 1.03,
1903
+ "learning_rate": 0.00018591340647115402,
1904
+ "loss": 0.3014,
1905
+ "step": 2800
1906
+ },
1907
+ {
1908
+ "epoch": 1.03,
1909
+ "eval_loss": 0.41257622838020325,
1910
+ "eval_runtime": 223.7363,
1911
+ "eval_samples_per_second": 21.378,
1912
+ "eval_steps_per_second": 5.346,
1913
+ "step": 2800
1914
+ },
1915
+ {
1916
+ "epoch": 1.03,
1917
+ "learning_rate": 0.0001858150615251905,
1918
+ "loss": 0.3013,
1919
+ "step": 2810
1920
+ },
1921
+ {
1922
+ "epoch": 1.03,
1923
+ "learning_rate": 0.00018571640067636662,
1924
+ "loss": 0.3364,
1925
+ "step": 2820
1926
+ },
1927
+ {
1928
+ "epoch": 1.04,
1929
+ "learning_rate": 0.00018561742428787324,
1930
+ "loss": 0.3159,
1931
+ "step": 2830
1932
+ },
1933
+ {
1934
+ "epoch": 1.04,
1935
+ "learning_rate": 0.00018552807605292504,
1936
+ "loss": 0.3037,
1937
+ "step": 2840
1938
+ },
1939
+ {
1940
+ "epoch": 1.04,
1941
+ "learning_rate": 0.00018542850114380946,
1942
+ "loss": 0.3087,
1943
+ "step": 2850
1944
+ },
1945
+ {
1946
+ "epoch": 1.05,
1947
+ "learning_rate": 0.00018532861175484162,
1948
+ "loss": 0.2858,
1949
+ "step": 2860
1950
+ },
1951
+ {
1952
+ "epoch": 1.05,
1953
+ "learning_rate": 0.00018522840825373492,
1954
+ "loss": 0.3208,
1955
+ "step": 2870
1956
+ },
1957
+ {
1958
+ "epoch": 1.06,
1959
+ "learning_rate": 0.00018512789100935906,
1960
+ "loss": 0.3117,
1961
+ "step": 2880
1962
+ },
1963
+ {
1964
+ "epoch": 1.06,
1965
+ "learning_rate": 0.00018502706039173856,
1966
+ "loss": 0.2921,
1967
+ "step": 2890
1968
+ },
1969
+ {
1970
+ "epoch": 1.06,
1971
+ "learning_rate": 0.0001849259167720517,
1972
+ "loss": 0.3354,
1973
+ "step": 2900
1974
+ },
1975
+ {
1976
+ "epoch": 1.06,
1977
+ "eval_loss": 0.4081941545009613,
1978
+ "eval_runtime": 223.8258,
1979
+ "eval_samples_per_second": 21.369,
1980
+ "eval_steps_per_second": 5.343,
1981
+ "step": 2900
1982
+ },
1983
+ {
1984
+ "epoch": 1.07,
1985
+ "learning_rate": 0.0001848244605226289,
1986
+ "loss": 0.2866,
1987
+ "step": 2910
1988
+ },
1989
+ {
1990
+ "epoch": 1.07,
1991
+ "learning_rate": 0.0001847226920169514,
1992
+ "loss": 0.308,
1993
+ "step": 2920
1994
+ },
1995
+ {
1996
+ "epoch": 1.07,
1997
+ "learning_rate": 0.00018462061162965,
1998
+ "loss": 0.3142,
1999
+ "step": 2930
2000
+ },
2001
+ {
2002
+ "epoch": 1.08,
2003
+ "learning_rate": 0.0001845182197365036,
2004
+ "loss": 0.3125,
2005
+ "step": 2940
2006
+ },
2007
+ {
2008
+ "epoch": 1.08,
2009
+ "learning_rate": 0.00018441551671443768,
2010
+ "loss": 0.307,
2011
+ "step": 2950
2012
+ },
2013
+ {
2014
+ "epoch": 1.08,
2015
+ "learning_rate": 0.00018431250294152323,
2016
+ "loss": 0.2804,
2017
+ "step": 2960
2018
+ },
2019
+ {
2020
+ "epoch": 1.09,
2021
+ "learning_rate": 0.00018420917879697507,
2022
+ "loss": 0.304,
2023
+ "step": 2970
2024
+ },
2025
+ {
2026
+ "epoch": 1.09,
2027
+ "learning_rate": 0.0001841055446611506,
2028
+ "loss": 0.3152,
2029
+ "step": 2980
2030
+ },
2031
+ {
2032
+ "epoch": 1.1,
2033
+ "learning_rate": 0.00018400160091554835,
2034
+ "loss": 0.3095,
2035
+ "step": 2990
2036
+ },
2037
+ {
2038
+ "epoch": 1.1,
2039
+ "learning_rate": 0.0001838973479428066,
2040
+ "loss": 0.2954,
2041
+ "step": 3000
2042
+ },
2043
+ {
2044
+ "epoch": 1.1,
2045
+ "eval_loss": 0.4158288836479187,
2046
+ "eval_runtime": 223.8063,
2047
+ "eval_samples_per_second": 21.371,
2048
+ "eval_steps_per_second": 5.344,
2049
+ "step": 3000
2050
+ },
2051
+ {
2052
+ "epoch": 1.1,
2053
+ "learning_rate": 0.00018379278612670193,
2054
+ "loss": 0.3167,
2055
+ "step": 3010
2056
+ },
2057
+ {
2058
+ "epoch": 1.11,
2059
+ "learning_rate": 0.00018368791585214784,
2060
+ "loss": 0.2872,
2061
+ "step": 3020
2062
+ },
2063
+ {
2064
+ "epoch": 1.11,
2065
+ "learning_rate": 0.00018358273750519337,
2066
+ "loss": 0.2999,
2067
+ "step": 3030
2068
+ },
2069
+ {
2070
+ "epoch": 1.11,
2071
+ "learning_rate": 0.00018347725147302158,
2072
+ "loss": 0.3229,
2073
+ "step": 3040
2074
+ },
2075
+ {
2076
+ "epoch": 1.12,
2077
+ "learning_rate": 0.00018337145814394825,
2078
+ "loss": 0.3332,
2079
+ "step": 3050
2080
+ },
2081
+ {
2082
+ "epoch": 1.12,
2083
+ "learning_rate": 0.0001832653579074203,
2084
+ "loss": 0.2818,
2085
+ "step": 3060
2086
+ },
2087
+ {
2088
+ "epoch": 1.12,
2089
+ "learning_rate": 0.00018315895115401457,
2090
+ "loss": 0.3198,
2091
+ "step": 3070
2092
+ },
2093
+ {
2094
+ "epoch": 1.13,
2095
+ "learning_rate": 0.00018305223827543604,
2096
+ "loss": 0.3504,
2097
+ "step": 3080
2098
+ },
2099
+ {
2100
+ "epoch": 1.13,
2101
+ "learning_rate": 0.0001829452196645168,
2102
+ "loss": 0.3392,
2103
+ "step": 3090
2104
+ },
2105
+ {
2106
+ "epoch": 1.14,
2107
+ "learning_rate": 0.00018283789571521436,
2108
+ "loss": 0.3023,
2109
+ "step": 3100
2110
+ },
2111
+ {
2112
+ "epoch": 1.14,
2113
+ "eval_loss": 0.40502193570137024,
2114
+ "eval_runtime": 223.8877,
2115
+ "eval_samples_per_second": 21.363,
2116
+ "eval_steps_per_second": 5.342,
2117
+ "step": 3100
2118
+ },
2119
+ {
2120
+ "epoch": 1.14,
2121
+ "learning_rate": 0.00018273026682261013,
2122
+ "loss": 0.3164,
2123
+ "step": 3110
2124
+ },
2125
+ {
2126
+ "epoch": 1.14,
2127
+ "learning_rate": 0.0001826223333829082,
2128
+ "loss": 0.2985,
2129
+ "step": 3120
2130
+ },
2131
+ {
2132
+ "epoch": 1.15,
2133
+ "learning_rate": 0.00018251409579343375,
2134
+ "loss": 0.2992,
2135
+ "step": 3130
2136
+ },
2137
+ {
2138
+ "epoch": 1.15,
2139
+ "learning_rate": 0.00018240555445263153,
2140
+ "loss": 0.3174,
2141
+ "step": 3140
2142
+ },
2143
+ {
2144
+ "epoch": 1.15,
2145
+ "learning_rate": 0.00018229670976006453,
2146
+ "loss": 0.3025,
2147
+ "step": 3150
2148
+ },
2149
+ {
2150
+ "epoch": 1.16,
2151
+ "learning_rate": 0.00018218756211641236,
2152
+ "loss": 0.3297,
2153
+ "step": 3160
2154
+ },
2155
+ {
2156
+ "epoch": 1.16,
2157
+ "learning_rate": 0.00018207811192346996,
2158
+ "loss": 0.3139,
2159
+ "step": 3170
2160
+ },
2161
+ {
2162
+ "epoch": 1.17,
2163
+ "learning_rate": 0.00018196835958414598,
2164
+ "loss": 0.2956,
2165
+ "step": 3180
2166
+ },
2167
+ {
2168
+ "epoch": 1.17,
2169
+ "learning_rate": 0.00018185830550246124,
2170
+ "loss": 0.3189,
2171
+ "step": 3190
2172
+ },
2173
+ {
2174
+ "epoch": 1.17,
2175
+ "learning_rate": 0.00018174795008354743,
2176
+ "loss": 0.2896,
2177
+ "step": 3200
2178
+ },
2179
+ {
2180
+ "epoch": 1.17,
2181
+ "eval_loss": 0.40529006719589233,
2182
+ "eval_runtime": 223.8738,
2183
+ "eval_samples_per_second": 21.365,
2184
+ "eval_steps_per_second": 5.342,
2185
+ "step": 3200
2186
+ },
2187
+ {
2188
+ "epoch": 1.18,
2189
+ "learning_rate": 0.00018163729373364554,
2190
+ "loss": 0.3085,
2191
+ "step": 3210
2192
+ },
2193
+ {
2194
+ "epoch": 1.18,
2195
+ "learning_rate": 0.0001815263368601043,
2196
+ "loss": 0.316,
2197
+ "step": 3220
2198
+ },
2199
+ {
2200
+ "epoch": 1.18,
2201
+ "learning_rate": 0.00018141507987137873,
2202
+ "loss": 0.326,
2203
+ "step": 3230
2204
+ },
2205
+ {
2206
+ "epoch": 1.19,
2207
+ "learning_rate": 0.00018130352317702865,
2208
+ "loss": 0.293,
2209
+ "step": 3240
2210
+ },
2211
+ {
2212
+ "epoch": 1.19,
2213
+ "learning_rate": 0.00018119166718771716,
2214
+ "loss": 0.2887,
2215
+ "step": 3250
2216
+ },
2217
+ {
2218
+ "epoch": 1.19,
2219
+ "learning_rate": 0.00018107951231520911,
2220
+ "loss": 0.3266,
2221
+ "step": 3260
2222
+ },
2223
+ {
2224
+ "epoch": 1.2,
2225
+ "learning_rate": 0.00018096705897236966,
2226
+ "loss": 0.3327,
2227
+ "step": 3270
2228
+ },
2229
+ {
2230
+ "epoch": 1.2,
2231
+ "learning_rate": 0.00018085430757316256,
2232
+ "loss": 0.2991,
2233
+ "step": 3280
2234
+ },
2235
+ {
2236
+ "epoch": 1.21,
2237
+ "learning_rate": 0.00018074125853264898,
2238
+ "loss": 0.3076,
2239
+ "step": 3290
2240
+ },
2241
+ {
2242
+ "epoch": 1.21,
2243
+ "learning_rate": 0.00018062791226698558,
2244
+ "loss": 0.3339,
2245
+ "step": 3300
2246
+ },
2247
+ {
2248
+ "epoch": 1.21,
2249
+ "eval_loss": 0.4053775668144226,
2250
+ "eval_runtime": 223.9954,
2251
+ "eval_samples_per_second": 21.353,
2252
+ "eval_steps_per_second": 5.339,
2253
+ "step": 3300
2254
+ },
2255
+ {
2256
+ "epoch": 1.21,
2257
+ "learning_rate": 0.00018051426919342317,
2258
+ "loss": 0.2935,
2259
+ "step": 3310
2260
+ },
2261
+ {
2262
+ "epoch": 1.22,
2263
+ "learning_rate": 0.00018040032973030536,
2264
+ "loss": 0.3497,
2265
+ "step": 3320
2266
+ },
2267
+ {
2268
+ "epoch": 1.22,
2269
+ "learning_rate": 0.00018028609429706664,
2270
+ "loss": 0.2897,
2271
+ "step": 3330
2272
+ },
2273
+ {
2274
+ "epoch": 1.22,
2275
+ "learning_rate": 0.00018017156331423114,
2276
+ "loss": 0.312,
2277
+ "step": 3340
2278
+ },
2279
+ {
2280
+ "epoch": 1.23,
2281
+ "learning_rate": 0.00018005673720341086,
2282
+ "loss": 0.303,
2283
+ "step": 3350
2284
+ },
2285
+ {
2286
+ "epoch": 1.23,
2287
+ "learning_rate": 0.00017994161638730432,
2288
+ "loss": 0.3253,
2289
+ "step": 3360
2290
+ },
2291
+ {
2292
+ "epoch": 1.23,
2293
+ "learning_rate": 0.00017982620128969488,
2294
+ "loss": 0.3322,
2295
+ "step": 3370
2296
+ },
2297
+ {
2298
+ "epoch": 1.24,
2299
+ "learning_rate": 0.0001797104923354492,
2300
+ "loss": 0.2988,
2301
+ "step": 3380
2302
+ },
2303
+ {
2304
+ "epoch": 1.24,
2305
+ "learning_rate": 0.00017959448995051575,
2306
+ "loss": 0.3094,
2307
+ "step": 3390
2308
+ },
2309
+ {
2310
+ "epoch": 1.25,
2311
+ "learning_rate": 0.00017947819456192306,
2312
+ "loss": 0.3118,
2313
+ "step": 3400
2314
+ },
2315
+ {
2316
+ "epoch": 1.25,
2317
+ "eval_loss": 0.3964312672615051,
2318
+ "eval_runtime": 223.8688,
2319
+ "eval_samples_per_second": 21.365,
2320
+ "eval_steps_per_second": 5.342,
2321
+ "step": 3400
2322
+ },
2323
+ {
2324
+ "epoch": 1.25,
2325
+ "learning_rate": 0.00017936160659777833,
2326
+ "loss": 0.2958,
2327
+ "step": 3410
2328
+ },
2329
+ {
2330
+ "epoch": 1.25,
2331
+ "learning_rate": 0.00017924472648726583,
2332
+ "loss": 0.3119,
2333
+ "step": 3420
2334
+ },
2335
+ {
2336
+ "epoch": 1.26,
2337
+ "learning_rate": 0.00017912755466064525,
2338
+ "loss": 0.3096,
2339
+ "step": 3430
2340
+ },
2341
+ {
2342
+ "epoch": 1.26,
2343
+ "learning_rate": 0.00017901009154925007,
2344
+ "loss": 0.2778,
2345
+ "step": 3440
2346
+ },
2347
+ {
2348
+ "epoch": 1.26,
2349
+ "learning_rate": 0.00017889233758548625,
2350
+ "loss": 0.2953,
2351
+ "step": 3450
2352
+ },
2353
+ {
2354
+ "epoch": 1.27,
2355
+ "learning_rate": 0.00017877429320283016,
2356
+ "loss": 0.3063,
2357
+ "step": 3460
2358
+ },
2359
+ {
2360
+ "epoch": 1.27,
2361
+ "learning_rate": 0.0001786559588358275,
2362
+ "loss": 0.2945,
2363
+ "step": 3470
2364
+ },
2365
+ {
2366
+ "epoch": 1.27,
2367
+ "learning_rate": 0.00017853733492009135,
2368
+ "loss": 0.3027,
2369
+ "step": 3480
2370
+ },
2371
+ {
2372
+ "epoch": 1.28,
2373
+ "learning_rate": 0.0001784184218923007,
2374
+ "loss": 0.2654,
2375
+ "step": 3490
2376
+ },
2377
+ {
2378
+ "epoch": 1.28,
2379
+ "learning_rate": 0.0001782992201901988,
2380
+ "loss": 0.3289,
2381
+ "step": 3500
2382
+ },
2383
+ {
2384
+ "epoch": 1.28,
2385
+ "eval_loss": 0.39914268255233765,
2386
+ "eval_runtime": 223.8716,
2387
+ "eval_samples_per_second": 21.365,
2388
+ "eval_steps_per_second": 5.342,
2389
+ "step": 3500
2390
+ },
2391
+ {
2392
+ "epoch": 1.29,
2393
+ "learning_rate": 0.0001781797302525916,
2394
+ "loss": 0.2894,
2395
+ "step": 3510
2396
+ },
2397
+ {
2398
+ "epoch": 1.29,
2399
+ "learning_rate": 0.00017805995251934614,
2400
+ "loss": 0.2776,
2401
+ "step": 3520
2402
+ },
2403
+ {
2404
+ "epoch": 1.29,
2405
+ "learning_rate": 0.00017793988743138877,
2406
+ "loss": 0.288,
2407
+ "step": 3530
2408
+ },
2409
+ {
2410
+ "epoch": 1.3,
2411
+ "learning_rate": 0.00017781953543070372,
2412
+ "loss": 0.3229,
2413
+ "step": 3540
2414
+ },
2415
+ {
2416
+ "epoch": 1.3,
2417
+ "learning_rate": 0.00017769889696033154,
2418
+ "loss": 0.285,
2419
+ "step": 3550
2420
+ },
2421
+ {
2422
+ "epoch": 1.3,
2423
+ "learning_rate": 0.0001775779724643671,
2424
+ "loss": 0.321,
2425
+ "step": 3560
2426
+ },
2427
+ {
2428
+ "epoch": 1.31,
2429
+ "learning_rate": 0.0001774567623879583,
2430
+ "loss": 0.3222,
2431
+ "step": 3570
2432
+ },
2433
+ {
2434
+ "epoch": 1.31,
2435
+ "learning_rate": 0.00017733526717730435,
2436
+ "loss": 0.3087,
2437
+ "step": 3580
2438
+ },
2439
+ {
2440
+ "epoch": 1.32,
2441
+ "learning_rate": 0.00017721348727965408,
2442
+ "loss": 0.2904,
2443
+ "step": 3590
2444
+ },
2445
+ {
2446
+ "epoch": 1.32,
2447
+ "learning_rate": 0.00017709142314330424,
2448
+ "loss": 0.2984,
2449
+ "step": 3600
2450
+ },
2451
+ {
2452
+ "epoch": 1.32,
2453
+ "eval_loss": 0.40588298439979553,
2454
+ "eval_runtime": 223.9372,
2455
+ "eval_samples_per_second": 21.359,
2456
+ "eval_steps_per_second": 5.341,
2457
+ "step": 3600
2458
+ },
2459
+ {
2460
+ "epoch": 1.32,
2461
+ "learning_rate": 0.00017696907521759804,
2462
+ "loss": 0.3216,
2463
+ "step": 3610
2464
+ },
2465
+ {
2466
+ "epoch": 1.33,
2467
+ "learning_rate": 0.00017684644395292326,
2468
+ "loss": 0.3019,
2469
+ "step": 3620
2470
+ },
2471
+ {
2472
+ "epoch": 1.33,
2473
+ "learning_rate": 0.00017672352980071078,
2474
+ "loss": 0.3272,
2475
+ "step": 3630
2476
+ },
2477
+ {
2478
+ "epoch": 1.33,
2479
+ "learning_rate": 0.00017660033321343285,
2480
+ "loss": 0.2892,
2481
+ "step": 3640
2482
+ },
2483
+ {
2484
+ "epoch": 1.34,
2485
+ "learning_rate": 0.0001764768546446014,
2486
+ "loss": 0.2921,
2487
+ "step": 3650
2488
+ },
2489
+ {
2490
+ "epoch": 1.34,
2491
+ "learning_rate": 0.00017635309454876636,
2492
+ "loss": 0.3105,
2493
+ "step": 3660
2494
+ },
2495
+ {
2496
+ "epoch": 1.34,
2497
+ "learning_rate": 0.00017622905338151408,
2498
+ "loss": 0.3205,
2499
+ "step": 3670
2500
+ },
2501
+ {
2502
+ "epoch": 1.35,
2503
+ "learning_rate": 0.00017610473159946556,
2504
+ "loss": 0.2838,
2505
+ "step": 3680
2506
+ },
2507
+ {
2508
+ "epoch": 1.35,
2509
+ "learning_rate": 0.00017598012966027482,
2510
+ "loss": 0.2762,
2511
+ "step": 3690
2512
+ },
2513
+ {
2514
+ "epoch": 1.36,
2515
+ "learning_rate": 0.0001758552480226271,
2516
+ "loss": 0.3277,
2517
+ "step": 3700
2518
+ },
2519
+ {
2520
+ "epoch": 1.36,
2521
+ "eval_loss": 0.3979549705982208,
2522
+ "eval_runtime": 223.8768,
2523
+ "eval_samples_per_second": 21.364,
2524
+ "eval_steps_per_second": 5.342,
2525
+ "step": 3700
2526
+ },
2527
+ {
2528
+ "epoch": 1.36,
2529
+ "learning_rate": 0.00017573008714623746,
2530
+ "loss": 0.2709,
2531
+ "step": 3710
2532
+ },
2533
+ {
2534
+ "epoch": 1.36,
2535
+ "learning_rate": 0.00017560464749184876,
2536
+ "loss": 0.2918,
2537
+ "step": 3720
2538
+ },
2539
+ {
2540
+ "epoch": 1.37,
2541
+ "learning_rate": 0.00017547892952123005,
2542
+ "loss": 0.3098,
2543
+ "step": 3730
2544
+ },
2545
+ {
2546
+ "epoch": 1.37,
2547
+ "learning_rate": 0.00017535293369717506,
2548
+ "loss": 0.3233,
2549
+ "step": 3740
2550
+ },
2551
+ {
2552
+ "epoch": 1.37,
2553
+ "learning_rate": 0.00017522666048350023,
2554
+ "loss": 0.31,
2555
+ "step": 3750
2556
+ },
2557
+ {
2558
+ "epoch": 1.38,
2559
+ "learning_rate": 0.00017510011034504324,
2560
+ "loss": 0.3337,
2561
+ "step": 3760
2562
+ },
2563
+ {
2564
+ "epoch": 1.38,
2565
+ "learning_rate": 0.00017497328374766112,
2566
+ "loss": 0.2845,
2567
+ "step": 3770
2568
+ },
2569
+ {
2570
+ "epoch": 1.38,
2571
+ "learning_rate": 0.00017484618115822857,
2572
+ "loss": 0.3073,
2573
+ "step": 3780
2574
+ },
2575
+ {
2576
+ "epoch": 1.39,
2577
+ "learning_rate": 0.00017471880304463638,
2578
+ "loss": 0.2977,
2579
+ "step": 3790
2580
+ },
2581
+ {
2582
+ "epoch": 1.39,
2583
+ "learning_rate": 0.0001745911498757895,
2584
+ "loss": 0.3011,
2585
+ "step": 3800
2586
+ },
2587
+ {
2588
+ "epoch": 1.39,
2589
+ "eval_loss": 0.4045478105545044,
2590
+ "eval_runtime": 223.7983,
2591
+ "eval_samples_per_second": 21.372,
2592
+ "eval_steps_per_second": 5.344,
2593
+ "step": 3800
2594
+ },
2595
+ {
2596
+ "epoch": 1.4,
2597
+ "learning_rate": 0.00017446322212160545,
2598
+ "loss": 0.3029,
2599
+ "step": 3810
2600
+ },
2601
+ {
2602
+ "epoch": 1.4,
2603
+ "learning_rate": 0.0001743350202530126,
2604
+ "loss": 0.3017,
2605
+ "step": 3820
2606
+ },
2607
+ {
2608
+ "epoch": 1.4,
2609
+ "learning_rate": 0.00017420654474194832,
2610
+ "loss": 0.3179,
2611
+ "step": 3830
2612
+ },
2613
+ {
2614
+ "epoch": 1.41,
2615
+ "learning_rate": 0.00017407779606135732,
2616
+ "loss": 0.2728,
2617
+ "step": 3840
2618
+ },
2619
+ {
2620
+ "epoch": 1.41,
2621
+ "learning_rate": 0.00017394877468518996,
2622
+ "loss": 0.286,
2623
+ "step": 3850
2624
+ },
2625
+ {
2626
+ "epoch": 1.41,
2627
+ "learning_rate": 0.00017381948108840042,
2628
+ "loss": 0.3009,
2629
+ "step": 3860
2630
+ },
2631
+ {
2632
+ "epoch": 1.42,
2633
+ "learning_rate": 0.00017368991574694495,
2634
+ "loss": 0.2946,
2635
+ "step": 3870
2636
+ },
2637
+ {
2638
+ "epoch": 1.42,
2639
+ "learning_rate": 0.0001735600791377802,
2640
+ "loss": 0.3229,
2641
+ "step": 3880
2642
+ },
2643
+ {
2644
+ "epoch": 1.43,
2645
+ "learning_rate": 0.00017342997173886134,
2646
+ "loss": 0.3372,
2647
+ "step": 3890
2648
+ },
2649
+ {
2650
+ "epoch": 1.43,
2651
+ "learning_rate": 0.00017329959402914046,
2652
+ "loss": 0.3194,
2653
+ "step": 3900
2654
+ },
2655
+ {
2656
+ "epoch": 1.43,
2657
+ "eval_loss": 0.40152063965797424,
2658
+ "eval_runtime": 223.8706,
2659
+ "eval_samples_per_second": 21.365,
2660
+ "eval_steps_per_second": 5.342,
2661
+ "step": 3900
2662
+ },
2663
+ {
2664
+ "epoch": 1.43,
2665
+ "learning_rate": 0.0001731689464885647,
2666
+ "loss": 0.2674,
2667
+ "step": 3910
2668
+ },
2669
+ {
2670
+ "epoch": 1.44,
2671
+ "learning_rate": 0.00017303802959807443,
2672
+ "loss": 0.2558,
2673
+ "step": 3920
2674
+ },
2675
+ {
2676
+ "epoch": 1.44,
2677
+ "learning_rate": 0.0001729068438396016,
2678
+ "loss": 0.3186,
2679
+ "step": 3930
2680
+ },
2681
+ {
2682
+ "epoch": 1.44,
2683
+ "learning_rate": 0.00017277538969606793,
2684
+ "loss": 0.3298,
2685
+ "step": 3940
2686
+ },
2687
+ {
2688
+ "epoch": 1.45,
2689
+ "learning_rate": 0.00017264366765138317,
2690
+ "loss": 0.2991,
2691
+ "step": 3950
2692
+ },
2693
+ {
2694
+ "epoch": 1.45,
2695
+ "learning_rate": 0.00017251167819044315,
2696
+ "loss": 0.2652,
2697
+ "step": 3960
2698
+ },
2699
+ {
2700
+ "epoch": 1.45,
2701
+ "learning_rate": 0.0001723794217991282,
2702
+ "loss": 0.3177,
2703
+ "step": 3970
2704
+ },
2705
+ {
2706
+ "epoch": 1.46,
2707
+ "learning_rate": 0.00017224689896430117,
2708
+ "loss": 0.3045,
2709
+ "step": 3980
2710
+ },
2711
+ {
2712
+ "epoch": 1.46,
2713
+ "learning_rate": 0.00017211411017380594,
2714
+ "loss": 0.3146,
2715
+ "step": 3990
2716
+ },
2717
+ {
2718
+ "epoch": 1.47,
2719
+ "learning_rate": 0.00017198105591646528,
2720
+ "loss": 0.2921,
2721
+ "step": 4000
2722
+ },
2723
+ {
2724
+ "epoch": 1.47,
2725
+ "eval_loss": 0.400907427072525,
2726
+ "eval_runtime": 224.0601,
2727
+ "eval_samples_per_second": 21.347,
2728
+ "eval_steps_per_second": 5.338,
2729
+ "step": 4000
2730
+ },
2731
+ {
2732
+ "epoch": 1.47,
2733
+ "learning_rate": 0.00017184773668207917,
2734
+ "loss": 0.3441,
2735
+ "step": 4010
2736
+ },
2737
+ {
2738
+ "epoch": 1.47,
2739
+ "learning_rate": 0.00017171415296142315,
2740
+ "loss": 0.2876,
2741
+ "step": 4020
2742
+ },
2743
+ {
2744
+ "epoch": 1.48,
2745
+ "learning_rate": 0.00017158030524624626,
2746
+ "loss": 0.2755,
2747
+ "step": 4030
2748
+ },
2749
+ {
2750
+ "epoch": 1.48,
2751
+ "learning_rate": 0.0001714461940292695,
2752
+ "loss": 0.3169,
2753
+ "step": 4040
2754
+ },
2755
+ {
2756
+ "epoch": 1.48,
2757
+ "learning_rate": 0.00017131181980418374,
2758
+ "loss": 0.2844,
2759
+ "step": 4050
2760
+ },
2761
+ {
2762
+ "epoch": 1.49,
2763
+ "learning_rate": 0.00017117718306564812,
2764
+ "loss": 0.2821,
2765
+ "step": 4060
2766
+ },
2767
+ {
2768
+ "epoch": 1.49,
2769
+ "learning_rate": 0.00017104228430928805,
2770
+ "loss": 0.3002,
2771
+ "step": 4070
2772
+ },
2773
+ {
2774
+ "epoch": 1.49,
2775
+ "learning_rate": 0.00017090712403169364,
2776
+ "loss": 0.3232,
2777
+ "step": 4080
2778
+ },
2779
+ {
2780
+ "epoch": 1.5,
2781
+ "learning_rate": 0.00017077170273041757,
2782
+ "loss": 0.3135,
2783
+ "step": 4090
2784
+ },
2785
+ {
2786
+ "epoch": 1.5,
2787
+ "learning_rate": 0.00017063602090397346,
2788
+ "loss": 0.2917,
2789
+ "step": 4100
2790
+ },
2791
+ {
2792
+ "epoch": 1.5,
2793
+ "eval_loss": 0.4019235670566559,
2794
+ "eval_runtime": 224.3869,
2795
+ "eval_samples_per_second": 21.316,
2796
+ "eval_steps_per_second": 5.33,
2797
+ "step": 4100
2798
+ },
2799
+ {
2800
+ "epoch": 1.51,
2801
+ "learning_rate": 0.00017050007905183398,
2802
+ "loss": 0.3089,
2803
+ "step": 4110
2804
+ },
2805
+ {
2806
+ "epoch": 1.51,
2807
+ "learning_rate": 0.000170363877674429,
2808
+ "loss": 0.3063,
2809
+ "step": 4120
2810
+ },
2811
+ {
2812
+ "epoch": 1.51,
2813
+ "learning_rate": 0.00017022741727314373,
2814
+ "loss": 0.3056,
2815
+ "step": 4130
2816
+ },
2817
+ {
2818
+ "epoch": 1.52,
2819
+ "learning_rate": 0.00017009069835031694,
2820
+ "loss": 0.3168,
2821
+ "step": 4140
2822
+ },
2823
+ {
2824
+ "epoch": 1.52,
2825
+ "learning_rate": 0.00016995372140923907,
2826
+ "loss": 0.3062,
2827
+ "step": 4150
2828
+ },
2829
+ {
2830
+ "epoch": 1.52,
2831
+ "learning_rate": 0.00016981648695415033,
2832
+ "loss": 0.2744,
2833
+ "step": 4160
2834
+ },
2835
+ {
2836
+ "epoch": 1.53,
2837
+ "learning_rate": 0.00016967899549023895,
2838
+ "loss": 0.2877,
2839
+ "step": 4170
2840
+ },
2841
+ {
2842
+ "epoch": 1.53,
2843
+ "learning_rate": 0.00016954124752363922,
2844
+ "loss": 0.3173,
2845
+ "step": 4180
2846
+ },
2847
+ {
2848
+ "epoch": 1.54,
2849
+ "learning_rate": 0.00016940324356142972,
2850
+ "loss": 0.2954,
2851
+ "step": 4190
2852
+ },
2853
+ {
2854
+ "epoch": 1.54,
2855
+ "learning_rate": 0.00016926498411163135,
2856
+ "loss": 0.2792,
2857
+ "step": 4200
2858
+ },
2859
+ {
2860
+ "epoch": 1.54,
2861
+ "eval_loss": 0.4046282172203064,
2862
+ "eval_runtime": 224.0328,
2863
+ "eval_samples_per_second": 21.35,
2864
+ "eval_steps_per_second": 5.339,
2865
+ "step": 4200
2866
+ },
2867
+ {
2868
+ "epoch": 1.54,
2869
+ "learning_rate": 0.00016912646968320552,
2870
+ "loss": 0.3168,
2871
+ "step": 4210
2872
+ },
2873
+ {
2874
+ "epoch": 1.55,
2875
+ "learning_rate": 0.00016898770078605226,
2876
+ "loss": 0.2854,
2877
+ "step": 4220
2878
+ },
2879
+ {
2880
+ "epoch": 1.55,
2881
+ "learning_rate": 0.00016884867793100843,
2882
+ "loss": 0.2979,
2883
+ "step": 4230
2884
+ },
2885
+ {
2886
+ "epoch": 1.55,
2887
+ "learning_rate": 0.00016870940162984566,
2888
+ "loss": 0.2656,
2889
+ "step": 4240
2890
+ },
2891
+ {
2892
+ "epoch": 1.56,
2893
+ "learning_rate": 0.00016856987239526863,
2894
+ "loss": 0.3169,
2895
+ "step": 4250
2896
+ },
2897
+ {
2898
+ "epoch": 1.56,
2899
+ "learning_rate": 0.00016843009074091306,
2900
+ "loss": 0.2872,
2901
+ "step": 4260
2902
+ },
2903
+ {
2904
+ "epoch": 1.56,
2905
+ "learning_rate": 0.00016829005718134397,
2906
+ "loss": 0.2954,
2907
+ "step": 4270
2908
+ },
2909
+ {
2910
+ "epoch": 1.57,
2911
+ "learning_rate": 0.00016814977223205362,
2912
+ "loss": 0.2885,
2913
+ "step": 4280
2914
+ },
2915
+ {
2916
+ "epoch": 1.57,
2917
+ "learning_rate": 0.00016800923640945974,
2918
+ "loss": 0.2927,
2919
+ "step": 4290
2920
+ },
2921
+ {
2922
+ "epoch": 1.58,
2923
+ "learning_rate": 0.0001678684502309035,
2924
+ "loss": 0.2886,
2925
+ "step": 4300
2926
+ },
2927
+ {
2928
+ "epoch": 1.58,
2929
+ "eval_loss": 0.4055146872997284,
2930
+ "eval_runtime": 224.1142,
2931
+ "eval_samples_per_second": 21.342,
2932
+ "eval_steps_per_second": 5.337,
2933
+ "step": 4300
2934
+ },
2935
+ {
2936
+ "epoch": 1.58,
2937
+ "learning_rate": 0.00016772741421464772,
2938
+ "loss": 0.3042,
2939
+ "step": 4310
2940
+ },
2941
+ {
2942
+ "epoch": 1.58,
2943
+ "learning_rate": 0.00016758612887987498,
2944
+ "loss": 0.2858,
2945
+ "step": 4320
2946
+ },
2947
+ {
2948
+ "epoch": 1.59,
2949
+ "learning_rate": 0.00016744459474668557,
2950
+ "loss": 0.3072,
2951
+ "step": 4330
2952
+ },
2953
+ {
2954
+ "epoch": 1.59,
2955
+ "learning_rate": 0.00016730281233609572,
2956
+ "loss": 0.3126,
2957
+ "step": 4340
2958
+ },
2959
+ {
2960
+ "epoch": 1.59,
2961
+ "learning_rate": 0.00016716078217003557,
2962
+ "loss": 0.2748,
2963
+ "step": 4350
2964
+ },
2965
+ {
2966
+ "epoch": 1.6,
2967
+ "learning_rate": 0.00016701850477134734,
2968
+ "loss": 0.2873,
2969
+ "step": 4360
2970
+ },
2971
+ {
2972
+ "epoch": 1.6,
2973
+ "learning_rate": 0.00016687598066378336,
2974
+ "loss": 0.3051,
2975
+ "step": 4370
2976
+ },
2977
+ {
2978
+ "epoch": 1.6,
2979
+ "learning_rate": 0.00016673321037200407,
2980
+ "loss": 0.3077,
2981
+ "step": 4380
2982
+ },
2983
+ {
2984
+ "epoch": 1.61,
2985
+ "learning_rate": 0.0001665901944215763,
2986
+ "loss": 0.2708,
2987
+ "step": 4390
2988
+ },
2989
+ {
2990
+ "epoch": 1.61,
2991
+ "learning_rate": 0.00016644693333897108,
2992
+ "loss": 0.2947,
2993
+ "step": 4400
2994
+ },
2995
+ {
2996
+ "epoch": 1.61,
2997
+ "eval_loss": 0.40514013171195984,
2998
+ "eval_runtime": 224.1094,
2999
+ "eval_samples_per_second": 21.342,
3000
+ "eval_steps_per_second": 5.337,
3001
+ "step": 4400
3002
+ },
3003
+ {
3004
+ "epoch": 1.62,
3005
+ "learning_rate": 0.0001663034276515619,
3006
+ "loss": 0.306,
3007
+ "step": 4410
3008
+ },
3009
+ {
3010
+ "epoch": 1.62,
3011
+ "learning_rate": 0.00016615967788762261,
3012
+ "loss": 0.3076,
3013
+ "step": 4420
3014
+ },
3015
+ {
3016
+ "epoch": 1.62,
3017
+ "learning_rate": 0.00016601568457632566,
3018
+ "loss": 0.2895,
3019
+ "step": 4430
3020
+ },
3021
+ {
3022
+ "epoch": 1.63,
3023
+ "learning_rate": 0.00016587144824773992,
3024
+ "loss": 0.2885,
3025
+ "step": 4440
3026
+ },
3027
+ {
3028
+ "epoch": 1.63,
3029
+ "learning_rate": 0.00016572696943282903,
3030
+ "loss": 0.3046,
3031
+ "step": 4450
3032
+ },
3033
+ {
3034
+ "epoch": 1.63,
3035
+ "learning_rate": 0.00016558224866344907,
3036
+ "loss": 0.3037,
3037
+ "step": 4460
3038
+ },
3039
+ {
3040
+ "epoch": 1.64,
3041
+ "learning_rate": 0.000165437286472347,
3042
+ "loss": 0.28,
3043
+ "step": 4470
3044
+ },
3045
+ {
3046
+ "epoch": 1.64,
3047
+ "learning_rate": 0.00016529208339315833,
3048
+ "loss": 0.2903,
3049
+ "step": 4480
3050
+ },
3051
+ {
3052
+ "epoch": 1.64,
3053
+ "learning_rate": 0.00016514663996040544,
3054
+ "loss": 0.2849,
3055
+ "step": 4490
3056
+ },
3057
+ {
3058
+ "epoch": 1.65,
3059
+ "learning_rate": 0.00016500095670949548,
3060
+ "loss": 0.2975,
3061
+ "step": 4500
3062
+ },
3063
+ {
3064
+ "epoch": 1.65,
3065
+ "eval_loss": 0.4067119359970093,
3066
+ "eval_runtime": 224.4914,
3067
+ "eval_samples_per_second": 21.306,
3068
+ "eval_steps_per_second": 5.328,
3069
+ "step": 4500
3070
+ },
3071
+ {
3072
+ "epoch": 1.65,
3073
+ "learning_rate": 0.00016485503417671836,
3074
+ "loss": 0.3115,
3075
+ "step": 4510
3076
+ },
3077
+ {
3078
+ "epoch": 1.66,
3079
+ "learning_rate": 0.00016470887289924492,
3080
+ "loss": 0.3409,
3081
+ "step": 4520
3082
+ },
3083
+ {
3084
+ "epoch": 1.66,
3085
+ "learning_rate": 0.00016456247341512485,
3086
+ "loss": 0.2604,
3087
+ "step": 4530
3088
+ },
3089
+ {
3090
+ "epoch": 1.66,
3091
+ "learning_rate": 0.00016441583626328467,
3092
+ "loss": 0.3226,
3093
+ "step": 4540
3094
+ },
3095
+ {
3096
+ "epoch": 1.67,
3097
+ "learning_rate": 0.00016426896198352587,
3098
+ "loss": 0.2816,
3099
+ "step": 4550
3100
+ },
3101
+ {
3102
+ "epoch": 1.67,
3103
+ "learning_rate": 0.00016412185111652278,
3104
+ "loss": 0.319,
3105
+ "step": 4560
3106
+ },
3107
+ {
3108
+ "epoch": 1.67,
3109
+ "learning_rate": 0.00016397450420382076,
3110
+ "loss": 0.2831,
3111
+ "step": 4570
3112
+ },
3113
+ {
3114
+ "epoch": 1.68,
3115
+ "learning_rate": 0.00016382692178783402,
3116
+ "loss": 0.2974,
3117
+ "step": 4580
3118
+ },
3119
+ {
3120
+ "epoch": 1.68,
3121
+ "learning_rate": 0.00016367910441184374,
3122
+ "loss": 0.2772,
3123
+ "step": 4590
3124
+ },
3125
+ {
3126
+ "epoch": 1.69,
3127
+ "learning_rate": 0.00016353105261999605,
3128
+ "loss": 0.3091,
3129
+ "step": 4600
3130
+ },
3131
+ {
3132
+ "epoch": 1.69,
3133
+ "eval_loss": 0.3946963846683502,
3134
+ "eval_runtime": 224.3647,
3135
+ "eval_samples_per_second": 21.318,
3136
+ "eval_steps_per_second": 5.331,
3137
+ "step": 4600
3138
+ },
3139
+ {
3140
+ "epoch": 1.69,
3141
+ "learning_rate": 0.00016338276695729994,
3142
+ "loss": 0.286,
3143
+ "step": 4610
3144
+ },
3145
+ {
3146
+ "epoch": 1.69,
3147
+ "learning_rate": 0.00016323424796962544,
3148
+ "loss": 0.2721,
3149
+ "step": 4620
3150
+ },
3151
+ {
3152
+ "epoch": 1.7,
3153
+ "learning_rate": 0.0001630854962037014,
3154
+ "loss": 0.2983,
3155
+ "step": 4630
3156
+ },
3157
+ {
3158
+ "epoch": 1.7,
3159
+ "learning_rate": 0.00016293651220711364,
3160
+ "loss": 0.2875,
3161
+ "step": 4640
3162
+ },
3163
+ {
3164
+ "epoch": 1.7,
3165
+ "learning_rate": 0.0001627872965283028,
3166
+ "loss": 0.2859,
3167
+ "step": 4650
3168
+ },
3169
+ {
3170
+ "epoch": 1.71,
3171
+ "learning_rate": 0.00016263784971656247,
3172
+ "loss": 0.2603,
3173
+ "step": 4660
3174
+ },
3175
+ {
3176
+ "epoch": 1.71,
3177
+ "learning_rate": 0.00016248817232203698,
3178
+ "loss": 0.3231,
3179
+ "step": 4670
3180
+ },
3181
+ {
3182
+ "epoch": 1.71,
3183
+ "learning_rate": 0.00016233826489571963,
3184
+ "loss": 0.2927,
3185
+ "step": 4680
3186
+ },
3187
+ {
3188
+ "epoch": 1.72,
3189
+ "learning_rate": 0.00016218812798945038,
3190
+ "loss": 0.297,
3191
+ "step": 4690
3192
+ },
3193
+ {
3194
+ "epoch": 1.72,
3195
+ "learning_rate": 0.00016203776215591403,
3196
+ "loss": 0.2908,
3197
+ "step": 4700
3198
+ },
3199
+ {
3200
+ "epoch": 1.72,
3201
+ "eval_loss": 0.403292179107666,
3202
+ "eval_runtime": 224.3888,
3203
+ "eval_samples_per_second": 21.316,
3204
+ "eval_steps_per_second": 5.33,
3205
+ "step": 4700
3206
+ },
3207
+ {
3208
+ "epoch": 1.73,
3209
+ "learning_rate": 0.0001618871679486381,
3210
+ "loss": 0.3094,
3211
+ "step": 4710
3212
+ },
3213
+ {
3214
+ "epoch": 1.73,
3215
+ "learning_rate": 0.00016173634592199076,
3216
+ "loss": 0.29,
3217
+ "step": 4720
3218
+ },
3219
+ {
3220
+ "epoch": 1.73,
3221
+ "learning_rate": 0.00016158529663117888,
3222
+ "loss": 0.3115,
3223
+ "step": 4730
3224
+ },
3225
+ {
3226
+ "epoch": 1.74,
3227
+ "learning_rate": 0.00016143402063224584,
3228
+ "loss": 0.3139,
3229
+ "step": 4740
3230
+ },
3231
+ {
3232
+ "epoch": 1.74,
3233
+ "learning_rate": 0.00016128251848206976,
3234
+ "loss": 0.302,
3235
+ "step": 4750
3236
+ },
3237
+ {
3238
+ "epoch": 1.74,
3239
+ "learning_rate": 0.00016113079073836107,
3240
+ "loss": 0.2777,
3241
+ "step": 4760
3242
+ },
3243
+ {
3244
+ "epoch": 1.75,
3245
+ "learning_rate": 0.0001609788379596608,
3246
+ "loss": 0.2906,
3247
+ "step": 4770
3248
+ },
3249
+ {
3250
+ "epoch": 1.75,
3251
+ "learning_rate": 0.00016082666070533832,
3252
+ "loss": 0.3028,
3253
+ "step": 4780
3254
+ },
3255
+ {
3256
+ "epoch": 1.75,
3257
+ "learning_rate": 0.0001606742595355893,
3258
+ "loss": 0.3159,
3259
+ "step": 4790
3260
+ },
3261
+ {
3262
+ "epoch": 1.76,
3263
+ "learning_rate": 0.0001605216350114338,
3264
+ "loss": 0.2864,
3265
+ "step": 4800
3266
+ },
3267
+ {
3268
+ "epoch": 1.76,
3269
+ "eval_loss": 0.40955105423927307,
3270
+ "eval_runtime": 224.3793,
3271
+ "eval_samples_per_second": 21.317,
3272
+ "eval_steps_per_second": 5.33,
3273
+ "step": 4800
3274
+ },
3275
+ {
3276
+ "epoch": 1.76,
3277
+ "learning_rate": 0.00016036878769471401,
3278
+ "loss": 0.3017,
3279
+ "step": 4810
3280
+ },
3281
+ {
3282
+ "epoch": 1.77,
3283
+ "learning_rate": 0.00016021571814809227,
3284
+ "loss": 0.2986,
3285
+ "step": 4820
3286
+ },
3287
+ {
3288
+ "epoch": 1.77,
3289
+ "learning_rate": 0.000160062426935049,
3290
+ "loss": 0.3077,
3291
+ "step": 4830
3292
+ },
3293
+ {
3294
+ "epoch": 1.77,
3295
+ "learning_rate": 0.00015990891461988065,
3296
+ "loss": 0.2758,
3297
+ "step": 4840
3298
+ },
3299
+ {
3300
+ "epoch": 1.78,
3301
+ "learning_rate": 0.00015975518176769755,
3302
+ "loss": 0.2592,
3303
+ "step": 4850
3304
+ },
3305
+ {
3306
+ "epoch": 1.78,
3307
+ "learning_rate": 0.0001596012289444219,
3308
+ "loss": 0.3021,
3309
+ "step": 4860
3310
+ },
3311
+ {
3312
+ "epoch": 1.78,
3313
+ "learning_rate": 0.00015944705671678565,
3314
+ "loss": 0.3047,
3315
+ "step": 4870
3316
+ },
3317
+ {
3318
+ "epoch": 1.79,
3319
+ "learning_rate": 0.00015929266565232846,
3320
+ "loss": 0.2785,
3321
+ "step": 4880
3322
+ },
3323
+ {
3324
+ "epoch": 1.79,
3325
+ "learning_rate": 0.00015913805631939546,
3326
+ "loss": 0.2839,
3327
+ "step": 4890
3328
+ },
3329
+ {
3330
+ "epoch": 1.8,
3331
+ "learning_rate": 0.00015898322928713544,
3332
+ "loss": 0.2788,
3333
+ "step": 4900
3334
+ },
3335
+ {
3336
+ "epoch": 1.8,
3337
+ "eval_loss": 0.40693148970603943,
3338
+ "eval_runtime": 224.6917,
3339
+ "eval_samples_per_second": 21.287,
3340
+ "eval_steps_per_second": 5.323,
3341
+ "step": 4900
3342
+ },
3343
+ {
3344
+ "epoch": 1.8,
3345
+ "learning_rate": 0.00015882818512549836,
3346
+ "loss": 0.2875,
3347
+ "step": 4910
3348
+ },
3349
+ {
3350
+ "epoch": 1.8,
3351
+ "learning_rate": 0.00015867292440523378,
3352
+ "loss": 0.2907,
3353
+ "step": 4920
3354
+ },
3355
+ {
3356
+ "epoch": 1.81,
3357
+ "learning_rate": 0.00015851744769788818,
3358
+ "loss": 0.2776,
3359
+ "step": 4930
3360
+ },
3361
+ {
3362
+ "epoch": 1.81,
3363
+ "learning_rate": 0.00015836175557580324,
3364
+ "loss": 0.2608,
3365
+ "step": 4940
3366
+ },
3367
+ {
3368
+ "epoch": 1.81,
3369
+ "learning_rate": 0.00015820584861211368,
3370
+ "loss": 0.2945,
3371
+ "step": 4950
3372
+ },
3373
+ {
3374
+ "epoch": 1.82,
3375
+ "learning_rate": 0.000158049727380745,
3376
+ "loss": 0.2771,
3377
+ "step": 4960
3378
+ },
3379
+ {
3380
+ "epoch": 1.82,
3381
+ "learning_rate": 0.0001578933924564115,
3382
+ "loss": 0.2802,
3383
+ "step": 4970
3384
+ },
3385
+ {
3386
+ "epoch": 1.82,
3387
+ "learning_rate": 0.0001577368444146142,
3388
+ "loss": 0.2849,
3389
+ "step": 4980
3390
+ },
3391
+ {
3392
+ "epoch": 1.83,
3393
+ "learning_rate": 0.00015758008383163854,
3394
+ "loss": 0.3013,
3395
+ "step": 4990
3396
+ },
3397
+ {
3398
+ "epoch": 1.83,
3399
+ "learning_rate": 0.0001574231112845524,
3400
+ "loss": 0.2942,
3401
+ "step": 5000
3402
+ },
3403
+ {
3404
+ "epoch": 1.83,
3405
+ "eval_loss": 0.399748831987381,
3406
+ "eval_runtime": 224.6973,
3407
+ "eval_samples_per_second": 21.286,
3408
+ "eval_steps_per_second": 5.323,
3409
+ "step": 5000
3410
+ },
3411
+ {
3412
+ "epoch": 1.84,
3413
+ "learning_rate": 0.00015726592735120393,
3414
+ "loss": 0.2883,
3415
+ "step": 5010
3416
+ },
3417
+ {
3418
+ "epoch": 1.84,
3419
+ "learning_rate": 0.0001571085326102195,
3420
+ "loss": 0.2639,
3421
+ "step": 5020
3422
+ },
3423
+ {
3424
+ "epoch": 1.84,
3425
+ "learning_rate": 0.0001569509276410015,
3426
+ "loss": 0.2624,
3427
+ "step": 5030
3428
+ },
3429
+ {
3430
+ "epoch": 1.85,
3431
+ "learning_rate": 0.00015679311302372614,
3432
+ "loss": 0.2823,
3433
+ "step": 5040
3434
+ },
3435
+ {
3436
+ "epoch": 1.85,
3437
+ "learning_rate": 0.0001566350893393414,
3438
+ "loss": 0.3026,
3439
+ "step": 5050
3440
+ },
3441
+ {
3442
+ "epoch": 1.85,
3443
+ "learning_rate": 0.00015647685716956494,
3444
+ "loss": 0.2803,
3445
+ "step": 5060
3446
+ },
3447
+ {
3448
+ "epoch": 1.86,
3449
+ "learning_rate": 0.00015631841709688184,
3450
+ "loss": 0.2991,
3451
+ "step": 5070
3452
+ },
3453
+ {
3454
+ "epoch": 1.86,
3455
+ "learning_rate": 0.00015615976970454257,
3456
+ "loss": 0.2734,
3457
+ "step": 5080
3458
+ },
3459
+ {
3460
+ "epoch": 1.86,
3461
+ "learning_rate": 0.00015600091557656072,
3462
+ "loss": 0.2674,
3463
+ "step": 5090
3464
+ },
3465
+ {
3466
+ "epoch": 1.87,
3467
+ "learning_rate": 0.000155841855297711,
3468
+ "loss": 0.2736,
3469
+ "step": 5100
3470
+ },
3471
+ {
3472
+ "epoch": 1.87,
3473
+ "eval_loss": 0.40985968708992004,
3474
+ "eval_runtime": 224.6419,
3475
+ "eval_samples_per_second": 21.292,
3476
+ "eval_steps_per_second": 5.324,
3477
+ "step": 5100
3478
+ },
3479
+ {
3480
+ "epoch": 1.87,
3481
+ "learning_rate": 0.0001556825894535269,
3482
+ "loss": 0.2828,
3483
+ "step": 5110
3484
+ },
3485
+ {
3486
+ "epoch": 1.88,
3487
+ "learning_rate": 0.00015552311863029875,
3488
+ "loss": 0.2872,
3489
+ "step": 5120
3490
+ },
3491
+ {
3492
+ "epoch": 1.88,
3493
+ "learning_rate": 0.00015536344341507129,
3494
+ "loss": 0.2669,
3495
+ "step": 5130
3496
+ },
3497
+ {
3498
+ "epoch": 1.88,
3499
+ "learning_rate": 0.0001552035643956419,
3500
+ "loss": 0.2708,
3501
+ "step": 5140
3502
+ },
3503
+ {
3504
+ "epoch": 1.89,
3505
+ "learning_rate": 0.00015504348216055798,
3506
+ "loss": 0.3048,
3507
+ "step": 5150
3508
+ },
3509
+ {
3510
+ "epoch": 1.89,
3511
+ "learning_rate": 0.00015488319729911512,
3512
+ "loss": 0.2964,
3513
+ "step": 5160
3514
+ },
3515
+ {
3516
+ "epoch": 1.89,
3517
+ "learning_rate": 0.00015472271040135483,
3518
+ "loss": 0.266,
3519
+ "step": 5170
3520
+ },
3521
+ {
3522
+ "epoch": 1.9,
3523
+ "learning_rate": 0.00015456202205806234,
3524
+ "loss": 0.2885,
3525
+ "step": 5180
3526
+ },
3527
+ {
3528
+ "epoch": 1.9,
3529
+ "learning_rate": 0.0001544011328607644,
3530
+ "loss": 0.3317,
3531
+ "step": 5190
3532
+ },
3533
+ {
3534
+ "epoch": 1.91,
3535
+ "learning_rate": 0.00015424004340172719,
3536
+ "loss": 0.2905,
3537
+ "step": 5200
3538
+ },
3539
+ {
3540
+ "epoch": 1.91,
3541
+ "eval_loss": 0.39825478196144104,
3542
+ "eval_runtime": 224.8976,
3543
+ "eval_samples_per_second": 21.267,
3544
+ "eval_steps_per_second": 5.318,
3545
+ "step": 5200
3546
+ },
3547
+ {
3548
+ "epoch": 1.91,
3549
+ "learning_rate": 0.0001540787542739541,
3550
+ "loss": 0.2663,
3551
+ "step": 5210
3552
+ },
3553
+ {
3554
+ "epoch": 1.91,
3555
+ "learning_rate": 0.00015391726607118345,
3556
+ "loss": 0.2838,
3557
+ "step": 5220
3558
+ },
3559
+ {
3560
+ "epoch": 1.92,
3561
+ "learning_rate": 0.00015375557938788657,
3562
+ "loss": 0.2962,
3563
+ "step": 5230
3564
+ },
3565
+ {
3566
+ "epoch": 1.92,
3567
+ "learning_rate": 0.0001535936948192653,
3568
+ "loss": 0.2884,
3569
+ "step": 5240
3570
+ },
3571
+ {
3572
+ "epoch": 1.92,
3573
+ "learning_rate": 0.00015343161296124994,
3574
+ "loss": 0.3026,
3575
+ "step": 5250
3576
+ },
3577
+ {
3578
+ "epoch": 1.93,
3579
+ "learning_rate": 0.00015326933441049714,
3580
+ "loss": 0.3234,
3581
+ "step": 5260
3582
+ },
3583
+ {
3584
+ "epoch": 1.93,
3585
+ "learning_rate": 0.00015310685976438753,
3586
+ "loss": 0.2547,
3587
+ "step": 5270
3588
+ },
3589
+ {
3590
+ "epoch": 1.93,
3591
+ "learning_rate": 0.00015294418962102363,
3592
+ "loss": 0.2896,
3593
+ "step": 5280
3594
+ },
3595
+ {
3596
+ "epoch": 1.94,
3597
+ "learning_rate": 0.00015278132457922764,
3598
+ "loss": 0.3173,
3599
+ "step": 5290
3600
+ },
3601
+ {
3602
+ "epoch": 1.94,
3603
+ "learning_rate": 0.00015261826523853926,
3604
+ "loss": 0.2848,
3605
+ "step": 5300
3606
+ },
3607
+ {
3608
+ "epoch": 1.94,
3609
+ "eval_loss": 0.4000326693058014,
3610
+ "eval_runtime": 224.8241,
3611
+ "eval_samples_per_second": 21.274,
3612
+ "eval_steps_per_second": 5.32,
3613
+ "step": 5300
3614
+ },
3615
+ {
3616
+ "epoch": 1.95,
3617
+ "learning_rate": 0.00015245501219921336,
3618
+ "loss": 0.2819,
3619
+ "step": 5310
3620
+ },
3621
+ {
3622
+ "epoch": 1.95,
3623
+ "learning_rate": 0.00015229156606221792,
3624
+ "loss": 0.2578,
3625
+ "step": 5320
3626
+ },
3627
+ {
3628
+ "epoch": 1.95,
3629
+ "learning_rate": 0.0001521279274292317,
3630
+ "loss": 0.2893,
3631
+ "step": 5330
3632
+ },
3633
+ {
3634
+ "epoch": 1.96,
3635
+ "learning_rate": 0.00015196409690264212,
3636
+ "loss": 0.2897,
3637
+ "step": 5340
3638
+ },
3639
+ {
3640
+ "epoch": 1.96,
3641
+ "learning_rate": 0.000151800075085543,
3642
+ "loss": 0.2718,
3643
+ "step": 5350
3644
+ },
3645
+ {
3646
+ "epoch": 1.96,
3647
+ "learning_rate": 0.0001516358625817323,
3648
+ "loss": 0.2908,
3649
+ "step": 5360
3650
+ },
3651
+ {
3652
+ "epoch": 1.97,
3653
+ "learning_rate": 0.00015147145999570998,
3654
+ "loss": 0.2694,
3655
+ "step": 5370
3656
+ },
3657
+ {
3658
+ "epoch": 1.97,
3659
+ "learning_rate": 0.0001513068679326757,
3660
+ "loss": 0.3028,
3661
+ "step": 5380
3662
+ },
3663
+ {
3664
+ "epoch": 1.97,
3665
+ "learning_rate": 0.00015114208699852663,
3666
+ "loss": 0.2707,
3667
+ "step": 5390
3668
+ }
3669
+ ],
3670
+ "logging_steps": 10,
3671
+ "max_steps": 16374,
3672
+ "num_input_tokens_seen": 0,
3673
+ "num_train_epochs": 6,
3674
+ "save_steps": 10,
3675
+ "total_flos": 1.756931929751421e+18,
3676
+ "train_batch_size": 4,
3677
+ "trial_name": null,
3678
+ "trial_params": null
3679
+ }
subject/checkpoint-5390/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1088a5ed93850145a58e0ae31a18a24ddce00e6340922c12cadad4f0f05c20aa
3
+ size 4856
subject/eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.98,
3
+ "eval_loss": 0.41133925318717957,
4
+ "eval_runtime": 345.8577,
5
+ "eval_samples_per_second": 13.829,
6
+ "eval_steps_per_second": 1.729
7
+ }
subject/runs/Jan26_17-21-12_platform/events.out.tfevents.1706289812.platform ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f4ef4c71b4db32103cb955de4270de38a019e97856e787857cb8aded9183080
3
+ size 104045
subject/runs/Jan29_08-21-25_platform/events.out.tfevents.1706516497.platform ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b35390fac49ca50db8bf4c1f53ee7ee3c6d8e9cbc1ce1341a837a3ce326c6cb
3
+ size 5332
subject/runs/Jan29_08-21-25_platform/events.out.tfevents.1706516849.platform ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f3495ec1898df0c84f5c93a2f30271324f26757c56d9bb6d53fd89b0dbb5d6f
3
+ size 311
subject/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": false,
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
+ }
subject/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
subject/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": 4096,
35
+ "pad_token": "<unk>",
36
+ "padding_side": "right",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "split_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
subject/train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.98,
3
+ "train_loss": 4.6051067095080594e-05,
4
+ "train_runtime": 5.408,
5
+ "train_samples_per_second": 15680.552,
6
+ "train_steps_per_second": 980.035
7
+ }
subject/trainer_log.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
subject/trainer_state.json ADDED
@@ -0,0 +1,3688 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.9752701960065946,
5
+ "eval_steps": 100,
6
+ "global_step": 5391,
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": 0.0001999998159397344,
14
+ "loss": 1.4113,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 0.00019999926375961516,
20
+ "loss": 0.8357,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 0.00019999834346167496,
26
+ "loss": 0.7122,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.01,
31
+ "learning_rate": 0.0001999970550493016,
32
+ "loss": 0.6653,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.02,
37
+ "learning_rate": 0.000199995398527238,
38
+ "loss": 0.6152,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.02,
43
+ "learning_rate": 0.00019999337390158218,
44
+ "loss": 0.5963,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.03,
49
+ "learning_rate": 0.00019999098117978715,
50
+ "loss": 0.5923,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.03,
55
+ "learning_rate": 0.00019998822037066105,
56
+ "loss": 0.5783,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.03,
61
+ "learning_rate": 0.00019998509148436697,
62
+ "loss": 0.5529,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.04,
67
+ "learning_rate": 0.000199981594532423,
68
+ "loss": 0.5382,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.04,
73
+ "eval_loss": 0.5621132850646973,
74
+ "eval_runtime": 398.8595,
75
+ "eval_samples_per_second": 11.992,
76
+ "eval_steps_per_second": 2.999,
77
+ "step": 100
78
+ },
79
+ {
80
+ "epoch": 0.04,
81
+ "learning_rate": 0.00019997772952770216,
82
+ "loss": 0.5489,
83
+ "step": 110
84
+ },
85
+ {
86
+ "epoch": 0.04,
87
+ "learning_rate": 0.00019997349648443225,
88
+ "loss": 0.5268,
89
+ "step": 120
90
+ },
91
+ {
92
+ "epoch": 0.05,
93
+ "learning_rate": 0.00019996889541819602,
94
+ "loss": 0.525,
95
+ "step": 130
96
+ },
97
+ {
98
+ "epoch": 0.05,
99
+ "learning_rate": 0.00019996392634593092,
100
+ "loss": 0.4926,
101
+ "step": 140
102
+ },
103
+ {
104
+ "epoch": 0.05,
105
+ "learning_rate": 0.00019995858928592916,
106
+ "loss": 0.5181,
107
+ "step": 150
108
+ },
109
+ {
110
+ "epoch": 0.06,
111
+ "learning_rate": 0.00019995288425783754,
112
+ "loss": 0.5225,
113
+ "step": 160
114
+ },
115
+ {
116
+ "epoch": 0.06,
117
+ "learning_rate": 0.00019994681128265743,
118
+ "loss": 0.4993,
119
+ "step": 170
120
+ },
121
+ {
122
+ "epoch": 0.07,
123
+ "learning_rate": 0.00019994037038274467,
124
+ "loss": 0.5387,
125
+ "step": 180
126
+ },
127
+ {
128
+ "epoch": 0.07,
129
+ "learning_rate": 0.0001999335615818096,
130
+ "loss": 0.465,
131
+ "step": 190
132
+ },
133
+ {
134
+ "epoch": 0.07,
135
+ "learning_rate": 0.00019992638490491676,
136
+ "loss": 0.4788,
137
+ "step": 200
138
+ },
139
+ {
140
+ "epoch": 0.07,
141
+ "eval_loss": 0.5201168060302734,
142
+ "eval_runtime": 398.3332,
143
+ "eval_samples_per_second": 12.008,
144
+ "eval_steps_per_second": 3.003,
145
+ "step": 200
146
+ },
147
+ {
148
+ "epoch": 0.08,
149
+ "learning_rate": 0.00019991884037848497,
150
+ "loss": 0.451,
151
+ "step": 210
152
+ },
153
+ {
154
+ "epoch": 0.08,
155
+ "learning_rate": 0.00019991092803028725,
156
+ "loss": 0.4586,
157
+ "step": 220
158
+ },
159
+ {
160
+ "epoch": 0.08,
161
+ "learning_rate": 0.00019990264788945052,
162
+ "loss": 0.4894,
163
+ "step": 230
164
+ },
165
+ {
166
+ "epoch": 0.09,
167
+ "learning_rate": 0.00019989399998645568,
168
+ "loss": 0.4666,
169
+ "step": 240
170
+ },
171
+ {
172
+ "epoch": 0.09,
173
+ "learning_rate": 0.00019988498435313744,
174
+ "loss": 0.4763,
175
+ "step": 250
176
+ },
177
+ {
178
+ "epoch": 0.1,
179
+ "learning_rate": 0.00019987560102268422,
180
+ "loss": 0.4539,
181
+ "step": 260
182
+ },
183
+ {
184
+ "epoch": 0.1,
185
+ "learning_rate": 0.00019986585002963793,
186
+ "loss": 0.5133,
187
+ "step": 270
188
+ },
189
+ {
190
+ "epoch": 0.1,
191
+ "learning_rate": 0.00019985573140989405,
192
+ "loss": 0.4704,
193
+ "step": 280
194
+ },
195
+ {
196
+ "epoch": 0.11,
197
+ "learning_rate": 0.00019984524520070125,
198
+ "loss": 0.4548,
199
+ "step": 290
200
+ },
201
+ {
202
+ "epoch": 0.11,
203
+ "learning_rate": 0.00019983439144066143,
204
+ "loss": 0.4682,
205
+ "step": 300
206
+ },
207
+ {
208
+ "epoch": 0.11,
209
+ "eval_loss": 0.497054785490036,
210
+ "eval_runtime": 399.4057,
211
+ "eval_samples_per_second": 11.975,
212
+ "eval_steps_per_second": 2.994,
213
+ "step": 300
214
+ },
215
+ {
216
+ "epoch": 0.11,
217
+ "learning_rate": 0.0001998231701697295,
218
+ "loss": 0.4867,
219
+ "step": 310
220
+ },
221
+ {
222
+ "epoch": 0.12,
223
+ "learning_rate": 0.0001998115814292133,
224
+ "loss": 0.4486,
225
+ "step": 320
226
+ },
227
+ {
228
+ "epoch": 0.12,
229
+ "learning_rate": 0.0001997996252617733,
230
+ "loss": 0.4472,
231
+ "step": 330
232
+ },
233
+ {
234
+ "epoch": 0.12,
235
+ "learning_rate": 0.00019978730171142268,
236
+ "loss": 0.4821,
237
+ "step": 340
238
+ },
239
+ {
240
+ "epoch": 0.13,
241
+ "learning_rate": 0.0001997746108235269,
242
+ "loss": 0.4475,
243
+ "step": 350
244
+ },
245
+ {
246
+ "epoch": 0.13,
247
+ "learning_rate": 0.00019976155264480377,
248
+ "loss": 0.4485,
249
+ "step": 360
250
+ },
251
+ {
252
+ "epoch": 0.14,
253
+ "learning_rate": 0.00019974812722332308,
254
+ "loss": 0.4547,
255
+ "step": 370
256
+ },
257
+ {
258
+ "epoch": 0.14,
259
+ "learning_rate": 0.0001997343346085066,
260
+ "loss": 0.4279,
261
+ "step": 380
262
+ },
263
+ {
264
+ "epoch": 0.14,
265
+ "learning_rate": 0.00019972017485112774,
266
+ "loss": 0.4748,
267
+ "step": 390
268
+ },
269
+ {
270
+ "epoch": 0.15,
271
+ "learning_rate": 0.0001997056480033115,
272
+ "loss": 0.4778,
273
+ "step": 400
274
+ },
275
+ {
276
+ "epoch": 0.15,
277
+ "eval_loss": 0.4730743169784546,
278
+ "eval_runtime": 403.2392,
279
+ "eval_samples_per_second": 11.861,
280
+ "eval_steps_per_second": 2.966,
281
+ "step": 400
282
+ },
283
+ {
284
+ "epoch": 0.15,
285
+ "learning_rate": 0.0001996907541185342,
286
+ "loss": 0.4393,
287
+ "step": 410
288
+ },
289
+ {
290
+ "epoch": 0.15,
291
+ "learning_rate": 0.00019967549325162324,
292
+ "loss": 0.4562,
293
+ "step": 420
294
+ },
295
+ {
296
+ "epoch": 0.16,
297
+ "learning_rate": 0.00019965986545875708,
298
+ "loss": 0.4389,
299
+ "step": 430
300
+ },
301
+ {
302
+ "epoch": 0.16,
303
+ "learning_rate": 0.0001996438707974648,
304
+ "loss": 0.4282,
305
+ "step": 440
306
+ },
307
+ {
308
+ "epoch": 0.16,
309
+ "learning_rate": 0.000199627509326626,
310
+ "loss": 0.4254,
311
+ "step": 450
312
+ },
313
+ {
314
+ "epoch": 0.17,
315
+ "learning_rate": 0.0001996107811064706,
316
+ "loss": 0.419,
317
+ "step": 460
318
+ },
319
+ {
320
+ "epoch": 0.17,
321
+ "learning_rate": 0.00019959368619857872,
322
+ "loss": 0.4261,
323
+ "step": 470
324
+ },
325
+ {
326
+ "epoch": 0.18,
327
+ "learning_rate": 0.0001995762246658801,
328
+ "loss": 0.4581,
329
+ "step": 480
330
+ },
331
+ {
332
+ "epoch": 0.18,
333
+ "learning_rate": 0.00019955839657265432,
334
+ "loss": 0.4333,
335
+ "step": 490
336
+ },
337
+ {
338
+ "epoch": 0.18,
339
+ "learning_rate": 0.00019954020198453018,
340
+ "loss": 0.4541,
341
+ "step": 500
342
+ },
343
+ {
344
+ "epoch": 0.18,
345
+ "eval_loss": 0.46874529123306274,
346
+ "eval_runtime": 403.7251,
347
+ "eval_samples_per_second": 11.847,
348
+ "eval_steps_per_second": 2.962,
349
+ "step": 500
350
+ },
351
+ {
352
+ "epoch": 0.19,
353
+ "learning_rate": 0.00019952164096848578,
354
+ "loss": 0.4482,
355
+ "step": 510
356
+ },
357
+ {
358
+ "epoch": 0.19,
359
+ "learning_rate": 0.00019950271359284795,
360
+ "loss": 0.4475,
361
+ "step": 520
362
+ },
363
+ {
364
+ "epoch": 0.19,
365
+ "learning_rate": 0.00019948341992729227,
366
+ "loss": 0.4339,
367
+ "step": 530
368
+ },
369
+ {
370
+ "epoch": 0.2,
371
+ "learning_rate": 0.00019946376004284272,
372
+ "loss": 0.4527,
373
+ "step": 540
374
+ },
375
+ {
376
+ "epoch": 0.2,
377
+ "learning_rate": 0.0001994437340118713,
378
+ "loss": 0.4009,
379
+ "step": 550
380
+ },
381
+ {
382
+ "epoch": 0.21,
383
+ "learning_rate": 0.000199423341908098,
384
+ "loss": 0.4496,
385
+ "step": 560
386
+ },
387
+ {
388
+ "epoch": 0.21,
389
+ "learning_rate": 0.0001994025838065903,
390
+ "loss": 0.4314,
391
+ "step": 570
392
+ },
393
+ {
394
+ "epoch": 0.21,
395
+ "learning_rate": 0.0001993814597837631,
396
+ "loss": 0.4454,
397
+ "step": 580
398
+ },
399
+ {
400
+ "epoch": 0.22,
401
+ "learning_rate": 0.00019935996991737818,
402
+ "loss": 0.4076,
403
+ "step": 590
404
+ },
405
+ {
406
+ "epoch": 0.22,
407
+ "learning_rate": 0.0001993381142865442,
408
+ "loss": 0.4786,
409
+ "step": 600
410
+ },
411
+ {
412
+ "epoch": 0.22,
413
+ "eval_loss": 0.4500006139278412,
414
+ "eval_runtime": 399.7496,
415
+ "eval_samples_per_second": 11.965,
416
+ "eval_steps_per_second": 2.992,
417
+ "step": 600
418
+ },
419
+ {
420
+ "epoch": 0.22,
421
+ "learning_rate": 0.00019931589297171628,
422
+ "loss": 0.4386,
423
+ "step": 610
424
+ },
425
+ {
426
+ "epoch": 0.23,
427
+ "learning_rate": 0.0001992933060546955,
428
+ "loss": 0.4805,
429
+ "step": 620
430
+ },
431
+ {
432
+ "epoch": 0.23,
433
+ "learning_rate": 0.00019927035361862904,
434
+ "loss": 0.4135,
435
+ "step": 630
436
+ },
437
+ {
438
+ "epoch": 0.23,
439
+ "learning_rate": 0.0001992470357480095,
440
+ "loss": 0.421,
441
+ "step": 640
442
+ },
443
+ {
444
+ "epoch": 0.24,
445
+ "learning_rate": 0.00019922335252867476,
446
+ "loss": 0.4224,
447
+ "step": 650
448
+ },
449
+ {
450
+ "epoch": 0.24,
451
+ "learning_rate": 0.00019919930404780766,
452
+ "loss": 0.4324,
453
+ "step": 660
454
+ },
455
+ {
456
+ "epoch": 0.25,
457
+ "learning_rate": 0.0001991748903939355,
458
+ "loss": 0.4062,
459
+ "step": 670
460
+ },
461
+ {
462
+ "epoch": 0.25,
463
+ "learning_rate": 0.00019915011165692997,
464
+ "loss": 0.4107,
465
+ "step": 680
466
+ },
467
+ {
468
+ "epoch": 0.25,
469
+ "learning_rate": 0.00019912496792800677,
470
+ "loss": 0.3953,
471
+ "step": 690
472
+ },
473
+ {
474
+ "epoch": 0.26,
475
+ "learning_rate": 0.00019909945929972502,
476
+ "loss": 0.3974,
477
+ "step": 700
478
+ },
479
+ {
480
+ "epoch": 0.26,
481
+ "eval_loss": 0.4456511437892914,
482
+ "eval_runtime": 403.5923,
483
+ "eval_samples_per_second": 11.851,
484
+ "eval_steps_per_second": 2.963,
485
+ "step": 700
486
+ },
487
+ {
488
+ "epoch": 0.26,
489
+ "learning_rate": 0.0001990735858659873,
490
+ "loss": 0.4006,
491
+ "step": 710
492
+ },
493
+ {
494
+ "epoch": 0.26,
495
+ "learning_rate": 0.000199047347722039,
496
+ "loss": 0.4258,
497
+ "step": 720
498
+ },
499
+ {
500
+ "epoch": 0.27,
501
+ "learning_rate": 0.00019902074496446815,
502
+ "loss": 0.4083,
503
+ "step": 730
504
+ },
505
+ {
506
+ "epoch": 0.27,
507
+ "learning_rate": 0.00019899377769120487,
508
+ "loss": 0.4097,
509
+ "step": 740
510
+ },
511
+ {
512
+ "epoch": 0.27,
513
+ "learning_rate": 0.00019896644600152135,
514
+ "loss": 0.4262,
515
+ "step": 750
516
+ },
517
+ {
518
+ "epoch": 0.28,
519
+ "learning_rate": 0.00019893874999603103,
520
+ "loss": 0.4204,
521
+ "step": 760
522
+ },
523
+ {
524
+ "epoch": 0.28,
525
+ "learning_rate": 0.0001989106897766887,
526
+ "loss": 0.4233,
527
+ "step": 770
528
+ },
529
+ {
530
+ "epoch": 0.29,
531
+ "learning_rate": 0.0001988822654467897,
532
+ "loss": 0.4107,
533
+ "step": 780
534
+ },
535
+ {
536
+ "epoch": 0.29,
537
+ "learning_rate": 0.00019885347711096993,
538
+ "loss": 0.4013,
539
+ "step": 790
540
+ },
541
+ {
542
+ "epoch": 0.29,
543
+ "learning_rate": 0.00019882432487520506,
544
+ "loss": 0.4142,
545
+ "step": 800
546
+ },
547
+ {
548
+ "epoch": 0.29,
549
+ "eval_loss": 0.4460389018058777,
550
+ "eval_runtime": 400.2256,
551
+ "eval_samples_per_second": 11.951,
552
+ "eval_steps_per_second": 2.988,
553
+ "step": 800
554
+ },
555
+ {
556
+ "epoch": 0.3,
557
+ "learning_rate": 0.0001987948088468105,
558
+ "loss": 0.4324,
559
+ "step": 810
560
+ },
561
+ {
562
+ "epoch": 0.3,
563
+ "learning_rate": 0.0001987649291344408,
564
+ "loss": 0.3975,
565
+ "step": 820
566
+ },
567
+ {
568
+ "epoch": 0.3,
569
+ "learning_rate": 0.00019873468584808934,
570
+ "loss": 0.4109,
571
+ "step": 830
572
+ },
573
+ {
574
+ "epoch": 0.31,
575
+ "learning_rate": 0.00019870407909908786,
576
+ "loss": 0.3987,
577
+ "step": 840
578
+ },
579
+ {
580
+ "epoch": 0.31,
581
+ "learning_rate": 0.00019867310900010605,
582
+ "loss": 0.3885,
583
+ "step": 850
584
+ },
585
+ {
586
+ "epoch": 0.32,
587
+ "learning_rate": 0.00019864177566515122,
588
+ "loss": 0.4122,
589
+ "step": 860
590
+ },
591
+ {
592
+ "epoch": 0.32,
593
+ "learning_rate": 0.00019861007920956786,
594
+ "loss": 0.405,
595
+ "step": 870
596
+ },
597
+ {
598
+ "epoch": 0.32,
599
+ "learning_rate": 0.00019857801975003704,
600
+ "loss": 0.404,
601
+ "step": 880
602
+ },
603
+ {
604
+ "epoch": 0.33,
605
+ "learning_rate": 0.0001985455974045763,
606
+ "loss": 0.382,
607
+ "step": 890
608
+ },
609
+ {
610
+ "epoch": 0.33,
611
+ "learning_rate": 0.0001985128122925389,
612
+ "loss": 0.4374,
613
+ "step": 900
614
+ },
615
+ {
616
+ "epoch": 0.33,
617
+ "eval_loss": 0.44203105568885803,
618
+ "eval_runtime": 401.2967,
619
+ "eval_samples_per_second": 11.919,
620
+ "eval_steps_per_second": 2.98,
621
+ "step": 900
622
+ },
623
+ {
624
+ "epoch": 0.33,
625
+ "learning_rate": 0.00019847966453461358,
626
+ "loss": 0.3859,
627
+ "step": 910
628
+ },
629
+ {
630
+ "epoch": 0.34,
631
+ "learning_rate": 0.00019844615425282405,
632
+ "loss": 0.4187,
633
+ "step": 920
634
+ },
635
+ {
636
+ "epoch": 0.34,
637
+ "learning_rate": 0.00019841228157052853,
638
+ "loss": 0.4078,
639
+ "step": 930
640
+ },
641
+ {
642
+ "epoch": 0.34,
643
+ "learning_rate": 0.0001983780466124193,
644
+ "loss": 0.4218,
645
+ "step": 940
646
+ },
647
+ {
648
+ "epoch": 0.35,
649
+ "learning_rate": 0.0001983434495045223,
650
+ "loss": 0.4109,
651
+ "step": 950
652
+ },
653
+ {
654
+ "epoch": 0.35,
655
+ "learning_rate": 0.00019830849037419656,
656
+ "loss": 0.4249,
657
+ "step": 960
658
+ },
659
+ {
660
+ "epoch": 0.36,
661
+ "learning_rate": 0.00019827316935013388,
662
+ "loss": 0.4073,
663
+ "step": 970
664
+ },
665
+ {
666
+ "epoch": 0.36,
667
+ "learning_rate": 0.0001982374865623581,
668
+ "loss": 0.4336,
669
+ "step": 980
670
+ },
671
+ {
672
+ "epoch": 0.36,
673
+ "learning_rate": 0.00019820144214222497,
674
+ "loss": 0.4056,
675
+ "step": 990
676
+ },
677
+ {
678
+ "epoch": 0.37,
679
+ "learning_rate": 0.00019816503622242137,
680
+ "loss": 0.4008,
681
+ "step": 1000
682
+ },
683
+ {
684
+ "epoch": 0.37,
685
+ "eval_loss": 0.4418930411338806,
686
+ "eval_runtime": 404.9805,
687
+ "eval_samples_per_second": 11.81,
688
+ "eval_steps_per_second": 2.953,
689
+ "step": 1000
690
+ },
691
+ {
692
+ "epoch": 0.37,
693
+ "learning_rate": 0.00019812826893696495,
694
+ "loss": 0.365,
695
+ "step": 1010
696
+ },
697
+ {
698
+ "epoch": 0.37,
699
+ "learning_rate": 0.00019809114042120367,
700
+ "loss": 0.4006,
701
+ "step": 1020
702
+ },
703
+ {
704
+ "epoch": 0.38,
705
+ "learning_rate": 0.0001980536508118152,
706
+ "loss": 0.3714,
707
+ "step": 1030
708
+ },
709
+ {
710
+ "epoch": 0.38,
711
+ "learning_rate": 0.00019801580024680652,
712
+ "loss": 0.3945,
713
+ "step": 1040
714
+ },
715
+ {
716
+ "epoch": 0.38,
717
+ "learning_rate": 0.00019797758886551324,
718
+ "loss": 0.4316,
719
+ "step": 1050
720
+ },
721
+ {
722
+ "epoch": 0.39,
723
+ "learning_rate": 0.0001979390168085994,
724
+ "loss": 0.3958,
725
+ "step": 1060
726
+ },
727
+ {
728
+ "epoch": 0.39,
729
+ "learning_rate": 0.00019790008421805664,
730
+ "loss": 0.419,
731
+ "step": 1070
732
+ },
733
+ {
734
+ "epoch": 0.4,
735
+ "learning_rate": 0.00019786079123720377,
736
+ "loss": 0.4048,
737
+ "step": 1080
738
+ },
739
+ {
740
+ "epoch": 0.4,
741
+ "learning_rate": 0.0001978211380106864,
742
+ "loss": 0.4002,
743
+ "step": 1090
744
+ },
745
+ {
746
+ "epoch": 0.4,
747
+ "learning_rate": 0.0001977811246844761,
748
+ "loss": 0.3979,
749
+ "step": 1100
750
+ },
751
+ {
752
+ "epoch": 0.4,
753
+ "eval_loss": 0.43328720331192017,
754
+ "eval_runtime": 398.734,
755
+ "eval_samples_per_second": 11.995,
756
+ "eval_steps_per_second": 2.999,
757
+ "step": 1100
758
+ },
759
+ {
760
+ "epoch": 0.41,
761
+ "learning_rate": 0.00019774075140587024,
762
+ "loss": 0.4047,
763
+ "step": 1110
764
+ },
765
+ {
766
+ "epoch": 0.41,
767
+ "learning_rate": 0.00019770001832349106,
768
+ "loss": 0.389,
769
+ "step": 1120
770
+ },
771
+ {
772
+ "epoch": 0.41,
773
+ "learning_rate": 0.00019765892558728542,
774
+ "loss": 0.3628,
775
+ "step": 1130
776
+ },
777
+ {
778
+ "epoch": 0.42,
779
+ "learning_rate": 0.0001976174733485242,
780
+ "loss": 0.3914,
781
+ "step": 1140
782
+ },
783
+ {
784
+ "epoch": 0.42,
785
+ "learning_rate": 0.00019757566175980146,
786
+ "loss": 0.3885,
787
+ "step": 1150
788
+ },
789
+ {
790
+ "epoch": 0.42,
791
+ "learning_rate": 0.00019753349097503437,
792
+ "loss": 0.4023,
793
+ "step": 1160
794
+ },
795
+ {
796
+ "epoch": 0.43,
797
+ "learning_rate": 0.0001974909611494622,
798
+ "loss": 0.4358,
799
+ "step": 1170
800
+ },
801
+ {
802
+ "epoch": 0.43,
803
+ "learning_rate": 0.00019744807243964597,
804
+ "loss": 0.3965,
805
+ "step": 1180
806
+ },
807
+ {
808
+ "epoch": 0.44,
809
+ "learning_rate": 0.00019740482500346779,
810
+ "loss": 0.3762,
811
+ "step": 1190
812
+ },
813
+ {
814
+ "epoch": 0.44,
815
+ "learning_rate": 0.0001973612190001304,
816
+ "loss": 0.4108,
817
+ "step": 1200
818
+ },
819
+ {
820
+ "epoch": 0.44,
821
+ "eval_loss": 0.4304308295249939,
822
+ "eval_runtime": 397.6793,
823
+ "eval_samples_per_second": 12.027,
824
+ "eval_steps_per_second": 3.007,
825
+ "step": 1200
826
+ },
827
+ {
828
+ "epoch": 0.44,
829
+ "learning_rate": 0.00019731725459015643,
830
+ "loss": 0.3838,
831
+ "step": 1210
832
+ },
833
+ {
834
+ "epoch": 0.45,
835
+ "learning_rate": 0.00019727293193538793,
836
+ "loss": 0.3782,
837
+ "step": 1220
838
+ },
839
+ {
840
+ "epoch": 0.45,
841
+ "learning_rate": 0.00019722825119898566,
842
+ "loss": 0.4034,
843
+ "step": 1230
844
+ },
845
+ {
846
+ "epoch": 0.45,
847
+ "learning_rate": 0.00019718321254542858,
848
+ "loss": 0.391,
849
+ "step": 1240
850
+ },
851
+ {
852
+ "epoch": 0.46,
853
+ "learning_rate": 0.0001971378161405132,
854
+ "loss": 0.4082,
855
+ "step": 1250
856
+ },
857
+ {
858
+ "epoch": 0.46,
859
+ "learning_rate": 0.0001970920621513531,
860
+ "loss": 0.3772,
861
+ "step": 1260
862
+ },
863
+ {
864
+ "epoch": 0.47,
865
+ "learning_rate": 0.00019704595074637805,
866
+ "loss": 0.3894,
867
+ "step": 1270
868
+ },
869
+ {
870
+ "epoch": 0.47,
871
+ "learning_rate": 0.00019699948209533355,
872
+ "loss": 0.3882,
873
+ "step": 1280
874
+ },
875
+ {
876
+ "epoch": 0.47,
877
+ "learning_rate": 0.00019695265636928032,
878
+ "loss": 0.3782,
879
+ "step": 1290
880
+ },
881
+ {
882
+ "epoch": 0.48,
883
+ "learning_rate": 0.0001969054737405934,
884
+ "loss": 0.3578,
885
+ "step": 1300
886
+ },
887
+ {
888
+ "epoch": 0.48,
889
+ "eval_loss": 0.42550337314605713,
890
+ "eval_runtime": 402.9262,
891
+ "eval_samples_per_second": 11.871,
892
+ "eval_steps_per_second": 2.968,
893
+ "step": 1300
894
+ },
895
+ {
896
+ "epoch": 0.48,
897
+ "learning_rate": 0.00019685793438296183,
898
+ "loss": 0.3644,
899
+ "step": 1310
900
+ },
901
+ {
902
+ "epoch": 0.48,
903
+ "learning_rate": 0.00019681003847138765,
904
+ "loss": 0.3702,
905
+ "step": 1320
906
+ },
907
+ {
908
+ "epoch": 0.49,
909
+ "learning_rate": 0.00019676178618218565,
910
+ "loss": 0.4213,
911
+ "step": 1330
912
+ },
913
+ {
914
+ "epoch": 0.49,
915
+ "learning_rate": 0.0001967131776929823,
916
+ "loss": 0.4126,
917
+ "step": 1340
918
+ },
919
+ {
920
+ "epoch": 0.49,
921
+ "learning_rate": 0.00019666421318271547,
922
+ "loss": 0.3643,
923
+ "step": 1350
924
+ },
925
+ {
926
+ "epoch": 0.5,
927
+ "learning_rate": 0.00019661489283163362,
928
+ "loss": 0.3854,
929
+ "step": 1360
930
+ },
931
+ {
932
+ "epoch": 0.5,
933
+ "learning_rate": 0.00019656521682129502,
934
+ "loss": 0.3933,
935
+ "step": 1370
936
+ },
937
+ {
938
+ "epoch": 0.51,
939
+ "learning_rate": 0.00019651518533456733,
940
+ "loss": 0.3841,
941
+ "step": 1380
942
+ },
943
+ {
944
+ "epoch": 0.51,
945
+ "learning_rate": 0.00019646479855562666,
946
+ "loss": 0.3677,
947
+ "step": 1390
948
+ },
949
+ {
950
+ "epoch": 0.51,
951
+ "learning_rate": 0.00019641405666995715,
952
+ "loss": 0.3895,
953
+ "step": 1400
954
+ },
955
+ {
956
+ "epoch": 0.51,
957
+ "eval_loss": 0.41961902379989624,
958
+ "eval_runtime": 402.2386,
959
+ "eval_samples_per_second": 11.891,
960
+ "eval_steps_per_second": 2.973,
961
+ "step": 1400
962
+ },
963
+ {
964
+ "epoch": 0.52,
965
+ "learning_rate": 0.00019636295986435003,
966
+ "loss": 0.3578,
967
+ "step": 1410
968
+ },
969
+ {
970
+ "epoch": 0.52,
971
+ "learning_rate": 0.00019631150832690318,
972
+ "loss": 0.3878,
973
+ "step": 1420
974
+ },
975
+ {
976
+ "epoch": 0.52,
977
+ "learning_rate": 0.00019625970224702025,
978
+ "loss": 0.3784,
979
+ "step": 1430
980
+ },
981
+ {
982
+ "epoch": 0.53,
983
+ "learning_rate": 0.00019620754181541008,
984
+ "loss": 0.4042,
985
+ "step": 1440
986
+ },
987
+ {
988
+ "epoch": 0.53,
989
+ "learning_rate": 0.0001961550272240859,
990
+ "loss": 0.361,
991
+ "step": 1450
992
+ },
993
+ {
994
+ "epoch": 0.53,
995
+ "learning_rate": 0.00019610215866636477,
996
+ "loss": 0.3877,
997
+ "step": 1460
998
+ },
999
+ {
1000
+ "epoch": 0.54,
1001
+ "learning_rate": 0.00019604893633686662,
1002
+ "loss": 0.3822,
1003
+ "step": 1470
1004
+ },
1005
+ {
1006
+ "epoch": 0.54,
1007
+ "learning_rate": 0.00019599536043151384,
1008
+ "loss": 0.3726,
1009
+ "step": 1480
1010
+ },
1011
+ {
1012
+ "epoch": 0.55,
1013
+ "learning_rate": 0.00019594143114753026,
1014
+ "loss": 0.3552,
1015
+ "step": 1490
1016
+ },
1017
+ {
1018
+ "epoch": 0.55,
1019
+ "learning_rate": 0.00019588714868344073,
1020
+ "loss": 0.3725,
1021
+ "step": 1500
1022
+ },
1023
+ {
1024
+ "epoch": 0.55,
1025
+ "eval_loss": 0.4203263521194458,
1026
+ "eval_runtime": 396.7561,
1027
+ "eval_samples_per_second": 12.055,
1028
+ "eval_steps_per_second": 3.014,
1029
+ "step": 1500
1030
+ },
1031
+ {
1032
+ "epoch": 0.55,
1033
+ "learning_rate": 0.00019583251323907006,
1034
+ "loss": 0.3993,
1035
+ "step": 1510
1036
+ },
1037
+ {
1038
+ "epoch": 0.56,
1039
+ "learning_rate": 0.0001957775250155426,
1040
+ "loss": 0.3557,
1041
+ "step": 1520
1042
+ },
1043
+ {
1044
+ "epoch": 0.56,
1045
+ "learning_rate": 0.0001957221842152813,
1046
+ "loss": 0.4027,
1047
+ "step": 1530
1048
+ },
1049
+ {
1050
+ "epoch": 0.56,
1051
+ "learning_rate": 0.00019566649104200696,
1052
+ "loss": 0.3547,
1053
+ "step": 1540
1054
+ },
1055
+ {
1056
+ "epoch": 0.57,
1057
+ "learning_rate": 0.00019561044570073763,
1058
+ "loss": 0.3978,
1059
+ "step": 1550
1060
+ },
1061
+ {
1062
+ "epoch": 0.57,
1063
+ "learning_rate": 0.00019555404839778767,
1064
+ "loss": 0.3919,
1065
+ "step": 1560
1066
+ },
1067
+ {
1068
+ "epoch": 0.58,
1069
+ "learning_rate": 0.00019549729934076717,
1070
+ "loss": 0.3713,
1071
+ "step": 1570
1072
+ },
1073
+ {
1074
+ "epoch": 0.58,
1075
+ "learning_rate": 0.00019544019873858102,
1076
+ "loss": 0.408,
1077
+ "step": 1580
1078
+ },
1079
+ {
1080
+ "epoch": 0.58,
1081
+ "learning_rate": 0.00019538274680142834,
1082
+ "loss": 0.3792,
1083
+ "step": 1590
1084
+ },
1085
+ {
1086
+ "epoch": 0.59,
1087
+ "learning_rate": 0.00019532494374080144,
1088
+ "loss": 0.3836,
1089
+ "step": 1600
1090
+ },
1091
+ {
1092
+ "epoch": 0.59,
1093
+ "eval_loss": 0.42036038637161255,
1094
+ "eval_runtime": 402.4927,
1095
+ "eval_samples_per_second": 11.883,
1096
+ "eval_steps_per_second": 2.971,
1097
+ "step": 1600
1098
+ },
1099
+ {
1100
+ "epoch": 0.59,
1101
+ "learning_rate": 0.00019526678976948525,
1102
+ "loss": 0.3411,
1103
+ "step": 1610
1104
+ },
1105
+ {
1106
+ "epoch": 0.59,
1107
+ "learning_rate": 0.0001952082851015565,
1108
+ "loss": 0.3863,
1109
+ "step": 1620
1110
+ },
1111
+ {
1112
+ "epoch": 0.6,
1113
+ "learning_rate": 0.00019514942995238287,
1114
+ "loss": 0.375,
1115
+ "step": 1630
1116
+ },
1117
+ {
1118
+ "epoch": 0.6,
1119
+ "learning_rate": 0.00019509022453862226,
1120
+ "loss": 0.3747,
1121
+ "step": 1640
1122
+ },
1123
+ {
1124
+ "epoch": 0.6,
1125
+ "learning_rate": 0.00019503066907822198,
1126
+ "loss": 0.36,
1127
+ "step": 1650
1128
+ },
1129
+ {
1130
+ "epoch": 0.61,
1131
+ "learning_rate": 0.00019497076379041786,
1132
+ "loss": 0.3919,
1133
+ "step": 1660
1134
+ },
1135
+ {
1136
+ "epoch": 0.61,
1137
+ "learning_rate": 0.00019491050889573357,
1138
+ "loss": 0.3699,
1139
+ "step": 1670
1140
+ },
1141
+ {
1142
+ "epoch": 0.62,
1143
+ "learning_rate": 0.00019484990461597978,
1144
+ "loss": 0.3829,
1145
+ "step": 1680
1146
+ },
1147
+ {
1148
+ "epoch": 0.62,
1149
+ "learning_rate": 0.00019478895117425323,
1150
+ "loss": 0.3589,
1151
+ "step": 1690
1152
+ },
1153
+ {
1154
+ "epoch": 0.62,
1155
+ "learning_rate": 0.00019472764879493616,
1156
+ "loss": 0.3784,
1157
+ "step": 1700
1158
+ },
1159
+ {
1160
+ "epoch": 0.62,
1161
+ "eval_loss": 0.4183010458946228,
1162
+ "eval_runtime": 404.1371,
1163
+ "eval_samples_per_second": 11.835,
1164
+ "eval_steps_per_second": 2.959,
1165
+ "step": 1700
1166
+ },
1167
+ {
1168
+ "epoch": 0.63,
1169
+ "learning_rate": 0.00019466599770369509,
1170
+ "loss": 0.3912,
1171
+ "step": 1710
1172
+ },
1173
+ {
1174
+ "epoch": 0.63,
1175
+ "learning_rate": 0.00019460399812748041,
1176
+ "loss": 0.374,
1177
+ "step": 1720
1178
+ },
1179
+ {
1180
+ "epoch": 0.63,
1181
+ "learning_rate": 0.0001945416502945253,
1182
+ "loss": 0.384,
1183
+ "step": 1730
1184
+ },
1185
+ {
1186
+ "epoch": 0.64,
1187
+ "learning_rate": 0.00019447895443434486,
1188
+ "loss": 0.3519,
1189
+ "step": 1740
1190
+ },
1191
+ {
1192
+ "epoch": 0.64,
1193
+ "learning_rate": 0.00019441591077773554,
1194
+ "loss": 0.3598,
1195
+ "step": 1750
1196
+ },
1197
+ {
1198
+ "epoch": 0.64,
1199
+ "learning_rate": 0.0001943525195567739,
1200
+ "loss": 0.3145,
1201
+ "step": 1760
1202
+ },
1203
+ {
1204
+ "epoch": 0.65,
1205
+ "learning_rate": 0.00019428878100481606,
1206
+ "loss": 0.3842,
1207
+ "step": 1770
1208
+ },
1209
+ {
1210
+ "epoch": 0.65,
1211
+ "learning_rate": 0.0001942246953564967,
1212
+ "loss": 0.3674,
1213
+ "step": 1780
1214
+ },
1215
+ {
1216
+ "epoch": 0.66,
1217
+ "learning_rate": 0.00019416026284772825,
1218
+ "loss": 0.3603,
1219
+ "step": 1790
1220
+ },
1221
+ {
1222
+ "epoch": 0.66,
1223
+ "learning_rate": 0.00019409548371570007,
1224
+ "loss": 0.369,
1225
+ "step": 1800
1226
+ },
1227
+ {
1228
+ "epoch": 0.66,
1229
+ "eval_loss": 0.4111176133155823,
1230
+ "eval_runtime": 402.9829,
1231
+ "eval_samples_per_second": 11.869,
1232
+ "eval_steps_per_second": 2.968,
1233
+ "step": 1800
1234
+ },
1235
+ {
1236
+ "epoch": 0.66,
1237
+ "learning_rate": 0.00019403035819887734,
1238
+ "loss": 0.3713,
1239
+ "step": 1810
1240
+ },
1241
+ {
1242
+ "epoch": 0.67,
1243
+ "learning_rate": 0.00019396488653700055,
1244
+ "loss": 0.3539,
1245
+ "step": 1820
1246
+ },
1247
+ {
1248
+ "epoch": 0.67,
1249
+ "learning_rate": 0.00019389906897108428,
1250
+ "loss": 0.3932,
1251
+ "step": 1830
1252
+ },
1253
+ {
1254
+ "epoch": 0.67,
1255
+ "learning_rate": 0.0001938329057434165,
1256
+ "loss": 0.3371,
1257
+ "step": 1840
1258
+ },
1259
+ {
1260
+ "epoch": 0.68,
1261
+ "learning_rate": 0.00019376639709755766,
1262
+ "loss": 0.3832,
1263
+ "step": 1850
1264
+ },
1265
+ {
1266
+ "epoch": 0.68,
1267
+ "learning_rate": 0.00019369954327833972,
1268
+ "loss": 0.3658,
1269
+ "step": 1860
1270
+ },
1271
+ {
1272
+ "epoch": 0.69,
1273
+ "learning_rate": 0.00019363234453186534,
1274
+ "loss": 0.3452,
1275
+ "step": 1870
1276
+ },
1277
+ {
1278
+ "epoch": 0.69,
1279
+ "learning_rate": 0.00019356480110550687,
1280
+ "loss": 0.3721,
1281
+ "step": 1880
1282
+ },
1283
+ {
1284
+ "epoch": 0.69,
1285
+ "learning_rate": 0.00019349691324790555,
1286
+ "loss": 0.3637,
1287
+ "step": 1890
1288
+ },
1289
+ {
1290
+ "epoch": 0.7,
1291
+ "learning_rate": 0.00019342868120897054,
1292
+ "loss": 0.3409,
1293
+ "step": 1900
1294
+ },
1295
+ {
1296
+ "epoch": 0.7,
1297
+ "eval_loss": 0.41198766231536865,
1298
+ "eval_runtime": 402.9688,
1299
+ "eval_samples_per_second": 11.869,
1300
+ "eval_steps_per_second": 2.968,
1301
+ "step": 1900
1302
+ },
1303
+ {
1304
+ "epoch": 0.7,
1305
+ "learning_rate": 0.00019336010523987796,
1306
+ "loss": 0.3424,
1307
+ "step": 1910
1308
+ },
1309
+ {
1310
+ "epoch": 0.7,
1311
+ "learning_rate": 0.00019329118559307,
1312
+ "loss": 0.3532,
1313
+ "step": 1920
1314
+ },
1315
+ {
1316
+ "epoch": 0.71,
1317
+ "learning_rate": 0.0001932219225222541,
1318
+ "loss": 0.3591,
1319
+ "step": 1930
1320
+ },
1321
+ {
1322
+ "epoch": 0.71,
1323
+ "learning_rate": 0.00019315231628240178,
1324
+ "loss": 0.3794,
1325
+ "step": 1940
1326
+ },
1327
+ {
1328
+ "epoch": 0.71,
1329
+ "learning_rate": 0.00019308236712974795,
1330
+ "loss": 0.3825,
1331
+ "step": 1950
1332
+ },
1333
+ {
1334
+ "epoch": 0.72,
1335
+ "learning_rate": 0.0001930120753217898,
1336
+ "loss": 0.3482,
1337
+ "step": 1960
1338
+ },
1339
+ {
1340
+ "epoch": 0.72,
1341
+ "learning_rate": 0.00019294144111728584,
1342
+ "loss": 0.3643,
1343
+ "step": 1970
1344
+ },
1345
+ {
1346
+ "epoch": 0.73,
1347
+ "learning_rate": 0.00019287046477625515,
1348
+ "loss": 0.3522,
1349
+ "step": 1980
1350
+ },
1351
+ {
1352
+ "epoch": 0.73,
1353
+ "learning_rate": 0.00019279914655997619,
1354
+ "loss": 0.3723,
1355
+ "step": 1990
1356
+ },
1357
+ {
1358
+ "epoch": 0.73,
1359
+ "learning_rate": 0.00019272748673098596,
1360
+ "loss": 0.388,
1361
+ "step": 2000
1362
+ },
1363
+ {
1364
+ "epoch": 0.73,
1365
+ "eval_loss": 0.41505494713783264,
1366
+ "eval_runtime": 404.7184,
1367
+ "eval_samples_per_second": 11.818,
1368
+ "eval_steps_per_second": 2.955,
1369
+ "step": 2000
1370
+ },
1371
+ {
1372
+ "epoch": 0.74,
1373
+ "learning_rate": 0.000192655485553079,
1374
+ "loss": 0.3739,
1375
+ "step": 2010
1376
+ },
1377
+ {
1378
+ "epoch": 0.74,
1379
+ "learning_rate": 0.00019258314329130641,
1380
+ "loss": 0.3625,
1381
+ "step": 2020
1382
+ },
1383
+ {
1384
+ "epoch": 0.74,
1385
+ "learning_rate": 0.00019251046021197496,
1386
+ "loss": 0.3174,
1387
+ "step": 2030
1388
+ },
1389
+ {
1390
+ "epoch": 0.75,
1391
+ "learning_rate": 0.00019243743658264593,
1392
+ "loss": 0.347,
1393
+ "step": 2040
1394
+ },
1395
+ {
1396
+ "epoch": 0.75,
1397
+ "learning_rate": 0.00019236407267213433,
1398
+ "loss": 0.3837,
1399
+ "step": 2050
1400
+ },
1401
+ {
1402
+ "epoch": 0.75,
1403
+ "learning_rate": 0.00019229036875050777,
1404
+ "loss": 0.3854,
1405
+ "step": 2060
1406
+ },
1407
+ {
1408
+ "epoch": 0.76,
1409
+ "learning_rate": 0.0001922163250890855,
1410
+ "loss": 0.3449,
1411
+ "step": 2070
1412
+ },
1413
+ {
1414
+ "epoch": 0.76,
1415
+ "learning_rate": 0.00019214194196043741,
1416
+ "loss": 0.3507,
1417
+ "step": 2080
1418
+ },
1419
+ {
1420
+ "epoch": 0.77,
1421
+ "learning_rate": 0.00019206721963838317,
1422
+ "loss": 0.3786,
1423
+ "step": 2090
1424
+ },
1425
+ {
1426
+ "epoch": 0.77,
1427
+ "learning_rate": 0.00019199215839799092,
1428
+ "loss": 0.3608,
1429
+ "step": 2100
1430
+ },
1431
+ {
1432
+ "epoch": 0.77,
1433
+ "eval_loss": 0.4092780351638794,
1434
+ "eval_runtime": 403.6583,
1435
+ "eval_samples_per_second": 11.849,
1436
+ "eval_steps_per_second": 2.963,
1437
+ "step": 2100
1438
+ },
1439
+ {
1440
+ "epoch": 0.77,
1441
+ "learning_rate": 0.0001919167585155765,
1442
+ "loss": 0.3233,
1443
+ "step": 2110
1444
+ },
1445
+ {
1446
+ "epoch": 0.78,
1447
+ "learning_rate": 0.00019184102026870235,
1448
+ "loss": 0.3679,
1449
+ "step": 2120
1450
+ },
1451
+ {
1452
+ "epoch": 0.78,
1453
+ "learning_rate": 0.0001917649439361765,
1454
+ "loss": 0.3426,
1455
+ "step": 2130
1456
+ },
1457
+ {
1458
+ "epoch": 0.78,
1459
+ "learning_rate": 0.00019168852979805162,
1460
+ "loss": 0.3272,
1461
+ "step": 2140
1462
+ },
1463
+ {
1464
+ "epoch": 0.79,
1465
+ "learning_rate": 0.00019161177813562379,
1466
+ "loss": 0.3572,
1467
+ "step": 2150
1468
+ },
1469
+ {
1470
+ "epoch": 0.79,
1471
+ "learning_rate": 0.0001915346892314316,
1472
+ "loss": 0.3429,
1473
+ "step": 2160
1474
+ },
1475
+ {
1476
+ "epoch": 0.8,
1477
+ "learning_rate": 0.0001914572633692552,
1478
+ "loss": 0.3325,
1479
+ "step": 2170
1480
+ },
1481
+ {
1482
+ "epoch": 0.8,
1483
+ "learning_rate": 0.00019137950083411505,
1484
+ "loss": 0.3444,
1485
+ "step": 2180
1486
+ },
1487
+ {
1488
+ "epoch": 0.8,
1489
+ "learning_rate": 0.000191301401912271,
1490
+ "loss": 0.351,
1491
+ "step": 2190
1492
+ },
1493
+ {
1494
+ "epoch": 0.81,
1495
+ "learning_rate": 0.00019122296689122123,
1496
+ "loss": 0.3171,
1497
+ "step": 2200
1498
+ },
1499
+ {
1500
+ "epoch": 0.81,
1501
+ "eval_loss": 0.4078885316848755,
1502
+ "eval_runtime": 403.041,
1503
+ "eval_samples_per_second": 11.867,
1504
+ "eval_steps_per_second": 2.967,
1505
+ "step": 2200
1506
+ },
1507
+ {
1508
+ "epoch": 0.81,
1509
+ "learning_rate": 0.0001911441960597012,
1510
+ "loss": 0.3626,
1511
+ "step": 2210
1512
+ },
1513
+ {
1514
+ "epoch": 0.81,
1515
+ "learning_rate": 0.0001910650897076824,
1516
+ "loss": 0.3529,
1517
+ "step": 2220
1518
+ },
1519
+ {
1520
+ "epoch": 0.82,
1521
+ "learning_rate": 0.00019098564812637165,
1522
+ "loss": 0.3727,
1523
+ "step": 2230
1524
+ },
1525
+ {
1526
+ "epoch": 0.82,
1527
+ "learning_rate": 0.0001909058716082097,
1528
+ "loss": 0.3111,
1529
+ "step": 2240
1530
+ },
1531
+ {
1532
+ "epoch": 0.82,
1533
+ "learning_rate": 0.0001908257604468703,
1534
+ "loss": 0.3846,
1535
+ "step": 2250
1536
+ },
1537
+ {
1538
+ "epoch": 0.83,
1539
+ "learning_rate": 0.00019074531493725906,
1540
+ "loss": 0.3612,
1541
+ "step": 2260
1542
+ },
1543
+ {
1544
+ "epoch": 0.83,
1545
+ "learning_rate": 0.0001906645353755124,
1546
+ "loss": 0.3573,
1547
+ "step": 2270
1548
+ },
1549
+ {
1550
+ "epoch": 0.84,
1551
+ "learning_rate": 0.00019058342205899656,
1552
+ "loss": 0.3902,
1553
+ "step": 2280
1554
+ },
1555
+ {
1556
+ "epoch": 0.84,
1557
+ "learning_rate": 0.0001905019752863062,
1558
+ "loss": 0.345,
1559
+ "step": 2290
1560
+ },
1561
+ {
1562
+ "epoch": 0.84,
1563
+ "learning_rate": 0.0001904201953572637,
1564
+ "loss": 0.3581,
1565
+ "step": 2300
1566
+ },
1567
+ {
1568
+ "epoch": 0.84,
1569
+ "eval_loss": 0.4118511974811554,
1570
+ "eval_runtime": 405.9796,
1571
+ "eval_samples_per_second": 11.781,
1572
+ "eval_steps_per_second": 2.946,
1573
+ "step": 2300
1574
+ },
1575
+ {
1576
+ "epoch": 0.85,
1577
+ "learning_rate": 0.00019033808257291768,
1578
+ "loss": 0.3303,
1579
+ "step": 2310
1580
+ },
1581
+ {
1582
+ "epoch": 0.85,
1583
+ "learning_rate": 0.00019025563723554223,
1584
+ "loss": 0.3432,
1585
+ "step": 2320
1586
+ },
1587
+ {
1588
+ "epoch": 0.85,
1589
+ "learning_rate": 0.00019017285964863554,
1590
+ "loss": 0.3436,
1591
+ "step": 2330
1592
+ },
1593
+ {
1594
+ "epoch": 0.86,
1595
+ "learning_rate": 0.0001900897501169189,
1596
+ "loss": 0.338,
1597
+ "step": 2340
1598
+ },
1599
+ {
1600
+ "epoch": 0.86,
1601
+ "learning_rate": 0.0001900063089463356,
1602
+ "loss": 0.3611,
1603
+ "step": 2350
1604
+ },
1605
+ {
1606
+ "epoch": 0.86,
1607
+ "learning_rate": 0.00018992253644404967,
1608
+ "loss": 0.3311,
1609
+ "step": 2360
1610
+ },
1611
+ {
1612
+ "epoch": 0.87,
1613
+ "learning_rate": 0.00018983843291844492,
1614
+ "loss": 0.3337,
1615
+ "step": 2370
1616
+ },
1617
+ {
1618
+ "epoch": 0.87,
1619
+ "learning_rate": 0.00018975399867912364,
1620
+ "loss": 0.358,
1621
+ "step": 2380
1622
+ },
1623
+ {
1624
+ "epoch": 0.88,
1625
+ "learning_rate": 0.0001896692340369057,
1626
+ "loss": 0.3802,
1627
+ "step": 2390
1628
+ },
1629
+ {
1630
+ "epoch": 0.88,
1631
+ "learning_rate": 0.00018958413930382704,
1632
+ "loss": 0.3389,
1633
+ "step": 2400
1634
+ },
1635
+ {
1636
+ "epoch": 0.88,
1637
+ "eval_loss": 0.41193756461143494,
1638
+ "eval_runtime": 402.9836,
1639
+ "eval_samples_per_second": 11.869,
1640
+ "eval_steps_per_second": 2.968,
1641
+ "step": 2400
1642
+ },
1643
+ {
1644
+ "epoch": 0.88,
1645
+ "learning_rate": 0.0001894987147931389,
1646
+ "loss": 0.3398,
1647
+ "step": 2410
1648
+ },
1649
+ {
1650
+ "epoch": 0.89,
1651
+ "learning_rate": 0.00018941296081930646,
1652
+ "loss": 0.3459,
1653
+ "step": 2420
1654
+ },
1655
+ {
1656
+ "epoch": 0.89,
1657
+ "learning_rate": 0.00018932687769800767,
1658
+ "loss": 0.3546,
1659
+ "step": 2430
1660
+ },
1661
+ {
1662
+ "epoch": 0.89,
1663
+ "learning_rate": 0.00018924046574613222,
1664
+ "loss": 0.3158,
1665
+ "step": 2440
1666
+ },
1667
+ {
1668
+ "epoch": 0.9,
1669
+ "learning_rate": 0.0001891537252817802,
1670
+ "loss": 0.3484,
1671
+ "step": 2450
1672
+ },
1673
+ {
1674
+ "epoch": 0.9,
1675
+ "learning_rate": 0.00018906665662426104,
1676
+ "loss": 0.3437,
1677
+ "step": 2460
1678
+ },
1679
+ {
1680
+ "epoch": 0.9,
1681
+ "learning_rate": 0.0001889792600940924,
1682
+ "loss": 0.3637,
1683
+ "step": 2470
1684
+ },
1685
+ {
1686
+ "epoch": 0.91,
1687
+ "learning_rate": 0.00018889153601299888,
1688
+ "loss": 0.3658,
1689
+ "step": 2480
1690
+ },
1691
+ {
1692
+ "epoch": 0.91,
1693
+ "learning_rate": 0.00018880348470391077,
1694
+ "loss": 0.3452,
1695
+ "step": 2490
1696
+ },
1697
+ {
1698
+ "epoch": 0.92,
1699
+ "learning_rate": 0.000188715106490963,
1700
+ "loss": 0.3302,
1701
+ "step": 2500
1702
+ },
1703
+ {
1704
+ "epoch": 0.92,
1705
+ "eval_loss": 0.4021553695201874,
1706
+ "eval_runtime": 399.2274,
1707
+ "eval_samples_per_second": 11.981,
1708
+ "eval_steps_per_second": 2.996,
1709
+ "step": 2500
1710
+ },
1711
+ {
1712
+ "epoch": 0.92,
1713
+ "learning_rate": 0.000188626401699494,
1714
+ "loss": 0.3511,
1715
+ "step": 2510
1716
+ },
1717
+ {
1718
+ "epoch": 0.92,
1719
+ "learning_rate": 0.00018853737065604426,
1720
+ "loss": 0.3232,
1721
+ "step": 2520
1722
+ },
1723
+ {
1724
+ "epoch": 0.93,
1725
+ "learning_rate": 0.00018844801368835532,
1726
+ "loss": 0.366,
1727
+ "step": 2530
1728
+ },
1729
+ {
1730
+ "epoch": 0.93,
1731
+ "learning_rate": 0.00018835833112536857,
1732
+ "loss": 0.3151,
1733
+ "step": 2540
1734
+ },
1735
+ {
1736
+ "epoch": 0.93,
1737
+ "learning_rate": 0.0001882683232972239,
1738
+ "loss": 0.3534,
1739
+ "step": 2550
1740
+ },
1741
+ {
1742
+ "epoch": 0.94,
1743
+ "learning_rate": 0.00018817799053525862,
1744
+ "loss": 0.3779,
1745
+ "step": 2560
1746
+ },
1747
+ {
1748
+ "epoch": 0.94,
1749
+ "learning_rate": 0.0001880873331720062,
1750
+ "loss": 0.3338,
1751
+ "step": 2570
1752
+ },
1753
+ {
1754
+ "epoch": 0.95,
1755
+ "learning_rate": 0.00018799635154119495,
1756
+ "loss": 0.3243,
1757
+ "step": 2580
1758
+ },
1759
+ {
1760
+ "epoch": 0.95,
1761
+ "learning_rate": 0.00018790504597774698,
1762
+ "loss": 0.3729,
1763
+ "step": 2590
1764
+ },
1765
+ {
1766
+ "epoch": 0.95,
1767
+ "learning_rate": 0.0001878134168177768,
1768
+ "loss": 0.3553,
1769
+ "step": 2600
1770
+ },
1771
+ {
1772
+ "epoch": 0.95,
1773
+ "eval_loss": 0.4054717421531677,
1774
+ "eval_runtime": 400.3737,
1775
+ "eval_samples_per_second": 11.946,
1776
+ "eval_steps_per_second": 2.987,
1777
+ "step": 2600
1778
+ },
1779
+ {
1780
+ "epoch": 0.96,
1781
+ "learning_rate": 0.00018772146439859015,
1782
+ "loss": 0.3268,
1783
+ "step": 2610
1784
+ },
1785
+ {
1786
+ "epoch": 0.96,
1787
+ "learning_rate": 0.00018762918905868277,
1788
+ "loss": 0.3304,
1789
+ "step": 2620
1790
+ },
1791
+ {
1792
+ "epoch": 0.96,
1793
+ "learning_rate": 0.00018753659113773913,
1794
+ "loss": 0.3606,
1795
+ "step": 2630
1796
+ },
1797
+ {
1798
+ "epoch": 0.97,
1799
+ "learning_rate": 0.0001874436709766312,
1800
+ "loss": 0.3421,
1801
+ "step": 2640
1802
+ },
1803
+ {
1804
+ "epoch": 0.97,
1805
+ "learning_rate": 0.00018735042891741718,
1806
+ "loss": 0.3629,
1807
+ "step": 2650
1808
+ },
1809
+ {
1810
+ "epoch": 0.97,
1811
+ "learning_rate": 0.0001872568653033402,
1812
+ "loss": 0.3636,
1813
+ "step": 2660
1814
+ },
1815
+ {
1816
+ "epoch": 0.98,
1817
+ "learning_rate": 0.00018716298047882714,
1818
+ "loss": 0.3485,
1819
+ "step": 2670
1820
+ },
1821
+ {
1822
+ "epoch": 0.98,
1823
+ "learning_rate": 0.00018706877478948735,
1824
+ "loss": 0.3169,
1825
+ "step": 2680
1826
+ },
1827
+ {
1828
+ "epoch": 0.99,
1829
+ "learning_rate": 0.00018697424858211126,
1830
+ "loss": 0.306,
1831
+ "step": 2690
1832
+ },
1833
+ {
1834
+ "epoch": 0.99,
1835
+ "learning_rate": 0.0001868794022046693,
1836
+ "loss": 0.3586,
1837
+ "step": 2700
1838
+ },
1839
+ {
1840
+ "epoch": 0.99,
1841
+ "eval_loss": 0.4048784375190735,
1842
+ "eval_runtime": 400.9367,
1843
+ "eval_samples_per_second": 11.93,
1844
+ "eval_steps_per_second": 2.983,
1845
+ "step": 2700
1846
+ },
1847
+ {
1848
+ "epoch": 0.99,
1849
+ "learning_rate": 0.00018678423600631042,
1850
+ "loss": 0.3311,
1851
+ "step": 2710
1852
+ },
1853
+ {
1854
+ "epoch": 1.0,
1855
+ "learning_rate": 0.00018668875033736094,
1856
+ "loss": 0.3066,
1857
+ "step": 2720
1858
+ },
1859
+ {
1860
+ "epoch": 1.0,
1861
+ "learning_rate": 0.00018659294554932324,
1862
+ "loss": 0.338,
1863
+ "step": 2730
1864
+ },
1865
+ {
1866
+ "epoch": 1.0,
1867
+ "learning_rate": 0.00018649682199487437,
1868
+ "loss": 0.3205,
1869
+ "step": 2740
1870
+ },
1871
+ {
1872
+ "epoch": 1.01,
1873
+ "learning_rate": 0.0001864003800278649,
1874
+ "loss": 0.3176,
1875
+ "step": 2750
1876
+ },
1877
+ {
1878
+ "epoch": 1.01,
1879
+ "learning_rate": 0.00018630362000331753,
1880
+ "loss": 0.2986,
1881
+ "step": 2760
1882
+ },
1883
+ {
1884
+ "epoch": 1.01,
1885
+ "learning_rate": 0.00018620654227742572,
1886
+ "loss": 0.3175,
1887
+ "step": 2770
1888
+ },
1889
+ {
1890
+ "epoch": 1.02,
1891
+ "learning_rate": 0.0001861091472075526,
1892
+ "loss": 0.3174,
1893
+ "step": 2780
1894
+ },
1895
+ {
1896
+ "epoch": 1.02,
1897
+ "learning_rate": 0.0001860114351522293,
1898
+ "loss": 0.2752,
1899
+ "step": 2790
1900
+ },
1901
+ {
1902
+ "epoch": 1.03,
1903
+ "learning_rate": 0.00018591340647115402,
1904
+ "loss": 0.3014,
1905
+ "step": 2800
1906
+ },
1907
+ {
1908
+ "epoch": 1.03,
1909
+ "eval_loss": 0.41257622838020325,
1910
+ "eval_runtime": 223.7363,
1911
+ "eval_samples_per_second": 21.378,
1912
+ "eval_steps_per_second": 5.346,
1913
+ "step": 2800
1914
+ },
1915
+ {
1916
+ "epoch": 1.03,
1917
+ "learning_rate": 0.0001858150615251905,
1918
+ "loss": 0.3013,
1919
+ "step": 2810
1920
+ },
1921
+ {
1922
+ "epoch": 1.03,
1923
+ "learning_rate": 0.00018571640067636662,
1924
+ "loss": 0.3364,
1925
+ "step": 2820
1926
+ },
1927
+ {
1928
+ "epoch": 1.04,
1929
+ "learning_rate": 0.00018561742428787324,
1930
+ "loss": 0.3159,
1931
+ "step": 2830
1932
+ },
1933
+ {
1934
+ "epoch": 1.04,
1935
+ "learning_rate": 0.00018552807605292504,
1936
+ "loss": 0.3037,
1937
+ "step": 2840
1938
+ },
1939
+ {
1940
+ "epoch": 1.04,
1941
+ "learning_rate": 0.00018542850114380946,
1942
+ "loss": 0.3087,
1943
+ "step": 2850
1944
+ },
1945
+ {
1946
+ "epoch": 1.05,
1947
+ "learning_rate": 0.00018532861175484162,
1948
+ "loss": 0.2858,
1949
+ "step": 2860
1950
+ },
1951
+ {
1952
+ "epoch": 1.05,
1953
+ "learning_rate": 0.00018522840825373492,
1954
+ "loss": 0.3208,
1955
+ "step": 2870
1956
+ },
1957
+ {
1958
+ "epoch": 1.06,
1959
+ "learning_rate": 0.00018512789100935906,
1960
+ "loss": 0.3117,
1961
+ "step": 2880
1962
+ },
1963
+ {
1964
+ "epoch": 1.06,
1965
+ "learning_rate": 0.00018502706039173856,
1966
+ "loss": 0.2921,
1967
+ "step": 2890
1968
+ },
1969
+ {
1970
+ "epoch": 1.06,
1971
+ "learning_rate": 0.0001849259167720517,
1972
+ "loss": 0.3354,
1973
+ "step": 2900
1974
+ },
1975
+ {
1976
+ "epoch": 1.06,
1977
+ "eval_loss": 0.4081941545009613,
1978
+ "eval_runtime": 223.8258,
1979
+ "eval_samples_per_second": 21.369,
1980
+ "eval_steps_per_second": 5.343,
1981
+ "step": 2900
1982
+ },
1983
+ {
1984
+ "epoch": 1.07,
1985
+ "learning_rate": 0.0001848244605226289,
1986
+ "loss": 0.2866,
1987
+ "step": 2910
1988
+ },
1989
+ {
1990
+ "epoch": 1.07,
1991
+ "learning_rate": 0.0001847226920169514,
1992
+ "loss": 0.308,
1993
+ "step": 2920
1994
+ },
1995
+ {
1996
+ "epoch": 1.07,
1997
+ "learning_rate": 0.00018462061162965,
1998
+ "loss": 0.3142,
1999
+ "step": 2930
2000
+ },
2001
+ {
2002
+ "epoch": 1.08,
2003
+ "learning_rate": 0.0001845182197365036,
2004
+ "loss": 0.3125,
2005
+ "step": 2940
2006
+ },
2007
+ {
2008
+ "epoch": 1.08,
2009
+ "learning_rate": 0.00018441551671443768,
2010
+ "loss": 0.307,
2011
+ "step": 2950
2012
+ },
2013
+ {
2014
+ "epoch": 1.08,
2015
+ "learning_rate": 0.00018431250294152323,
2016
+ "loss": 0.2804,
2017
+ "step": 2960
2018
+ },
2019
+ {
2020
+ "epoch": 1.09,
2021
+ "learning_rate": 0.00018420917879697507,
2022
+ "loss": 0.304,
2023
+ "step": 2970
2024
+ },
2025
+ {
2026
+ "epoch": 1.09,
2027
+ "learning_rate": 0.0001841055446611506,
2028
+ "loss": 0.3152,
2029
+ "step": 2980
2030
+ },
2031
+ {
2032
+ "epoch": 1.1,
2033
+ "learning_rate": 0.00018400160091554835,
2034
+ "loss": 0.3095,
2035
+ "step": 2990
2036
+ },
2037
+ {
2038
+ "epoch": 1.1,
2039
+ "learning_rate": 0.0001838973479428066,
2040
+ "loss": 0.2954,
2041
+ "step": 3000
2042
+ },
2043
+ {
2044
+ "epoch": 1.1,
2045
+ "eval_loss": 0.4158288836479187,
2046
+ "eval_runtime": 223.8063,
2047
+ "eval_samples_per_second": 21.371,
2048
+ "eval_steps_per_second": 5.344,
2049
+ "step": 3000
2050
+ },
2051
+ {
2052
+ "epoch": 1.1,
2053
+ "learning_rate": 0.00018379278612670193,
2054
+ "loss": 0.3167,
2055
+ "step": 3010
2056
+ },
2057
+ {
2058
+ "epoch": 1.11,
2059
+ "learning_rate": 0.00018368791585214784,
2060
+ "loss": 0.2872,
2061
+ "step": 3020
2062
+ },
2063
+ {
2064
+ "epoch": 1.11,
2065
+ "learning_rate": 0.00018358273750519337,
2066
+ "loss": 0.2999,
2067
+ "step": 3030
2068
+ },
2069
+ {
2070
+ "epoch": 1.11,
2071
+ "learning_rate": 0.00018347725147302158,
2072
+ "loss": 0.3229,
2073
+ "step": 3040
2074
+ },
2075
+ {
2076
+ "epoch": 1.12,
2077
+ "learning_rate": 0.00018337145814394825,
2078
+ "loss": 0.3332,
2079
+ "step": 3050
2080
+ },
2081
+ {
2082
+ "epoch": 1.12,
2083
+ "learning_rate": 0.0001832653579074203,
2084
+ "loss": 0.2818,
2085
+ "step": 3060
2086
+ },
2087
+ {
2088
+ "epoch": 1.12,
2089
+ "learning_rate": 0.00018315895115401457,
2090
+ "loss": 0.3198,
2091
+ "step": 3070
2092
+ },
2093
+ {
2094
+ "epoch": 1.13,
2095
+ "learning_rate": 0.00018305223827543604,
2096
+ "loss": 0.3504,
2097
+ "step": 3080
2098
+ },
2099
+ {
2100
+ "epoch": 1.13,
2101
+ "learning_rate": 0.0001829452196645168,
2102
+ "loss": 0.3392,
2103
+ "step": 3090
2104
+ },
2105
+ {
2106
+ "epoch": 1.14,
2107
+ "learning_rate": 0.00018283789571521436,
2108
+ "loss": 0.3023,
2109
+ "step": 3100
2110
+ },
2111
+ {
2112
+ "epoch": 1.14,
2113
+ "eval_loss": 0.40502193570137024,
2114
+ "eval_runtime": 223.8877,
2115
+ "eval_samples_per_second": 21.363,
2116
+ "eval_steps_per_second": 5.342,
2117
+ "step": 3100
2118
+ },
2119
+ {
2120
+ "epoch": 1.14,
2121
+ "learning_rate": 0.00018273026682261013,
2122
+ "loss": 0.3164,
2123
+ "step": 3110
2124
+ },
2125
+ {
2126
+ "epoch": 1.14,
2127
+ "learning_rate": 0.0001826223333829082,
2128
+ "loss": 0.2985,
2129
+ "step": 3120
2130
+ },
2131
+ {
2132
+ "epoch": 1.15,
2133
+ "learning_rate": 0.00018251409579343375,
2134
+ "loss": 0.2992,
2135
+ "step": 3130
2136
+ },
2137
+ {
2138
+ "epoch": 1.15,
2139
+ "learning_rate": 0.00018240555445263153,
2140
+ "loss": 0.3174,
2141
+ "step": 3140
2142
+ },
2143
+ {
2144
+ "epoch": 1.15,
2145
+ "learning_rate": 0.00018229670976006453,
2146
+ "loss": 0.3025,
2147
+ "step": 3150
2148
+ },
2149
+ {
2150
+ "epoch": 1.16,
2151
+ "learning_rate": 0.00018218756211641236,
2152
+ "loss": 0.3297,
2153
+ "step": 3160
2154
+ },
2155
+ {
2156
+ "epoch": 1.16,
2157
+ "learning_rate": 0.00018207811192346996,
2158
+ "loss": 0.3139,
2159
+ "step": 3170
2160
+ },
2161
+ {
2162
+ "epoch": 1.17,
2163
+ "learning_rate": 0.00018196835958414598,
2164
+ "loss": 0.2956,
2165
+ "step": 3180
2166
+ },
2167
+ {
2168
+ "epoch": 1.17,
2169
+ "learning_rate": 0.00018185830550246124,
2170
+ "loss": 0.3189,
2171
+ "step": 3190
2172
+ },
2173
+ {
2174
+ "epoch": 1.17,
2175
+ "learning_rate": 0.00018174795008354743,
2176
+ "loss": 0.2896,
2177
+ "step": 3200
2178
+ },
2179
+ {
2180
+ "epoch": 1.17,
2181
+ "eval_loss": 0.40529006719589233,
2182
+ "eval_runtime": 223.8738,
2183
+ "eval_samples_per_second": 21.365,
2184
+ "eval_steps_per_second": 5.342,
2185
+ "step": 3200
2186
+ },
2187
+ {
2188
+ "epoch": 1.18,
2189
+ "learning_rate": 0.00018163729373364554,
2190
+ "loss": 0.3085,
2191
+ "step": 3210
2192
+ },
2193
+ {
2194
+ "epoch": 1.18,
2195
+ "learning_rate": 0.0001815263368601043,
2196
+ "loss": 0.316,
2197
+ "step": 3220
2198
+ },
2199
+ {
2200
+ "epoch": 1.18,
2201
+ "learning_rate": 0.00018141507987137873,
2202
+ "loss": 0.326,
2203
+ "step": 3230
2204
+ },
2205
+ {
2206
+ "epoch": 1.19,
2207
+ "learning_rate": 0.00018130352317702865,
2208
+ "loss": 0.293,
2209
+ "step": 3240
2210
+ },
2211
+ {
2212
+ "epoch": 1.19,
2213
+ "learning_rate": 0.00018119166718771716,
2214
+ "loss": 0.2887,
2215
+ "step": 3250
2216
+ },
2217
+ {
2218
+ "epoch": 1.19,
2219
+ "learning_rate": 0.00018107951231520911,
2220
+ "loss": 0.3266,
2221
+ "step": 3260
2222
+ },
2223
+ {
2224
+ "epoch": 1.2,
2225
+ "learning_rate": 0.00018096705897236966,
2226
+ "loss": 0.3327,
2227
+ "step": 3270
2228
+ },
2229
+ {
2230
+ "epoch": 1.2,
2231
+ "learning_rate": 0.00018085430757316256,
2232
+ "loss": 0.2991,
2233
+ "step": 3280
2234
+ },
2235
+ {
2236
+ "epoch": 1.21,
2237
+ "learning_rate": 0.00018074125853264898,
2238
+ "loss": 0.3076,
2239
+ "step": 3290
2240
+ },
2241
+ {
2242
+ "epoch": 1.21,
2243
+ "learning_rate": 0.00018062791226698558,
2244
+ "loss": 0.3339,
2245
+ "step": 3300
2246
+ },
2247
+ {
2248
+ "epoch": 1.21,
2249
+ "eval_loss": 0.4053775668144226,
2250
+ "eval_runtime": 223.9954,
2251
+ "eval_samples_per_second": 21.353,
2252
+ "eval_steps_per_second": 5.339,
2253
+ "step": 3300
2254
+ },
2255
+ {
2256
+ "epoch": 1.21,
2257
+ "learning_rate": 0.00018051426919342317,
2258
+ "loss": 0.2935,
2259
+ "step": 3310
2260
+ },
2261
+ {
2262
+ "epoch": 1.22,
2263
+ "learning_rate": 0.00018040032973030536,
2264
+ "loss": 0.3497,
2265
+ "step": 3320
2266
+ },
2267
+ {
2268
+ "epoch": 1.22,
2269
+ "learning_rate": 0.00018028609429706664,
2270
+ "loss": 0.2897,
2271
+ "step": 3330
2272
+ },
2273
+ {
2274
+ "epoch": 1.22,
2275
+ "learning_rate": 0.00018017156331423114,
2276
+ "loss": 0.312,
2277
+ "step": 3340
2278
+ },
2279
+ {
2280
+ "epoch": 1.23,
2281
+ "learning_rate": 0.00018005673720341086,
2282
+ "loss": 0.303,
2283
+ "step": 3350
2284
+ },
2285
+ {
2286
+ "epoch": 1.23,
2287
+ "learning_rate": 0.00017994161638730432,
2288
+ "loss": 0.3253,
2289
+ "step": 3360
2290
+ },
2291
+ {
2292
+ "epoch": 1.23,
2293
+ "learning_rate": 0.00017982620128969488,
2294
+ "loss": 0.3322,
2295
+ "step": 3370
2296
+ },
2297
+ {
2298
+ "epoch": 1.24,
2299
+ "learning_rate": 0.0001797104923354492,
2300
+ "loss": 0.2988,
2301
+ "step": 3380
2302
+ },
2303
+ {
2304
+ "epoch": 1.24,
2305
+ "learning_rate": 0.00017959448995051575,
2306
+ "loss": 0.3094,
2307
+ "step": 3390
2308
+ },
2309
+ {
2310
+ "epoch": 1.25,
2311
+ "learning_rate": 0.00017947819456192306,
2312
+ "loss": 0.3118,
2313
+ "step": 3400
2314
+ },
2315
+ {
2316
+ "epoch": 1.25,
2317
+ "eval_loss": 0.3964312672615051,
2318
+ "eval_runtime": 223.8688,
2319
+ "eval_samples_per_second": 21.365,
2320
+ "eval_steps_per_second": 5.342,
2321
+ "step": 3400
2322
+ },
2323
+ {
2324
+ "epoch": 1.25,
2325
+ "learning_rate": 0.00017936160659777833,
2326
+ "loss": 0.2958,
2327
+ "step": 3410
2328
+ },
2329
+ {
2330
+ "epoch": 1.25,
2331
+ "learning_rate": 0.00017924472648726583,
2332
+ "loss": 0.3119,
2333
+ "step": 3420
2334
+ },
2335
+ {
2336
+ "epoch": 1.26,
2337
+ "learning_rate": 0.00017912755466064525,
2338
+ "loss": 0.3096,
2339
+ "step": 3430
2340
+ },
2341
+ {
2342
+ "epoch": 1.26,
2343
+ "learning_rate": 0.00017901009154925007,
2344
+ "loss": 0.2778,
2345
+ "step": 3440
2346
+ },
2347
+ {
2348
+ "epoch": 1.26,
2349
+ "learning_rate": 0.00017889233758548625,
2350
+ "loss": 0.2953,
2351
+ "step": 3450
2352
+ },
2353
+ {
2354
+ "epoch": 1.27,
2355
+ "learning_rate": 0.00017877429320283016,
2356
+ "loss": 0.3063,
2357
+ "step": 3460
2358
+ },
2359
+ {
2360
+ "epoch": 1.27,
2361
+ "learning_rate": 0.0001786559588358275,
2362
+ "loss": 0.2945,
2363
+ "step": 3470
2364
+ },
2365
+ {
2366
+ "epoch": 1.27,
2367
+ "learning_rate": 0.00017853733492009135,
2368
+ "loss": 0.3027,
2369
+ "step": 3480
2370
+ },
2371
+ {
2372
+ "epoch": 1.28,
2373
+ "learning_rate": 0.0001784184218923007,
2374
+ "loss": 0.2654,
2375
+ "step": 3490
2376
+ },
2377
+ {
2378
+ "epoch": 1.28,
2379
+ "learning_rate": 0.0001782992201901988,
2380
+ "loss": 0.3289,
2381
+ "step": 3500
2382
+ },
2383
+ {
2384
+ "epoch": 1.28,
2385
+ "eval_loss": 0.39914268255233765,
2386
+ "eval_runtime": 223.8716,
2387
+ "eval_samples_per_second": 21.365,
2388
+ "eval_steps_per_second": 5.342,
2389
+ "step": 3500
2390
+ },
2391
+ {
2392
+ "epoch": 1.29,
2393
+ "learning_rate": 0.0001781797302525916,
2394
+ "loss": 0.2894,
2395
+ "step": 3510
2396
+ },
2397
+ {
2398
+ "epoch": 1.29,
2399
+ "learning_rate": 0.00017805995251934614,
2400
+ "loss": 0.2776,
2401
+ "step": 3520
2402
+ },
2403
+ {
2404
+ "epoch": 1.29,
2405
+ "learning_rate": 0.00017793988743138877,
2406
+ "loss": 0.288,
2407
+ "step": 3530
2408
+ },
2409
+ {
2410
+ "epoch": 1.3,
2411
+ "learning_rate": 0.00017781953543070372,
2412
+ "loss": 0.3229,
2413
+ "step": 3540
2414
+ },
2415
+ {
2416
+ "epoch": 1.3,
2417
+ "learning_rate": 0.00017769889696033154,
2418
+ "loss": 0.285,
2419
+ "step": 3550
2420
+ },
2421
+ {
2422
+ "epoch": 1.3,
2423
+ "learning_rate": 0.0001775779724643671,
2424
+ "loss": 0.321,
2425
+ "step": 3560
2426
+ },
2427
+ {
2428
+ "epoch": 1.31,
2429
+ "learning_rate": 0.0001774567623879583,
2430
+ "loss": 0.3222,
2431
+ "step": 3570
2432
+ },
2433
+ {
2434
+ "epoch": 1.31,
2435
+ "learning_rate": 0.00017733526717730435,
2436
+ "loss": 0.3087,
2437
+ "step": 3580
2438
+ },
2439
+ {
2440
+ "epoch": 1.32,
2441
+ "learning_rate": 0.00017721348727965408,
2442
+ "loss": 0.2904,
2443
+ "step": 3590
2444
+ },
2445
+ {
2446
+ "epoch": 1.32,
2447
+ "learning_rate": 0.00017709142314330424,
2448
+ "loss": 0.2984,
2449
+ "step": 3600
2450
+ },
2451
+ {
2452
+ "epoch": 1.32,
2453
+ "eval_loss": 0.40588298439979553,
2454
+ "eval_runtime": 223.9372,
2455
+ "eval_samples_per_second": 21.359,
2456
+ "eval_steps_per_second": 5.341,
2457
+ "step": 3600
2458
+ },
2459
+ {
2460
+ "epoch": 1.32,
2461
+ "learning_rate": 0.00017696907521759804,
2462
+ "loss": 0.3216,
2463
+ "step": 3610
2464
+ },
2465
+ {
2466
+ "epoch": 1.33,
2467
+ "learning_rate": 0.00017684644395292326,
2468
+ "loss": 0.3019,
2469
+ "step": 3620
2470
+ },
2471
+ {
2472
+ "epoch": 1.33,
2473
+ "learning_rate": 0.00017672352980071078,
2474
+ "loss": 0.3272,
2475
+ "step": 3630
2476
+ },
2477
+ {
2478
+ "epoch": 1.33,
2479
+ "learning_rate": 0.00017660033321343285,
2480
+ "loss": 0.2892,
2481
+ "step": 3640
2482
+ },
2483
+ {
2484
+ "epoch": 1.34,
2485
+ "learning_rate": 0.0001764768546446014,
2486
+ "loss": 0.2921,
2487
+ "step": 3650
2488
+ },
2489
+ {
2490
+ "epoch": 1.34,
2491
+ "learning_rate": 0.00017635309454876636,
2492
+ "loss": 0.3105,
2493
+ "step": 3660
2494
+ },
2495
+ {
2496
+ "epoch": 1.34,
2497
+ "learning_rate": 0.00017622905338151408,
2498
+ "loss": 0.3205,
2499
+ "step": 3670
2500
+ },
2501
+ {
2502
+ "epoch": 1.35,
2503
+ "learning_rate": 0.00017610473159946556,
2504
+ "loss": 0.2838,
2505
+ "step": 3680
2506
+ },
2507
+ {
2508
+ "epoch": 1.35,
2509
+ "learning_rate": 0.00017598012966027482,
2510
+ "loss": 0.2762,
2511
+ "step": 3690
2512
+ },
2513
+ {
2514
+ "epoch": 1.36,
2515
+ "learning_rate": 0.0001758552480226271,
2516
+ "loss": 0.3277,
2517
+ "step": 3700
2518
+ },
2519
+ {
2520
+ "epoch": 1.36,
2521
+ "eval_loss": 0.3979549705982208,
2522
+ "eval_runtime": 223.8768,
2523
+ "eval_samples_per_second": 21.364,
2524
+ "eval_steps_per_second": 5.342,
2525
+ "step": 3700
2526
+ },
2527
+ {
2528
+ "epoch": 1.36,
2529
+ "learning_rate": 0.00017573008714623746,
2530
+ "loss": 0.2709,
2531
+ "step": 3710
2532
+ },
2533
+ {
2534
+ "epoch": 1.36,
2535
+ "learning_rate": 0.00017560464749184876,
2536
+ "loss": 0.2918,
2537
+ "step": 3720
2538
+ },
2539
+ {
2540
+ "epoch": 1.37,
2541
+ "learning_rate": 0.00017547892952123005,
2542
+ "loss": 0.3098,
2543
+ "step": 3730
2544
+ },
2545
+ {
2546
+ "epoch": 1.37,
2547
+ "learning_rate": 0.00017535293369717506,
2548
+ "loss": 0.3233,
2549
+ "step": 3740
2550
+ },
2551
+ {
2552
+ "epoch": 1.37,
2553
+ "learning_rate": 0.00017522666048350023,
2554
+ "loss": 0.31,
2555
+ "step": 3750
2556
+ },
2557
+ {
2558
+ "epoch": 1.38,
2559
+ "learning_rate": 0.00017510011034504324,
2560
+ "loss": 0.3337,
2561
+ "step": 3760
2562
+ },
2563
+ {
2564
+ "epoch": 1.38,
2565
+ "learning_rate": 0.00017497328374766112,
2566
+ "loss": 0.2845,
2567
+ "step": 3770
2568
+ },
2569
+ {
2570
+ "epoch": 1.38,
2571
+ "learning_rate": 0.00017484618115822857,
2572
+ "loss": 0.3073,
2573
+ "step": 3780
2574
+ },
2575
+ {
2576
+ "epoch": 1.39,
2577
+ "learning_rate": 0.00017471880304463638,
2578
+ "loss": 0.2977,
2579
+ "step": 3790
2580
+ },
2581
+ {
2582
+ "epoch": 1.39,
2583
+ "learning_rate": 0.0001745911498757895,
2584
+ "loss": 0.3011,
2585
+ "step": 3800
2586
+ },
2587
+ {
2588
+ "epoch": 1.39,
2589
+ "eval_loss": 0.4045478105545044,
2590
+ "eval_runtime": 223.7983,
2591
+ "eval_samples_per_second": 21.372,
2592
+ "eval_steps_per_second": 5.344,
2593
+ "step": 3800
2594
+ },
2595
+ {
2596
+ "epoch": 1.4,
2597
+ "learning_rate": 0.00017446322212160545,
2598
+ "loss": 0.3029,
2599
+ "step": 3810
2600
+ },
2601
+ {
2602
+ "epoch": 1.4,
2603
+ "learning_rate": 0.0001743350202530126,
2604
+ "loss": 0.3017,
2605
+ "step": 3820
2606
+ },
2607
+ {
2608
+ "epoch": 1.4,
2609
+ "learning_rate": 0.00017420654474194832,
2610
+ "loss": 0.3179,
2611
+ "step": 3830
2612
+ },
2613
+ {
2614
+ "epoch": 1.41,
2615
+ "learning_rate": 0.00017407779606135732,
2616
+ "loss": 0.2728,
2617
+ "step": 3840
2618
+ },
2619
+ {
2620
+ "epoch": 1.41,
2621
+ "learning_rate": 0.00017394877468518996,
2622
+ "loss": 0.286,
2623
+ "step": 3850
2624
+ },
2625
+ {
2626
+ "epoch": 1.41,
2627
+ "learning_rate": 0.00017381948108840042,
2628
+ "loss": 0.3009,
2629
+ "step": 3860
2630
+ },
2631
+ {
2632
+ "epoch": 1.42,
2633
+ "learning_rate": 0.00017368991574694495,
2634
+ "loss": 0.2946,
2635
+ "step": 3870
2636
+ },
2637
+ {
2638
+ "epoch": 1.42,
2639
+ "learning_rate": 0.0001735600791377802,
2640
+ "loss": 0.3229,
2641
+ "step": 3880
2642
+ },
2643
+ {
2644
+ "epoch": 1.43,
2645
+ "learning_rate": 0.00017342997173886134,
2646
+ "loss": 0.3372,
2647
+ "step": 3890
2648
+ },
2649
+ {
2650
+ "epoch": 1.43,
2651
+ "learning_rate": 0.00017329959402914046,
2652
+ "loss": 0.3194,
2653
+ "step": 3900
2654
+ },
2655
+ {
2656
+ "epoch": 1.43,
2657
+ "eval_loss": 0.40152063965797424,
2658
+ "eval_runtime": 223.8706,
2659
+ "eval_samples_per_second": 21.365,
2660
+ "eval_steps_per_second": 5.342,
2661
+ "step": 3900
2662
+ },
2663
+ {
2664
+ "epoch": 1.43,
2665
+ "learning_rate": 0.0001731689464885647,
2666
+ "loss": 0.2674,
2667
+ "step": 3910
2668
+ },
2669
+ {
2670
+ "epoch": 1.44,
2671
+ "learning_rate": 0.00017303802959807443,
2672
+ "loss": 0.2558,
2673
+ "step": 3920
2674
+ },
2675
+ {
2676
+ "epoch": 1.44,
2677
+ "learning_rate": 0.0001729068438396016,
2678
+ "loss": 0.3186,
2679
+ "step": 3930
2680
+ },
2681
+ {
2682
+ "epoch": 1.44,
2683
+ "learning_rate": 0.00017277538969606793,
2684
+ "loss": 0.3298,
2685
+ "step": 3940
2686
+ },
2687
+ {
2688
+ "epoch": 1.45,
2689
+ "learning_rate": 0.00017264366765138317,
2690
+ "loss": 0.2991,
2691
+ "step": 3950
2692
+ },
2693
+ {
2694
+ "epoch": 1.45,
2695
+ "learning_rate": 0.00017251167819044315,
2696
+ "loss": 0.2652,
2697
+ "step": 3960
2698
+ },
2699
+ {
2700
+ "epoch": 1.45,
2701
+ "learning_rate": 0.0001723794217991282,
2702
+ "loss": 0.3177,
2703
+ "step": 3970
2704
+ },
2705
+ {
2706
+ "epoch": 1.46,
2707
+ "learning_rate": 0.00017224689896430117,
2708
+ "loss": 0.3045,
2709
+ "step": 3980
2710
+ },
2711
+ {
2712
+ "epoch": 1.46,
2713
+ "learning_rate": 0.00017211411017380594,
2714
+ "loss": 0.3146,
2715
+ "step": 3990
2716
+ },
2717
+ {
2718
+ "epoch": 1.47,
2719
+ "learning_rate": 0.00017198105591646528,
2720
+ "loss": 0.2921,
2721
+ "step": 4000
2722
+ },
2723
+ {
2724
+ "epoch": 1.47,
2725
+ "eval_loss": 0.400907427072525,
2726
+ "eval_runtime": 224.0601,
2727
+ "eval_samples_per_second": 21.347,
2728
+ "eval_steps_per_second": 5.338,
2729
+ "step": 4000
2730
+ },
2731
+ {
2732
+ "epoch": 1.47,
2733
+ "learning_rate": 0.00017184773668207917,
2734
+ "loss": 0.3441,
2735
+ "step": 4010
2736
+ },
2737
+ {
2738
+ "epoch": 1.47,
2739
+ "learning_rate": 0.00017171415296142315,
2740
+ "loss": 0.2876,
2741
+ "step": 4020
2742
+ },
2743
+ {
2744
+ "epoch": 1.48,
2745
+ "learning_rate": 0.00017158030524624626,
2746
+ "loss": 0.2755,
2747
+ "step": 4030
2748
+ },
2749
+ {
2750
+ "epoch": 1.48,
2751
+ "learning_rate": 0.0001714461940292695,
2752
+ "loss": 0.3169,
2753
+ "step": 4040
2754
+ },
2755
+ {
2756
+ "epoch": 1.48,
2757
+ "learning_rate": 0.00017131181980418374,
2758
+ "loss": 0.2844,
2759
+ "step": 4050
2760
+ },
2761
+ {
2762
+ "epoch": 1.49,
2763
+ "learning_rate": 0.00017117718306564812,
2764
+ "loss": 0.2821,
2765
+ "step": 4060
2766
+ },
2767
+ {
2768
+ "epoch": 1.49,
2769
+ "learning_rate": 0.00017104228430928805,
2770
+ "loss": 0.3002,
2771
+ "step": 4070
2772
+ },
2773
+ {
2774
+ "epoch": 1.49,
2775
+ "learning_rate": 0.00017090712403169364,
2776
+ "loss": 0.3232,
2777
+ "step": 4080
2778
+ },
2779
+ {
2780
+ "epoch": 1.5,
2781
+ "learning_rate": 0.00017077170273041757,
2782
+ "loss": 0.3135,
2783
+ "step": 4090
2784
+ },
2785
+ {
2786
+ "epoch": 1.5,
2787
+ "learning_rate": 0.00017063602090397346,
2788
+ "loss": 0.2917,
2789
+ "step": 4100
2790
+ },
2791
+ {
2792
+ "epoch": 1.5,
2793
+ "eval_loss": 0.4019235670566559,
2794
+ "eval_runtime": 224.3869,
2795
+ "eval_samples_per_second": 21.316,
2796
+ "eval_steps_per_second": 5.33,
2797
+ "step": 4100
2798
+ },
2799
+ {
2800
+ "epoch": 1.51,
2801
+ "learning_rate": 0.00017050007905183398,
2802
+ "loss": 0.3089,
2803
+ "step": 4110
2804
+ },
2805
+ {
2806
+ "epoch": 1.51,
2807
+ "learning_rate": 0.000170363877674429,
2808
+ "loss": 0.3063,
2809
+ "step": 4120
2810
+ },
2811
+ {
2812
+ "epoch": 1.51,
2813
+ "learning_rate": 0.00017022741727314373,
2814
+ "loss": 0.3056,
2815
+ "step": 4130
2816
+ },
2817
+ {
2818
+ "epoch": 1.52,
2819
+ "learning_rate": 0.00017009069835031694,
2820
+ "loss": 0.3168,
2821
+ "step": 4140
2822
+ },
2823
+ {
2824
+ "epoch": 1.52,
2825
+ "learning_rate": 0.00016995372140923907,
2826
+ "loss": 0.3062,
2827
+ "step": 4150
2828
+ },
2829
+ {
2830
+ "epoch": 1.52,
2831
+ "learning_rate": 0.00016981648695415033,
2832
+ "loss": 0.2744,
2833
+ "step": 4160
2834
+ },
2835
+ {
2836
+ "epoch": 1.53,
2837
+ "learning_rate": 0.00016967899549023895,
2838
+ "loss": 0.2877,
2839
+ "step": 4170
2840
+ },
2841
+ {
2842
+ "epoch": 1.53,
2843
+ "learning_rate": 0.00016954124752363922,
2844
+ "loss": 0.3173,
2845
+ "step": 4180
2846
+ },
2847
+ {
2848
+ "epoch": 1.54,
2849
+ "learning_rate": 0.00016940324356142972,
2850
+ "loss": 0.2954,
2851
+ "step": 4190
2852
+ },
2853
+ {
2854
+ "epoch": 1.54,
2855
+ "learning_rate": 0.00016926498411163135,
2856
+ "loss": 0.2792,
2857
+ "step": 4200
2858
+ },
2859
+ {
2860
+ "epoch": 1.54,
2861
+ "eval_loss": 0.4046282172203064,
2862
+ "eval_runtime": 224.0328,
2863
+ "eval_samples_per_second": 21.35,
2864
+ "eval_steps_per_second": 5.339,
2865
+ "step": 4200
2866
+ },
2867
+ {
2868
+ "epoch": 1.54,
2869
+ "learning_rate": 0.00016912646968320552,
2870
+ "loss": 0.3168,
2871
+ "step": 4210
2872
+ },
2873
+ {
2874
+ "epoch": 1.55,
2875
+ "learning_rate": 0.00016898770078605226,
2876
+ "loss": 0.2854,
2877
+ "step": 4220
2878
+ },
2879
+ {
2880
+ "epoch": 1.55,
2881
+ "learning_rate": 0.00016884867793100843,
2882
+ "loss": 0.2979,
2883
+ "step": 4230
2884
+ },
2885
+ {
2886
+ "epoch": 1.55,
2887
+ "learning_rate": 0.00016870940162984566,
2888
+ "loss": 0.2656,
2889
+ "step": 4240
2890
+ },
2891
+ {
2892
+ "epoch": 1.56,
2893
+ "learning_rate": 0.00016856987239526863,
2894
+ "loss": 0.3169,
2895
+ "step": 4250
2896
+ },
2897
+ {
2898
+ "epoch": 1.56,
2899
+ "learning_rate": 0.00016843009074091306,
2900
+ "loss": 0.2872,
2901
+ "step": 4260
2902
+ },
2903
+ {
2904
+ "epoch": 1.56,
2905
+ "learning_rate": 0.00016829005718134397,
2906
+ "loss": 0.2954,
2907
+ "step": 4270
2908
+ },
2909
+ {
2910
+ "epoch": 1.57,
2911
+ "learning_rate": 0.00016814977223205362,
2912
+ "loss": 0.2885,
2913
+ "step": 4280
2914
+ },
2915
+ {
2916
+ "epoch": 1.57,
2917
+ "learning_rate": 0.00016800923640945974,
2918
+ "loss": 0.2927,
2919
+ "step": 4290
2920
+ },
2921
+ {
2922
+ "epoch": 1.58,
2923
+ "learning_rate": 0.0001678684502309035,
2924
+ "loss": 0.2886,
2925
+ "step": 4300
2926
+ },
2927
+ {
2928
+ "epoch": 1.58,
2929
+ "eval_loss": 0.4055146872997284,
2930
+ "eval_runtime": 224.1142,
2931
+ "eval_samples_per_second": 21.342,
2932
+ "eval_steps_per_second": 5.337,
2933
+ "step": 4300
2934
+ },
2935
+ {
2936
+ "epoch": 1.58,
2937
+ "learning_rate": 0.00016772741421464772,
2938
+ "loss": 0.3042,
2939
+ "step": 4310
2940
+ },
2941
+ {
2942
+ "epoch": 1.58,
2943
+ "learning_rate": 0.00016758612887987498,
2944
+ "loss": 0.2858,
2945
+ "step": 4320
2946
+ },
2947
+ {
2948
+ "epoch": 1.59,
2949
+ "learning_rate": 0.00016744459474668557,
2950
+ "loss": 0.3072,
2951
+ "step": 4330
2952
+ },
2953
+ {
2954
+ "epoch": 1.59,
2955
+ "learning_rate": 0.00016730281233609572,
2956
+ "loss": 0.3126,
2957
+ "step": 4340
2958
+ },
2959
+ {
2960
+ "epoch": 1.59,
2961
+ "learning_rate": 0.00016716078217003557,
2962
+ "loss": 0.2748,
2963
+ "step": 4350
2964
+ },
2965
+ {
2966
+ "epoch": 1.6,
2967
+ "learning_rate": 0.00016701850477134734,
2968
+ "loss": 0.2873,
2969
+ "step": 4360
2970
+ },
2971
+ {
2972
+ "epoch": 1.6,
2973
+ "learning_rate": 0.00016687598066378336,
2974
+ "loss": 0.3051,
2975
+ "step": 4370
2976
+ },
2977
+ {
2978
+ "epoch": 1.6,
2979
+ "learning_rate": 0.00016673321037200407,
2980
+ "loss": 0.3077,
2981
+ "step": 4380
2982
+ },
2983
+ {
2984
+ "epoch": 1.61,
2985
+ "learning_rate": 0.0001665901944215763,
2986
+ "loss": 0.2708,
2987
+ "step": 4390
2988
+ },
2989
+ {
2990
+ "epoch": 1.61,
2991
+ "learning_rate": 0.00016644693333897108,
2992
+ "loss": 0.2947,
2993
+ "step": 4400
2994
+ },
2995
+ {
2996
+ "epoch": 1.61,
2997
+ "eval_loss": 0.40514013171195984,
2998
+ "eval_runtime": 224.1094,
2999
+ "eval_samples_per_second": 21.342,
3000
+ "eval_steps_per_second": 5.337,
3001
+ "step": 4400
3002
+ },
3003
+ {
3004
+ "epoch": 1.62,
3005
+ "learning_rate": 0.0001663034276515619,
3006
+ "loss": 0.306,
3007
+ "step": 4410
3008
+ },
3009
+ {
3010
+ "epoch": 1.62,
3011
+ "learning_rate": 0.00016615967788762261,
3012
+ "loss": 0.3076,
3013
+ "step": 4420
3014
+ },
3015
+ {
3016
+ "epoch": 1.62,
3017
+ "learning_rate": 0.00016601568457632566,
3018
+ "loss": 0.2895,
3019
+ "step": 4430
3020
+ },
3021
+ {
3022
+ "epoch": 1.63,
3023
+ "learning_rate": 0.00016587144824773992,
3024
+ "loss": 0.2885,
3025
+ "step": 4440
3026
+ },
3027
+ {
3028
+ "epoch": 1.63,
3029
+ "learning_rate": 0.00016572696943282903,
3030
+ "loss": 0.3046,
3031
+ "step": 4450
3032
+ },
3033
+ {
3034
+ "epoch": 1.63,
3035
+ "learning_rate": 0.00016558224866344907,
3036
+ "loss": 0.3037,
3037
+ "step": 4460
3038
+ },
3039
+ {
3040
+ "epoch": 1.64,
3041
+ "learning_rate": 0.000165437286472347,
3042
+ "loss": 0.28,
3043
+ "step": 4470
3044
+ },
3045
+ {
3046
+ "epoch": 1.64,
3047
+ "learning_rate": 0.00016529208339315833,
3048
+ "loss": 0.2903,
3049
+ "step": 4480
3050
+ },
3051
+ {
3052
+ "epoch": 1.64,
3053
+ "learning_rate": 0.00016514663996040544,
3054
+ "loss": 0.2849,
3055
+ "step": 4490
3056
+ },
3057
+ {
3058
+ "epoch": 1.65,
3059
+ "learning_rate": 0.00016500095670949548,
3060
+ "loss": 0.2975,
3061
+ "step": 4500
3062
+ },
3063
+ {
3064
+ "epoch": 1.65,
3065
+ "eval_loss": 0.4067119359970093,
3066
+ "eval_runtime": 224.4914,
3067
+ "eval_samples_per_second": 21.306,
3068
+ "eval_steps_per_second": 5.328,
3069
+ "step": 4500
3070
+ },
3071
+ {
3072
+ "epoch": 1.65,
3073
+ "learning_rate": 0.00016485503417671836,
3074
+ "loss": 0.3115,
3075
+ "step": 4510
3076
+ },
3077
+ {
3078
+ "epoch": 1.66,
3079
+ "learning_rate": 0.00016470887289924492,
3080
+ "loss": 0.3409,
3081
+ "step": 4520
3082
+ },
3083
+ {
3084
+ "epoch": 1.66,
3085
+ "learning_rate": 0.00016456247341512485,
3086
+ "loss": 0.2604,
3087
+ "step": 4530
3088
+ },
3089
+ {
3090
+ "epoch": 1.66,
3091
+ "learning_rate": 0.00016441583626328467,
3092
+ "loss": 0.3226,
3093
+ "step": 4540
3094
+ },
3095
+ {
3096
+ "epoch": 1.67,
3097
+ "learning_rate": 0.00016426896198352587,
3098
+ "loss": 0.2816,
3099
+ "step": 4550
3100
+ },
3101
+ {
3102
+ "epoch": 1.67,
3103
+ "learning_rate": 0.00016412185111652278,
3104
+ "loss": 0.319,
3105
+ "step": 4560
3106
+ },
3107
+ {
3108
+ "epoch": 1.67,
3109
+ "learning_rate": 0.00016397450420382076,
3110
+ "loss": 0.2831,
3111
+ "step": 4570
3112
+ },
3113
+ {
3114
+ "epoch": 1.68,
3115
+ "learning_rate": 0.00016382692178783402,
3116
+ "loss": 0.2974,
3117
+ "step": 4580
3118
+ },
3119
+ {
3120
+ "epoch": 1.68,
3121
+ "learning_rate": 0.00016367910441184374,
3122
+ "loss": 0.2772,
3123
+ "step": 4590
3124
+ },
3125
+ {
3126
+ "epoch": 1.69,
3127
+ "learning_rate": 0.00016353105261999605,
3128
+ "loss": 0.3091,
3129
+ "step": 4600
3130
+ },
3131
+ {
3132
+ "epoch": 1.69,
3133
+ "eval_loss": 0.3946963846683502,
3134
+ "eval_runtime": 224.3647,
3135
+ "eval_samples_per_second": 21.318,
3136
+ "eval_steps_per_second": 5.331,
3137
+ "step": 4600
3138
+ },
3139
+ {
3140
+ "epoch": 1.69,
3141
+ "learning_rate": 0.00016338276695729994,
3142
+ "loss": 0.286,
3143
+ "step": 4610
3144
+ },
3145
+ {
3146
+ "epoch": 1.69,
3147
+ "learning_rate": 0.00016323424796962544,
3148
+ "loss": 0.2721,
3149
+ "step": 4620
3150
+ },
3151
+ {
3152
+ "epoch": 1.7,
3153
+ "learning_rate": 0.0001630854962037014,
3154
+ "loss": 0.2983,
3155
+ "step": 4630
3156
+ },
3157
+ {
3158
+ "epoch": 1.7,
3159
+ "learning_rate": 0.00016293651220711364,
3160
+ "loss": 0.2875,
3161
+ "step": 4640
3162
+ },
3163
+ {
3164
+ "epoch": 1.7,
3165
+ "learning_rate": 0.0001627872965283028,
3166
+ "loss": 0.2859,
3167
+ "step": 4650
3168
+ },
3169
+ {
3170
+ "epoch": 1.71,
3171
+ "learning_rate": 0.00016263784971656247,
3172
+ "loss": 0.2603,
3173
+ "step": 4660
3174
+ },
3175
+ {
3176
+ "epoch": 1.71,
3177
+ "learning_rate": 0.00016248817232203698,
3178
+ "loss": 0.3231,
3179
+ "step": 4670
3180
+ },
3181
+ {
3182
+ "epoch": 1.71,
3183
+ "learning_rate": 0.00016233826489571963,
3184
+ "loss": 0.2927,
3185
+ "step": 4680
3186
+ },
3187
+ {
3188
+ "epoch": 1.72,
3189
+ "learning_rate": 0.00016218812798945038,
3190
+ "loss": 0.297,
3191
+ "step": 4690
3192
+ },
3193
+ {
3194
+ "epoch": 1.72,
3195
+ "learning_rate": 0.00016203776215591403,
3196
+ "loss": 0.2908,
3197
+ "step": 4700
3198
+ },
3199
+ {
3200
+ "epoch": 1.72,
3201
+ "eval_loss": 0.403292179107666,
3202
+ "eval_runtime": 224.3888,
3203
+ "eval_samples_per_second": 21.316,
3204
+ "eval_steps_per_second": 5.33,
3205
+ "step": 4700
3206
+ },
3207
+ {
3208
+ "epoch": 1.73,
3209
+ "learning_rate": 0.0001618871679486381,
3210
+ "loss": 0.3094,
3211
+ "step": 4710
3212
+ },
3213
+ {
3214
+ "epoch": 1.73,
3215
+ "learning_rate": 0.00016173634592199076,
3216
+ "loss": 0.29,
3217
+ "step": 4720
3218
+ },
3219
+ {
3220
+ "epoch": 1.73,
3221
+ "learning_rate": 0.00016158529663117888,
3222
+ "loss": 0.3115,
3223
+ "step": 4730
3224
+ },
3225
+ {
3226
+ "epoch": 1.74,
3227
+ "learning_rate": 0.00016143402063224584,
3228
+ "loss": 0.3139,
3229
+ "step": 4740
3230
+ },
3231
+ {
3232
+ "epoch": 1.74,
3233
+ "learning_rate": 0.00016128251848206976,
3234
+ "loss": 0.302,
3235
+ "step": 4750
3236
+ },
3237
+ {
3238
+ "epoch": 1.74,
3239
+ "learning_rate": 0.00016113079073836107,
3240
+ "loss": 0.2777,
3241
+ "step": 4760
3242
+ },
3243
+ {
3244
+ "epoch": 1.75,
3245
+ "learning_rate": 0.0001609788379596608,
3246
+ "loss": 0.2906,
3247
+ "step": 4770
3248
+ },
3249
+ {
3250
+ "epoch": 1.75,
3251
+ "learning_rate": 0.00016082666070533832,
3252
+ "loss": 0.3028,
3253
+ "step": 4780
3254
+ },
3255
+ {
3256
+ "epoch": 1.75,
3257
+ "learning_rate": 0.0001606742595355893,
3258
+ "loss": 0.3159,
3259
+ "step": 4790
3260
+ },
3261
+ {
3262
+ "epoch": 1.76,
3263
+ "learning_rate": 0.0001605216350114338,
3264
+ "loss": 0.2864,
3265
+ "step": 4800
3266
+ },
3267
+ {
3268
+ "epoch": 1.76,
3269
+ "eval_loss": 0.40955105423927307,
3270
+ "eval_runtime": 224.3793,
3271
+ "eval_samples_per_second": 21.317,
3272
+ "eval_steps_per_second": 5.33,
3273
+ "step": 4800
3274
+ },
3275
+ {
3276
+ "epoch": 1.76,
3277
+ "learning_rate": 0.00016036878769471401,
3278
+ "loss": 0.3017,
3279
+ "step": 4810
3280
+ },
3281
+ {
3282
+ "epoch": 1.77,
3283
+ "learning_rate": 0.00016021571814809227,
3284
+ "loss": 0.2986,
3285
+ "step": 4820
3286
+ },
3287
+ {
3288
+ "epoch": 1.77,
3289
+ "learning_rate": 0.000160062426935049,
3290
+ "loss": 0.3077,
3291
+ "step": 4830
3292
+ },
3293
+ {
3294
+ "epoch": 1.77,
3295
+ "learning_rate": 0.00015990891461988065,
3296
+ "loss": 0.2758,
3297
+ "step": 4840
3298
+ },
3299
+ {
3300
+ "epoch": 1.78,
3301
+ "learning_rate": 0.00015975518176769755,
3302
+ "loss": 0.2592,
3303
+ "step": 4850
3304
+ },
3305
+ {
3306
+ "epoch": 1.78,
3307
+ "learning_rate": 0.0001596012289444219,
3308
+ "loss": 0.3021,
3309
+ "step": 4860
3310
+ },
3311
+ {
3312
+ "epoch": 1.78,
3313
+ "learning_rate": 0.00015944705671678565,
3314
+ "loss": 0.3047,
3315
+ "step": 4870
3316
+ },
3317
+ {
3318
+ "epoch": 1.79,
3319
+ "learning_rate": 0.00015929266565232846,
3320
+ "loss": 0.2785,
3321
+ "step": 4880
3322
+ },
3323
+ {
3324
+ "epoch": 1.79,
3325
+ "learning_rate": 0.00015913805631939546,
3326
+ "loss": 0.2839,
3327
+ "step": 4890
3328
+ },
3329
+ {
3330
+ "epoch": 1.8,
3331
+ "learning_rate": 0.00015898322928713544,
3332
+ "loss": 0.2788,
3333
+ "step": 4900
3334
+ },
3335
+ {
3336
+ "epoch": 1.8,
3337
+ "eval_loss": 0.40693148970603943,
3338
+ "eval_runtime": 224.6917,
3339
+ "eval_samples_per_second": 21.287,
3340
+ "eval_steps_per_second": 5.323,
3341
+ "step": 4900
3342
+ },
3343
+ {
3344
+ "epoch": 1.8,
3345
+ "learning_rate": 0.00015882818512549836,
3346
+ "loss": 0.2875,
3347
+ "step": 4910
3348
+ },
3349
+ {
3350
+ "epoch": 1.8,
3351
+ "learning_rate": 0.00015867292440523378,
3352
+ "loss": 0.2907,
3353
+ "step": 4920
3354
+ },
3355
+ {
3356
+ "epoch": 1.81,
3357
+ "learning_rate": 0.00015851744769788818,
3358
+ "loss": 0.2776,
3359
+ "step": 4930
3360
+ },
3361
+ {
3362
+ "epoch": 1.81,
3363
+ "learning_rate": 0.00015836175557580324,
3364
+ "loss": 0.2608,
3365
+ "step": 4940
3366
+ },
3367
+ {
3368
+ "epoch": 1.81,
3369
+ "learning_rate": 0.00015820584861211368,
3370
+ "loss": 0.2945,
3371
+ "step": 4950
3372
+ },
3373
+ {
3374
+ "epoch": 1.82,
3375
+ "learning_rate": 0.000158049727380745,
3376
+ "loss": 0.2771,
3377
+ "step": 4960
3378
+ },
3379
+ {
3380
+ "epoch": 1.82,
3381
+ "learning_rate": 0.0001578933924564115,
3382
+ "loss": 0.2802,
3383
+ "step": 4970
3384
+ },
3385
+ {
3386
+ "epoch": 1.82,
3387
+ "learning_rate": 0.0001577368444146142,
3388
+ "loss": 0.2849,
3389
+ "step": 4980
3390
+ },
3391
+ {
3392
+ "epoch": 1.83,
3393
+ "learning_rate": 0.00015758008383163854,
3394
+ "loss": 0.3013,
3395
+ "step": 4990
3396
+ },
3397
+ {
3398
+ "epoch": 1.83,
3399
+ "learning_rate": 0.0001574231112845524,
3400
+ "loss": 0.2942,
3401
+ "step": 5000
3402
+ },
3403
+ {
3404
+ "epoch": 1.83,
3405
+ "eval_loss": 0.399748831987381,
3406
+ "eval_runtime": 224.6973,
3407
+ "eval_samples_per_second": 21.286,
3408
+ "eval_steps_per_second": 5.323,
3409
+ "step": 5000
3410
+ },
3411
+ {
3412
+ "epoch": 1.84,
3413
+ "learning_rate": 0.00015726592735120393,
3414
+ "loss": 0.2883,
3415
+ "step": 5010
3416
+ },
3417
+ {
3418
+ "epoch": 1.84,
3419
+ "learning_rate": 0.0001571085326102195,
3420
+ "loss": 0.2639,
3421
+ "step": 5020
3422
+ },
3423
+ {
3424
+ "epoch": 1.84,
3425
+ "learning_rate": 0.0001569509276410015,
3426
+ "loss": 0.2624,
3427
+ "step": 5030
3428
+ },
3429
+ {
3430
+ "epoch": 1.85,
3431
+ "learning_rate": 0.00015679311302372614,
3432
+ "loss": 0.2823,
3433
+ "step": 5040
3434
+ },
3435
+ {
3436
+ "epoch": 1.85,
3437
+ "learning_rate": 0.0001566350893393414,
3438
+ "loss": 0.3026,
3439
+ "step": 5050
3440
+ },
3441
+ {
3442
+ "epoch": 1.85,
3443
+ "learning_rate": 0.00015647685716956494,
3444
+ "loss": 0.2803,
3445
+ "step": 5060
3446
+ },
3447
+ {
3448
+ "epoch": 1.86,
3449
+ "learning_rate": 0.00015631841709688184,
3450
+ "loss": 0.2991,
3451
+ "step": 5070
3452
+ },
3453
+ {
3454
+ "epoch": 1.86,
3455
+ "learning_rate": 0.00015615976970454257,
3456
+ "loss": 0.2734,
3457
+ "step": 5080
3458
+ },
3459
+ {
3460
+ "epoch": 1.86,
3461
+ "learning_rate": 0.00015600091557656072,
3462
+ "loss": 0.2674,
3463
+ "step": 5090
3464
+ },
3465
+ {
3466
+ "epoch": 1.87,
3467
+ "learning_rate": 0.000155841855297711,
3468
+ "loss": 0.2736,
3469
+ "step": 5100
3470
+ },
3471
+ {
3472
+ "epoch": 1.87,
3473
+ "eval_loss": 0.40985968708992004,
3474
+ "eval_runtime": 224.6419,
3475
+ "eval_samples_per_second": 21.292,
3476
+ "eval_steps_per_second": 5.324,
3477
+ "step": 5100
3478
+ },
3479
+ {
3480
+ "epoch": 1.87,
3481
+ "learning_rate": 0.0001556825894535269,
3482
+ "loss": 0.2828,
3483
+ "step": 5110
3484
+ },
3485
+ {
3486
+ "epoch": 1.88,
3487
+ "learning_rate": 0.00015552311863029875,
3488
+ "loss": 0.2872,
3489
+ "step": 5120
3490
+ },
3491
+ {
3492
+ "epoch": 1.88,
3493
+ "learning_rate": 0.00015536344341507129,
3494
+ "loss": 0.2669,
3495
+ "step": 5130
3496
+ },
3497
+ {
3498
+ "epoch": 1.88,
3499
+ "learning_rate": 0.0001552035643956419,
3500
+ "loss": 0.2708,
3501
+ "step": 5140
3502
+ },
3503
+ {
3504
+ "epoch": 1.89,
3505
+ "learning_rate": 0.00015504348216055798,
3506
+ "loss": 0.3048,
3507
+ "step": 5150
3508
+ },
3509
+ {
3510
+ "epoch": 1.89,
3511
+ "learning_rate": 0.00015488319729911512,
3512
+ "loss": 0.2964,
3513
+ "step": 5160
3514
+ },
3515
+ {
3516
+ "epoch": 1.89,
3517
+ "learning_rate": 0.00015472271040135483,
3518
+ "loss": 0.266,
3519
+ "step": 5170
3520
+ },
3521
+ {
3522
+ "epoch": 1.9,
3523
+ "learning_rate": 0.00015456202205806234,
3524
+ "loss": 0.2885,
3525
+ "step": 5180
3526
+ },
3527
+ {
3528
+ "epoch": 1.9,
3529
+ "learning_rate": 0.0001544011328607644,
3530
+ "loss": 0.3317,
3531
+ "step": 5190
3532
+ },
3533
+ {
3534
+ "epoch": 1.91,
3535
+ "learning_rate": 0.00015424004340172719,
3536
+ "loss": 0.2905,
3537
+ "step": 5200
3538
+ },
3539
+ {
3540
+ "epoch": 1.91,
3541
+ "eval_loss": 0.39825478196144104,
3542
+ "eval_runtime": 224.8976,
3543
+ "eval_samples_per_second": 21.267,
3544
+ "eval_steps_per_second": 5.318,
3545
+ "step": 5200
3546
+ },
3547
+ {
3548
+ "epoch": 1.91,
3549
+ "learning_rate": 0.0001540787542739541,
3550
+ "loss": 0.2663,
3551
+ "step": 5210
3552
+ },
3553
+ {
3554
+ "epoch": 1.91,
3555
+ "learning_rate": 0.00015391726607118345,
3556
+ "loss": 0.2838,
3557
+ "step": 5220
3558
+ },
3559
+ {
3560
+ "epoch": 1.92,
3561
+ "learning_rate": 0.00015375557938788657,
3562
+ "loss": 0.2962,
3563
+ "step": 5230
3564
+ },
3565
+ {
3566
+ "epoch": 1.92,
3567
+ "learning_rate": 0.0001535936948192653,
3568
+ "loss": 0.2884,
3569
+ "step": 5240
3570
+ },
3571
+ {
3572
+ "epoch": 1.92,
3573
+ "learning_rate": 0.00015343161296124994,
3574
+ "loss": 0.3026,
3575
+ "step": 5250
3576
+ },
3577
+ {
3578
+ "epoch": 1.93,
3579
+ "learning_rate": 0.00015326933441049714,
3580
+ "loss": 0.3234,
3581
+ "step": 5260
3582
+ },
3583
+ {
3584
+ "epoch": 1.93,
3585
+ "learning_rate": 0.00015310685976438753,
3586
+ "loss": 0.2547,
3587
+ "step": 5270
3588
+ },
3589
+ {
3590
+ "epoch": 1.93,
3591
+ "learning_rate": 0.00015294418962102363,
3592
+ "loss": 0.2896,
3593
+ "step": 5280
3594
+ },
3595
+ {
3596
+ "epoch": 1.94,
3597
+ "learning_rate": 0.00015278132457922764,
3598
+ "loss": 0.3173,
3599
+ "step": 5290
3600
+ },
3601
+ {
3602
+ "epoch": 1.94,
3603
+ "learning_rate": 0.00015261826523853926,
3604
+ "loss": 0.2848,
3605
+ "step": 5300
3606
+ },
3607
+ {
3608
+ "epoch": 1.94,
3609
+ "eval_loss": 0.4000326693058014,
3610
+ "eval_runtime": 224.8241,
3611
+ "eval_samples_per_second": 21.274,
3612
+ "eval_steps_per_second": 5.32,
3613
+ "step": 5300
3614
+ },
3615
+ {
3616
+ "epoch": 1.95,
3617
+ "learning_rate": 0.00015245501219921336,
3618
+ "loss": 0.2819,
3619
+ "step": 5310
3620
+ },
3621
+ {
3622
+ "epoch": 1.95,
3623
+ "learning_rate": 0.00015229156606221792,
3624
+ "loss": 0.2578,
3625
+ "step": 5320
3626
+ },
3627
+ {
3628
+ "epoch": 1.95,
3629
+ "learning_rate": 0.0001521279274292317,
3630
+ "loss": 0.2893,
3631
+ "step": 5330
3632
+ },
3633
+ {
3634
+ "epoch": 1.96,
3635
+ "learning_rate": 0.00015196409690264212,
3636
+ "loss": 0.2897,
3637
+ "step": 5340
3638
+ },
3639
+ {
3640
+ "epoch": 1.96,
3641
+ "learning_rate": 0.000151800075085543,
3642
+ "loss": 0.2718,
3643
+ "step": 5350
3644
+ },
3645
+ {
3646
+ "epoch": 1.96,
3647
+ "learning_rate": 0.0001516358625817323,
3648
+ "loss": 0.2908,
3649
+ "step": 5360
3650
+ },
3651
+ {
3652
+ "epoch": 1.97,
3653
+ "learning_rate": 0.00015147145999570998,
3654
+ "loss": 0.2694,
3655
+ "step": 5370
3656
+ },
3657
+ {
3658
+ "epoch": 1.97,
3659
+ "learning_rate": 0.0001513068679326757,
3660
+ "loss": 0.3028,
3661
+ "step": 5380
3662
+ },
3663
+ {
3664
+ "epoch": 1.97,
3665
+ "learning_rate": 0.00015114208699852663,
3666
+ "loss": 0.2707,
3667
+ "step": 5390
3668
+ },
3669
+ {
3670
+ "epoch": 1.98,
3671
+ "step": 5391,
3672
+ "total_flos": 1.7572375333880463e+18,
3673
+ "train_loss": 4.6051067095080594e-05,
3674
+ "train_runtime": 5.408,
3675
+ "train_samples_per_second": 15680.552,
3676
+ "train_steps_per_second": 980.035
3677
+ }
3678
+ ],
3679
+ "logging_steps": 10,
3680
+ "max_steps": 5300,
3681
+ "num_input_tokens_seen": 0,
3682
+ "num_train_epochs": 2,
3683
+ "save_steps": 10,
3684
+ "total_flos": 1.7572375333880463e+18,
3685
+ "train_batch_size": 4,
3686
+ "trial_name": null,
3687
+ "trial_params": null
3688
+ }
subject/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92c228a1db36a91126dfb1a7f6e564260da73b3eb7065106838b10950e25fab0
3
+ size 4920
subject/training_eval_loss.png ADDED
subject/training_loss.png ADDED