Model save
Browse files- README.md +16 -11
- all_results.json +7 -7
- model.safetensors +1 -1
- test_results.json +7 -7
- trainer_state.json +449 -231
- val_results.json +7 -7
README.md
CHANGED
|
@@ -18,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [facebook/timesformer-base-finetuned-k400](https://huggingface.co/facebook/timesformer-base-finetuned-k400) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
-
- F1: 0.
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
@@ -40,23 +40,28 @@ More information needed
|
|
| 40 |
|
| 41 |
The following hyperparameters were used during training:
|
| 42 |
- learning_rate: 5e-05
|
| 43 |
-
- train_batch_size:
|
| 44 |
-
- eval_batch_size:
|
| 45 |
- seed: 42
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
-
- training_steps:
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|
| 54 |
|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [facebook/timesformer-base-finetuned-k400](https://huggingface.co/facebook/timesformer-base-finetuned-k400) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.4027
|
| 22 |
+
- Accuracy: 0.8838
|
| 23 |
+
- F1: 0.8838
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 40 |
|
| 41 |
The following hyperparameters were used during training:
|
| 42 |
- learning_rate: 5e-05
|
| 43 |
+
- train_batch_size: 12
|
| 44 |
+
- eval_batch_size: 12
|
| 45 |
- seed: 42
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- training_steps: 610
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|
| 54 |
|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|
|
| 55 |
+
| 0.6712 | 0.1016 | 62 | 0.8671 | 0.6680 | 0.6623 |
|
| 56 |
+
| 0.3119 | 1.1016 | 124 | 0.5911 | 0.7884 | 0.7887 |
|
| 57 |
+
| 0.2505 | 2.1016 | 186 | 0.5297 | 0.8008 | 0.8002 |
|
| 58 |
+
| 0.207 | 3.1016 | 248 | 0.5970 | 0.7801 | 0.7787 |
|
| 59 |
+
| 0.1743 | 4.1016 | 310 | 0.5612 | 0.8050 | 0.7984 |
|
| 60 |
+
| 0.1005 | 5.1016 | 372 | 0.4027 | 0.8838 | 0.8838 |
|
| 61 |
+
| 0.0147 | 6.1016 | 434 | 0.4360 | 0.8589 | 0.8573 |
|
| 62 |
+
| 0.0573 | 7.1016 | 496 | 0.4451 | 0.8714 | 0.8697 |
|
| 63 |
+
| 0.0143 | 8.1016 | 558 | 0.4099 | 0.8672 | 0.8666 |
|
| 64 |
+
| 0.1311 | 9.0852 | 610 | 0.4056 | 0.8755 | 0.8752 |
|
| 65 |
|
| 66 |
|
| 67 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_f1": 0.
|
| 5 |
-
"eval_loss": 0.
|
| 6 |
-
"eval_runtime":
|
| 7 |
-
"eval_samples_per_second": 3.
|
| 8 |
-
"eval_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.085245901639345,
|
| 3 |
+
"eval_accuracy": 0.8838174273858921,
|
| 4 |
+
"eval_f1": 0.8838017754864972,
|
| 5 |
+
"eval_loss": 0.40273845195770264,
|
| 6 |
+
"eval_runtime": 76.8071,
|
| 7 |
+
"eval_samples_per_second": 3.138,
|
| 8 |
+
"eval_steps_per_second": 0.273
|
| 9 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 485078408
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6771e05fca54be28cdcd593bce1ae9495f8bb0e6f7177ed9683d50368ee125d1
|
| 3 |
size 485078408
|
test_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_f1": 0.
|
| 5 |
-
"eval_loss": 0.
|
| 6 |
-
"eval_runtime":
|
| 7 |
-
"eval_samples_per_second": 3.
|
| 8 |
-
"eval_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.085245901639345,
|
| 3 |
+
"eval_accuracy": 0.8337078651685393,
|
| 4 |
+
"eval_f1": 0.8322580739791249,
|
| 5 |
+
"eval_loss": 0.6123429536819458,
|
| 6 |
+
"eval_runtime": 140.7823,
|
| 7 |
+
"eval_samples_per_second": 3.161,
|
| 8 |
+
"eval_steps_per_second": 0.27
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,364 +1,582 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "videomae-surf-analytics-runpod/checkpoint-
|
| 4 |
-
"epoch":
|
| 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": 1.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 4.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate": 4.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"
|
| 56 |
-
"
|
| 57 |
-
"
|
| 58 |
-
"
|
|
|
|
|
|
|
|
|
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch":
|
| 62 |
-
"
|
| 63 |
-
"
|
| 64 |
-
"
|
| 65 |
-
"
|
| 66 |
-
"eval_samples_per_second": 3.126,
|
| 67 |
-
"eval_steps_per_second": 0.324,
|
| 68 |
-
"step": 75
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch": 1.
|
| 72 |
-
"grad_norm":
|
| 73 |
-
"learning_rate": 4.
|
| 74 |
-
"loss": 0.
|
| 75 |
"step": 80
|
| 76 |
},
|
| 77 |
{
|
| 78 |
-
"epoch": 1.
|
| 79 |
-
"grad_norm": 3.
|
| 80 |
-
"learning_rate": 4.
|
| 81 |
-
"loss": 0.
|
| 82 |
"step": 90
|
| 83 |
},
|
| 84 |
{
|
| 85 |
-
"epoch": 1.
|
| 86 |
-
"grad_norm":
|
| 87 |
-
"learning_rate": 4.
|
| 88 |
-
"loss": 0.
|
| 89 |
"step": 100
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch": 1.
|
| 93 |
-
"grad_norm":
|
| 94 |
-
"learning_rate":
|
| 95 |
-
"loss": 0.
|
| 96 |
"step": 110
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch": 1.
|
| 100 |
-
"grad_norm":
|
| 101 |
-
"learning_rate":
|
| 102 |
-
"loss": 0.
|
| 103 |
"step": 120
|
| 104 |
},
|
| 105 |
{
|
| 106 |
-
"epoch": 1.
|
| 107 |
-
"
|
| 108 |
-
"
|
| 109 |
-
"
|
| 110 |
-
"
|
|
|
|
|
|
|
|
|
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch":
|
| 114 |
-
"grad_norm":
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss": 0.
|
| 117 |
-
"step":
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch":
|
| 121 |
-
"grad_norm":
|
| 122 |
-
"learning_rate":
|
| 123 |
-
"loss": 0.
|
| 124 |
-
"step":
|
| 125 |
},
|
| 126 |
{
|
| 127 |
-
"epoch":
|
| 128 |
-
"
|
| 129 |
-
"
|
| 130 |
-
"
|
| 131 |
-
"eval_runtime": 78.2702,
|
| 132 |
-
"eval_samples_per_second": 3.079,
|
| 133 |
-
"eval_steps_per_second": 0.319,
|
| 134 |
"step": 150
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch": 2.
|
| 138 |
-
"grad_norm":
|
| 139 |
-
"learning_rate":
|
| 140 |
-
"loss": 0.
|
| 141 |
"step": 160
|
| 142 |
},
|
| 143 |
{
|
| 144 |
-
"epoch": 2.
|
| 145 |
-
"grad_norm":
|
| 146 |
-
"learning_rate":
|
| 147 |
-
"loss": 0.
|
| 148 |
"step": 170
|
| 149 |
},
|
| 150 |
{
|
| 151 |
-
"epoch": 2.
|
| 152 |
-
"grad_norm":
|
| 153 |
-
"learning_rate":
|
| 154 |
-
"loss": 0.
|
| 155 |
"step": 180
|
| 156 |
},
|
| 157 |
{
|
| 158 |
-
"epoch": 2.
|
| 159 |
-
"
|
| 160 |
-
"
|
| 161 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 162 |
"step": 190
|
| 163 |
},
|
| 164 |
{
|
| 165 |
-
"epoch":
|
| 166 |
-
"grad_norm": 7.
|
| 167 |
-
"learning_rate":
|
| 168 |
-
"loss": 0.
|
| 169 |
"step": 200
|
| 170 |
},
|
| 171 |
{
|
| 172 |
-
"epoch":
|
| 173 |
-
"grad_norm":
|
| 174 |
-
"learning_rate":
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 210
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch":
|
| 180 |
-
"grad_norm":
|
| 181 |
-
"learning_rate":
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 220
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch":
|
| 187 |
-
"
|
| 188 |
-
"
|
| 189 |
-
"
|
| 190 |
-
"eval_runtime": 75.07,
|
| 191 |
-
"eval_samples_per_second": 3.21,
|
| 192 |
-
"eval_steps_per_second": 0.333,
|
| 193 |
-
"step": 225
|
| 194 |
-
},
|
| 195 |
-
{
|
| 196 |
-
"epoch": 3.0135135135135136,
|
| 197 |
-
"grad_norm": 8.580607414245605,
|
| 198 |
-
"learning_rate": 2.102102102102102e-05,
|
| 199 |
-
"loss": 0.8979,
|
| 200 |
"step": 230
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch": 3.
|
| 204 |
-
"grad_norm":
|
| 205 |
-
"learning_rate":
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 240
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch": 3.
|
| 211 |
-
"
|
| 212 |
-
"
|
| 213 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 214 |
"step": 250
|
| 215 |
},
|
| 216 |
{
|
| 217 |
-
"epoch":
|
| 218 |
-
"grad_norm":
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 260
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch":
|
| 225 |
-
"grad_norm":
|
| 226 |
-
"learning_rate":
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 270
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch":
|
| 232 |
-
"grad_norm":
|
| 233 |
-
"learning_rate":
|
| 234 |
-
"loss": 0.
|
| 235 |
"step": 280
|
| 236 |
},
|
| 237 |
{
|
| 238 |
-
"epoch":
|
| 239 |
-
"grad_norm":
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 290
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"grad_norm":
|
| 247 |
-
"learning_rate":
|
| 248 |
-
"loss": 0.
|
| 249 |
"step": 300
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch":
|
| 253 |
-
"
|
| 254 |
-
"
|
| 255 |
-
"
|
| 256 |
-
"
|
| 257 |
-
"eval_samples_per_second": 3.261,
|
| 258 |
-
"eval_steps_per_second": 0.338,
|
| 259 |
-
"step": 300
|
| 260 |
},
|
| 261 |
{
|
| 262 |
-
"epoch": 4.
|
| 263 |
-
"
|
| 264 |
-
"
|
| 265 |
-
"
|
|
|
|
|
|
|
|
|
|
| 266 |
"step": 310
|
| 267 |
},
|
| 268 |
{
|
| 269 |
-
"epoch":
|
| 270 |
-
"grad_norm": 0.
|
| 271 |
-
"learning_rate":
|
| 272 |
-
"loss": 0.
|
| 273 |
"step": 320
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch":
|
| 277 |
-
"grad_norm": 0.
|
| 278 |
-
"learning_rate":
|
| 279 |
-
"loss": 0.
|
| 280 |
"step": 330
|
| 281 |
},
|
| 282 |
{
|
| 283 |
-
"epoch":
|
| 284 |
-
"grad_norm": 0.
|
| 285 |
-
"learning_rate":
|
| 286 |
-
"loss": 0.
|
| 287 |
"step": 340
|
| 288 |
},
|
| 289 |
{
|
| 290 |
-
"epoch":
|
| 291 |
-
"grad_norm":
|
| 292 |
-
"learning_rate":
|
| 293 |
-
"loss": 0.
|
| 294 |
"step": 350
|
| 295 |
},
|
| 296 |
{
|
| 297 |
-
"epoch":
|
| 298 |
-
"grad_norm":
|
| 299 |
-
"learning_rate":
|
| 300 |
-
"loss": 0.
|
| 301 |
"step": 360
|
| 302 |
},
|
| 303 |
{
|
| 304 |
-
"epoch":
|
| 305 |
-
"grad_norm": 0.
|
| 306 |
-
"learning_rate":
|
| 307 |
-
"loss": 0.
|
| 308 |
"step": 370
|
| 309 |
},
|
| 310 |
{
|
| 311 |
-
"epoch":
|
| 312 |
-
"eval_accuracy": 0.
|
| 313 |
-
"eval_f1": 0.
|
| 314 |
-
"eval_loss": 0.
|
| 315 |
-
"eval_runtime":
|
| 316 |
-
"eval_samples_per_second": 3.
|
| 317 |
-
"eval_steps_per_second": 0.
|
| 318 |
-
"step":
|
| 319 |
},
|
| 320 |
{
|
| 321 |
-
"epoch":
|
| 322 |
-
"
|
| 323 |
-
"
|
| 324 |
-
"
|
| 325 |
-
"
|
| 326 |
-
"train_samples_per_second": 1.959,
|
| 327 |
-
"train_steps_per_second": 0.196
|
| 328 |
},
|
| 329 |
{
|
| 330 |
-
"epoch":
|
| 331 |
-
"
|
| 332 |
-
"
|
| 333 |
-
"
|
| 334 |
-
"
|
| 335 |
-
"eval_samples_per_second": 2.929,
|
| 336 |
-
"eval_steps_per_second": 0.296,
|
| 337 |
-
"step": 370
|
| 338 |
},
|
| 339 |
{
|
| 340 |
-
"epoch":
|
| 341 |
-
"
|
| 342 |
-
"
|
| 343 |
-
"
|
| 344 |
-
"
|
| 345 |
-
"eval_samples_per_second": 3.24,
|
| 346 |
-
"eval_steps_per_second": 0.328,
|
| 347 |
-
"step": 370
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch":
|
| 351 |
-
"
|
| 352 |
-
"
|
| 353 |
-
"
|
| 354 |
-
"
|
| 355 |
-
|
| 356 |
-
|
| 357 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 358 |
}
|
| 359 |
],
|
| 360 |
"logging_steps": 10,
|
| 361 |
-
"max_steps":
|
| 362 |
"num_input_tokens_seen": 0,
|
| 363 |
"num_train_epochs": 9223372036854775807,
|
| 364 |
"save_steps": 500,
|
|
@@ -374,8 +592,8 @@
|
|
| 374 |
"attributes": {}
|
| 375 |
}
|
| 376 |
},
|
| 377 |
-
"total_flos":
|
| 378 |
-
"train_batch_size":
|
| 379 |
"trial_name": null,
|
| 380 |
"trial_params": null
|
| 381 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8838017754864972,
|
| 3 |
+
"best_model_checkpoint": "videomae-surf-analytics-runpod/checkpoint-372",
|
| 4 |
+
"epoch": 9.085245901639345,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 610,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.01639344262295082,
|
| 13 |
+
"grad_norm": 14.228123664855957,
|
| 14 |
+
"learning_rate": 8.196721311475409e-06,
|
| 15 |
+
"loss": 1.5374,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.03278688524590164,
|
| 20 |
+
"grad_norm": 8.263284683227539,
|
| 21 |
+
"learning_rate": 1.6393442622950818e-05,
|
| 22 |
+
"loss": 1.33,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.04918032786885246,
|
| 27 |
+
"grad_norm": 10.129837989807129,
|
| 28 |
+
"learning_rate": 2.459016393442623e-05,
|
| 29 |
+
"loss": 1.1214,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.06557377049180328,
|
| 34 |
+
"grad_norm": 5.447906017303467,
|
| 35 |
+
"learning_rate": 3.2786885245901635e-05,
|
| 36 |
+
"loss": 1.023,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.08196721311475409,
|
| 41 |
+
"grad_norm": 5.764439105987549,
|
| 42 |
+
"learning_rate": 4.098360655737705e-05,
|
| 43 |
+
"loss": 0.7497,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.09836065573770492,
|
| 48 |
+
"grad_norm": 6.531442165374756,
|
| 49 |
+
"learning_rate": 4.918032786885246e-05,
|
| 50 |
+
"loss": 0.6712,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.10163934426229508,
|
| 55 |
+
"eval_accuracy": 0.6680497925311203,
|
| 56 |
+
"eval_f1": 0.6623298693885865,
|
| 57 |
+
"eval_loss": 0.8671284317970276,
|
| 58 |
+
"eval_runtime": 81.9529,
|
| 59 |
+
"eval_samples_per_second": 2.941,
|
| 60 |
+
"eval_steps_per_second": 0.256,
|
| 61 |
+
"step": 62
|
| 62 |
},
|
| 63 |
{
|
| 64 |
+
"epoch": 1.0131147540983607,
|
| 65 |
+
"grad_norm": 6.3692240715026855,
|
| 66 |
+
"learning_rate": 4.918032786885246e-05,
|
| 67 |
+
"loss": 0.5055,
|
| 68 |
+
"step": 70
|
|
|
|
|
|
|
|
|
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 1.0295081967213116,
|
| 72 |
+
"grad_norm": 7.524365425109863,
|
| 73 |
+
"learning_rate": 4.8269581056466304e-05,
|
| 74 |
+
"loss": 0.4465,
|
| 75 |
"step": 80
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"epoch": 1.0459016393442624,
|
| 79 |
+
"grad_norm": 3.4210591316223145,
|
| 80 |
+
"learning_rate": 4.7358834244080144e-05,
|
| 81 |
+
"loss": 0.3804,
|
| 82 |
"step": 90
|
| 83 |
},
|
| 84 |
{
|
| 85 |
+
"epoch": 1.0622950819672132,
|
| 86 |
+
"grad_norm": 7.561025619506836,
|
| 87 |
+
"learning_rate": 4.644808743169399e-05,
|
| 88 |
+
"loss": 0.4557,
|
| 89 |
"step": 100
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 1.0786885245901638,
|
| 93 |
+
"grad_norm": 2.567615509033203,
|
| 94 |
+
"learning_rate": 4.553734061930783e-05,
|
| 95 |
+
"loss": 0.3558,
|
| 96 |
"step": 110
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 1.0950819672131147,
|
| 100 |
+
"grad_norm": 5.366397380828857,
|
| 101 |
+
"learning_rate": 4.462659380692168e-05,
|
| 102 |
+
"loss": 0.3119,
|
| 103 |
"step": 120
|
| 104 |
},
|
| 105 |
{
|
| 106 |
+
"epoch": 1.1016393442622952,
|
| 107 |
+
"eval_accuracy": 0.7883817427385892,
|
| 108 |
+
"eval_f1": 0.7887466325291007,
|
| 109 |
+
"eval_loss": 0.5910844802856445,
|
| 110 |
+
"eval_runtime": 82.6936,
|
| 111 |
+
"eval_samples_per_second": 2.914,
|
| 112 |
+
"eval_steps_per_second": 0.254,
|
| 113 |
+
"step": 124
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 2.0098360655737704,
|
| 117 |
+
"grad_norm": 2.0202689170837402,
|
| 118 |
+
"learning_rate": 4.371584699453552e-05,
|
| 119 |
+
"loss": 0.2248,
|
| 120 |
+
"step": 130
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 2.0262295081967214,
|
| 124 |
+
"grad_norm": 5.155245780944824,
|
| 125 |
+
"learning_rate": 4.280510018214937e-05,
|
| 126 |
+
"loss": 0.2797,
|
| 127 |
+
"step": 140
|
| 128 |
},
|
| 129 |
{
|
| 130 |
+
"epoch": 2.042622950819672,
|
| 131 |
+
"grad_norm": 8.79983139038086,
|
| 132 |
+
"learning_rate": 4.189435336976321e-05,
|
| 133 |
+
"loss": 0.2696,
|
|
|
|
|
|
|
|
|
|
| 134 |
"step": 150
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 2.059016393442623,
|
| 138 |
+
"grad_norm": 2.6221516132354736,
|
| 139 |
+
"learning_rate": 4.098360655737705e-05,
|
| 140 |
+
"loss": 0.2733,
|
| 141 |
"step": 160
|
| 142 |
},
|
| 143 |
{
|
| 144 |
+
"epoch": 2.0754098360655737,
|
| 145 |
+
"grad_norm": 1.32801353931427,
|
| 146 |
+
"learning_rate": 4.007285974499089e-05,
|
| 147 |
+
"loss": 0.159,
|
| 148 |
"step": 170
|
| 149 |
},
|
| 150 |
{
|
| 151 |
+
"epoch": 2.091803278688525,
|
| 152 |
+
"grad_norm": 7.7792792320251465,
|
| 153 |
+
"learning_rate": 3.916211293260474e-05,
|
| 154 |
+
"loss": 0.2505,
|
| 155 |
"step": 180
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 2.101639344262295,
|
| 159 |
+
"eval_accuracy": 0.8008298755186722,
|
| 160 |
+
"eval_f1": 0.8002014090167812,
|
| 161 |
+
"eval_loss": 0.5296825170516968,
|
| 162 |
+
"eval_runtime": 80.6025,
|
| 163 |
+
"eval_samples_per_second": 2.99,
|
| 164 |
+
"eval_steps_per_second": 0.261,
|
| 165 |
+
"step": 186
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 3.0065573770491802,
|
| 169 |
+
"grad_norm": 6.463658332824707,
|
| 170 |
+
"learning_rate": 3.825136612021858e-05,
|
| 171 |
+
"loss": 0.3667,
|
| 172 |
"step": 190
|
| 173 |
},
|
| 174 |
{
|
| 175 |
+
"epoch": 3.0229508196721313,
|
| 176 |
+
"grad_norm": 7.710232257843018,
|
| 177 |
+
"learning_rate": 3.7340619307832425e-05,
|
| 178 |
+
"loss": 0.1461,
|
| 179 |
"step": 200
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 3.039344262295082,
|
| 183 |
+
"grad_norm": 8.827251434326172,
|
| 184 |
+
"learning_rate": 3.6429872495446266e-05,
|
| 185 |
+
"loss": 0.1328,
|
| 186 |
"step": 210
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 3.055737704918033,
|
| 190 |
+
"grad_norm": 14.754895210266113,
|
| 191 |
+
"learning_rate": 3.551912568306011e-05,
|
| 192 |
+
"loss": 0.4213,
|
| 193 |
"step": 220
|
| 194 |
},
|
| 195 |
{
|
| 196 |
+
"epoch": 3.0721311475409836,
|
| 197 |
+
"grad_norm": 7.798481464385986,
|
| 198 |
+
"learning_rate": 3.4608378870673954e-05,
|
| 199 |
+
"loss": 0.2256,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 200 |
"step": 230
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 3.088524590163934,
|
| 204 |
+
"grad_norm": 1.4291377067565918,
|
| 205 |
+
"learning_rate": 3.36976320582878e-05,
|
| 206 |
+
"loss": 0.207,
|
| 207 |
"step": 240
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 3.101639344262295,
|
| 211 |
+
"eval_accuracy": 0.7800829875518672,
|
| 212 |
+
"eval_f1": 0.778687519310884,
|
| 213 |
+
"eval_loss": 0.5970368981361389,
|
| 214 |
+
"eval_runtime": 76.1019,
|
| 215 |
+
"eval_samples_per_second": 3.167,
|
| 216 |
+
"eval_steps_per_second": 0.276,
|
| 217 |
+
"step": 248
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 4.00327868852459,
|
| 221 |
+
"grad_norm": 6.123374938964844,
|
| 222 |
+
"learning_rate": 3.2786885245901635e-05,
|
| 223 |
+
"loss": 0.1879,
|
| 224 |
"step": 250
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 4.019672131147541,
|
| 228 |
+
"grad_norm": 2.9475815296173096,
|
| 229 |
+
"learning_rate": 3.187613843351548e-05,
|
| 230 |
+
"loss": 0.0801,
|
| 231 |
"step": 260
|
| 232 |
},
|
| 233 |
{
|
| 234 |
+
"epoch": 4.036065573770492,
|
| 235 |
+
"grad_norm": 0.0647067278623581,
|
| 236 |
+
"learning_rate": 3.096539162112932e-05,
|
| 237 |
+
"loss": 0.0925,
|
| 238 |
"step": 270
|
| 239 |
},
|
| 240 |
{
|
| 241 |
+
"epoch": 4.052459016393443,
|
| 242 |
+
"grad_norm": 2.0591495037078857,
|
| 243 |
+
"learning_rate": 3.005464480874317e-05,
|
| 244 |
+
"loss": 0.1952,
|
| 245 |
"step": 280
|
| 246 |
},
|
| 247 |
{
|
| 248 |
+
"epoch": 4.0688524590163935,
|
| 249 |
+
"grad_norm": 9.379964828491211,
|
| 250 |
+
"learning_rate": 2.9143897996357018e-05,
|
| 251 |
+
"loss": 0.2213,
|
| 252 |
"step": 290
|
| 253 |
},
|
| 254 |
{
|
| 255 |
+
"epoch": 4.085245901639344,
|
| 256 |
+
"grad_norm": 0.2886744737625122,
|
| 257 |
+
"learning_rate": 2.823315118397086e-05,
|
| 258 |
+
"loss": 0.1651,
|
| 259 |
"step": 300
|
| 260 |
},
|
| 261 |
{
|
| 262 |
+
"epoch": 4.101639344262295,
|
| 263 |
+
"grad_norm": 10.280256271362305,
|
| 264 |
+
"learning_rate": 2.7322404371584703e-05,
|
| 265 |
+
"loss": 0.1743,
|
| 266 |
+
"step": 310
|
|
|
|
|
|
|
|
|
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 4.101639344262295,
|
| 270 |
+
"eval_accuracy": 0.8049792531120332,
|
| 271 |
+
"eval_f1": 0.7984134204702362,
|
| 272 |
+
"eval_loss": 0.5612084865570068,
|
| 273 |
+
"eval_runtime": 76.7892,
|
| 274 |
+
"eval_samples_per_second": 3.138,
|
| 275 |
+
"eval_steps_per_second": 0.273,
|
| 276 |
"step": 310
|
| 277 |
},
|
| 278 |
{
|
| 279 |
+
"epoch": 5.016393442622951,
|
| 280 |
+
"grad_norm": 0.10767544060945511,
|
| 281 |
+
"learning_rate": 2.6411657559198543e-05,
|
| 282 |
+
"loss": 0.1745,
|
| 283 |
"step": 320
|
| 284 |
},
|
| 285 |
{
|
| 286 |
+
"epoch": 5.032786885245901,
|
| 287 |
+
"grad_norm": 0.31735455989837646,
|
| 288 |
+
"learning_rate": 2.550091074681239e-05,
|
| 289 |
+
"loss": 0.0893,
|
| 290 |
"step": 330
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 5.049180327868853,
|
| 294 |
+
"grad_norm": 0.657698392868042,
|
| 295 |
+
"learning_rate": 2.459016393442623e-05,
|
| 296 |
+
"loss": 0.0209,
|
| 297 |
"step": 340
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 5.065573770491803,
|
| 301 |
+
"grad_norm": 6.128834247589111,
|
| 302 |
+
"learning_rate": 2.3679417122040072e-05,
|
| 303 |
+
"loss": 0.0931,
|
| 304 |
"step": 350
|
| 305 |
},
|
| 306 |
{
|
| 307 |
+
"epoch": 5.081967213114754,
|
| 308 |
+
"grad_norm": 3.8232431411743164,
|
| 309 |
+
"learning_rate": 2.2768670309653916e-05,
|
| 310 |
+
"loss": 0.1862,
|
| 311 |
"step": 360
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 5.098360655737705,
|
| 315 |
+
"grad_norm": 0.28946903347969055,
|
| 316 |
+
"learning_rate": 2.185792349726776e-05,
|
| 317 |
+
"loss": 0.1005,
|
| 318 |
"step": 370
|
| 319 |
},
|
| 320 |
{
|
| 321 |
+
"epoch": 5.101639344262295,
|
| 322 |
+
"eval_accuracy": 0.8838174273858921,
|
| 323 |
+
"eval_f1": 0.8838017754864972,
|
| 324 |
+
"eval_loss": 0.40273839235305786,
|
| 325 |
+
"eval_runtime": 79.1005,
|
| 326 |
+
"eval_samples_per_second": 3.047,
|
| 327 |
+
"eval_steps_per_second": 0.265,
|
| 328 |
+
"step": 372
|
| 329 |
},
|
| 330 |
{
|
| 331 |
+
"epoch": 6.0131147540983605,
|
| 332 |
+
"grad_norm": 7.813482761383057,
|
| 333 |
+
"learning_rate": 2.0947176684881604e-05,
|
| 334 |
+
"loss": 0.1164,
|
| 335 |
+
"step": 380
|
|
|
|
|
|
|
| 336 |
},
|
| 337 |
{
|
| 338 |
+
"epoch": 6.029508196721311,
|
| 339 |
+
"grad_norm": 4.695272445678711,
|
| 340 |
+
"learning_rate": 2.0036429872495445e-05,
|
| 341 |
+
"loss": 0.128,
|
| 342 |
+
"step": 390
|
|
|
|
|
|
|
|
|
|
| 343 |
},
|
| 344 |
{
|
| 345 |
+
"epoch": 6.045901639344263,
|
| 346 |
+
"grad_norm": 0.22056636214256287,
|
| 347 |
+
"learning_rate": 1.912568306010929e-05,
|
| 348 |
+
"loss": 0.1201,
|
| 349 |
+
"step": 400
|
|
|
|
|
|
|
|
|
|
| 350 |
},
|
| 351 |
{
|
| 352 |
+
"epoch": 6.062295081967213,
|
| 353 |
+
"grad_norm": 0.24654638767242432,
|
| 354 |
+
"learning_rate": 1.8214936247723133e-05,
|
| 355 |
+
"loss": 0.0823,
|
| 356 |
+
"step": 410
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 6.078688524590164,
|
| 360 |
+
"grad_norm": 1.5276825428009033,
|
| 361 |
+
"learning_rate": 1.7304189435336977e-05,
|
| 362 |
+
"loss": 0.0994,
|
| 363 |
+
"step": 420
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 6.0950819672131145,
|
| 367 |
+
"grad_norm": 5.22976016998291,
|
| 368 |
+
"learning_rate": 1.6393442622950818e-05,
|
| 369 |
+
"loss": 0.0147,
|
| 370 |
+
"step": 430
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 6.101639344262295,
|
| 374 |
+
"eval_accuracy": 0.8589211618257261,
|
| 375 |
+
"eval_f1": 0.8572866763193951,
|
| 376 |
+
"eval_loss": 0.4360053241252899,
|
| 377 |
+
"eval_runtime": 76.601,
|
| 378 |
+
"eval_samples_per_second": 3.146,
|
| 379 |
+
"eval_steps_per_second": 0.274,
|
| 380 |
+
"step": 434
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 7.00983606557377,
|
| 384 |
+
"grad_norm": 0.01562822423875332,
|
| 385 |
+
"learning_rate": 1.548269581056466e-05,
|
| 386 |
+
"loss": 0.0178,
|
| 387 |
+
"step": 440
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 7.026229508196721,
|
| 391 |
+
"grad_norm": 0.15828734636306763,
|
| 392 |
+
"learning_rate": 1.4571948998178509e-05,
|
| 393 |
+
"loss": 0.1265,
|
| 394 |
+
"step": 450
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 7.0426229508196725,
|
| 398 |
+
"grad_norm": 0.9054508805274963,
|
| 399 |
+
"learning_rate": 1.3661202185792351e-05,
|
| 400 |
+
"loss": 0.0251,
|
| 401 |
+
"step": 460
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 7.059016393442623,
|
| 405 |
+
"grad_norm": 0.2960349917411804,
|
| 406 |
+
"learning_rate": 1.2750455373406195e-05,
|
| 407 |
+
"loss": 0.0241,
|
| 408 |
+
"step": 470
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 7.075409836065574,
|
| 412 |
+
"grad_norm": 0.08049295842647552,
|
| 413 |
+
"learning_rate": 1.1839708561020036e-05,
|
| 414 |
+
"loss": 0.0518,
|
| 415 |
+
"step": 480
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 7.091803278688524,
|
| 419 |
+
"grad_norm": 0.25148022174835205,
|
| 420 |
+
"learning_rate": 1.092896174863388e-05,
|
| 421 |
+
"loss": 0.0573,
|
| 422 |
+
"step": 490
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 7.101639344262295,
|
| 426 |
+
"eval_accuracy": 0.8713692946058091,
|
| 427 |
+
"eval_f1": 0.8697143392786835,
|
| 428 |
+
"eval_loss": 0.4450831115245819,
|
| 429 |
+
"eval_runtime": 75.7767,
|
| 430 |
+
"eval_samples_per_second": 3.18,
|
| 431 |
+
"eval_steps_per_second": 0.277,
|
| 432 |
+
"step": 496
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 8.00655737704918,
|
| 436 |
+
"grad_norm": 0.028504155576229095,
|
| 437 |
+
"learning_rate": 1.0018214936247722e-05,
|
| 438 |
+
"loss": 0.0705,
|
| 439 |
+
"step": 500
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 8.02295081967213,
|
| 443 |
+
"grad_norm": 0.7975661158561707,
|
| 444 |
+
"learning_rate": 9.107468123861566e-06,
|
| 445 |
+
"loss": 0.0547,
|
| 446 |
+
"step": 510
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 8.039344262295081,
|
| 450 |
+
"grad_norm": 0.035094812512397766,
|
| 451 |
+
"learning_rate": 8.196721311475409e-06,
|
| 452 |
+
"loss": 0.0484,
|
| 453 |
+
"step": 520
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 8.055737704918032,
|
| 457 |
+
"grad_norm": 4.1898651123046875,
|
| 458 |
+
"learning_rate": 7.2859744990892545e-06,
|
| 459 |
+
"loss": 0.0834,
|
| 460 |
+
"step": 530
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 8.072131147540984,
|
| 464 |
+
"grad_norm": 0.8504851460456848,
|
| 465 |
+
"learning_rate": 6.375227686703098e-06,
|
| 466 |
+
"loss": 0.1502,
|
| 467 |
+
"step": 540
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 8.088524590163935,
|
| 471 |
+
"grad_norm": 0.8113920092582703,
|
| 472 |
+
"learning_rate": 5.46448087431694e-06,
|
| 473 |
+
"loss": 0.0143,
|
| 474 |
+
"step": 550
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 8.101639344262296,
|
| 478 |
+
"eval_accuracy": 0.8672199170124482,
|
| 479 |
+
"eval_f1": 0.8666497215968316,
|
| 480 |
+
"eval_loss": 0.40985623002052307,
|
| 481 |
+
"eval_runtime": 79.8008,
|
| 482 |
+
"eval_samples_per_second": 3.02,
|
| 483 |
+
"eval_steps_per_second": 0.263,
|
| 484 |
+
"step": 558
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 9.00327868852459,
|
| 488 |
+
"grad_norm": 6.550829887390137,
|
| 489 |
+
"learning_rate": 4.553734061930783e-06,
|
| 490 |
+
"loss": 0.067,
|
| 491 |
+
"step": 560
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 9.01967213114754,
|
| 495 |
+
"grad_norm": 2.0722174644470215,
|
| 496 |
+
"learning_rate": 3.6429872495446273e-06,
|
| 497 |
+
"loss": 0.0158,
|
| 498 |
+
"step": 570
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 9.036065573770491,
|
| 502 |
+
"grad_norm": 0.09152109175920486,
|
| 503 |
+
"learning_rate": 2.73224043715847e-06,
|
| 504 |
+
"loss": 0.0203,
|
| 505 |
+
"step": 580
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 9.052459016393442,
|
| 509 |
+
"grad_norm": 0.12251006811857224,
|
| 510 |
+
"learning_rate": 1.8214936247723136e-06,
|
| 511 |
+
"loss": 0.0169,
|
| 512 |
+
"step": 590
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 9.068852459016393,
|
| 516 |
+
"grad_norm": 0.1367557793855667,
|
| 517 |
+
"learning_rate": 9.107468123861568e-07,
|
| 518 |
+
"loss": 0.0043,
|
| 519 |
+
"step": 600
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 9.085245901639345,
|
| 523 |
+
"grad_norm": 6.975312232971191,
|
| 524 |
+
"learning_rate": 0.0,
|
| 525 |
+
"loss": 0.1311,
|
| 526 |
+
"step": 610
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 9.085245901639345,
|
| 530 |
+
"eval_accuracy": 0.8755186721991701,
|
| 531 |
+
"eval_f1": 0.8751571381793353,
|
| 532 |
+
"eval_loss": 0.40564292669296265,
|
| 533 |
+
"eval_runtime": 76.4754,
|
| 534 |
+
"eval_samples_per_second": 3.151,
|
| 535 |
+
"eval_steps_per_second": 0.275,
|
| 536 |
+
"step": 610
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 9.085245901639345,
|
| 540 |
+
"step": 610,
|
| 541 |
+
"total_flos": 6.389823367042892e+18,
|
| 542 |
+
"train_loss": 0.24862837887933997,
|
| 543 |
+
"train_runtime": 3777.5727,
|
| 544 |
+
"train_samples_per_second": 1.938,
|
| 545 |
+
"train_steps_per_second": 0.161
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 9.085245901639345,
|
| 549 |
+
"eval_accuracy": 0.9824561403508771,
|
| 550 |
+
"eval_f1": 0.9824594539682324,
|
| 551 |
+
"eval_loss": 0.04523608461022377,
|
| 552 |
+
"eval_runtime": 250.7162,
|
| 553 |
+
"eval_samples_per_second": 2.956,
|
| 554 |
+
"eval_steps_per_second": 0.247,
|
| 555 |
+
"step": 610
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 9.085245901639345,
|
| 559 |
+
"eval_accuracy": 0.8337078651685393,
|
| 560 |
+
"eval_f1": 0.8322580739791249,
|
| 561 |
+
"eval_loss": 0.6123429536819458,
|
| 562 |
+
"eval_runtime": 140.7823,
|
| 563 |
+
"eval_samples_per_second": 3.161,
|
| 564 |
+
"eval_steps_per_second": 0.27,
|
| 565 |
+
"step": 610
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 9.085245901639345,
|
| 569 |
+
"eval_accuracy": 0.8838174273858921,
|
| 570 |
+
"eval_f1": 0.8838017754864972,
|
| 571 |
+
"eval_loss": 0.40273845195770264,
|
| 572 |
+
"eval_runtime": 76.8071,
|
| 573 |
+
"eval_samples_per_second": 3.138,
|
| 574 |
+
"eval_steps_per_second": 0.273,
|
| 575 |
+
"step": 610
|
| 576 |
}
|
| 577 |
],
|
| 578 |
"logging_steps": 10,
|
| 579 |
+
"max_steps": 610,
|
| 580 |
"num_input_tokens_seen": 0,
|
| 581 |
"num_train_epochs": 9223372036854775807,
|
| 582 |
"save_steps": 500,
|
|
|
|
| 592 |
"attributes": {}
|
| 593 |
}
|
| 594 |
},
|
| 595 |
+
"total_flos": 6.389823367042892e+18,
|
| 596 |
+
"train_batch_size": 12,
|
| 597 |
"trial_name": null,
|
| 598 |
"trial_params": null
|
| 599 |
}
|
val_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_f1": 0.
|
| 5 |
-
"eval_loss": 0.
|
| 6 |
-
"eval_runtime":
|
| 7 |
-
"eval_samples_per_second": 3.
|
| 8 |
-
"eval_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.085245901639345,
|
| 3 |
+
"eval_accuracy": 0.8838174273858921,
|
| 4 |
+
"eval_f1": 0.8838017754864972,
|
| 5 |
+
"eval_loss": 0.40273845195770264,
|
| 6 |
+
"eval_runtime": 76.8071,
|
| 7 |
+
"eval_samples_per_second": 3.138,
|
| 8 |
+
"eval_steps_per_second": 0.273
|
| 9 |
}
|