Model save
Browse files- README.md +25 -26
- adapter_model.safetensors +1 -1
- all_results.json +6 -11
- runs/Jun10_02-23-10_48ddfe8e991f/events.out.tfevents.1717986227.48ddfe8e991f.24991.0 +2 -2
- train_results.json +6 -6
- trainer_state.json +749 -1829
README.md
CHANGED
|
@@ -2,13 +2,12 @@
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
- trl
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
base_model: google/gemma-2b
|
| 10 |
datasets:
|
| 11 |
-
-
|
| 12 |
model-index:
|
| 13 |
- name: gemma2b-summarize-gemini1_5flash-64k
|
| 14 |
results: []
|
|
@@ -19,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
|
| 20 |
# gemma2b-summarize-gemini1_5flash-64k
|
| 21 |
|
| 22 |
-
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
-
- Loss: 2.
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
@@ -41,14 +40,14 @@ More information needed
|
|
| 41 |
|
| 42 |
The following hyperparameters were used during training:
|
| 43 |
- learning_rate: 0.0002
|
| 44 |
-
- train_batch_size:
|
| 45 |
-
- eval_batch_size:
|
| 46 |
- seed: 42
|
| 47 |
- distributed_type: multi-GPU
|
| 48 |
-
- num_devices:
|
| 49 |
- gradient_accumulation_steps: 2
|
| 50 |
-
- total_train_batch_size:
|
| 51 |
-
- total_eval_batch_size:
|
| 52 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
- lr_scheduler_type: cosine
|
| 54 |
- lr_scheduler_warmup_ratio: 0.1
|
|
@@ -58,27 +57,27 @@ The following hyperparameters were used during training:
|
|
| 58 |
|
| 59 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 60 |
|:-------------:|:-------:|:----:|:---------------:|
|
| 61 |
-
| 1.
|
| 62 |
-
| 1.
|
| 63 |
-
| 1.
|
| 64 |
-
|
|
| 65 |
-
|
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
| 76 |
|
| 77 |
|
| 78 |
### Framework versions
|
| 79 |
|
| 80 |
- PEFT 0.11.1
|
| 81 |
-
- Transformers 4.
|
| 82 |
-
- Pytorch 2.
|
| 83 |
- Datasets 2.19.2
|
| 84 |
- Tokenizers 0.19.1
|
|
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
| 8 |
base_model: google/gemma-2b
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: gemma2b-summarize-gemini1_5flash-64k
|
| 13 |
results: []
|
|
|
|
| 18 |
|
| 19 |
# gemma2b-summarize-gemini1_5flash-64k
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 2.7185
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 40 |
|
| 41 |
The following hyperparameters were used during training:
|
| 42 |
- learning_rate: 0.0002
|
| 43 |
+
- train_batch_size: 16
|
| 44 |
+
- eval_batch_size: 16
|
| 45 |
- seed: 42
|
| 46 |
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 8
|
| 48 |
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 256
|
| 50 |
+
- total_eval_batch_size: 128
|
| 51 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
- lr_scheduler_type: cosine
|
| 53 |
- lr_scheduler_warmup_ratio: 0.1
|
|
|
|
| 57 |
|
| 58 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
|:-------------:|:-------:|:----:|:---------------:|
|
| 60 |
+
| 1.518 | 0.9905 | 52 | 2.7709 |
|
| 61 |
+
| 1.1423 | 2.0 | 105 | 2.6595 |
|
| 62 |
+
| 1.0681 | 2.9905 | 157 | 2.6406 |
|
| 63 |
+
| 1.0335 | 4.0 | 210 | 2.6427 |
|
| 64 |
+
| 1.0079 | 4.9905 | 262 | 2.6459 |
|
| 65 |
+
| 0.9837 | 6.0 | 315 | 2.6574 |
|
| 66 |
+
| 0.966 | 6.9905 | 367 | 2.6700 |
|
| 67 |
+
| 0.9474 | 8.0 | 420 | 2.6799 |
|
| 68 |
+
| 0.9406 | 8.9905 | 472 | 2.6883 |
|
| 69 |
+
| 0.9245 | 10.0 | 525 | 2.6975 |
|
| 70 |
+
| 0.9208 | 10.9905 | 577 | 2.7079 |
|
| 71 |
+
| 0.9195 | 12.0 | 630 | 2.7148 |
|
| 72 |
+
| 0.9212 | 12.9905 | 682 | 2.7154 |
|
| 73 |
+
| 0.9136 | 14.0 | 735 | 2.7181 |
|
| 74 |
+
| 0.9103 | 14.8571 | 780 | 2.7185 |
|
| 75 |
|
| 76 |
|
| 77 |
### Framework versions
|
| 78 |
|
| 79 |
- PEFT 0.11.1
|
| 80 |
+
- Transformers 4.41.2
|
| 81 |
+
- Pytorch 2.3.1+cu121
|
| 82 |
- Datasets 2.19.2
|
| 83 |
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 39256960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2ae6086e7cdaa1c7742eb4042577d161b6afc3838df568f2b317918f4e82a95d
|
| 3 |
size 39256960
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 14.
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"eval_samples_per_second": 19.634,
|
| 7 |
-
"eval_steps_per_second": 1.963,
|
| 8 |
-
"total_flos": 1.2217365722824704e+18,
|
| 9 |
-
"train_loss": 1.0227742999027938,
|
| 10 |
-
"train_runtime": 5355.4835,
|
| 11 |
"train_samples": 63353,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 14.857142857142858,
|
| 3 |
+
"total_flos": 1.2277516310308454e+18,
|
| 4 |
+
"train_loss": 1.077159938445458,
|
| 5 |
+
"train_runtime": 4175.1629,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"train_samples": 63353,
|
| 7 |
+
"train_samples_per_second": 47.883,
|
| 8 |
+
"train_steps_per_second": 0.187
|
| 9 |
}
|
runs/Jun10_02-23-10_48ddfe8e991f/events.out.tfevents.1717986227.48ddfe8e991f.24991.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5119fd0853b4e3e482596fc5841bf36b582f1f8f34c1c401585c1877c033ad9a
|
| 3 |
+
size 43037
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 14.
|
| 3 |
-
"total_flos": 1.
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 63353,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 14.857142857142858,
|
| 3 |
+
"total_flos": 1.2277516310308454e+18,
|
| 4 |
+
"train_loss": 1.077159938445458,
|
| 5 |
+
"train_runtime": 4175.1629,
|
| 6 |
"train_samples": 63353,
|
| 7 |
+
"train_samples_per_second": 47.883,
|
| 8 |
+
"train_steps_per_second": 0.187
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,2341 +1,1261 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 14.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 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.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 3.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 3.
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 2.
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 3.
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm": 2.
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 2.
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm": 2.
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 2.
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm": 2.
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 2.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm": 1.
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss":
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss":
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss":
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"
|
| 91 |
-
"
|
| 92 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch":
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
-
"learning_rate":
|
| 99 |
-
"loss": 1.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch":
|
| 104 |
-
"grad_norm": 0.
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch":
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch":
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"loss": 1.
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch":
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate": 0.
|
| 127 |
-
"loss": 1.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch":
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
-
"learning_rate": 0.
|
| 134 |
-
"loss": 1.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch":
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
-
"learning_rate": 0.
|
| 141 |
-
"loss": 1.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch":
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
-
"learning_rate": 0.
|
| 148 |
-
"loss": 1.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch":
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
-
"learning_rate": 0.
|
| 155 |
-
"loss": 1.
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0
|
| 160 |
-
"
|
| 161 |
-
"
|
| 162 |
-
"
|
| 163 |
-
"
|
| 164 |
-
"step": 104
|
| 165 |
},
|
| 166 |
{
|
| 167 |
-
"epoch":
|
| 168 |
-
"
|
| 169 |
-
"
|
| 170 |
-
"
|
|
|
|
| 171 |
"step": 105
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch":
|
| 175 |
-
"grad_norm": 0.
|
| 176 |
-
"learning_rate": 0.
|
| 177 |
-
"loss": 1.
|
| 178 |
"step": 110
|
| 179 |
},
|
| 180 |
{
|
| 181 |
-
"epoch":
|
| 182 |
-
"grad_norm": 0.
|
| 183 |
-
"learning_rate": 0.
|
| 184 |
-
"loss": 1.
|
| 185 |
"step": 115
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch":
|
| 189 |
-
"grad_norm": 0.
|
| 190 |
-
"learning_rate": 0.
|
| 191 |
-
"loss": 1.
|
| 192 |
"step": 120
|
| 193 |
},
|
| 194 |
{
|
| 195 |
-
"epoch":
|
| 196 |
-
"grad_norm":
|
| 197 |
-
"learning_rate": 0.
|
| 198 |
-
"loss": 1.
|
| 199 |
"step": 125
|
| 200 |
},
|
| 201 |
{
|
| 202 |
-
"epoch":
|
| 203 |
-
"grad_norm": 0.
|
| 204 |
-
"learning_rate": 0.
|
| 205 |
-
"loss": 1.
|
| 206 |
"step": 130
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch":
|
| 210 |
-
"grad_norm": 0.
|
| 211 |
-
"learning_rate": 0.
|
| 212 |
-
"loss": 1.
|
| 213 |
"step": 135
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"epoch":
|
| 217 |
-
"grad_norm": 0.
|
| 218 |
-
"learning_rate": 0.
|
| 219 |
-
"loss": 1.
|
| 220 |
"step": 140
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"epoch":
|
| 224 |
-
"grad_norm": 0.
|
| 225 |
-
"learning_rate": 0.
|
| 226 |
-
"loss": 1.
|
| 227 |
"step": 145
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch":
|
| 231 |
-
"grad_norm": 0.
|
| 232 |
-
"learning_rate": 0.
|
| 233 |
-
"loss": 1.
|
| 234 |
"step": 150
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch":
|
| 238 |
-
"grad_norm": 0.
|
| 239 |
-
"learning_rate": 0.
|
| 240 |
-
"loss": 1.
|
| 241 |
"step": 155
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch":
|
| 245 |
-
"
|
| 246 |
-
"
|
| 247 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 248 |
"step": 160
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"grad_norm":
|
| 253 |
-
"learning_rate": 0.
|
| 254 |
-
"loss": 1.
|
| 255 |
"step": 165
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch":
|
| 259 |
-
"grad_norm": 0.
|
| 260 |
-
"learning_rate": 0.
|
| 261 |
-
"loss": 1.
|
| 262 |
"step": 170
|
| 263 |
},
|
| 264 |
{
|
| 265 |
-
"epoch":
|
| 266 |
-
"grad_norm": 0.
|
| 267 |
-
"learning_rate": 0.
|
| 268 |
-
"loss": 1.
|
| 269 |
"step": 175
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch":
|
| 273 |
-
"grad_norm": 0.
|
| 274 |
-
"learning_rate": 0.
|
| 275 |
-
"loss": 1.
|
| 276 |
"step": 180
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch":
|
| 280 |
-
"grad_norm": 0.
|
| 281 |
-
"learning_rate": 0.
|
| 282 |
-
"loss": 1.
|
| 283 |
"step": 185
|
| 284 |
},
|
| 285 |
{
|
| 286 |
-
"epoch":
|
| 287 |
-
"grad_norm": 0.
|
| 288 |
-
"learning_rate": 0.
|
| 289 |
-
"loss": 1.
|
| 290 |
"step": 190
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch":
|
| 294 |
-
"grad_norm": 0.
|
| 295 |
-
"learning_rate": 0.
|
| 296 |
-
"loss": 1.
|
| 297 |
"step": 195
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch":
|
| 301 |
-
"grad_norm": 0.
|
| 302 |
-
"learning_rate": 0.
|
| 303 |
-
"loss": 1.
|
| 304 |
"step": 200
|
| 305 |
},
|
| 306 |
{
|
| 307 |
-
"epoch":
|
| 308 |
-
"grad_norm": 0.
|
| 309 |
-
"learning_rate": 0.
|
| 310 |
-
"loss": 1.
|
| 311 |
"step": 205
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch":
|
| 315 |
-
"
|
| 316 |
-
"
|
| 317 |
-
"
|
| 318 |
-
"
|
| 319 |
-
"step": 209
|
| 320 |
},
|
| 321 |
{
|
| 322 |
-
"epoch":
|
| 323 |
-
"
|
| 324 |
-
"
|
| 325 |
-
"
|
|
|
|
| 326 |
"step": 210
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch":
|
| 330 |
-
"grad_norm": 0.
|
| 331 |
-
"learning_rate": 0.
|
| 332 |
-
"loss": 1.
|
| 333 |
"step": 215
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch":
|
| 337 |
-
"grad_norm": 0.
|
| 338 |
-
"learning_rate": 0.
|
| 339 |
-
"loss": 1.
|
| 340 |
"step": 220
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"epoch":
|
| 344 |
-
"grad_norm": 0.
|
| 345 |
-
"learning_rate": 0.
|
| 346 |
-
"loss": 1.
|
| 347 |
"step": 225
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch":
|
| 351 |
-
"grad_norm": 0.
|
| 352 |
-
"learning_rate": 0.
|
| 353 |
-
"loss": 1.
|
| 354 |
"step": 230
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch":
|
| 358 |
-
"grad_norm": 0.
|
| 359 |
-
"learning_rate": 0.
|
| 360 |
-
"loss": 1.
|
| 361 |
"step": 235
|
| 362 |
},
|
| 363 |
{
|
| 364 |
-
"epoch":
|
| 365 |
-
"grad_norm": 0.
|
| 366 |
-
"learning_rate": 0.
|
| 367 |
-
"loss": 1.
|
| 368 |
"step": 240
|
| 369 |
},
|
| 370 |
{
|
| 371 |
-
"epoch":
|
| 372 |
-
"grad_norm": 0.
|
| 373 |
-
"learning_rate": 0.
|
| 374 |
-
"loss": 1.
|
| 375 |
"step": 245
|
| 376 |
},
|
| 377 |
{
|
| 378 |
-
"epoch":
|
| 379 |
-
"grad_norm": 0.
|
| 380 |
-
"learning_rate": 0.
|
| 381 |
-
"loss": 1.
|
| 382 |
"step": 250
|
| 383 |
},
|
| 384 |
{
|
| 385 |
-
"epoch":
|
| 386 |
-
"grad_norm": 0.
|
| 387 |
-
"learning_rate": 0.
|
| 388 |
-
"loss": 1.
|
| 389 |
"step": 255
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch":
|
| 393 |
-
"grad_norm": 0.
|
| 394 |
-
"learning_rate": 0.
|
| 395 |
-
"loss": 1.
|
| 396 |
"step": 260
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch":
|
| 400 |
-
"
|
| 401 |
-
"
|
| 402 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 403 |
"step": 265
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"epoch":
|
| 407 |
-
"grad_norm": 0.
|
| 408 |
-
"learning_rate": 0.
|
| 409 |
-
"loss":
|
| 410 |
"step": 270
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"epoch":
|
| 414 |
-
"grad_norm":
|
| 415 |
-
"learning_rate": 0.
|
| 416 |
-
"loss":
|
| 417 |
"step": 275
|
| 418 |
},
|
| 419 |
{
|
| 420 |
-
"epoch":
|
| 421 |
-
"grad_norm": 0.
|
| 422 |
-
"learning_rate": 0.
|
| 423 |
-
"loss":
|
| 424 |
"step": 280
|
| 425 |
},
|
| 426 |
{
|
| 427 |
-
"epoch":
|
| 428 |
-
"grad_norm": 0.
|
| 429 |
-
"learning_rate": 0.
|
| 430 |
-
"loss":
|
| 431 |
"step": 285
|
| 432 |
},
|
| 433 |
{
|
| 434 |
-
"epoch":
|
| 435 |
-
"grad_norm": 0.
|
| 436 |
-
"learning_rate": 0.
|
| 437 |
-
"loss":
|
| 438 |
"step": 290
|
| 439 |
},
|
| 440 |
{
|
| 441 |
-
"epoch":
|
| 442 |
-
"grad_norm": 0.
|
| 443 |
-
"learning_rate": 0.
|
| 444 |
-
"loss":
|
| 445 |
"step": 295
|
| 446 |
},
|
| 447 |
{
|
| 448 |
-
"epoch":
|
| 449 |
-
"grad_norm": 0.
|
| 450 |
-
"learning_rate": 0.
|
| 451 |
-
"loss":
|
| 452 |
"step": 300
|
| 453 |
},
|
| 454 |
{
|
| 455 |
-
"epoch":
|
| 456 |
-
"grad_norm": 0.
|
| 457 |
-
"learning_rate": 0.
|
| 458 |
-
"loss":
|
| 459 |
"step": 305
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch":
|
| 463 |
-
"grad_norm": 0.
|
| 464 |
-
"learning_rate": 0.
|
| 465 |
-
"loss":
|
| 466 |
"step": 310
|
| 467 |
},
|
| 468 |
{
|
| 469 |
-
"epoch":
|
| 470 |
-
"
|
| 471 |
-
"
|
| 472 |
-
"
|
| 473 |
-
"
|
| 474 |
-
"step": 313
|
| 475 |
},
|
| 476 |
{
|
| 477 |
-
"epoch":
|
| 478 |
-
"
|
| 479 |
-
"
|
| 480 |
-
"
|
|
|
|
| 481 |
"step": 315
|
| 482 |
},
|
| 483 |
{
|
| 484 |
-
"epoch":
|
| 485 |
-
"grad_norm": 0.
|
| 486 |
-
"learning_rate": 0.
|
| 487 |
-
"loss":
|
| 488 |
"step": 320
|
| 489 |
},
|
| 490 |
{
|
| 491 |
-
"epoch":
|
| 492 |
-
"grad_norm": 0.
|
| 493 |
-
"learning_rate": 0.
|
| 494 |
-
"loss":
|
| 495 |
"step": 325
|
| 496 |
},
|
| 497 |
{
|
| 498 |
-
"epoch":
|
| 499 |
-
"grad_norm": 0.
|
| 500 |
-
"learning_rate": 0.
|
| 501 |
-
"loss":
|
| 502 |
"step": 330
|
| 503 |
},
|
| 504 |
{
|
| 505 |
-
"epoch":
|
| 506 |
-
"grad_norm": 0.
|
| 507 |
-
"learning_rate": 0.
|
| 508 |
-
"loss":
|
| 509 |
"step": 335
|
| 510 |
},
|
| 511 |
{
|
| 512 |
-
"epoch":
|
| 513 |
-
"grad_norm": 0.
|
| 514 |
-
"learning_rate": 0.
|
| 515 |
-
"loss":
|
| 516 |
"step": 340
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch":
|
| 520 |
-
"grad_norm": 0.
|
| 521 |
-
"learning_rate": 0.
|
| 522 |
-
"loss":
|
| 523 |
"step": 345
|
| 524 |
},
|
| 525 |
{
|
| 526 |
-
"epoch":
|
| 527 |
-
"grad_norm": 0.
|
| 528 |
-
"learning_rate": 0.
|
| 529 |
-
"loss":
|
| 530 |
"step": 350
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch":
|
| 534 |
-
"grad_norm": 0.
|
| 535 |
-
"learning_rate": 0.
|
| 536 |
-
"loss": 0.
|
| 537 |
"step": 355
|
| 538 |
},
|
| 539 |
{
|
| 540 |
-
"epoch":
|
| 541 |
-
"grad_norm": 0.
|
| 542 |
-
"learning_rate": 0.
|
| 543 |
-
"loss": 0.
|
| 544 |
"step": 360
|
| 545 |
},
|
| 546 |
{
|
| 547 |
-
"epoch":
|
| 548 |
-
"grad_norm": 0.
|
| 549 |
-
"learning_rate": 0.
|
| 550 |
-
"loss":
|
| 551 |
"step": 365
|
| 552 |
},
|
| 553 |
{
|
| 554 |
-
"epoch":
|
| 555 |
-
"
|
| 556 |
-
"
|
| 557 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 558 |
"step": 370
|
| 559 |
},
|
| 560 |
{
|
| 561 |
-
"epoch":
|
| 562 |
-
"grad_norm": 0.
|
| 563 |
-
"learning_rate": 0.
|
| 564 |
-
"loss": 0.
|
| 565 |
"step": 375
|
| 566 |
},
|
| 567 |
{
|
| 568 |
-
"epoch":
|
| 569 |
-
"grad_norm": 0.
|
| 570 |
-
"learning_rate": 0.
|
| 571 |
-
"loss":
|
| 572 |
"step": 380
|
| 573 |
},
|
| 574 |
{
|
| 575 |
-
"epoch":
|
| 576 |
-
"grad_norm": 0.
|
| 577 |
-
"learning_rate": 0.
|
| 578 |
-
"loss":
|
| 579 |
"step": 385
|
| 580 |
},
|
| 581 |
{
|
| 582 |
-
"epoch":
|
| 583 |
-
"grad_norm": 0.
|
| 584 |
-
"learning_rate": 0.
|
| 585 |
-
"loss": 0.
|
| 586 |
"step": 390
|
| 587 |
},
|
| 588 |
{
|
| 589 |
-
"epoch":
|
| 590 |
-
"grad_norm": 0.
|
| 591 |
-
"learning_rate": 0.
|
| 592 |
-
"loss": 0.
|
| 593 |
"step": 395
|
| 594 |
},
|
| 595 |
{
|
| 596 |
-
"epoch":
|
| 597 |
-
"grad_norm": 0.
|
| 598 |
-
"learning_rate": 0.
|
| 599 |
-
"loss": 0.
|
| 600 |
"step": 400
|
| 601 |
},
|
| 602 |
{
|
| 603 |
-
"epoch":
|
| 604 |
-
"grad_norm": 0.
|
| 605 |
-
"learning_rate": 0.
|
| 606 |
-
"loss": 0.
|
| 607 |
"step": 405
|
| 608 |
},
|
| 609 |
{
|
| 610 |
-
"epoch":
|
| 611 |
-
"grad_norm": 0.
|
| 612 |
-
"learning_rate": 0.
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 410
|
| 615 |
},
|
| 616 |
{
|
| 617 |
-
"epoch":
|
| 618 |
-
"grad_norm": 0.
|
| 619 |
-
"learning_rate": 0.
|
| 620 |
-
"loss": 0.
|
| 621 |
"step": 415
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"epoch":
|
| 625 |
-
"
|
| 626 |
-
"
|
| 627 |
-
"
|
| 628 |
-
"
|
| 629 |
-
"step": 418
|
| 630 |
},
|
| 631 |
{
|
| 632 |
-
"epoch":
|
| 633 |
-
"
|
| 634 |
-
"
|
| 635 |
-
"
|
|
|
|
| 636 |
"step": 420
|
| 637 |
},
|
| 638 |
{
|
| 639 |
-
"epoch":
|
| 640 |
-
"grad_norm": 0.
|
| 641 |
-
"learning_rate": 0.
|
| 642 |
-
"loss": 0.
|
| 643 |
"step": 425
|
| 644 |
},
|
| 645 |
{
|
| 646 |
-
"epoch":
|
| 647 |
-
"grad_norm": 0.
|
| 648 |
-
"learning_rate":
|
| 649 |
-
"loss": 0.
|
| 650 |
"step": 430
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch":
|
| 654 |
-
"grad_norm": 0.
|
| 655 |
-
"learning_rate":
|
| 656 |
-
"loss": 0.
|
| 657 |
"step": 435
|
| 658 |
},
|
| 659 |
{
|
| 660 |
-
"epoch":
|
| 661 |
-
"grad_norm": 0.
|
| 662 |
-
"learning_rate":
|
| 663 |
-
"loss": 0.
|
| 664 |
"step": 440
|
| 665 |
},
|
| 666 |
{
|
| 667 |
-
"epoch":
|
| 668 |
-
"grad_norm": 0.
|
| 669 |
-
"learning_rate":
|
| 670 |
-
"loss": 0.
|
| 671 |
"step": 445
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"epoch":
|
| 675 |
-
"grad_norm": 0.
|
| 676 |
-
"learning_rate":
|
| 677 |
-
"loss": 0.
|
| 678 |
"step": 450
|
| 679 |
},
|
| 680 |
{
|
| 681 |
-
"epoch":
|
| 682 |
-
"grad_norm": 0.
|
| 683 |
-
"learning_rate":
|
| 684 |
-
"loss": 0.
|
| 685 |
"step": 455
|
| 686 |
},
|
| 687 |
{
|
| 688 |
-
"epoch":
|
| 689 |
-
"grad_norm": 0.
|
| 690 |
-
"learning_rate":
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 460
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch":
|
| 696 |
-
"grad_norm": 0.
|
| 697 |
-
"learning_rate":
|
| 698 |
-
"loss": 0.
|
| 699 |
"step": 465
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch":
|
| 703 |
-
"grad_norm": 0.
|
| 704 |
-
"learning_rate":
|
| 705 |
-
"loss": 0.
|
| 706 |
"step": 470
|
| 707 |
},
|
| 708 |
{
|
| 709 |
-
"epoch":
|
| 710 |
-
"
|
| 711 |
-
"
|
| 712 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 713 |
"step": 475
|
| 714 |
},
|
| 715 |
{
|
| 716 |
-
"epoch":
|
| 717 |
-
"grad_norm": 0.
|
| 718 |
-
"learning_rate":
|
| 719 |
-
"loss": 0.
|
| 720 |
"step": 480
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"epoch":
|
| 724 |
-
"grad_norm": 0.
|
| 725 |
-
"learning_rate":
|
| 726 |
-
"loss": 0.
|
| 727 |
"step": 485
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"epoch":
|
| 731 |
-
"grad_norm": 0.
|
| 732 |
-
"learning_rate": 0.0001733486566452446,
|
| 733 |
-
"loss": 0.9705,
|
| 734 |
-
"step": 490
|
| 735 |
-
},
|
| 736 |
-
{
|
| 737 |
-
"epoch": 4.7368421052631575,
|
| 738 |
-
"grad_norm": 0.353515625,
|
| 739 |
-
"learning_rate": 0.00017258362865354426,
|
| 740 |
-
"loss": 0.985,
|
| 741 |
-
"step": 495
|
| 742 |
-
},
|
| 743 |
-
{
|
| 744 |
-
"epoch": 4.784688995215311,
|
| 745 |
-
"grad_norm": 0.380859375,
|
| 746 |
-
"learning_rate": 0.0001718095153532274,
|
| 747 |
-
"loss": 0.9817,
|
| 748 |
-
"step": 500
|
| 749 |
-
},
|
| 750 |
-
{
|
| 751 |
-
"epoch": 4.832535885167464,
|
| 752 |
-
"grad_norm": 0.474609375,
|
| 753 |
-
"learning_rate": 0.00017102641364022457,
|
| 754 |
-
"loss": 0.9686,
|
| 755 |
-
"step": 505
|
| 756 |
-
},
|
| 757 |
-
{
|
| 758 |
-
"epoch": 4.880382775119617,
|
| 759 |
-
"grad_norm": 0.384765625,
|
| 760 |
-
"learning_rate": 0.00017023442153554777,
|
| 761 |
-
"loss": 0.9838,
|
| 762 |
-
"step": 510
|
| 763 |
-
},
|
| 764 |
-
{
|
| 765 |
-
"epoch": 4.92822966507177,
|
| 766 |
-
"grad_norm": 0.53515625,
|
| 767 |
-
"learning_rate": 0.00016943363817302135,
|
| 768 |
-
"loss": 0.9774,
|
| 769 |
-
"step": 515
|
| 770 |
-
},
|
| 771 |
-
{
|
| 772 |
-
"epoch": 4.976076555023924,
|
| 773 |
-
"grad_norm": 0.50390625,
|
| 774 |
-
"learning_rate": 0.0001686241637868734,
|
| 775 |
-
"loss": 0.9775,
|
| 776 |
-
"step": 520
|
| 777 |
-
},
|
| 778 |
-
{
|
| 779 |
-
"epoch": 4.9952153110047846,
|
| 780 |
-
"eval_loss": 2.476422071456909,
|
| 781 |
-
"eval_runtime": 0.6478,
|
| 782 |
-
"eval_samples_per_second": 15.436,
|
| 783 |
-
"eval_steps_per_second": 1.544,
|
| 784 |
-
"step": 522
|
| 785 |
-
},
|
| 786 |
-
{
|
| 787 |
-
"epoch": 5.023923444976076,
|
| 788 |
-
"grad_norm": 0.625,
|
| 789 |
-
"learning_rate": 0.0001678060996991891,
|
| 790 |
-
"loss": 0.9684,
|
| 791 |
-
"step": 525
|
| 792 |
-
},
|
| 793 |
-
{
|
| 794 |
-
"epoch": 5.07177033492823,
|
| 795 |
-
"grad_norm": 0.33203125,
|
| 796 |
-
"learning_rate": 0.00016697954830722868,
|
| 797 |
-
"loss": 0.9614,
|
| 798 |
-
"step": 530
|
| 799 |
-
},
|
| 800 |
-
{
|
| 801 |
-
"epoch": 5.119617224880383,
|
| 802 |
-
"grad_norm": 0.376953125,
|
| 803 |
-
"learning_rate": 0.00016614461307061,
|
| 804 |
-
"loss": 0.963,
|
| 805 |
-
"step": 535
|
| 806 |
-
},
|
| 807 |
-
{
|
| 808 |
-
"epoch": 5.167464114832536,
|
| 809 |
-
"grad_norm": 0.439453125,
|
| 810 |
-
"learning_rate": 0.0001653013984983585,
|
| 811 |
-
"loss": 0.9558,
|
| 812 |
-
"step": 540
|
| 813 |
-
},
|
| 814 |
-
{
|
| 815 |
-
"epoch": 5.215311004784689,
|
| 816 |
-
"grad_norm": 0.375,
|
| 817 |
-
"learning_rate": 0.00016445001013582608,
|
| 818 |
-
"loss": 0.9533,
|
| 819 |
-
"step": 545
|
| 820 |
-
},
|
| 821 |
-
{
|
| 822 |
-
"epoch": 5.2631578947368425,
|
| 823 |
-
"grad_norm": 0.40234375,
|
| 824 |
-
"learning_rate": 0.0001635905545514795,
|
| 825 |
-
"loss": 0.9664,
|
| 826 |
-
"step": 550
|
| 827 |
-
},
|
| 828 |
-
{
|
| 829 |
-
"epoch": 5.311004784688995,
|
| 830 |
-
"grad_norm": 0.380859375,
|
| 831 |
-
"learning_rate": 0.00016272313932356162,
|
| 832 |
-
"loss": 0.9552,
|
| 833 |
-
"step": 555
|
| 834 |
-
},
|
| 835 |
-
{
|
| 836 |
-
"epoch": 5.358851674641148,
|
| 837 |
-
"grad_norm": 0.416015625,
|
| 838 |
-
"learning_rate": 0.0001618478730266255,
|
| 839 |
-
"loss": 0.9505,
|
| 840 |
-
"step": 560
|
| 841 |
-
},
|
| 842 |
-
{
|
| 843 |
-
"epoch": 5.4066985645933014,
|
| 844 |
-
"grad_norm": 0.39453125,
|
| 845 |
-
"learning_rate": 0.00016096486521794434,
|
| 846 |
-
"loss": 0.964,
|
| 847 |
-
"step": 565
|
| 848 |
-
},
|
| 849 |
-
{
|
| 850 |
-
"epoch": 5.454545454545454,
|
| 851 |
-
"grad_norm": 0.416015625,
|
| 852 |
-
"learning_rate": 0.0001600742264237979,
|
| 853 |
-
"loss": 0.9558,
|
| 854 |
-
"step": 570
|
| 855 |
-
},
|
| 856 |
-
{
|
| 857 |
-
"epoch": 5.502392344497608,
|
| 858 |
-
"grad_norm": 0.5625,
|
| 859 |
-
"learning_rate": 0.0001591760681256382,
|
| 860 |
-
"loss": 0.9573,
|
| 861 |
-
"step": 575
|
| 862 |
-
},
|
| 863 |
-
{
|
| 864 |
-
"epoch": 5.55023923444976,
|
| 865 |
-
"grad_norm": 0.3515625,
|
| 866 |
-
"learning_rate": 0.00015827050274613513,
|
| 867 |
-
"loss": 0.9587,
|
| 868 |
-
"step": 580
|
| 869 |
-
},
|
| 870 |
-
{
|
| 871 |
-
"epoch": 5.598086124401914,
|
| 872 |
-
"grad_norm": 0.421875,
|
| 873 |
-
"learning_rate": 0.0001573576436351046,
|
| 874 |
-
"loss": 0.9638,
|
| 875 |
-
"step": 585
|
| 876 |
-
},
|
| 877 |
-
{
|
| 878 |
-
"epoch": 5.645933014354067,
|
| 879 |
-
"grad_norm": 0.4140625,
|
| 880 |
-
"learning_rate": 0.0001564376050553205,
|
| 881 |
-
"loss": 0.9513,
|
| 882 |
-
"step": 590
|
| 883 |
-
},
|
| 884 |
-
{
|
| 885 |
-
"epoch": 5.69377990430622,
|
| 886 |
-
"grad_norm": 0.62109375,
|
| 887 |
-
"learning_rate": 0.0001555105021682123,
|
| 888 |
-
"loss": 0.9627,
|
| 889 |
-
"step": 595
|
| 890 |
-
},
|
| 891 |
-
{
|
| 892 |
-
"epoch": 5.741626794258373,
|
| 893 |
-
"grad_norm": 0.62890625,
|
| 894 |
-
"learning_rate": 0.00015457645101945046,
|
| 895 |
-
"loss": 0.9546,
|
| 896 |
-
"step": 600
|
| 897 |
-
},
|
| 898 |
-
{
|
| 899 |
-
"epoch": 5.7894736842105265,
|
| 900 |
-
"grad_norm": 0.361328125,
|
| 901 |
-
"learning_rate": 0.00015363556852442085,
|
| 902 |
-
"loss": 0.9481,
|
| 903 |
-
"step": 605
|
| 904 |
-
},
|
| 905 |
-
{
|
| 906 |
-
"epoch": 5.837320574162679,
|
| 907 |
-
"grad_norm": 0.462890625,
|
| 908 |
-
"learning_rate": 0.00015268797245359035,
|
| 909 |
-
"loss": 0.9629,
|
| 910 |
-
"step": 610
|
| 911 |
-
},
|
| 912 |
-
{
|
| 913 |
-
"epoch": 5.885167464114833,
|
| 914 |
-
"grad_norm": 0.33984375,
|
| 915 |
-
"learning_rate": 0.00015173378141776568,
|
| 916 |
-
"loss": 0.9519,
|
| 917 |
-
"step": 615
|
| 918 |
-
},
|
| 919 |
-
{
|
| 920 |
-
"epoch": 5.9330143540669855,
|
| 921 |
-
"grad_norm": 0.423828125,
|
| 922 |
-
"learning_rate": 0.0001507731148532468,
|
| 923 |
-
"loss": 0.9584,
|
| 924 |
-
"step": 620
|
| 925 |
-
},
|
| 926 |
-
{
|
| 927 |
-
"epoch": 5.980861244019139,
|
| 928 |
-
"grad_norm": 0.3828125,
|
| 929 |
-
"learning_rate": 0.00014980609300687683,
|
| 930 |
-
"loss": 0.9584,
|
| 931 |
-
"step": 625
|
| 932 |
-
},
|
| 933 |
-
{
|
| 934 |
-
"epoch": 6.0,
|
| 935 |
-
"eval_loss": 2.487025737762451,
|
| 936 |
-
"eval_runtime": 0.5347,
|
| 937 |
-
"eval_samples_per_second": 18.7,
|
| 938 |
-
"eval_steps_per_second": 1.87,
|
| 939 |
-
"step": 627
|
| 940 |
-
},
|
| 941 |
-
{
|
| 942 |
-
"epoch": 6.028708133971292,
|
| 943 |
-
"grad_norm": 0.361328125,
|
| 944 |
-
"learning_rate": 0.00014883283692099112,
|
| 945 |
-
"loss": 0.9435,
|
| 946 |
-
"step": 630
|
| 947 |
-
},
|
| 948 |
-
{
|
| 949 |
-
"epoch": 6.076555023923445,
|
| 950 |
-
"grad_norm": 0.388671875,
|
| 951 |
-
"learning_rate": 0.000147853468418266,
|
| 952 |
-
"loss": 0.9462,
|
| 953 |
-
"step": 635
|
| 954 |
-
},
|
| 955 |
-
{
|
| 956 |
-
"epoch": 6.124401913875598,
|
| 957 |
-
"grad_norm": 0.443359375,
|
| 958 |
-
"learning_rate": 0.00014686811008647038,
|
| 959 |
-
"loss": 0.9372,
|
| 960 |
-
"step": 640
|
| 961 |
-
},
|
| 962 |
-
{
|
| 963 |
-
"epoch": 6.172248803827751,
|
| 964 |
-
"grad_norm": 0.37109375,
|
| 965 |
-
"learning_rate": 0.00014587688526312143,
|
| 966 |
-
"loss": 0.9405,
|
| 967 |
-
"step": 645
|
| 968 |
-
},
|
| 969 |
-
{
|
| 970 |
-
"epoch": 6.220095693779904,
|
| 971 |
-
"grad_norm": 0.474609375,
|
| 972 |
-
"learning_rate": 0.00014487991802004623,
|
| 973 |
-
"loss": 0.942,
|
| 974 |
-
"step": 650
|
| 975 |
-
},
|
| 976 |
-
{
|
| 977 |
-
"epoch": 6.267942583732057,
|
| 978 |
-
"grad_norm": 0.3984375,
|
| 979 |
-
"learning_rate": 0.00014387733314785193,
|
| 980 |
-
"loss": 0.9495,
|
| 981 |
-
"step": 655
|
| 982 |
-
},
|
| 983 |
-
{
|
| 984 |
-
"epoch": 6.315789473684211,
|
| 985 |
-
"grad_norm": 0.38671875,
|
| 986 |
-
"learning_rate": 0.00014286925614030542,
|
| 987 |
-
"loss": 0.9415,
|
| 988 |
-
"step": 660
|
| 989 |
-
},
|
| 990 |
-
{
|
| 991 |
-
"epoch": 6.363636363636363,
|
| 992 |
-
"grad_norm": 0.390625,
|
| 993 |
-
"learning_rate": 0.00014185581317862546,
|
| 994 |
-
"loss": 0.94,
|
| 995 |
-
"step": 665
|
| 996 |
-
},
|
| 997 |
-
{
|
| 998 |
-
"epoch": 6.411483253588517,
|
| 999 |
-
"grad_norm": 0.48046875,
|
| 1000 |
-
"learning_rate": 0.00014083713111568842,
|
| 1001 |
-
"loss": 0.9344,
|
| 1002 |
-
"step": 670
|
| 1003 |
-
},
|
| 1004 |
-
{
|
| 1005 |
-
"epoch": 6.45933014354067,
|
| 1006 |
-
"grad_norm": 0.439453125,
|
| 1007 |
-
"learning_rate": 0.0001398133374601501,
|
| 1008 |
-
"loss": 0.9438,
|
| 1009 |
-
"step": 675
|
| 1010 |
-
},
|
| 1011 |
-
{
|
| 1012 |
-
"epoch": 6.507177033492823,
|
| 1013 |
-
"grad_norm": 0.333984375,
|
| 1014 |
-
"learning_rate": 0.0001387845603604855,
|
| 1015 |
-
"loss": 0.941,
|
| 1016 |
-
"step": 680
|
| 1017 |
-
},
|
| 1018 |
-
{
|
| 1019 |
-
"epoch": 6.555023923444976,
|
| 1020 |
-
"grad_norm": 0.5078125,
|
| 1021 |
-
"learning_rate": 0.00013775092858894837,
|
| 1022 |
-
"loss": 0.9433,
|
| 1023 |
-
"step": 685
|
| 1024 |
-
},
|
| 1025 |
-
{
|
| 1026 |
-
"epoch": 6.6028708133971294,
|
| 1027 |
-
"grad_norm": 0.3515625,
|
| 1028 |
-
"learning_rate": 0.00013671257152545277,
|
| 1029 |
-
"loss": 0.9433,
|
| 1030 |
-
"step": 690
|
| 1031 |
-
},
|
| 1032 |
-
{
|
| 1033 |
-
"epoch": 6.650717703349282,
|
| 1034 |
-
"grad_norm": 0.451171875,
|
| 1035 |
-
"learning_rate": 0.00013566961914137867,
|
| 1036 |
-
"loss": 0.9435,
|
| 1037 |
-
"step": 695
|
| 1038 |
-
},
|
| 1039 |
-
{
|
| 1040 |
-
"epoch": 6.698564593301436,
|
| 1041 |
-
"grad_norm": 0.41015625,
|
| 1042 |
-
"learning_rate": 0.00013462220198330328,
|
| 1043 |
-
"loss": 0.9431,
|
| 1044 |
-
"step": 700
|
| 1045 |
-
},
|
| 1046 |
-
{
|
| 1047 |
-
"epoch": 6.746411483253588,
|
| 1048 |
-
"grad_norm": 0.44140625,
|
| 1049 |
-
"learning_rate": 0.0001335704511566605,
|
| 1050 |
-
"loss": 0.9409,
|
| 1051 |
-
"step": 705
|
| 1052 |
-
},
|
| 1053 |
-
{
|
| 1054 |
-
"epoch": 6.794258373205742,
|
| 1055 |
-
"grad_norm": 0.546875,
|
| 1056 |
-
"learning_rate": 0.0001325144983093305,
|
| 1057 |
-
"loss": 0.9437,
|
| 1058 |
-
"step": 710
|
| 1059 |
-
},
|
| 1060 |
-
{
|
| 1061 |
-
"epoch": 6.842105263157895,
|
| 1062 |
-
"grad_norm": 0.412109375,
|
| 1063 |
-
"learning_rate": 0.00013145447561516138,
|
| 1064 |
-
"loss": 0.9491,
|
| 1065 |
-
"step": 715
|
| 1066 |
-
},
|
| 1067 |
-
{
|
| 1068 |
-
"epoch": 6.889952153110048,
|
| 1069 |
-
"grad_norm": 0.38671875,
|
| 1070 |
-
"learning_rate": 0.0001303905157574247,
|
| 1071 |
-
"loss": 0.9445,
|
| 1072 |
-
"step": 720
|
| 1073 |
-
},
|
| 1074 |
-
{
|
| 1075 |
-
"epoch": 6.937799043062201,
|
| 1076 |
-
"grad_norm": 0.380859375,
|
| 1077 |
-
"learning_rate": 0.00012932275191220776,
|
| 1078 |
-
"loss": 0.9315,
|
| 1079 |
-
"step": 725
|
| 1080 |
-
},
|
| 1081 |
-
{
|
| 1082 |
-
"epoch": 6.985645933014354,
|
| 1083 |
-
"grad_norm": 0.36328125,
|
| 1084 |
-
"learning_rate": 0.0001282513177317437,
|
| 1085 |
-
"loss": 0.9368,
|
| 1086 |
-
"step": 730
|
| 1087 |
-
},
|
| 1088 |
-
{
|
| 1089 |
-
"epoch": 6.9952153110047846,
|
| 1090 |
-
"eval_loss": 2.497544765472412,
|
| 1091 |
-
"eval_runtime": 0.6485,
|
| 1092 |
-
"eval_samples_per_second": 15.419,
|
| 1093 |
-
"eval_steps_per_second": 1.542,
|
| 1094 |
-
"step": 731
|
| 1095 |
-
},
|
| 1096 |
-
{
|
| 1097 |
-
"epoch": 7.033492822966507,
|
| 1098 |
-
"grad_norm": 0.69921875,
|
| 1099 |
-
"learning_rate": 0.00012717634732768243,
|
| 1100 |
-
"loss": 0.9238,
|
| 1101 |
-
"step": 735
|
| 1102 |
-
},
|
| 1103 |
-
{
|
| 1104 |
-
"epoch": 7.08133971291866,
|
| 1105 |
-
"grad_norm": 0.39453125,
|
| 1106 |
-
"learning_rate": 0.00012609797525430373,
|
| 1107 |
-
"loss": 0.9235,
|
| 1108 |
-
"step": 740
|
| 1109 |
-
},
|
| 1110 |
-
{
|
| 1111 |
-
"epoch": 7.1291866028708135,
|
| 1112 |
-
"grad_norm": 0.3984375,
|
| 1113 |
-
"learning_rate": 0.00012501633649167495,
|
| 1114 |
-
"loss": 0.9148,
|
| 1115 |
-
"step": 745
|
| 1116 |
-
},
|
| 1117 |
-
{
|
| 1118 |
-
"epoch": 7.177033492822966,
|
| 1119 |
-
"grad_norm": 0.455078125,
|
| 1120 |
-
"learning_rate": 0.0001239315664287558,
|
| 1121 |
-
"loss": 0.9182,
|
| 1122 |
-
"step": 750
|
| 1123 |
-
},
|
| 1124 |
-
{
|
| 1125 |
-
"epoch": 7.22488038277512,
|
| 1126 |
-
"grad_norm": 0.435546875,
|
| 1127 |
-
"learning_rate": 0.00012284380084645139,
|
| 1128 |
-
"loss": 0.9237,
|
| 1129 |
-
"step": 755
|
| 1130 |
-
},
|
| 1131 |
-
{
|
| 1132 |
-
"epoch": 7.2727272727272725,
|
| 1133 |
-
"grad_norm": 0.443359375,
|
| 1134 |
-
"learning_rate": 0.00012175317590061674,
|
| 1135 |
-
"loss": 0.93,
|
| 1136 |
-
"step": 760
|
| 1137 |
-
},
|
| 1138 |
-
{
|
| 1139 |
-
"epoch": 7.320574162679426,
|
| 1140 |
-
"grad_norm": 0.75,
|
| 1141 |
-
"learning_rate": 0.00012065982810501404,
|
| 1142 |
-
"loss": 0.9268,
|
| 1143 |
-
"step": 765
|
| 1144 |
-
},
|
| 1145 |
-
{
|
| 1146 |
-
"epoch": 7.368421052631579,
|
| 1147 |
-
"grad_norm": 0.48828125,
|
| 1148 |
-
"learning_rate": 0.00011956389431422507,
|
| 1149 |
-
"loss": 0.9214,
|
| 1150 |
-
"step": 770
|
| 1151 |
-
},
|
| 1152 |
-
{
|
| 1153 |
-
"epoch": 7.416267942583732,
|
| 1154 |
-
"grad_norm": 0.6015625,
|
| 1155 |
-
"learning_rate": 0.00011846551170652127,
|
| 1156 |
-
"loss": 0.931,
|
| 1157 |
-
"step": 775
|
| 1158 |
-
},
|
| 1159 |
-
{
|
| 1160 |
-
"epoch": 7.464114832535885,
|
| 1161 |
-
"grad_norm": 0.453125,
|
| 1162 |
-
"learning_rate": 0.00011736481776669306,
|
| 1163 |
-
"loss": 0.9316,
|
| 1164 |
-
"step": 780
|
| 1165 |
-
},
|
| 1166 |
-
{
|
| 1167 |
-
"epoch": 7.511961722488039,
|
| 1168 |
-
"grad_norm": 0.369140625,
|
| 1169 |
-
"learning_rate": 0.0001162619502688407,
|
| 1170 |
-
"loss": 0.9272,
|
| 1171 |
-
"step": 785
|
| 1172 |
-
},
|
| 1173 |
-
{
|
| 1174 |
-
"epoch": 7.559808612440191,
|
| 1175 |
-
"grad_norm": 0.435546875,
|
| 1176 |
-
"learning_rate": 0.00011515704725912926,
|
| 1177 |
-
"loss": 0.9219,
|
| 1178 |
-
"step": 790
|
| 1179 |
-
},
|
| 1180 |
-
{
|
| 1181 |
-
"epoch": 7.607655502392344,
|
| 1182 |
-
"grad_norm": 0.42578125,
|
| 1183 |
-
"learning_rate": 0.00011405024703850929,
|
| 1184 |
-
"loss": 0.9363,
|
| 1185 |
-
"step": 795
|
| 1186 |
-
},
|
| 1187 |
-
{
|
| 1188 |
-
"epoch": 7.655502392344498,
|
| 1189 |
-
"grad_norm": 0.5390625,
|
| 1190 |
-
"learning_rate": 0.00011294168814540553,
|
| 1191 |
-
"loss": 0.9388,
|
| 1192 |
-
"step": 800
|
| 1193 |
-
},
|
| 1194 |
-
{
|
| 1195 |
-
"epoch": 7.703349282296651,
|
| 1196 |
-
"grad_norm": 0.431640625,
|
| 1197 |
-
"learning_rate": 0.00011183150933837632,
|
| 1198 |
-
"loss": 0.9284,
|
| 1199 |
-
"step": 805
|
| 1200 |
-
},
|
| 1201 |
-
{
|
| 1202 |
-
"epoch": 7.751196172248804,
|
| 1203 |
-
"grad_norm": 0.353515625,
|
| 1204 |
-
"learning_rate": 0.00011071984957874479,
|
| 1205 |
-
"loss": 0.9222,
|
| 1206 |
-
"step": 810
|
| 1207 |
-
},
|
| 1208 |
-
{
|
| 1209 |
-
"epoch": 7.7990430622009566,
|
| 1210 |
-
"grad_norm": 0.42578125,
|
| 1211 |
-
"learning_rate": 0.00010960684801320536,
|
| 1212 |
-
"loss": 0.9335,
|
| 1213 |
-
"step": 815
|
| 1214 |
-
},
|
| 1215 |
-
{
|
| 1216 |
-
"epoch": 7.84688995215311,
|
| 1217 |
-
"grad_norm": 0.6171875,
|
| 1218 |
-
"learning_rate": 0.00010849264395640649,
|
| 1219 |
-
"loss": 0.9382,
|
| 1220 |
-
"step": 820
|
| 1221 |
-
},
|
| 1222 |
-
{
|
| 1223 |
-
"epoch": 7.894736842105263,
|
| 1224 |
-
"grad_norm": 0.66015625,
|
| 1225 |
-
"learning_rate": 0.00010737737687351284,
|
| 1226 |
-
"loss": 0.9414,
|
| 1227 |
-
"step": 825
|
| 1228 |
-
},
|
| 1229 |
-
{
|
| 1230 |
-
"epoch": 7.942583732057416,
|
| 1231 |
-
"grad_norm": 0.5234375,
|
| 1232 |
-
"learning_rate": 0.0001062611863627482,
|
| 1233 |
-
"loss": 0.9374,
|
| 1234 |
-
"step": 830
|
| 1235 |
-
},
|
| 1236 |
-
{
|
| 1237 |
-
"epoch": 7.990430622009569,
|
| 1238 |
-
"grad_norm": 0.33984375,
|
| 1239 |
-
"learning_rate": 0.00010514421213792205,
|
| 1240 |
-
"loss": 0.93,
|
| 1241 |
-
"step": 835
|
| 1242 |
-
},
|
| 1243 |
-
{
|
| 1244 |
-
"epoch": 8.0,
|
| 1245 |
-
"eval_loss": 2.506725311279297,
|
| 1246 |
-
"eval_runtime": 0.5363,
|
| 1247 |
-
"eval_samples_per_second": 18.647,
|
| 1248 |
-
"eval_steps_per_second": 1.865,
|
| 1249 |
-
"step": 836
|
| 1250 |
-
},
|
| 1251 |
-
{
|
| 1252 |
-
"epoch": 8.038277511961722,
|
| 1253 |
-
"grad_norm": 0.390625,
|
| 1254 |
-
"learning_rate": 0.00010402659401094152,
|
| 1255 |
-
"loss": 0.9129,
|
| 1256 |
-
"step": 840
|
| 1257 |
-
},
|
| 1258 |
-
{
|
| 1259 |
-
"epoch": 8.086124401913876,
|
| 1260 |
-
"grad_norm": 0.35546875,
|
| 1261 |
-
"learning_rate": 0.00010290847187431113,
|
| 1262 |
-
"loss": 0.9107,
|
| 1263 |
-
"step": 845
|
| 1264 |
-
},
|
| 1265 |
-
{
|
| 1266 |
-
"epoch": 8.133971291866029,
|
| 1267 |
-
"grad_norm": 0.36328125,
|
| 1268 |
-
"learning_rate": 0.00010178998568362243,
|
| 1269 |
-
"loss": 0.9226,
|
| 1270 |
-
"step": 850
|
| 1271 |
-
},
|
| 1272 |
-
{
|
| 1273 |
-
"epoch": 8.181818181818182,
|
| 1274 |
-
"grad_norm": 0.345703125,
|
| 1275 |
-
"learning_rate": 0.00010067127544003563,
|
| 1276 |
-
"loss": 0.9184,
|
| 1277 |
-
"step": 855
|
| 1278 |
-
},
|
| 1279 |
-
{
|
| 1280 |
-
"epoch": 8.229665071770334,
|
| 1281 |
-
"grad_norm": 0.35546875,
|
| 1282 |
-
"learning_rate": 9.955248117275566e-05,
|
| 1283 |
-
"loss": 0.915,
|
| 1284 |
-
"step": 860
|
| 1285 |
-
},
|
| 1286 |
-
{
|
| 1287 |
-
"epoch": 8.277511961722489,
|
| 1288 |
-
"grad_norm": 0.396484375,
|
| 1289 |
-
"learning_rate": 9.843374292150488e-05,
|
| 1290 |
-
"loss": 0.9234,
|
| 1291 |
-
"step": 865
|
| 1292 |
-
},
|
| 1293 |
-
{
|
| 1294 |
-
"epoch": 8.325358851674642,
|
| 1295 |
-
"grad_norm": 0.3671875,
|
| 1296 |
-
"learning_rate": 9.73152007189939e-05,
|
| 1297 |
-
"loss": 0.9169,
|
| 1298 |
-
"step": 870
|
| 1299 |
-
},
|
| 1300 |
-
{
|
| 1301 |
-
"epoch": 8.373205741626794,
|
| 1302 |
-
"grad_norm": 0.373046875,
|
| 1303 |
-
"learning_rate": 9.619699457339405e-05,
|
| 1304 |
-
"loss": 0.9131,
|
| 1305 |
-
"step": 875
|
| 1306 |
-
},
|
| 1307 |
-
{
|
| 1308 |
-
"epoch": 8.421052631578947,
|
| 1309 |
-
"grad_norm": 0.3671875,
|
| 1310 |
-
"learning_rate": 9.507926445081219e-05,
|
| 1311 |
-
"loss": 0.9189,
|
| 1312 |
-
"step": 880
|
| 1313 |
-
},
|
| 1314 |
-
{
|
| 1315 |
-
"epoch": 8.4688995215311,
|
| 1316 |
-
"grad_norm": 0.359375,
|
| 1317 |
-
"learning_rate": 9.396215025777139e-05,
|
| 1318 |
-
"loss": 0.9125,
|
| 1319 |
-
"step": 885
|
| 1320 |
-
},
|
| 1321 |
-
{
|
| 1322 |
-
"epoch": 8.516746411483254,
|
| 1323 |
-
"grad_norm": 0.341796875,
|
| 1324 |
-
"learning_rate": 9.284579182369867e-05,
|
| 1325 |
-
"loss": 0.9089,
|
| 1326 |
-
"step": 890
|
| 1327 |
-
},
|
| 1328 |
-
{
|
| 1329 |
-
"epoch": 8.564593301435407,
|
| 1330 |
-
"grad_norm": 0.4453125,
|
| 1331 |
-
"learning_rate": 9.173032888342244e-05,
|
| 1332 |
-
"loss": 0.9221,
|
| 1333 |
-
"step": 895
|
| 1334 |
-
},
|
| 1335 |
-
{
|
| 1336 |
-
"epoch": 8.61244019138756,
|
| 1337 |
-
"grad_norm": 0.5546875,
|
| 1338 |
-
"learning_rate": 9.061590105968208e-05,
|
| 1339 |
-
"loss": 0.9155,
|
| 1340 |
-
"step": 900
|
| 1341 |
-
},
|
| 1342 |
-
{
|
| 1343 |
-
"epoch": 8.660287081339712,
|
| 1344 |
-
"grad_norm": 0.55859375,
|
| 1345 |
-
"learning_rate": 8.950264784565112e-05,
|
| 1346 |
-
"loss": 0.9286,
|
| 1347 |
-
"step": 905
|
| 1348 |
-
},
|
| 1349 |
-
{
|
| 1350 |
-
"epoch": 8.708133971291867,
|
| 1351 |
-
"grad_norm": 0.36328125,
|
| 1352 |
-
"learning_rate": 8.839070858747697e-05,
|
| 1353 |
-
"loss": 0.9199,
|
| 1354 |
-
"step": 910
|
| 1355 |
-
},
|
| 1356 |
-
{
|
| 1357 |
-
"epoch": 8.75598086124402,
|
| 1358 |
-
"grad_norm": 0.365234375,
|
| 1359 |
-
"learning_rate": 8.728022246683894e-05,
|
| 1360 |
-
"loss": 0.908,
|
| 1361 |
-
"step": 915
|
| 1362 |
-
},
|
| 1363 |
-
{
|
| 1364 |
-
"epoch": 8.803827751196172,
|
| 1365 |
-
"grad_norm": 0.345703125,
|
| 1366 |
-
"learning_rate": 8.617132848352671e-05,
|
| 1367 |
-
"loss": 0.9169,
|
| 1368 |
-
"step": 920
|
| 1369 |
-
},
|
| 1370 |
-
{
|
| 1371 |
-
"epoch": 8.851674641148325,
|
| 1372 |
-
"grad_norm": 0.390625,
|
| 1373 |
-
"learning_rate": 8.506416543804182e-05,
|
| 1374 |
-
"loss": 0.9238,
|
| 1375 |
-
"step": 925
|
| 1376 |
-
},
|
| 1377 |
-
{
|
| 1378 |
-
"epoch": 8.89952153110048,
|
| 1379 |
-
"grad_norm": 0.5703125,
|
| 1380 |
-
"learning_rate": 8.395887191422397e-05,
|
| 1381 |
-
"loss": 0.9209,
|
| 1382 |
-
"step": 930
|
| 1383 |
-
},
|
| 1384 |
-
{
|
| 1385 |
-
"epoch": 8.947368421052632,
|
| 1386 |
-
"grad_norm": 0.46875,
|
| 1387 |
-
"learning_rate": 8.285558626190447e-05,
|
| 1388 |
-
"loss": 0.9189,
|
| 1389 |
-
"step": 935
|
| 1390 |
-
},
|
| 1391 |
-
{
|
| 1392 |
-
"epoch": 8.995215311004785,
|
| 1393 |
-
"grad_norm": 0.337890625,
|
| 1394 |
-
"learning_rate": 8.175444657958876e-05,
|
| 1395 |
-
"loss": 0.9195,
|
| 1396 |
-
"step": 940
|
| 1397 |
-
},
|
| 1398 |
-
{
|
| 1399 |
-
"epoch": 8.995215311004785,
|
| 1400 |
-
"eval_loss": 2.5168216228485107,
|
| 1401 |
-
"eval_runtime": 0.6511,
|
| 1402 |
-
"eval_samples_per_second": 15.36,
|
| 1403 |
-
"eval_steps_per_second": 1.536,
|
| 1404 |
-
"step": 940
|
| 1405 |
-
},
|
| 1406 |
-
{
|
| 1407 |
-
"epoch": 9.043062200956937,
|
| 1408 |
-
"grad_norm": 0.40625,
|
| 1409 |
-
"learning_rate": 8.065559069717088e-05,
|
| 1410 |
-
"loss": 0.9021,
|
| 1411 |
-
"step": 945
|
| 1412 |
-
},
|
| 1413 |
-
{
|
| 1414 |
-
"epoch": 9.090909090909092,
|
| 1415 |
-
"grad_norm": 0.41015625,
|
| 1416 |
-
"learning_rate": 7.955915615868111e-05,
|
| 1417 |
-
"loss": 0.9103,
|
| 1418 |
-
"step": 950
|
| 1419 |
-
},
|
| 1420 |
-
{
|
| 1421 |
-
"epoch": 9.138755980861244,
|
| 1422 |
-
"grad_norm": 0.392578125,
|
| 1423 |
-
"learning_rate": 7.846528020506957e-05,
|
| 1424 |
-
"loss": 0.8983,
|
| 1425 |
-
"step": 955
|
| 1426 |
-
},
|
| 1427 |
-
{
|
| 1428 |
-
"epoch": 9.186602870813397,
|
| 1429 |
-
"grad_norm": 0.36328125,
|
| 1430 |
-
"learning_rate": 7.73740997570278e-05,
|
| 1431 |
-
"loss": 0.9037,
|
| 1432 |
-
"step": 960
|
| 1433 |
-
},
|
| 1434 |
-
{
|
| 1435 |
-
"epoch": 9.23444976076555,
|
| 1436 |
-
"grad_norm": 0.353515625,
|
| 1437 |
-
"learning_rate": 7.628575139785024e-05,
|
| 1438 |
-
"loss": 0.9036,
|
| 1439 |
-
"step": 965
|
| 1440 |
-
},
|
| 1441 |
-
{
|
| 1442 |
-
"epoch": 9.282296650717702,
|
| 1443 |
-
"grad_norm": 0.388671875,
|
| 1444 |
-
"learning_rate": 7.520037135633816e-05,
|
| 1445 |
-
"loss": 0.9051,
|
| 1446 |
-
"step": 970
|
| 1447 |
-
},
|
| 1448 |
-
{
|
| 1449 |
-
"epoch": 9.330143540669857,
|
| 1450 |
-
"grad_norm": 0.439453125,
|
| 1451 |
-
"learning_rate": 7.411809548974792e-05,
|
| 1452 |
-
"loss": 0.9186,
|
| 1453 |
-
"step": 975
|
| 1454 |
-
},
|
| 1455 |
-
{
|
| 1456 |
-
"epoch": 9.37799043062201,
|
| 1457 |
-
"grad_norm": 0.365234375,
|
| 1458 |
"learning_rate": 7.303905926678564e-05,
|
| 1459 |
-
"loss": 0.
|
| 1460 |
-
"step":
|
| 1461 |
-
},
|
| 1462 |
-
{
|
| 1463 |
-
"epoch": 9.425837320574162,
|
| 1464 |
-
"grad_norm": 0.375,
|
| 1465 |
-
"learning_rate": 7.196339775065042e-05,
|
| 1466 |
-
"loss": 0.9069,
|
| 1467 |
-
"step": 985
|
| 1468 |
},
|
| 1469 |
{
|
| 1470 |
-
"epoch": 9.
|
| 1471 |
-
"grad_norm": 0.
|
| 1472 |
"learning_rate": 7.089124558212871e-05,
|
| 1473 |
-
"loss": 0.
|
| 1474 |
-
"step":
|
| 1475 |
-
},
|
| 1476 |
-
{
|
| 1477 |
-
"epoch": 9.52153110047847,
|
| 1478 |
-
"grad_norm": 0.38671875,
|
| 1479 |
-
"learning_rate": 6.982273696274106e-05,
|
| 1480 |
-
"loss": 0.9121,
|
| 1481 |
-
"step": 995
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
-
"epoch": 9.
|
| 1485 |
-
"grad_norm": 0.
|
| 1486 |
"learning_rate": 6.875800563794425e-05,
|
| 1487 |
-
"loss": 0.
|
| 1488 |
-
"step":
|
| 1489 |
-
},
|
| 1490 |
-
{
|
| 1491 |
-
"epoch": 9.617224880382775,
|
| 1492 |
-
"grad_norm": 0.361328125,
|
| 1493 |
-
"learning_rate": 6.769718488039023e-05,
|
| 1494 |
-
"loss": 0.9101,
|
| 1495 |
-
"step": 1005
|
| 1496 |
},
|
| 1497 |
{
|
| 1498 |
-
"epoch": 9.
|
| 1499 |
-
"grad_norm": 0.
|
| 1500 |
"learning_rate": 6.664040747324437e-05,
|
| 1501 |
-
"loss": 0.
|
| 1502 |
-
"step":
|
| 1503 |
-
},
|
| 1504 |
-
{
|
| 1505 |
-
"epoch": 9.712918660287082,
|
| 1506 |
-
"grad_norm": 0.375,
|
| 1507 |
-
"learning_rate": 6.558780569356507e-05,
|
| 1508 |
-
"loss": 0.9106,
|
| 1509 |
-
"step": 1015
|
| 1510 |
},
|
| 1511 |
{
|
| 1512 |
-
"epoch": 9.
|
| 1513 |
-
"grad_norm": 0.
|
| 1514 |
"learning_rate": 6.453951129574644e-05,
|
| 1515 |
-
"loss": 0.
|
| 1516 |
-
"step":
|
| 1517 |
-
},
|
| 1518 |
-
{
|
| 1519 |
-
"epoch": 9.808612440191387,
|
| 1520 |
-
"grad_norm": 0.40234375,
|
| 1521 |
-
"learning_rate": 6.349565549502676e-05,
|
| 1522 |
-
"loss": 0.9033,
|
| 1523 |
-
"step": 1025
|
| 1524 |
},
|
| 1525 |
{
|
| 1526 |
-
"epoch": 9.
|
| 1527 |
-
"grad_norm": 0.
|
| 1528 |
"learning_rate": 6.245636895106402e-05,
|
| 1529 |
-
"loss": 0.
|
| 1530 |
-
"step":
|
| 1531 |
-
},
|
| 1532 |
-
{
|
| 1533 |
-
"epoch": 9.904306220095695,
|
| 1534 |
-
"grad_norm": 0.357421875,
|
| 1535 |
-
"learning_rate": 6.142178175158149e-05,
|
| 1536 |
-
"loss": 0.9082,
|
| 1537 |
-
"step": 1035
|
| 1538 |
},
|
| 1539 |
{
|
| 1540 |
-
"epoch": 9.
|
| 1541 |
-
"grad_norm": 0.
|
| 1542 |
"learning_rate": 6.039202339608432e-05,
|
| 1543 |
-
"loss": 0.
|
| 1544 |
-
"step":
|
| 1545 |
-
},
|
| 1546 |
-
{
|
| 1547 |
-
"epoch": 10.0,
|
| 1548 |
-
"grad_norm": 0.50390625,
|
| 1549 |
-
"learning_rate": 5.9367222779650334e-05,
|
| 1550 |
-
"loss": 0.912,
|
| 1551 |
-
"step": 1045
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
"epoch": 10.0,
|
| 1555 |
-
"
|
| 1556 |
-
"eval_runtime": 0.5317,
|
| 1557 |
-
"eval_samples_per_second": 18.809,
|
| 1558 |
-
"eval_steps_per_second": 1.881,
|
| 1559 |
-
"step": 1045
|
| 1560 |
-
},
|
| 1561 |
-
{
|
| 1562 |
-
"epoch": 10.047846889952153,
|
| 1563 |
-
"grad_norm": 0.439453125,
|
| 1564 |
"learning_rate": 5.834750817679606e-05,
|
| 1565 |
-
"loss": 0.
|
| 1566 |
-
"step":
|
| 1567 |
},
|
| 1568 |
{
|
| 1569 |
-
"epoch": 10.
|
| 1570 |
-
"
|
| 1571 |
-
"
|
| 1572 |
-
"
|
| 1573 |
-
"
|
|
|
|
| 1574 |
},
|
| 1575 |
{
|
| 1576 |
-
"epoch": 10.
|
| 1577 |
-
"grad_norm": 0.
|
| 1578 |
"learning_rate": 5.6323846910828735e-05,
|
| 1579 |
-
"loss": 0.
|
| 1580 |
-
"step":
|
| 1581 |
-
},
|
| 1582 |
-
{
|
| 1583 |
-
"epoch": 10.191387559808613,
|
| 1584 |
-
"grad_norm": 0.349609375,
|
| 1585 |
-
"learning_rate": 5.5320153549837415e-05,
|
| 1586 |
-
"loss": 0.9026,
|
| 1587 |
-
"step": 1065
|
| 1588 |
},
|
| 1589 |
{
|
| 1590 |
-
"epoch": 10.
|
| 1591 |
-
"grad_norm": 0.
|
| 1592 |
"learning_rate": 5.432205277496327e-05,
|
| 1593 |
-
"loss": 0.
|
| 1594 |
-
"step":
|
| 1595 |
-
},
|
| 1596 |
-
{
|
| 1597 |
-
"epoch": 10.287081339712918,
|
| 1598 |
-
"grad_norm": 0.359375,
|
| 1599 |
-
"learning_rate": 5.33296695186977e-05,
|
| 1600 |
-
"loss": 0.9,
|
| 1601 |
-
"step": 1075
|
| 1602 |
},
|
| 1603 |
{
|
| 1604 |
-
"epoch": 10.
|
| 1605 |
-
"grad_norm": 0.
|
| 1606 |
"learning_rate": 5.234312799786921e-05,
|
| 1607 |
-
"loss": 0.
|
| 1608 |
-
"step":
|
| 1609 |
-
},
|
| 1610 |
-
{
|
| 1611 |
-
"epoch": 10.382775119617225,
|
| 1612 |
-
"grad_norm": 0.41015625,
|
| 1613 |
-
"learning_rate": 5.1362551698094964e-05,
|
| 1614 |
-
"loss": 0.8969,
|
| 1615 |
-
"step": 1085
|
| 1616 |
},
|
| 1617 |
{
|
| 1618 |
-
"epoch": 10.
|
| 1619 |
-
"grad_norm": 0.
|
| 1620 |
"learning_rate": 5.0388063358324134e-05,
|
| 1621 |
-
"loss": 0.
|
| 1622 |
-
"step":
|
| 1623 |
-
},
|
| 1624 |
-
{
|
| 1625 |
-
"epoch": 10.47846889952153,
|
| 1626 |
-
"grad_norm": 0.3515625,
|
| 1627 |
-
"learning_rate": 4.9419784955474524e-05,
|
| 1628 |
-
"loss": 0.8991,
|
| 1629 |
-
"step": 1095
|
| 1630 |
},
|
| 1631 |
{
|
| 1632 |
-
"epoch": 10.
|
| 1633 |
-
"grad_norm": 0.
|
| 1634 |
"learning_rate": 4.845783768916482e-05,
|
| 1635 |
-
"loss": 0.
|
| 1636 |
-
"step":
|
| 1637 |
-
},
|
| 1638 |
-
{
|
| 1639 |
-
"epoch": 10.574162679425838,
|
| 1640 |
-
"grad_norm": 0.353515625,
|
| 1641 |
-
"learning_rate": 4.7502341966544e-05,
|
| 1642 |
-
"loss": 0.9012,
|
| 1643 |
-
"step": 1105
|
| 1644 |
},
|
| 1645 |
{
|
| 1646 |
-
"epoch": 10.
|
| 1647 |
-
"grad_norm": 0.
|
| 1648 |
"learning_rate": 4.6553417387219886e-05,
|
| 1649 |
-
"loss": 0.
|
| 1650 |
-
"step":
|
| 1651 |
-
},
|
| 1652 |
-
{
|
| 1653 |
-
"epoch": 10.669856459330143,
|
| 1654 |
-
"grad_norm": 0.333984375,
|
| 1655 |
-
"learning_rate": 4.5611182728288895e-05,
|
| 1656 |
-
"loss": 0.9073,
|
| 1657 |
-
"step": 1115
|
| 1658 |
},
|
| 1659 |
{
|
| 1660 |
-
"epoch": 10.
|
| 1661 |
-
"grad_norm": 0.
|
| 1662 |
"learning_rate": 4.467575592946864e-05,
|
| 1663 |
-
"loss": 0.
|
| 1664 |
-
"step":
|
| 1665 |
-
},
|
| 1666 |
-
{
|
| 1667 |
-
"epoch": 10.76555023923445,
|
| 1668 |
-
"grad_norm": 0.349609375,
|
| 1669 |
-
"learning_rate": 4.374725407833532e-05,
|
| 1670 |
-
"loss": 0.902,
|
| 1671 |
-
"step": 1125
|
| 1672 |
},
|
| 1673 |
{
|
| 1674 |
-
"epoch": 10.
|
| 1675 |
-
"grad_norm": 0.
|
| 1676 |
"learning_rate": 4.282579339566802e-05,
|
| 1677 |
-
"loss": 0.
|
| 1678 |
-
"step":
|
| 1679 |
-
},
|
| 1680 |
-
{
|
| 1681 |
-
"epoch": 10.861244019138756,
|
| 1682 |
-
"grad_norm": 0.37109375,
|
| 1683 |
-
"learning_rate": 4.1911489220901236e-05,
|
| 1684 |
-
"loss": 0.9002,
|
| 1685 |
-
"step": 1135
|
| 1686 |
},
|
| 1687 |
{
|
| 1688 |
-
"epoch": 10.
|
| 1689 |
-
"grad_norm": 0.
|
| 1690 |
"learning_rate": 4.100445599768774e-05,
|
| 1691 |
-
"loss": 0.
|
| 1692 |
-
"step":
|
| 1693 |
-
},
|
| 1694 |
-
{
|
| 1695 |
-
"epoch": 10.956937799043063,
|
| 1696 |
-
"grad_norm": 0.34375,
|
| 1697 |
-
"learning_rate": 4.0104807259573716e-05,
|
| 1698 |
-
"loss": 0.9003,
|
| 1699 |
-
"step": 1145
|
| 1700 |
-
},
|
| 1701 |
-
{
|
| 1702 |
-
"epoch": 10.995215311004785,
|
| 1703 |
-
"eval_loss": 2.535585403442383,
|
| 1704 |
-
"eval_runtime": 0.6791,
|
| 1705 |
-
"eval_samples_per_second": 14.726,
|
| 1706 |
-
"eval_steps_per_second": 1.473,
|
| 1707 |
-
"step": 1149
|
| 1708 |
},
|
| 1709 |
{
|
| 1710 |
-
"epoch":
|
| 1711 |
-
"grad_norm": 0.
|
| 1712 |
"learning_rate": 3.9212655615787804e-05,
|
| 1713 |
-
"loss": 0.
|
| 1714 |
-
"step":
|
| 1715 |
},
|
| 1716 |
{
|
| 1717 |
-
"epoch":
|
| 1718 |
-
"
|
| 1719 |
-
"
|
| 1720 |
-
"
|
| 1721 |
-
"
|
|
|
|
| 1722 |
},
|
| 1723 |
{
|
| 1724 |
-
"epoch": 11.
|
| 1725 |
-
"grad_norm": 0.
|
| 1726 |
"learning_rate": 3.745128934207225e-05,
|
| 1727 |
-
"loss": 0.
|
| 1728 |
-
"step":
|
| 1729 |
-
},
|
| 1730 |
-
{
|
| 1731 |
-
"epoch": 11.148325358851675,
|
| 1732 |
-
"grad_norm": 0.388671875,
|
| 1733 |
-
"learning_rate": 3.6582295182742964e-05,
|
| 1734 |
-
"loss": 0.8935,
|
| 1735 |
-
"step": 1165
|
| 1736 |
},
|
| 1737 |
{
|
| 1738 |
-
"epoch": 11.
|
| 1739 |
-
"grad_norm": 0.
|
| 1740 |
"learning_rate": 3.5721239031346066e-05,
|
| 1741 |
-
"loss": 0.
|
| 1742 |
-
"step":
|
| 1743 |
-
},
|
| 1744 |
-
{
|
| 1745 |
-
"epoch": 11.24401913875598,
|
| 1746 |
-
"grad_norm": 0.3359375,
|
| 1747 |
-
"learning_rate": 3.4868228666467704e-05,
|
| 1748 |
-
"loss": 0.8972,
|
| 1749 |
-
"step": 1175
|
| 1750 |
},
|
| 1751 |
{
|
| 1752 |
-
"epoch": 11.
|
| 1753 |
-
"grad_norm": 0.
|
| 1754 |
"learning_rate": 3.402337085960119e-05,
|
| 1755 |
-
"loss": 0.
|
| 1756 |
-
"step":
|
| 1757 |
-
},
|
| 1758 |
-
{
|
| 1759 |
-
"epoch": 11.339712918660288,
|
| 1760 |
-
"grad_norm": 0.341796875,
|
| 1761 |
-
"learning_rate": 3.318677136178228e-05,
|
| 1762 |
-
"loss": 0.8943,
|
| 1763 |
-
"step": 1185
|
| 1764 |
},
|
| 1765 |
{
|
| 1766 |
-
"epoch": 11.
|
| 1767 |
-
"grad_norm": 0.
|
| 1768 |
"learning_rate": 3.235853489035241e-05,
|
| 1769 |
-
"loss": 0.
|
| 1770 |
-
"step":
|
| 1771 |
-
},
|
| 1772 |
-
{
|
| 1773 |
-
"epoch": 11.435406698564593,
|
| 1774 |
-
"grad_norm": 0.384765625,
|
| 1775 |
-
"learning_rate": 3.153876511585122e-05,
|
| 1776 |
-
"loss": 0.8917,
|
| 1777 |
-
"step": 1195
|
| 1778 |
},
|
| 1779 |
{
|
| 1780 |
-
"epoch": 11.
|
| 1781 |
-
"grad_norm": 0.
|
| 1782 |
"learning_rate": 3.072756464904006e-05,
|
| 1783 |
-
"loss": 0.
|
| 1784 |
-
"step":
|
| 1785 |
-
},
|
| 1786 |
-
{
|
| 1787 |
-
"epoch": 11.5311004784689,
|
| 1788 |
-
"grad_norm": 0.337890625,
|
| 1789 |
-
"learning_rate": 2.9925035028058134e-05,
|
| 1790 |
-
"loss": 0.8951,
|
| 1791 |
-
"step": 1205
|
| 1792 |
},
|
| 1793 |
{
|
| 1794 |
-
"epoch": 11.
|
| 1795 |
-
"grad_norm": 0.
|
| 1796 |
"learning_rate": 2.9131276705713006e-05,
|
| 1797 |
-
"loss": 0.
|
| 1798 |
-
"step":
|
| 1799 |
-
},
|
| 1800 |
-
{
|
| 1801 |
-
"epoch": 11.626794258373206,
|
| 1802 |
-
"grad_norm": 0.345703125,
|
| 1803 |
-
"learning_rate": 2.8346389036906828e-05,
|
| 1804 |
-
"loss": 0.899,
|
| 1805 |
-
"step": 1215
|
| 1806 |
},
|
| 1807 |
{
|
| 1808 |
-
"epoch": 11.
|
| 1809 |
-
"grad_norm": 0.
|
| 1810 |
"learning_rate": 2.7570470266200176e-05,
|
| 1811 |
-
"loss": 0.
|
| 1812 |
-
"step":
|
| 1813 |
-
},
|
| 1814 |
-
{
|
| 1815 |
-
"epoch": 11.722488038277511,
|
| 1816 |
-
"grad_norm": 0.33984375,
|
| 1817 |
-
"learning_rate": 2.68036175155147e-05,
|
| 1818 |
-
"loss": 0.8917,
|
| 1819 |
-
"step": 1225
|
| 1820 |
},
|
| 1821 |
{
|
| 1822 |
-
"epoch": 11.
|
| 1823 |
-
"grad_norm": 0.
|
| 1824 |
"learning_rate": 2.6045926771976303e-05,
|
| 1825 |
-
"loss": 0.
|
| 1826 |
-
"step":
|
| 1827 |
-
},
|
| 1828 |
-
{
|
| 1829 |
-
"epoch": 11.818181818181818,
|
| 1830 |
-
"grad_norm": 0.3671875,
|
| 1831 |
-
"learning_rate": 2.529749287590042e-05,
|
| 1832 |
-
"loss": 0.9028,
|
| 1833 |
-
"step": 1235
|
| 1834 |
},
|
| 1835 |
{
|
| 1836 |
-
"epoch": 11.
|
| 1837 |
-
"grad_norm": 0.
|
| 1838 |
"learning_rate": 2.4558409508920986e-05,
|
| 1839 |
-
"loss": 0.
|
| 1840 |
-
"step":
|
| 1841 |
-
},
|
| 1842 |
-
{
|
| 1843 |
-
"epoch": 11.913875598086124,
|
| 1844 |
-
"grad_norm": 0.349609375,
|
| 1845 |
-
"learning_rate": 2.382876918226409e-05,
|
| 1846 |
-
"loss": 0.8938,
|
| 1847 |
-
"step": 1245
|
| 1848 |
},
|
| 1849 |
{
|
| 1850 |
-
"epoch": 11.
|
| 1851 |
-
"grad_norm": 0.
|
| 1852 |
"learning_rate": 2.3108663225168435e-05,
|
| 1853 |
-
"loss": 0.
|
| 1854 |
-
"step":
|
| 1855 |
},
|
| 1856 |
{
|
| 1857 |
"epoch": 12.0,
|
| 1858 |
-
"
|
| 1859 |
-
"eval_runtime": 0.536,
|
| 1860 |
-
"eval_samples_per_second": 18.658,
|
| 1861 |
-
"eval_steps_per_second": 1.866,
|
| 1862 |
-
"step": 1254
|
| 1863 |
-
},
|
| 1864 |
-
{
|
| 1865 |
-
"epoch": 12.009569377990431,
|
| 1866 |
-
"grad_norm": 0.33984375,
|
| 1867 |
-
"learning_rate": 2.239818177345364e-05,
|
| 1868 |
-
"loss": 0.9023,
|
| 1869 |
-
"step": 1255
|
| 1870 |
-
},
|
| 1871 |
-
{
|
| 1872 |
-
"epoch": 12.057416267942584,
|
| 1873 |
-
"grad_norm": 0.337890625,
|
| 1874 |
"learning_rate": 2.1697413758237784e-05,
|
| 1875 |
-
"loss": 0.
|
| 1876 |
-
"step":
|
| 1877 |
},
|
| 1878 |
{
|
| 1879 |
-
"epoch": 12.
|
| 1880 |
-
"
|
| 1881 |
-
"
|
| 1882 |
-
"
|
| 1883 |
-
"
|
|
|
|
| 1884 |
},
|
| 1885 |
{
|
| 1886 |
-
"epoch": 12.
|
| 1887 |
-
"grad_norm": 0.
|
| 1888 |
"learning_rate": 2.032536767163141e-05,
|
| 1889 |
-
"loss": 0.
|
| 1890 |
-
"step":
|
| 1891 |
-
},
|
| 1892 |
-
{
|
| 1893 |
-
"epoch": 12.200956937799043,
|
| 1894 |
-
"grad_norm": 0.3671875,
|
| 1895 |
-
"learning_rate": 1.965426133954854e-05,
|
| 1896 |
-
"loss": 0.9023,
|
| 1897 |
-
"step": 1275
|
| 1898 |
},
|
| 1899 |
{
|
| 1900 |
-
"epoch": 12.
|
| 1901 |
-
"grad_norm": 0.
|
| 1902 |
"learning_rate": 1.8993211901083353e-05,
|
| 1903 |
-
"loss": 0.
|
| 1904 |
-
"step":
|
| 1905 |
-
},
|
| 1906 |
-
{
|
| 1907 |
-
"epoch": 12.296650717703349,
|
| 1908 |
-
"grad_norm": 0.328125,
|
| 1909 |
-
"learning_rate": 1.8342302099938057e-05,
|
| 1910 |
-
"loss": 0.8925,
|
| 1911 |
-
"step": 1285
|
| 1912 |
},
|
| 1913 |
{
|
| 1914 |
-
"epoch": 12.
|
| 1915 |
-
"grad_norm": 0.
|
| 1916 |
"learning_rate": 1.7701613410634365e-05,
|
| 1917 |
-
"loss": 0.
|
| 1918 |
-
"step":
|
| 1919 |
-
},
|
| 1920 |
-
{
|
| 1921 |
-
"epoch": 12.392344497607656,
|
| 1922 |
-
"grad_norm": 0.33203125,
|
| 1923 |
-
"learning_rate": 1.7071226028315113e-05,
|
| 1924 |
-
"loss": 0.8922,
|
| 1925 |
-
"step": 1295
|
| 1926 |
},
|
| 1927 |
{
|
| 1928 |
-
"epoch": 12.
|
| 1929 |
-
"grad_norm": 0.
|
| 1930 |
"learning_rate": 1.6451218858706374e-05,
|
| 1931 |
-
"loss": 0.
|
| 1932 |
-
"step":
|
| 1933 |
-
},
|
| 1934 |
-
{
|
| 1935 |
-
"epoch": 12.488038277511961,
|
| 1936 |
-
"grad_norm": 0.337890625,
|
| 1937 |
-
"learning_rate": 1.584166950824061e-05,
|
| 1938 |
-
"loss": 0.8992,
|
| 1939 |
-
"step": 1305
|
| 1940 |
},
|
| 1941 |
{
|
| 1942 |
-
"epoch": 12.
|
| 1943 |
-
"grad_norm": 0.
|
| 1944 |
"learning_rate": 1.5242654274342894e-05,
|
| 1945 |
-
"loss": 0.
|
| 1946 |
-
"step":
|
| 1947 |
-
},
|
| 1948 |
-
{
|
| 1949 |
-
"epoch": 12.583732057416269,
|
| 1950 |
-
"grad_norm": 0.353515625,
|
| 1951 |
-
"learning_rate": 1.4654248135880621e-05,
|
| 1952 |
-
"loss": 0.8942,
|
| 1953 |
-
"step": 1315
|
| 1954 |
},
|
| 1955 |
{
|
| 1956 |
-
"epoch": 12.
|
| 1957 |
-
"grad_norm": 0.
|
| 1958 |
"learning_rate": 1.4076524743778319e-05,
|
| 1959 |
-
"loss": 0.
|
| 1960 |
-
"step":
|
| 1961 |
-
},
|
| 1962 |
-
{
|
| 1963 |
-
"epoch": 12.679425837320574,
|
| 1964 |
-
"grad_norm": 0.33203125,
|
| 1965 |
-
"learning_rate": 1.350955641179893e-05,
|
| 1966 |
-
"loss": 0.8981,
|
| 1967 |
-
"step": 1325
|
| 1968 |
},
|
| 1969 |
{
|
| 1970 |
-
"epoch": 12.
|
| 1971 |
-
"grad_norm": 0.
|
| 1972 |
"learning_rate": 1.295341410749208e-05,
|
| 1973 |
-
"loss": 0.
|
| 1974 |
-
"step":
|
| 1975 |
-
},
|
| 1976 |
-
{
|
| 1977 |
-
"epoch": 12.775119617224881,
|
| 1978 |
-
"grad_norm": 0.34375,
|
| 1979 |
-
"learning_rate": 1.2408167443311214e-05,
|
| 1980 |
-
"loss": 0.8945,
|
| 1981 |
-
"step": 1335
|
| 1982 |
},
|
| 1983 |
{
|
| 1984 |
-
"epoch": 12.
|
| 1985 |
-
"grad_norm": 0.
|
| 1986 |
"learning_rate": 1.1873884667900125e-05,
|
| 1987 |
-
"loss": 0.
|
| 1988 |
-
"step":
|
| 1989 |
-
},
|
| 1990 |
-
{
|
| 1991 |
-
"epoch": 12.870813397129186,
|
| 1992 |
-
"grad_norm": 0.34375,
|
| 1993 |
-
"learning_rate": 1.1350632657550253e-05,
|
| 1994 |
-
"loss": 0.8922,
|
| 1995 |
-
"step": 1345
|
| 1996 |
},
|
| 1997 |
{
|
| 1998 |
-
"epoch": 12.
|
| 1999 |
-
"grad_norm": 0.
|
| 2000 |
"learning_rate": 1.083847690782972e-05,
|
| 2001 |
-
"loss": 0.
|
| 2002 |
-
"step":
|
| 2003 |
-
},
|
| 2004 |
-
{
|
| 2005 |
-
"epoch": 12.966507177033494,
|
| 2006 |
-
"grad_norm": 0.349609375,
|
| 2007 |
-
"learning_rate": 1.0337481525385362e-05,
|
| 2008 |
-
"loss": 0.9006,
|
| 2009 |
-
"step": 1355
|
| 2010 |
-
},
|
| 2011 |
-
{
|
| 2012 |
-
"epoch": 12.995215311004785,
|
| 2013 |
-
"eval_loss": 2.5425758361816406,
|
| 2014 |
-
"eval_runtime": 0.6353,
|
| 2015 |
-
"eval_samples_per_second": 15.741,
|
| 2016 |
-
"eval_steps_per_second": 1.574,
|
| 2017 |
-
"step": 1358
|
| 2018 |
},
|
| 2019 |
{
|
| 2020 |
-
"epoch":
|
| 2021 |
-
"grad_norm": 0.
|
| 2022 |
"learning_rate": 9.8477092199184e-06,
|
| 2023 |
-
"loss": 0.
|
| 2024 |
-
"step":
|
| 2025 |
},
|
| 2026 |
{
|
| 2027 |
-
"epoch":
|
| 2028 |
-
"
|
| 2029 |
-
"
|
| 2030 |
-
"
|
| 2031 |
-
"
|
|
|
|
| 2032 |
},
|
| 2033 |
{
|
| 2034 |
-
"epoch": 13.
|
| 2035 |
-
"grad_norm": 0.
|
| 2036 |
"learning_rate": 8.902077647072881e-06,
|
| 2037 |
-
"loss": 0.
|
| 2038 |
-
"step":
|
| 2039 |
-
},
|
| 2040 |
-
{
|
| 2041 |
-
"epoch": 13.157894736842104,
|
| 2042 |
-
"grad_norm": 0.34375,
|
| 2043 |
-
"learning_rate": 8.446336744604378e-06,
|
| 2044 |
-
"loss": 0.888,
|
| 2045 |
-
"step": 1375
|
| 2046 |
},
|
| 2047 |
{
|
| 2048 |
-
"epoch": 13.
|
| 2049 |
-
"grad_norm": 0.
|
| 2050 |
"learning_rate": 8.002055634117578e-06,
|
| 2051 |
-
"loss": 0.
|
| 2052 |
-
"step":
|
| 2053 |
-
},
|
| 2054 |
-
{
|
| 2055 |
-
"epoch": 13.253588516746412,
|
| 2056 |
-
"grad_norm": 0.34375,
|
| 2057 |
-
"learning_rate": 7.569289926375933e-06,
|
| 2058 |
-
"loss": 0.8931,
|
| 2059 |
-
"step": 1385
|
| 2060 |
},
|
| 2061 |
{
|
| 2062 |
-
"epoch": 13.
|
| 2063 |
-
"grad_norm": 0.
|
| 2064 |
"learning_rate": 7.148093790757371e-06,
|
| 2065 |
-
"loss": 0.
|
| 2066 |
-
"step":
|
| 2067 |
-
},
|
| 2068 |
-
{
|
| 2069 |
-
"epoch": 13.349282296650717,
|
| 2070 |
-
"grad_norm": 0.330078125,
|
| 2071 |
-
"learning_rate": 6.738519948473976e-06,
|
| 2072 |
-
"loss": 0.8914,
|
| 2073 |
-
"step": 1395
|
| 2074 |
},
|
| 2075 |
{
|
| 2076 |
-
"epoch": 13.
|
| 2077 |
-
"grad_norm": 0.
|
| 2078 |
"learning_rate": 6.3406196659728465e-06,
|
| 2079 |
-
"loss": 0.
|
| 2080 |
-
"step":
|
| 2081 |
-
},
|
| 2082 |
-
{
|
| 2083 |
-
"epoch": 13.444976076555024,
|
| 2084 |
-
"grad_norm": 0.330078125,
|
| 2085 |
-
"learning_rate": 5.954442748519073e-06,
|
| 2086 |
-
"loss": 0.8908,
|
| 2087 |
-
"step": 1405
|
| 2088 |
},
|
| 2089 |
{
|
| 2090 |
-
"epoch": 13.
|
| 2091 |
-
"grad_norm": 0.
|
| 2092 |
"learning_rate": 5.580037533961546e-06,
|
| 2093 |
-
"loss": 0.
|
| 2094 |
-
"step":
|
| 2095 |
-
},
|
| 2096 |
-
{
|
| 2097 |
-
"epoch": 13.54066985645933,
|
| 2098 |
-
"grad_norm": 0.349609375,
|
| 2099 |
-
"learning_rate": 5.217450886682584e-06,
|
| 2100 |
-
"loss": 0.8958,
|
| 2101 |
-
"step": 1415
|
| 2102 |
},
|
| 2103 |
{
|
| 2104 |
-
"epoch": 13.
|
| 2105 |
-
"grad_norm": 0.
|
| 2106 |
"learning_rate": 4.866728191731829e-06,
|
| 2107 |
-
"loss": 0.
|
| 2108 |
-
"step":
|
| 2109 |
-
},
|
| 2110 |
-
{
|
| 2111 |
-
"epoch": 13.636363636363637,
|
| 2112 |
-
"grad_norm": 0.330078125,
|
| 2113 |
-
"learning_rate": 4.527913349145441e-06,
|
| 2114 |
-
"loss": 0.8995,
|
| 2115 |
-
"step": 1425
|
| 2116 |
},
|
| 2117 |
{
|
| 2118 |
-
"epoch": 13.
|
| 2119 |
-
"grad_norm": 0.
|
| 2120 |
"learning_rate": 4.20104876845111e-06,
|
| 2121 |
-
"loss": 0.
|
| 2122 |
-
"step":
|
| 2123 |
-
},
|
| 2124 |
-
{
|
| 2125 |
-
"epoch": 13.732057416267942,
|
| 2126 |
-
"grad_norm": 0.33203125,
|
| 2127 |
-
"learning_rate": 3.886175363359646e-06,
|
| 2128 |
-
"loss": 0.8964,
|
| 2129 |
-
"step": 1435
|
| 2130 |
},
|
| 2131 |
{
|
| 2132 |
-
"epoch": 13.
|
| 2133 |
-
"grad_norm": 0.
|
| 2134 |
"learning_rate": 3.5833325466437694e-06,
|
| 2135 |
-
"loss": 0.
|
| 2136 |
-
"step":
|
| 2137 |
-
},
|
| 2138 |
-
{
|
| 2139 |
-
"epoch": 13.82775119617225,
|
| 2140 |
-
"grad_norm": 0.341796875,
|
| 2141 |
-
"learning_rate": 3.2925582252048338e-06,
|
| 2142 |
-
"loss": 0.8903,
|
| 2143 |
-
"step": 1445
|
| 2144 |
},
|
| 2145 |
{
|
| 2146 |
-
"epoch": 13.
|
| 2147 |
-
"grad_norm": 0.
|
| 2148 |
"learning_rate": 3.013888795328057e-06,
|
| 2149 |
-
"loss": 0.
|
| 2150 |
-
"step":
|
| 2151 |
-
},
|
| 2152 |
-
{
|
| 2153 |
-
"epoch": 13.923444976076555,
|
| 2154 |
-
"grad_norm": 0.330078125,
|
| 2155 |
-
"learning_rate": 2.7473591381266708e-06,
|
| 2156 |
-
"loss": 0.8921,
|
| 2157 |
-
"step": 1455
|
| 2158 |
},
|
| 2159 |
{
|
| 2160 |
-
"epoch": 13.
|
| 2161 |
-
"grad_norm": 0.
|
| 2162 |
"learning_rate": 2.4930026151759766e-06,
|
| 2163 |
-
"loss": 0.
|
| 2164 |
-
"step":
|
| 2165 |
},
|
| 2166 |
{
|
| 2167 |
"epoch": 14.0,
|
| 2168 |
-
"
|
| 2169 |
-
"eval_runtime": 0.5348,
|
| 2170 |
-
"eval_samples_per_second": 18.7,
|
| 2171 |
-
"eval_steps_per_second": 1.87,
|
| 2172 |
-
"step": 1463
|
| 2173 |
-
},
|
| 2174 |
-
{
|
| 2175 |
-
"epoch": 14.019138755980862,
|
| 2176 |
-
"grad_norm": 0.3359375,
|
| 2177 |
-
"learning_rate": 2.250851064337367e-06,
|
| 2178 |
-
"loss": 0.8857,
|
| 2179 |
-
"step": 1465
|
| 2180 |
-
},
|
| 2181 |
-
{
|
| 2182 |
-
"epoch": 14.066985645933014,
|
| 2183 |
-
"grad_norm": 0.330078125,
|
| 2184 |
"learning_rate": 2.0209347957732328e-06,
|
| 2185 |
-
"loss": 0.
|
| 2186 |
-
"step":
|
| 2187 |
},
|
| 2188 |
{
|
| 2189 |
-
"epoch": 14.
|
| 2190 |
-
"
|
| 2191 |
-
"
|
| 2192 |
-
"
|
| 2193 |
-
"
|
|
|
|
| 2194 |
},
|
| 2195 |
{
|
| 2196 |
-
"epoch": 14.
|
| 2197 |
-
"grad_norm": 0.
|
| 2198 |
"learning_rate": 1.5979216850509848e-06,
|
| 2199 |
-
"loss": 0.
|
| 2200 |
-
"step":
|
| 2201 |
-
},
|
| 2202 |
-
{
|
| 2203 |
-
"epoch": 14.210526315789474,
|
| 2204 |
-
"grad_norm": 0.345703125,
|
| 2205 |
-
"learning_rate": 1.404877791536141e-06,
|
| 2206 |
-
"loss": 0.8934,
|
| 2207 |
-
"step": 1485
|
| 2208 |
},
|
| 2209 |
{
|
| 2210 |
-
"epoch": 14.
|
| 2211 |
-
"grad_norm": 0.
|
| 2212 |
"learning_rate": 1.2241750709546917e-06,
|
| 2213 |
-
"loss": 0.
|
| 2214 |
-
"step":
|
| 2215 |
-
},
|
| 2216 |
-
{
|
| 2217 |
-
"epoch": 14.30622009569378,
|
| 2218 |
-
"grad_norm": 0.341796875,
|
| 2219 |
-
"learning_rate": 1.055836141905553e-06,
|
| 2220 |
-
"loss": 0.8992,
|
| 2221 |
-
"step": 1495
|
| 2222 |
},
|
| 2223 |
{
|
| 2224 |
-
"epoch": 14.
|
| 2225 |
-
"grad_norm": 0.
|
| 2226 |
"learning_rate": 8.998820754091531e-07,
|
| 2227 |
-
"loss": 0.
|
| 2228 |
-
"step":
|
| 2229 |
-
},
|
| 2230 |
-
{
|
| 2231 |
-
"epoch": 14.401913875598087,
|
| 2232 |
-
"grad_norm": 0.3515625,
|
| 2233 |
-
"learning_rate": 7.563323922699983e-07,
|
| 2234 |
-
"loss": 0.8897,
|
| 2235 |
-
"step": 1505
|
| 2236 |
},
|
| 2237 |
{
|
| 2238 |
-
"epoch": 14.
|
| 2239 |
-
"grad_norm": 0.
|
| 2240 |
"learning_rate": 6.25205060633205e-07,
|
| 2241 |
-
"loss": 0.
|
| 2242 |
-
"step":
|
| 2243 |
-
},
|
| 2244 |
-
{
|
| 2245 |
-
"epoch": 14.497607655502392,
|
| 2246 |
-
"grad_norm": 0.3359375,
|
| 2247 |
-
"learning_rate": 5.065164937354428e-07,
|
| 2248 |
-
"loss": 0.9045,
|
| 2249 |
-
"step": 1515
|
| 2250 |
},
|
| 2251 |
{
|
| 2252 |
-
"epoch": 14.
|
| 2253 |
-
"grad_norm": 0.
|
| 2254 |
"learning_rate": 4.0028154785050063e-07,
|
| 2255 |
-
"loss": 0.
|
| 2256 |
-
"step":
|
| 2257 |
-
},
|
| 2258 |
-
{
|
| 2259 |
-
"epoch": 14.593301435406698,
|
| 2260 |
-
"grad_norm": 0.3515625,
|
| 2261 |
-
"learning_rate": 3.065135204296965e-07,
|
| 2262 |
-
"loss": 0.8995,
|
| 2263 |
-
"step": 1525
|
| 2264 |
},
|
| 2265 |
{
|
| 2266 |
-
"epoch": 14.
|
| 2267 |
-
"grad_norm": 0.
|
| 2268 |
"learning_rate": 2.2522414843748618e-07,
|
| 2269 |
-
"loss": 0.
|
| 2270 |
-
"step":
|
| 2271 |
-
},
|
| 2272 |
-
{
|
| 2273 |
-
"epoch": 14.688995215311005,
|
| 2274 |
-
"grad_norm": 0.353515625,
|
| 2275 |
-
"learning_rate": 1.5642360688225e-07,
|
| 2276 |
-
"loss": 0.8905,
|
| 2277 |
-
"step": 1535
|
| 2278 |
},
|
| 2279 |
{
|
| 2280 |
-
"epoch": 14.
|
| 2281 |
-
"grad_norm": 0.
|
| 2282 |
"learning_rate": 1.0012050754277802e-07,
|
| 2283 |
-
"loss": 0.
|
| 2284 |
-
"step":
|
| 2285 |
-
},
|
| 2286 |
-
{
|
| 2287 |
-
"epoch": 14.78468899521531,
|
| 2288 |
-
"grad_norm": 0.34375,
|
| 2289 |
-
"learning_rate": 5.632189789027687e-08,
|
| 2290 |
-
"loss": 0.8943,
|
| 2291 |
-
"step": 1545
|
| 2292 |
},
|
| 2293 |
{
|
| 2294 |
-
"epoch": 14.
|
| 2295 |
-
"grad_norm": 0.
|
| 2296 |
"learning_rate": 2.5033260206275277e-08,
|
| 2297 |
-
"loss": 0.
|
| 2298 |
-
"step":
|
| 2299 |
-
},
|
| 2300 |
-
{
|
| 2301 |
-
"epoch": 14.880382775119617,
|
| 2302 |
-
"grad_norm": 0.326171875,
|
| 2303 |
-
"learning_rate": 6.25851089636198e-09,
|
| 2304 |
-
"loss": 0.8944,
|
| 2305 |
-
"step": 1555
|
| 2306 |
},
|
| 2307 |
{
|
| 2308 |
-
"epoch": 14.
|
| 2309 |
-
"grad_norm": 0.
|
| 2310 |
"learning_rate": 0.0,
|
| 2311 |
-
"loss": 0.
|
| 2312 |
-
"step":
|
| 2313 |
},
|
| 2314 |
{
|
| 2315 |
-
"epoch": 14.
|
| 2316 |
-
"eval_loss": 2.
|
| 2317 |
-
"eval_runtime": 0.
|
| 2318 |
-
"eval_samples_per_second":
|
| 2319 |
-
"eval_steps_per_second":
|
| 2320 |
-
"step":
|
| 2321 |
},
|
| 2322 |
{
|
| 2323 |
-
"epoch": 14.
|
| 2324 |
-
"step":
|
| 2325 |
-
"total_flos": 1.
|
| 2326 |
-
"train_loss": 1.
|
| 2327 |
-
"train_runtime":
|
| 2328 |
-
"train_samples_per_second":
|
| 2329 |
-
"train_steps_per_second": 0.
|
| 2330 |
}
|
| 2331 |
],
|
| 2332 |
"logging_steps": 5,
|
| 2333 |
-
"max_steps":
|
| 2334 |
"num_input_tokens_seen": 0,
|
| 2335 |
"num_train_epochs": 15,
|
| 2336 |
"save_steps": 100,
|
| 2337 |
-
"
|
| 2338 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2339 |
"trial_name": null,
|
| 2340 |
"trial_params": null
|
| 2341 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 14.857142857142858,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 780,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.01904761904761905,
|
| 13 |
+
"grad_norm": 3.15625,
|
| 14 |
+
"learning_rate": 2.564102564102564e-06,
|
| 15 |
+
"loss": 3.3205,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.09523809523809523,
|
| 20 |
+
"grad_norm": 4.34375,
|
| 21 |
+
"learning_rate": 1.282051282051282e-05,
|
| 22 |
+
"loss": 3.3126,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.19047619047619047,
|
| 27 |
+
"grad_norm": 2.609375,
|
| 28 |
+
"learning_rate": 2.564102564102564e-05,
|
| 29 |
+
"loss": 3.2719,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.2857142857142857,
|
| 34 |
+
"grad_norm": 2.828125,
|
| 35 |
+
"learning_rate": 3.846153846153846e-05,
|
| 36 |
+
"loss": 3.1205,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.38095238095238093,
|
| 41 |
+
"grad_norm": 2.03125,
|
| 42 |
+
"learning_rate": 5.128205128205128e-05,
|
| 43 |
+
"loss": 2.8114,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.47619047619047616,
|
| 48 |
+
"grad_norm": 2.125,
|
| 49 |
+
"learning_rate": 6.410256410256412e-05,
|
| 50 |
+
"loss": 2.5384,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.5714285714285714,
|
| 55 |
+
"grad_norm": 6.03125,
|
| 56 |
+
"learning_rate": 7.692307692307693e-05,
|
| 57 |
+
"loss": 2.3201,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.6666666666666666,
|
| 62 |
+
"grad_norm": 2.0625,
|
| 63 |
+
"learning_rate": 8.974358974358975e-05,
|
| 64 |
+
"loss": 2.1053,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.7619047619047619,
|
| 69 |
+
"grad_norm": 1.359375,
|
| 70 |
+
"learning_rate": 0.00010256410256410256,
|
| 71 |
+
"loss": 1.8717,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.8571428571428571,
|
| 76 |
+
"grad_norm": 0.54296875,
|
| 77 |
+
"learning_rate": 0.00011538461538461538,
|
| 78 |
+
"loss": 1.6592,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.9523809523809523,
|
| 83 |
+
"grad_norm": 0.427734375,
|
| 84 |
+
"learning_rate": 0.00012820512820512823,
|
| 85 |
+
"loss": 1.518,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.9904761904761905,
|
| 90 |
+
"eval_loss": 2.7708818912506104,
|
| 91 |
+
"eval_runtime": 0.4987,
|
| 92 |
+
"eval_samples_per_second": 42.112,
|
| 93 |
+
"eval_steps_per_second": 2.005,
|
| 94 |
+
"step": 52
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.0476190476190477,
|
| 98 |
+
"grad_norm": 0.396484375,
|
| 99 |
+
"learning_rate": 0.00014102564102564104,
|
| 100 |
+
"loss": 1.426,
|
| 101 |
"step": 55
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 1.1428571428571428,
|
| 105 |
+
"grad_norm": 0.51953125,
|
| 106 |
+
"learning_rate": 0.00015384615384615385,
|
| 107 |
+
"loss": 1.3513,
|
| 108 |
"step": 60
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 1.2380952380952381,
|
| 112 |
+
"grad_norm": 0.75390625,
|
| 113 |
+
"learning_rate": 0.0001666666666666667,
|
| 114 |
+
"loss": 1.3036,
|
| 115 |
"step": 65
|
| 116 |
},
|
| 117 |
{
|
| 118 |
+
"epoch": 1.3333333333333333,
|
| 119 |
+
"grad_norm": 0.703125,
|
| 120 |
+
"learning_rate": 0.0001794871794871795,
|
| 121 |
+
"loss": 1.2627,
|
| 122 |
"step": 70
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 1.4285714285714286,
|
| 126 |
+
"grad_norm": 0.6015625,
|
| 127 |
+
"learning_rate": 0.00019230769230769233,
|
| 128 |
+
"loss": 1.2335,
|
| 129 |
"step": 75
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 1.5238095238095237,
|
| 133 |
+
"grad_norm": 0.31640625,
|
| 134 |
+
"learning_rate": 0.00019999599453798524,
|
| 135 |
+
"loss": 1.2094,
|
| 136 |
"step": 80
|
| 137 |
},
|
| 138 |
{
|
| 139 |
+
"epoch": 1.619047619047619,
|
| 140 |
+
"grad_norm": 0.359375,
|
| 141 |
+
"learning_rate": 0.0001999509367752813,
|
| 142 |
+
"loss": 1.1957,
|
| 143 |
"step": 85
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 1.7142857142857144,
|
| 147 |
+
"grad_norm": 0.80859375,
|
| 148 |
+
"learning_rate": 0.00019985583705641418,
|
| 149 |
+
"loss": 1.1772,
|
| 150 |
"step": 90
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"epoch": 1.8095238095238095,
|
| 154 |
+
"grad_norm": 0.5,
|
| 155 |
+
"learning_rate": 0.0001997107429945041,
|
| 156 |
+
"loss": 1.1595,
|
| 157 |
"step": 95
|
| 158 |
},
|
| 159 |
{
|
| 160 |
+
"epoch": 1.9047619047619047,
|
| 161 |
+
"grad_norm": 0.48046875,
|
| 162 |
+
"learning_rate": 0.0001995157272330992,
|
| 163 |
+
"loss": 1.1451,
|
| 164 |
"step": 100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 2.0,
|
| 168 |
+
"grad_norm": 0.408203125,
|
| 169 |
+
"learning_rate": 0.0001992708874098054,
|
| 170 |
+
"loss": 1.1423,
|
| 171 |
+
"step": 105
|
|
|
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 2.0,
|
| 175 |
+
"eval_loss": 2.659493923187256,
|
| 176 |
+
"eval_runtime": 0.4877,
|
| 177 |
+
"eval_samples_per_second": 43.055,
|
| 178 |
+
"eval_steps_per_second": 2.05,
|
| 179 |
"step": 105
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 2.0952380952380953,
|
| 183 |
+
"grad_norm": 0.419921875,
|
| 184 |
+
"learning_rate": 0.00019897634610740287,
|
| 185 |
+
"loss": 1.1131,
|
| 186 |
"step": 110
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 2.1904761904761907,
|
| 190 |
+
"grad_norm": 0.50390625,
|
| 191 |
+
"learning_rate": 0.00019863225079247285,
|
| 192 |
+
"loss": 1.1121,
|
| 193 |
"step": 115
|
| 194 |
},
|
| 195 |
{
|
| 196 |
+
"epoch": 2.2857142857142856,
|
| 197 |
+
"grad_norm": 0.333984375,
|
| 198 |
+
"learning_rate": 0.00019823877374156647,
|
| 199 |
+
"loss": 1.103,
|
| 200 |
"step": 120
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 2.380952380952381,
|
| 204 |
+
"grad_norm": 0.5859375,
|
| 205 |
+
"learning_rate": 0.00019779611195495177,
|
| 206 |
+
"loss": 1.1023,
|
| 207 |
"step": 125
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 2.4761904761904763,
|
| 211 |
+
"grad_norm": 0.95703125,
|
| 212 |
+
"learning_rate": 0.00019730448705798239,
|
| 213 |
+
"loss": 1.086,
|
| 214 |
"step": 130
|
| 215 |
},
|
| 216 |
{
|
| 217 |
+
"epoch": 2.571428571428571,
|
| 218 |
+
"grad_norm": 0.392578125,
|
| 219 |
+
"learning_rate": 0.00019676414519013781,
|
| 220 |
+
"loss": 1.0886,
|
| 221 |
"step": 135
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 2.6666666666666665,
|
| 225 |
+
"grad_norm": 0.70703125,
|
| 226 |
+
"learning_rate": 0.0001961753568817896,
|
| 227 |
+
"loss": 1.0818,
|
| 228 |
"step": 140
|
| 229 |
},
|
| 230 |
{
|
| 231 |
+
"epoch": 2.761904761904762,
|
| 232 |
+
"grad_norm": 0.49609375,
|
| 233 |
+
"learning_rate": 0.0001955384169187563,
|
| 234 |
+
"loss": 1.0799,
|
| 235 |
"step": 145
|
| 236 |
},
|
| 237 |
{
|
| 238 |
+
"epoch": 2.857142857142857,
|
| 239 |
+
"grad_norm": 0.71484375,
|
| 240 |
+
"learning_rate": 0.00019485364419471454,
|
| 241 |
+
"loss": 1.0792,
|
| 242 |
"step": 150
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 2.9523809523809526,
|
| 246 |
+
"grad_norm": 0.267578125,
|
| 247 |
+
"learning_rate": 0.00019412138155154002,
|
| 248 |
+
"loss": 1.0681,
|
| 249 |
"step": 155
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 2.9904761904761905,
|
| 253 |
+
"eval_loss": 2.6406373977661133,
|
| 254 |
+
"eval_runtime": 0.4936,
|
| 255 |
+
"eval_samples_per_second": 42.54,
|
| 256 |
+
"eval_steps_per_second": 2.026,
|
| 257 |
+
"step": 157
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.0476190476190474,
|
| 261 |
+
"grad_norm": 0.291015625,
|
| 262 |
+
"learning_rate": 0.0001933419956076584,
|
| 263 |
+
"loss": 1.0607,
|
| 264 |
"step": 160
|
| 265 |
},
|
| 266 |
{
|
| 267 |
+
"epoch": 3.142857142857143,
|
| 268 |
+
"grad_norm": 0.384765625,
|
| 269 |
+
"learning_rate": 0.00019251587657449236,
|
| 270 |
+
"loss": 1.0467,
|
| 271 |
"step": 165
|
| 272 |
},
|
| 273 |
{
|
| 274 |
+
"epoch": 3.238095238095238,
|
| 275 |
+
"grad_norm": 0.2578125,
|
| 276 |
+
"learning_rate": 0.00019164343806109632,
|
| 277 |
+
"loss": 1.0429,
|
| 278 |
"step": 170
|
| 279 |
},
|
| 280 |
{
|
| 281 |
+
"epoch": 3.3333333333333335,
|
| 282 |
+
"grad_norm": 0.28125,
|
| 283 |
+
"learning_rate": 0.00019072511686707663,
|
| 284 |
+
"loss": 1.0477,
|
| 285 |
"step": 175
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 3.4285714285714284,
|
| 289 |
+
"grad_norm": 0.23046875,
|
| 290 |
+
"learning_rate": 0.0001897613727639014,
|
| 291 |
+
"loss": 1.0407,
|
| 292 |
"step": 180
|
| 293 |
},
|
| 294 |
{
|
| 295 |
+
"epoch": 3.5238095238095237,
|
| 296 |
+
"grad_norm": 0.390625,
|
| 297 |
+
"learning_rate": 0.00018875268826470872,
|
| 298 |
+
"loss": 1.0408,
|
| 299 |
"step": 185
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 3.619047619047619,
|
| 303 |
+
"grad_norm": 0.3046875,
|
| 304 |
+
"learning_rate": 0.00018769956838272936,
|
| 305 |
+
"loss": 1.0392,
|
| 306 |
"step": 190
|
| 307 |
},
|
| 308 |
{
|
| 309 |
+
"epoch": 3.7142857142857144,
|
| 310 |
+
"grad_norm": 0.3203125,
|
| 311 |
+
"learning_rate": 0.00018660254037844388,
|
| 312 |
+
"loss": 1.0393,
|
| 313 |
"step": 195
|
| 314 |
},
|
| 315 |
{
|
| 316 |
+
"epoch": 3.8095238095238093,
|
| 317 |
+
"grad_norm": 0.32421875,
|
| 318 |
+
"learning_rate": 0.00018546215349560203,
|
| 319 |
+
"loss": 1.0385,
|
| 320 |
"step": 200
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"epoch": 3.9047619047619047,
|
| 324 |
+
"grad_norm": 0.828125,
|
| 325 |
+
"learning_rate": 0.00018427897868623534,
|
| 326 |
+
"loss": 1.0383,
|
| 327 |
"step": 205
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 4.0,
|
| 331 |
+
"grad_norm": 0.4375,
|
| 332 |
+
"learning_rate": 0.00018305360832480117,
|
| 333 |
+
"loss": 1.0335,
|
| 334 |
+
"step": 210
|
|
|
|
| 335 |
},
|
| 336 |
{
|
| 337 |
+
"epoch": 4.0,
|
| 338 |
+
"eval_loss": 2.642651319503784,
|
| 339 |
+
"eval_runtime": 0.4862,
|
| 340 |
+
"eval_samples_per_second": 43.193,
|
| 341 |
+
"eval_steps_per_second": 2.057,
|
| 342 |
"step": 210
|
| 343 |
},
|
| 344 |
{
|
| 345 |
+
"epoch": 4.095238095238095,
|
| 346 |
+
"grad_norm": 0.291015625,
|
| 347 |
+
"learning_rate": 0.00018178665591160172,
|
| 348 |
+
"loss": 1.0188,
|
| 349 |
"step": 215
|
| 350 |
},
|
| 351 |
{
|
| 352 |
+
"epoch": 4.190476190476191,
|
| 353 |
+
"grad_norm": 0.27734375,
|
| 354 |
+
"learning_rate": 0.00018047875576562557,
|
| 355 |
+
"loss": 1.0134,
|
| 356 |
"step": 220
|
| 357 |
},
|
| 358 |
{
|
| 359 |
+
"epoch": 4.285714285714286,
|
| 360 |
+
"grad_norm": 0.255859375,
|
| 361 |
+
"learning_rate": 0.0001791305627069662,
|
| 362 |
+
"loss": 1.0109,
|
| 363 |
"step": 225
|
| 364 |
},
|
| 365 |
{
|
| 366 |
+
"epoch": 4.380952380952381,
|
| 367 |
+
"grad_norm": 0.265625,
|
| 368 |
+
"learning_rate": 0.0001777427517289766,
|
| 369 |
+
"loss": 1.009,
|
| 370 |
"step": 230
|
| 371 |
},
|
| 372 |
{
|
| 373 |
+
"epoch": 4.476190476190476,
|
| 374 |
+
"grad_norm": 0.2890625,
|
| 375 |
+
"learning_rate": 0.00017631601766032336,
|
| 376 |
+
"loss": 1.0116,
|
| 377 |
"step": 235
|
| 378 |
},
|
| 379 |
{
|
| 380 |
+
"epoch": 4.571428571428571,
|
| 381 |
+
"grad_norm": 0.5,
|
| 382 |
+
"learning_rate": 0.00017485107481711012,
|
| 383 |
+
"loss": 1.0067,
|
| 384 |
"step": 240
|
| 385 |
},
|
| 386 |
{
|
| 387 |
+
"epoch": 4.666666666666667,
|
| 388 |
+
"grad_norm": 0.58203125,
|
| 389 |
+
"learning_rate": 0.0001733486566452446,
|
| 390 |
+
"loss": 1.0112,
|
| 391 |
"step": 245
|
| 392 |
},
|
| 393 |
{
|
| 394 |
+
"epoch": 4.761904761904762,
|
| 395 |
+
"grad_norm": 0.40234375,
|
| 396 |
+
"learning_rate": 0.0001718095153532274,
|
| 397 |
+
"loss": 1.0096,
|
| 398 |
"step": 250
|
| 399 |
},
|
| 400 |
{
|
| 401 |
+
"epoch": 4.857142857142857,
|
| 402 |
+
"grad_norm": 0.67578125,
|
| 403 |
+
"learning_rate": 0.00017023442153554777,
|
| 404 |
+
"loss": 1.0046,
|
| 405 |
"step": 255
|
| 406 |
},
|
| 407 |
{
|
| 408 |
+
"epoch": 4.9523809523809526,
|
| 409 |
+
"grad_norm": 0.76953125,
|
| 410 |
+
"learning_rate": 0.0001686241637868734,
|
| 411 |
+
"loss": 1.0079,
|
| 412 |
"step": 260
|
| 413 |
},
|
| 414 |
{
|
| 415 |
+
"epoch": 4.9904761904761905,
|
| 416 |
+
"eval_loss": 2.645949363708496,
|
| 417 |
+
"eval_runtime": 0.4992,
|
| 418 |
+
"eval_samples_per_second": 42.069,
|
| 419 |
+
"eval_steps_per_second": 2.003,
|
| 420 |
+
"step": 262
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 5.0476190476190474,
|
| 424 |
+
"grad_norm": 0.310546875,
|
| 425 |
+
"learning_rate": 0.00016697954830722868,
|
| 426 |
+
"loss": 0.9948,
|
| 427 |
"step": 265
|
| 428 |
},
|
| 429 |
{
|
| 430 |
+
"epoch": 5.142857142857143,
|
| 431 |
+
"grad_norm": 0.34375,
|
| 432 |
+
"learning_rate": 0.0001653013984983585,
|
| 433 |
+
"loss": 0.9868,
|
| 434 |
"step": 270
|
| 435 |
},
|
| 436 |
{
|
| 437 |
+
"epoch": 5.238095238095238,
|
| 438 |
+
"grad_norm": 0.5625,
|
| 439 |
+
"learning_rate": 0.0001635905545514795,
|
| 440 |
+
"loss": 0.9845,
|
| 441 |
"step": 275
|
| 442 |
},
|
| 443 |
{
|
| 444 |
+
"epoch": 5.333333333333333,
|
| 445 |
+
"grad_norm": 0.5859375,
|
| 446 |
+
"learning_rate": 0.0001618478730266255,
|
| 447 |
+
"loss": 0.9812,
|
| 448 |
"step": 280
|
| 449 |
},
|
| 450 |
{
|
| 451 |
+
"epoch": 5.428571428571429,
|
| 452 |
+
"grad_norm": 0.65625,
|
| 453 |
+
"learning_rate": 0.0001600742264237979,
|
| 454 |
+
"loss": 0.9887,
|
| 455 |
"step": 285
|
| 456 |
},
|
| 457 |
{
|
| 458 |
+
"epoch": 5.523809523809524,
|
| 459 |
+
"grad_norm": 0.357421875,
|
| 460 |
+
"learning_rate": 0.00015827050274613513,
|
| 461 |
+
"loss": 0.9863,
|
| 462 |
"step": 290
|
| 463 |
},
|
| 464 |
{
|
| 465 |
+
"epoch": 5.619047619047619,
|
| 466 |
+
"grad_norm": 0.30078125,
|
| 467 |
+
"learning_rate": 0.0001564376050553205,
|
| 468 |
+
"loss": 0.9862,
|
| 469 |
"step": 295
|
| 470 |
},
|
| 471 |
{
|
| 472 |
+
"epoch": 5.714285714285714,
|
| 473 |
+
"grad_norm": 0.275390625,
|
| 474 |
+
"learning_rate": 0.00015457645101945046,
|
| 475 |
+
"loss": 0.9865,
|
| 476 |
"step": 300
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 5.809523809523809,
|
| 480 |
+
"grad_norm": 0.337890625,
|
| 481 |
+
"learning_rate": 0.00015268797245359035,
|
| 482 |
+
"loss": 0.9929,
|
| 483 |
"step": 305
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 5.904761904761905,
|
| 487 |
+
"grad_norm": 0.294921875,
|
| 488 |
+
"learning_rate": 0.0001507731148532468,
|
| 489 |
+
"loss": 0.9868,
|
| 490 |
"step": 310
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"epoch": 6.0,
|
| 494 |
+
"grad_norm": 0.32421875,
|
| 495 |
+
"learning_rate": 0.00014883283692099112,
|
| 496 |
+
"loss": 0.9837,
|
| 497 |
+
"step": 315
|
|
|
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 6.0,
|
| 501 |
+
"eval_loss": 2.657383441925049,
|
| 502 |
+
"eval_runtime": 0.4893,
|
| 503 |
+
"eval_samples_per_second": 42.915,
|
| 504 |
+
"eval_steps_per_second": 2.044,
|
| 505 |
"step": 315
|
| 506 |
},
|
| 507 |
{
|
| 508 |
+
"epoch": 6.095238095238095,
|
| 509 |
+
"grad_norm": 0.294921875,
|
| 510 |
+
"learning_rate": 0.00014686811008647038,
|
| 511 |
+
"loss": 0.9669,
|
| 512 |
"step": 320
|
| 513 |
},
|
| 514 |
{
|
| 515 |
+
"epoch": 6.190476190476191,
|
| 516 |
+
"grad_norm": 0.287109375,
|
| 517 |
+
"learning_rate": 0.00014487991802004623,
|
| 518 |
+
"loss": 0.9674,
|
| 519 |
"step": 325
|
| 520 |
},
|
| 521 |
{
|
| 522 |
+
"epoch": 6.285714285714286,
|
| 523 |
+
"grad_norm": 0.3046875,
|
| 524 |
+
"learning_rate": 0.00014286925614030542,
|
| 525 |
+
"loss": 0.9712,
|
| 526 |
"step": 330
|
| 527 |
},
|
| 528 |
{
|
| 529 |
+
"epoch": 6.380952380952381,
|
| 530 |
+
"grad_norm": 0.28515625,
|
| 531 |
+
"learning_rate": 0.00014083713111568842,
|
| 532 |
+
"loss": 0.9663,
|
| 533 |
"step": 335
|
| 534 |
},
|
| 535 |
{
|
| 536 |
+
"epoch": 6.476190476190476,
|
| 537 |
+
"grad_norm": 0.462890625,
|
| 538 |
+
"learning_rate": 0.0001387845603604855,
|
| 539 |
+
"loss": 0.9674,
|
| 540 |
"step": 340
|
| 541 |
},
|
| 542 |
{
|
| 543 |
+
"epoch": 6.571428571428571,
|
| 544 |
+
"grad_norm": 0.54296875,
|
| 545 |
+
"learning_rate": 0.00013671257152545277,
|
| 546 |
+
"loss": 0.9732,
|
| 547 |
"step": 345
|
| 548 |
},
|
| 549 |
{
|
| 550 |
+
"epoch": 6.666666666666667,
|
| 551 |
+
"grad_norm": 0.490234375,
|
| 552 |
+
"learning_rate": 0.00013462220198330328,
|
| 553 |
+
"loss": 0.9743,
|
| 554 |
"step": 350
|
| 555 |
},
|
| 556 |
{
|
| 557 |
+
"epoch": 6.761904761904762,
|
| 558 |
+
"grad_norm": 0.296875,
|
| 559 |
+
"learning_rate": 0.0001325144983093305,
|
| 560 |
+
"loss": 0.9704,
|
| 561 |
"step": 355
|
| 562 |
},
|
| 563 |
{
|
| 564 |
+
"epoch": 6.857142857142857,
|
| 565 |
+
"grad_norm": 0.326171875,
|
| 566 |
+
"learning_rate": 0.0001303905157574247,
|
| 567 |
+
"loss": 0.9617,
|
| 568 |
"step": 360
|
| 569 |
},
|
| 570 |
{
|
| 571 |
+
"epoch": 6.9523809523809526,
|
| 572 |
+
"grad_norm": 0.3984375,
|
| 573 |
+
"learning_rate": 0.0001282513177317437,
|
| 574 |
+
"loss": 0.966,
|
| 575 |
"step": 365
|
| 576 |
},
|
| 577 |
{
|
| 578 |
+
"epoch": 6.9904761904761905,
|
| 579 |
+
"eval_loss": 2.6700327396392822,
|
| 580 |
+
"eval_runtime": 0.5365,
|
| 581 |
+
"eval_samples_per_second": 39.146,
|
| 582 |
+
"eval_steps_per_second": 1.864,
|
| 583 |
+
"step": 367
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 7.0476190476190474,
|
| 587 |
+
"grad_norm": 0.294921875,
|
| 588 |
+
"learning_rate": 0.00012609797525430373,
|
| 589 |
+
"loss": 0.9592,
|
| 590 |
"step": 370
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 7.142857142857143,
|
| 594 |
+
"grad_norm": 0.3203125,
|
| 595 |
+
"learning_rate": 0.0001239315664287558,
|
| 596 |
+
"loss": 0.9535,
|
| 597 |
"step": 375
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 7.238095238095238,
|
| 601 |
+
"grad_norm": 0.26953125,
|
| 602 |
+
"learning_rate": 0.00012175317590061674,
|
| 603 |
+
"loss": 0.9528,
|
| 604 |
"step": 380
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 7.333333333333333,
|
| 608 |
+
"grad_norm": 0.416015625,
|
| 609 |
+
"learning_rate": 0.00011956389431422507,
|
| 610 |
+
"loss": 0.9572,
|
| 611 |
"step": 385
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 7.428571428571429,
|
| 615 |
+
"grad_norm": 0.341796875,
|
| 616 |
+
"learning_rate": 0.00011736481776669306,
|
| 617 |
+
"loss": 0.9461,
|
| 618 |
"step": 390
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 7.523809523809524,
|
| 622 |
+
"grad_norm": 0.30078125,
|
| 623 |
+
"learning_rate": 0.00011515704725912926,
|
| 624 |
+
"loss": 0.9513,
|
| 625 |
"step": 395
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 7.619047619047619,
|
| 629 |
+
"grad_norm": 0.3125,
|
| 630 |
+
"learning_rate": 0.00011294168814540553,
|
| 631 |
+
"loss": 0.9566,
|
| 632 |
"step": 400
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 7.714285714285714,
|
| 636 |
+
"grad_norm": 0.609375,
|
| 637 |
+
"learning_rate": 0.00011071984957874479,
|
| 638 |
+
"loss": 0.9594,
|
| 639 |
"step": 405
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 7.809523809523809,
|
| 643 |
+
"grad_norm": 0.296875,
|
| 644 |
+
"learning_rate": 0.00010849264395640649,
|
| 645 |
+
"loss": 0.9554,
|
| 646 |
"step": 410
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"epoch": 7.904761904761905,
|
| 650 |
+
"grad_norm": 0.33203125,
|
| 651 |
+
"learning_rate": 0.0001062611863627482,
|
| 652 |
+
"loss": 0.9479,
|
| 653 |
"step": 415
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 8.0,
|
| 657 |
+
"grad_norm": 0.353515625,
|
| 658 |
+
"learning_rate": 0.00010402659401094152,
|
| 659 |
+
"loss": 0.9474,
|
| 660 |
+
"step": 420
|
|
|
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 8.0,
|
| 664 |
+
"eval_loss": 2.6798880100250244,
|
| 665 |
+
"eval_runtime": 0.4848,
|
| 666 |
+
"eval_samples_per_second": 43.314,
|
| 667 |
+
"eval_steps_per_second": 2.063,
|
| 668 |
"step": 420
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 8.095238095238095,
|
| 672 |
+
"grad_norm": 0.404296875,
|
| 673 |
+
"learning_rate": 0.00010178998568362243,
|
| 674 |
+
"loss": 0.9432,
|
| 675 |
"step": 425
|
| 676 |
},
|
| 677 |
{
|
| 678 |
+
"epoch": 8.19047619047619,
|
| 679 |
+
"grad_norm": 0.26953125,
|
| 680 |
+
"learning_rate": 9.955248117275566e-05,
|
| 681 |
+
"loss": 0.9398,
|
| 682 |
"step": 430
|
| 683 |
},
|
| 684 |
{
|
| 685 |
+
"epoch": 8.285714285714286,
|
| 686 |
+
"grad_norm": 0.326171875,
|
| 687 |
+
"learning_rate": 9.73152007189939e-05,
|
| 688 |
+
"loss": 0.9348,
|
| 689 |
"step": 435
|
| 690 |
},
|
| 691 |
{
|
| 692 |
+
"epoch": 8.380952380952381,
|
| 693 |
+
"grad_norm": 0.30078125,
|
| 694 |
+
"learning_rate": 9.507926445081219e-05,
|
| 695 |
+
"loss": 0.9416,
|
| 696 |
"step": 440
|
| 697 |
},
|
| 698 |
{
|
| 699 |
+
"epoch": 8.476190476190476,
|
| 700 |
+
"grad_norm": 0.28515625,
|
| 701 |
+
"learning_rate": 9.284579182369867e-05,
|
| 702 |
+
"loss": 0.9352,
|
| 703 |
"step": 445
|
| 704 |
},
|
| 705 |
{
|
| 706 |
+
"epoch": 8.571428571428571,
|
| 707 |
+
"grad_norm": 0.283203125,
|
| 708 |
+
"learning_rate": 9.061590105968208e-05,
|
| 709 |
+
"loss": 0.9427,
|
| 710 |
"step": 450
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"epoch": 8.666666666666666,
|
| 714 |
+
"grad_norm": 0.283203125,
|
| 715 |
+
"learning_rate": 8.839070858747697e-05,
|
| 716 |
+
"loss": 0.9413,
|
| 717 |
"step": 455
|
| 718 |
},
|
| 719 |
{
|
| 720 |
+
"epoch": 8.761904761904763,
|
| 721 |
+
"grad_norm": 0.294921875,
|
| 722 |
+
"learning_rate": 8.617132848352671e-05,
|
| 723 |
+
"loss": 0.9409,
|
| 724 |
"step": 460
|
| 725 |
},
|
| 726 |
{
|
| 727 |
+
"epoch": 8.857142857142858,
|
| 728 |
+
"grad_norm": 0.2890625,
|
| 729 |
+
"learning_rate": 8.395887191422397e-05,
|
| 730 |
+
"loss": 0.942,
|
| 731 |
"step": 465
|
| 732 |
},
|
| 733 |
{
|
| 734 |
+
"epoch": 8.952380952380953,
|
| 735 |
+
"grad_norm": 0.298828125,
|
| 736 |
+
"learning_rate": 8.175444657958876e-05,
|
| 737 |
+
"loss": 0.9406,
|
| 738 |
"step": 470
|
| 739 |
},
|
| 740 |
{
|
| 741 |
+
"epoch": 8.99047619047619,
|
| 742 |
+
"eval_loss": 2.688331365585327,
|
| 743 |
+
"eval_runtime": 0.5482,
|
| 744 |
+
"eval_samples_per_second": 38.309,
|
| 745 |
+
"eval_steps_per_second": 1.824,
|
| 746 |
+
"step": 472
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 9.047619047619047,
|
| 750 |
+
"grad_norm": 0.302734375,
|
| 751 |
+
"learning_rate": 7.955915615868111e-05,
|
| 752 |
+
"loss": 0.9375,
|
| 753 |
"step": 475
|
| 754 |
},
|
| 755 |
{
|
| 756 |
+
"epoch": 9.142857142857142,
|
| 757 |
+
"grad_norm": 0.275390625,
|
| 758 |
+
"learning_rate": 7.73740997570278e-05,
|
| 759 |
+
"loss": 0.9322,
|
| 760 |
"step": 480
|
| 761 |
},
|
| 762 |
{
|
| 763 |
+
"epoch": 9.238095238095237,
|
| 764 |
+
"grad_norm": 0.26171875,
|
| 765 |
+
"learning_rate": 7.520037135633816e-05,
|
| 766 |
+
"loss": 0.9292,
|
| 767 |
"step": 485
|
| 768 |
},
|
| 769 |
{
|
| 770 |
+
"epoch": 9.333333333333334,
|
| 771 |
+
"grad_norm": 0.287109375,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 772 |
"learning_rate": 7.303905926678564e-05,
|
| 773 |
+
"loss": 0.9269,
|
| 774 |
+
"step": 490
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 775 |
},
|
| 776 |
{
|
| 777 |
+
"epoch": 9.428571428571429,
|
| 778 |
+
"grad_norm": 0.341796875,
|
| 779 |
"learning_rate": 7.089124558212871e-05,
|
| 780 |
+
"loss": 0.9331,
|
| 781 |
+
"step": 495
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 782 |
},
|
| 783 |
{
|
| 784 |
+
"epoch": 9.523809523809524,
|
| 785 |
+
"grad_norm": 0.28125,
|
| 786 |
"learning_rate": 6.875800563794425e-05,
|
| 787 |
+
"loss": 0.9344,
|
| 788 |
+
"step": 500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 789 |
},
|
| 790 |
{
|
| 791 |
+
"epoch": 9.619047619047619,
|
| 792 |
+
"grad_norm": 0.28515625,
|
| 793 |
"learning_rate": 6.664040747324437e-05,
|
| 794 |
+
"loss": 0.9294,
|
| 795 |
+
"step": 505
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 796 |
},
|
| 797 |
{
|
| 798 |
+
"epoch": 9.714285714285714,
|
| 799 |
+
"grad_norm": 0.267578125,
|
| 800 |
"learning_rate": 6.453951129574644e-05,
|
| 801 |
+
"loss": 0.9347,
|
| 802 |
+
"step": 510
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 803 |
},
|
| 804 |
{
|
| 805 |
+
"epoch": 9.80952380952381,
|
| 806 |
+
"grad_norm": 0.30078125,
|
| 807 |
"learning_rate": 6.245636895106402e-05,
|
| 808 |
+
"loss": 0.9282,
|
| 809 |
+
"step": 515
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 810 |
},
|
| 811 |
{
|
| 812 |
+
"epoch": 9.904761904761905,
|
| 813 |
+
"grad_norm": 0.353515625,
|
| 814 |
"learning_rate": 6.039202339608432e-05,
|
| 815 |
+
"loss": 0.9299,
|
| 816 |
+
"step": 520
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 817 |
},
|
| 818 |
{
|
| 819 |
"epoch": 10.0,
|
| 820 |
+
"grad_norm": 0.27734375,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 821 |
"learning_rate": 5.834750817679606e-05,
|
| 822 |
+
"loss": 0.9245,
|
| 823 |
+
"step": 525
|
| 824 |
},
|
| 825 |
{
|
| 826 |
+
"epoch": 10.0,
|
| 827 |
+
"eval_loss": 2.6975326538085938,
|
| 828 |
+
"eval_runtime": 0.485,
|
| 829 |
+
"eval_samples_per_second": 43.303,
|
| 830 |
+
"eval_steps_per_second": 2.062,
|
| 831 |
+
"step": 525
|
| 832 |
},
|
| 833 |
{
|
| 834 |
+
"epoch": 10.095238095238095,
|
| 835 |
+
"grad_norm": 0.30078125,
|
| 836 |
"learning_rate": 5.6323846910828735e-05,
|
| 837 |
+
"loss": 0.9233,
|
| 838 |
+
"step": 530
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 839 |
},
|
| 840 |
{
|
| 841 |
+
"epoch": 10.19047619047619,
|
| 842 |
+
"grad_norm": 0.265625,
|
| 843 |
"learning_rate": 5.432205277496327e-05,
|
| 844 |
+
"loss": 0.9235,
|
| 845 |
+
"step": 535
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 846 |
},
|
| 847 |
{
|
| 848 |
+
"epoch": 10.285714285714286,
|
| 849 |
+
"grad_norm": 0.2734375,
|
| 850 |
"learning_rate": 5.234312799786921e-05,
|
| 851 |
+
"loss": 0.9194,
|
| 852 |
+
"step": 540
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 853 |
},
|
| 854 |
{
|
| 855 |
+
"epoch": 10.380952380952381,
|
| 856 |
+
"grad_norm": 0.29296875,
|
| 857 |
"learning_rate": 5.0388063358324134e-05,
|
| 858 |
+
"loss": 0.9235,
|
| 859 |
+
"step": 545
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 860 |
},
|
| 861 |
{
|
| 862 |
+
"epoch": 10.476190476190476,
|
| 863 |
+
"grad_norm": 0.279296875,
|
| 864 |
"learning_rate": 4.845783768916482e-05,
|
| 865 |
+
"loss": 0.9217,
|
| 866 |
+
"step": 550
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 867 |
},
|
| 868 |
{
|
| 869 |
+
"epoch": 10.571428571428571,
|
| 870 |
+
"grad_norm": 0.265625,
|
| 871 |
"learning_rate": 4.6553417387219886e-05,
|
| 872 |
+
"loss": 0.9243,
|
| 873 |
+
"step": 555
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 874 |
},
|
| 875 |
{
|
| 876 |
+
"epoch": 10.666666666666666,
|
| 877 |
+
"grad_norm": 0.306640625,
|
| 878 |
"learning_rate": 4.467575592946864e-05,
|
| 879 |
+
"loss": 0.9262,
|
| 880 |
+
"step": 560
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 881 |
},
|
| 882 |
{
|
| 883 |
+
"epoch": 10.761904761904763,
|
| 884 |
+
"grad_norm": 0.255859375,
|
| 885 |
"learning_rate": 4.282579339566802e-05,
|
| 886 |
+
"loss": 0.9206,
|
| 887 |
+
"step": 565
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 888 |
},
|
| 889 |
{
|
| 890 |
+
"epoch": 10.857142857142858,
|
| 891 |
+
"grad_norm": 0.2578125,
|
| 892 |
"learning_rate": 4.100445599768774e-05,
|
| 893 |
+
"loss": 0.9256,
|
| 894 |
+
"step": 570
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 895 |
},
|
| 896 |
{
|
| 897 |
+
"epoch": 10.952380952380953,
|
| 898 |
+
"grad_norm": 0.265625,
|
| 899 |
"learning_rate": 3.9212655615787804e-05,
|
| 900 |
+
"loss": 0.9208,
|
| 901 |
+
"step": 575
|
| 902 |
},
|
| 903 |
{
|
| 904 |
+
"epoch": 10.99047619047619,
|
| 905 |
+
"eval_loss": 2.707897663116455,
|
| 906 |
+
"eval_runtime": 0.647,
|
| 907 |
+
"eval_samples_per_second": 32.456,
|
| 908 |
+
"eval_steps_per_second": 1.546,
|
| 909 |
+
"step": 577
|
| 910 |
},
|
| 911 |
{
|
| 912 |
+
"epoch": 11.047619047619047,
|
| 913 |
+
"grad_norm": 0.25,
|
| 914 |
"learning_rate": 3.745128934207225e-05,
|
| 915 |
+
"loss": 0.9226,
|
| 916 |
+
"step": 580
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 917 |
},
|
| 918 |
{
|
| 919 |
+
"epoch": 11.142857142857142,
|
| 920 |
+
"grad_norm": 0.263671875,
|
| 921 |
"learning_rate": 3.5721239031346066e-05,
|
| 922 |
+
"loss": 0.9121,
|
| 923 |
+
"step": 585
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 924 |
},
|
| 925 |
{
|
| 926 |
+
"epoch": 11.238095238095237,
|
| 927 |
+
"grad_norm": 0.259765625,
|
| 928 |
"learning_rate": 3.402337085960119e-05,
|
| 929 |
+
"loss": 0.9155,
|
| 930 |
+
"step": 590
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 931 |
},
|
| 932 |
{
|
| 933 |
+
"epoch": 11.333333333333334,
|
| 934 |
+
"grad_norm": 0.2578125,
|
| 935 |
"learning_rate": 3.235853489035241e-05,
|
| 936 |
+
"loss": 0.9194,
|
| 937 |
+
"step": 595
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 938 |
},
|
| 939 |
{
|
| 940 |
+
"epoch": 11.428571428571429,
|
| 941 |
+
"grad_norm": 0.26953125,
|
| 942 |
"learning_rate": 3.072756464904006e-05,
|
| 943 |
+
"loss": 0.9202,
|
| 944 |
+
"step": 600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 945 |
},
|
| 946 |
{
|
| 947 |
+
"epoch": 11.523809523809524,
|
| 948 |
+
"grad_norm": 0.27734375,
|
| 949 |
"learning_rate": 2.9131276705713006e-05,
|
| 950 |
+
"loss": 0.9186,
|
| 951 |
+
"step": 605
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 952 |
},
|
| 953 |
{
|
| 954 |
+
"epoch": 11.619047619047619,
|
| 955 |
+
"grad_norm": 0.259765625,
|
| 956 |
"learning_rate": 2.7570470266200176e-05,
|
| 957 |
+
"loss": 0.9195,
|
| 958 |
+
"step": 610
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 959 |
},
|
| 960 |
{
|
| 961 |
+
"epoch": 11.714285714285714,
|
| 962 |
+
"grad_norm": 0.2470703125,
|
| 963 |
"learning_rate": 2.6045926771976303e-05,
|
| 964 |
+
"loss": 0.9223,
|
| 965 |
+
"step": 615
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 966 |
},
|
| 967 |
{
|
| 968 |
+
"epoch": 11.80952380952381,
|
| 969 |
+
"grad_norm": 0.25,
|
| 970 |
"learning_rate": 2.4558409508920986e-05,
|
| 971 |
+
"loss": 0.9139,
|
| 972 |
+
"step": 620
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 973 |
},
|
| 974 |
{
|
| 975 |
+
"epoch": 11.904761904761905,
|
| 976 |
+
"grad_norm": 0.248046875,
|
| 977 |
"learning_rate": 2.3108663225168435e-05,
|
| 978 |
+
"loss": 0.9117,
|
| 979 |
+
"step": 625
|
| 980 |
},
|
| 981 |
{
|
| 982 |
"epoch": 12.0,
|
| 983 |
+
"grad_norm": 0.263671875,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 984 |
"learning_rate": 2.1697413758237784e-05,
|
| 985 |
+
"loss": 0.9195,
|
| 986 |
+
"step": 630
|
| 987 |
},
|
| 988 |
{
|
| 989 |
+
"epoch": 12.0,
|
| 990 |
+
"eval_loss": 2.714789867401123,
|
| 991 |
+
"eval_runtime": 0.4906,
|
| 992 |
+
"eval_samples_per_second": 42.808,
|
| 993 |
+
"eval_steps_per_second": 2.038,
|
| 994 |
+
"step": 630
|
| 995 |
},
|
| 996 |
{
|
| 997 |
+
"epoch": 12.095238095238095,
|
| 998 |
+
"grad_norm": 0.2392578125,
|
| 999 |
"learning_rate": 2.032536767163141e-05,
|
| 1000 |
+
"loss": 0.9167,
|
| 1001 |
+
"step": 635
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
+
"epoch": 12.19047619047619,
|
| 1005 |
+
"grad_norm": 0.25,
|
| 1006 |
"learning_rate": 1.8993211901083353e-05,
|
| 1007 |
+
"loss": 0.9144,
|
| 1008 |
+
"step": 640
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1009 |
},
|
| 1010 |
{
|
| 1011 |
+
"epoch": 12.285714285714286,
|
| 1012 |
+
"grad_norm": 0.2578125,
|
| 1013 |
"learning_rate": 1.7701613410634365e-05,
|
| 1014 |
+
"loss": 0.915,
|
| 1015 |
+
"step": 645
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1016 |
},
|
| 1017 |
{
|
| 1018 |
+
"epoch": 12.380952380952381,
|
| 1019 |
+
"grad_norm": 0.248046875,
|
| 1020 |
"learning_rate": 1.6451218858706374e-05,
|
| 1021 |
+
"loss": 0.9153,
|
| 1022 |
+
"step": 650
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1023 |
},
|
| 1024 |
{
|
| 1025 |
+
"epoch": 12.476190476190476,
|
| 1026 |
+
"grad_norm": 0.251953125,
|
| 1027 |
"learning_rate": 1.5242654274342894e-05,
|
| 1028 |
+
"loss": 0.9133,
|
| 1029 |
+
"step": 655
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1030 |
},
|
| 1031 |
{
|
| 1032 |
+
"epoch": 12.571428571428571,
|
| 1033 |
+
"grad_norm": 0.25,
|
| 1034 |
"learning_rate": 1.4076524743778319e-05,
|
| 1035 |
+
"loss": 0.9134,
|
| 1036 |
+
"step": 660
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1037 |
},
|
| 1038 |
{
|
| 1039 |
+
"epoch": 12.666666666666666,
|
| 1040 |
+
"grad_norm": 0.248046875,
|
| 1041 |
"learning_rate": 1.295341410749208e-05,
|
| 1042 |
+
"loss": 0.9126,
|
| 1043 |
+
"step": 665
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1044 |
},
|
| 1045 |
{
|
| 1046 |
+
"epoch": 12.761904761904763,
|
| 1047 |
+
"grad_norm": 0.25,
|
| 1048 |
"learning_rate": 1.1873884667900125e-05,
|
| 1049 |
+
"loss": 0.9094,
|
| 1050 |
+
"step": 670
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
+
"epoch": 12.857142857142858,
|
| 1054 |
+
"grad_norm": 0.236328125,
|
| 1055 |
"learning_rate": 1.083847690782972e-05,
|
| 1056 |
+
"loss": 0.9183,
|
| 1057 |
+
"step": 675
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
+
"epoch": 12.952380952380953,
|
| 1061 |
+
"grad_norm": 0.255859375,
|
| 1062 |
"learning_rate": 9.8477092199184e-06,
|
| 1063 |
+
"loss": 0.9212,
|
| 1064 |
+
"step": 680
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
+
"epoch": 12.99047619047619,
|
| 1068 |
+
"eval_loss": 2.7153878211975098,
|
| 1069 |
+
"eval_runtime": 0.5024,
|
| 1070 |
+
"eval_samples_per_second": 41.796,
|
| 1071 |
+
"eval_steps_per_second": 1.99,
|
| 1072 |
+
"step": 682
|
| 1073 |
},
|
| 1074 |
{
|
| 1075 |
+
"epoch": 13.047619047619047,
|
| 1076 |
+
"grad_norm": 0.2431640625,
|
| 1077 |
"learning_rate": 8.902077647072881e-06,
|
| 1078 |
+
"loss": 0.9139,
|
| 1079 |
+
"step": 685
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
+
"epoch": 13.142857142857142,
|
| 1083 |
+
"grad_norm": 0.2431640625,
|
| 1084 |
"learning_rate": 8.002055634117578e-06,
|
| 1085 |
+
"loss": 0.9123,
|
| 1086 |
+
"step": 690
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1087 |
},
|
| 1088 |
{
|
| 1089 |
+
"epoch": 13.238095238095237,
|
| 1090 |
+
"grad_norm": 0.2470703125,
|
| 1091 |
"learning_rate": 7.148093790757371e-06,
|
| 1092 |
+
"loss": 0.9152,
|
| 1093 |
+
"step": 695
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
+
"epoch": 13.333333333333334,
|
| 1097 |
+
"grad_norm": 0.2578125,
|
| 1098 |
"learning_rate": 6.3406196659728465e-06,
|
| 1099 |
+
"loss": 0.9087,
|
| 1100 |
+
"step": 700
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
+
"epoch": 13.428571428571429,
|
| 1104 |
+
"grad_norm": 0.23828125,
|
| 1105 |
"learning_rate": 5.580037533961546e-06,
|
| 1106 |
+
"loss": 0.9212,
|
| 1107 |
+
"step": 705
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
+
"epoch": 13.523809523809524,
|
| 1111 |
+
"grad_norm": 0.2373046875,
|
| 1112 |
"learning_rate": 4.866728191731829e-06,
|
| 1113 |
+
"loss": 0.909,
|
| 1114 |
+
"step": 710
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1115 |
},
|
| 1116 |
{
|
| 1117 |
+
"epoch": 13.619047619047619,
|
| 1118 |
+
"grad_norm": 0.2431640625,
|
| 1119 |
"learning_rate": 4.20104876845111e-06,
|
| 1120 |
+
"loss": 0.9146,
|
| 1121 |
+
"step": 715
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1122 |
},
|
| 1123 |
{
|
| 1124 |
+
"epoch": 13.714285714285714,
|
| 1125 |
+
"grad_norm": 0.2431640625,
|
| 1126 |
"learning_rate": 3.5833325466437694e-06,
|
| 1127 |
+
"loss": 0.9107,
|
| 1128 |
+
"step": 720
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1129 |
},
|
| 1130 |
{
|
| 1131 |
+
"epoch": 13.80952380952381,
|
| 1132 |
+
"grad_norm": 0.2451171875,
|
| 1133 |
"learning_rate": 3.013888795328057e-06,
|
| 1134 |
+
"loss": 0.9136,
|
| 1135 |
+
"step": 725
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
+
"epoch": 13.904761904761905,
|
| 1139 |
+
"grad_norm": 0.236328125,
|
| 1140 |
"learning_rate": 2.4930026151759766e-06,
|
| 1141 |
+
"loss": 0.9147,
|
| 1142 |
+
"step": 730
|
| 1143 |
},
|
| 1144 |
{
|
| 1145 |
"epoch": 14.0,
|
| 1146 |
+
"grad_norm": 0.2470703125,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1147 |
"learning_rate": 2.0209347957732328e-06,
|
| 1148 |
+
"loss": 0.9136,
|
| 1149 |
+
"step": 735
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
+
"epoch": 14.0,
|
| 1153 |
+
"eval_loss": 2.7180798053741455,
|
| 1154 |
+
"eval_runtime": 0.4838,
|
| 1155 |
+
"eval_samples_per_second": 43.41,
|
| 1156 |
+
"eval_steps_per_second": 2.067,
|
| 1157 |
+
"step": 735
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
+
"epoch": 14.095238095238095,
|
| 1161 |
+
"grad_norm": 0.23828125,
|
| 1162 |
"learning_rate": 1.5979216850509848e-06,
|
| 1163 |
+
"loss": 0.9092,
|
| 1164 |
+
"step": 740
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
+
"epoch": 14.19047619047619,
|
| 1168 |
+
"grad_norm": 0.2578125,
|
| 1169 |
"learning_rate": 1.2241750709546917e-06,
|
| 1170 |
+
"loss": 0.9159,
|
| 1171 |
+
"step": 745
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
+
"epoch": 14.285714285714286,
|
| 1175 |
+
"grad_norm": 0.236328125,
|
| 1176 |
"learning_rate": 8.998820754091531e-07,
|
| 1177 |
+
"loss": 0.9153,
|
| 1178 |
+
"step": 750
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
+
"epoch": 14.380952380952381,
|
| 1182 |
+
"grad_norm": 0.244140625,
|
| 1183 |
"learning_rate": 6.25205060633205e-07,
|
| 1184 |
+
"loss": 0.9114,
|
| 1185 |
+
"step": 755
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
+
"epoch": 14.476190476190476,
|
| 1189 |
+
"grad_norm": 0.244140625,
|
| 1190 |
"learning_rate": 4.0028154785050063e-07,
|
| 1191 |
+
"loss": 0.9163,
|
| 1192 |
+
"step": 760
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1193 |
},
|
| 1194 |
{
|
| 1195 |
+
"epoch": 14.571428571428571,
|
| 1196 |
+
"grad_norm": 0.236328125,
|
| 1197 |
"learning_rate": 2.2522414843748618e-07,
|
| 1198 |
+
"loss": 0.9153,
|
| 1199 |
+
"step": 765
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
+
"epoch": 14.666666666666666,
|
| 1203 |
+
"grad_norm": 0.23828125,
|
| 1204 |
"learning_rate": 1.0012050754277802e-07,
|
| 1205 |
+
"loss": 0.912,
|
| 1206 |
+
"step": 770
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
+
"epoch": 14.761904761904763,
|
| 1210 |
+
"grad_norm": 0.25,
|
| 1211 |
"learning_rate": 2.5033260206275277e-08,
|
| 1212 |
+
"loss": 0.9155,
|
| 1213 |
+
"step": 775
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1214 |
},
|
| 1215 |
{
|
| 1216 |
+
"epoch": 14.857142857142858,
|
| 1217 |
+
"grad_norm": 0.240234375,
|
| 1218 |
"learning_rate": 0.0,
|
| 1219 |
+
"loss": 0.9103,
|
| 1220 |
+
"step": 780
|
| 1221 |
},
|
| 1222 |
{
|
| 1223 |
+
"epoch": 14.857142857142858,
|
| 1224 |
+
"eval_loss": 2.71852707862854,
|
| 1225 |
+
"eval_runtime": 0.4854,
|
| 1226 |
+
"eval_samples_per_second": 43.26,
|
| 1227 |
+
"eval_steps_per_second": 2.06,
|
| 1228 |
+
"step": 780
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
+
"epoch": 14.857142857142858,
|
| 1232 |
+
"step": 780,
|
| 1233 |
+
"total_flos": 1.2277516310308454e+18,
|
| 1234 |
+
"train_loss": 1.077159938445458,
|
| 1235 |
+
"train_runtime": 4175.1629,
|
| 1236 |
+
"train_samples_per_second": 47.883,
|
| 1237 |
+
"train_steps_per_second": 0.187
|
| 1238 |
}
|
| 1239 |
],
|
| 1240 |
"logging_steps": 5,
|
| 1241 |
+
"max_steps": 780,
|
| 1242 |
"num_input_tokens_seen": 0,
|
| 1243 |
"num_train_epochs": 15,
|
| 1244 |
"save_steps": 100,
|
| 1245 |
+
"stateful_callbacks": {
|
| 1246 |
+
"TrainerControl": {
|
| 1247 |
+
"args": {
|
| 1248 |
+
"should_epoch_stop": false,
|
| 1249 |
+
"should_evaluate": false,
|
| 1250 |
+
"should_log": false,
|
| 1251 |
+
"should_save": true,
|
| 1252 |
+
"should_training_stop": false
|
| 1253 |
+
},
|
| 1254 |
+
"attributes": {}
|
| 1255 |
+
}
|
| 1256 |
+
},
|
| 1257 |
+
"total_flos": 1.2277516310308454e+18,
|
| 1258 |
+
"train_batch_size": 16,
|
| 1259 |
"trial_name": null,
|
| 1260 |
"trial_params": null
|
| 1261 |
}
|