Commit
·
e0cae81
1
Parent(s):
84b0181
yashcode00/wav2vec2-large-xlsr-indian-language-classification-featureExtractor
Browse files- README.md +14 -6
- all_results.json +11 -11
- eval_results.json +6 -6
- pytorch_model.bin +1 -1
- train_results.json +6 -6
- trainer_state.json +337 -721
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [yashcode00/wav2vec2-large-xlsr-indian-language-classification-featureExtractor](https://huggingface.co/yashcode00/wav2vec2-large-xlsr-indian-language-classification-featureExtractor) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -45,15 +45,23 @@ The following hyperparameters were used during training:
|
|
| 45 |
- total_train_batch_size: 64
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
-
- num_epochs:
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 53 |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
|
| 58 |
|
| 59 |
### Framework versions
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [yashcode00/wav2vec2-large-xlsr-indian-language-classification-featureExtractor](https://huggingface.co/yashcode00/wav2vec2-large-xlsr-indian-language-classification-featureExtractor) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.3117
|
| 21 |
+
- Accuracy: 0.9323
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 45 |
- total_train_batch_size: 64
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
+
- num_epochs: 60
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 53 |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
| 54 |
+
| 0.0559 | 5.28 | 1000 | 0.3097 | 0.9191 |
|
| 55 |
+
| 0.047 | 10.56 | 2000 | 0.3482 | 0.9191 |
|
| 56 |
+
| 0.0402 | 15.84 | 3000 | 0.3890 | 0.9080 |
|
| 57 |
+
| 0.0328 | 21.12 | 4000 | 0.3746 | 0.9150 |
|
| 58 |
+
| 0.0189 | 26.4 | 5000 | 0.4274 | 0.9113 |
|
| 59 |
+
| 0.0187 | 31.68 | 6000 | 0.4131 | 0.9101 |
|
| 60 |
+
| 0.0203 | 36.96 | 7000 | 0.3643 | 0.9237 |
|
| 61 |
+
| 0.0147 | 42.24 | 8000 | 0.3574 | 0.9295 |
|
| 62 |
+
| 0.0148 | 47.52 | 9000 | 0.3653 | 0.9220 |
|
| 63 |
+
| 0.0137 | 52.81 | 10000 | 0.3257 | 0.9352 |
|
| 64 |
+
| 0.0174 | 58.09 | 11000 | 0.3097 | 0.9340 |
|
| 65 |
|
| 66 |
|
| 67 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 50.
|
| 6 |
"eval_samples": 2424,
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second":
|
| 9 |
-
"total_flos":
|
| 10 |
-
"train_loss": 0.
|
| 11 |
-
"train_runtime":
|
| 12 |
"train_samples": 12120,
|
| 13 |
-
"train_samples_per_second":
|
| 14 |
-
"train_steps_per_second": 0.
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 59.88,
|
| 3 |
+
"eval_accuracy": 0.9323432445526123,
|
| 4 |
+
"eval_loss": 0.3117374777793884,
|
| 5 |
+
"eval_runtime": 50.8983,
|
| 6 |
"eval_samples": 2424,
|
| 7 |
+
"eval_samples_per_second": 47.624,
|
| 8 |
+
"eval_steps_per_second": 5.953,
|
| 9 |
+
"total_flos": 2.20514636224512e+19,
|
| 10 |
+
"train_loss": 0.030625741817122836,
|
| 11 |
+
"train_runtime": 25244.9182,
|
| 12 |
"train_samples": 12120,
|
| 13 |
+
"train_samples_per_second": 28.806,
|
| 14 |
+
"train_steps_per_second": 0.449
|
| 15 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 50.
|
| 6 |
"eval_samples": 2424,
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 59.88,
|
| 3 |
+
"eval_accuracy": 0.9323432445526123,
|
| 4 |
+
"eval_loss": 0.3117374777793884,
|
| 5 |
+
"eval_runtime": 50.8983,
|
| 6 |
"eval_samples": 2424,
|
| 7 |
+
"eval_samples_per_second": 47.624,
|
| 8 |
+
"eval_steps_per_second": 5.953
|
| 9 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1266146037
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41ca0eeca582f52ad187db0b010f7bf152237a85fc348fd2af50256d74874d7a
|
| 3 |
size 1266146037
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 12120,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 59.88,
|
| 3 |
+
"total_flos": 2.20514636224512e+19,
|
| 4 |
+
"train_loss": 0.030625741817122836,
|
| 5 |
+
"train_runtime": 25244.9182,
|
| 6 |
"train_samples": 12120,
|
| 7 |
+
"train_samples_per_second": 28.806,
|
| 8 |
+
"train_steps_per_second": 0.449
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1198 +1,814 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps":
|
| 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": 0.53,
|
| 13 |
-
"learning_rate": 4.
|
| 14 |
-
"loss": 0.
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 1.06,
|
| 19 |
-
"learning_rate": 4.
|
| 20 |
-
"loss": 0.
|
| 21 |
"step": 200
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 1.58,
|
| 25 |
-
"learning_rate": 4.
|
| 26 |
-
"loss": 0.
|
| 27 |
"step": 300
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 2.11,
|
| 31 |
-
"learning_rate": 4.
|
| 32 |
-
"loss": 0.
|
| 33 |
"step": 400
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 2.64,
|
| 37 |
-
"learning_rate": 4.
|
| 38 |
-
"loss": 0.
|
| 39 |
"step": 500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 3.17,
|
| 43 |
-
"learning_rate": 4.
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 600
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 3.7,
|
| 49 |
-
"learning_rate": 4.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 700
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 4.22,
|
| 55 |
-
"learning_rate": 4.
|
| 56 |
-
"loss": 0.
|
| 57 |
"step": 800
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 4.75,
|
| 61 |
-
"learning_rate": 4.
|
| 62 |
-
"loss": 0.
|
| 63 |
"step": 900
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 5.28,
|
| 67 |
-
"learning_rate": 4.
|
| 68 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 69 |
"step": 1000
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 5.81,
|
| 73 |
-
"learning_rate": 4.
|
| 74 |
-
"loss": 0.
|
| 75 |
"step": 1100
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 6.34,
|
| 79 |
-
"learning_rate": 4.
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 1200
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 6.86,
|
| 85 |
-
"learning_rate": 4.
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 1300
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 7.39,
|
| 91 |
-
"learning_rate": 4.
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 1400
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 7.92,
|
| 97 |
-
"learning_rate": 4.
|
| 98 |
-
"loss": 0.
|
| 99 |
"step": 1500
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 8.45,
|
| 103 |
-
"learning_rate": 4.
|
| 104 |
-
"loss": 0.
|
| 105 |
"step": 1600
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 8.98,
|
| 109 |
-
"learning_rate": 4.
|
| 110 |
-
"loss": 0.
|
| 111 |
"step": 1700
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 9.5,
|
| 115 |
-
"learning_rate": 4.
|
| 116 |
-
"loss": 0.
|
| 117 |
"step": 1800
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 10.03,
|
| 121 |
-
"learning_rate": 4.
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 1900
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 10.56,
|
| 127 |
-
"learning_rate": 4.
|
| 128 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 129 |
"step": 2000
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 11.09,
|
| 133 |
-
"learning_rate": 4.
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 2100
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 11.62,
|
| 139 |
-
"learning_rate": 4.
|
| 140 |
-
"loss": 0.
|
| 141 |
"step": 2200
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 12.15,
|
| 145 |
-
"learning_rate":
|
| 146 |
-
"loss": 0.
|
| 147 |
"step": 2300
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 12.67,
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 2400
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 13.2,
|
| 157 |
-
"learning_rate":
|
| 158 |
-
"loss": 0.
|
| 159 |
"step": 2500
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 13.73,
|
| 163 |
-
"learning_rate":
|
| 164 |
-
"loss": 0.
|
| 165 |
"step": 2600
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 14.26,
|
| 169 |
-
"learning_rate":
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 2700
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 14.79,
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 2800
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 15.31,
|
| 181 |
-
"learning_rate":
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 2900
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 15.84,
|
| 187 |
-
"learning_rate":
|
| 188 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 189 |
"step": 3000
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 16.37,
|
| 193 |
-
"learning_rate":
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 3100
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 16.9,
|
| 199 |
-
"learning_rate":
|
| 200 |
-
"loss": 0.
|
| 201 |
"step": 3200
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 17.43,
|
| 205 |
-
"learning_rate":
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 3300
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 17.95,
|
| 211 |
-
"learning_rate":
|
| 212 |
-
"loss": 0.
|
| 213 |
"step": 3400
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 18.48,
|
| 217 |
-
"learning_rate":
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 3500
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 19.01,
|
| 223 |
-
"learning_rate":
|
| 224 |
-
"loss": 0.
|
| 225 |
"step": 3600
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"epoch": 19.54,
|
| 229 |
-
"learning_rate":
|
| 230 |
-
"loss": 0.
|
| 231 |
"step": 3700
|
| 232 |
},
|
| 233 |
{
|
| 234 |
"epoch": 20.07,
|
| 235 |
-
"learning_rate": 3.
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 3800
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 20.59,
|
| 241 |
-
"learning_rate": 3.
|
| 242 |
-
"loss": 0.
|
| 243 |
"step": 3900
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 21.12,
|
| 247 |
-
"learning_rate": 3.
|
| 248 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 249 |
"step": 4000
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 21.65,
|
| 253 |
-
"learning_rate": 3.
|
| 254 |
-
"loss": 0.
|
| 255 |
"step": 4100
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 22.18,
|
| 259 |
-
"learning_rate": 3.
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 4200
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 22.71,
|
| 265 |
-
"learning_rate": 3.
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 4300
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 23.23,
|
| 271 |
-
"learning_rate": 3.
|
| 272 |
-
"loss": 0.
|
| 273 |
"step": 4400
|
| 274 |
},
|
| 275 |
{
|
| 276 |
"epoch": 23.76,
|
| 277 |
-
"learning_rate": 3.
|
| 278 |
-
"loss": 0.
|
| 279 |
"step": 4500
|
| 280 |
},
|
| 281 |
{
|
| 282 |
"epoch": 24.29,
|
| 283 |
-
"learning_rate":
|
| 284 |
-
"loss": 0.
|
| 285 |
"step": 4600
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 24.82,
|
| 289 |
-
"learning_rate":
|
| 290 |
-
"loss": 0.
|
| 291 |
"step": 4700
|
| 292 |
},
|
| 293 |
{
|
| 294 |
"epoch": 25.35,
|
| 295 |
-
"learning_rate":
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 4800
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 25.87,
|
| 301 |
-
"learning_rate":
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 4900
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 26.4,
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 5000
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 26.4,
|
| 313 |
-
"eval_accuracy": 0.
|
| 314 |
-
"eval_loss": 0.
|
| 315 |
-
"eval_runtime":
|
| 316 |
-
"eval_samples_per_second":
|
| 317 |
-
"eval_steps_per_second":
|
| 318 |
"step": 5000
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 26.93,
|
| 322 |
-
"learning_rate":
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 5100
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 27.46,
|
| 328 |
-
"learning_rate":
|
| 329 |
-
"loss": 0.
|
| 330 |
"step": 5200
|
| 331 |
},
|
| 332 |
{
|
| 333 |
"epoch": 27.99,
|
| 334 |
-
"learning_rate":
|
| 335 |
-
"loss": 0.
|
| 336 |
"step": 5300
|
| 337 |
},
|
| 338 |
{
|
| 339 |
"epoch": 28.51,
|
| 340 |
-
"learning_rate":
|
| 341 |
-
"loss": 0.
|
| 342 |
"step": 5400
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 29.04,
|
| 346 |
-
"learning_rate":
|
| 347 |
-
"loss": 0.
|
| 348 |
"step": 5500
|
| 349 |
},
|
| 350 |
{
|
| 351 |
"epoch": 29.57,
|
| 352 |
-
"learning_rate":
|
| 353 |
-
"loss": 0.
|
| 354 |
"step": 5600
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 30.1,
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 5700
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 30.63,
|
| 364 |
-
"learning_rate":
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 5800
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 31.16,
|
| 370 |
-
"learning_rate":
|
| 371 |
-
"loss": 0.
|
| 372 |
"step": 5900
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 31.68,
|
| 376 |
-
"learning_rate":
|
| 377 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 378 |
"step": 6000
|
| 379 |
},
|
| 380 |
{
|
| 381 |
"epoch": 32.21,
|
| 382 |
-
"learning_rate":
|
| 383 |
-
"loss": 0.
|
| 384 |
"step": 6100
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 32.74,
|
| 388 |
-
"learning_rate":
|
| 389 |
-
"loss": 0.
|
| 390 |
"step": 6200
|
| 391 |
},
|
| 392 |
{
|
| 393 |
"epoch": 33.27,
|
| 394 |
-
"learning_rate":
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 6300
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 33.8,
|
| 400 |
-
"learning_rate":
|
| 401 |
-
"loss": 0.
|
| 402 |
"step": 6400
|
| 403 |
},
|
| 404 |
{
|
| 405 |
"epoch": 34.32,
|
| 406 |
-
"learning_rate":
|
| 407 |
-
"loss": 0.
|
| 408 |
"step": 6500
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 34.85,
|
| 412 |
-
"learning_rate":
|
| 413 |
-
"loss": 0.
|
| 414 |
"step": 6600
|
| 415 |
},
|
| 416 |
{
|
| 417 |
"epoch": 35.38,
|
| 418 |
-
"learning_rate":
|
| 419 |
-
"loss": 0.
|
| 420 |
"step": 6700
|
| 421 |
},
|
| 422 |
{
|
| 423 |
"epoch": 35.91,
|
| 424 |
-
"learning_rate":
|
| 425 |
-
"loss": 0.
|
| 426 |
"step": 6800
|
| 427 |
},
|
| 428 |
{
|
| 429 |
"epoch": 36.44,
|
| 430 |
-
"learning_rate":
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 6900
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 36.96,
|
| 436 |
-
"learning_rate":
|
| 437 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 438 |
"step": 7000
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 37.49,
|
| 442 |
-
"learning_rate":
|
| 443 |
-
"loss": 0.
|
| 444 |
"step": 7100
|
| 445 |
},
|
| 446 |
{
|
| 447 |
"epoch": 38.02,
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss": 0.
|
| 450 |
"step": 7200
|
| 451 |
},
|
| 452 |
{
|
| 453 |
"epoch": 38.55,
|
| 454 |
-
"learning_rate":
|
| 455 |
-
"loss": 0.
|
| 456 |
"step": 7300
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 39.08,
|
| 460 |
-
"learning_rate":
|
| 461 |
-
"loss": 0.
|
| 462 |
"step": 7400
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 39.6,
|
| 466 |
-
"learning_rate":
|
| 467 |
-
"loss": 0.
|
| 468 |
"step": 7500
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 40.13,
|
| 472 |
-
"learning_rate":
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 7600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 40.66,
|
| 478 |
-
"learning_rate":
|
| 479 |
-
"loss": 0.
|
| 480 |
"step": 7700
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 41.19,
|
| 484 |
-
"learning_rate":
|
| 485 |
-
"loss": 0.
|
| 486 |
"step": 7800
|
| 487 |
},
|
| 488 |
{
|
| 489 |
"epoch": 41.72,
|
| 490 |
-
"learning_rate":
|
| 491 |
-
"loss": 0.
|
| 492 |
"step": 7900
|
| 493 |
},
|
| 494 |
{
|
| 495 |
"epoch": 42.24,
|
| 496 |
-
"learning_rate":
|
| 497 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 498 |
"step": 8000
|
| 499 |
},
|
| 500 |
{
|
| 501 |
"epoch": 42.77,
|
| 502 |
-
"learning_rate":
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 8100
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 43.3,
|
| 508 |
-
"learning_rate":
|
| 509 |
-
"loss": 0.
|
| 510 |
"step": 8200
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 43.83,
|
| 514 |
-
"learning_rate":
|
| 515 |
-
"loss": 0.
|
| 516 |
"step": 8300
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 44.36,
|
| 520 |
-
"learning_rate":
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 8400
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 44.88,
|
| 526 |
-
"learning_rate":
|
| 527 |
-
"loss": 0.
|
| 528 |
"step": 8500
|
| 529 |
},
|
| 530 |
{
|
| 531 |
"epoch": 45.41,
|
| 532 |
-
"learning_rate":
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 8600
|
| 535 |
},
|
| 536 |
{
|
| 537 |
"epoch": 45.94,
|
| 538 |
-
"learning_rate":
|
| 539 |
-
"loss": 0.
|
| 540 |
"step": 8700
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 46.47,
|
| 544 |
-
"learning_rate":
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 8800
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 47.0,
|
| 550 |
-
"learning_rate":
|
| 551 |
-
"loss": 0.
|
| 552 |
"step": 8900
|
| 553 |
},
|
| 554 |
{
|
| 555 |
"epoch": 47.52,
|
| 556 |
-
"learning_rate":
|
| 557 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 558 |
"step": 9000
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 48.05,
|
| 562 |
-
"learning_rate":
|
| 563 |
-
"loss": 0.
|
| 564 |
"step": 9100
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 48.58,
|
| 568 |
-
"learning_rate":
|
| 569 |
-
"loss": 0.
|
| 570 |
"step": 9200
|
| 571 |
},
|
| 572 |
{
|
| 573 |
"epoch": 49.11,
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 9300
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 49.64,
|
| 580 |
-
"learning_rate":
|
| 581 |
-
"loss": 0.
|
| 582 |
"step": 9400
|
| 583 |
},
|
| 584 |
{
|
| 585 |
"epoch": 50.17,
|
| 586 |
-
"learning_rate":
|
| 587 |
-
"loss": 0.
|
| 588 |
"step": 9500
|
| 589 |
},
|
| 590 |
{
|
| 591 |
"epoch": 50.69,
|
| 592 |
-
"learning_rate":
|
| 593 |
-
"loss": 0.
|
| 594 |
"step": 9600
|
| 595 |
},
|
| 596 |
{
|
| 597 |
"epoch": 51.22,
|
| 598 |
-
"learning_rate":
|
| 599 |
-
"loss": 0.
|
| 600 |
"step": 9700
|
| 601 |
},
|
| 602 |
{
|
| 603 |
"epoch": 51.75,
|
| 604 |
-
"learning_rate":
|
| 605 |
-
"loss": 0.
|
| 606 |
"step": 9800
|
| 607 |
},
|
| 608 |
{
|
| 609 |
"epoch": 52.28,
|
| 610 |
-
"learning_rate":
|
| 611 |
-
"loss": 0.
|
| 612 |
"step": 9900
|
| 613 |
},
|
| 614 |
{
|
| 615 |
"epoch": 52.81,
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 10000
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 52.81,
|
| 622 |
-
"eval_accuracy": 0.
|
| 623 |
-
"eval_loss": 0.
|
| 624 |
-
"eval_runtime": 50.
|
| 625 |
-
"eval_samples_per_second":
|
| 626 |
-
"eval_steps_per_second":
|
| 627 |
"step": 10000
|
| 628 |
},
|
| 629 |
{
|
| 630 |
"epoch": 53.33,
|
| 631 |
-
"learning_rate":
|
| 632 |
-
"loss": 0.
|
| 633 |
"step": 10100
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 53.86,
|
| 637 |
-
"learning_rate":
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 10200
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 54.39,
|
| 643 |
-
"learning_rate":
|
| 644 |
-
"loss": 0.
|
| 645 |
"step": 10300
|
| 646 |
},
|
| 647 |
{
|
| 648 |
"epoch": 54.92,
|
| 649 |
-
"learning_rate":
|
| 650 |
-
"loss": 0.
|
| 651 |
"step": 10400
|
| 652 |
},
|
| 653 |
{
|
| 654 |
"epoch": 55.45,
|
| 655 |
-
"learning_rate":
|
| 656 |
-
"loss": 0.
|
| 657 |
"step": 10500
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 55.97,
|
| 661 |
-
"learning_rate":
|
| 662 |
-
"loss": 0.
|
| 663 |
"step": 10600
|
| 664 |
},
|
| 665 |
{
|
| 666 |
"epoch": 56.5,
|
| 667 |
-
"learning_rate": 2.
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 10700
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 57.03,
|
| 673 |
-
"learning_rate": 2.
|
| 674 |
-
"loss": 0.
|
| 675 |
"step": 10800
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 57.56,
|
| 679 |
-
"learning_rate":
|
| 680 |
-
"loss": 0.
|
| 681 |
"step": 10900
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 58.09,
|
| 685 |
-
"learning_rate":
|
| 686 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 687 |
"step": 11000
|
| 688 |
},
|
| 689 |
{
|
| 690 |
"epoch": 58.61,
|
| 691 |
-
"learning_rate":
|
| 692 |
-
"loss": 0.
|
| 693 |
"step": 11100
|
| 694 |
},
|
| 695 |
{
|
| 696 |
"epoch": 59.14,
|
| 697 |
-
"learning_rate":
|
| 698 |
-
"loss": 0.
|
| 699 |
"step": 11200
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 59.67,
|
| 703 |
-
"learning_rate": 2.
|
| 704 |
-
"loss": 0.
|
| 705 |
"step": 11300
|
| 706 |
},
|
| 707 |
{
|
| 708 |
-
"epoch":
|
| 709 |
-
"
|
| 710 |
-
"
|
| 711 |
-
"
|
| 712 |
-
|
| 713 |
-
|
| 714 |
-
"
|
| 715 |
-
"learning_rate": 1.9595238095238097e-05,
|
| 716 |
-
"loss": 0.0297,
|
| 717 |
-
"step": 11500
|
| 718 |
-
},
|
| 719 |
-
{
|
| 720 |
-
"epoch": 61.25,
|
| 721 |
-
"learning_rate": 1.9330687830687833e-05,
|
| 722 |
-
"loss": 0.0375,
|
| 723 |
-
"step": 11600
|
| 724 |
-
},
|
| 725 |
-
{
|
| 726 |
-
"epoch": 61.78,
|
| 727 |
-
"learning_rate": 1.906613756613757e-05,
|
| 728 |
-
"loss": 0.0262,
|
| 729 |
-
"step": 11700
|
| 730 |
-
},
|
| 731 |
-
{
|
| 732 |
-
"epoch": 62.31,
|
| 733 |
-
"learning_rate": 1.8801587301587305e-05,
|
| 734 |
-
"loss": 0.0333,
|
| 735 |
-
"step": 11800
|
| 736 |
-
},
|
| 737 |
-
{
|
| 738 |
-
"epoch": 62.84,
|
| 739 |
-
"learning_rate": 1.8537037037037037e-05,
|
| 740 |
-
"loss": 0.025,
|
| 741 |
-
"step": 11900
|
| 742 |
-
},
|
| 743 |
-
{
|
| 744 |
-
"epoch": 63.37,
|
| 745 |
-
"learning_rate": 1.8272486772486774e-05,
|
| 746 |
-
"loss": 0.0245,
|
| 747 |
-
"step": 12000
|
| 748 |
-
},
|
| 749 |
-
{
|
| 750 |
-
"epoch": 63.89,
|
| 751 |
-
"learning_rate": 1.800793650793651e-05,
|
| 752 |
-
"loss": 0.0261,
|
| 753 |
-
"step": 12100
|
| 754 |
-
},
|
| 755 |
-
{
|
| 756 |
-
"epoch": 64.42,
|
| 757 |
-
"learning_rate": 1.7743386243386246e-05,
|
| 758 |
-
"loss": 0.0277,
|
| 759 |
-
"step": 12200
|
| 760 |
-
},
|
| 761 |
-
{
|
| 762 |
-
"epoch": 64.95,
|
| 763 |
-
"learning_rate": 1.7478835978835982e-05,
|
| 764 |
-
"loss": 0.0306,
|
| 765 |
-
"step": 12300
|
| 766 |
-
},
|
| 767 |
-
{
|
| 768 |
-
"epoch": 65.48,
|
| 769 |
-
"learning_rate": 1.7214285714285715e-05,
|
| 770 |
-
"loss": 0.0287,
|
| 771 |
-
"step": 12400
|
| 772 |
-
},
|
| 773 |
-
{
|
| 774 |
-
"epoch": 66.01,
|
| 775 |
-
"learning_rate": 1.694973544973545e-05,
|
| 776 |
-
"loss": 0.0222,
|
| 777 |
-
"step": 12500
|
| 778 |
-
},
|
| 779 |
-
{
|
| 780 |
-
"epoch": 66.53,
|
| 781 |
-
"learning_rate": 1.6685185185185187e-05,
|
| 782 |
-
"loss": 0.0302,
|
| 783 |
-
"step": 12600
|
| 784 |
-
},
|
| 785 |
-
{
|
| 786 |
-
"epoch": 67.06,
|
| 787 |
-
"learning_rate": 1.6420634920634923e-05,
|
| 788 |
-
"loss": 0.0252,
|
| 789 |
-
"step": 12700
|
| 790 |
-
},
|
| 791 |
-
{
|
| 792 |
-
"epoch": 67.59,
|
| 793 |
-
"learning_rate": 1.615608465608466e-05,
|
| 794 |
-
"loss": 0.0221,
|
| 795 |
-
"step": 12800
|
| 796 |
-
},
|
| 797 |
-
{
|
| 798 |
-
"epoch": 68.12,
|
| 799 |
-
"learning_rate": 1.5891534391534395e-05,
|
| 800 |
-
"loss": 0.0383,
|
| 801 |
-
"step": 12900
|
| 802 |
-
},
|
| 803 |
-
{
|
| 804 |
-
"epoch": 68.65,
|
| 805 |
-
"learning_rate": 1.5626984126984128e-05,
|
| 806 |
-
"loss": 0.0242,
|
| 807 |
-
"step": 13000
|
| 808 |
-
},
|
| 809 |
-
{
|
| 810 |
-
"epoch": 69.17,
|
| 811 |
-
"learning_rate": 1.5362433862433864e-05,
|
| 812 |
-
"loss": 0.0194,
|
| 813 |
-
"step": 13100
|
| 814 |
-
},
|
| 815 |
-
{
|
| 816 |
-
"epoch": 69.7,
|
| 817 |
-
"learning_rate": 1.50978835978836e-05,
|
| 818 |
-
"loss": 0.0245,
|
| 819 |
-
"step": 13200
|
| 820 |
-
},
|
| 821 |
-
{
|
| 822 |
-
"epoch": 70.23,
|
| 823 |
-
"learning_rate": 1.4833333333333336e-05,
|
| 824 |
-
"loss": 0.0263,
|
| 825 |
-
"step": 13300
|
| 826 |
-
},
|
| 827 |
-
{
|
| 828 |
-
"epoch": 70.76,
|
| 829 |
-
"learning_rate": 1.456878306878307e-05,
|
| 830 |
-
"loss": 0.0276,
|
| 831 |
-
"step": 13400
|
| 832 |
-
},
|
| 833 |
-
{
|
| 834 |
-
"epoch": 71.29,
|
| 835 |
-
"learning_rate": 1.4304232804232806e-05,
|
| 836 |
-
"loss": 0.0213,
|
| 837 |
-
"step": 13500
|
| 838 |
-
},
|
| 839 |
-
{
|
| 840 |
-
"epoch": 71.82,
|
| 841 |
-
"learning_rate": 1.4042328042328043e-05,
|
| 842 |
-
"loss": 0.0296,
|
| 843 |
-
"step": 13600
|
| 844 |
-
},
|
| 845 |
-
{
|
| 846 |
-
"epoch": 72.34,
|
| 847 |
-
"learning_rate": 1.3777777777777778e-05,
|
| 848 |
-
"loss": 0.0231,
|
| 849 |
-
"step": 13700
|
| 850 |
-
},
|
| 851 |
-
{
|
| 852 |
-
"epoch": 72.87,
|
| 853 |
-
"learning_rate": 1.3513227513227514e-05,
|
| 854 |
-
"loss": 0.0241,
|
| 855 |
-
"step": 13800
|
| 856 |
-
},
|
| 857 |
-
{
|
| 858 |
-
"epoch": 73.4,
|
| 859 |
-
"learning_rate": 1.324867724867725e-05,
|
| 860 |
-
"loss": 0.0283,
|
| 861 |
-
"step": 13900
|
| 862 |
-
},
|
| 863 |
-
{
|
| 864 |
-
"epoch": 73.93,
|
| 865 |
-
"learning_rate": 1.2984126984126984e-05,
|
| 866 |
-
"loss": 0.0231,
|
| 867 |
-
"step": 14000
|
| 868 |
-
},
|
| 869 |
-
{
|
| 870 |
-
"epoch": 74.46,
|
| 871 |
-
"learning_rate": 1.271957671957672e-05,
|
| 872 |
-
"loss": 0.0179,
|
| 873 |
-
"step": 14100
|
| 874 |
-
},
|
| 875 |
-
{
|
| 876 |
-
"epoch": 74.98,
|
| 877 |
-
"learning_rate": 1.2455026455026456e-05,
|
| 878 |
-
"loss": 0.0208,
|
| 879 |
-
"step": 14200
|
| 880 |
-
},
|
| 881 |
-
{
|
| 882 |
-
"epoch": 75.51,
|
| 883 |
-
"learning_rate": 1.219047619047619e-05,
|
| 884 |
-
"loss": 0.0277,
|
| 885 |
-
"step": 14300
|
| 886 |
-
},
|
| 887 |
-
{
|
| 888 |
-
"epoch": 76.04,
|
| 889 |
-
"learning_rate": 1.1925925925925927e-05,
|
| 890 |
-
"loss": 0.0195,
|
| 891 |
-
"step": 14400
|
| 892 |
-
},
|
| 893 |
-
{
|
| 894 |
-
"epoch": 76.57,
|
| 895 |
-
"learning_rate": 1.1661375661375661e-05,
|
| 896 |
-
"loss": 0.0188,
|
| 897 |
-
"step": 14500
|
| 898 |
-
},
|
| 899 |
-
{
|
| 900 |
-
"epoch": 77.1,
|
| 901 |
-
"learning_rate": 1.1396825396825397e-05,
|
| 902 |
-
"loss": 0.0184,
|
| 903 |
-
"step": 14600
|
| 904 |
-
},
|
| 905 |
-
{
|
| 906 |
-
"epoch": 77.62,
|
| 907 |
-
"learning_rate": 1.1132275132275133e-05,
|
| 908 |
-
"loss": 0.0165,
|
| 909 |
-
"step": 14700
|
| 910 |
-
},
|
| 911 |
-
{
|
| 912 |
-
"epoch": 78.15,
|
| 913 |
-
"learning_rate": 1.0867724867724868e-05,
|
| 914 |
-
"loss": 0.0245,
|
| 915 |
-
"step": 14800
|
| 916 |
-
},
|
| 917 |
-
{
|
| 918 |
-
"epoch": 78.68,
|
| 919 |
-
"learning_rate": 1.0603174603174604e-05,
|
| 920 |
-
"loss": 0.0331,
|
| 921 |
-
"step": 14900
|
| 922 |
-
},
|
| 923 |
-
{
|
| 924 |
-
"epoch": 79.21,
|
| 925 |
-
"learning_rate": 1.033862433862434e-05,
|
| 926 |
-
"loss": 0.0173,
|
| 927 |
-
"step": 15000
|
| 928 |
-
},
|
| 929 |
-
{
|
| 930 |
-
"epoch": 79.21,
|
| 931 |
-
"eval_accuracy": 0.8968647122383118,
|
| 932 |
-
"eval_loss": 0.5708499550819397,
|
| 933 |
-
"eval_runtime": 49.6061,
|
| 934 |
-
"eval_samples_per_second": 48.865,
|
| 935 |
-
"eval_steps_per_second": 6.108,
|
| 936 |
-
"step": 15000
|
| 937 |
-
},
|
| 938 |
-
{
|
| 939 |
-
"epoch": 79.74,
|
| 940 |
-
"learning_rate": 1.0074074074074074e-05,
|
| 941 |
-
"loss": 0.0243,
|
| 942 |
-
"step": 15100
|
| 943 |
-
},
|
| 944 |
-
{
|
| 945 |
-
"epoch": 80.26,
|
| 946 |
-
"learning_rate": 9.80952380952381e-06,
|
| 947 |
-
"loss": 0.0203,
|
| 948 |
-
"step": 15200
|
| 949 |
-
},
|
| 950 |
-
{
|
| 951 |
-
"epoch": 80.79,
|
| 952 |
-
"learning_rate": 9.544973544973545e-06,
|
| 953 |
-
"loss": 0.018,
|
| 954 |
-
"step": 15300
|
| 955 |
-
},
|
| 956 |
-
{
|
| 957 |
-
"epoch": 81.32,
|
| 958 |
-
"learning_rate": 9.280423280423281e-06,
|
| 959 |
-
"loss": 0.0239,
|
| 960 |
-
"step": 15400
|
| 961 |
-
},
|
| 962 |
-
{
|
| 963 |
-
"epoch": 81.85,
|
| 964 |
-
"learning_rate": 9.015873015873017e-06,
|
| 965 |
-
"loss": 0.0176,
|
| 966 |
-
"step": 15500
|
| 967 |
-
},
|
| 968 |
-
{
|
| 969 |
-
"epoch": 82.38,
|
| 970 |
-
"learning_rate": 8.751322751322751e-06,
|
| 971 |
-
"loss": 0.0231,
|
| 972 |
-
"step": 15600
|
| 973 |
-
},
|
| 974 |
-
{
|
| 975 |
-
"epoch": 82.9,
|
| 976 |
-
"learning_rate": 8.486772486772487e-06,
|
| 977 |
-
"loss": 0.0181,
|
| 978 |
-
"step": 15700
|
| 979 |
-
},
|
| 980 |
-
{
|
| 981 |
-
"epoch": 83.43,
|
| 982 |
-
"learning_rate": 8.222222222222223e-06,
|
| 983 |
-
"loss": 0.0221,
|
| 984 |
-
"step": 15800
|
| 985 |
-
},
|
| 986 |
-
{
|
| 987 |
-
"epoch": 83.96,
|
| 988 |
-
"learning_rate": 7.957671957671958e-06,
|
| 989 |
-
"loss": 0.0132,
|
| 990 |
-
"step": 15900
|
| 991 |
-
},
|
| 992 |
-
{
|
| 993 |
-
"epoch": 84.49,
|
| 994 |
-
"learning_rate": 7.693121693121694e-06,
|
| 995 |
-
"loss": 0.0127,
|
| 996 |
-
"step": 16000
|
| 997 |
-
},
|
| 998 |
-
{
|
| 999 |
-
"epoch": 85.02,
|
| 1000 |
-
"learning_rate": 7.428571428571429e-06,
|
| 1001 |
-
"loss": 0.0178,
|
| 1002 |
-
"step": 16100
|
| 1003 |
-
},
|
| 1004 |
-
{
|
| 1005 |
-
"epoch": 85.54,
|
| 1006 |
-
"learning_rate": 7.1640211640211644e-06,
|
| 1007 |
-
"loss": 0.0176,
|
| 1008 |
-
"step": 16200
|
| 1009 |
-
},
|
| 1010 |
-
{
|
| 1011 |
-
"epoch": 86.07,
|
| 1012 |
-
"learning_rate": 6.8994708994709e-06,
|
| 1013 |
-
"loss": 0.0169,
|
| 1014 |
-
"step": 16300
|
| 1015 |
-
},
|
| 1016 |
-
{
|
| 1017 |
-
"epoch": 86.6,
|
| 1018 |
-
"learning_rate": 6.634920634920636e-06,
|
| 1019 |
-
"loss": 0.0163,
|
| 1020 |
-
"step": 16400
|
| 1021 |
-
},
|
| 1022 |
-
{
|
| 1023 |
-
"epoch": 87.13,
|
| 1024 |
-
"learning_rate": 6.370370370370371e-06,
|
| 1025 |
-
"loss": 0.015,
|
| 1026 |
-
"step": 16500
|
| 1027 |
-
},
|
| 1028 |
-
{
|
| 1029 |
-
"epoch": 87.66,
|
| 1030 |
-
"learning_rate": 6.105820105820106e-06,
|
| 1031 |
-
"loss": 0.022,
|
| 1032 |
-
"step": 16600
|
| 1033 |
-
},
|
| 1034 |
-
{
|
| 1035 |
-
"epoch": 88.18,
|
| 1036 |
-
"learning_rate": 5.841269841269842e-06,
|
| 1037 |
-
"loss": 0.0115,
|
| 1038 |
-
"step": 16700
|
| 1039 |
-
},
|
| 1040 |
-
{
|
| 1041 |
-
"epoch": 88.71,
|
| 1042 |
-
"learning_rate": 5.576719576719577e-06,
|
| 1043 |
-
"loss": 0.0148,
|
| 1044 |
-
"step": 16800
|
| 1045 |
-
},
|
| 1046 |
-
{
|
| 1047 |
-
"epoch": 89.24,
|
| 1048 |
-
"learning_rate": 5.312169312169312e-06,
|
| 1049 |
-
"loss": 0.0153,
|
| 1050 |
-
"step": 16900
|
| 1051 |
-
},
|
| 1052 |
-
{
|
| 1053 |
-
"epoch": 89.77,
|
| 1054 |
-
"learning_rate": 5.047619047619047e-06,
|
| 1055 |
-
"loss": 0.0194,
|
| 1056 |
-
"step": 17000
|
| 1057 |
-
},
|
| 1058 |
-
{
|
| 1059 |
-
"epoch": 90.3,
|
| 1060 |
-
"learning_rate": 4.783068783068783e-06,
|
| 1061 |
-
"loss": 0.0165,
|
| 1062 |
-
"step": 17100
|
| 1063 |
-
},
|
| 1064 |
-
{
|
| 1065 |
-
"epoch": 90.83,
|
| 1066 |
-
"learning_rate": 4.5185185185185185e-06,
|
| 1067 |
-
"loss": 0.013,
|
| 1068 |
-
"step": 17200
|
| 1069 |
-
},
|
| 1070 |
-
{
|
| 1071 |
-
"epoch": 91.35,
|
| 1072 |
-
"learning_rate": 4.253968253968254e-06,
|
| 1073 |
-
"loss": 0.0212,
|
| 1074 |
-
"step": 17300
|
| 1075 |
-
},
|
| 1076 |
-
{
|
| 1077 |
-
"epoch": 91.88,
|
| 1078 |
-
"learning_rate": 3.989417989417989e-06,
|
| 1079 |
-
"loss": 0.0192,
|
| 1080 |
-
"step": 17400
|
| 1081 |
-
},
|
| 1082 |
-
{
|
| 1083 |
-
"epoch": 92.41,
|
| 1084 |
-
"learning_rate": 3.7248677248677246e-06,
|
| 1085 |
-
"loss": 0.0152,
|
| 1086 |
-
"step": 17500
|
| 1087 |
-
},
|
| 1088 |
-
{
|
| 1089 |
-
"epoch": 92.94,
|
| 1090 |
-
"learning_rate": 3.4603174603174603e-06,
|
| 1091 |
-
"loss": 0.02,
|
| 1092 |
-
"step": 17600
|
| 1093 |
-
},
|
| 1094 |
-
{
|
| 1095 |
-
"epoch": 93.47,
|
| 1096 |
-
"learning_rate": 3.1957671957671955e-06,
|
| 1097 |
-
"loss": 0.0089,
|
| 1098 |
-
"step": 17700
|
| 1099 |
-
},
|
| 1100 |
-
{
|
| 1101 |
-
"epoch": 93.99,
|
| 1102 |
-
"learning_rate": 2.9312169312169316e-06,
|
| 1103 |
-
"loss": 0.0124,
|
| 1104 |
-
"step": 17800
|
| 1105 |
-
},
|
| 1106 |
-
{
|
| 1107 |
-
"epoch": 94.52,
|
| 1108 |
-
"learning_rate": 2.666666666666667e-06,
|
| 1109 |
-
"loss": 0.019,
|
| 1110 |
-
"step": 17900
|
| 1111 |
-
},
|
| 1112 |
-
{
|
| 1113 |
-
"epoch": 95.05,
|
| 1114 |
-
"learning_rate": 2.402116402116402e-06,
|
| 1115 |
-
"loss": 0.0151,
|
| 1116 |
-
"step": 18000
|
| 1117 |
-
},
|
| 1118 |
-
{
|
| 1119 |
-
"epoch": 95.58,
|
| 1120 |
-
"learning_rate": 2.1375661375661377e-06,
|
| 1121 |
-
"loss": 0.0184,
|
| 1122 |
-
"step": 18100
|
| 1123 |
-
},
|
| 1124 |
-
{
|
| 1125 |
-
"epoch": 96.11,
|
| 1126 |
-
"learning_rate": 1.873015873015873e-06,
|
| 1127 |
-
"loss": 0.0146,
|
| 1128 |
-
"step": 18200
|
| 1129 |
-
},
|
| 1130 |
-
{
|
| 1131 |
-
"epoch": 96.63,
|
| 1132 |
-
"learning_rate": 1.6084656084656084e-06,
|
| 1133 |
-
"loss": 0.0227,
|
| 1134 |
-
"step": 18300
|
| 1135 |
-
},
|
| 1136 |
-
{
|
| 1137 |
-
"epoch": 97.16,
|
| 1138 |
-
"learning_rate": 1.343915343915344e-06,
|
| 1139 |
-
"loss": 0.0149,
|
| 1140 |
-
"step": 18400
|
| 1141 |
-
},
|
| 1142 |
-
{
|
| 1143 |
-
"epoch": 97.69,
|
| 1144 |
-
"learning_rate": 1.0793650793650795e-06,
|
| 1145 |
-
"loss": 0.015,
|
| 1146 |
-
"step": 18500
|
| 1147 |
-
},
|
| 1148 |
-
{
|
| 1149 |
-
"epoch": 98.22,
|
| 1150 |
-
"learning_rate": 8.148148148148147e-07,
|
| 1151 |
-
"loss": 0.0166,
|
| 1152 |
-
"step": 18600
|
| 1153 |
-
},
|
| 1154 |
-
{
|
| 1155 |
-
"epoch": 98.75,
|
| 1156 |
-
"learning_rate": 5.502645502645503e-07,
|
| 1157 |
-
"loss": 0.0167,
|
| 1158 |
-
"step": 18700
|
| 1159 |
-
},
|
| 1160 |
-
{
|
| 1161 |
-
"epoch": 99.27,
|
| 1162 |
-
"learning_rate": 2.8571428571428575e-07,
|
| 1163 |
-
"loss": 0.0152,
|
| 1164 |
-
"step": 18800
|
| 1165 |
-
},
|
| 1166 |
-
{
|
| 1167 |
-
"epoch": 99.8,
|
| 1168 |
-
"learning_rate": 2.1164021164021167e-08,
|
| 1169 |
-
"loss": 0.0164,
|
| 1170 |
-
"step": 18900
|
| 1171 |
-
},
|
| 1172 |
-
{
|
| 1173 |
-
"epoch": 99.8,
|
| 1174 |
-
"step": 18900,
|
| 1175 |
-
"total_flos": 3.6752439370752e+19,
|
| 1176 |
-
"train_loss": 0.05161126141825681,
|
| 1177 |
-
"train_runtime": 41136.7074,
|
| 1178 |
-
"train_samples_per_second": 29.463,
|
| 1179 |
-
"train_steps_per_second": 0.459
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
-
"epoch":
|
| 1183 |
-
"eval_accuracy": 0.
|
| 1184 |
-
"eval_loss": 0.
|
| 1185 |
-
"eval_runtime": 50.
|
| 1186 |
-
"eval_samples_per_second":
|
| 1187 |
-
"eval_steps_per_second":
|
| 1188 |
-
"step":
|
| 1189 |
}
|
| 1190 |
],
|
| 1191 |
"logging_steps": 100,
|
| 1192 |
-
"max_steps":
|
| 1193 |
-
"num_train_epochs":
|
| 1194 |
"save_steps": 2000,
|
| 1195 |
-
"total_flos":
|
| 1196 |
"trial_name": null,
|
| 1197 |
"trial_params": null
|
| 1198 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 59.881188118811885,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 11340,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.53,
|
| 13 |
+
"learning_rate": 4.956349206349207e-05,
|
| 14 |
+
"loss": 0.0798,
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 1.06,
|
| 19 |
+
"learning_rate": 4.912257495590829e-05,
|
| 20 |
+
"loss": 0.0816,
|
| 21 |
"step": 200
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 1.58,
|
| 25 |
+
"learning_rate": 4.868165784832452e-05,
|
| 26 |
+
"loss": 0.0761,
|
| 27 |
"step": 300
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 2.11,
|
| 31 |
+
"learning_rate": 4.824514991181658e-05,
|
| 32 |
+
"loss": 0.0723,
|
| 33 |
"step": 400
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 2.64,
|
| 37 |
+
"learning_rate": 4.7804232804232806e-05,
|
| 38 |
+
"loss": 0.0601,
|
| 39 |
"step": 500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 3.17,
|
| 43 |
+
"learning_rate": 4.736331569664903e-05,
|
| 44 |
+
"loss": 0.0593,
|
| 45 |
"step": 600
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 3.7,
|
| 49 |
+
"learning_rate": 4.692239858906526e-05,
|
| 50 |
+
"loss": 0.0692,
|
| 51 |
"step": 700
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 4.22,
|
| 55 |
+
"learning_rate": 4.648148148148148e-05,
|
| 56 |
+
"loss": 0.0587,
|
| 57 |
"step": 800
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 4.75,
|
| 61 |
+
"learning_rate": 4.604056437389771e-05,
|
| 62 |
+
"loss": 0.0598,
|
| 63 |
"step": 900
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 5.28,
|
| 67 |
+
"learning_rate": 4.559964726631393e-05,
|
| 68 |
+
"loss": 0.0559,
|
| 69 |
+
"step": 1000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 5.28,
|
| 73 |
+
"eval_accuracy": 0.9191418886184692,
|
| 74 |
+
"eval_loss": 0.3096904754638672,
|
| 75 |
+
"eval_runtime": 51.3243,
|
| 76 |
+
"eval_samples_per_second": 47.229,
|
| 77 |
+
"eval_steps_per_second": 5.904,
|
| 78 |
"step": 1000
|
| 79 |
},
|
| 80 |
{
|
| 81 |
"epoch": 5.81,
|
| 82 |
+
"learning_rate": 4.515873015873016e-05,
|
| 83 |
+
"loss": 0.0552,
|
| 84 |
"step": 1100
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 6.34,
|
| 88 |
+
"learning_rate": 4.471781305114639e-05,
|
| 89 |
+
"loss": 0.0519,
|
| 90 |
"step": 1200
|
| 91 |
},
|
| 92 |
{
|
| 93 |
"epoch": 6.86,
|
| 94 |
+
"learning_rate": 4.428130511463845e-05,
|
| 95 |
+
"loss": 0.0536,
|
| 96 |
"step": 1300
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 7.39,
|
| 100 |
+
"learning_rate": 4.3840388007054675e-05,
|
| 101 |
+
"loss": 0.0438,
|
| 102 |
"step": 1400
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 7.92,
|
| 106 |
+
"learning_rate": 4.33994708994709e-05,
|
| 107 |
+
"loss": 0.0436,
|
| 108 |
"step": 1500
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 8.45,
|
| 112 |
+
"learning_rate": 4.295855379188713e-05,
|
| 113 |
+
"loss": 0.0447,
|
| 114 |
"step": 1600
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 8.98,
|
| 118 |
+
"learning_rate": 4.2517636684303355e-05,
|
| 119 |
+
"loss": 0.0535,
|
| 120 |
"step": 1700
|
| 121 |
},
|
| 122 |
{
|
| 123 |
"epoch": 9.5,
|
| 124 |
+
"learning_rate": 4.207671957671958e-05,
|
| 125 |
+
"loss": 0.0467,
|
| 126 |
"step": 1800
|
| 127 |
},
|
| 128 |
{
|
| 129 |
"epoch": 10.03,
|
| 130 |
+
"learning_rate": 4.16358024691358e-05,
|
| 131 |
+
"loss": 0.0557,
|
| 132 |
"step": 1900
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 10.56,
|
| 136 |
+
"learning_rate": 4.1194885361552036e-05,
|
| 137 |
+
"loss": 0.047,
|
| 138 |
+
"step": 2000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 10.56,
|
| 142 |
+
"eval_accuracy": 0.9191418886184692,
|
| 143 |
+
"eval_loss": 0.34823155403137207,
|
| 144 |
+
"eval_runtime": 51.4649,
|
| 145 |
+
"eval_samples_per_second": 47.1,
|
| 146 |
+
"eval_steps_per_second": 5.888,
|
| 147 |
"step": 2000
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 11.09,
|
| 151 |
+
"learning_rate": 4.0753968253968256e-05,
|
| 152 |
+
"loss": 0.0397,
|
| 153 |
"step": 2100
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 11.62,
|
| 157 |
+
"learning_rate": 4.031305114638448e-05,
|
| 158 |
+
"loss": 0.0475,
|
| 159 |
"step": 2200
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 12.15,
|
| 163 |
+
"learning_rate": 3.987213403880071e-05,
|
| 164 |
+
"loss": 0.0478,
|
| 165 |
"step": 2300
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 12.67,
|
| 169 |
+
"learning_rate": 3.9431216931216936e-05,
|
| 170 |
+
"loss": 0.0421,
|
| 171 |
"step": 2400
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 13.2,
|
| 175 |
+
"learning_rate": 3.8994708994709e-05,
|
| 176 |
+
"loss": 0.0408,
|
| 177 |
"step": 2500
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 13.73,
|
| 181 |
+
"learning_rate": 3.8553791887125224e-05,
|
| 182 |
+
"loss": 0.0368,
|
| 183 |
"step": 2600
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 14.26,
|
| 187 |
+
"learning_rate": 3.8112874779541445e-05,
|
| 188 |
+
"loss": 0.0379,
|
| 189 |
"step": 2700
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 14.79,
|
| 193 |
+
"learning_rate": 3.767195767195768e-05,
|
| 194 |
+
"loss": 0.052,
|
| 195 |
"step": 2800
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 15.31,
|
| 199 |
+
"learning_rate": 3.72310405643739e-05,
|
| 200 |
+
"loss": 0.0424,
|
| 201 |
"step": 2900
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 15.84,
|
| 205 |
+
"learning_rate": 3.6790123456790125e-05,
|
| 206 |
+
"loss": 0.0402,
|
| 207 |
+
"step": 3000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 15.84,
|
| 211 |
+
"eval_accuracy": 0.9080032706260681,
|
| 212 |
+
"eval_loss": 0.3889801502227783,
|
| 213 |
+
"eval_runtime": 50.7451,
|
| 214 |
+
"eval_samples_per_second": 47.768,
|
| 215 |
+
"eval_steps_per_second": 5.971,
|
| 216 |
"step": 3000
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 16.37,
|
| 220 |
+
"learning_rate": 3.634920634920635e-05,
|
| 221 |
+
"loss": 0.0417,
|
| 222 |
"step": 3100
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 16.9,
|
| 226 |
+
"learning_rate": 3.590828924162258e-05,
|
| 227 |
+
"loss": 0.0419,
|
| 228 |
"step": 3200
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 17.43,
|
| 232 |
+
"learning_rate": 3.54673721340388e-05,
|
| 233 |
+
"loss": 0.0439,
|
| 234 |
"step": 3300
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 17.95,
|
| 238 |
+
"learning_rate": 3.502645502645503e-05,
|
| 239 |
+
"loss": 0.0446,
|
| 240 |
"step": 3400
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 18.48,
|
| 244 |
+
"learning_rate": 3.458553791887125e-05,
|
| 245 |
+
"loss": 0.0339,
|
| 246 |
"step": 3500
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 19.01,
|
| 250 |
+
"learning_rate": 3.414462081128748e-05,
|
| 251 |
+
"loss": 0.0301,
|
| 252 |
"step": 3600
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 19.54,
|
| 256 |
+
"learning_rate": 3.3703703703703706e-05,
|
| 257 |
+
"loss": 0.0277,
|
| 258 |
"step": 3700
|
| 259 |
},
|
| 260 |
{
|
| 261 |
"epoch": 20.07,
|
| 262 |
+
"learning_rate": 3.326278659611993e-05,
|
| 263 |
+
"loss": 0.0333,
|
| 264 |
"step": 3800
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 20.59,
|
| 268 |
+
"learning_rate": 3.282186948853615e-05,
|
| 269 |
+
"loss": 0.0298,
|
| 270 |
"step": 3900
|
| 271 |
},
|
| 272 |
{
|
| 273 |
"epoch": 21.12,
|
| 274 |
+
"learning_rate": 3.2380952380952386e-05,
|
| 275 |
+
"loss": 0.0328,
|
| 276 |
+
"step": 4000
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 21.12,
|
| 280 |
+
"eval_accuracy": 0.9150164723396301,
|
| 281 |
+
"eval_loss": 0.37457939982414246,
|
| 282 |
+
"eval_runtime": 50.3165,
|
| 283 |
+
"eval_samples_per_second": 48.175,
|
| 284 |
+
"eval_steps_per_second": 6.022,
|
| 285 |
"step": 4000
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 21.65,
|
| 289 |
+
"learning_rate": 3.1940035273368606e-05,
|
| 290 |
+
"loss": 0.0298,
|
| 291 |
"step": 4100
|
| 292 |
},
|
| 293 |
{
|
| 294 |
"epoch": 22.18,
|
| 295 |
+
"learning_rate": 3.149911816578483e-05,
|
| 296 |
+
"loss": 0.0314,
|
| 297 |
"step": 4200
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 22.71,
|
| 301 |
+
"learning_rate": 3.105820105820106e-05,
|
| 302 |
+
"loss": 0.0253,
|
| 303 |
"step": 4300
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 23.23,
|
| 307 |
+
"learning_rate": 3.061728395061729e-05,
|
| 308 |
+
"loss": 0.0339,
|
| 309 |
"step": 4400
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 23.76,
|
| 313 |
+
"learning_rate": 3.017636684303351e-05,
|
| 314 |
+
"loss": 0.0266,
|
| 315 |
"step": 4500
|
| 316 |
},
|
| 317 |
{
|
| 318 |
"epoch": 24.29,
|
| 319 |
+
"learning_rate": 2.973544973544974e-05,
|
| 320 |
+
"loss": 0.0361,
|
| 321 |
"step": 4600
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 24.82,
|
| 325 |
+
"learning_rate": 2.929453262786596e-05,
|
| 326 |
+
"loss": 0.0305,
|
| 327 |
"step": 4700
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 25.35,
|
| 331 |
+
"learning_rate": 2.885361552028219e-05,
|
| 332 |
+
"loss": 0.0294,
|
| 333 |
"step": 4800
|
| 334 |
},
|
| 335 |
{
|
| 336 |
"epoch": 25.87,
|
| 337 |
+
"learning_rate": 2.8412698412698414e-05,
|
| 338 |
+
"loss": 0.0339,
|
| 339 |
"step": 4900
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 26.4,
|
| 343 |
+
"learning_rate": 2.797178130511464e-05,
|
| 344 |
+
"loss": 0.0189,
|
| 345 |
"step": 5000
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 26.4,
|
| 349 |
+
"eval_accuracy": 0.9113036394119263,
|
| 350 |
+
"eval_loss": 0.42735978960990906,
|
| 351 |
+
"eval_runtime": 49.4145,
|
| 352 |
+
"eval_samples_per_second": 49.054,
|
| 353 |
+
"eval_steps_per_second": 6.132,
|
| 354 |
"step": 5000
|
| 355 |
},
|
| 356 |
{
|
| 357 |
"epoch": 26.93,
|
| 358 |
+
"learning_rate": 2.7530864197530864e-05,
|
| 359 |
+
"loss": 0.0285,
|
| 360 |
"step": 5100
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 27.46,
|
| 364 |
+
"learning_rate": 2.7089947089947094e-05,
|
| 365 |
+
"loss": 0.0307,
|
| 366 |
"step": 5200
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 27.99,
|
| 370 |
+
"learning_rate": 2.6649029982363318e-05,
|
| 371 |
+
"loss": 0.0269,
|
| 372 |
"step": 5300
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 28.51,
|
| 376 |
+
"learning_rate": 2.6208112874779544e-05,
|
| 377 |
+
"loss": 0.0292,
|
| 378 |
"step": 5400
|
| 379 |
},
|
| 380 |
{
|
| 381 |
"epoch": 29.04,
|
| 382 |
+
"learning_rate": 2.5767195767195768e-05,
|
| 383 |
+
"loss": 0.032,
|
| 384 |
"step": 5500
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 29.57,
|
| 388 |
+
"learning_rate": 2.5326278659611995e-05,
|
| 389 |
+
"loss": 0.0297,
|
| 390 |
"step": 5600
|
| 391 |
},
|
| 392 |
{
|
| 393 |
"epoch": 30.1,
|
| 394 |
+
"learning_rate": 2.4885361552028218e-05,
|
| 395 |
+
"loss": 0.027,
|
| 396 |
"step": 5700
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 30.63,
|
| 400 |
+
"learning_rate": 2.4444444444444445e-05,
|
| 401 |
+
"loss": 0.0258,
|
| 402 |
"step": 5800
|
| 403 |
},
|
| 404 |
{
|
| 405 |
"epoch": 31.16,
|
| 406 |
+
"learning_rate": 2.4003527336860672e-05,
|
| 407 |
+
"loss": 0.0237,
|
| 408 |
"step": 5900
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 31.68,
|
| 412 |
+
"learning_rate": 2.3562610229276895e-05,
|
| 413 |
+
"loss": 0.0187,
|
| 414 |
+
"step": 6000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 31.68,
|
| 418 |
+
"eval_accuracy": 0.9100660085678101,
|
| 419 |
+
"eval_loss": 0.4131234884262085,
|
| 420 |
+
"eval_runtime": 49.5018,
|
| 421 |
+
"eval_samples_per_second": 48.968,
|
| 422 |
+
"eval_steps_per_second": 6.121,
|
| 423 |
"step": 6000
|
| 424 |
},
|
| 425 |
{
|
| 426 |
"epoch": 32.21,
|
| 427 |
+
"learning_rate": 2.3121693121693122e-05,
|
| 428 |
+
"loss": 0.0255,
|
| 429 |
"step": 6100
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 32.74,
|
| 433 |
+
"learning_rate": 2.268077601410935e-05,
|
| 434 |
+
"loss": 0.0244,
|
| 435 |
"step": 6200
|
| 436 |
},
|
| 437 |
{
|
| 438 |
"epoch": 33.27,
|
| 439 |
+
"learning_rate": 2.2239858906525572e-05,
|
| 440 |
+
"loss": 0.0273,
|
| 441 |
"step": 6300
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 33.8,
|
| 445 |
+
"learning_rate": 2.17989417989418e-05,
|
| 446 |
+
"loss": 0.0214,
|
| 447 |
"step": 6400
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 34.32,
|
| 451 |
+
"learning_rate": 2.1358024691358026e-05,
|
| 452 |
+
"loss": 0.0265,
|
| 453 |
"step": 6500
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 34.85,
|
| 457 |
+
"learning_rate": 2.091710758377425e-05,
|
| 458 |
+
"loss": 0.0276,
|
| 459 |
"step": 6600
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 35.38,
|
| 463 |
+
"learning_rate": 2.0476190476190476e-05,
|
| 464 |
+
"loss": 0.0153,
|
| 465 |
"step": 6700
|
| 466 |
},
|
| 467 |
{
|
| 468 |
"epoch": 35.91,
|
| 469 |
+
"learning_rate": 2.0035273368606703e-05,
|
| 470 |
+
"loss": 0.0246,
|
| 471 |
"step": 6800
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 36.44,
|
| 475 |
+
"learning_rate": 1.959435626102293e-05,
|
| 476 |
+
"loss": 0.0266,
|
| 477 |
"step": 6900
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 36.96,
|
| 481 |
+
"learning_rate": 1.9153439153439153e-05,
|
| 482 |
+
"loss": 0.0203,
|
| 483 |
+
"step": 7000
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 36.96,
|
| 487 |
+
"eval_accuracy": 0.9236798882484436,
|
| 488 |
+
"eval_loss": 0.3643423020839691,
|
| 489 |
+
"eval_runtime": 49.3507,
|
| 490 |
+
"eval_samples_per_second": 49.118,
|
| 491 |
+
"eval_steps_per_second": 6.14,
|
| 492 |
"step": 7000
|
| 493 |
},
|
| 494 |
{
|
| 495 |
"epoch": 37.49,
|
| 496 |
+
"learning_rate": 1.871252204585538e-05,
|
| 497 |
+
"loss": 0.0225,
|
| 498 |
"step": 7100
|
| 499 |
},
|
| 500 |
{
|
| 501 |
"epoch": 38.02,
|
| 502 |
+
"learning_rate": 1.8271604938271607e-05,
|
| 503 |
+
"loss": 0.0296,
|
| 504 |
"step": 7200
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 38.55,
|
| 508 |
+
"learning_rate": 1.783068783068783e-05,
|
| 509 |
+
"loss": 0.0181,
|
| 510 |
"step": 7300
|
| 511 |
},
|
| 512 |
{
|
| 513 |
"epoch": 39.08,
|
| 514 |
+
"learning_rate": 1.7389770723104057e-05,
|
| 515 |
+
"loss": 0.0184,
|
| 516 |
"step": 7400
|
| 517 |
},
|
| 518 |
{
|
| 519 |
"epoch": 39.6,
|
| 520 |
+
"learning_rate": 1.6948853615520284e-05,
|
| 521 |
+
"loss": 0.0191,
|
| 522 |
"step": 7500
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 40.13,
|
| 526 |
+
"learning_rate": 1.6507936507936507e-05,
|
| 527 |
+
"loss": 0.0224,
|
| 528 |
"step": 7600
|
| 529 |
},
|
| 530 |
{
|
| 531 |
"epoch": 40.66,
|
| 532 |
+
"learning_rate": 1.6067019400352734e-05,
|
| 533 |
+
"loss": 0.0161,
|
| 534 |
"step": 7700
|
| 535 |
},
|
| 536 |
{
|
| 537 |
"epoch": 41.19,
|
| 538 |
+
"learning_rate": 1.562610229276896e-05,
|
| 539 |
+
"loss": 0.0211,
|
| 540 |
"step": 7800
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 41.72,
|
| 544 |
+
"learning_rate": 1.5185185185185186e-05,
|
| 545 |
+
"loss": 0.0165,
|
| 546 |
"step": 7900
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 42.24,
|
| 550 |
+
"learning_rate": 1.4744268077601411e-05,
|
| 551 |
+
"loss": 0.0147,
|
| 552 |
+
"step": 8000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 42.24,
|
| 556 |
+
"eval_accuracy": 0.9294554591178894,
|
| 557 |
+
"eval_loss": 0.3574332892894745,
|
| 558 |
+
"eval_runtime": 49.7962,
|
| 559 |
+
"eval_samples_per_second": 48.678,
|
| 560 |
+
"eval_steps_per_second": 6.085,
|
| 561 |
"step": 8000
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 42.77,
|
| 565 |
+
"learning_rate": 1.4303350970017638e-05,
|
| 566 |
+
"loss": 0.0235,
|
| 567 |
"step": 8100
|
| 568 |
},
|
| 569 |
{
|
| 570 |
"epoch": 43.3,
|
| 571 |
+
"learning_rate": 1.3862433862433863e-05,
|
| 572 |
+
"loss": 0.0207,
|
| 573 |
"step": 8200
|
| 574 |
},
|
| 575 |
{
|
| 576 |
"epoch": 43.83,
|
| 577 |
+
"learning_rate": 1.3421516754850088e-05,
|
| 578 |
+
"loss": 0.0139,
|
| 579 |
"step": 8300
|
| 580 |
},
|
| 581 |
{
|
| 582 |
"epoch": 44.36,
|
| 583 |
+
"learning_rate": 1.2980599647266315e-05,
|
| 584 |
+
"loss": 0.0168,
|
| 585 |
"step": 8400
|
| 586 |
},
|
| 587 |
{
|
| 588 |
"epoch": 44.88,
|
| 589 |
+
"learning_rate": 1.253968253968254e-05,
|
| 590 |
+
"loss": 0.0146,
|
| 591 |
"step": 8500
|
| 592 |
},
|
| 593 |
{
|
| 594 |
"epoch": 45.41,
|
| 595 |
+
"learning_rate": 1.2098765432098767e-05,
|
| 596 |
+
"loss": 0.0149,
|
| 597 |
"step": 8600
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 45.94,
|
| 601 |
+
"learning_rate": 1.1657848324514992e-05,
|
| 602 |
+
"loss": 0.0155,
|
| 603 |
"step": 8700
|
| 604 |
},
|
| 605 |
{
|
| 606 |
"epoch": 46.47,
|
| 607 |
+
"learning_rate": 1.1221340388007055e-05,
|
| 608 |
+
"loss": 0.0196,
|
| 609 |
"step": 8800
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 47.0,
|
| 613 |
+
"learning_rate": 1.0780423280423282e-05,
|
| 614 |
+
"loss": 0.0229,
|
| 615 |
"step": 8900
|
| 616 |
},
|
| 617 |
{
|
| 618 |
"epoch": 47.52,
|
| 619 |
+
"learning_rate": 1.0339506172839507e-05,
|
| 620 |
+
"loss": 0.0148,
|
| 621 |
+
"step": 9000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 47.52,
|
| 625 |
+
"eval_accuracy": 0.9220296740531921,
|
| 626 |
+
"eval_loss": 0.36532989144325256,
|
| 627 |
+
"eval_runtime": 50.0277,
|
| 628 |
+
"eval_samples_per_second": 48.453,
|
| 629 |
+
"eval_steps_per_second": 6.057,
|
| 630 |
"step": 9000
|
| 631 |
},
|
| 632 |
{
|
| 633 |
"epoch": 48.05,
|
| 634 |
+
"learning_rate": 9.898589065255732e-06,
|
| 635 |
+
"loss": 0.0133,
|
| 636 |
"step": 9100
|
| 637 |
},
|
| 638 |
{
|
| 639 |
"epoch": 48.58,
|
| 640 |
+
"learning_rate": 9.457671957671959e-06,
|
| 641 |
+
"loss": 0.0131,
|
| 642 |
"step": 9200
|
| 643 |
},
|
| 644 |
{
|
| 645 |
"epoch": 49.11,
|
| 646 |
+
"learning_rate": 9.016754850088184e-06,
|
| 647 |
+
"loss": 0.0121,
|
| 648 |
"step": 9300
|
| 649 |
},
|
| 650 |
{
|
| 651 |
"epoch": 49.64,
|
| 652 |
+
"learning_rate": 8.575837742504409e-06,
|
| 653 |
+
"loss": 0.0168,
|
| 654 |
"step": 9400
|
| 655 |
},
|
| 656 |
{
|
| 657 |
"epoch": 50.17,
|
| 658 |
+
"learning_rate": 8.134920634920636e-06,
|
| 659 |
+
"loss": 0.0148,
|
| 660 |
"step": 9500
|
| 661 |
},
|
| 662 |
{
|
| 663 |
"epoch": 50.69,
|
| 664 |
+
"learning_rate": 7.694003527336861e-06,
|
| 665 |
+
"loss": 0.0129,
|
| 666 |
"step": 9600
|
| 667 |
},
|
| 668 |
{
|
| 669 |
"epoch": 51.22,
|
| 670 |
+
"learning_rate": 7.253086419753087e-06,
|
| 671 |
+
"loss": 0.012,
|
| 672 |
"step": 9700
|
| 673 |
},
|
| 674 |
{
|
| 675 |
"epoch": 51.75,
|
| 676 |
+
"learning_rate": 6.812169312169313e-06,
|
| 677 |
+
"loss": 0.0112,
|
| 678 |
"step": 9800
|
| 679 |
},
|
| 680 |
{
|
| 681 |
"epoch": 52.28,
|
| 682 |
+
"learning_rate": 6.371252204585539e-06,
|
| 683 |
+
"loss": 0.0193,
|
| 684 |
"step": 9900
|
| 685 |
},
|
| 686 |
{
|
| 687 |
"epoch": 52.81,
|
| 688 |
+
"learning_rate": 5.930335097001764e-06,
|
| 689 |
+
"loss": 0.0137,
|
| 690 |
"step": 10000
|
| 691 |
},
|
| 692 |
{
|
| 693 |
"epoch": 52.81,
|
| 694 |
+
"eval_accuracy": 0.9352310299873352,
|
| 695 |
+
"eval_loss": 0.3257134258747101,
|
| 696 |
+
"eval_runtime": 50.886,
|
| 697 |
+
"eval_samples_per_second": 47.636,
|
| 698 |
+
"eval_steps_per_second": 5.954,
|
| 699 |
"step": 10000
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 53.33,
|
| 703 |
+
"learning_rate": 5.489417989417989e-06,
|
| 704 |
+
"loss": 0.0171,
|
| 705 |
"step": 10100
|
| 706 |
},
|
| 707 |
{
|
| 708 |
"epoch": 53.86,
|
| 709 |
+
"learning_rate": 5.048500881834215e-06,
|
| 710 |
+
"loss": 0.0169,
|
| 711 |
"step": 10200
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 54.39,
|
| 715 |
+
"learning_rate": 4.611992945326279e-06,
|
| 716 |
+
"loss": 0.0136,
|
| 717 |
"step": 10300
|
| 718 |
},
|
| 719 |
{
|
| 720 |
"epoch": 54.92,
|
| 721 |
+
"learning_rate": 4.171075837742505e-06,
|
| 722 |
+
"loss": 0.0152,
|
| 723 |
"step": 10400
|
| 724 |
},
|
| 725 |
{
|
| 726 |
"epoch": 55.45,
|
| 727 |
+
"learning_rate": 3.7301587301587305e-06,
|
| 728 |
+
"loss": 0.015,
|
| 729 |
"step": 10500
|
| 730 |
},
|
| 731 |
{
|
| 732 |
"epoch": 55.97,
|
| 733 |
+
"learning_rate": 3.289241622574956e-06,
|
| 734 |
+
"loss": 0.0136,
|
| 735 |
"step": 10600
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 56.5,
|
| 739 |
+
"learning_rate": 2.848324514991182e-06,
|
| 740 |
+
"loss": 0.0156,
|
| 741 |
"step": 10700
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 57.03,
|
| 745 |
+
"learning_rate": 2.4074074074074075e-06,
|
| 746 |
+
"loss": 0.0137,
|
| 747 |
"step": 10800
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 57.56,
|
| 751 |
+
"learning_rate": 1.9664902998236335e-06,
|
| 752 |
+
"loss": 0.0147,
|
| 753 |
"step": 10900
|
| 754 |
},
|
| 755 |
{
|
| 756 |
"epoch": 58.09,
|
| 757 |
+
"learning_rate": 1.525573192239859e-06,
|
| 758 |
+
"loss": 0.0174,
|
| 759 |
+
"step": 11000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 58.09,
|
| 763 |
+
"eval_accuracy": 0.933993399143219,
|
| 764 |
+
"eval_loss": 0.30968689918518066,
|
| 765 |
+
"eval_runtime": 51.0931,
|
| 766 |
+
"eval_samples_per_second": 47.443,
|
| 767 |
+
"eval_steps_per_second": 5.93,
|
| 768 |
"step": 11000
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 58.61,
|
| 772 |
+
"learning_rate": 1.0846560846560847e-06,
|
| 773 |
+
"loss": 0.0163,
|
| 774 |
"step": 11100
|
| 775 |
},
|
| 776 |
{
|
| 777 |
"epoch": 59.14,
|
| 778 |
+
"learning_rate": 6.437389770723105e-07,
|
| 779 |
+
"loss": 0.0096,
|
| 780 |
"step": 11200
|
| 781 |
},
|
| 782 |
{
|
| 783 |
"epoch": 59.67,
|
| 784 |
+
"learning_rate": 2.0282186948853617e-07,
|
| 785 |
+
"loss": 0.0121,
|
| 786 |
"step": 11300
|
| 787 |
},
|
| 788 |
{
|
| 789 |
+
"epoch": 59.88,
|
| 790 |
+
"step": 11340,
|
| 791 |
+
"total_flos": 2.20514636224512e+19,
|
| 792 |
+
"train_loss": 0.030625741817122836,
|
| 793 |
+
"train_runtime": 25244.9182,
|
| 794 |
+
"train_samples_per_second": 28.806,
|
| 795 |
+
"train_steps_per_second": 0.449
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 796 |
},
|
| 797 |
{
|
| 798 |
+
"epoch": 59.88,
|
| 799 |
+
"eval_accuracy": 0.9323432445526123,
|
| 800 |
+
"eval_loss": 0.3117374777793884,
|
| 801 |
+
"eval_runtime": 50.8983,
|
| 802 |
+
"eval_samples_per_second": 47.624,
|
| 803 |
+
"eval_steps_per_second": 5.953,
|
| 804 |
+
"step": 11340
|
| 805 |
}
|
| 806 |
],
|
| 807 |
"logging_steps": 100,
|
| 808 |
+
"max_steps": 11340,
|
| 809 |
+
"num_train_epochs": 60,
|
| 810 |
"save_steps": 2000,
|
| 811 |
+
"total_flos": 2.20514636224512e+19,
|
| 812 |
"trial_name": null,
|
| 813 |
"trial_params": null
|
| 814 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4155
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34a256b8a5d2d883823dcb395af53cbfffa6c0546324fd46a6cbd61fd66d4518
|
| 3 |
size 4155
|