End of training
Browse files- README.md +5 -3
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Oct11_09-57-14_ed9849b3ed7e/events.out.tfevents.1728642265.ed9849b3ed7e.18938.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +991 -411
README.md
CHANGED
|
@@ -3,6 +3,8 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: timm/mobilenetv3_large_100.miil_in21k
|
| 5 |
tags:
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
@@ -16,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# test-timm
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [timm/mobilenetv3_large_100.miil_in21k](https://huggingface.co/timm/mobilenetv3_large_100.miil_in21k) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: timm/mobilenetv3_large_100.miil_in21k
|
| 5 |
tags:
|
| 6 |
+
- image-classification
|
| 7 |
+
- vision
|
| 8 |
- generated_from_trainer
|
| 9 |
metrics:
|
| 10 |
- accuracy
|
|
|
|
| 18 |
|
| 19 |
# test-timm
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [timm/mobilenetv3_large_100.miil_in21k](https://huggingface.co/timm/mobilenetv3_large_100.miil_in21k) on the davanstrien/zenodo-presentations-open-labels dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.4904
|
| 24 |
+
- Accuracy: 0.7874
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second":
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy": 0.7874015748031497,
|
| 4 |
+
"eval_loss": 0.4903908967971802,
|
| 5 |
+
"eval_runtime": 3.9654,
|
| 6 |
+
"eval_samples_per_second": 64.055,
|
| 7 |
+
"eval_steps_per_second": 1.009,
|
| 8 |
+
"total_flos": 2.72467378584576e+17,
|
| 9 |
+
"train_loss": 0.4791690407628598,
|
| 10 |
+
"train_runtime": 1616.3187,
|
| 11 |
+
"train_samples_per_second": 44.391,
|
| 12 |
+
"train_steps_per_second": 0.711
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy": 0.7874015748031497,
|
| 4 |
+
"eval_loss": 0.4903908967971802,
|
| 5 |
+
"eval_runtime": 3.9654,
|
| 6 |
+
"eval_samples_per_second": 64.055,
|
| 7 |
+
"eval_steps_per_second": 1.009
|
| 8 |
}
|
runs/Oct11_09-57-14_ed9849b3ed7e/events.out.tfevents.1728642265.ed9849b3ed7e.18938.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a78099521846bf932e10ca606f522731882787ea3ebe82eb67d5e94cfa500323
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"total_flos": 2.72467378584576e+17,
|
| 4 |
+
"train_loss": 0.4791690407628598,
|
| 5 |
+
"train_runtime": 1616.3187,
|
| 6 |
+
"train_samples_per_second": 44.391,
|
| 7 |
+
"train_steps_per_second": 0.711
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,702 +1,1282 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./beans_outputs/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": 3.
|
| 14 |
-
"learning_rate": 1.
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 1.
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0
|
| 27 |
-
"
|
| 28 |
-
"
|
| 29 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch":
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 1.
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0
|
| 41 |
-
"
|
| 42 |
-
"
|
| 43 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch":
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate": 1.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0
|
| 55 |
-
"
|
| 56 |
-
"
|
| 57 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch":
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate": 1.
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch":
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 1.
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0
|
| 76 |
-
"
|
| 77 |
-
"
|
| 78 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch":
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate": 1.
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0
|
| 90 |
-
"
|
| 91 |
-
"
|
| 92 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch":
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate": 1.
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0
|
| 104 |
-
"
|
| 105 |
-
"
|
| 106 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch":
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate": 1.
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch":
|
| 118 |
-
"grad_norm": 2.
|
| 119 |
-
"learning_rate": 1.
|
| 120 |
-
"loss": 0.
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0
|
| 125 |
-
"
|
| 126 |
-
"
|
| 127 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 128 |
"step": 170
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch":
|
| 132 |
-
"grad_norm":
|
| 133 |
-
"learning_rate": 1.
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch":
|
| 139 |
-
"eval_accuracy": 0.
|
| 140 |
-
"eval_loss": 0.
|
| 141 |
-
"eval_runtime":
|
| 142 |
-
"eval_samples_per_second":
|
| 143 |
-
"eval_steps_per_second":
|
| 144 |
-
"step":
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch":
|
| 148 |
-
"grad_norm": 3.
|
| 149 |
-
"learning_rate": 1.
|
| 150 |
-
"loss": 0.
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
-
"epoch":
|
| 155 |
-
"grad_norm": 2.
|
| 156 |
-
"learning_rate": 1.
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch":
|
| 162 |
-
"
|
| 163 |
-
"
|
| 164 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 165 |
"step": 210
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"epoch":
|
| 169 |
-
"grad_norm":
|
| 170 |
-
"learning_rate": 1.
|
| 171 |
-
"loss": 0.
|
| 172 |
"step": 220
|
| 173 |
},
|
| 174 |
{
|
| 175 |
-
"epoch":
|
| 176 |
-
"grad_norm":
|
| 177 |
-
"learning_rate": 1.
|
| 178 |
-
"loss": 0.
|
| 179 |
"step": 230
|
| 180 |
},
|
| 181 |
{
|
| 182 |
-
"epoch":
|
| 183 |
-
"
|
| 184 |
-
"
|
| 185 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 186 |
"step": 240
|
| 187 |
},
|
| 188 |
{
|
| 189 |
-
"epoch":
|
| 190 |
-
"grad_norm":
|
| 191 |
-
"learning_rate": 1.
|
| 192 |
-
"loss": 0.
|
| 193 |
"step": 250
|
| 194 |
},
|
| 195 |
{
|
| 196 |
-
"epoch":
|
| 197 |
-
"
|
| 198 |
-
"
|
| 199 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 200 |
"step": 260
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch":
|
| 204 |
-
"grad_norm":
|
| 205 |
-
"learning_rate": 1.
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 270
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch":
|
| 211 |
-
"
|
| 212 |
-
"
|
| 213 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 214 |
"step": 280
|
| 215 |
},
|
| 216 |
{
|
| 217 |
-
"epoch":
|
| 218 |
-
"grad_norm": 2.
|
| 219 |
-
"learning_rate": 1.
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 290
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch":
|
| 225 |
-
"
|
| 226 |
-
"
|
| 227 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 228 |
"step": 300
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch":
|
| 232 |
-
"grad_norm":
|
| 233 |
-
"learning_rate": 1.
|
| 234 |
-
"loss": 0.
|
| 235 |
"step": 310
|
| 236 |
},
|
| 237 |
{
|
| 238 |
-
"epoch":
|
| 239 |
-
"grad_norm":
|
| 240 |
-
"learning_rate": 1.
|
| 241 |
-
"loss": 0.
|
| 242 |
"step": 320
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"
|
| 247 |
-
"
|
| 248 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 249 |
"step": 330
|
| 250 |
},
|
| 251 |
{
|
| 252 |
-
"epoch":
|
| 253 |
-
"grad_norm":
|
| 254 |
-
"learning_rate": 1.
|
| 255 |
-
"loss": 0.
|
| 256 |
"step": 340
|
| 257 |
},
|
| 258 |
{
|
| 259 |
-
"epoch":
|
| 260 |
-
"
|
| 261 |
-
"
|
| 262 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 263 |
"step": 350
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"grad_norm": 2.
|
| 268 |
-
"learning_rate": 1.
|
| 269 |
-
"loss": 0.
|
| 270 |
"step": 360
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch":
|
| 274 |
-
"eval_accuracy": 0.
|
| 275 |
-
"eval_loss": 0.
|
| 276 |
-
"eval_runtime":
|
| 277 |
-
"eval_samples_per_second":
|
| 278 |
-
"eval_steps_per_second":
|
| 279 |
-
"step":
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch":
|
| 283 |
-
"grad_norm":
|
| 284 |
-
"learning_rate": 1.
|
| 285 |
-
"loss": 0.
|
| 286 |
"step": 370
|
| 287 |
},
|
| 288 |
{
|
| 289 |
-
"epoch":
|
| 290 |
-
"grad_norm": 2.
|
| 291 |
-
"learning_rate": 1.
|
| 292 |
-
"loss": 0.
|
| 293 |
"step": 380
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch":
|
| 297 |
-
"grad_norm":
|
| 298 |
-
"learning_rate": 1.
|
| 299 |
-
"loss": 0.
|
| 300 |
"step": 390
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"epoch":
|
| 304 |
-
"
|
| 305 |
-
"
|
| 306 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 307 |
"step": 400
|
| 308 |
},
|
| 309 |
{
|
| 310 |
-
"epoch":
|
| 311 |
-
"grad_norm": 2.
|
| 312 |
-
"learning_rate": 1.
|
| 313 |
-
"loss": 0.
|
| 314 |
"step": 410
|
| 315 |
},
|
| 316 |
{
|
| 317 |
-
"epoch":
|
| 318 |
-
"
|
| 319 |
-
"
|
| 320 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 321 |
"step": 420
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch":
|
| 325 |
-
"grad_norm":
|
| 326 |
-
"learning_rate": 1.
|
| 327 |
-
"loss": 0.
|
| 328 |
"step": 430
|
| 329 |
},
|
| 330 |
{
|
| 331 |
-
"epoch":
|
| 332 |
-
"
|
| 333 |
-
"
|
| 334 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 335 |
"step": 440
|
| 336 |
},
|
| 337 |
{
|
| 338 |
-
"epoch":
|
| 339 |
-
"grad_norm":
|
| 340 |
-
"learning_rate":
|
| 341 |
-
"loss": 0.
|
| 342 |
"step": 450
|
| 343 |
},
|
| 344 |
{
|
| 345 |
-
"epoch":
|
| 346 |
-
"grad_norm":
|
| 347 |
-
"learning_rate":
|
| 348 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 349 |
"step": 460
|
| 350 |
},
|
| 351 |
{
|
| 352 |
-
"epoch":
|
| 353 |
-
"grad_norm":
|
| 354 |
-
"learning_rate":
|
| 355 |
-
"loss": 0.
|
| 356 |
"step": 470
|
| 357 |
},
|
| 358 |
{
|
| 359 |
-
"epoch":
|
| 360 |
-
"grad_norm": 2.
|
| 361 |
-
"learning_rate":
|
| 362 |
-
"loss": 0.
|
| 363 |
"step": 480
|
| 364 |
},
|
| 365 |
{
|
| 366 |
-
"epoch":
|
| 367 |
-
"
|
| 368 |
-
"
|
| 369 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 370 |
"step": 490
|
| 371 |
},
|
| 372 |
{
|
| 373 |
-
"epoch":
|
| 374 |
-
"grad_norm": 3.
|
| 375 |
-
"learning_rate":
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 500
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch":
|
| 381 |
-
"
|
| 382 |
-
"
|
| 383 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 384 |
"step": 510
|
| 385 |
},
|
| 386 |
{
|
| 387 |
-
"epoch":
|
| 388 |
-
"grad_norm":
|
| 389 |
-
"learning_rate":
|
| 390 |
-
"loss": 0.
|
| 391 |
"step": 520
|
| 392 |
},
|
| 393 |
{
|
| 394 |
-
"epoch":
|
| 395 |
-
"
|
| 396 |
-
"
|
| 397 |
-
"
|
| 398 |
-
"
|
|
|
|
|
|
|
| 399 |
},
|
| 400 |
{
|
| 401 |
-
"epoch":
|
| 402 |
-
"grad_norm":
|
| 403 |
-
"learning_rate":
|
| 404 |
-
"loss": 0.
|
| 405 |
-
"step":
|
| 406 |
},
|
| 407 |
{
|
| 408 |
-
"epoch":
|
| 409 |
-
"
|
| 410 |
-
"
|
| 411 |
-
"
|
| 412 |
-
"eval_samples_per_second": 48.94,
|
| 413 |
-
"eval_steps_per_second": 6.166,
|
| 414 |
"step": 540
|
| 415 |
},
|
| 416 |
{
|
| 417 |
-
"epoch":
|
| 418 |
-
"grad_norm":
|
| 419 |
-
"learning_rate":
|
| 420 |
-
"loss": 0.
|
| 421 |
"step": 550
|
| 422 |
},
|
| 423 |
{
|
| 424 |
-
"epoch":
|
| 425 |
-
"
|
| 426 |
-
"
|
| 427 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 428 |
"step": 560
|
| 429 |
},
|
| 430 |
{
|
| 431 |
-
"epoch":
|
| 432 |
-
"grad_norm":
|
| 433 |
-
"learning_rate":
|
| 434 |
-
"loss": 0.
|
| 435 |
"step": 570
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch":
|
| 439 |
-
"
|
| 440 |
-
"
|
| 441 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 442 |
"step": 580
|
| 443 |
},
|
| 444 |
{
|
| 445 |
-
"epoch":
|
| 446 |
-
"grad_norm":
|
| 447 |
-
"learning_rate":
|
| 448 |
-
"loss": 0.
|
| 449 |
"step": 590
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch":
|
| 453 |
-
"
|
| 454 |
-
"
|
| 455 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 456 |
"step": 600
|
| 457 |
},
|
| 458 |
{
|
| 459 |
-
"epoch":
|
| 460 |
-
"grad_norm": 5.
|
| 461 |
-
"learning_rate":
|
| 462 |
-
"loss": 0.
|
| 463 |
"step": 610
|
| 464 |
},
|
| 465 |
{
|
| 466 |
-
"epoch":
|
| 467 |
-
"grad_norm":
|
| 468 |
-
"learning_rate":
|
| 469 |
-
"loss": 0.
|
| 470 |
"step": 620
|
| 471 |
},
|
| 472 |
{
|
| 473 |
-
"epoch":
|
| 474 |
-
"
|
| 475 |
-
"
|
| 476 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 477 |
"step": 630
|
| 478 |
},
|
| 479 |
{
|
| 480 |
-
"epoch":
|
| 481 |
-
"grad_norm":
|
| 482 |
-
"learning_rate":
|
| 483 |
-
"loss": 0.
|
| 484 |
"step": 640
|
| 485 |
},
|
| 486 |
{
|
| 487 |
-
"epoch":
|
| 488 |
-
"
|
| 489 |
-
"
|
| 490 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 491 |
"step": 650
|
| 492 |
},
|
| 493 |
{
|
| 494 |
-
"epoch":
|
| 495 |
-
"grad_norm":
|
| 496 |
-
"learning_rate":
|
| 497 |
-
"loss": 0.
|
| 498 |
"step": 660
|
| 499 |
},
|
| 500 |
{
|
| 501 |
-
"epoch":
|
| 502 |
-
"
|
| 503 |
-
"
|
| 504 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 505 |
"step": 670
|
| 506 |
},
|
| 507 |
{
|
| 508 |
-
"epoch":
|
| 509 |
-
"grad_norm":
|
| 510 |
-
"learning_rate":
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 680
|
| 513 |
},
|
| 514 |
{
|
| 515 |
-
"epoch":
|
| 516 |
-
"grad_norm":
|
| 517 |
-
"learning_rate":
|
| 518 |
-
"loss": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 519 |
"step": 690
|
| 520 |
},
|
| 521 |
{
|
| 522 |
-
"epoch":
|
| 523 |
-
"grad_norm": 2.
|
| 524 |
-
"learning_rate":
|
| 525 |
-
"loss": 0.
|
| 526 |
"step": 700
|
| 527 |
},
|
| 528 |
{
|
| 529 |
-
"epoch":
|
| 530 |
-
"grad_norm":
|
| 531 |
-
"learning_rate":
|
| 532 |
-
"loss": 0.
|
| 533 |
"step": 710
|
| 534 |
},
|
| 535 |
{
|
| 536 |
-
"epoch":
|
| 537 |
-
"
|
| 538 |
-
"
|
| 539 |
-
"
|
| 540 |
-
"
|
|
|
|
|
|
|
| 541 |
},
|
| 542 |
{
|
| 543 |
-
"epoch":
|
| 544 |
-
"
|
| 545 |
-
"
|
| 546 |
-
"
|
| 547 |
-
"eval_samples_per_second": 51.774,
|
| 548 |
-
"eval_steps_per_second": 6.523,
|
| 549 |
"step": 720
|
| 550 |
},
|
| 551 |
{
|
| 552 |
-
"epoch":
|
| 553 |
-
"grad_norm":
|
| 554 |
-
"learning_rate":
|
| 555 |
-
"loss": 0.
|
| 556 |
"step": 730
|
| 557 |
},
|
| 558 |
{
|
| 559 |
-
"epoch":
|
| 560 |
-
"
|
| 561 |
-
"
|
| 562 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 563 |
"step": 740
|
| 564 |
},
|
| 565 |
{
|
| 566 |
-
"epoch":
|
| 567 |
-
"grad_norm":
|
| 568 |
-
"learning_rate":
|
| 569 |
-
"loss": 0.
|
| 570 |
"step": 750
|
| 571 |
},
|
| 572 |
{
|
| 573 |
-
"epoch":
|
| 574 |
-
"
|
| 575 |
-
"
|
| 576 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 577 |
"step": 760
|
| 578 |
},
|
| 579 |
{
|
| 580 |
-
"epoch":
|
| 581 |
-
"grad_norm": 2.
|
| 582 |
-
"learning_rate":
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 770
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch":
|
| 588 |
-
"grad_norm": 2.
|
| 589 |
-
"learning_rate":
|
| 590 |
-
"loss": 0.
|
| 591 |
"step": 780
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch":
|
| 595 |
-
"
|
| 596 |
-
"
|
| 597 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 598 |
"step": 790
|
| 599 |
},
|
| 600 |
{
|
| 601 |
-
"epoch":
|
| 602 |
-
"grad_norm": 4.
|
| 603 |
-
"learning_rate":
|
| 604 |
-
"loss": 0.
|
| 605 |
"step": 800
|
| 606 |
},
|
| 607 |
{
|
| 608 |
-
"epoch":
|
| 609 |
-
"
|
| 610 |
-
"
|
| 611 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 612 |
"step": 810
|
| 613 |
},
|
| 614 |
{
|
| 615 |
-
"epoch":
|
| 616 |
-
"grad_norm":
|
| 617 |
-
"learning_rate":
|
| 618 |
-
"loss": 0.
|
| 619 |
"step": 820
|
| 620 |
},
|
| 621 |
{
|
| 622 |
-
"epoch":
|
| 623 |
-
"
|
| 624 |
-
"
|
| 625 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 626 |
"step": 830
|
| 627 |
},
|
| 628 |
{
|
| 629 |
-
"epoch":
|
| 630 |
-
"grad_norm": 2.
|
| 631 |
-
"learning_rate":
|
| 632 |
-
"loss": 0.
|
| 633 |
"step": 840
|
| 634 |
},
|
| 635 |
{
|
| 636 |
-
"epoch":
|
| 637 |
-
"grad_norm": 2.
|
| 638 |
-
"learning_rate":
|
| 639 |
-
"loss": 0.
|
| 640 |
"step": 850
|
| 641 |
},
|
| 642 |
{
|
| 643 |
-
"epoch":
|
| 644 |
-
"
|
| 645 |
-
"
|
| 646 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 647 |
"step": 860
|
| 648 |
},
|
| 649 |
{
|
| 650 |
-
"epoch":
|
| 651 |
-
"grad_norm":
|
| 652 |
-
"learning_rate":
|
| 653 |
-
"loss": 0.
|
| 654 |
"step": 870
|
| 655 |
},
|
| 656 |
{
|
| 657 |
-
"epoch":
|
| 658 |
-
"
|
| 659 |
-
"
|
| 660 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 661 |
"step": 880
|
| 662 |
},
|
| 663 |
{
|
| 664 |
-
"epoch":
|
| 665 |
-
"grad_norm": 2.
|
| 666 |
-
"learning_rate":
|
| 667 |
-
"loss": 0.
|
| 668 |
"step": 890
|
| 669 |
},
|
| 670 |
{
|
| 671 |
-
"epoch":
|
| 672 |
-
"
|
| 673 |
-
"
|
| 674 |
-
"
|
| 675 |
-
"
|
|
|
|
|
|
|
| 676 |
},
|
| 677 |
{
|
| 678 |
-
"epoch":
|
| 679 |
-
"
|
| 680 |
-
"
|
| 681 |
-
"
|
| 682 |
-
"eval_samples_per_second": 45.923,
|
| 683 |
-
"eval_steps_per_second": 5.786,
|
| 684 |
"step": 900
|
| 685 |
},
|
| 686 |
{
|
| 687 |
-
"epoch":
|
| 688 |
-
"
|
| 689 |
-
"
|
| 690 |
-
"
|
| 691 |
-
"
|
| 692 |
-
|
| 693 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 694 |
}
|
| 695 |
],
|
| 696 |
"logging_steps": 10,
|
| 697 |
-
"max_steps":
|
| 698 |
"num_input_tokens_seen": 0,
|
| 699 |
-
"num_train_epochs":
|
| 700 |
"save_steps": 500,
|
| 701 |
"stateful_callbacks": {
|
| 702 |
"TrainerControl": {
|
|
@@ -710,8 +1290,8 @@
|
|
| 710 |
"attributes": {}
|
| 711 |
}
|
| 712 |
},
|
| 713 |
-
"total_flos":
|
| 714 |
-
"train_batch_size":
|
| 715 |
"trial_name": null,
|
| 716 |
"trial_params": null
|
| 717 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.4903908967971802,
|
| 3 |
+
"best_model_checkpoint": "./beans_outputs/checkpoint-621",
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1150,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.43478260869565216,
|
| 13 |
+
"grad_norm": 3.492309093475342,
|
| 14 |
+
"learning_rate": 1.9826086956521742e-05,
|
| 15 |
+
"loss": 0.7305,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.8695652173913043,
|
| 20 |
+
"grad_norm": 4.003854751586914,
|
| 21 |
+
"learning_rate": 1.965217391304348e-05,
|
| 22 |
+
"loss": 0.6794,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_accuracy": 0.6062992125984252,
|
| 28 |
+
"eval_loss": 0.6559741497039795,
|
| 29 |
+
"eval_runtime": 4.6669,
|
| 30 |
+
"eval_samples_per_second": 54.425,
|
| 31 |
+
"eval_steps_per_second": 0.857,
|
| 32 |
+
"step": 23
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.3043478260869565,
|
| 36 |
+
"grad_norm": 4.675185680389404,
|
| 37 |
+
"learning_rate": 1.947826086956522e-05,
|
| 38 |
+
"loss": 0.6399,
|
| 39 |
"step": 30
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 1.7391304347826086,
|
| 43 |
+
"grad_norm": 4.350035667419434,
|
| 44 |
+
"learning_rate": 1.9304347826086957e-05,
|
| 45 |
+
"loss": 0.6215,
|
| 46 |
"step": 40
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"epoch": 2.0,
|
| 50 |
+
"eval_accuracy": 0.7362204724409449,
|
| 51 |
+
"eval_loss": 0.5833372473716736,
|
| 52 |
+
"eval_runtime": 3.7107,
|
| 53 |
+
"eval_samples_per_second": 68.451,
|
| 54 |
+
"eval_steps_per_second": 1.078,
|
| 55 |
+
"step": 46
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 2.1739130434782608,
|
| 59 |
+
"grad_norm": 3.180147886276245,
|
| 60 |
+
"learning_rate": 1.9130434782608697e-05,
|
| 61 |
+
"loss": 0.5964,
|
| 62 |
"step": 50
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 2.608695652173913,
|
| 66 |
+
"grad_norm": 3.246190309524536,
|
| 67 |
+
"learning_rate": 1.8956521739130434e-05,
|
| 68 |
+
"loss": 0.5784,
|
| 69 |
"step": 60
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 3.0,
|
| 73 |
+
"eval_accuracy": 0.7598425196850394,
|
| 74 |
+
"eval_loss": 0.5489528179168701,
|
| 75 |
+
"eval_runtime": 4.3517,
|
| 76 |
+
"eval_samples_per_second": 58.367,
|
| 77 |
+
"eval_steps_per_second": 0.919,
|
| 78 |
+
"step": 69
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.0434782608695654,
|
| 82 |
+
"grad_norm": 2.3400914669036865,
|
| 83 |
+
"learning_rate": 1.8782608695652175e-05,
|
| 84 |
+
"loss": 0.5412,
|
| 85 |
"step": 70
|
| 86 |
},
|
| 87 |
{
|
| 88 |
+
"epoch": 3.4782608695652173,
|
| 89 |
+
"grad_norm": 3.5264837741851807,
|
| 90 |
+
"learning_rate": 1.8608695652173912e-05,
|
| 91 |
+
"loss": 0.5659,
|
| 92 |
"step": 80
|
| 93 |
},
|
| 94 |
{
|
| 95 |
+
"epoch": 3.9130434782608696,
|
| 96 |
+
"grad_norm": 4.993140697479248,
|
| 97 |
+
"learning_rate": 1.8434782608695653e-05,
|
| 98 |
+
"loss": 0.5347,
|
| 99 |
"step": 90
|
| 100 |
},
|
| 101 |
{
|
| 102 |
+
"epoch": 4.0,
|
| 103 |
+
"eval_accuracy": 0.7637795275590551,
|
| 104 |
+
"eval_loss": 0.5305963754653931,
|
| 105 |
+
"eval_runtime": 3.9321,
|
| 106 |
+
"eval_samples_per_second": 64.596,
|
| 107 |
+
"eval_steps_per_second": 1.017,
|
| 108 |
+
"step": 92
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 4.3478260869565215,
|
| 112 |
+
"grad_norm": 2.20806622505188,
|
| 113 |
+
"learning_rate": 1.8260869565217393e-05,
|
| 114 |
+
"loss": 0.5086,
|
| 115 |
"step": 100
|
| 116 |
},
|
| 117 |
{
|
| 118 |
+
"epoch": 4.782608695652174,
|
| 119 |
+
"grad_norm": 4.256261825561523,
|
| 120 |
+
"learning_rate": 1.808695652173913e-05,
|
| 121 |
+
"loss": 0.5307,
|
| 122 |
"step": 110
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 5.0,
|
| 126 |
+
"eval_accuracy": 0.7637795275590551,
|
| 127 |
+
"eval_loss": 0.5235078930854797,
|
| 128 |
+
"eval_runtime": 3.7141,
|
| 129 |
+
"eval_samples_per_second": 68.389,
|
| 130 |
+
"eval_steps_per_second": 1.077,
|
| 131 |
+
"step": 115
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.217391304347826,
|
| 135 |
+
"grad_norm": 2.6543545722961426,
|
| 136 |
+
"learning_rate": 1.791304347826087e-05,
|
| 137 |
+
"loss": 0.5085,
|
| 138 |
"step": 120
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 5.6521739130434785,
|
| 142 |
+
"grad_norm": 4.274487495422363,
|
| 143 |
+
"learning_rate": 1.773913043478261e-05,
|
| 144 |
+
"loss": 0.5391,
|
| 145 |
"step": 130
|
| 146 |
},
|
| 147 |
{
|
| 148 |
+
"epoch": 6.0,
|
| 149 |
+
"eval_accuracy": 0.7677165354330708,
|
| 150 |
+
"eval_loss": 0.5090441703796387,
|
| 151 |
+
"eval_runtime": 4.3438,
|
| 152 |
+
"eval_samples_per_second": 58.475,
|
| 153 |
+
"eval_steps_per_second": 0.921,
|
| 154 |
+
"step": 138
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 6.086956521739131,
|
| 158 |
+
"grad_norm": 3.147414445877075,
|
| 159 |
+
"learning_rate": 1.756521739130435e-05,
|
| 160 |
+
"loss": 0.4977,
|
| 161 |
"step": 140
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 6.521739130434782,
|
| 165 |
+
"grad_norm": 4.254673004150391,
|
| 166 |
+
"learning_rate": 1.739130434782609e-05,
|
| 167 |
+
"loss": 0.5297,
|
| 168 |
"step": 150
|
| 169 |
},
|
| 170 |
{
|
| 171 |
+
"epoch": 6.956521739130435,
|
| 172 |
+
"grad_norm": 2.083784818649292,
|
| 173 |
+
"learning_rate": 1.721739130434783e-05,
|
| 174 |
+
"loss": 0.48,
|
| 175 |
"step": 160
|
| 176 |
},
|
| 177 |
{
|
| 178 |
+
"epoch": 7.0,
|
| 179 |
+
"eval_accuracy": 0.7716535433070866,
|
| 180 |
+
"eval_loss": 0.5108471512794495,
|
| 181 |
+
"eval_runtime": 4.1877,
|
| 182 |
+
"eval_samples_per_second": 60.653,
|
| 183 |
+
"eval_steps_per_second": 0.955,
|
| 184 |
+
"step": 161
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 7.391304347826087,
|
| 188 |
+
"grad_norm": 4.193545818328857,
|
| 189 |
+
"learning_rate": 1.7043478260869566e-05,
|
| 190 |
+
"loss": 0.4826,
|
| 191 |
"step": 170
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 7.826086956521739,
|
| 195 |
+
"grad_norm": 2.05076003074646,
|
| 196 |
+
"learning_rate": 1.6869565217391307e-05,
|
| 197 |
+
"loss": 0.473,
|
| 198 |
"step": 180
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 8.0,
|
| 202 |
+
"eval_accuracy": 0.7755905511811023,
|
| 203 |
+
"eval_loss": 0.5028324127197266,
|
| 204 |
+
"eval_runtime": 3.7021,
|
| 205 |
+
"eval_samples_per_second": 68.61,
|
| 206 |
+
"eval_steps_per_second": 1.08,
|
| 207 |
+
"step": 184
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 8.26086956521739,
|
| 211 |
+
"grad_norm": 3.007233142852783,
|
| 212 |
+
"learning_rate": 1.6695652173913044e-05,
|
| 213 |
+
"loss": 0.5255,
|
| 214 |
"step": 190
|
| 215 |
},
|
| 216 |
{
|
| 217 |
+
"epoch": 8.695652173913043,
|
| 218 |
+
"grad_norm": 2.196945905685425,
|
| 219 |
+
"learning_rate": 1.6521739130434785e-05,
|
| 220 |
+
"loss": 0.5014,
|
| 221 |
"step": 200
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 9.0,
|
| 225 |
+
"eval_accuracy": 0.7716535433070866,
|
| 226 |
+
"eval_loss": 0.5054498314857483,
|
| 227 |
+
"eval_runtime": 4.225,
|
| 228 |
+
"eval_samples_per_second": 60.119,
|
| 229 |
+
"eval_steps_per_second": 0.947,
|
| 230 |
+
"step": 207
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 9.130434782608695,
|
| 234 |
+
"grad_norm": 2.184353828430176,
|
| 235 |
+
"learning_rate": 1.6347826086956525e-05,
|
| 236 |
+
"loss": 0.5044,
|
| 237 |
"step": 210
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 9.565217391304348,
|
| 241 |
+
"grad_norm": 4.106619358062744,
|
| 242 |
+
"learning_rate": 1.6173913043478262e-05,
|
| 243 |
+
"loss": 0.4822,
|
| 244 |
"step": 220
|
| 245 |
},
|
| 246 |
{
|
| 247 |
+
"epoch": 10.0,
|
| 248 |
+
"grad_norm": 4.000082969665527,
|
| 249 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 250 |
+
"loss": 0.496,
|
| 251 |
"step": 230
|
| 252 |
},
|
| 253 |
{
|
| 254 |
+
"epoch": 10.0,
|
| 255 |
+
"eval_accuracy": 0.7716535433070866,
|
| 256 |
+
"eval_loss": 0.5039955973625183,
|
| 257 |
+
"eval_runtime": 5.3498,
|
| 258 |
+
"eval_samples_per_second": 47.478,
|
| 259 |
+
"eval_steps_per_second": 0.748,
|
| 260 |
+
"step": 230
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 10.434782608695652,
|
| 264 |
+
"grad_norm": 5.726933002471924,
|
| 265 |
+
"learning_rate": 1.582608695652174e-05,
|
| 266 |
+
"loss": 0.5101,
|
| 267 |
"step": 240
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 10.869565217391305,
|
| 271 |
+
"grad_norm": 4.100568771362305,
|
| 272 |
+
"learning_rate": 1.565217391304348e-05,
|
| 273 |
+
"loss": 0.4688,
|
| 274 |
"step": 250
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 11.0,
|
| 278 |
+
"eval_accuracy": 0.7677165354330708,
|
| 279 |
+
"eval_loss": 0.4972316324710846,
|
| 280 |
+
"eval_runtime": 3.7607,
|
| 281 |
+
"eval_samples_per_second": 67.54,
|
| 282 |
+
"eval_steps_per_second": 1.064,
|
| 283 |
+
"step": 253
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 11.304347826086957,
|
| 287 |
+
"grad_norm": 2.6119587421417236,
|
| 288 |
+
"learning_rate": 1.5478260869565217e-05,
|
| 289 |
+
"loss": 0.485,
|
| 290 |
"step": 260
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"epoch": 11.73913043478261,
|
| 294 |
+
"grad_norm": 3.003861427307129,
|
| 295 |
+
"learning_rate": 1.5304347826086958e-05,
|
| 296 |
+
"loss": 0.4943,
|
| 297 |
"step": 270
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 12.0,
|
| 301 |
+
"eval_accuracy": 0.7637795275590551,
|
| 302 |
+
"eval_loss": 0.49771231412887573,
|
| 303 |
+
"eval_runtime": 4.9203,
|
| 304 |
+
"eval_samples_per_second": 51.622,
|
| 305 |
+
"eval_steps_per_second": 0.813,
|
| 306 |
+
"step": 276
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 12.173913043478262,
|
| 310 |
+
"grad_norm": 2.9490270614624023,
|
| 311 |
+
"learning_rate": 1.5130434782608697e-05,
|
| 312 |
+
"loss": 0.4505,
|
| 313 |
"step": 280
|
| 314 |
},
|
| 315 |
{
|
| 316 |
+
"epoch": 12.608695652173914,
|
| 317 |
+
"grad_norm": 2.8131847381591797,
|
| 318 |
+
"learning_rate": 1.4956521739130436e-05,
|
| 319 |
+
"loss": 0.5012,
|
| 320 |
"step": 290
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"epoch": 13.0,
|
| 324 |
+
"eval_accuracy": 0.7716535433070866,
|
| 325 |
+
"eval_loss": 0.5057242512702942,
|
| 326 |
+
"eval_runtime": 3.7024,
|
| 327 |
+
"eval_samples_per_second": 68.605,
|
| 328 |
+
"eval_steps_per_second": 1.08,
|
| 329 |
+
"step": 299
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 13.043478260869565,
|
| 333 |
+
"grad_norm": 17.65978240966797,
|
| 334 |
+
"learning_rate": 1.4782608695652174e-05,
|
| 335 |
+
"loss": 0.4768,
|
| 336 |
"step": 300
|
| 337 |
},
|
| 338 |
{
|
| 339 |
+
"epoch": 13.478260869565217,
|
| 340 |
+
"grad_norm": 2.085587978363037,
|
| 341 |
+
"learning_rate": 1.4608695652173915e-05,
|
| 342 |
+
"loss": 0.4729,
|
| 343 |
"step": 310
|
| 344 |
},
|
| 345 |
{
|
| 346 |
+
"epoch": 13.91304347826087,
|
| 347 |
+
"grad_norm": 4.59744119644165,
|
| 348 |
+
"learning_rate": 1.4434782608695654e-05,
|
| 349 |
+
"loss": 0.4639,
|
| 350 |
"step": 320
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"epoch": 14.0,
|
| 354 |
+
"eval_accuracy": 0.7716535433070866,
|
| 355 |
+
"eval_loss": 0.5010089874267578,
|
| 356 |
+
"eval_runtime": 3.7018,
|
| 357 |
+
"eval_samples_per_second": 68.616,
|
| 358 |
+
"eval_steps_per_second": 1.081,
|
| 359 |
+
"step": 322
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 14.347826086956522,
|
| 363 |
+
"grad_norm": 2.4057395458221436,
|
| 364 |
+
"learning_rate": 1.4260869565217392e-05,
|
| 365 |
+
"loss": 0.4751,
|
| 366 |
"step": 330
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 14.782608695652174,
|
| 370 |
+
"grad_norm": 3.549567222595215,
|
| 371 |
+
"learning_rate": 1.4086956521739133e-05,
|
| 372 |
+
"loss": 0.4709,
|
| 373 |
"step": 340
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 15.0,
|
| 377 |
+
"eval_accuracy": 0.7795275590551181,
|
| 378 |
+
"eval_loss": 0.4948899447917938,
|
| 379 |
+
"eval_runtime": 4.9714,
|
| 380 |
+
"eval_samples_per_second": 51.092,
|
| 381 |
+
"eval_steps_per_second": 0.805,
|
| 382 |
+
"step": 345
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 15.217391304347826,
|
| 386 |
+
"grad_norm": 6.705427646636963,
|
| 387 |
+
"learning_rate": 1.391304347826087e-05,
|
| 388 |
+
"loss": 0.4379,
|
| 389 |
"step": 350
|
| 390 |
},
|
| 391 |
{
|
| 392 |
+
"epoch": 15.652173913043478,
|
| 393 |
+
"grad_norm": 2.444533348083496,
|
| 394 |
+
"learning_rate": 1.373913043478261e-05,
|
| 395 |
+
"loss": 0.4888,
|
| 396 |
"step": 360
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 16.0,
|
| 400 |
+
"eval_accuracy": 0.7834645669291339,
|
| 401 |
+
"eval_loss": 0.49550917744636536,
|
| 402 |
+
"eval_runtime": 3.6768,
|
| 403 |
+
"eval_samples_per_second": 69.081,
|
| 404 |
+
"eval_steps_per_second": 1.088,
|
| 405 |
+
"step": 368
|
| 406 |
},
|
| 407 |
{
|
| 408 |
+
"epoch": 16.08695652173913,
|
| 409 |
+
"grad_norm": 5.470461845397949,
|
| 410 |
+
"learning_rate": 1.3565217391304348e-05,
|
| 411 |
+
"loss": 0.4952,
|
| 412 |
"step": 370
|
| 413 |
},
|
| 414 |
{
|
| 415 |
+
"epoch": 16.52173913043478,
|
| 416 |
+
"grad_norm": 2.0678608417510986,
|
| 417 |
+
"learning_rate": 1.3391304347826088e-05,
|
| 418 |
+
"loss": 0.4784,
|
| 419 |
"step": 380
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 16.956521739130434,
|
| 423 |
+
"grad_norm": 6.63480806350708,
|
| 424 |
+
"learning_rate": 1.3217391304347827e-05,
|
| 425 |
+
"loss": 0.4594,
|
| 426 |
"step": 390
|
| 427 |
},
|
| 428 |
{
|
| 429 |
+
"epoch": 17.0,
|
| 430 |
+
"eval_accuracy": 0.7716535433070866,
|
| 431 |
+
"eval_loss": 0.49856194853782654,
|
| 432 |
+
"eval_runtime": 3.7219,
|
| 433 |
+
"eval_samples_per_second": 68.245,
|
| 434 |
+
"eval_steps_per_second": 1.075,
|
| 435 |
+
"step": 391
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 17.391304347826086,
|
| 439 |
+
"grad_norm": 4.448991298675537,
|
| 440 |
+
"learning_rate": 1.3043478260869566e-05,
|
| 441 |
+
"loss": 0.4607,
|
| 442 |
"step": 400
|
| 443 |
},
|
| 444 |
{
|
| 445 |
+
"epoch": 17.82608695652174,
|
| 446 |
+
"grad_norm": 2.716780424118042,
|
| 447 |
+
"learning_rate": 1.2869565217391305e-05,
|
| 448 |
+
"loss": 0.4745,
|
| 449 |
"step": 410
|
| 450 |
},
|
| 451 |
{
|
| 452 |
+
"epoch": 18.0,
|
| 453 |
+
"eval_accuracy": 0.7677165354330708,
|
| 454 |
+
"eval_loss": 0.501070499420166,
|
| 455 |
+
"eval_runtime": 4.5054,
|
| 456 |
+
"eval_samples_per_second": 56.377,
|
| 457 |
+
"eval_steps_per_second": 0.888,
|
| 458 |
+
"step": 414
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 18.26086956521739,
|
| 462 |
+
"grad_norm": 2.406355857849121,
|
| 463 |
+
"learning_rate": 1.2695652173913045e-05,
|
| 464 |
+
"loss": 0.4639,
|
| 465 |
"step": 420
|
| 466 |
},
|
| 467 |
{
|
| 468 |
+
"epoch": 18.695652173913043,
|
| 469 |
+
"grad_norm": 5.627669811248779,
|
| 470 |
+
"learning_rate": 1.2521739130434784e-05,
|
| 471 |
+
"loss": 0.4667,
|
| 472 |
"step": 430
|
| 473 |
},
|
| 474 |
{
|
| 475 |
+
"epoch": 19.0,
|
| 476 |
+
"eval_accuracy": 0.7755905511811023,
|
| 477 |
+
"eval_loss": 0.4928434491157532,
|
| 478 |
+
"eval_runtime": 4.0475,
|
| 479 |
+
"eval_samples_per_second": 62.756,
|
| 480 |
+
"eval_steps_per_second": 0.988,
|
| 481 |
+
"step": 437
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 19.130434782608695,
|
| 485 |
+
"grad_norm": 4.074652671813965,
|
| 486 |
+
"learning_rate": 1.2347826086956523e-05,
|
| 487 |
+
"loss": 0.4671,
|
| 488 |
"step": 440
|
| 489 |
},
|
| 490 |
{
|
| 491 |
+
"epoch": 19.565217391304348,
|
| 492 |
+
"grad_norm": 5.88148832321167,
|
| 493 |
+
"learning_rate": 1.2173913043478263e-05,
|
| 494 |
+
"loss": 0.4442,
|
| 495 |
"step": 450
|
| 496 |
},
|
| 497 |
{
|
| 498 |
+
"epoch": 20.0,
|
| 499 |
+
"grad_norm": 3.00347900390625,
|
| 500 |
+
"learning_rate": 1.2e-05,
|
| 501 |
+
"loss": 0.4551,
|
| 502 |
+
"step": 460
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 20.0,
|
| 506 |
+
"eval_accuracy": 0.7795275590551181,
|
| 507 |
+
"eval_loss": 0.5055357217788696,
|
| 508 |
+
"eval_runtime": 3.6885,
|
| 509 |
+
"eval_samples_per_second": 68.862,
|
| 510 |
+
"eval_steps_per_second": 1.084,
|
| 511 |
"step": 460
|
| 512 |
},
|
| 513 |
{
|
| 514 |
+
"epoch": 20.434782608695652,
|
| 515 |
+
"grad_norm": 10.164237976074219,
|
| 516 |
+
"learning_rate": 1.182608695652174e-05,
|
| 517 |
+
"loss": 0.4657,
|
| 518 |
"step": 470
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 20.869565217391305,
|
| 522 |
+
"grad_norm": 2.1962711811065674,
|
| 523 |
+
"learning_rate": 1.1652173913043478e-05,
|
| 524 |
+
"loss": 0.4657,
|
| 525 |
"step": 480
|
| 526 |
},
|
| 527 |
{
|
| 528 |
+
"epoch": 21.0,
|
| 529 |
+
"eval_accuracy": 0.7755905511811023,
|
| 530 |
+
"eval_loss": 0.4928124248981476,
|
| 531 |
+
"eval_runtime": 4.4478,
|
| 532 |
+
"eval_samples_per_second": 57.107,
|
| 533 |
+
"eval_steps_per_second": 0.899,
|
| 534 |
+
"step": 483
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 21.304347826086957,
|
| 538 |
+
"grad_norm": 5.0302228927612305,
|
| 539 |
+
"learning_rate": 1.1478260869565218e-05,
|
| 540 |
+
"loss": 0.4564,
|
| 541 |
"step": 490
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 21.73913043478261,
|
| 545 |
+
"grad_norm": 3.5275819301605225,
|
| 546 |
+
"learning_rate": 1.1304347826086957e-05,
|
| 547 |
+
"loss": 0.4818,
|
| 548 |
"step": 500
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 22.0,
|
| 552 |
+
"eval_accuracy": 0.7755905511811023,
|
| 553 |
+
"eval_loss": 0.5001721978187561,
|
| 554 |
+
"eval_runtime": 4.0355,
|
| 555 |
+
"eval_samples_per_second": 62.942,
|
| 556 |
+
"eval_steps_per_second": 0.991,
|
| 557 |
+
"step": 506
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 22.17391304347826,
|
| 561 |
+
"grad_norm": 6.920666694641113,
|
| 562 |
+
"learning_rate": 1.1130434782608696e-05,
|
| 563 |
+
"loss": 0.4608,
|
| 564 |
"step": 510
|
| 565 |
},
|
| 566 |
{
|
| 567 |
+
"epoch": 22.608695652173914,
|
| 568 |
+
"grad_norm": 2.2840707302093506,
|
| 569 |
+
"learning_rate": 1.0956521739130435e-05,
|
| 570 |
+
"loss": 0.4633,
|
| 571 |
"step": 520
|
| 572 |
},
|
| 573 |
{
|
| 574 |
+
"epoch": 23.0,
|
| 575 |
+
"eval_accuracy": 0.7834645669291339,
|
| 576 |
+
"eval_loss": 0.49459317326545715,
|
| 577 |
+
"eval_runtime": 3.7179,
|
| 578 |
+
"eval_samples_per_second": 68.319,
|
| 579 |
+
"eval_steps_per_second": 1.076,
|
| 580 |
+
"step": 529
|
| 581 |
},
|
| 582 |
{
|
| 583 |
+
"epoch": 23.043478260869566,
|
| 584 |
+
"grad_norm": 6.509201526641846,
|
| 585 |
+
"learning_rate": 1.0782608695652175e-05,
|
| 586 |
+
"loss": 0.4694,
|
| 587 |
+
"step": 530
|
| 588 |
},
|
| 589 |
{
|
| 590 |
+
"epoch": 23.47826086956522,
|
| 591 |
+
"grad_norm": 2.403275489807129,
|
| 592 |
+
"learning_rate": 1.0608695652173914e-05,
|
| 593 |
+
"loss": 0.4874,
|
|
|
|
|
|
|
| 594 |
"step": 540
|
| 595 |
},
|
| 596 |
{
|
| 597 |
+
"epoch": 23.91304347826087,
|
| 598 |
+
"grad_norm": 2.1320598125457764,
|
| 599 |
+
"learning_rate": 1.0434782608695653e-05,
|
| 600 |
+
"loss": 0.4779,
|
| 601 |
"step": 550
|
| 602 |
},
|
| 603 |
{
|
| 604 |
+
"epoch": 24.0,
|
| 605 |
+
"eval_accuracy": 0.7795275590551181,
|
| 606 |
+
"eval_loss": 0.49417200684547424,
|
| 607 |
+
"eval_runtime": 4.3215,
|
| 608 |
+
"eval_samples_per_second": 58.776,
|
| 609 |
+
"eval_steps_per_second": 0.926,
|
| 610 |
+
"step": 552
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 24.347826086956523,
|
| 614 |
+
"grad_norm": 3.7421488761901855,
|
| 615 |
+
"learning_rate": 1.0260869565217393e-05,
|
| 616 |
+
"loss": 0.4579,
|
| 617 |
"step": 560
|
| 618 |
},
|
| 619 |
{
|
| 620 |
+
"epoch": 24.782608695652176,
|
| 621 |
+
"grad_norm": 3.07060170173645,
|
| 622 |
+
"learning_rate": 1.008695652173913e-05,
|
| 623 |
+
"loss": 0.4718,
|
| 624 |
"step": 570
|
| 625 |
},
|
| 626 |
{
|
| 627 |
+
"epoch": 25.0,
|
| 628 |
+
"eval_accuracy": 0.7834645669291339,
|
| 629 |
+
"eval_loss": 0.49625155329704285,
|
| 630 |
+
"eval_runtime": 5.5612,
|
| 631 |
+
"eval_samples_per_second": 45.674,
|
| 632 |
+
"eval_steps_per_second": 0.719,
|
| 633 |
+
"step": 575
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 25.217391304347824,
|
| 637 |
+
"grad_norm": 4.446998596191406,
|
| 638 |
+
"learning_rate": 9.913043478260871e-06,
|
| 639 |
+
"loss": 0.443,
|
| 640 |
"step": 580
|
| 641 |
},
|
| 642 |
{
|
| 643 |
+
"epoch": 25.652173913043477,
|
| 644 |
+
"grad_norm": 2.4786624908447266,
|
| 645 |
+
"learning_rate": 9.73913043478261e-06,
|
| 646 |
+
"loss": 0.4511,
|
| 647 |
"step": 590
|
| 648 |
},
|
| 649 |
{
|
| 650 |
+
"epoch": 26.0,
|
| 651 |
+
"eval_accuracy": 0.7716535433070866,
|
| 652 |
+
"eval_loss": 0.5011107325553894,
|
| 653 |
+
"eval_runtime": 3.7637,
|
| 654 |
+
"eval_samples_per_second": 67.487,
|
| 655 |
+
"eval_steps_per_second": 1.063,
|
| 656 |
+
"step": 598
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 26.08695652173913,
|
| 660 |
+
"grad_norm": 5.552999019622803,
|
| 661 |
+
"learning_rate": 9.565217391304349e-06,
|
| 662 |
+
"loss": 0.4631,
|
| 663 |
"step": 600
|
| 664 |
},
|
| 665 |
{
|
| 666 |
+
"epoch": 26.52173913043478,
|
| 667 |
+
"grad_norm": 5.050811290740967,
|
| 668 |
+
"learning_rate": 9.391304347826087e-06,
|
| 669 |
+
"loss": 0.4564,
|
| 670 |
"step": 610
|
| 671 |
},
|
| 672 |
{
|
| 673 |
+
"epoch": 26.956521739130434,
|
| 674 |
+
"grad_norm": 2.0711512565612793,
|
| 675 |
+
"learning_rate": 9.217391304347826e-06,
|
| 676 |
+
"loss": 0.4798,
|
| 677 |
"step": 620
|
| 678 |
},
|
| 679 |
{
|
| 680 |
+
"epoch": 27.0,
|
| 681 |
+
"eval_accuracy": 0.7874015748031497,
|
| 682 |
+
"eval_loss": 0.4903908967971802,
|
| 683 |
+
"eval_runtime": 4.9056,
|
| 684 |
+
"eval_samples_per_second": 51.777,
|
| 685 |
+
"eval_steps_per_second": 0.815,
|
| 686 |
+
"step": 621
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 27.391304347826086,
|
| 690 |
+
"grad_norm": 4.117509365081787,
|
| 691 |
+
"learning_rate": 9.043478260869565e-06,
|
| 692 |
+
"loss": 0.4411,
|
| 693 |
"step": 630
|
| 694 |
},
|
| 695 |
{
|
| 696 |
+
"epoch": 27.82608695652174,
|
| 697 |
+
"grad_norm": 4.448685646057129,
|
| 698 |
+
"learning_rate": 8.869565217391306e-06,
|
| 699 |
+
"loss": 0.4868,
|
| 700 |
"step": 640
|
| 701 |
},
|
| 702 |
{
|
| 703 |
+
"epoch": 28.0,
|
| 704 |
+
"eval_accuracy": 0.7834645669291339,
|
| 705 |
+
"eval_loss": 0.4982087016105652,
|
| 706 |
+
"eval_runtime": 3.7322,
|
| 707 |
+
"eval_samples_per_second": 68.057,
|
| 708 |
+
"eval_steps_per_second": 1.072,
|
| 709 |
+
"step": 644
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 28.26086956521739,
|
| 713 |
+
"grad_norm": 3.0993807315826416,
|
| 714 |
+
"learning_rate": 8.695652173913044e-06,
|
| 715 |
+
"loss": 0.4414,
|
| 716 |
"step": 650
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 28.695652173913043,
|
| 720 |
+
"grad_norm": 4.982347011566162,
|
| 721 |
+
"learning_rate": 8.521739130434783e-06,
|
| 722 |
+
"loss": 0.4653,
|
| 723 |
"step": 660
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 29.0,
|
| 727 |
+
"eval_accuracy": 0.7874015748031497,
|
| 728 |
+
"eval_loss": 0.498798668384552,
|
| 729 |
+
"eval_runtime": 3.7347,
|
| 730 |
+
"eval_samples_per_second": 68.012,
|
| 731 |
+
"eval_steps_per_second": 1.071,
|
| 732 |
+
"step": 667
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 29.130434782608695,
|
| 736 |
+
"grad_norm": 3.081833600997925,
|
| 737 |
+
"learning_rate": 8.347826086956522e-06,
|
| 738 |
+
"loss": 0.4503,
|
| 739 |
"step": 670
|
| 740 |
},
|
| 741 |
{
|
| 742 |
+
"epoch": 29.565217391304348,
|
| 743 |
+
"grad_norm": 4.352429389953613,
|
| 744 |
+
"learning_rate": 8.173913043478263e-06,
|
| 745 |
+
"loss": 0.4674,
|
| 746 |
"step": 680
|
| 747 |
},
|
| 748 |
{
|
| 749 |
+
"epoch": 30.0,
|
| 750 |
+
"grad_norm": 5.281393051147461,
|
| 751 |
+
"learning_rate": 8.000000000000001e-06,
|
| 752 |
+
"loss": 0.4613,
|
| 753 |
+
"step": 690
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 30.0,
|
| 757 |
+
"eval_accuracy": 0.7795275590551181,
|
| 758 |
+
"eval_loss": 0.49851593375205994,
|
| 759 |
+
"eval_runtime": 4.8766,
|
| 760 |
+
"eval_samples_per_second": 52.085,
|
| 761 |
+
"eval_steps_per_second": 0.82,
|
| 762 |
"step": 690
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 30.434782608695652,
|
| 766 |
+
"grad_norm": 2.2079997062683105,
|
| 767 |
+
"learning_rate": 7.82608695652174e-06,
|
| 768 |
+
"loss": 0.4574,
|
| 769 |
"step": 700
|
| 770 |
},
|
| 771 |
{
|
| 772 |
+
"epoch": 30.869565217391305,
|
| 773 |
+
"grad_norm": 4.6935858726501465,
|
| 774 |
+
"learning_rate": 7.652173913043479e-06,
|
| 775 |
+
"loss": 0.4675,
|
| 776 |
"step": 710
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"epoch": 31.0,
|
| 780 |
+
"eval_accuracy": 0.7716535433070866,
|
| 781 |
+
"eval_loss": 0.5060083270072937,
|
| 782 |
+
"eval_runtime": 3.7305,
|
| 783 |
+
"eval_samples_per_second": 68.087,
|
| 784 |
+
"eval_steps_per_second": 1.072,
|
| 785 |
+
"step": 713
|
| 786 |
},
|
| 787 |
{
|
| 788 |
+
"epoch": 31.304347826086957,
|
| 789 |
+
"grad_norm": 4.8790602684021,
|
| 790 |
+
"learning_rate": 7.478260869565218e-06,
|
| 791 |
+
"loss": 0.4802,
|
|
|
|
|
|
|
| 792 |
"step": 720
|
| 793 |
},
|
| 794 |
{
|
| 795 |
+
"epoch": 31.73913043478261,
|
| 796 |
+
"grad_norm": 5.6365485191345215,
|
| 797 |
+
"learning_rate": 7.304347826086957e-06,
|
| 798 |
+
"loss": 0.4587,
|
| 799 |
"step": 730
|
| 800 |
},
|
| 801 |
{
|
| 802 |
+
"epoch": 32.0,
|
| 803 |
+
"eval_accuracy": 0.7716535433070866,
|
| 804 |
+
"eval_loss": 0.5059147477149963,
|
| 805 |
+
"eval_runtime": 3.7699,
|
| 806 |
+
"eval_samples_per_second": 67.376,
|
| 807 |
+
"eval_steps_per_second": 1.061,
|
| 808 |
+
"step": 736
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 32.17391304347826,
|
| 812 |
+
"grad_norm": 5.480165004730225,
|
| 813 |
+
"learning_rate": 7.130434782608696e-06,
|
| 814 |
+
"loss": 0.4541,
|
| 815 |
"step": 740
|
| 816 |
},
|
| 817 |
{
|
| 818 |
+
"epoch": 32.608695652173914,
|
| 819 |
+
"grad_norm": 2.053098440170288,
|
| 820 |
+
"learning_rate": 6.956521739130435e-06,
|
| 821 |
+
"loss": 0.464,
|
| 822 |
"step": 750
|
| 823 |
},
|
| 824 |
{
|
| 825 |
+
"epoch": 33.0,
|
| 826 |
+
"eval_accuracy": 0.7795275590551181,
|
| 827 |
+
"eval_loss": 0.5041583180427551,
|
| 828 |
+
"eval_runtime": 4.9981,
|
| 829 |
+
"eval_samples_per_second": 50.82,
|
| 830 |
+
"eval_steps_per_second": 0.8,
|
| 831 |
+
"step": 759
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 33.04347826086956,
|
| 835 |
+
"grad_norm": 3.6429481506347656,
|
| 836 |
+
"learning_rate": 6.782608695652174e-06,
|
| 837 |
+
"loss": 0.454,
|
| 838 |
"step": 760
|
| 839 |
},
|
| 840 |
{
|
| 841 |
+
"epoch": 33.47826086956522,
|
| 842 |
+
"grad_norm": 2.436143636703491,
|
| 843 |
+
"learning_rate": 6.6086956521739135e-06,
|
| 844 |
+
"loss": 0.4612,
|
| 845 |
"step": 770
|
| 846 |
},
|
| 847 |
{
|
| 848 |
+
"epoch": 33.91304347826087,
|
| 849 |
+
"grad_norm": 2.5793776512145996,
|
| 850 |
+
"learning_rate": 6.434782608695652e-06,
|
| 851 |
+
"loss": 0.4374,
|
| 852 |
"step": 780
|
| 853 |
},
|
| 854 |
{
|
| 855 |
+
"epoch": 34.0,
|
| 856 |
+
"eval_accuracy": 0.7677165354330708,
|
| 857 |
+
"eval_loss": 0.5063456296920776,
|
| 858 |
+
"eval_runtime": 3.7117,
|
| 859 |
+
"eval_samples_per_second": 68.432,
|
| 860 |
+
"eval_steps_per_second": 1.078,
|
| 861 |
+
"step": 782
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 34.34782608695652,
|
| 865 |
+
"grad_norm": 3.71374773979187,
|
| 866 |
+
"learning_rate": 6.260869565217392e-06,
|
| 867 |
+
"loss": 0.4667,
|
| 868 |
"step": 790
|
| 869 |
},
|
| 870 |
{
|
| 871 |
+
"epoch": 34.78260869565217,
|
| 872 |
+
"grad_norm": 4.282368183135986,
|
| 873 |
+
"learning_rate": 6.086956521739132e-06,
|
| 874 |
+
"loss": 0.4864,
|
| 875 |
"step": 800
|
| 876 |
},
|
| 877 |
{
|
| 878 |
+
"epoch": 35.0,
|
| 879 |
+
"eval_accuracy": 0.7677165354330708,
|
| 880 |
+
"eval_loss": 0.5039507150650024,
|
| 881 |
+
"eval_runtime": 3.6837,
|
| 882 |
+
"eval_samples_per_second": 68.952,
|
| 883 |
+
"eval_steps_per_second": 1.086,
|
| 884 |
+
"step": 805
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 35.21739130434783,
|
| 888 |
+
"grad_norm": 2.896638870239258,
|
| 889 |
+
"learning_rate": 5.91304347826087e-06,
|
| 890 |
+
"loss": 0.4922,
|
| 891 |
"step": 810
|
| 892 |
},
|
| 893 |
{
|
| 894 |
+
"epoch": 35.65217391304348,
|
| 895 |
+
"grad_norm": 2.2342097759246826,
|
| 896 |
+
"learning_rate": 5.739130434782609e-06,
|
| 897 |
+
"loss": 0.4354,
|
| 898 |
"step": 820
|
| 899 |
},
|
| 900 |
{
|
| 901 |
+
"epoch": 36.0,
|
| 902 |
+
"eval_accuracy": 0.7716535433070866,
|
| 903 |
+
"eval_loss": 0.5108994841575623,
|
| 904 |
+
"eval_runtime": 4.9899,
|
| 905 |
+
"eval_samples_per_second": 50.902,
|
| 906 |
+
"eval_steps_per_second": 0.802,
|
| 907 |
+
"step": 828
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 36.08695652173913,
|
| 911 |
+
"grad_norm": 8.385408401489258,
|
| 912 |
+
"learning_rate": 5.565217391304348e-06,
|
| 913 |
+
"loss": 0.4585,
|
| 914 |
"step": 830
|
| 915 |
},
|
| 916 |
{
|
| 917 |
+
"epoch": 36.52173913043478,
|
| 918 |
+
"grad_norm": 2.839411497116089,
|
| 919 |
+
"learning_rate": 5.391304347826088e-06,
|
| 920 |
+
"loss": 0.4497,
|
| 921 |
"step": 840
|
| 922 |
},
|
| 923 |
{
|
| 924 |
+
"epoch": 36.95652173913044,
|
| 925 |
+
"grad_norm": 2.479076623916626,
|
| 926 |
+
"learning_rate": 5.2173913043478265e-06,
|
| 927 |
+
"loss": 0.4655,
|
| 928 |
"step": 850
|
| 929 |
},
|
| 930 |
{
|
| 931 |
+
"epoch": 37.0,
|
| 932 |
+
"eval_accuracy": 0.7716535433070866,
|
| 933 |
+
"eval_loss": 0.510716438293457,
|
| 934 |
+
"eval_runtime": 3.6997,
|
| 935 |
+
"eval_samples_per_second": 68.653,
|
| 936 |
+
"eval_steps_per_second": 1.081,
|
| 937 |
+
"step": 851
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 37.391304347826086,
|
| 941 |
+
"grad_norm": 2.271686553955078,
|
| 942 |
+
"learning_rate": 5.043478260869565e-06,
|
| 943 |
+
"loss": 0.4462,
|
| 944 |
"step": 860
|
| 945 |
},
|
| 946 |
{
|
| 947 |
+
"epoch": 37.82608695652174,
|
| 948 |
+
"grad_norm": 3.4210402965545654,
|
| 949 |
+
"learning_rate": 4.869565217391305e-06,
|
| 950 |
+
"loss": 0.4691,
|
| 951 |
"step": 870
|
| 952 |
},
|
| 953 |
{
|
| 954 |
+
"epoch": 38.0,
|
| 955 |
+
"eval_accuracy": 0.7677165354330708,
|
| 956 |
+
"eval_loss": 0.5093376636505127,
|
| 957 |
+
"eval_runtime": 3.7287,
|
| 958 |
+
"eval_samples_per_second": 68.119,
|
| 959 |
+
"eval_steps_per_second": 1.073,
|
| 960 |
+
"step": 874
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 38.26086956521739,
|
| 964 |
+
"grad_norm": 5.694761276245117,
|
| 965 |
+
"learning_rate": 4.695652173913044e-06,
|
| 966 |
+
"loss": 0.4592,
|
| 967 |
"step": 880
|
| 968 |
},
|
| 969 |
{
|
| 970 |
+
"epoch": 38.69565217391305,
|
| 971 |
+
"grad_norm": 2.2949883937835693,
|
| 972 |
+
"learning_rate": 4.5217391304347826e-06,
|
| 973 |
+
"loss": 0.4826,
|
| 974 |
"step": 890
|
| 975 |
},
|
| 976 |
{
|
| 977 |
+
"epoch": 39.0,
|
| 978 |
+
"eval_accuracy": 0.7716535433070866,
|
| 979 |
+
"eval_loss": 0.5044277906417847,
|
| 980 |
+
"eval_runtime": 4.9781,
|
| 981 |
+
"eval_samples_per_second": 51.024,
|
| 982 |
+
"eval_steps_per_second": 0.804,
|
| 983 |
+
"step": 897
|
| 984 |
},
|
| 985 |
{
|
| 986 |
+
"epoch": 39.130434782608695,
|
| 987 |
+
"grad_norm": 3.4144210815429688,
|
| 988 |
+
"learning_rate": 4.347826086956522e-06,
|
| 989 |
+
"loss": 0.4407,
|
|
|
|
|
|
|
| 990 |
"step": 900
|
| 991 |
},
|
| 992 |
{
|
| 993 |
+
"epoch": 39.56521739130435,
|
| 994 |
+
"grad_norm": 2.22868013381958,
|
| 995 |
+
"learning_rate": 4.173913043478261e-06,
|
| 996 |
+
"loss": 0.4482,
|
| 997 |
+
"step": 910
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 40.0,
|
| 1001 |
+
"grad_norm": 3.2193689346313477,
|
| 1002 |
+
"learning_rate": 4.000000000000001e-06,
|
| 1003 |
+
"loss": 0.4577,
|
| 1004 |
+
"step": 920
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 40.0,
|
| 1008 |
+
"eval_accuracy": 0.7795275590551181,
|
| 1009 |
+
"eval_loss": 0.4999626874923706,
|
| 1010 |
+
"eval_runtime": 3.6952,
|
| 1011 |
+
"eval_samples_per_second": 68.738,
|
| 1012 |
+
"eval_steps_per_second": 1.082,
|
| 1013 |
+
"step": 920
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 40.43478260869565,
|
| 1017 |
+
"grad_norm": 4.500718593597412,
|
| 1018 |
+
"learning_rate": 3.8260869565217395e-06,
|
| 1019 |
+
"loss": 0.4585,
|
| 1020 |
+
"step": 930
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 40.869565217391305,
|
| 1024 |
+
"grad_norm": 1.9281222820281982,
|
| 1025 |
+
"learning_rate": 3.6521739130434787e-06,
|
| 1026 |
+
"loss": 0.4636,
|
| 1027 |
+
"step": 940
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 41.0,
|
| 1031 |
+
"eval_accuracy": 0.7716535433070866,
|
| 1032 |
+
"eval_loss": 0.4962589144706726,
|
| 1033 |
+
"eval_runtime": 3.6977,
|
| 1034 |
+
"eval_samples_per_second": 68.69,
|
| 1035 |
+
"eval_steps_per_second": 1.082,
|
| 1036 |
+
"step": 943
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 41.30434782608695,
|
| 1040 |
+
"grad_norm": 2.193452835083008,
|
| 1041 |
+
"learning_rate": 3.4782608695652175e-06,
|
| 1042 |
+
"loss": 0.4306,
|
| 1043 |
+
"step": 950
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 41.73913043478261,
|
| 1047 |
+
"grad_norm": 2.2370336055755615,
|
| 1048 |
+
"learning_rate": 3.3043478260869567e-06,
|
| 1049 |
+
"loss": 0.4361,
|
| 1050 |
+
"step": 960
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 42.0,
|
| 1054 |
+
"eval_accuracy": 0.7716535433070866,
|
| 1055 |
+
"eval_loss": 0.4958040118217468,
|
| 1056 |
+
"eval_runtime": 4.9548,
|
| 1057 |
+
"eval_samples_per_second": 51.264,
|
| 1058 |
+
"eval_steps_per_second": 0.807,
|
| 1059 |
+
"step": 966
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 42.17391304347826,
|
| 1063 |
+
"grad_norm": 3.6354355812072754,
|
| 1064 |
+
"learning_rate": 3.130434782608696e-06,
|
| 1065 |
+
"loss": 0.4514,
|
| 1066 |
+
"step": 970
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 42.608695652173914,
|
| 1070 |
+
"grad_norm": 1.8955118656158447,
|
| 1071 |
+
"learning_rate": 2.956521739130435e-06,
|
| 1072 |
+
"loss": 0.4534,
|
| 1073 |
+
"step": 980
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 43.0,
|
| 1077 |
+
"eval_accuracy": 0.7795275590551181,
|
| 1078 |
+
"eval_loss": 0.5007808208465576,
|
| 1079 |
+
"eval_runtime": 3.7121,
|
| 1080 |
+
"eval_samples_per_second": 68.424,
|
| 1081 |
+
"eval_steps_per_second": 1.078,
|
| 1082 |
+
"step": 989
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 43.04347826086956,
|
| 1086 |
+
"grad_norm": 2.2034902572631836,
|
| 1087 |
+
"learning_rate": 2.782608695652174e-06,
|
| 1088 |
+
"loss": 0.4176,
|
| 1089 |
+
"step": 990
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 43.47826086956522,
|
| 1093 |
+
"grad_norm": 4.387076377868652,
|
| 1094 |
+
"learning_rate": 2.6086956521739132e-06,
|
| 1095 |
+
"loss": 0.4748,
|
| 1096 |
+
"step": 1000
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 43.91304347826087,
|
| 1100 |
+
"grad_norm": 5.444644927978516,
|
| 1101 |
+
"learning_rate": 2.4347826086956525e-06,
|
| 1102 |
+
"loss": 0.4559,
|
| 1103 |
+
"step": 1010
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 44.0,
|
| 1107 |
+
"eval_accuracy": 0.7795275590551181,
|
| 1108 |
+
"eval_loss": 0.5025174021720886,
|
| 1109 |
+
"eval_runtime": 3.7093,
|
| 1110 |
+
"eval_samples_per_second": 68.476,
|
| 1111 |
+
"eval_steps_per_second": 1.078,
|
| 1112 |
+
"step": 1012
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 44.34782608695652,
|
| 1116 |
+
"grad_norm": 2.2067017555236816,
|
| 1117 |
+
"learning_rate": 2.2608695652173913e-06,
|
| 1118 |
+
"loss": 0.4882,
|
| 1119 |
+
"step": 1020
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 44.78260869565217,
|
| 1123 |
+
"grad_norm": 3.562736988067627,
|
| 1124 |
+
"learning_rate": 2.0869565217391305e-06,
|
| 1125 |
+
"loss": 0.4189,
|
| 1126 |
+
"step": 1030
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 45.0,
|
| 1130 |
+
"eval_accuracy": 0.7755905511811023,
|
| 1131 |
+
"eval_loss": 0.5014046430587769,
|
| 1132 |
+
"eval_runtime": 4.9992,
|
| 1133 |
+
"eval_samples_per_second": 50.808,
|
| 1134 |
+
"eval_steps_per_second": 0.8,
|
| 1135 |
+
"step": 1035
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 45.21739130434783,
|
| 1139 |
+
"grad_norm": 10.402663230895996,
|
| 1140 |
+
"learning_rate": 1.9130434782608697e-06,
|
| 1141 |
+
"loss": 0.4432,
|
| 1142 |
+
"step": 1040
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 45.65217391304348,
|
| 1146 |
+
"grad_norm": 4.949878215789795,
|
| 1147 |
+
"learning_rate": 1.7391304347826088e-06,
|
| 1148 |
+
"loss": 0.4861,
|
| 1149 |
+
"step": 1050
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 46.0,
|
| 1153 |
+
"eval_accuracy": 0.7677165354330708,
|
| 1154 |
+
"eval_loss": 0.5003762245178223,
|
| 1155 |
+
"eval_runtime": 3.7019,
|
| 1156 |
+
"eval_samples_per_second": 68.614,
|
| 1157 |
+
"eval_steps_per_second": 1.081,
|
| 1158 |
+
"step": 1058
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 46.08695652173913,
|
| 1162 |
+
"grad_norm": 1.938593864440918,
|
| 1163 |
+
"learning_rate": 1.565217391304348e-06,
|
| 1164 |
+
"loss": 0.4326,
|
| 1165 |
+
"step": 1060
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 46.52173913043478,
|
| 1169 |
+
"grad_norm": 3.236699342727661,
|
| 1170 |
+
"learning_rate": 1.391304347826087e-06,
|
| 1171 |
+
"loss": 0.4726,
|
| 1172 |
+
"step": 1070
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 46.95652173913044,
|
| 1176 |
+
"grad_norm": 3.047184944152832,
|
| 1177 |
+
"learning_rate": 1.2173913043478262e-06,
|
| 1178 |
+
"loss": 0.4709,
|
| 1179 |
+
"step": 1080
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 47.0,
|
| 1183 |
+
"eval_accuracy": 0.7795275590551181,
|
| 1184 |
+
"eval_loss": 0.5004997849464417,
|
| 1185 |
+
"eval_runtime": 3.7143,
|
| 1186 |
+
"eval_samples_per_second": 68.384,
|
| 1187 |
+
"eval_steps_per_second": 1.077,
|
| 1188 |
+
"step": 1081
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 47.391304347826086,
|
| 1192 |
+
"grad_norm": 2.8639461994171143,
|
| 1193 |
+
"learning_rate": 1.0434782608695653e-06,
|
| 1194 |
+
"loss": 0.4649,
|
| 1195 |
+
"step": 1090
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 47.82608695652174,
|
| 1199 |
+
"grad_norm": 3.7704715728759766,
|
| 1200 |
+
"learning_rate": 8.695652173913044e-07,
|
| 1201 |
+
"loss": 0.4726,
|
| 1202 |
+
"step": 1100
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 48.0,
|
| 1206 |
+
"eval_accuracy": 0.7716535433070866,
|
| 1207 |
+
"eval_loss": 0.5007592439651489,
|
| 1208 |
+
"eval_runtime": 4.8498,
|
| 1209 |
+
"eval_samples_per_second": 52.373,
|
| 1210 |
+
"eval_steps_per_second": 0.825,
|
| 1211 |
+
"step": 1104
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 48.26086956521739,
|
| 1215 |
+
"grad_norm": 4.941337585449219,
|
| 1216 |
+
"learning_rate": 6.956521739130435e-07,
|
| 1217 |
+
"loss": 0.4314,
|
| 1218 |
+
"step": 1110
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 48.69565217391305,
|
| 1222 |
+
"grad_norm": 3.2265655994415283,
|
| 1223 |
+
"learning_rate": 5.217391304347826e-07,
|
| 1224 |
+
"loss": 0.4441,
|
| 1225 |
+
"step": 1120
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 49.0,
|
| 1229 |
+
"eval_accuracy": 0.7755905511811023,
|
| 1230 |
+
"eval_loss": 0.4987953305244446,
|
| 1231 |
+
"eval_runtime": 3.6681,
|
| 1232 |
+
"eval_samples_per_second": 69.246,
|
| 1233 |
+
"eval_steps_per_second": 1.09,
|
| 1234 |
+
"step": 1127
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 49.130434782608695,
|
| 1238 |
+
"grad_norm": 3.7678611278533936,
|
| 1239 |
+
"learning_rate": 3.4782608695652175e-07,
|
| 1240 |
+
"loss": 0.4571,
|
| 1241 |
+
"step": 1130
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 49.56521739130435,
|
| 1245 |
+
"grad_norm": 3.657460927963257,
|
| 1246 |
+
"learning_rate": 1.7391304347826088e-07,
|
| 1247 |
+
"loss": 0.4558,
|
| 1248 |
+
"step": 1140
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 50.0,
|
| 1252 |
+
"grad_norm": 3.096832513809204,
|
| 1253 |
+
"learning_rate": 0.0,
|
| 1254 |
+
"loss": 0.4579,
|
| 1255 |
+
"step": 1150
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 50.0,
|
| 1259 |
+
"eval_accuracy": 0.7755905511811023,
|
| 1260 |
+
"eval_loss": 0.499985009431839,
|
| 1261 |
+
"eval_runtime": 3.8189,
|
| 1262 |
+
"eval_samples_per_second": 66.512,
|
| 1263 |
+
"eval_steps_per_second": 1.047,
|
| 1264 |
+
"step": 1150
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 50.0,
|
| 1268 |
+
"step": 1150,
|
| 1269 |
+
"total_flos": 2.72467378584576e+17,
|
| 1270 |
+
"train_loss": 0.4791690407628598,
|
| 1271 |
+
"train_runtime": 1616.3187,
|
| 1272 |
+
"train_samples_per_second": 44.391,
|
| 1273 |
+
"train_steps_per_second": 0.711
|
| 1274 |
}
|
| 1275 |
],
|
| 1276 |
"logging_steps": 10,
|
| 1277 |
+
"max_steps": 1150,
|
| 1278 |
"num_input_tokens_seen": 0,
|
| 1279 |
+
"num_train_epochs": 50,
|
| 1280 |
"save_steps": 500,
|
| 1281 |
"stateful_callbacks": {
|
| 1282 |
"TrainerControl": {
|
|
|
|
| 1290 |
"attributes": {}
|
| 1291 |
}
|
| 1292 |
},
|
| 1293 |
+
"total_flos": 2.72467378584576e+17,
|
| 1294 |
+
"train_batch_size": 64,
|
| 1295 |
"trial_name": null,
|
| 1296 |
"trial_params": null
|
| 1297 |
}
|