End of training
Browse files- README.md +9 -9
- all_results.json +14 -6
- eval_results.json +11 -0
- runs/Jul25_04-38-12_d85b605ca0e9/events.out.tfevents.1721882854.d85b605ca0e9.546.2 +3 -0
- train_results.json +6 -6
- trainer_state.json +1114 -418
README.md
CHANGED
|
@@ -25,16 +25,16 @@ model-index:
|
|
| 25 |
metrics:
|
| 26 |
- name: Accuracy
|
| 27 |
type: accuracy
|
| 28 |
-
value: 0.
|
| 29 |
- name: F1
|
| 30 |
type: f1
|
| 31 |
-
value: 0.
|
| 32 |
- name: Precision
|
| 33 |
type: precision
|
| 34 |
-
value: 0.
|
| 35 |
- name: Recall
|
| 36 |
type: recall
|
| 37 |
-
value: 0.
|
| 38 |
---
|
| 39 |
|
| 40 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -44,11 +44,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 44 |
|
| 45 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the webdataset dataset.
|
| 46 |
It achieves the following results on the evaluation set:
|
| 47 |
-
- Loss: 0.
|
| 48 |
-
- Accuracy: 0.
|
| 49 |
-
- F1: 0.
|
| 50 |
-
- Precision: 0.
|
| 51 |
-
- Recall: 0.
|
| 52 |
|
| 53 |
## Model description
|
| 54 |
|
|
|
|
| 25 |
metrics:
|
| 26 |
- name: Accuracy
|
| 27 |
type: accuracy
|
| 28 |
+
value: 0.963716814159292
|
| 29 |
- name: F1
|
| 30 |
type: f1
|
| 31 |
+
value: 0.9118279569892475
|
| 32 |
- name: Precision
|
| 33 |
type: precision
|
| 34 |
+
value: 0.905982905982906
|
| 35 |
- name: Recall
|
| 36 |
type: recall
|
| 37 |
+
value: 0.9177489177489178
|
| 38 |
---
|
| 39 |
|
| 40 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 44 |
|
| 45 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the webdataset dataset.
|
| 46 |
It achieves the following results on the evaluation set:
|
| 47 |
+
- Loss: 0.0965
|
| 48 |
+
- Accuracy: 0.9637
|
| 49 |
+
- F1: 0.9118
|
| 50 |
+
- Precision: 0.9060
|
| 51 |
+
- Recall: 0.9177
|
| 52 |
|
| 53 |
## Model description
|
| 54 |
|
all_results.json
CHANGED
|
@@ -1,8 +1,16 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 0.963716814159292,
|
| 4 |
+
"eval_f1": 0.9118279569892475,
|
| 5 |
+
"eval_loss": 0.09645664691925049,
|
| 6 |
+
"eval_precision": 0.905982905982906,
|
| 7 |
+
"eval_recall": 0.9177489177489178,
|
| 8 |
+
"eval_runtime": 0.9893,
|
| 9 |
+
"eval_samples_per_second": 114.226,
|
| 10 |
+
"eval_steps_per_second": 15.163,
|
| 11 |
+
"total_flos": 2.352825493649326e+18,
|
| 12 |
+
"train_loss": 0.05003170374160012,
|
| 13 |
+
"train_runtime": 517.118,
|
| 14 |
+
"train_samples_per_second": 58.71,
|
| 15 |
+
"train_steps_per_second": 3.713
|
| 16 |
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"eval_accuracy": 0.963716814159292,
|
| 4 |
+
"eval_f1": 0.9118279569892475,
|
| 5 |
+
"eval_loss": 0.09645664691925049,
|
| 6 |
+
"eval_precision": 0.905982905982906,
|
| 7 |
+
"eval_recall": 0.9177489177489178,
|
| 8 |
+
"eval_runtime": 0.9893,
|
| 9 |
+
"eval_samples_per_second": 114.226,
|
| 10 |
+
"eval_steps_per_second": 15.163
|
| 11 |
+
}
|
runs/Jul25_04-38-12_d85b605ca0e9/events.out.tfevents.1721882854.d85b605ca0e9.546.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1d05bca1c639e8d629fa575e09355eba8e8b6b2e13660194cfc1044aab08ad6
|
| 3 |
+
size 560
|
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": 3.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 30.0,
|
| 3 |
+
"total_flos": 2.352825493649326e+18,
|
| 4 |
+
"train_loss": 0.05003170374160012,
|
| 5 |
+
"train_runtime": 517.118,
|
| 6 |
+
"train_samples_per_second": 58.71,
|
| 7 |
+
"train_steps_per_second": 3.713
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,903 +1,1599 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned_v2024-7-24-frost/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 100,
|
| 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": 0.15625,
|
| 13 |
-
"grad_norm": 0.
|
| 14 |
-
"learning_rate": 1.
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.3125,
|
| 20 |
-
"grad_norm": 0.
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.46875,
|
| 27 |
-
"grad_norm": 0.
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.625,
|
| 34 |
-
"grad_norm": 0.
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.78125,
|
| 41 |
-
"grad_norm": 0.
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.9375,
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 1.09375,
|
| 55 |
-
"grad_norm": 0.
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 1.25,
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 1.40625,
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 1.5625,
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 1.5625,
|
| 83 |
-
"eval_accuracy": 0.
|
| 84 |
-
"eval_f1": 0.
|
| 85 |
-
"eval_loss": 0.
|
| 86 |
-
"eval_precision": 0.
|
| 87 |
-
"eval_recall": 0.
|
| 88 |
-
"eval_runtime": 0.
|
| 89 |
-
"eval_samples_per_second":
|
| 90 |
-
"eval_steps_per_second": 16.
|
| 91 |
"step": 100
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 1.71875,
|
| 95 |
-
"grad_norm": 0.
|
| 96 |
-
"learning_rate": 0.
|
| 97 |
-
"loss": 0.
|
| 98 |
"step": 110
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 1.875,
|
| 102 |
-
"grad_norm": 0.
|
| 103 |
-
"learning_rate": 0.
|
| 104 |
-
"loss": 0.
|
| 105 |
"step": 120
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 2.03125,
|
| 109 |
-
"grad_norm": 0.
|
| 110 |
-
"learning_rate": 0.
|
| 111 |
-
"loss": 0.
|
| 112 |
"step": 130
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"epoch": 2.1875,
|
| 116 |
-
"grad_norm": 0.
|
| 117 |
-
"learning_rate": 0.
|
| 118 |
-
"loss": 0.
|
| 119 |
"step": 140
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 2.34375,
|
| 123 |
-
"grad_norm": 0.
|
| 124 |
-
"learning_rate": 0.
|
| 125 |
-
"loss": 0.
|
| 126 |
"step": 150
|
| 127 |
},
|
| 128 |
{
|
| 129 |
"epoch": 2.5,
|
| 130 |
-
"grad_norm": 0.
|
| 131 |
-
"learning_rate": 0.
|
| 132 |
-
"loss": 0.
|
| 133 |
"step": 160
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"epoch": 2.65625,
|
| 137 |
-
"grad_norm": 0.
|
| 138 |
-
"learning_rate": 0.
|
| 139 |
-
"loss": 0.
|
| 140 |
"step": 170
|
| 141 |
},
|
| 142 |
{
|
| 143 |
"epoch": 2.8125,
|
| 144 |
-
"grad_norm": 0.
|
| 145 |
-
"learning_rate": 0.
|
| 146 |
-
"loss": 0.
|
| 147 |
"step": 180
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 2.96875,
|
| 151 |
-
"grad_norm":
|
| 152 |
-
"learning_rate": 0.
|
| 153 |
-
"loss": 0.
|
| 154 |
"step": 190
|
| 155 |
},
|
| 156 |
{
|
| 157 |
"epoch": 3.125,
|
| 158 |
-
"grad_norm":
|
| 159 |
-
"learning_rate": 0.
|
| 160 |
-
"loss": 0.
|
| 161 |
"step": 200
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 3.125,
|
| 165 |
-
"eval_accuracy": 0.
|
| 166 |
-
"eval_f1": 0.
|
| 167 |
-
"eval_loss": 0.
|
| 168 |
-
"eval_precision": 0.
|
| 169 |
-
"eval_recall": 0.
|
| 170 |
-
"eval_runtime": 0.
|
| 171 |
-
"eval_samples_per_second":
|
| 172 |
-
"eval_steps_per_second":
|
| 173 |
"step": 200
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 3.28125,
|
| 177 |
-
"grad_norm": 0.
|
| 178 |
-
"learning_rate": 0.
|
| 179 |
-
"loss": 0.
|
| 180 |
"step": 210
|
| 181 |
},
|
| 182 |
{
|
| 183 |
"epoch": 3.4375,
|
| 184 |
-
"grad_norm": 0.
|
| 185 |
-
"learning_rate": 0.
|
| 186 |
-
"loss": 0.
|
| 187 |
"step": 220
|
| 188 |
},
|
| 189 |
{
|
| 190 |
"epoch": 3.59375,
|
| 191 |
-
"grad_norm": 0.
|
| 192 |
-
"learning_rate": 0.
|
| 193 |
-
"loss": 0.
|
| 194 |
"step": 230
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 3.75,
|
| 198 |
-
"grad_norm": 0.
|
| 199 |
-
"learning_rate": 0.
|
| 200 |
-
"loss": 0.
|
| 201 |
"step": 240
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 3.90625,
|
| 205 |
-
"grad_norm": 0.
|
| 206 |
-
"learning_rate": 0.
|
| 207 |
-
"loss": 0.
|
| 208 |
"step": 250
|
| 209 |
},
|
| 210 |
{
|
| 211 |
"epoch": 4.0625,
|
| 212 |
-
"grad_norm": 0.
|
| 213 |
-
"learning_rate": 0.
|
| 214 |
-
"loss": 0.
|
| 215 |
"step": 260
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 4.21875,
|
| 219 |
-
"grad_norm": 0.
|
| 220 |
-
"learning_rate": 0.
|
| 221 |
-
"loss": 0.
|
| 222 |
"step": 270
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 4.375,
|
| 226 |
-
"grad_norm": 0.
|
| 227 |
-
"learning_rate": 0.
|
| 228 |
-
"loss": 0.
|
| 229 |
"step": 280
|
| 230 |
},
|
| 231 |
{
|
| 232 |
"epoch": 4.53125,
|
| 233 |
-
"grad_norm": 0.
|
| 234 |
-
"learning_rate": 0.
|
| 235 |
-
"loss": 0.
|
| 236 |
"step": 290
|
| 237 |
},
|
| 238 |
{
|
| 239 |
"epoch": 4.6875,
|
| 240 |
-
"grad_norm": 0.
|
| 241 |
-
"learning_rate": 0.
|
| 242 |
-
"loss": 0.
|
| 243 |
"step": 300
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 4.6875,
|
| 247 |
-
"eval_accuracy": 0.
|
| 248 |
-
"eval_f1": 0.
|
| 249 |
-
"eval_loss": 0.
|
| 250 |
-
"eval_precision": 0.
|
| 251 |
-
"eval_recall": 0.
|
| 252 |
-
"eval_runtime":
|
| 253 |
-
"eval_samples_per_second":
|
| 254 |
-
"eval_steps_per_second":
|
| 255 |
"step": 300
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 4.84375,
|
| 259 |
-
"grad_norm": 0.
|
| 260 |
-
"learning_rate": 0.
|
| 261 |
-
"loss": 0.
|
| 262 |
"step": 310
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 5.0,
|
| 266 |
-
"grad_norm": 0.
|
| 267 |
-
"learning_rate": 0.
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 320
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 5.15625,
|
| 273 |
-
"grad_norm": 0.
|
| 274 |
-
"learning_rate": 0.
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 330
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 5.3125,
|
| 280 |
-
"grad_norm": 0.
|
| 281 |
-
"learning_rate": 0.
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 340
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 5.46875,
|
| 287 |
-
"grad_norm": 0.
|
| 288 |
-
"learning_rate": 0.
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 350
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 5.625,
|
| 294 |
-
"grad_norm": 0.
|
| 295 |
-
"learning_rate": 0.
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 360
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 5.78125,
|
| 301 |
-
"grad_norm": 0.
|
| 302 |
-
"learning_rate": 0.
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 370
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 5.9375,
|
| 308 |
-
"grad_norm": 0.
|
| 309 |
-
"learning_rate": 0.
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 380
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 6.09375,
|
| 315 |
-
"grad_norm": 0.
|
| 316 |
-
"learning_rate": 0.
|
| 317 |
-
"loss": 0.
|
| 318 |
"step": 390
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 6.25,
|
| 322 |
-
"grad_norm": 0.
|
| 323 |
-
"learning_rate": 0.
|
| 324 |
-
"loss": 0.
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 6.25,
|
| 329 |
-
"eval_accuracy": 0.
|
| 330 |
-
"eval_f1": 0.
|
| 331 |
-
"eval_loss": 0.
|
| 332 |
-
"eval_precision": 0.
|
| 333 |
-
"eval_recall": 0.
|
| 334 |
-
"eval_runtime":
|
| 335 |
-
"eval_samples_per_second":
|
| 336 |
-
"eval_steps_per_second":
|
| 337 |
"step": 400
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 6.40625,
|
| 341 |
-
"grad_norm": 0.
|
| 342 |
-
"learning_rate": 0.
|
| 343 |
-
"loss": 0.
|
| 344 |
"step": 410
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 6.5625,
|
| 348 |
-
"grad_norm": 0.
|
| 349 |
-
"learning_rate": 0.
|
| 350 |
-
"loss": 0.
|
| 351 |
"step": 420
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 6.71875,
|
| 355 |
-
"grad_norm": 0.
|
| 356 |
-
"learning_rate": 0.
|
| 357 |
-
"loss": 0.
|
| 358 |
"step": 430
|
| 359 |
},
|
| 360 |
{
|
| 361 |
"epoch": 6.875,
|
| 362 |
-
"grad_norm": 0.
|
| 363 |
-
"learning_rate": 0.
|
| 364 |
-
"loss": 0.
|
| 365 |
"step": 440
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 7.03125,
|
| 369 |
-
"grad_norm": 0.
|
| 370 |
-
"learning_rate": 0.
|
| 371 |
-
"loss": 0.
|
| 372 |
"step": 450
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 7.1875,
|
| 376 |
-
"grad_norm": 0.
|
| 377 |
-
"learning_rate": 0.
|
| 378 |
-
"loss": 0.
|
| 379 |
"step": 460
|
| 380 |
},
|
| 381 |
{
|
| 382 |
"epoch": 7.34375,
|
| 383 |
-
"grad_norm": 0.
|
| 384 |
-
"learning_rate": 0.
|
| 385 |
-
"loss": 0.
|
| 386 |
"step": 470
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 7.5,
|
| 390 |
-
"grad_norm": 0.
|
| 391 |
-
"learning_rate": 0.
|
| 392 |
-
"loss": 0.
|
| 393 |
"step": 480
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 7.65625,
|
| 397 |
-
"grad_norm": 0.
|
| 398 |
-
"learning_rate": 0.
|
| 399 |
-
"loss": 0.
|
| 400 |
"step": 490
|
| 401 |
},
|
| 402 |
{
|
| 403 |
"epoch": 7.8125,
|
| 404 |
-
"grad_norm": 0.
|
| 405 |
-
"learning_rate": 0.
|
| 406 |
-
"loss": 0.
|
| 407 |
"step": 500
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 7.8125,
|
| 411 |
-
"eval_accuracy": 0.
|
| 412 |
-
"eval_f1": 0.
|
| 413 |
-
"eval_loss": 0.
|
| 414 |
-
"eval_precision": 0.
|
| 415 |
-
"eval_recall": 0.
|
| 416 |
-
"eval_runtime":
|
| 417 |
-
"eval_samples_per_second":
|
| 418 |
-
"eval_steps_per_second":
|
| 419 |
"step": 500
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 7.96875,
|
| 423 |
-
"grad_norm": 0.
|
| 424 |
-
"learning_rate": 0.
|
| 425 |
-
"loss": 0.
|
| 426 |
"step": 510
|
| 427 |
},
|
| 428 |
{
|
| 429 |
"epoch": 8.125,
|
| 430 |
-
"grad_norm": 0.
|
| 431 |
-
"learning_rate": 0.
|
| 432 |
-
"loss": 0.
|
| 433 |
"step": 520
|
| 434 |
},
|
| 435 |
{
|
| 436 |
"epoch": 8.28125,
|
| 437 |
-
"grad_norm": 0.
|
| 438 |
-
"learning_rate": 0.
|
| 439 |
-
"loss": 0.
|
| 440 |
"step": 530
|
| 441 |
},
|
| 442 |
{
|
| 443 |
"epoch": 8.4375,
|
| 444 |
-
"grad_norm": 0.
|
| 445 |
-
"learning_rate": 0.
|
| 446 |
-
"loss": 0.
|
| 447 |
"step": 540
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 8.59375,
|
| 451 |
-
"grad_norm": 0.
|
| 452 |
-
"learning_rate": 0.
|
| 453 |
-
"loss": 0.
|
| 454 |
"step": 550
|
| 455 |
},
|
| 456 |
{
|
| 457 |
"epoch": 8.75,
|
| 458 |
-
"grad_norm": 0.
|
| 459 |
-
"learning_rate": 0.
|
| 460 |
-
"loss": 0.
|
| 461 |
"step": 560
|
| 462 |
},
|
| 463 |
{
|
| 464 |
"epoch": 8.90625,
|
| 465 |
-
"grad_norm": 0.
|
| 466 |
-
"learning_rate": 0.
|
| 467 |
-
"loss": 0.
|
| 468 |
"step": 570
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 9.0625,
|
| 472 |
-
"grad_norm": 0.
|
| 473 |
-
"learning_rate": 0.
|
| 474 |
-
"loss": 0.
|
| 475 |
"step": 580
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 9.21875,
|
| 479 |
-
"grad_norm": 0.
|
| 480 |
-
"learning_rate": 0.
|
| 481 |
-
"loss": 0.
|
| 482 |
"step": 590
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 9.375,
|
| 486 |
-
"grad_norm": 0.
|
| 487 |
-
"learning_rate":
|
| 488 |
-
"loss": 0.
|
| 489 |
"step": 600
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 9.375,
|
| 493 |
-
"eval_accuracy": 0.
|
| 494 |
-
"eval_f1": 0.
|
| 495 |
-
"eval_loss": 0.
|
| 496 |
-
"eval_precision": 0.
|
| 497 |
-
"eval_recall": 0.
|
| 498 |
-
"eval_runtime":
|
| 499 |
-
"eval_samples_per_second":
|
| 500 |
-
"eval_steps_per_second":
|
| 501 |
"step": 600
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 9.53125,
|
| 505 |
-
"grad_norm": 0.
|
| 506 |
-
"learning_rate":
|
| 507 |
-
"loss": 0.
|
| 508 |
"step": 610
|
| 509 |
},
|
| 510 |
{
|
| 511 |
"epoch": 9.6875,
|
| 512 |
-
"grad_norm": 0.
|
| 513 |
-
"learning_rate":
|
| 514 |
-
"loss": 0.
|
| 515 |
"step": 620
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 9.84375,
|
| 519 |
-
"grad_norm": 0.
|
| 520 |
-
"learning_rate":
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 630
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 10.0,
|
| 526 |
-
"grad_norm": 0.
|
| 527 |
-
"learning_rate":
|
| 528 |
-
"loss": 0.
|
| 529 |
"step": 640
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 10.15625,
|
| 533 |
-
"grad_norm": 0.
|
| 534 |
-
"learning_rate":
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 650
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 10.3125,
|
| 540 |
-
"grad_norm": 0.
|
| 541 |
-
"learning_rate":
|
| 542 |
-
"loss": 0.
|
| 543 |
"step": 660
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 10.46875,
|
| 547 |
-
"grad_norm": 0.
|
| 548 |
-
"learning_rate":
|
| 549 |
-
"loss": 0.
|
| 550 |
"step": 670
|
| 551 |
},
|
| 552 |
{
|
| 553 |
"epoch": 10.625,
|
| 554 |
-
"grad_norm": 0.
|
| 555 |
-
"learning_rate":
|
| 556 |
-
"loss": 0.
|
| 557 |
"step": 680
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 10.78125,
|
| 561 |
-
"grad_norm": 0.
|
| 562 |
-
"learning_rate":
|
| 563 |
-
"loss": 0.
|
| 564 |
"step": 690
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 10.9375,
|
| 568 |
-
"grad_norm": 0.
|
| 569 |
-
"learning_rate":
|
| 570 |
-
"loss": 0.
|
| 571 |
"step": 700
|
| 572 |
},
|
| 573 |
{
|
| 574 |
"epoch": 10.9375,
|
| 575 |
-
"eval_accuracy": 0.
|
| 576 |
-
"eval_f1": 0.
|
| 577 |
-
"eval_loss": 0.
|
| 578 |
-
"eval_precision": 0.
|
| 579 |
-
"eval_recall": 0.
|
| 580 |
-
"eval_runtime":
|
| 581 |
-
"eval_samples_per_second":
|
| 582 |
-
"eval_steps_per_second":
|
| 583 |
"step": 700
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 11.09375,
|
| 587 |
-
"grad_norm": 0.
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 710
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 11.25,
|
| 594 |
-
"grad_norm": 0.
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 720
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 11.40625,
|
| 601 |
-
"grad_norm":
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 730
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 11.5625,
|
| 608 |
-
"grad_norm": 0.
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 740
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 11.71875,
|
| 615 |
-
"grad_norm": 0.
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 750
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 11.875,
|
| 622 |
-
"grad_norm": 0.
|
| 623 |
-
"learning_rate":
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 760
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 12.03125,
|
| 629 |
-
"grad_norm": 0.
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 770
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 12.1875,
|
| 636 |
-
"grad_norm": 0.
|
| 637 |
-
"learning_rate":
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 780
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 12.34375,
|
| 643 |
-
"grad_norm": 0.
|
| 644 |
-
"learning_rate":
|
| 645 |
-
"loss": 0.
|
| 646 |
"step": 790
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 12.5,
|
| 650 |
-
"grad_norm": 0.
|
| 651 |
-
"learning_rate":
|
| 652 |
-
"loss": 0.
|
| 653 |
"step": 800
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 12.5,
|
| 657 |
-
"eval_accuracy": 0.
|
| 658 |
-
"eval_f1": 0.
|
| 659 |
-
"eval_loss": 0.
|
| 660 |
-
"eval_precision": 0.
|
| 661 |
-
"eval_recall": 0.
|
| 662 |
-
"eval_runtime": 0.
|
| 663 |
-
"eval_samples_per_second":
|
| 664 |
-
"eval_steps_per_second": 16.
|
| 665 |
"step": 800
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 12.65625,
|
| 669 |
-
"grad_norm": 0.
|
| 670 |
-
"learning_rate":
|
| 671 |
-
"loss": 0.
|
| 672 |
"step": 810
|
| 673 |
},
|
| 674 |
{
|
| 675 |
"epoch": 12.8125,
|
| 676 |
-
"grad_norm": 0.
|
| 677 |
-
"learning_rate":
|
| 678 |
-
"loss": 0.
|
| 679 |
"step": 820
|
| 680 |
},
|
| 681 |
{
|
| 682 |
"epoch": 12.96875,
|
| 683 |
-
"grad_norm": 0.
|
| 684 |
-
"learning_rate":
|
| 685 |
-
"loss": 0.
|
| 686 |
"step": 830
|
| 687 |
},
|
| 688 |
{
|
| 689 |
"epoch": 13.125,
|
| 690 |
-
"grad_norm": 0.
|
| 691 |
-
"learning_rate":
|
| 692 |
-
"loss": 0.
|
| 693 |
"step": 840
|
| 694 |
},
|
| 695 |
{
|
| 696 |
"epoch": 13.28125,
|
| 697 |
-
"grad_norm": 0.
|
| 698 |
-
"learning_rate":
|
| 699 |
-
"loss": 0.
|
| 700 |
"step": 850
|
| 701 |
},
|
| 702 |
{
|
| 703 |
"epoch": 13.4375,
|
| 704 |
-
"grad_norm": 0.
|
| 705 |
-
"learning_rate":
|
| 706 |
-
"loss": 0.
|
| 707 |
"step": 860
|
| 708 |
},
|
| 709 |
{
|
| 710 |
"epoch": 13.59375,
|
| 711 |
-
"grad_norm": 0.
|
| 712 |
-
"learning_rate":
|
| 713 |
-
"loss": 0.
|
| 714 |
"step": 870
|
| 715 |
},
|
| 716 |
{
|
| 717 |
"epoch": 13.75,
|
| 718 |
-
"grad_norm": 0.
|
| 719 |
-
"learning_rate":
|
| 720 |
-
"loss": 0.
|
| 721 |
"step": 880
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 13.90625,
|
| 725 |
-
"grad_norm": 0.
|
| 726 |
-
"learning_rate":
|
| 727 |
-
"loss": 0.
|
| 728 |
"step": 890
|
| 729 |
},
|
| 730 |
{
|
| 731 |
"epoch": 14.0625,
|
| 732 |
-
"grad_norm": 0.
|
| 733 |
-
"learning_rate":
|
| 734 |
-
"loss": 0.
|
| 735 |
"step": 900
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 14.0625,
|
| 739 |
-
"eval_accuracy": 0.
|
| 740 |
-
"eval_f1": 0.
|
| 741 |
-
"eval_loss": 0.
|
| 742 |
-
"eval_precision": 0.
|
| 743 |
-
"eval_recall": 0.
|
| 744 |
-
"eval_runtime":
|
| 745 |
-
"eval_samples_per_second":
|
| 746 |
-
"eval_steps_per_second":
|
| 747 |
"step": 900
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 14.21875,
|
| 751 |
-
"grad_norm": 0.
|
| 752 |
-
"learning_rate":
|
| 753 |
-
"loss": 0.
|
| 754 |
"step": 910
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 14.375,
|
| 758 |
-
"grad_norm": 0.
|
| 759 |
-
"learning_rate":
|
| 760 |
-
"loss": 0.
|
| 761 |
"step": 920
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 14.53125,
|
| 765 |
-
"grad_norm": 0.
|
| 766 |
-
"learning_rate":
|
| 767 |
-
"loss": 0.
|
| 768 |
"step": 930
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 14.6875,
|
| 772 |
-
"grad_norm": 0.
|
| 773 |
-
"learning_rate":
|
| 774 |
-
"loss": 0.
|
| 775 |
"step": 940
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 14.84375,
|
| 779 |
-
"grad_norm": 0.
|
| 780 |
-
"learning_rate":
|
| 781 |
-
"loss": 0.
|
| 782 |
"step": 950
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 15.0,
|
| 786 |
-
"grad_norm": 0.
|
| 787 |
-
"learning_rate":
|
| 788 |
-
"loss": 0.
|
| 789 |
"step": 960
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 15.15625,
|
| 793 |
-
"grad_norm": 0.
|
| 794 |
-
"learning_rate":
|
| 795 |
-
"loss": 0.
|
| 796 |
"step": 970
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 15.3125,
|
| 800 |
-
"grad_norm": 0.
|
| 801 |
-
"learning_rate":
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 980
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 15.46875,
|
| 807 |
-
"grad_norm": 0.
|
| 808 |
-
"learning_rate":
|
| 809 |
-
"loss": 0.
|
| 810 |
"step": 990
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 15.625,
|
| 814 |
-
"grad_norm": 0.
|
| 815 |
-
"learning_rate":
|
| 816 |
-
"loss": 0.
|
| 817 |
"step": 1000
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 15.625,
|
| 821 |
-
"eval_accuracy": 0.
|
| 822 |
-
"eval_f1": 0.
|
| 823 |
-
"eval_loss": 0.
|
| 824 |
-
"eval_precision": 0.
|
| 825 |
-
"eval_recall": 0.
|
| 826 |
-
"eval_runtime":
|
| 827 |
-
"eval_samples_per_second":
|
| 828 |
-
"eval_steps_per_second":
|
| 829 |
"step": 1000
|
| 830 |
},
|
| 831 |
{
|
| 832 |
"epoch": 15.78125,
|
| 833 |
-
"grad_norm": 0.
|
| 834 |
-
"learning_rate":
|
| 835 |
-
"loss": 0.
|
| 836 |
"step": 1010
|
| 837 |
},
|
| 838 |
{
|
| 839 |
"epoch": 15.9375,
|
| 840 |
-
"grad_norm": 0.
|
| 841 |
-
"learning_rate":
|
| 842 |
-
"loss": 0.
|
| 843 |
"step": 1020
|
| 844 |
},
|
| 845 |
{
|
| 846 |
"epoch": 16.09375,
|
| 847 |
-
"grad_norm": 0.
|
| 848 |
-
"learning_rate":
|
| 849 |
-
"loss": 0.
|
| 850 |
"step": 1030
|
| 851 |
},
|
| 852 |
{
|
| 853 |
"epoch": 16.25,
|
| 854 |
-
"grad_norm": 0.
|
| 855 |
-
"learning_rate":
|
| 856 |
-
"loss": 0.
|
| 857 |
"step": 1040
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 16.40625,
|
| 861 |
-
"grad_norm": 0.
|
| 862 |
-
"learning_rate":
|
| 863 |
-
"loss": 0.
|
| 864 |
"step": 1050
|
| 865 |
},
|
| 866 |
{
|
| 867 |
"epoch": 16.5625,
|
| 868 |
-
"grad_norm":
|
| 869 |
-
"learning_rate":
|
| 870 |
-
"loss": 0.
|
| 871 |
"step": 1060
|
| 872 |
},
|
| 873 |
{
|
| 874 |
"epoch": 16.71875,
|
| 875 |
-
"grad_norm": 0.
|
| 876 |
-
"learning_rate":
|
| 877 |
-
"loss": 0.
|
| 878 |
"step": 1070
|
| 879 |
},
|
| 880 |
{
|
| 881 |
"epoch": 16.875,
|
| 882 |
-
"grad_norm": 0.
|
| 883 |
-
"learning_rate":
|
| 884 |
-
"loss": 0.
|
| 885 |
"step": 1080
|
| 886 |
},
|
| 887 |
{
|
| 888 |
-
"epoch": 17.
|
| 889 |
-
"
|
| 890 |
-
"
|
| 891 |
-
"
|
| 892 |
-
"
|
| 893 |
-
|
| 894 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 895 |
}
|
| 896 |
],
|
| 897 |
"logging_steps": 10,
|
| 898 |
-
"max_steps":
|
| 899 |
"num_input_tokens_seen": 0,
|
| 900 |
-
"num_train_epochs":
|
| 901 |
"save_steps": 500,
|
| 902 |
"stateful_callbacks": {
|
| 903 |
"TrainerControl": {
|
|
@@ -911,7 +1607,7 @@
|
|
| 911 |
"attributes": {}
|
| 912 |
}
|
| 913 |
},
|
| 914 |
-
"total_flos":
|
| 915 |
"train_batch_size": 16,
|
| 916 |
"trial_name": null,
|
| 917 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.09645664691925049,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned_v2024-7-24-frost/checkpoint-500",
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 1920,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.15625,
|
| 13 |
+
"grad_norm": 0.3123115003108978,
|
| 14 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 15 |
+
"loss": 0.0743,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.3125,
|
| 20 |
+
"grad_norm": 0.10650705546140671,
|
| 21 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 22 |
+
"loss": 0.0699,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.46875,
|
| 27 |
+
"grad_norm": 0.40636828541755676,
|
| 28 |
+
"learning_rate": 3.125e-05,
|
| 29 |
+
"loss": 0.0732,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.625,
|
| 34 |
+
"grad_norm": 0.42912840843200684,
|
| 35 |
+
"learning_rate": 4.166666666666667e-05,
|
| 36 |
+
"loss": 0.075,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.78125,
|
| 41 |
+
"grad_norm": 0.3166373372077942,
|
| 42 |
+
"learning_rate": 5.208333333333334e-05,
|
| 43 |
+
"loss": 0.0695,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.9375,
|
| 48 |
+
"grad_norm": 0.8551476001739502,
|
| 49 |
+
"learning_rate": 6.25e-05,
|
| 50 |
+
"loss": 0.0883,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 1.09375,
|
| 55 |
+
"grad_norm": 0.18066875636577606,
|
| 56 |
+
"learning_rate": 7.291666666666667e-05,
|
| 57 |
+
"loss": 0.0699,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 1.25,
|
| 62 |
+
"grad_norm": 0.28325945138931274,
|
| 63 |
+
"learning_rate": 8.333333333333334e-05,
|
| 64 |
+
"loss": 0.0627,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 1.40625,
|
| 69 |
+
"grad_norm": 0.3701513409614563,
|
| 70 |
+
"learning_rate": 9.375e-05,
|
| 71 |
+
"loss": 0.0866,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 1.5625,
|
| 76 |
+
"grad_norm": 0.35587912797927856,
|
| 77 |
+
"learning_rate": 0.00010416666666666667,
|
| 78 |
+
"loss": 0.0728,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 1.5625,
|
| 83 |
+
"eval_accuracy": 0.984070796460177,
|
| 84 |
+
"eval_f1": 0.9606986899563319,
|
| 85 |
+
"eval_loss": 0.06593623757362366,
|
| 86 |
+
"eval_precision": 0.9691629955947136,
|
| 87 |
+
"eval_recall": 0.9523809523809523,
|
| 88 |
+
"eval_runtime": 0.9039,
|
| 89 |
+
"eval_samples_per_second": 125.009,
|
| 90 |
+
"eval_steps_per_second": 16.594,
|
| 91 |
"step": 100
|
| 92 |
},
|
| 93 |
{
|
| 94 |
"epoch": 1.71875,
|
| 95 |
+
"grad_norm": 0.23759329319000244,
|
| 96 |
+
"learning_rate": 0.00011458333333333333,
|
| 97 |
+
"loss": 0.0653,
|
| 98 |
"step": 110
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 1.875,
|
| 102 |
+
"grad_norm": 0.6092272996902466,
|
| 103 |
+
"learning_rate": 0.000125,
|
| 104 |
+
"loss": 0.1015,
|
| 105 |
"step": 120
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 2.03125,
|
| 109 |
+
"grad_norm": 0.15887708961963654,
|
| 110 |
+
"learning_rate": 0.0001354166666666667,
|
| 111 |
+
"loss": 0.09,
|
| 112 |
"step": 130
|
| 113 |
},
|
| 114 |
{
|
| 115 |
"epoch": 2.1875,
|
| 116 |
+
"grad_norm": 0.3399417996406555,
|
| 117 |
+
"learning_rate": 0.00014583333333333335,
|
| 118 |
+
"loss": 0.0847,
|
| 119 |
"step": 140
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 2.34375,
|
| 123 |
+
"grad_norm": 0.2599344253540039,
|
| 124 |
+
"learning_rate": 0.00015625,
|
| 125 |
+
"loss": 0.0722,
|
| 126 |
"step": 150
|
| 127 |
},
|
| 128 |
{
|
| 129 |
"epoch": 2.5,
|
| 130 |
+
"grad_norm": 0.20714014768600464,
|
| 131 |
+
"learning_rate": 0.0001666666666666667,
|
| 132 |
+
"loss": 0.0915,
|
| 133 |
"step": 160
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"epoch": 2.65625,
|
| 137 |
+
"grad_norm": 0.7900287508964539,
|
| 138 |
+
"learning_rate": 0.00017708333333333335,
|
| 139 |
+
"loss": 0.1008,
|
| 140 |
"step": 170
|
| 141 |
},
|
| 142 |
{
|
| 143 |
"epoch": 2.8125,
|
| 144 |
+
"grad_norm": 0.23315797746181488,
|
| 145 |
+
"learning_rate": 0.0001875,
|
| 146 |
+
"loss": 0.1142,
|
| 147 |
"step": 180
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 2.96875,
|
| 151 |
+
"grad_norm": 1.258319616317749,
|
| 152 |
+
"learning_rate": 0.0001979166666666667,
|
| 153 |
+
"loss": 0.1027,
|
| 154 |
"step": 190
|
| 155 |
},
|
| 156 |
{
|
| 157 |
"epoch": 3.125,
|
| 158 |
+
"grad_norm": 1.0168662071228027,
|
| 159 |
+
"learning_rate": 0.0001990740740740741,
|
| 160 |
+
"loss": 0.0871,
|
| 161 |
"step": 200
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 3.125,
|
| 165 |
+
"eval_accuracy": 0.9566371681415929,
|
| 166 |
+
"eval_f1": 0.8941684665226782,
|
| 167 |
+
"eval_loss": 0.12436065077781677,
|
| 168 |
+
"eval_precision": 0.8922413793103449,
|
| 169 |
+
"eval_recall": 0.8961038961038961,
|
| 170 |
+
"eval_runtime": 0.8601,
|
| 171 |
+
"eval_samples_per_second": 131.383,
|
| 172 |
+
"eval_steps_per_second": 17.44,
|
| 173 |
"step": 200
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"epoch": 3.28125,
|
| 177 |
+
"grad_norm": 0.38566353917121887,
|
| 178 |
+
"learning_rate": 0.0001979166666666667,
|
| 179 |
+
"loss": 0.1166,
|
| 180 |
"step": 210
|
| 181 |
},
|
| 182 |
{
|
| 183 |
"epoch": 3.4375,
|
| 184 |
+
"grad_norm": 0.4687894284725189,
|
| 185 |
+
"learning_rate": 0.00019675925925925926,
|
| 186 |
+
"loss": 0.108,
|
| 187 |
"step": 220
|
| 188 |
},
|
| 189 |
{
|
| 190 |
"epoch": 3.59375,
|
| 191 |
+
"grad_norm": 0.5190223455429077,
|
| 192 |
+
"learning_rate": 0.00019560185185185186,
|
| 193 |
+
"loss": 0.0901,
|
| 194 |
"step": 230
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 3.75,
|
| 198 |
+
"grad_norm": 0.5094243288040161,
|
| 199 |
+
"learning_rate": 0.00019444444444444446,
|
| 200 |
+
"loss": 0.1144,
|
| 201 |
"step": 240
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 3.90625,
|
| 205 |
+
"grad_norm": 0.5921277403831482,
|
| 206 |
+
"learning_rate": 0.00019328703703703706,
|
| 207 |
+
"loss": 0.1196,
|
| 208 |
"step": 250
|
| 209 |
},
|
| 210 |
{
|
| 211 |
"epoch": 4.0625,
|
| 212 |
+
"grad_norm": 0.23840609192848206,
|
| 213 |
+
"learning_rate": 0.00019212962962962963,
|
| 214 |
+
"loss": 0.1122,
|
| 215 |
"step": 260
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 4.21875,
|
| 219 |
+
"grad_norm": 0.9276812672615051,
|
| 220 |
+
"learning_rate": 0.00019097222222222223,
|
| 221 |
+
"loss": 0.1147,
|
| 222 |
"step": 270
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 4.375,
|
| 226 |
+
"grad_norm": 0.7325614094734192,
|
| 227 |
+
"learning_rate": 0.00018981481481481483,
|
| 228 |
+
"loss": 0.1075,
|
| 229 |
"step": 280
|
| 230 |
},
|
| 231 |
{
|
| 232 |
"epoch": 4.53125,
|
| 233 |
+
"grad_norm": 0.5574468374252319,
|
| 234 |
+
"learning_rate": 0.00018865740740740743,
|
| 235 |
+
"loss": 0.0958,
|
| 236 |
"step": 290
|
| 237 |
},
|
| 238 |
{
|
| 239 |
"epoch": 4.6875,
|
| 240 |
+
"grad_norm": 0.3893429934978485,
|
| 241 |
+
"learning_rate": 0.0001875,
|
| 242 |
+
"loss": 0.0999,
|
| 243 |
"step": 300
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 4.6875,
|
| 247 |
+
"eval_accuracy": 0.963716814159292,
|
| 248 |
+
"eval_f1": 0.9125799573560768,
|
| 249 |
+
"eval_loss": 0.10427873581647873,
|
| 250 |
+
"eval_precision": 0.8991596638655462,
|
| 251 |
+
"eval_recall": 0.9264069264069265,
|
| 252 |
+
"eval_runtime": 0.8493,
|
| 253 |
+
"eval_samples_per_second": 133.049,
|
| 254 |
+
"eval_steps_per_second": 17.661,
|
| 255 |
"step": 300
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 4.84375,
|
| 259 |
+
"grad_norm": 0.49028488993644714,
|
| 260 |
+
"learning_rate": 0.0001863425925925926,
|
| 261 |
+
"loss": 0.1087,
|
| 262 |
"step": 310
|
| 263 |
},
|
| 264 |
{
|
| 265 |
"epoch": 5.0,
|
| 266 |
+
"grad_norm": 0.6510241627693176,
|
| 267 |
+
"learning_rate": 0.0001851851851851852,
|
| 268 |
+
"loss": 0.0949,
|
| 269 |
"step": 320
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 5.15625,
|
| 273 |
+
"grad_norm": 0.3244408667087555,
|
| 274 |
+
"learning_rate": 0.00018402777777777778,
|
| 275 |
+
"loss": 0.0957,
|
| 276 |
"step": 330
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 5.3125,
|
| 280 |
+
"grad_norm": 0.32894158363342285,
|
| 281 |
+
"learning_rate": 0.00018287037037037038,
|
| 282 |
+
"loss": 0.0761,
|
| 283 |
"step": 340
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 5.46875,
|
| 287 |
+
"grad_norm": 0.4168912470340729,
|
| 288 |
+
"learning_rate": 0.00018171296296296297,
|
| 289 |
+
"loss": 0.1014,
|
| 290 |
"step": 350
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 5.625,
|
| 294 |
+
"grad_norm": 0.30746978521347046,
|
| 295 |
+
"learning_rate": 0.00018055555555555557,
|
| 296 |
+
"loss": 0.0789,
|
| 297 |
"step": 360
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 5.78125,
|
| 301 |
+
"grad_norm": 0.3337535262107849,
|
| 302 |
+
"learning_rate": 0.00017939814814814815,
|
| 303 |
+
"loss": 0.0891,
|
| 304 |
"step": 370
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 5.9375,
|
| 308 |
+
"grad_norm": 0.2659320533275604,
|
| 309 |
+
"learning_rate": 0.00017824074074074075,
|
| 310 |
+
"loss": 0.0798,
|
| 311 |
"step": 380
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 6.09375,
|
| 315 |
+
"grad_norm": 0.28791913390159607,
|
| 316 |
+
"learning_rate": 0.00017708333333333335,
|
| 317 |
+
"loss": 0.0961,
|
| 318 |
"step": 390
|
| 319 |
},
|
| 320 |
{
|
| 321 |
"epoch": 6.25,
|
| 322 |
+
"grad_norm": 0.41803187131881714,
|
| 323 |
+
"learning_rate": 0.00017592592592592595,
|
| 324 |
+
"loss": 0.0743,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 6.25,
|
| 329 |
+
"eval_accuracy": 0.9610619469026549,
|
| 330 |
+
"eval_f1": 0.9043478260869565,
|
| 331 |
+
"eval_loss": 0.10431604832410812,
|
| 332 |
+
"eval_precision": 0.9082969432314411,
|
| 333 |
+
"eval_recall": 0.9004329004329005,
|
| 334 |
+
"eval_runtime": 1.3126,
|
| 335 |
+
"eval_samples_per_second": 86.086,
|
| 336 |
+
"eval_steps_per_second": 11.427,
|
| 337 |
"step": 400
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 6.40625,
|
| 341 |
+
"grad_norm": 0.398034930229187,
|
| 342 |
+
"learning_rate": 0.00017476851851851852,
|
| 343 |
+
"loss": 0.0798,
|
| 344 |
"step": 410
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 6.5625,
|
| 348 |
+
"grad_norm": 0.533364474773407,
|
| 349 |
+
"learning_rate": 0.00017361111111111112,
|
| 350 |
+
"loss": 0.0808,
|
| 351 |
"step": 420
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 6.71875,
|
| 355 |
+
"grad_norm": 0.6189862489700317,
|
| 356 |
+
"learning_rate": 0.00017245370370370372,
|
| 357 |
+
"loss": 0.091,
|
| 358 |
"step": 430
|
| 359 |
},
|
| 360 |
{
|
| 361 |
"epoch": 6.875,
|
| 362 |
+
"grad_norm": 0.31593209505081177,
|
| 363 |
+
"learning_rate": 0.00017129629629629632,
|
| 364 |
+
"loss": 0.0729,
|
| 365 |
"step": 440
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 7.03125,
|
| 369 |
+
"grad_norm": 0.3167741000652313,
|
| 370 |
+
"learning_rate": 0.0001701388888888889,
|
| 371 |
+
"loss": 0.0796,
|
| 372 |
"step": 450
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 7.1875,
|
| 376 |
+
"grad_norm": 0.30901169776916504,
|
| 377 |
+
"learning_rate": 0.0001689814814814815,
|
| 378 |
+
"loss": 0.0867,
|
| 379 |
"step": 460
|
| 380 |
},
|
| 381 |
{
|
| 382 |
"epoch": 7.34375,
|
| 383 |
+
"grad_norm": 0.13378705084323883,
|
| 384 |
+
"learning_rate": 0.0001678240740740741,
|
| 385 |
+
"loss": 0.0701,
|
| 386 |
"step": 470
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 7.5,
|
| 390 |
+
"grad_norm": 0.15507709980010986,
|
| 391 |
+
"learning_rate": 0.0001666666666666667,
|
| 392 |
+
"loss": 0.0789,
|
| 393 |
"step": 480
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 7.65625,
|
| 397 |
+
"grad_norm": 0.21113860607147217,
|
| 398 |
+
"learning_rate": 0.00016550925925925926,
|
| 399 |
+
"loss": 0.0647,
|
| 400 |
"step": 490
|
| 401 |
},
|
| 402 |
{
|
| 403 |
"epoch": 7.8125,
|
| 404 |
+
"grad_norm": 0.15848499536514282,
|
| 405 |
+
"learning_rate": 0.00016435185185185186,
|
| 406 |
+
"loss": 0.0655,
|
| 407 |
"step": 500
|
| 408 |
},
|
| 409 |
{
|
| 410 |
"epoch": 7.8125,
|
| 411 |
+
"eval_accuracy": 0.963716814159292,
|
| 412 |
+
"eval_f1": 0.9118279569892475,
|
| 413 |
+
"eval_loss": 0.09645664691925049,
|
| 414 |
+
"eval_precision": 0.905982905982906,
|
| 415 |
+
"eval_recall": 0.9177489177489178,
|
| 416 |
+
"eval_runtime": 2.2304,
|
| 417 |
+
"eval_samples_per_second": 50.664,
|
| 418 |
+
"eval_steps_per_second": 6.725,
|
| 419 |
"step": 500
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 7.96875,
|
| 423 |
+
"grad_norm": 0.19086486101150513,
|
| 424 |
+
"learning_rate": 0.00016319444444444446,
|
| 425 |
+
"loss": 0.0502,
|
| 426 |
"step": 510
|
| 427 |
},
|
| 428 |
{
|
| 429 |
"epoch": 8.125,
|
| 430 |
+
"grad_norm": 0.4851354956626892,
|
| 431 |
+
"learning_rate": 0.00016203703703703706,
|
| 432 |
+
"loss": 0.0646,
|
| 433 |
"step": 520
|
| 434 |
},
|
| 435 |
{
|
| 436 |
"epoch": 8.28125,
|
| 437 |
+
"grad_norm": 0.43803560733795166,
|
| 438 |
+
"learning_rate": 0.00016087962962962963,
|
| 439 |
+
"loss": 0.0668,
|
| 440 |
"step": 530
|
| 441 |
},
|
| 442 |
{
|
| 443 |
"epoch": 8.4375,
|
| 444 |
+
"grad_norm": 0.26552197337150574,
|
| 445 |
+
"learning_rate": 0.00015972222222222223,
|
| 446 |
+
"loss": 0.0549,
|
| 447 |
"step": 540
|
| 448 |
},
|
| 449 |
{
|
| 450 |
"epoch": 8.59375,
|
| 451 |
+
"grad_norm": 0.18909405171871185,
|
| 452 |
+
"learning_rate": 0.00015856481481481483,
|
| 453 |
+
"loss": 0.07,
|
| 454 |
"step": 550
|
| 455 |
},
|
| 456 |
{
|
| 457 |
"epoch": 8.75,
|
| 458 |
+
"grad_norm": 0.2485276311635971,
|
| 459 |
+
"learning_rate": 0.00015740740740740743,
|
| 460 |
+
"loss": 0.0525,
|
| 461 |
"step": 560
|
| 462 |
},
|
| 463 |
{
|
| 464 |
"epoch": 8.90625,
|
| 465 |
+
"grad_norm": 0.4424391984939575,
|
| 466 |
+
"learning_rate": 0.00015625,
|
| 467 |
+
"loss": 0.0775,
|
| 468 |
"step": 570
|
| 469 |
},
|
| 470 |
{
|
| 471 |
"epoch": 9.0625,
|
| 472 |
+
"grad_norm": 0.2584344446659088,
|
| 473 |
+
"learning_rate": 0.0001550925925925926,
|
| 474 |
+
"loss": 0.064,
|
| 475 |
"step": 580
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 9.21875,
|
| 479 |
+
"grad_norm": 0.6115286946296692,
|
| 480 |
+
"learning_rate": 0.0001539351851851852,
|
| 481 |
+
"loss": 0.0599,
|
| 482 |
"step": 590
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 9.375,
|
| 486 |
+
"grad_norm": 0.5155323147773743,
|
| 487 |
+
"learning_rate": 0.00015277777777777777,
|
| 488 |
+
"loss": 0.0559,
|
| 489 |
"step": 600
|
| 490 |
},
|
| 491 |
{
|
| 492 |
"epoch": 9.375,
|
| 493 |
+
"eval_accuracy": 0.9619469026548673,
|
| 494 |
+
"eval_f1": 0.9087048832271762,
|
| 495 |
+
"eval_loss": 0.10384609550237656,
|
| 496 |
+
"eval_precision": 0.8916666666666667,
|
| 497 |
+
"eval_recall": 0.9264069264069265,
|
| 498 |
+
"eval_runtime": 2.2545,
|
| 499 |
+
"eval_samples_per_second": 50.123,
|
| 500 |
+
"eval_steps_per_second": 6.653,
|
| 501 |
"step": 600
|
| 502 |
},
|
| 503 |
{
|
| 504 |
"epoch": 9.53125,
|
| 505 |
+
"grad_norm": 0.40844494104385376,
|
| 506 |
+
"learning_rate": 0.00015162037037037037,
|
| 507 |
+
"loss": 0.072,
|
| 508 |
"step": 610
|
| 509 |
},
|
| 510 |
{
|
| 511 |
"epoch": 9.6875,
|
| 512 |
+
"grad_norm": 0.2091340720653534,
|
| 513 |
+
"learning_rate": 0.00015046296296296297,
|
| 514 |
+
"loss": 0.0513,
|
| 515 |
"step": 620
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 9.84375,
|
| 519 |
+
"grad_norm": 0.22117160260677338,
|
| 520 |
+
"learning_rate": 0.00014930555555555557,
|
| 521 |
+
"loss": 0.0623,
|
| 522 |
"step": 630
|
| 523 |
},
|
| 524 |
{
|
| 525 |
"epoch": 10.0,
|
| 526 |
+
"grad_norm": 0.31236401200294495,
|
| 527 |
+
"learning_rate": 0.00014814814814814815,
|
| 528 |
+
"loss": 0.0534,
|
| 529 |
"step": 640
|
| 530 |
},
|
| 531 |
{
|
| 532 |
"epoch": 10.15625,
|
| 533 |
+
"grad_norm": 0.31281912326812744,
|
| 534 |
+
"learning_rate": 0.00014699074074074075,
|
| 535 |
+
"loss": 0.0644,
|
| 536 |
"step": 650
|
| 537 |
},
|
| 538 |
{
|
| 539 |
"epoch": 10.3125,
|
| 540 |
+
"grad_norm": 0.5201927423477173,
|
| 541 |
+
"learning_rate": 0.00014583333333333335,
|
| 542 |
+
"loss": 0.057,
|
| 543 |
"step": 660
|
| 544 |
},
|
| 545 |
{
|
| 546 |
"epoch": 10.46875,
|
| 547 |
+
"grad_norm": 0.2596763074398041,
|
| 548 |
+
"learning_rate": 0.00014467592592592594,
|
| 549 |
+
"loss": 0.0542,
|
| 550 |
"step": 670
|
| 551 |
},
|
| 552 |
{
|
| 553 |
"epoch": 10.625,
|
| 554 |
+
"grad_norm": 0.3063810467720032,
|
| 555 |
+
"learning_rate": 0.00014351851851851852,
|
| 556 |
+
"loss": 0.0389,
|
| 557 |
"step": 680
|
| 558 |
},
|
| 559 |
{
|
| 560 |
"epoch": 10.78125,
|
| 561 |
+
"grad_norm": 0.48713332414627075,
|
| 562 |
+
"learning_rate": 0.00014236111111111112,
|
| 563 |
+
"loss": 0.0742,
|
| 564 |
"step": 690
|
| 565 |
},
|
| 566 |
{
|
| 567 |
"epoch": 10.9375,
|
| 568 |
+
"grad_norm": 0.21316884458065033,
|
| 569 |
+
"learning_rate": 0.00014120370370370372,
|
| 570 |
+
"loss": 0.0517,
|
| 571 |
"step": 700
|
| 572 |
},
|
| 573 |
{
|
| 574 |
"epoch": 10.9375,
|
| 575 |
+
"eval_accuracy": 0.9584070796460177,
|
| 576 |
+
"eval_f1": 0.8997867803837952,
|
| 577 |
+
"eval_loss": 0.09719711542129517,
|
| 578 |
+
"eval_precision": 0.8865546218487395,
|
| 579 |
+
"eval_recall": 0.9134199134199135,
|
| 580 |
+
"eval_runtime": 0.8764,
|
| 581 |
+
"eval_samples_per_second": 128.937,
|
| 582 |
+
"eval_steps_per_second": 17.115,
|
| 583 |
"step": 700
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 11.09375,
|
| 587 |
+
"grad_norm": 0.9216361045837402,
|
| 588 |
+
"learning_rate": 0.00014004629629629632,
|
| 589 |
+
"loss": 0.0623,
|
| 590 |
"step": 710
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 11.25,
|
| 594 |
+
"grad_norm": 0.31130528450012207,
|
| 595 |
+
"learning_rate": 0.0001388888888888889,
|
| 596 |
+
"loss": 0.0641,
|
| 597 |
"step": 720
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 11.40625,
|
| 601 |
+
"grad_norm": 1.03948974609375,
|
| 602 |
+
"learning_rate": 0.0001377314814814815,
|
| 603 |
+
"loss": 0.0594,
|
| 604 |
"step": 730
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 11.5625,
|
| 608 |
+
"grad_norm": 0.12757237255573273,
|
| 609 |
+
"learning_rate": 0.0001365740740740741,
|
| 610 |
+
"loss": 0.0572,
|
| 611 |
"step": 740
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 11.71875,
|
| 615 |
+
"grad_norm": 0.25488346815109253,
|
| 616 |
+
"learning_rate": 0.0001354166666666667,
|
| 617 |
+
"loss": 0.0533,
|
| 618 |
"step": 750
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 11.875,
|
| 622 |
+
"grad_norm": 0.2517576813697815,
|
| 623 |
+
"learning_rate": 0.00013425925925925926,
|
| 624 |
+
"loss": 0.0557,
|
| 625 |
"step": 760
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 12.03125,
|
| 629 |
+
"grad_norm": 0.14332328736782074,
|
| 630 |
+
"learning_rate": 0.00013310185185185186,
|
| 631 |
+
"loss": 0.0433,
|
| 632 |
"step": 770
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 12.1875,
|
| 636 |
+
"grad_norm": 0.7062014937400818,
|
| 637 |
+
"learning_rate": 0.00013194444444444446,
|
| 638 |
+
"loss": 0.0569,
|
| 639 |
"step": 780
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 12.34375,
|
| 643 |
+
"grad_norm": 0.727057158946991,
|
| 644 |
+
"learning_rate": 0.00013078703703703706,
|
| 645 |
+
"loss": 0.0443,
|
| 646 |
"step": 790
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 12.5,
|
| 650 |
+
"grad_norm": 0.17331984639167786,
|
| 651 |
+
"learning_rate": 0.00012962962962962963,
|
| 652 |
+
"loss": 0.0407,
|
| 653 |
"step": 800
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 12.5,
|
| 657 |
+
"eval_accuracy": 0.963716814159292,
|
| 658 |
+
"eval_f1": 0.9110629067245118,
|
| 659 |
+
"eval_loss": 0.11198227852582932,
|
| 660 |
+
"eval_precision": 0.9130434782608695,
|
| 661 |
+
"eval_recall": 0.9090909090909091,
|
| 662 |
+
"eval_runtime": 0.9019,
|
| 663 |
+
"eval_samples_per_second": 125.294,
|
| 664 |
+
"eval_steps_per_second": 16.632,
|
| 665 |
"step": 800
|
| 666 |
},
|
| 667 |
{
|
| 668 |
"epoch": 12.65625,
|
| 669 |
+
"grad_norm": 0.104576975107193,
|
| 670 |
+
"learning_rate": 0.00012847222222222223,
|
| 671 |
+
"loss": 0.0465,
|
| 672 |
"step": 810
|
| 673 |
},
|
| 674 |
{
|
| 675 |
"epoch": 12.8125,
|
| 676 |
+
"grad_norm": 0.6552168726921082,
|
| 677 |
+
"learning_rate": 0.00012731481481481483,
|
| 678 |
+
"loss": 0.0536,
|
| 679 |
"step": 820
|
| 680 |
},
|
| 681 |
{
|
| 682 |
"epoch": 12.96875,
|
| 683 |
+
"grad_norm": 0.39452189207077026,
|
| 684 |
+
"learning_rate": 0.00012615740740740743,
|
| 685 |
+
"loss": 0.0514,
|
| 686 |
"step": 830
|
| 687 |
},
|
| 688 |
{
|
| 689 |
"epoch": 13.125,
|
| 690 |
+
"grad_norm": 0.16756129264831543,
|
| 691 |
+
"learning_rate": 0.000125,
|
| 692 |
+
"loss": 0.0417,
|
| 693 |
"step": 840
|
| 694 |
},
|
| 695 |
{
|
| 696 |
"epoch": 13.28125,
|
| 697 |
+
"grad_norm": 0.13866697251796722,
|
| 698 |
+
"learning_rate": 0.00012384259259259258,
|
| 699 |
+
"loss": 0.0419,
|
| 700 |
"step": 850
|
| 701 |
},
|
| 702 |
{
|
| 703 |
"epoch": 13.4375,
|
| 704 |
+
"grad_norm": 0.9053749442100525,
|
| 705 |
+
"learning_rate": 0.0001226851851851852,
|
| 706 |
+
"loss": 0.0548,
|
| 707 |
"step": 860
|
| 708 |
},
|
| 709 |
{
|
| 710 |
"epoch": 13.59375,
|
| 711 |
+
"grad_norm": 0.43149927258491516,
|
| 712 |
+
"learning_rate": 0.00012152777777777777,
|
| 713 |
+
"loss": 0.0503,
|
| 714 |
"step": 870
|
| 715 |
},
|
| 716 |
{
|
| 717 |
"epoch": 13.75,
|
| 718 |
+
"grad_norm": 0.49532395601272583,
|
| 719 |
+
"learning_rate": 0.00012037037037037037,
|
| 720 |
+
"loss": 0.0476,
|
| 721 |
"step": 880
|
| 722 |
},
|
| 723 |
{
|
| 724 |
"epoch": 13.90625,
|
| 725 |
+
"grad_norm": 0.12025842815637589,
|
| 726 |
+
"learning_rate": 0.00011921296296296296,
|
| 727 |
+
"loss": 0.049,
|
| 728 |
"step": 890
|
| 729 |
},
|
| 730 |
{
|
| 731 |
"epoch": 14.0625,
|
| 732 |
+
"grad_norm": 0.8570975065231323,
|
| 733 |
+
"learning_rate": 0.00011805555555555556,
|
| 734 |
+
"loss": 0.0513,
|
| 735 |
"step": 900
|
| 736 |
},
|
| 737 |
{
|
| 738 |
"epoch": 14.0625,
|
| 739 |
+
"eval_accuracy": 0.9557522123893806,
|
| 740 |
+
"eval_f1": 0.8893805309734513,
|
| 741 |
+
"eval_loss": 0.1092919260263443,
|
| 742 |
+
"eval_precision": 0.9095022624434389,
|
| 743 |
+
"eval_recall": 0.8701298701298701,
|
| 744 |
+
"eval_runtime": 1.2237,
|
| 745 |
+
"eval_samples_per_second": 92.344,
|
| 746 |
+
"eval_steps_per_second": 12.258,
|
| 747 |
"step": 900
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 14.21875,
|
| 751 |
+
"grad_norm": 0.4120664596557617,
|
| 752 |
+
"learning_rate": 0.00011689814814814815,
|
| 753 |
+
"loss": 0.0552,
|
| 754 |
"step": 910
|
| 755 |
},
|
| 756 |
{
|
| 757 |
"epoch": 14.375,
|
| 758 |
+
"grad_norm": 0.24265483021736145,
|
| 759 |
+
"learning_rate": 0.00011574074074074075,
|
| 760 |
+
"loss": 0.0434,
|
| 761 |
"step": 920
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 14.53125,
|
| 765 |
+
"grad_norm": 0.23618777096271515,
|
| 766 |
+
"learning_rate": 0.00011458333333333333,
|
| 767 |
+
"loss": 0.0482,
|
| 768 |
"step": 930
|
| 769 |
},
|
| 770 |
{
|
| 771 |
"epoch": 14.6875,
|
| 772 |
+
"grad_norm": 0.1366555392742157,
|
| 773 |
+
"learning_rate": 0.00011342592592592593,
|
| 774 |
+
"loss": 0.045,
|
| 775 |
"step": 940
|
| 776 |
},
|
| 777 |
{
|
| 778 |
"epoch": 14.84375,
|
| 779 |
+
"grad_norm": 0.1841152310371399,
|
| 780 |
+
"learning_rate": 0.00011226851851851852,
|
| 781 |
+
"loss": 0.0539,
|
| 782 |
"step": 950
|
| 783 |
},
|
| 784 |
{
|
| 785 |
"epoch": 15.0,
|
| 786 |
+
"grad_norm": 0.6849538087844849,
|
| 787 |
+
"learning_rate": 0.00011111111111111112,
|
| 788 |
+
"loss": 0.0363,
|
| 789 |
"step": 960
|
| 790 |
},
|
| 791 |
{
|
| 792 |
"epoch": 15.15625,
|
| 793 |
+
"grad_norm": 0.5442699790000916,
|
| 794 |
+
"learning_rate": 0.0001099537037037037,
|
| 795 |
+
"loss": 0.0372,
|
| 796 |
"step": 970
|
| 797 |
},
|
| 798 |
{
|
| 799 |
"epoch": 15.3125,
|
| 800 |
+
"grad_norm": 0.3825988173484802,
|
| 801 |
+
"learning_rate": 0.0001087962962962963,
|
| 802 |
+
"loss": 0.0405,
|
| 803 |
"step": 980
|
| 804 |
},
|
| 805 |
{
|
| 806 |
"epoch": 15.46875,
|
| 807 |
+
"grad_norm": 0.0459093414247036,
|
| 808 |
+
"learning_rate": 0.00010763888888888889,
|
| 809 |
+
"loss": 0.0386,
|
| 810 |
"step": 990
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 15.625,
|
| 814 |
+
"grad_norm": 0.2602522373199463,
|
| 815 |
+
"learning_rate": 0.00010648148148148149,
|
| 816 |
+
"loss": 0.0378,
|
| 817 |
"step": 1000
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 15.625,
|
| 821 |
+
"eval_accuracy": 0.9548672566371681,
|
| 822 |
+
"eval_f1": 0.8888888888888888,
|
| 823 |
+
"eval_loss": 0.11969945579767227,
|
| 824 |
+
"eval_precision": 0.8947368421052632,
|
| 825 |
+
"eval_recall": 0.8831168831168831,
|
| 826 |
+
"eval_runtime": 0.8768,
|
| 827 |
+
"eval_samples_per_second": 128.871,
|
| 828 |
+
"eval_steps_per_second": 17.107,
|
| 829 |
"step": 1000
|
| 830 |
},
|
| 831 |
{
|
| 832 |
"epoch": 15.78125,
|
| 833 |
+
"grad_norm": 0.07926033437252045,
|
| 834 |
+
"learning_rate": 0.00010532407407407407,
|
| 835 |
+
"loss": 0.0419,
|
| 836 |
"step": 1010
|
| 837 |
},
|
| 838 |
{
|
| 839 |
"epoch": 15.9375,
|
| 840 |
+
"grad_norm": 0.2084084302186966,
|
| 841 |
+
"learning_rate": 0.00010416666666666667,
|
| 842 |
+
"loss": 0.0336,
|
| 843 |
"step": 1020
|
| 844 |
},
|
| 845 |
{
|
| 846 |
"epoch": 16.09375,
|
| 847 |
+
"grad_norm": 0.11587415635585785,
|
| 848 |
+
"learning_rate": 0.00010300925925925926,
|
| 849 |
+
"loss": 0.0293,
|
| 850 |
"step": 1030
|
| 851 |
},
|
| 852 |
{
|
| 853 |
"epoch": 16.25,
|
| 854 |
+
"grad_norm": 0.4128260314464569,
|
| 855 |
+
"learning_rate": 0.00010185185185185186,
|
| 856 |
+
"loss": 0.0346,
|
| 857 |
"step": 1040
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 16.40625,
|
| 861 |
+
"grad_norm": 0.2051563411951065,
|
| 862 |
+
"learning_rate": 0.00010069444444444445,
|
| 863 |
+
"loss": 0.0404,
|
| 864 |
"step": 1050
|
| 865 |
},
|
| 866 |
{
|
| 867 |
"epoch": 16.5625,
|
| 868 |
+
"grad_norm": 1.0257600545883179,
|
| 869 |
+
"learning_rate": 9.953703703703704e-05,
|
| 870 |
+
"loss": 0.0521,
|
| 871 |
"step": 1060
|
| 872 |
},
|
| 873 |
{
|
| 874 |
"epoch": 16.71875,
|
| 875 |
+
"grad_norm": 0.13610199093818665,
|
| 876 |
+
"learning_rate": 9.837962962962963e-05,
|
| 877 |
+
"loss": 0.0513,
|
| 878 |
"step": 1070
|
| 879 |
},
|
| 880 |
{
|
| 881 |
"epoch": 16.875,
|
| 882 |
+
"grad_norm": 0.5424107909202576,
|
| 883 |
+
"learning_rate": 9.722222222222223e-05,
|
| 884 |
+
"loss": 0.0662,
|
| 885 |
"step": 1080
|
| 886 |
},
|
| 887 |
{
|
| 888 |
+
"epoch": 17.03125,
|
| 889 |
+
"grad_norm": 0.1417212188243866,
|
| 890 |
+
"learning_rate": 9.606481481481482e-05,
|
| 891 |
+
"loss": 0.0364,
|
| 892 |
+
"step": 1090
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 17.1875,
|
| 896 |
+
"grad_norm": 0.15864621102809906,
|
| 897 |
+
"learning_rate": 9.490740740740742e-05,
|
| 898 |
+
"loss": 0.0487,
|
| 899 |
+
"step": 1100
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 17.1875,
|
| 903 |
+
"eval_accuracy": 0.9646017699115044,
|
| 904 |
+
"eval_f1": 0.9137931034482759,
|
| 905 |
+
"eval_loss": 0.09552007168531418,
|
| 906 |
+
"eval_precision": 0.9098712446351931,
|
| 907 |
+
"eval_recall": 0.9177489177489178,
|
| 908 |
+
"eval_runtime": 0.8603,
|
| 909 |
+
"eval_samples_per_second": 131.352,
|
| 910 |
+
"eval_steps_per_second": 17.436,
|
| 911 |
+
"step": 1100
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 17.34375,
|
| 915 |
+
"grad_norm": 0.5382766127586365,
|
| 916 |
+
"learning_rate": 9.375e-05,
|
| 917 |
+
"loss": 0.0315,
|
| 918 |
+
"step": 1110
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 17.5,
|
| 922 |
+
"grad_norm": 0.16078180074691772,
|
| 923 |
+
"learning_rate": 9.25925925925926e-05,
|
| 924 |
+
"loss": 0.0245,
|
| 925 |
+
"step": 1120
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 17.65625,
|
| 929 |
+
"grad_norm": 0.05187100172042847,
|
| 930 |
+
"learning_rate": 9.143518518518519e-05,
|
| 931 |
+
"loss": 0.0342,
|
| 932 |
+
"step": 1130
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 17.8125,
|
| 936 |
+
"grad_norm": 0.4016551077365875,
|
| 937 |
+
"learning_rate": 9.027777777777779e-05,
|
| 938 |
+
"loss": 0.0285,
|
| 939 |
+
"step": 1140
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 17.96875,
|
| 943 |
+
"grad_norm": 0.22832362353801727,
|
| 944 |
+
"learning_rate": 8.912037037037037e-05,
|
| 945 |
+
"loss": 0.0379,
|
| 946 |
+
"step": 1150
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 18.125,
|
| 950 |
+
"grad_norm": 1.3720444440841675,
|
| 951 |
+
"learning_rate": 8.796296296296297e-05,
|
| 952 |
+
"loss": 0.0369,
|
| 953 |
+
"step": 1160
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 18.28125,
|
| 957 |
+
"grad_norm": 0.23146755993366241,
|
| 958 |
+
"learning_rate": 8.680555555555556e-05,
|
| 959 |
+
"loss": 0.0343,
|
| 960 |
+
"step": 1170
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 18.4375,
|
| 964 |
+
"grad_norm": 0.2672041356563568,
|
| 965 |
+
"learning_rate": 8.564814814814816e-05,
|
| 966 |
+
"loss": 0.0353,
|
| 967 |
+
"step": 1180
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 18.59375,
|
| 971 |
+
"grad_norm": 0.17212288081645966,
|
| 972 |
+
"learning_rate": 8.449074074074074e-05,
|
| 973 |
+
"loss": 0.0315,
|
| 974 |
+
"step": 1190
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 18.75,
|
| 978 |
+
"grad_norm": 0.12989170849323273,
|
| 979 |
+
"learning_rate": 8.333333333333334e-05,
|
| 980 |
+
"loss": 0.0272,
|
| 981 |
+
"step": 1200
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 18.75,
|
| 985 |
+
"eval_accuracy": 0.9566371681415929,
|
| 986 |
+
"eval_f1": 0.8927789934354485,
|
| 987 |
+
"eval_loss": 0.10875095427036285,
|
| 988 |
+
"eval_precision": 0.9026548672566371,
|
| 989 |
+
"eval_recall": 0.8831168831168831,
|
| 990 |
+
"eval_runtime": 1.2152,
|
| 991 |
+
"eval_samples_per_second": 92.985,
|
| 992 |
+
"eval_steps_per_second": 12.343,
|
| 993 |
+
"step": 1200
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 18.90625,
|
| 997 |
+
"grad_norm": 0.15251386165618896,
|
| 998 |
+
"learning_rate": 8.217592592592593e-05,
|
| 999 |
+
"loss": 0.0353,
|
| 1000 |
+
"step": 1210
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 19.0625,
|
| 1004 |
+
"grad_norm": 0.1500956416130066,
|
| 1005 |
+
"learning_rate": 8.101851851851853e-05,
|
| 1006 |
+
"loss": 0.0258,
|
| 1007 |
+
"step": 1220
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 19.21875,
|
| 1011 |
+
"grad_norm": 0.16236737370491028,
|
| 1012 |
+
"learning_rate": 7.986111111111112e-05,
|
| 1013 |
+
"loss": 0.0318,
|
| 1014 |
+
"step": 1230
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 19.375,
|
| 1018 |
+
"grad_norm": 0.5188699960708618,
|
| 1019 |
+
"learning_rate": 7.870370370370372e-05,
|
| 1020 |
+
"loss": 0.0388,
|
| 1021 |
+
"step": 1240
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 19.53125,
|
| 1025 |
+
"grad_norm": 0.14171747863292694,
|
| 1026 |
+
"learning_rate": 7.75462962962963e-05,
|
| 1027 |
+
"loss": 0.0371,
|
| 1028 |
+
"step": 1250
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 19.6875,
|
| 1032 |
+
"grad_norm": 0.355496883392334,
|
| 1033 |
+
"learning_rate": 7.638888888888889e-05,
|
| 1034 |
+
"loss": 0.0278,
|
| 1035 |
+
"step": 1260
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 19.84375,
|
| 1039 |
+
"grad_norm": 0.30447283387184143,
|
| 1040 |
+
"learning_rate": 7.523148148148149e-05,
|
| 1041 |
+
"loss": 0.0307,
|
| 1042 |
+
"step": 1270
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 20.0,
|
| 1046 |
+
"grad_norm": 1.8779629468917847,
|
| 1047 |
+
"learning_rate": 7.407407407407407e-05,
|
| 1048 |
+
"loss": 0.0378,
|
| 1049 |
+
"step": 1280
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 20.15625,
|
| 1053 |
+
"grad_norm": 0.24115116894245148,
|
| 1054 |
+
"learning_rate": 7.291666666666667e-05,
|
| 1055 |
+
"loss": 0.0234,
|
| 1056 |
+
"step": 1290
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 20.3125,
|
| 1060 |
+
"grad_norm": 0.07465353608131409,
|
| 1061 |
+
"learning_rate": 7.175925925925926e-05,
|
| 1062 |
+
"loss": 0.0241,
|
| 1063 |
+
"step": 1300
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 20.3125,
|
| 1067 |
+
"eval_accuracy": 0.963716814159292,
|
| 1068 |
+
"eval_f1": 0.9114470842332613,
|
| 1069 |
+
"eval_loss": 0.09792255610227585,
|
| 1070 |
+
"eval_precision": 0.9094827586206896,
|
| 1071 |
+
"eval_recall": 0.9134199134199135,
|
| 1072 |
+
"eval_runtime": 0.8811,
|
| 1073 |
+
"eval_samples_per_second": 128.253,
|
| 1074 |
+
"eval_steps_per_second": 17.025,
|
| 1075 |
+
"step": 1300
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 20.46875,
|
| 1079 |
+
"grad_norm": 0.44642359018325806,
|
| 1080 |
+
"learning_rate": 7.060185185185186e-05,
|
| 1081 |
+
"loss": 0.0271,
|
| 1082 |
+
"step": 1310
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 20.625,
|
| 1086 |
+
"grad_norm": 0.16677480936050415,
|
| 1087 |
+
"learning_rate": 6.944444444444444e-05,
|
| 1088 |
+
"loss": 0.0199,
|
| 1089 |
+
"step": 1320
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 20.78125,
|
| 1093 |
+
"grad_norm": 0.05306961014866829,
|
| 1094 |
+
"learning_rate": 6.828703703703704e-05,
|
| 1095 |
+
"loss": 0.028,
|
| 1096 |
+
"step": 1330
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 20.9375,
|
| 1100 |
+
"grad_norm": 0.7962948679924011,
|
| 1101 |
+
"learning_rate": 6.712962962962963e-05,
|
| 1102 |
+
"loss": 0.0238,
|
| 1103 |
+
"step": 1340
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 21.09375,
|
| 1107 |
+
"grad_norm": 0.19253899157047272,
|
| 1108 |
+
"learning_rate": 6.597222222222223e-05,
|
| 1109 |
+
"loss": 0.0367,
|
| 1110 |
+
"step": 1350
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 21.25,
|
| 1114 |
+
"grad_norm": 0.22666649520397186,
|
| 1115 |
+
"learning_rate": 6.481481481481482e-05,
|
| 1116 |
+
"loss": 0.021,
|
| 1117 |
+
"step": 1360
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 21.40625,
|
| 1121 |
+
"grad_norm": 0.09341959655284882,
|
| 1122 |
+
"learning_rate": 6.365740740740742e-05,
|
| 1123 |
+
"loss": 0.0217,
|
| 1124 |
+
"step": 1370
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 21.5625,
|
| 1128 |
+
"grad_norm": 0.40562504529953003,
|
| 1129 |
+
"learning_rate": 6.25e-05,
|
| 1130 |
+
"loss": 0.0268,
|
| 1131 |
+
"step": 1380
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 21.71875,
|
| 1135 |
+
"grad_norm": 0.20743058621883392,
|
| 1136 |
+
"learning_rate": 6.13425925925926e-05,
|
| 1137 |
+
"loss": 0.0394,
|
| 1138 |
+
"step": 1390
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 21.875,
|
| 1142 |
+
"grad_norm": 0.16062897443771362,
|
| 1143 |
+
"learning_rate": 6.018518518518519e-05,
|
| 1144 |
+
"loss": 0.0311,
|
| 1145 |
+
"step": 1400
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 21.875,
|
| 1149 |
+
"eval_accuracy": 0.9654867256637168,
|
| 1150 |
+
"eval_f1": 0.9157667386609072,
|
| 1151 |
+
"eval_loss": 0.11342811584472656,
|
| 1152 |
+
"eval_precision": 0.9137931034482759,
|
| 1153 |
+
"eval_recall": 0.9177489177489178,
|
| 1154 |
+
"eval_runtime": 0.8884,
|
| 1155 |
+
"eval_samples_per_second": 127.2,
|
| 1156 |
+
"eval_steps_per_second": 16.885,
|
| 1157 |
+
"step": 1400
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 22.03125,
|
| 1161 |
+
"grad_norm": 0.08394443988800049,
|
| 1162 |
+
"learning_rate": 5.902777777777778e-05,
|
| 1163 |
+
"loss": 0.0312,
|
| 1164 |
+
"step": 1410
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 22.1875,
|
| 1168 |
+
"grad_norm": 0.6736553311347961,
|
| 1169 |
+
"learning_rate": 5.787037037037037e-05,
|
| 1170 |
+
"loss": 0.0383,
|
| 1171 |
+
"step": 1420
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 22.34375,
|
| 1175 |
+
"grad_norm": 0.563914954662323,
|
| 1176 |
+
"learning_rate": 5.6712962962962965e-05,
|
| 1177 |
+
"loss": 0.0287,
|
| 1178 |
+
"step": 1430
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 22.5,
|
| 1182 |
+
"grad_norm": 0.08304356783628464,
|
| 1183 |
+
"learning_rate": 5.555555555555556e-05,
|
| 1184 |
+
"loss": 0.026,
|
| 1185 |
+
"step": 1440
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 22.65625,
|
| 1189 |
+
"grad_norm": 0.6314889788627625,
|
| 1190 |
+
"learning_rate": 5.439814814814815e-05,
|
| 1191 |
+
"loss": 0.0337,
|
| 1192 |
+
"step": 1450
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 22.8125,
|
| 1196 |
+
"grad_norm": 0.1526585817337036,
|
| 1197 |
+
"learning_rate": 5.3240740740740744e-05,
|
| 1198 |
+
"loss": 0.0386,
|
| 1199 |
+
"step": 1460
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 22.96875,
|
| 1203 |
+
"grad_norm": 0.4352094829082489,
|
| 1204 |
+
"learning_rate": 5.208333333333334e-05,
|
| 1205 |
+
"loss": 0.0225,
|
| 1206 |
+
"step": 1470
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 23.125,
|
| 1210 |
+
"grad_norm": 0.07802680879831314,
|
| 1211 |
+
"learning_rate": 5.092592592592593e-05,
|
| 1212 |
+
"loss": 0.028,
|
| 1213 |
+
"step": 1480
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 23.28125,
|
| 1217 |
+
"grad_norm": 0.06631523370742798,
|
| 1218 |
+
"learning_rate": 4.976851851851852e-05,
|
| 1219 |
+
"loss": 0.0216,
|
| 1220 |
+
"step": 1490
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 23.4375,
|
| 1224 |
+
"grad_norm": 0.4568875730037689,
|
| 1225 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 1226 |
+
"loss": 0.0303,
|
| 1227 |
+
"step": 1500
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 23.4375,
|
| 1231 |
+
"eval_accuracy": 0.9628318584070796,
|
| 1232 |
+
"eval_f1": 0.9078947368421053,
|
| 1233 |
+
"eval_loss": 0.10922601819038391,
|
| 1234 |
+
"eval_precision": 0.92,
|
| 1235 |
+
"eval_recall": 0.8961038961038961,
|
| 1236 |
+
"eval_runtime": 1.1366,
|
| 1237 |
+
"eval_samples_per_second": 99.417,
|
| 1238 |
+
"eval_steps_per_second": 13.197,
|
| 1239 |
+
"step": 1500
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 23.59375,
|
| 1243 |
+
"grad_norm": 0.16732257604599,
|
| 1244 |
+
"learning_rate": 4.745370370370371e-05,
|
| 1245 |
+
"loss": 0.0182,
|
| 1246 |
+
"step": 1510
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 23.75,
|
| 1250 |
+
"grad_norm": 0.8489612340927124,
|
| 1251 |
+
"learning_rate": 4.62962962962963e-05,
|
| 1252 |
+
"loss": 0.0419,
|
| 1253 |
+
"step": 1520
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 23.90625,
|
| 1257 |
+
"grad_norm": 0.23256537318229675,
|
| 1258 |
+
"learning_rate": 4.5138888888888894e-05,
|
| 1259 |
+
"loss": 0.0344,
|
| 1260 |
+
"step": 1530
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 24.0625,
|
| 1264 |
+
"grad_norm": 0.23274816572666168,
|
| 1265 |
+
"learning_rate": 4.3981481481481486e-05,
|
| 1266 |
+
"loss": 0.0283,
|
| 1267 |
+
"step": 1540
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 24.21875,
|
| 1271 |
+
"grad_norm": 0.1935439109802246,
|
| 1272 |
+
"learning_rate": 4.282407407407408e-05,
|
| 1273 |
+
"loss": 0.0293,
|
| 1274 |
+
"step": 1550
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 24.375,
|
| 1278 |
+
"grad_norm": 0.4433891773223877,
|
| 1279 |
+
"learning_rate": 4.166666666666667e-05,
|
| 1280 |
+
"loss": 0.0341,
|
| 1281 |
+
"step": 1560
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 24.53125,
|
| 1285 |
+
"grad_norm": 0.3329981863498688,
|
| 1286 |
+
"learning_rate": 4.0509259259259265e-05,
|
| 1287 |
+
"loss": 0.022,
|
| 1288 |
+
"step": 1570
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 24.6875,
|
| 1292 |
+
"grad_norm": 0.774336040019989,
|
| 1293 |
+
"learning_rate": 3.935185185185186e-05,
|
| 1294 |
+
"loss": 0.0304,
|
| 1295 |
+
"step": 1580
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 24.84375,
|
| 1299 |
+
"grad_norm": 0.45676717162132263,
|
| 1300 |
+
"learning_rate": 3.8194444444444444e-05,
|
| 1301 |
+
"loss": 0.0213,
|
| 1302 |
+
"step": 1590
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 25.0,
|
| 1306 |
+
"grad_norm": 0.18224991858005524,
|
| 1307 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 1308 |
+
"loss": 0.0225,
|
| 1309 |
+
"step": 1600
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 25.0,
|
| 1313 |
+
"eval_accuracy": 0.9628318584070796,
|
| 1314 |
+
"eval_f1": 0.908296943231441,
|
| 1315 |
+
"eval_loss": 0.11213955283164978,
|
| 1316 |
+
"eval_precision": 0.9162995594713657,
|
| 1317 |
+
"eval_recall": 0.9004329004329005,
|
| 1318 |
+
"eval_runtime": 1.05,
|
| 1319 |
+
"eval_samples_per_second": 107.622,
|
| 1320 |
+
"eval_steps_per_second": 14.286,
|
| 1321 |
+
"step": 1600
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 25.15625,
|
| 1325 |
+
"grad_norm": 0.45673811435699463,
|
| 1326 |
+
"learning_rate": 3.587962962962963e-05,
|
| 1327 |
+
"loss": 0.0253,
|
| 1328 |
+
"step": 1610
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 25.3125,
|
| 1332 |
+
"grad_norm": 0.09005212038755417,
|
| 1333 |
+
"learning_rate": 3.472222222222222e-05,
|
| 1334 |
+
"loss": 0.0127,
|
| 1335 |
+
"step": 1620
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 25.46875,
|
| 1339 |
+
"grad_norm": 0.20682398974895477,
|
| 1340 |
+
"learning_rate": 3.3564814814814815e-05,
|
| 1341 |
+
"loss": 0.0231,
|
| 1342 |
+
"step": 1630
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 25.625,
|
| 1346 |
+
"grad_norm": 0.7664525508880615,
|
| 1347 |
+
"learning_rate": 3.240740740740741e-05,
|
| 1348 |
+
"loss": 0.0174,
|
| 1349 |
+
"step": 1640
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 25.78125,
|
| 1353 |
+
"grad_norm": 0.20978455245494843,
|
| 1354 |
+
"learning_rate": 3.125e-05,
|
| 1355 |
+
"loss": 0.0203,
|
| 1356 |
+
"step": 1650
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 25.9375,
|
| 1360 |
+
"grad_norm": 0.5540274977684021,
|
| 1361 |
+
"learning_rate": 3.0092592592592593e-05,
|
| 1362 |
+
"loss": 0.0205,
|
| 1363 |
+
"step": 1660
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 26.09375,
|
| 1367 |
+
"grad_norm": 0.1240416169166565,
|
| 1368 |
+
"learning_rate": 2.8935185185185186e-05,
|
| 1369 |
+
"loss": 0.0168,
|
| 1370 |
+
"step": 1670
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 26.25,
|
| 1374 |
+
"grad_norm": 0.04385749623179436,
|
| 1375 |
+
"learning_rate": 2.777777777777778e-05,
|
| 1376 |
+
"loss": 0.0162,
|
| 1377 |
+
"step": 1680
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 26.40625,
|
| 1381 |
+
"grad_norm": 0.1799972951412201,
|
| 1382 |
+
"learning_rate": 2.6620370370370372e-05,
|
| 1383 |
+
"loss": 0.0196,
|
| 1384 |
+
"step": 1690
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 26.5625,
|
| 1388 |
+
"grad_norm": 0.24593585729599,
|
| 1389 |
+
"learning_rate": 2.5462962962962965e-05,
|
| 1390 |
+
"loss": 0.0292,
|
| 1391 |
+
"step": 1700
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 26.5625,
|
| 1395 |
+
"eval_accuracy": 0.9619469026548673,
|
| 1396 |
+
"eval_f1": 0.9071274298056156,
|
| 1397 |
+
"eval_loss": 0.11493521183729172,
|
| 1398 |
+
"eval_precision": 0.9051724137931034,
|
| 1399 |
+
"eval_recall": 0.9090909090909091,
|
| 1400 |
+
"eval_runtime": 0.8719,
|
| 1401 |
+
"eval_samples_per_second": 129.607,
|
| 1402 |
+
"eval_steps_per_second": 17.204,
|
| 1403 |
+
"step": 1700
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 26.71875,
|
| 1407 |
+
"grad_norm": 0.2344673126935959,
|
| 1408 |
+
"learning_rate": 2.4305555555555558e-05,
|
| 1409 |
+
"loss": 0.0177,
|
| 1410 |
+
"step": 1710
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 26.875,
|
| 1414 |
+
"grad_norm": 0.263621062040329,
|
| 1415 |
+
"learning_rate": 2.314814814814815e-05,
|
| 1416 |
+
"loss": 0.0288,
|
| 1417 |
+
"step": 1720
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 27.03125,
|
| 1421 |
+
"grad_norm": 0.27248746156692505,
|
| 1422 |
+
"learning_rate": 2.1990740740740743e-05,
|
| 1423 |
+
"loss": 0.0158,
|
| 1424 |
+
"step": 1730
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 27.1875,
|
| 1428 |
+
"grad_norm": 0.35065901279449463,
|
| 1429 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 1430 |
+
"loss": 0.0198,
|
| 1431 |
+
"step": 1740
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 27.34375,
|
| 1435 |
+
"grad_norm": 0.23319651186466217,
|
| 1436 |
+
"learning_rate": 1.967592592592593e-05,
|
| 1437 |
+
"loss": 0.0208,
|
| 1438 |
+
"step": 1750
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 27.5,
|
| 1442 |
+
"grad_norm": 0.15196481347084045,
|
| 1443 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 1444 |
+
"loss": 0.0161,
|
| 1445 |
+
"step": 1760
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 27.65625,
|
| 1449 |
+
"grad_norm": 0.17273353040218353,
|
| 1450 |
+
"learning_rate": 1.736111111111111e-05,
|
| 1451 |
+
"loss": 0.0213,
|
| 1452 |
+
"step": 1770
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 27.8125,
|
| 1456 |
+
"grad_norm": 0.31511059403419495,
|
| 1457 |
+
"learning_rate": 1.6203703703703704e-05,
|
| 1458 |
+
"loss": 0.012,
|
| 1459 |
+
"step": 1780
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 27.96875,
|
| 1463 |
+
"grad_norm": 0.09265203773975372,
|
| 1464 |
+
"learning_rate": 1.5046296296296297e-05,
|
| 1465 |
+
"loss": 0.0218,
|
| 1466 |
+
"step": 1790
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 28.125,
|
| 1470 |
+
"grad_norm": 0.2638147473335266,
|
| 1471 |
+
"learning_rate": 1.388888888888889e-05,
|
| 1472 |
+
"loss": 0.0261,
|
| 1473 |
+
"step": 1800
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 28.125,
|
| 1477 |
+
"eval_accuracy": 0.9619469026548673,
|
| 1478 |
+
"eval_f1": 0.9079229122055675,
|
| 1479 |
+
"eval_loss": 0.11067904531955719,
|
| 1480 |
+
"eval_precision": 0.8983050847457628,
|
| 1481 |
+
"eval_recall": 0.9177489177489178,
|
| 1482 |
+
"eval_runtime": 0.9722,
|
| 1483 |
+
"eval_samples_per_second": 116.235,
|
| 1484 |
+
"eval_steps_per_second": 15.429,
|
| 1485 |
+
"step": 1800
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 28.28125,
|
| 1489 |
+
"grad_norm": 0.2734526991844177,
|
| 1490 |
+
"learning_rate": 1.2731481481481482e-05,
|
| 1491 |
+
"loss": 0.0175,
|
| 1492 |
+
"step": 1810
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 28.4375,
|
| 1496 |
+
"grad_norm": 0.06026133522391319,
|
| 1497 |
+
"learning_rate": 1.1574074074074075e-05,
|
| 1498 |
+
"loss": 0.0168,
|
| 1499 |
+
"step": 1820
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 28.59375,
|
| 1503 |
+
"grad_norm": 0.02611556649208069,
|
| 1504 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 1505 |
+
"loss": 0.0177,
|
| 1506 |
+
"step": 1830
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 28.75,
|
| 1510 |
+
"grad_norm": 0.23434928059577942,
|
| 1511 |
+
"learning_rate": 9.259259259259259e-06,
|
| 1512 |
+
"loss": 0.0252,
|
| 1513 |
+
"step": 1840
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 28.90625,
|
| 1517 |
+
"grad_norm": 0.07623735815286636,
|
| 1518 |
+
"learning_rate": 8.101851851851852e-06,
|
| 1519 |
+
"loss": 0.0235,
|
| 1520 |
+
"step": 1850
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 29.0625,
|
| 1524 |
+
"grad_norm": 0.05061192065477371,
|
| 1525 |
+
"learning_rate": 6.944444444444445e-06,
|
| 1526 |
+
"loss": 0.0185,
|
| 1527 |
+
"step": 1860
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 29.21875,
|
| 1531 |
+
"grad_norm": 0.03355779871344566,
|
| 1532 |
+
"learning_rate": 5.787037037037038e-06,
|
| 1533 |
+
"loss": 0.0157,
|
| 1534 |
+
"step": 1870
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 29.375,
|
| 1538 |
+
"grad_norm": 0.023396974429488182,
|
| 1539 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 1540 |
+
"loss": 0.016,
|
| 1541 |
+
"step": 1880
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 29.53125,
|
| 1545 |
+
"grad_norm": 0.31754446029663086,
|
| 1546 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 1547 |
+
"loss": 0.0213,
|
| 1548 |
+
"step": 1890
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 29.6875,
|
| 1552 |
+
"grad_norm": 0.04646694287657738,
|
| 1553 |
+
"learning_rate": 2.3148148148148148e-06,
|
| 1554 |
+
"loss": 0.0166,
|
| 1555 |
+
"step": 1900
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 29.6875,
|
| 1559 |
+
"eval_accuracy": 0.9610619469026549,
|
| 1560 |
+
"eval_f1": 0.9051724137931034,
|
| 1561 |
+
"eval_loss": 0.11101004481315613,
|
| 1562 |
+
"eval_precision": 0.9012875536480687,
|
| 1563 |
+
"eval_recall": 0.9090909090909091,
|
| 1564 |
+
"eval_runtime": 0.8549,
|
| 1565 |
+
"eval_samples_per_second": 132.182,
|
| 1566 |
+
"eval_steps_per_second": 17.546,
|
| 1567 |
+
"step": 1900
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 29.84375,
|
| 1571 |
+
"grad_norm": 0.05433151125907898,
|
| 1572 |
+
"learning_rate": 1.1574074074074074e-06,
|
| 1573 |
+
"loss": 0.0174,
|
| 1574 |
+
"step": 1910
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 30.0,
|
| 1578 |
+
"grad_norm": 0.3085578382015228,
|
| 1579 |
+
"learning_rate": 0.0,
|
| 1580 |
+
"loss": 0.0208,
|
| 1581 |
+
"step": 1920
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 30.0,
|
| 1585 |
+
"step": 1920,
|
| 1586 |
+
"total_flos": 2.352825493649326e+18,
|
| 1587 |
+
"train_loss": 0.05003170374160012,
|
| 1588 |
+
"train_runtime": 517.118,
|
| 1589 |
+
"train_samples_per_second": 58.71,
|
| 1590 |
+
"train_steps_per_second": 3.713
|
| 1591 |
}
|
| 1592 |
],
|
| 1593 |
"logging_steps": 10,
|
| 1594 |
+
"max_steps": 1920,
|
| 1595 |
"num_input_tokens_seen": 0,
|
| 1596 |
+
"num_train_epochs": 30,
|
| 1597 |
"save_steps": 500,
|
| 1598 |
"stateful_callbacks": {
|
| 1599 |
"TrainerControl": {
|
|
|
|
| 1607 |
"attributes": {}
|
| 1608 |
}
|
| 1609 |
},
|
| 1610 |
+
"total_flos": 2.352825493649326e+18,
|
| 1611 |
"train_batch_size": 16,
|
| 1612 |
"trial_name": null,
|
| 1613 |
"trial_params": null
|