🍻 cheers
Browse files- README.md +5 -4
- all_results.json +12 -12
- eval_results.json +7 -7
- runs/May26_21-06-28_35d69dd4580d/events.out.tfevents.1748295298.35d69dd4580d.35.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +1443 -650
README.md
CHANGED
|
@@ -3,6 +3,7 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: microsoft/swin-base-simmim-window6-192
|
| 5 |
tags:
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
@@ -16,11 +17,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# swin-ena24
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [microsoft/swin-base-simmim-window6-192](https://huggingface.co/microsoft/swin-base-simmim-window6-192) on
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss:
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
-
- F1 Macro: 0.
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: microsoft/swin-base-simmim-window6-192
|
| 5 |
tags:
|
| 6 |
+
- image-classification
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
|
|
|
| 17 |
|
| 18 |
# swin-ena24
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [microsoft/swin-base-simmim-window6-192](https://huggingface.co/microsoft/swin-base-simmim-window6-192) on the ena24 dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 2.4677
|
| 23 |
+
- Accuracy: 0.5146
|
| 24 |
+
- F1 Macro: 0.4328
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_f1_macro": 0.
|
| 5 |
-
"eval_loss": 2.
|
| 6 |
-
"eval_runtime": 12.
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second": 5.
|
| 9 |
-
"total_flos":
|
| 10 |
-
"train_loss": 0.
|
| 11 |
-
"train_runtime":
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 1.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 7.0,
|
| 3 |
+
"eval_accuracy": 0.5146484375,
|
| 4 |
+
"eval_f1_macro": 0.432831730682114,
|
| 5 |
+
"eval_loss": 2.467733144760132,
|
| 6 |
+
"eval_runtime": 12.1136,
|
| 7 |
+
"eval_samples_per_second": 84.533,
|
| 8 |
+
"eval_steps_per_second": 5.283,
|
| 9 |
+
"total_flos": 2.5585840915697664e+18,
|
| 10 |
+
"train_loss": 0.4836694428009911,
|
| 11 |
+
"train_runtime": 1538.4023,
|
| 12 |
+
"train_samples_per_second": 28.894,
|
| 13 |
+
"train_steps_per_second": 1.806
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_f1_macro": 0.
|
| 5 |
-
"eval_loss": 2.
|
| 6 |
-
"eval_runtime": 12.
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second": 5.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 7.0,
|
| 3 |
+
"eval_accuracy": 0.5146484375,
|
| 4 |
+
"eval_f1_macro": 0.432831730682114,
|
| 5 |
+
"eval_loss": 2.467733144760132,
|
| 6 |
+
"eval_runtime": 12.1136,
|
| 7 |
+
"eval_samples_per_second": 84.533,
|
| 8 |
+
"eval_steps_per_second": 5.283
|
| 9 |
}
|
runs/May26_21-06-28_35d69dd4580d/events.out.tfevents.1748295298.35d69dd4580d.35.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:88af87eb188944eb9bcf29288441ae963bd95b848a78b14454c88cc062382539
|
| 3 |
+
size 463
|
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": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 7.0,
|
| 3 |
+
"total_flos": 2.5585840915697664e+18,
|
| 4 |
+
"train_loss": 0.4836694428009911,
|
| 5 |
+
"train_runtime": 1538.4023,
|
| 6 |
+
"train_samples_per_second": 28.894,
|
| 7 |
+
"train_steps_per_second": 1.806
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1444 +1,2237 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 2.
|
| 4 |
-
"best_model_checkpoint": "./swin-ena24/checkpoint-
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 100,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.02518891687657431,
|
| 14 |
-
"grad_norm":
|
| 15 |
-
"learning_rate": 0.
|
| 16 |
-
"loss":
|
| 17 |
"step": 10
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.05037783375314862,
|
| 21 |
-
"grad_norm":
|
| 22 |
-
"learning_rate": 0.
|
| 23 |
-
"loss":
|
| 24 |
"step": 20
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.07556675062972293,
|
| 28 |
-
"grad_norm":
|
| 29 |
-
"learning_rate": 0.
|
| 30 |
-
"loss":
|
| 31 |
"step": 30
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.10075566750629723,
|
| 35 |
-
"grad_norm":
|
| 36 |
-
"learning_rate": 0.
|
| 37 |
-
"loss":
|
| 38 |
"step": 40
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.12594458438287154,
|
| 42 |
-
"grad_norm":
|
| 43 |
-
"learning_rate": 0.
|
| 44 |
-
"loss":
|
| 45 |
"step": 50
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.15113350125944586,
|
| 49 |
-
"grad_norm":
|
| 50 |
-
"learning_rate": 0.
|
| 51 |
-
"loss":
|
| 52 |
"step": 60
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.17632241813602015,
|
| 56 |
-
"grad_norm":
|
| 57 |
-
"learning_rate": 0.
|
| 58 |
-
"loss":
|
| 59 |
"step": 70
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.20151133501259447,
|
| 63 |
-
"grad_norm":
|
| 64 |
-
"learning_rate": 0.
|
| 65 |
-
"loss":
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.22670025188916876,
|
| 70 |
-
"grad_norm":
|
| 71 |
-
"learning_rate": 0.
|
| 72 |
-
"loss":
|
| 73 |
"step": 90
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.2518891687657431,
|
| 77 |
-
"grad_norm":
|
| 78 |
-
"learning_rate": 0.
|
| 79 |
-
"loss":
|
| 80 |
"step": 100
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.2518891687657431,
|
| 84 |
-
"eval_accuracy": 0.
|
| 85 |
-
"eval_f1_macro": 0.
|
| 86 |
-
"eval_loss":
|
| 87 |
-
"eval_runtime":
|
| 88 |
-
"eval_samples_per_second":
|
| 89 |
-
"eval_steps_per_second": 5.
|
| 90 |
"step": 100
|
| 91 |
},
|
| 92 |
{
|
| 93 |
"epoch": 0.2770780856423174,
|
| 94 |
-
"grad_norm":
|
| 95 |
-
"learning_rate": 0.
|
| 96 |
-
"loss":
|
| 97 |
"step": 110
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 0.3022670025188917,
|
| 101 |
-
"grad_norm":
|
| 102 |
-
"learning_rate": 0.
|
| 103 |
-
"loss":
|
| 104 |
"step": 120
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 0.327455919395466,
|
| 108 |
-
"grad_norm":
|
| 109 |
-
"learning_rate": 0.
|
| 110 |
-
"loss":
|
| 111 |
"step": 130
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 0.3526448362720403,
|
| 115 |
-
"grad_norm":
|
| 116 |
-
"learning_rate": 0.
|
| 117 |
-
"loss":
|
| 118 |
"step": 140
|
| 119 |
},
|
| 120 |
{
|
| 121 |
"epoch": 0.3778337531486146,
|
| 122 |
-
"grad_norm":
|
| 123 |
-
"learning_rate": 0.
|
| 124 |
-
"loss":
|
| 125 |
"step": 150
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 0.40302267002518893,
|
| 129 |
-
"grad_norm":
|
| 130 |
-
"learning_rate": 0.
|
| 131 |
-
"loss":
|
| 132 |
"step": 160
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 0.4282115869017632,
|
| 136 |
-
"grad_norm":
|
| 137 |
-
"learning_rate": 0.
|
| 138 |
-
"loss": 1.
|
| 139 |
"step": 170
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 0.4534005037783375,
|
| 143 |
-
"grad_norm":
|
| 144 |
-
"learning_rate": 0.
|
| 145 |
-
"loss": 1.
|
| 146 |
"step": 180
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 0.47858942065491183,
|
| 150 |
-
"grad_norm":
|
| 151 |
-
"learning_rate": 0.
|
| 152 |
-
"loss": 1.
|
| 153 |
"step": 190
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 0.5037783375314862,
|
| 157 |
-
"grad_norm":
|
| 158 |
-
"learning_rate": 0.
|
| 159 |
-
"loss":
|
| 160 |
"step": 200
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 0.5037783375314862,
|
| 164 |
-
"eval_accuracy": 0.
|
| 165 |
-
"eval_f1_macro": 0.
|
| 166 |
-
"eval_loss": 2.
|
| 167 |
-
"eval_runtime":
|
| 168 |
-
"eval_samples_per_second":
|
| 169 |
-
"eval_steps_per_second": 5.
|
| 170 |
"step": 200
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.5289672544080605,
|
| 174 |
-
"grad_norm":
|
| 175 |
-
"learning_rate": 0.
|
| 176 |
-
"loss":
|
| 177 |
"step": 210
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.5541561712846348,
|
| 181 |
-
"grad_norm":
|
| 182 |
-
"learning_rate": 0.
|
| 183 |
-
"loss":
|
| 184 |
"step": 220
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.5793450881612091,
|
| 188 |
-
"grad_norm":
|
| 189 |
-
"learning_rate": 0.
|
| 190 |
-
"loss":
|
| 191 |
"step": 230
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.6045340050377834,
|
| 195 |
-
"grad_norm":
|
| 196 |
-
"learning_rate": 0.
|
| 197 |
-
"loss":
|
| 198 |
"step": 240
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.6297229219143576,
|
| 202 |
-
"grad_norm":
|
| 203 |
-
"learning_rate": 0.
|
| 204 |
-
"loss":
|
| 205 |
"step": 250
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.654911838790932,
|
| 209 |
-
"grad_norm":
|
| 210 |
-
"learning_rate": 0.
|
| 211 |
-
"loss":
|
| 212 |
"step": 260
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.6801007556675063,
|
| 216 |
-
"grad_norm":
|
| 217 |
-
"learning_rate": 0.
|
| 218 |
-
"loss":
|
| 219 |
"step": 270
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.7052896725440806,
|
| 223 |
-
"grad_norm":
|
| 224 |
-
"learning_rate": 0.
|
| 225 |
-
"loss":
|
| 226 |
"step": 280
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.7304785894206549,
|
| 230 |
-
"grad_norm":
|
| 231 |
-
"learning_rate": 0.
|
| 232 |
-
"loss":
|
| 233 |
"step": 290
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.7556675062972292,
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate": 0.
|
| 239 |
-
"loss":
|
| 240 |
"step": 300
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.7556675062972292,
|
| 244 |
-
"eval_accuracy": 0.
|
| 245 |
-
"eval_f1_macro": 0.
|
| 246 |
-
"eval_loss": 2.
|
| 247 |
-
"eval_runtime":
|
| 248 |
-
"eval_samples_per_second":
|
| 249 |
-
"eval_steps_per_second": 5.
|
| 250 |
"step": 300
|
| 251 |
},
|
| 252 |
{
|
| 253 |
"epoch": 0.7808564231738035,
|
| 254 |
-
"grad_norm":
|
| 255 |
-
"learning_rate": 0.
|
| 256 |
-
"loss":
|
| 257 |
"step": 310
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.8060453400503779,
|
| 261 |
-
"grad_norm":
|
| 262 |
-
"learning_rate": 0.
|
| 263 |
-
"loss":
|
| 264 |
"step": 320
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 0.8312342569269522,
|
| 268 |
-
"grad_norm":
|
| 269 |
-
"learning_rate": 0.
|
| 270 |
-
"loss":
|
| 271 |
"step": 330
|
| 272 |
},
|
| 273 |
{
|
| 274 |
"epoch": 0.8564231738035264,
|
| 275 |
-
"grad_norm":
|
| 276 |
-
"learning_rate": 0.
|
| 277 |
-
"loss":
|
| 278 |
"step": 340
|
| 279 |
},
|
| 280 |
{
|
| 281 |
"epoch": 0.8816120906801007,
|
| 282 |
-
"grad_norm":
|
| 283 |
-
"learning_rate": 0.
|
| 284 |
-
"loss":
|
| 285 |
"step": 350
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 0.906801007556675,
|
| 289 |
-
"grad_norm":
|
| 290 |
-
"learning_rate": 0.
|
| 291 |
-
"loss":
|
| 292 |
"step": 360
|
| 293 |
},
|
| 294 |
{
|
| 295 |
"epoch": 0.9319899244332494,
|
| 296 |
-
"grad_norm":
|
| 297 |
-
"learning_rate": 0.
|
| 298 |
-
"loss":
|
| 299 |
"step": 370
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 0.9571788413098237,
|
| 303 |
-
"grad_norm":
|
| 304 |
-
"learning_rate": 0.
|
| 305 |
-
"loss":
|
| 306 |
"step": 380
|
| 307 |
},
|
| 308 |
{
|
| 309 |
"epoch": 0.982367758186398,
|
| 310 |
-
"grad_norm":
|
| 311 |
-
"learning_rate": 0.
|
| 312 |
-
"loss":
|
| 313 |
"step": 390
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 1.0075566750629723,
|
| 317 |
-
"grad_norm":
|
| 318 |
-
"learning_rate": 0.
|
| 319 |
-
"loss":
|
| 320 |
"step": 400
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 1.0075566750629723,
|
| 324 |
-
"eval_accuracy": 0.
|
| 325 |
-
"eval_f1_macro": 0.
|
| 326 |
-
"eval_loss": 2.
|
| 327 |
-
"eval_runtime":
|
| 328 |
-
"eval_samples_per_second":
|
| 329 |
-
"eval_steps_per_second": 5.
|
| 330 |
"step": 400
|
| 331 |
},
|
| 332 |
{
|
| 333 |
"epoch": 1.0327455919395465,
|
| 334 |
-
"grad_norm":
|
| 335 |
-
"learning_rate": 0.
|
| 336 |
-
"loss":
|
| 337 |
"step": 410
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 1.057934508816121,
|
| 341 |
-
"grad_norm":
|
| 342 |
-
"learning_rate": 0.
|
| 343 |
-
"loss": 0.
|
| 344 |
"step": 420
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 1.0831234256926952,
|
| 348 |
-
"grad_norm":
|
| 349 |
-
"learning_rate": 0.
|
| 350 |
-
"loss": 0.
|
| 351 |
"step": 430
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 1.1083123425692696,
|
| 355 |
-
"grad_norm":
|
| 356 |
-
"learning_rate": 0.
|
| 357 |
-
"loss":
|
| 358 |
"step": 440
|
| 359 |
},
|
| 360 |
{
|
| 361 |
"epoch": 1.1335012594458438,
|
| 362 |
-
"grad_norm":
|
| 363 |
-
"learning_rate": 0.
|
| 364 |
-
"loss": 0.
|
| 365 |
"step": 450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 1.1586901763224182,
|
| 369 |
-
"grad_norm":
|
| 370 |
-
"learning_rate": 0.
|
| 371 |
-
"loss":
|
| 372 |
"step": 460
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 1.1838790931989924,
|
| 376 |
-
"grad_norm":
|
| 377 |
-
"learning_rate": 0.
|
| 378 |
-
"loss": 0.
|
| 379 |
"step": 470
|
| 380 |
},
|
| 381 |
{
|
| 382 |
"epoch": 1.2090680100755669,
|
| 383 |
-
"grad_norm":
|
| 384 |
-
"learning_rate": 0.
|
| 385 |
-
"loss":
|
| 386 |
"step": 480
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 1.234256926952141,
|
| 390 |
-
"grad_norm":
|
| 391 |
-
"learning_rate": 0.
|
| 392 |
-
"loss": 0.
|
| 393 |
"step": 490
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 1.2594458438287153,
|
| 397 |
-
"grad_norm":
|
| 398 |
-
"learning_rate": 0.
|
| 399 |
-
"loss": 0.
|
| 400 |
"step": 500
|
| 401 |
},
|
| 402 |
{
|
| 403 |
"epoch": 1.2594458438287153,
|
| 404 |
-
"eval_accuracy": 0.
|
| 405 |
-
"eval_f1_macro": 0.
|
| 406 |
-
"eval_loss": 2.
|
| 407 |
-
"eval_runtime":
|
| 408 |
-
"eval_samples_per_second":
|
| 409 |
-
"eval_steps_per_second": 5.
|
| 410 |
"step": 500
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 1.2846347607052897,
|
| 414 |
-
"grad_norm":
|
| 415 |
-
"learning_rate": 0.
|
| 416 |
-
"loss": 0.
|
| 417 |
"step": 510
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 1.309823677581864,
|
| 421 |
-
"grad_norm":
|
| 422 |
-
"learning_rate": 0.
|
| 423 |
-
"loss":
|
| 424 |
"step": 520
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 1.3350125944584383,
|
| 428 |
-
"grad_norm":
|
| 429 |
-
"learning_rate": 0.
|
| 430 |
-
"loss": 0.
|
| 431 |
"step": 530
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 1.3602015113350125,
|
| 435 |
-
"grad_norm":
|
| 436 |
-
"learning_rate": 0.
|
| 437 |
-
"loss": 0.
|
| 438 |
"step": 540
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 1.385390428211587,
|
| 442 |
-
"grad_norm":
|
| 443 |
-
"learning_rate": 0.
|
| 444 |
-
"loss": 0.
|
| 445 |
"step": 550
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 1.4105793450881612,
|
| 449 |
-
"grad_norm":
|
| 450 |
-
"learning_rate": 0.
|
| 451 |
-
"loss": 0.
|
| 452 |
"step": 560
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 1.4357682619647356,
|
| 456 |
-
"grad_norm":
|
| 457 |
-
"learning_rate": 0.
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 570
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 1.4609571788413098,
|
| 463 |
-
"grad_norm":
|
| 464 |
-
"learning_rate": 0.
|
| 465 |
-
"loss": 0.
|
| 466 |
"step": 580
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 1.486146095717884,
|
| 470 |
-
"grad_norm":
|
| 471 |
-
"learning_rate": 0.
|
| 472 |
-
"loss": 0.
|
| 473 |
"step": 590
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 1.5113350125944585,
|
| 477 |
-
"grad_norm":
|
| 478 |
-
"learning_rate": 0.
|
| 479 |
-
"loss": 0.
|
| 480 |
"step": 600
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 1.5113350125944585,
|
| 484 |
-
"eval_accuracy": 0.
|
| 485 |
-
"eval_f1_macro": 0.
|
| 486 |
-
"eval_loss": 2.
|
| 487 |
-
"eval_runtime":
|
| 488 |
-
"eval_samples_per_second":
|
| 489 |
-
"eval_steps_per_second": 5.
|
| 490 |
"step": 600
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 1.536523929471033,
|
| 494 |
-
"grad_norm":
|
| 495 |
-
"learning_rate": 0.
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 610
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 1.561712846347607,
|
| 501 |
-
"grad_norm":
|
| 502 |
-
"learning_rate": 0.
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 620
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 1.5869017632241813,
|
| 508 |
-
"grad_norm":
|
| 509 |
-
"learning_rate": 0.
|
| 510 |
-
"loss": 0.
|
| 511 |
"step": 630
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 1.6120906801007555,
|
| 515 |
-
"grad_norm":
|
| 516 |
-
"learning_rate": 0.
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 640
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 1.63727959697733,
|
| 522 |
-
"grad_norm":
|
| 523 |
-
"learning_rate": 0.
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 650
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 1.6624685138539044,
|
| 529 |
-
"grad_norm":
|
| 530 |
-
"learning_rate": 0.
|
| 531 |
-
"loss": 0.
|
| 532 |
"step": 660
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 1.6876574307304786,
|
| 536 |
-
"grad_norm":
|
| 537 |
-
"learning_rate": 0.
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 670
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 1.7128463476070528,
|
| 543 |
-
"grad_norm":
|
| 544 |
-
"learning_rate": 0.
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 680
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 1.7380352644836272,
|
| 550 |
-
"grad_norm":
|
| 551 |
-
"learning_rate": 0.
|
| 552 |
-
"loss": 0.
|
| 553 |
"step": 690
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 1.7632241813602016,
|
| 557 |
-
"grad_norm":
|
| 558 |
-
"learning_rate": 0.
|
| 559 |
-
"loss": 0.
|
| 560 |
"step": 700
|
| 561 |
},
|
| 562 |
{
|
| 563 |
"epoch": 1.7632241813602016,
|
| 564 |
-
"eval_accuracy": 0.
|
| 565 |
-
"eval_f1_macro": 0.
|
| 566 |
-
"eval_loss": 3.
|
| 567 |
-
"eval_runtime": 11.
|
| 568 |
-
"eval_samples_per_second":
|
| 569 |
-
"eval_steps_per_second": 5.
|
| 570 |
"step": 700
|
| 571 |
},
|
| 572 |
{
|
| 573 |
"epoch": 1.7884130982367759,
|
| 574 |
-
"grad_norm":
|
| 575 |
-
"learning_rate": 0.
|
| 576 |
-
"loss": 0.
|
| 577 |
"step": 710
|
| 578 |
},
|
| 579 |
{
|
| 580 |
"epoch": 1.81360201511335,
|
| 581 |
-
"grad_norm":
|
| 582 |
-
"learning_rate": 0.
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 720
|
| 585 |
},
|
| 586 |
{
|
| 587 |
"epoch": 1.8387909319899243,
|
| 588 |
-
"grad_norm":
|
| 589 |
-
"learning_rate": 0.
|
| 590 |
-
"loss": 0.
|
| 591 |
"step": 730
|
| 592 |
},
|
| 593 |
{
|
| 594 |
"epoch": 1.8639798488664987,
|
| 595 |
-
"grad_norm":
|
| 596 |
-
"learning_rate": 0.
|
| 597 |
-
"loss": 0.
|
| 598 |
"step": 740
|
| 599 |
},
|
| 600 |
{
|
| 601 |
"epoch": 1.8891687657430731,
|
| 602 |
-
"grad_norm":
|
| 603 |
-
"learning_rate": 0.
|
| 604 |
-
"loss": 0.
|
| 605 |
"step": 750
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 1.9143576826196473,
|
| 609 |
-
"grad_norm":
|
| 610 |
-
"learning_rate": 0.
|
| 611 |
-
"loss": 0.
|
| 612 |
"step": 760
|
| 613 |
},
|
| 614 |
{
|
| 615 |
"epoch": 1.9395465994962215,
|
| 616 |
-
"grad_norm":
|
| 617 |
-
"learning_rate": 0.
|
| 618 |
-
"loss": 0.
|
| 619 |
"step": 770
|
| 620 |
},
|
| 621 |
{
|
| 622 |
"epoch": 1.964735516372796,
|
| 623 |
-
"grad_norm":
|
| 624 |
-
"learning_rate": 0.
|
| 625 |
-
"loss": 0.
|
| 626 |
"step": 780
|
| 627 |
},
|
| 628 |
{
|
| 629 |
"epoch": 1.9899244332493704,
|
| 630 |
-
"grad_norm":
|
| 631 |
-
"learning_rate": 0.
|
| 632 |
-
"loss": 0.
|
| 633 |
"step": 790
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 2.0151133501259446,
|
| 637 |
-
"grad_norm":
|
| 638 |
-
"learning_rate": 0.
|
| 639 |
-
"loss": 0.
|
| 640 |
"step": 800
|
| 641 |
},
|
| 642 |
{
|
| 643 |
"epoch": 2.0151133501259446,
|
| 644 |
"eval_accuracy": 0.484375,
|
| 645 |
-
"eval_f1_macro": 0.
|
| 646 |
-
"eval_loss": 2.
|
| 647 |
-
"eval_runtime":
|
| 648 |
-
"eval_samples_per_second":
|
| 649 |
-
"eval_steps_per_second": 5.
|
| 650 |
"step": 800
|
| 651 |
},
|
| 652 |
{
|
| 653 |
"epoch": 2.040302267002519,
|
| 654 |
-
"grad_norm":
|
| 655 |
-
"learning_rate": 0.
|
| 656 |
-
"loss": 0.
|
| 657 |
"step": 810
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 2.065491183879093,
|
| 661 |
-
"grad_norm":
|
| 662 |
-
"learning_rate": 0.
|
| 663 |
-
"loss": 0.
|
| 664 |
"step": 820
|
| 665 |
},
|
| 666 |
{
|
| 667 |
"epoch": 2.0906801007556677,
|
| 668 |
-
"grad_norm":
|
| 669 |
-
"learning_rate": 0.
|
| 670 |
-
"loss": 0.
|
| 671 |
"step": 830
|
| 672 |
},
|
| 673 |
{
|
| 674 |
"epoch": 2.115869017632242,
|
| 675 |
-
"grad_norm":
|
| 676 |
-
"learning_rate": 0.
|
| 677 |
-
"loss": 0.
|
| 678 |
"step": 840
|
| 679 |
},
|
| 680 |
{
|
| 681 |
"epoch": 2.141057934508816,
|
| 682 |
-
"grad_norm":
|
| 683 |
-
"learning_rate": 0.
|
| 684 |
-
"loss": 0.
|
| 685 |
"step": 850
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 2.1662468513853903,
|
| 689 |
-
"grad_norm":
|
| 690 |
-
"learning_rate": 0.
|
| 691 |
-
"loss": 0.
|
| 692 |
"step": 860
|
| 693 |
},
|
| 694 |
{
|
| 695 |
"epoch": 2.1914357682619645,
|
| 696 |
-
"grad_norm":
|
| 697 |
-
"learning_rate": 0.
|
| 698 |
-
"loss": 0.
|
| 699 |
"step": 870
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 2.216624685138539,
|
| 703 |
-
"grad_norm":
|
| 704 |
-
"learning_rate": 0.
|
| 705 |
-
"loss": 0.
|
| 706 |
"step": 880
|
| 707 |
},
|
| 708 |
{
|
| 709 |
"epoch": 2.2418136020151134,
|
| 710 |
-
"grad_norm":
|
| 711 |
-
"learning_rate": 0.
|
| 712 |
-
"loss": 0.
|
| 713 |
"step": 890
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 2.2670025188916876,
|
| 717 |
-
"grad_norm":
|
| 718 |
-
"learning_rate":
|
| 719 |
-
"loss": 0.
|
| 720 |
"step": 900
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 2.2670025188916876,
|
| 724 |
-
"eval_accuracy": 0.
|
| 725 |
-
"eval_f1_macro": 0.
|
| 726 |
-
"eval_loss": 2.
|
| 727 |
-
"eval_runtime": 12.
|
| 728 |
-
"eval_samples_per_second":
|
| 729 |
-
"eval_steps_per_second": 5.
|
| 730 |
"step": 900
|
| 731 |
},
|
| 732 |
{
|
| 733 |
"epoch": 2.292191435768262,
|
| 734 |
-
"grad_norm":
|
| 735 |
-
"learning_rate":
|
| 736 |
-
"loss": 0.
|
| 737 |
"step": 910
|
| 738 |
},
|
| 739 |
{
|
| 740 |
"epoch": 2.3173803526448364,
|
| 741 |
-
"grad_norm":
|
| 742 |
-
"learning_rate":
|
| 743 |
-
"loss": 0.
|
| 744 |
"step": 920
|
| 745 |
},
|
| 746 |
{
|
| 747 |
"epoch": 2.3425692695214106,
|
| 748 |
-
"grad_norm":
|
| 749 |
-
"learning_rate":
|
| 750 |
-
"loss": 0.
|
| 751 |
"step": 930
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 2.367758186397985,
|
| 755 |
-
"grad_norm":
|
| 756 |
-
"learning_rate":
|
| 757 |
-
"loss": 0.
|
| 758 |
"step": 940
|
| 759 |
},
|
| 760 |
{
|
| 761 |
"epoch": 2.392947103274559,
|
| 762 |
-
"grad_norm":
|
| 763 |
-
"learning_rate":
|
| 764 |
-
"loss": 0.
|
| 765 |
"step": 950
|
| 766 |
},
|
| 767 |
{
|
| 768 |
"epoch": 2.4181360201511337,
|
| 769 |
-
"grad_norm":
|
| 770 |
-
"learning_rate":
|
| 771 |
-
"loss": 0.
|
| 772 |
"step": 960
|
| 773 |
},
|
| 774 |
{
|
| 775 |
"epoch": 2.443324937027708,
|
| 776 |
-
"grad_norm":
|
| 777 |
-
"learning_rate":
|
| 778 |
-
"loss": 0.
|
| 779 |
"step": 970
|
| 780 |
},
|
| 781 |
{
|
| 782 |
"epoch": 2.468513853904282,
|
| 783 |
-
"grad_norm":
|
| 784 |
-
"learning_rate":
|
| 785 |
-
"loss": 0.
|
| 786 |
"step": 980
|
| 787 |
},
|
| 788 |
{
|
| 789 |
"epoch": 2.4937027707808563,
|
| 790 |
-
"grad_norm":
|
| 791 |
-
"learning_rate":
|
| 792 |
-
"loss": 0.
|
| 793 |
"step": 990
|
| 794 |
},
|
| 795 |
{
|
| 796 |
"epoch": 2.5188916876574305,
|
| 797 |
-
"grad_norm":
|
| 798 |
-
"learning_rate":
|
| 799 |
-
"loss": 0.
|
| 800 |
"step": 1000
|
| 801 |
},
|
| 802 |
{
|
| 803 |
"epoch": 2.5188916876574305,
|
| 804 |
-
"eval_accuracy": 0.
|
| 805 |
-
"eval_f1_macro": 0.
|
| 806 |
-
"eval_loss": 2.
|
| 807 |
-
"eval_runtime":
|
| 808 |
-
"eval_samples_per_second":
|
| 809 |
-
"eval_steps_per_second": 5.
|
| 810 |
"step": 1000
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 2.544080604534005,
|
| 814 |
-
"grad_norm":
|
| 815 |
-
"learning_rate":
|
| 816 |
-
"loss": 0.
|
| 817 |
"step": 1010
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 2.5692695214105794,
|
| 821 |
-
"grad_norm":
|
| 822 |
-
"learning_rate":
|
| 823 |
-
"loss": 0.
|
| 824 |
"step": 1020
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 2.5944584382871536,
|
| 828 |
-
"grad_norm":
|
| 829 |
-
"learning_rate":
|
| 830 |
-
"loss": 0.
|
| 831 |
"step": 1030
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 2.619647355163728,
|
| 835 |
-
"grad_norm":
|
| 836 |
-
"learning_rate":
|
| 837 |
-
"loss": 0.
|
| 838 |
"step": 1040
|
| 839 |
},
|
| 840 |
{
|
| 841 |
"epoch": 2.644836272040302,
|
| 842 |
-
"grad_norm":
|
| 843 |
-
"learning_rate":
|
| 844 |
-
"loss": 0.
|
| 845 |
"step": 1050
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 2.6700251889168767,
|
| 849 |
-
"grad_norm":
|
| 850 |
-
"learning_rate":
|
| 851 |
-
"loss": 0.
|
| 852 |
"step": 1060
|
| 853 |
},
|
| 854 |
{
|
| 855 |
"epoch": 2.695214105793451,
|
| 856 |
-
"grad_norm":
|
| 857 |
-
"learning_rate":
|
| 858 |
-
"loss": 0.
|
| 859 |
"step": 1070
|
| 860 |
},
|
| 861 |
{
|
| 862 |
"epoch": 2.720403022670025,
|
| 863 |
-
"grad_norm":
|
| 864 |
-
"learning_rate":
|
| 865 |
-
"loss": 0.
|
| 866 |
"step": 1080
|
| 867 |
},
|
| 868 |
{
|
| 869 |
"epoch": 2.7455919395465997,
|
| 870 |
-
"grad_norm":
|
| 871 |
-
"learning_rate":
|
| 872 |
-
"loss": 0.
|
| 873 |
"step": 1090
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 2.770780856423174,
|
| 877 |
-
"grad_norm":
|
| 878 |
-
"learning_rate":
|
| 879 |
-
"loss": 0.
|
| 880 |
"step": 1100
|
| 881 |
},
|
| 882 |
{
|
| 883 |
"epoch": 2.770780856423174,
|
| 884 |
-
"eval_accuracy": 0.
|
| 885 |
-
"eval_f1_macro": 0.
|
| 886 |
-
"eval_loss": 2.
|
| 887 |
-
"eval_runtime":
|
| 888 |
-
"eval_samples_per_second":
|
| 889 |
-
"eval_steps_per_second": 5.
|
| 890 |
"step": 1100
|
| 891 |
},
|
| 892 |
{
|
| 893 |
"epoch": 2.795969773299748,
|
| 894 |
-
"grad_norm":
|
| 895 |
-
"learning_rate":
|
| 896 |
-
"loss": 0.
|
| 897 |
"step": 1110
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 2.8211586901763224,
|
| 901 |
-
"grad_norm":
|
| 902 |
-
"learning_rate":
|
| 903 |
-
"loss": 0.
|
| 904 |
"step": 1120
|
| 905 |
},
|
| 906 |
{
|
| 907 |
"epoch": 2.8463476070528966,
|
| 908 |
-
"grad_norm":
|
| 909 |
-
"learning_rate":
|
| 910 |
-
"loss": 0.
|
| 911 |
"step": 1130
|
| 912 |
},
|
| 913 |
{
|
| 914 |
"epoch": 2.8715365239294712,
|
| 915 |
-
"grad_norm":
|
| 916 |
-
"learning_rate":
|
| 917 |
-
"loss": 0.
|
| 918 |
"step": 1140
|
| 919 |
},
|
| 920 |
{
|
| 921 |
"epoch": 2.8967254408060454,
|
| 922 |
-
"grad_norm":
|
| 923 |
-
"learning_rate":
|
| 924 |
-
"loss": 0.
|
| 925 |
"step": 1150
|
| 926 |
},
|
| 927 |
{
|
| 928 |
"epoch": 2.9219143576826196,
|
| 929 |
-
"grad_norm":
|
| 930 |
-
"learning_rate":
|
| 931 |
-
"loss": 0.
|
| 932 |
"step": 1160
|
| 933 |
},
|
| 934 |
{
|
| 935 |
"epoch": 2.947103274559194,
|
| 936 |
-
"grad_norm":
|
| 937 |
-
"learning_rate":
|
| 938 |
-
"loss": 0.
|
| 939 |
"step": 1170
|
| 940 |
},
|
| 941 |
{
|
| 942 |
"epoch": 2.972292191435768,
|
| 943 |
-
"grad_norm":
|
| 944 |
-
"learning_rate":
|
| 945 |
-
"loss": 0.
|
| 946 |
"step": 1180
|
| 947 |
},
|
| 948 |
{
|
| 949 |
"epoch": 2.9974811083123427,
|
| 950 |
-
"grad_norm":
|
| 951 |
-
"learning_rate":
|
| 952 |
-
"loss": 0.
|
| 953 |
"step": 1190
|
| 954 |
},
|
| 955 |
{
|
| 956 |
"epoch": 3.022670025188917,
|
| 957 |
-
"grad_norm":
|
| 958 |
-
"learning_rate":
|
| 959 |
-
"loss": 0.
|
| 960 |
"step": 1200
|
| 961 |
},
|
| 962 |
{
|
| 963 |
"epoch": 3.022670025188917,
|
| 964 |
-
"eval_accuracy": 0.
|
| 965 |
-
"eval_f1_macro": 0.
|
| 966 |
-
"eval_loss":
|
| 967 |
-
"eval_runtime":
|
| 968 |
-
"eval_samples_per_second":
|
| 969 |
-
"eval_steps_per_second": 5.
|
| 970 |
"step": 1200
|
| 971 |
},
|
| 972 |
{
|
| 973 |
"epoch": 3.047858942065491,
|
| 974 |
-
"grad_norm":
|
| 975 |
-
"learning_rate":
|
| 976 |
-
"loss": 0.
|
| 977 |
"step": 1210
|
| 978 |
},
|
| 979 |
{
|
| 980 |
"epoch": 3.0730478589420653,
|
| 981 |
-
"grad_norm":
|
| 982 |
-
"learning_rate":
|
| 983 |
-
"loss": 0.
|
| 984 |
"step": 1220
|
| 985 |
},
|
| 986 |
{
|
| 987 |
"epoch": 3.09823677581864,
|
| 988 |
-
"grad_norm":
|
| 989 |
-
"learning_rate":
|
| 990 |
-
"loss": 0.
|
| 991 |
"step": 1230
|
| 992 |
},
|
| 993 |
{
|
| 994 |
"epoch": 3.123425692695214,
|
| 995 |
-
"grad_norm":
|
| 996 |
-
"learning_rate":
|
| 997 |
-
"loss": 0.
|
| 998 |
"step": 1240
|
| 999 |
},
|
| 1000 |
{
|
| 1001 |
"epoch": 3.1486146095717884,
|
| 1002 |
-
"grad_norm":
|
| 1003 |
-
"learning_rate":
|
| 1004 |
-
"loss": 0.
|
| 1005 |
"step": 1250
|
| 1006 |
},
|
| 1007 |
{
|
| 1008 |
"epoch": 3.1738035264483626,
|
| 1009 |
-
"grad_norm":
|
| 1010 |
-
"learning_rate":
|
| 1011 |
-
"loss": 0.
|
| 1012 |
"step": 1260
|
| 1013 |
},
|
| 1014 |
{
|
| 1015 |
"epoch": 3.1989924433249373,
|
| 1016 |
-
"grad_norm":
|
| 1017 |
-
"learning_rate":
|
| 1018 |
-
"loss": 0.
|
| 1019 |
"step": 1270
|
| 1020 |
},
|
| 1021 |
{
|
| 1022 |
"epoch": 3.2241813602015115,
|
| 1023 |
-
"grad_norm":
|
| 1024 |
-
"learning_rate":
|
| 1025 |
-
"loss": 0.
|
| 1026 |
"step": 1280
|
| 1027 |
},
|
| 1028 |
{
|
| 1029 |
"epoch": 3.2493702770780857,
|
| 1030 |
-
"grad_norm":
|
| 1031 |
-
"learning_rate":
|
| 1032 |
-
"loss": 0.
|
| 1033 |
"step": 1290
|
| 1034 |
},
|
| 1035 |
{
|
| 1036 |
"epoch": 3.27455919395466,
|
| 1037 |
-
"grad_norm":
|
| 1038 |
-
"learning_rate":
|
| 1039 |
-
"loss": 0.
|
| 1040 |
"step": 1300
|
| 1041 |
},
|
| 1042 |
{
|
| 1043 |
"epoch": 3.27455919395466,
|
| 1044 |
-
"eval_accuracy": 0.
|
| 1045 |
-
"eval_f1_macro": 0.
|
| 1046 |
-
"eval_loss":
|
| 1047 |
-
"eval_runtime":
|
| 1048 |
-
"eval_samples_per_second":
|
| 1049 |
-
"eval_steps_per_second": 5.
|
| 1050 |
"step": 1300
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
"epoch": 3.299748110831234,
|
| 1054 |
-
"grad_norm":
|
| 1055 |
-
"learning_rate":
|
| 1056 |
-
"loss": 0.
|
| 1057 |
"step": 1310
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
"epoch": 3.3249370277078087,
|
| 1061 |
-
"grad_norm":
|
| 1062 |
-
"learning_rate":
|
| 1063 |
-
"loss": 0.
|
| 1064 |
"step": 1320
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
"epoch": 3.350125944584383,
|
| 1068 |
-
"grad_norm":
|
| 1069 |
-
"learning_rate":
|
| 1070 |
-
"loss": 0.
|
| 1071 |
"step": 1330
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
"epoch": 3.375314861460957,
|
| 1075 |
-
"grad_norm":
|
| 1076 |
-
"learning_rate":
|
| 1077 |
-
"loss": 0.
|
| 1078 |
"step": 1340
|
| 1079 |
},
|
| 1080 |
{
|
| 1081 |
"epoch": 3.4005037783375314,
|
| 1082 |
-
"grad_norm":
|
| 1083 |
-
"learning_rate":
|
| 1084 |
-
"loss": 0.
|
| 1085 |
"step": 1350
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
"epoch": 3.4256926952141056,
|
| 1089 |
-
"grad_norm":
|
| 1090 |
-
"learning_rate":
|
| 1091 |
-
"loss": 0.
|
| 1092 |
"step": 1360
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
"epoch": 3.4508816120906802,
|
| 1096 |
-
"grad_norm":
|
| 1097 |
-
"learning_rate":
|
| 1098 |
-
"loss": 0.
|
| 1099 |
"step": 1370
|
| 1100 |
},
|
| 1101 |
{
|
| 1102 |
"epoch": 3.4760705289672544,
|
| 1103 |
-
"grad_norm":
|
| 1104 |
-
"learning_rate":
|
| 1105 |
-
"loss": 0.
|
| 1106 |
"step": 1380
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
"epoch": 3.5012594458438286,
|
| 1110 |
-
"grad_norm":
|
| 1111 |
-
"learning_rate":
|
| 1112 |
-
"loss": 0.
|
| 1113 |
"step": 1390
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
"epoch": 3.5264483627204033,
|
| 1117 |
-
"grad_norm":
|
| 1118 |
-
"learning_rate":
|
| 1119 |
-
"loss": 0.
|
| 1120 |
"step": 1400
|
| 1121 |
},
|
| 1122 |
{
|
| 1123 |
"epoch": 3.5264483627204033,
|
| 1124 |
-
"eval_accuracy": 0.
|
| 1125 |
-
"eval_f1_macro": 0.
|
| 1126 |
-
"eval_loss":
|
| 1127 |
-
"eval_runtime":
|
| 1128 |
-
"eval_samples_per_second":
|
| 1129 |
-
"eval_steps_per_second": 5.
|
| 1130 |
"step": 1400
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
"epoch": 3.551637279596977,
|
| 1134 |
-
"grad_norm":
|
| 1135 |
-
"learning_rate":
|
| 1136 |
-
"loss": 0.
|
| 1137 |
"step": 1410
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
"epoch": 3.5768261964735517,
|
| 1141 |
-
"grad_norm":
|
| 1142 |
-
"learning_rate":
|
| 1143 |
-
"loss": 0.
|
| 1144 |
"step": 1420
|
| 1145 |
},
|
| 1146 |
{
|
| 1147 |
"epoch": 3.602015113350126,
|
| 1148 |
-
"grad_norm":
|
| 1149 |
-
"learning_rate":
|
| 1150 |
-
"loss": 0.
|
| 1151 |
"step": 1430
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
"epoch": 3.6272040302267,
|
| 1155 |
-
"grad_norm":
|
| 1156 |
-
"learning_rate":
|
| 1157 |
-
"loss": 0.
|
| 1158 |
"step": 1440
|
| 1159 |
},
|
| 1160 |
{
|
| 1161 |
"epoch": 3.652392947103275,
|
| 1162 |
-
"grad_norm":
|
| 1163 |
-
"learning_rate":
|
| 1164 |
-
"loss": 0.
|
| 1165 |
"step": 1450
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
"epoch": 3.677581863979849,
|
| 1169 |
-
"grad_norm":
|
| 1170 |
-
"learning_rate":
|
| 1171 |
-
"loss": 0.
|
| 1172 |
"step": 1460
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
"epoch": 3.702770780856423,
|
| 1176 |
-
"grad_norm":
|
| 1177 |
-
"learning_rate":
|
| 1178 |
-
"loss": 0.
|
| 1179 |
"step": 1470
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
"epoch": 3.7279596977329974,
|
| 1183 |
-
"grad_norm":
|
| 1184 |
-
"learning_rate":
|
| 1185 |
-
"loss": 0.
|
| 1186 |
"step": 1480
|
| 1187 |
},
|
| 1188 |
{
|
| 1189 |
"epoch": 3.7531486146095716,
|
| 1190 |
-
"grad_norm":
|
| 1191 |
-
"learning_rate":
|
| 1192 |
-
"loss": 0.
|
| 1193 |
"step": 1490
|
| 1194 |
},
|
| 1195 |
{
|
| 1196 |
"epoch": 3.7783375314861463,
|
| 1197 |
-
"grad_norm":
|
| 1198 |
-
"learning_rate":
|
| 1199 |
-
"loss": 0.
|
| 1200 |
"step": 1500
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
"epoch": 3.7783375314861463,
|
| 1204 |
-
"eval_accuracy": 0.
|
| 1205 |
-
"eval_f1_macro": 0.
|
| 1206 |
-
"eval_loss":
|
| 1207 |
-
"eval_runtime":
|
| 1208 |
-
"eval_samples_per_second":
|
| 1209 |
-
"eval_steps_per_second": 5.
|
| 1210 |
"step": 1500
|
| 1211 |
},
|
| 1212 |
{
|
| 1213 |
"epoch": 3.8035264483627205,
|
| 1214 |
-
"grad_norm":
|
| 1215 |
-
"learning_rate":
|
| 1216 |
-
"loss": 0.
|
| 1217 |
"step": 1510
|
| 1218 |
},
|
| 1219 |
{
|
| 1220 |
"epoch": 3.8287153652392947,
|
| 1221 |
-
"grad_norm":
|
| 1222 |
-
"learning_rate":
|
| 1223 |
-
"loss": 0.
|
| 1224 |
"step": 1520
|
| 1225 |
},
|
| 1226 |
{
|
| 1227 |
"epoch": 3.853904282115869,
|
| 1228 |
-
"grad_norm":
|
| 1229 |
-
"learning_rate":
|
| 1230 |
-
"loss": 0.
|
| 1231 |
"step": 1530
|
| 1232 |
},
|
| 1233 |
{
|
| 1234 |
"epoch": 3.879093198992443,
|
| 1235 |
-
"grad_norm":
|
| 1236 |
-
"learning_rate":
|
| 1237 |
-
"loss": 0.
|
| 1238 |
"step": 1540
|
| 1239 |
},
|
| 1240 |
{
|
| 1241 |
"epoch": 3.9042821158690177,
|
| 1242 |
-
"grad_norm":
|
| 1243 |
-
"learning_rate":
|
| 1244 |
-
"loss": 0.
|
| 1245 |
"step": 1550
|
| 1246 |
},
|
| 1247 |
{
|
| 1248 |
"epoch": 3.929471032745592,
|
| 1249 |
-
"grad_norm":
|
| 1250 |
-
"learning_rate":
|
| 1251 |
-
"loss": 0.
|
| 1252 |
"step": 1560
|
| 1253 |
},
|
| 1254 |
{
|
| 1255 |
"epoch": 3.954659949622166,
|
| 1256 |
-
"grad_norm":
|
| 1257 |
-
"learning_rate":
|
| 1258 |
-
"loss": 0.
|
| 1259 |
"step": 1570
|
| 1260 |
},
|
| 1261 |
{
|
| 1262 |
"epoch": 3.979848866498741,
|
| 1263 |
-
"grad_norm":
|
| 1264 |
-
"learning_rate":
|
| 1265 |
-
"loss": 0.
|
| 1266 |
"step": 1580
|
| 1267 |
},
|
| 1268 |
{
|
| 1269 |
"epoch": 4.005037783375315,
|
| 1270 |
-
"grad_norm":
|
| 1271 |
-
"learning_rate":
|
| 1272 |
-
"loss": 0.
|
| 1273 |
"step": 1590
|
| 1274 |
},
|
| 1275 |
{
|
| 1276 |
"epoch": 4.030226700251889,
|
| 1277 |
-
"grad_norm":
|
| 1278 |
-
"learning_rate":
|
| 1279 |
-
"loss": 0.
|
| 1280 |
"step": 1600
|
| 1281 |
},
|
| 1282 |
{
|
| 1283 |
"epoch": 4.030226700251889,
|
| 1284 |
-
"eval_accuracy": 0.
|
| 1285 |
-
"eval_f1_macro": 0.
|
| 1286 |
-
"eval_loss":
|
| 1287 |
-
"eval_runtime":
|
| 1288 |
-
"eval_samples_per_second":
|
| 1289 |
-
"eval_steps_per_second": 5.
|
| 1290 |
"step": 1600
|
| 1291 |
},
|
| 1292 |
{
|
| 1293 |
"epoch": 4.055415617128464,
|
| 1294 |
-
"grad_norm":
|
| 1295 |
-
"learning_rate":
|
| 1296 |
-
"loss": 0.
|
| 1297 |
"step": 1610
|
| 1298 |
},
|
| 1299 |
{
|
| 1300 |
"epoch": 4.080604534005038,
|
| 1301 |
-
"grad_norm":
|
| 1302 |
-
"learning_rate":
|
| 1303 |
-
"loss": 0.
|
| 1304 |
"step": 1620
|
| 1305 |
},
|
| 1306 |
{
|
| 1307 |
"epoch": 4.105793450881612,
|
| 1308 |
-
"grad_norm":
|
| 1309 |
-
"learning_rate":
|
| 1310 |
-
"loss": 0.
|
| 1311 |
"step": 1630
|
| 1312 |
},
|
| 1313 |
{
|
| 1314 |
"epoch": 4.130982367758186,
|
| 1315 |
-
"grad_norm":
|
| 1316 |
-
"learning_rate":
|
| 1317 |
-
"loss": 0.
|
| 1318 |
"step": 1640
|
| 1319 |
},
|
| 1320 |
{
|
| 1321 |
"epoch": 4.156171284634761,
|
| 1322 |
-
"grad_norm":
|
| 1323 |
-
"learning_rate":
|
| 1324 |
-
"loss": 0.
|
| 1325 |
"step": 1650
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
"epoch": 4.181360201511335,
|
| 1329 |
-
"grad_norm":
|
| 1330 |
-
"learning_rate":
|
| 1331 |
-
"loss": 0.
|
| 1332 |
"step": 1660
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
"epoch": 4.206549118387909,
|
| 1336 |
-
"grad_norm":
|
| 1337 |
-
"learning_rate":
|
| 1338 |
-
"loss": 0.
|
| 1339 |
"step": 1670
|
| 1340 |
},
|
| 1341 |
{
|
| 1342 |
"epoch": 4.231738035264484,
|
| 1343 |
-
"grad_norm":
|
| 1344 |
-
"learning_rate":
|
| 1345 |
-
"loss": 0.
|
| 1346 |
"step": 1680
|
| 1347 |
},
|
| 1348 |
{
|
| 1349 |
"epoch": 4.2569269521410575,
|
| 1350 |
-
"grad_norm":
|
| 1351 |
-
"learning_rate":
|
| 1352 |
-
"loss": 0.
|
| 1353 |
"step": 1690
|
| 1354 |
},
|
| 1355 |
{
|
| 1356 |
"epoch": 4.282115869017632,
|
| 1357 |
-
"grad_norm":
|
| 1358 |
-
"learning_rate":
|
| 1359 |
-
"loss": 0.
|
| 1360 |
"step": 1700
|
| 1361 |
},
|
| 1362 |
{
|
| 1363 |
"epoch": 4.282115869017632,
|
| 1364 |
-
"eval_accuracy": 0.
|
| 1365 |
-
"eval_f1_macro": 0.
|
| 1366 |
-
"eval_loss": 3.
|
| 1367 |
-
"eval_runtime":
|
| 1368 |
-
"eval_samples_per_second":
|
| 1369 |
-
"eval_steps_per_second": 5.
|
| 1370 |
"step": 1700
|
| 1371 |
},
|
| 1372 |
{
|
| 1373 |
"epoch": 4.307304785894207,
|
| 1374 |
-
"grad_norm":
|
| 1375 |
-
"learning_rate":
|
| 1376 |
-
"loss": 0.
|
| 1377 |
"step": 1710
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
"epoch": 4.332493702770781,
|
| 1381 |
-
"grad_norm":
|
| 1382 |
-
"learning_rate": 7.
|
| 1383 |
-
"loss": 0.
|
| 1384 |
"step": 1720
|
| 1385 |
},
|
| 1386 |
{
|
| 1387 |
"epoch": 4.357682619647355,
|
| 1388 |
-
"grad_norm":
|
| 1389 |
-
"learning_rate":
|
| 1390 |
-
"loss": 0.
|
| 1391 |
"step": 1730
|
| 1392 |
},
|
| 1393 |
{
|
| 1394 |
"epoch": 4.382871536523929,
|
| 1395 |
-
"grad_norm":
|
| 1396 |
-
"learning_rate":
|
| 1397 |
-
"loss": 0.
|
| 1398 |
"step": 1740
|
| 1399 |
},
|
| 1400 |
{
|
| 1401 |
"epoch": 4.408060453400504,
|
| 1402 |
-
"grad_norm":
|
| 1403 |
-
"learning_rate":
|
| 1404 |
-
"loss": 0.
|
| 1405 |
"step": 1750
|
| 1406 |
},
|
| 1407 |
{
|
| 1408 |
"epoch": 4.433249370277078,
|
| 1409 |
-
"grad_norm":
|
| 1410 |
-
"learning_rate":
|
| 1411 |
-
"loss": 0.
|
| 1412 |
"step": 1760
|
| 1413 |
},
|
| 1414 |
{
|
| 1415 |
"epoch": 4.458438287153652,
|
| 1416 |
-
"grad_norm":
|
| 1417 |
-
"learning_rate":
|
| 1418 |
-
"loss": 0.
|
| 1419 |
"step": 1770
|
| 1420 |
},
|
| 1421 |
{
|
| 1422 |
"epoch": 4.483627204030227,
|
| 1423 |
-
"grad_norm":
|
| 1424 |
-
"learning_rate":
|
| 1425 |
-
"loss": 0.
|
| 1426 |
"step": 1780
|
| 1427 |
},
|
| 1428 |
{
|
| 1429 |
-
"epoch": 4.
|
| 1430 |
-
"
|
| 1431 |
-
"
|
| 1432 |
-
"
|
| 1433 |
-
"
|
| 1434 |
-
|
| 1435 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1436 |
}
|
| 1437 |
],
|
| 1438 |
"logging_steps": 10,
|
| 1439 |
-
"max_steps":
|
| 1440 |
"num_input_tokens_seen": 0,
|
| 1441 |
-
"num_train_epochs":
|
| 1442 |
"save_steps": 100,
|
| 1443 |
"stateful_callbacks": {
|
| 1444 |
"TrainerControl": {
|
|
@@ -1452,7 +2245,7 @@
|
|
| 1452 |
"attributes": {}
|
| 1453 |
}
|
| 1454 |
},
|
| 1455 |
-
"total_flos":
|
| 1456 |
"train_batch_size": 16,
|
| 1457 |
"trial_name": null,
|
| 1458 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 1100,
|
| 3 |
+
"best_metric": 2.467733144760132,
|
| 4 |
+
"best_model_checkpoint": "./swin-ena24/checkpoint-1100",
|
| 5 |
+
"epoch": 7.0,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 2779,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.02518891687657431,
|
| 14 |
+
"grad_norm": 254123.9375,
|
| 15 |
+
"learning_rate": 0.00019935228499460238,
|
| 16 |
+
"loss": 2.9691,
|
| 17 |
"step": 10
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.05037783375314862,
|
| 21 |
+
"grad_norm": 426010.46875,
|
| 22 |
+
"learning_rate": 0.00019863260165527168,
|
| 23 |
+
"loss": 2.8919,
|
| 24 |
"step": 20
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.07556675062972293,
|
| 28 |
+
"grad_norm": 438250.125,
|
| 29 |
+
"learning_rate": 0.000197912918315941,
|
| 30 |
+
"loss": 2.7195,
|
| 31 |
"step": 30
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.10075566750629723,
|
| 35 |
+
"grad_norm": 643403.375,
|
| 36 |
+
"learning_rate": 0.00019719323497661032,
|
| 37 |
+
"loss": 2.567,
|
| 38 |
"step": 40
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.12594458438287154,
|
| 42 |
+
"grad_norm": 416875.5,
|
| 43 |
+
"learning_rate": 0.0001964735516372796,
|
| 44 |
+
"loss": 2.6196,
|
| 45 |
"step": 50
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.15113350125944586,
|
| 49 |
+
"grad_norm": 412576.78125,
|
| 50 |
+
"learning_rate": 0.0001957538682979489,
|
| 51 |
+
"loss": 2.4359,
|
| 52 |
"step": 60
|
| 53 |
},
|
| 54 |
{
|
| 55 |
"epoch": 0.17632241813602015,
|
| 56 |
+
"grad_norm": 371569.9375,
|
| 57 |
+
"learning_rate": 0.00019503418495861824,
|
| 58 |
+
"loss": 2.4815,
|
| 59 |
"step": 70
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 0.20151133501259447,
|
| 63 |
+
"grad_norm": 297408.5625,
|
| 64 |
+
"learning_rate": 0.0001943145016192875,
|
| 65 |
+
"loss": 2.508,
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 0.22670025188916876,
|
| 70 |
+
"grad_norm": 405330.71875,
|
| 71 |
+
"learning_rate": 0.00019359481827995682,
|
| 72 |
+
"loss": 2.385,
|
| 73 |
"step": 90
|
| 74 |
},
|
| 75 |
{
|
| 76 |
"epoch": 0.2518891687657431,
|
| 77 |
+
"grad_norm": 578770.75,
|
| 78 |
+
"learning_rate": 0.00019287513494062612,
|
| 79 |
+
"loss": 1.9888,
|
| 80 |
"step": 100
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.2518891687657431,
|
| 84 |
+
"eval_accuracy": 0.1630859375,
|
| 85 |
+
"eval_f1_macro": 0.08928735942932195,
|
| 86 |
+
"eval_loss": 3.412175178527832,
|
| 87 |
+
"eval_runtime": 11.0442,
|
| 88 |
+
"eval_samples_per_second": 92.718,
|
| 89 |
+
"eval_steps_per_second": 5.795,
|
| 90 |
"step": 100
|
| 91 |
},
|
| 92 |
{
|
| 93 |
"epoch": 0.2770780856423174,
|
| 94 |
+
"grad_norm": 510874.125,
|
| 95 |
+
"learning_rate": 0.00019215545160129545,
|
| 96 |
+
"loss": 2.188,
|
| 97 |
"step": 110
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 0.3022670025188917,
|
| 101 |
+
"grad_norm": 513859.9375,
|
| 102 |
+
"learning_rate": 0.00019143576826196473,
|
| 103 |
+
"loss": 2.1743,
|
| 104 |
"step": 120
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 0.327455919395466,
|
| 108 |
+
"grad_norm": 527887.5,
|
| 109 |
+
"learning_rate": 0.00019071608492263404,
|
| 110 |
+
"loss": 2.2732,
|
| 111 |
"step": 130
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 0.3526448362720403,
|
| 115 |
+
"grad_norm": 460399.375,
|
| 116 |
+
"learning_rate": 0.00018999640158330337,
|
| 117 |
+
"loss": 2.1564,
|
| 118 |
"step": 140
|
| 119 |
},
|
| 120 |
{
|
| 121 |
"epoch": 0.3778337531486146,
|
| 122 |
+
"grad_norm": 569056.8125,
|
| 123 |
+
"learning_rate": 0.00018927671824397267,
|
| 124 |
+
"loss": 1.8225,
|
| 125 |
"step": 150
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 0.40302267002518893,
|
| 129 |
+
"grad_norm": 909920.6875,
|
| 130 |
+
"learning_rate": 0.00018855703490464195,
|
| 131 |
+
"loss": 1.771,
|
| 132 |
"step": 160
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 0.4282115869017632,
|
| 136 |
+
"grad_norm": 494884.5,
|
| 137 |
+
"learning_rate": 0.00018783735156531128,
|
| 138 |
+
"loss": 1.8349,
|
| 139 |
"step": 170
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 0.4534005037783375,
|
| 143 |
+
"grad_norm": 550597.375,
|
| 144 |
+
"learning_rate": 0.0001871176682259806,
|
| 145 |
+
"loss": 1.5891,
|
| 146 |
"step": 180
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 0.47858942065491183,
|
| 150 |
+
"grad_norm": 462934.75,
|
| 151 |
+
"learning_rate": 0.00018639798488664987,
|
| 152 |
+
"loss": 1.8926,
|
| 153 |
"step": 190
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 0.5037783375314862,
|
| 157 |
+
"grad_norm": 724803.5,
|
| 158 |
+
"learning_rate": 0.00018567830154731917,
|
| 159 |
+
"loss": 1.6111,
|
| 160 |
"step": 200
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 0.5037783375314862,
|
| 164 |
+
"eval_accuracy": 0.2578125,
|
| 165 |
+
"eval_f1_macro": 0.15353256242114557,
|
| 166 |
+
"eval_loss": 2.9077086448669434,
|
| 167 |
+
"eval_runtime": 11.368,
|
| 168 |
+
"eval_samples_per_second": 90.078,
|
| 169 |
+
"eval_steps_per_second": 5.63,
|
| 170 |
"step": 200
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.5289672544080605,
|
| 174 |
+
"grad_norm": 559555.8125,
|
| 175 |
+
"learning_rate": 0.0001849586182079885,
|
| 176 |
+
"loss": 1.6636,
|
| 177 |
"step": 210
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.5541561712846348,
|
| 181 |
+
"grad_norm": 1021320.3125,
|
| 182 |
+
"learning_rate": 0.0001842389348686578,
|
| 183 |
+
"loss": 1.462,
|
| 184 |
"step": 220
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.5793450881612091,
|
| 188 |
+
"grad_norm": 692250.375,
|
| 189 |
+
"learning_rate": 0.0001835192515293271,
|
| 190 |
+
"loss": 1.5007,
|
| 191 |
"step": 230
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.6045340050377834,
|
| 195 |
+
"grad_norm": 774498.1875,
|
| 196 |
+
"learning_rate": 0.00018279956818999642,
|
| 197 |
+
"loss": 1.5122,
|
| 198 |
"step": 240
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.6297229219143576,
|
| 202 |
+
"grad_norm": 497269.53125,
|
| 203 |
+
"learning_rate": 0.00018207988485066572,
|
| 204 |
+
"loss": 1.3379,
|
| 205 |
"step": 250
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.654911838790932,
|
| 209 |
+
"grad_norm": 786711.4375,
|
| 210 |
+
"learning_rate": 0.00018136020151133503,
|
| 211 |
+
"loss": 1.5591,
|
| 212 |
"step": 260
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.6801007556675063,
|
| 216 |
+
"grad_norm": 561642.375,
|
| 217 |
+
"learning_rate": 0.00018064051817200433,
|
| 218 |
+
"loss": 1.3499,
|
| 219 |
"step": 270
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.7052896725440806,
|
| 223 |
+
"grad_norm": 372517.21875,
|
| 224 |
+
"learning_rate": 0.00017992083483267364,
|
| 225 |
+
"loss": 1.3296,
|
| 226 |
"step": 280
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.7304785894206549,
|
| 230 |
+
"grad_norm": 783448.4375,
|
| 231 |
+
"learning_rate": 0.00017920115149334294,
|
| 232 |
+
"loss": 1.2106,
|
| 233 |
"step": 290
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.7556675062972292,
|
| 237 |
+
"grad_norm": 748525.0,
|
| 238 |
+
"learning_rate": 0.00017848146815401222,
|
| 239 |
+
"loss": 1.1276,
|
| 240 |
"step": 300
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.7556675062972292,
|
| 244 |
+
"eval_accuracy": 0.357421875,
|
| 245 |
+
"eval_f1_macro": 0.28272669387224364,
|
| 246 |
+
"eval_loss": 2.6503801345825195,
|
| 247 |
+
"eval_runtime": 11.5522,
|
| 248 |
+
"eval_samples_per_second": 88.641,
|
| 249 |
+
"eval_steps_per_second": 5.54,
|
| 250 |
"step": 300
|
| 251 |
},
|
| 252 |
{
|
| 253 |
"epoch": 0.7808564231738035,
|
| 254 |
+
"grad_norm": 481374.1875,
|
| 255 |
+
"learning_rate": 0.00017776178481468155,
|
| 256 |
+
"loss": 1.1727,
|
| 257 |
"step": 310
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.8060453400503779,
|
| 261 |
+
"grad_norm": 1038731.1875,
|
| 262 |
+
"learning_rate": 0.00017704210147535086,
|
| 263 |
+
"loss": 1.1783,
|
| 264 |
"step": 320
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 0.8312342569269522,
|
| 268 |
+
"grad_norm": 369614.8125,
|
| 269 |
+
"learning_rate": 0.00017632241813602016,
|
| 270 |
+
"loss": 1.4703,
|
| 271 |
"step": 330
|
| 272 |
},
|
| 273 |
{
|
| 274 |
"epoch": 0.8564231738035264,
|
| 275 |
+
"grad_norm": 411708.9375,
|
| 276 |
+
"learning_rate": 0.00017560273479668947,
|
| 277 |
+
"loss": 1.0237,
|
| 278 |
"step": 340
|
| 279 |
},
|
| 280 |
{
|
| 281 |
"epoch": 0.8816120906801007,
|
| 282 |
+
"grad_norm": 777554.375,
|
| 283 |
+
"learning_rate": 0.00017488305145735877,
|
| 284 |
+
"loss": 1.3113,
|
| 285 |
"step": 350
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 0.906801007556675,
|
| 289 |
+
"grad_norm": 589113.125,
|
| 290 |
+
"learning_rate": 0.00017416336811802808,
|
| 291 |
+
"loss": 1.4282,
|
| 292 |
"step": 360
|
| 293 |
},
|
| 294 |
{
|
| 295 |
"epoch": 0.9319899244332494,
|
| 296 |
+
"grad_norm": 537991.9375,
|
| 297 |
+
"learning_rate": 0.00017344368477869738,
|
| 298 |
+
"loss": 1.0802,
|
| 299 |
"step": 370
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 0.9571788413098237,
|
| 303 |
+
"grad_norm": 745339.875,
|
| 304 |
+
"learning_rate": 0.0001727240014393667,
|
| 305 |
+
"loss": 1.1078,
|
| 306 |
"step": 380
|
| 307 |
},
|
| 308 |
{
|
| 309 |
"epoch": 0.982367758186398,
|
| 310 |
+
"grad_norm": 515597.875,
|
| 311 |
+
"learning_rate": 0.000172004318100036,
|
| 312 |
+
"loss": 1.3261,
|
| 313 |
"step": 390
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 1.0075566750629723,
|
| 317 |
+
"grad_norm": 438800.96875,
|
| 318 |
+
"learning_rate": 0.0001712846347607053,
|
| 319 |
+
"loss": 1.0234,
|
| 320 |
"step": 400
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 1.0075566750629723,
|
| 324 |
+
"eval_accuracy": 0.390625,
|
| 325 |
+
"eval_f1_macro": 0.31557618159267287,
|
| 326 |
+
"eval_loss": 2.572810173034668,
|
| 327 |
+
"eval_runtime": 11.7819,
|
| 328 |
+
"eval_samples_per_second": 86.913,
|
| 329 |
+
"eval_steps_per_second": 5.432,
|
| 330 |
"step": 400
|
| 331 |
},
|
| 332 |
{
|
| 333 |
"epoch": 1.0327455919395465,
|
| 334 |
+
"grad_norm": 360076.0,
|
| 335 |
+
"learning_rate": 0.0001705649514213746,
|
| 336 |
+
"loss": 1.0415,
|
| 337 |
"step": 410
|
| 338 |
},
|
| 339 |
{
|
| 340 |
"epoch": 1.057934508816121,
|
| 341 |
+
"grad_norm": 468955.34375,
|
| 342 |
+
"learning_rate": 0.0001698452680820439,
|
| 343 |
+
"loss": 0.8579,
|
| 344 |
"step": 420
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 1.0831234256926952,
|
| 348 |
+
"grad_norm": 765571.4375,
|
| 349 |
+
"learning_rate": 0.0001691255847427132,
|
| 350 |
+
"loss": 0.9408,
|
| 351 |
"step": 430
|
| 352 |
},
|
| 353 |
{
|
| 354 |
"epoch": 1.1083123425692696,
|
| 355 |
+
"grad_norm": 682154.8125,
|
| 356 |
+
"learning_rate": 0.00016840590140338252,
|
| 357 |
+
"loss": 1.0962,
|
| 358 |
"step": 440
|
| 359 |
},
|
| 360 |
{
|
| 361 |
"epoch": 1.1335012594458438,
|
| 362 |
+
"grad_norm": 861960.5,
|
| 363 |
+
"learning_rate": 0.00016768621806405182,
|
| 364 |
+
"loss": 0.8124,
|
| 365 |
"step": 450
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 1.1586901763224182,
|
| 369 |
+
"grad_norm": 583161.8125,
|
| 370 |
+
"learning_rate": 0.00016696653472472113,
|
| 371 |
+
"loss": 1.1014,
|
| 372 |
"step": 460
|
| 373 |
},
|
| 374 |
{
|
| 375 |
"epoch": 1.1838790931989924,
|
| 376 |
+
"grad_norm": 390205.40625,
|
| 377 |
+
"learning_rate": 0.00016624685138539046,
|
| 378 |
+
"loss": 0.8475,
|
| 379 |
"step": 470
|
| 380 |
},
|
| 381 |
{
|
| 382 |
"epoch": 1.2090680100755669,
|
| 383 |
+
"grad_norm": 293254.3125,
|
| 384 |
+
"learning_rate": 0.00016552716804605974,
|
| 385 |
+
"loss": 1.0482,
|
| 386 |
"step": 480
|
| 387 |
},
|
| 388 |
{
|
| 389 |
"epoch": 1.234256926952141,
|
| 390 |
+
"grad_norm": 571449.25,
|
| 391 |
+
"learning_rate": 0.00016480748470672904,
|
| 392 |
+
"loss": 0.8776,
|
| 393 |
"step": 490
|
| 394 |
},
|
| 395 |
{
|
| 396 |
"epoch": 1.2594458438287153,
|
| 397 |
+
"grad_norm": 362773.25,
|
| 398 |
+
"learning_rate": 0.00016408780136739835,
|
| 399 |
+
"loss": 0.8909,
|
| 400 |
"step": 500
|
| 401 |
},
|
| 402 |
{
|
| 403 |
"epoch": 1.2594458438287153,
|
| 404 |
+
"eval_accuracy": 0.421875,
|
| 405 |
+
"eval_f1_macro": 0.3388453334601594,
|
| 406 |
+
"eval_loss": 2.500704765319824,
|
| 407 |
+
"eval_runtime": 11.4557,
|
| 408 |
+
"eval_samples_per_second": 89.388,
|
| 409 |
+
"eval_steps_per_second": 5.587,
|
| 410 |
"step": 500
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 1.2846347607052897,
|
| 414 |
+
"grad_norm": 351467.09375,
|
| 415 |
+
"learning_rate": 0.00016336811802806765,
|
| 416 |
+
"loss": 0.8244,
|
| 417 |
"step": 510
|
| 418 |
},
|
| 419 |
{
|
| 420 |
"epoch": 1.309823677581864,
|
| 421 |
+
"grad_norm": 894287.0625,
|
| 422 |
+
"learning_rate": 0.00016264843468873696,
|
| 423 |
+
"loss": 1.048,
|
| 424 |
"step": 520
|
| 425 |
},
|
| 426 |
{
|
| 427 |
"epoch": 1.3350125944584383,
|
| 428 |
+
"grad_norm": 474119.75,
|
| 429 |
+
"learning_rate": 0.00016192875134940626,
|
| 430 |
+
"loss": 0.9196,
|
| 431 |
"step": 530
|
| 432 |
},
|
| 433 |
{
|
| 434 |
"epoch": 1.3602015113350125,
|
| 435 |
+
"grad_norm": 1069011.125,
|
| 436 |
+
"learning_rate": 0.0001612090680100756,
|
| 437 |
+
"loss": 0.804,
|
| 438 |
"step": 540
|
| 439 |
},
|
| 440 |
{
|
| 441 |
"epoch": 1.385390428211587,
|
| 442 |
+
"grad_norm": 587531.6875,
|
| 443 |
+
"learning_rate": 0.00016048938467074487,
|
| 444 |
+
"loss": 0.6696,
|
| 445 |
"step": 550
|
| 446 |
},
|
| 447 |
{
|
| 448 |
"epoch": 1.4105793450881612,
|
| 449 |
+
"grad_norm": 879147.25,
|
| 450 |
+
"learning_rate": 0.00015976970133141418,
|
| 451 |
+
"loss": 0.7831,
|
| 452 |
"step": 560
|
| 453 |
},
|
| 454 |
{
|
| 455 |
"epoch": 1.4357682619647356,
|
| 456 |
+
"grad_norm": 219215.859375,
|
| 457 |
+
"learning_rate": 0.0001590500179920835,
|
| 458 |
+
"loss": 0.5681,
|
| 459 |
"step": 570
|
| 460 |
},
|
| 461 |
{
|
| 462 |
"epoch": 1.4609571788413098,
|
| 463 |
+
"grad_norm": 447798.375,
|
| 464 |
+
"learning_rate": 0.00015833033465275279,
|
| 465 |
+
"loss": 0.8663,
|
| 466 |
"step": 580
|
| 467 |
},
|
| 468 |
{
|
| 469 |
"epoch": 1.486146095717884,
|
| 470 |
+
"grad_norm": 708370.625,
|
| 471 |
+
"learning_rate": 0.0001576106513134221,
|
| 472 |
+
"loss": 0.7913,
|
| 473 |
"step": 590
|
| 474 |
},
|
| 475 |
{
|
| 476 |
"epoch": 1.5113350125944585,
|
| 477 |
+
"grad_norm": 822008.0625,
|
| 478 |
+
"learning_rate": 0.0001568909679740914,
|
| 479 |
+
"loss": 0.8008,
|
| 480 |
"step": 600
|
| 481 |
},
|
| 482 |
{
|
| 483 |
"epoch": 1.5113350125944585,
|
| 484 |
+
"eval_accuracy": 0.404296875,
|
| 485 |
+
"eval_f1_macro": 0.3618779716937303,
|
| 486 |
+
"eval_loss": 2.7039053440093994,
|
| 487 |
+
"eval_runtime": 11.8087,
|
| 488 |
+
"eval_samples_per_second": 86.716,
|
| 489 |
+
"eval_steps_per_second": 5.42,
|
| 490 |
"step": 600
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 1.536523929471033,
|
| 494 |
+
"grad_norm": 467169.0625,
|
| 495 |
+
"learning_rate": 0.00015617128463476073,
|
| 496 |
+
"loss": 0.7902,
|
| 497 |
"step": 610
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 1.561712846347607,
|
| 501 |
+
"grad_norm": 385000.3125,
|
| 502 |
+
"learning_rate": 0.00015545160129543,
|
| 503 |
+
"loss": 0.7146,
|
| 504 |
"step": 620
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 1.5869017632241813,
|
| 508 |
+
"grad_norm": 366275.21875,
|
| 509 |
+
"learning_rate": 0.0001547319179560993,
|
| 510 |
+
"loss": 0.6558,
|
| 511 |
"step": 630
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 1.6120906801007555,
|
| 515 |
+
"grad_norm": 432902.1875,
|
| 516 |
+
"learning_rate": 0.00015401223461676864,
|
| 517 |
+
"loss": 0.7669,
|
| 518 |
"step": 640
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 1.63727959697733,
|
| 522 |
+
"grad_norm": 426131.71875,
|
| 523 |
+
"learning_rate": 0.00015329255127743795,
|
| 524 |
+
"loss": 0.7121,
|
| 525 |
"step": 650
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 1.6624685138539044,
|
| 529 |
+
"grad_norm": 426302.0,
|
| 530 |
+
"learning_rate": 0.00015257286793810722,
|
| 531 |
+
"loss": 0.9564,
|
| 532 |
"step": 660
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 1.6876574307304786,
|
| 536 |
+
"grad_norm": 405949.34375,
|
| 537 |
+
"learning_rate": 0.00015185318459877656,
|
| 538 |
+
"loss": 0.6072,
|
| 539 |
"step": 670
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 1.7128463476070528,
|
| 543 |
+
"grad_norm": 319128.53125,
|
| 544 |
+
"learning_rate": 0.00015113350125944586,
|
| 545 |
+
"loss": 0.5537,
|
| 546 |
"step": 680
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 1.7380352644836272,
|
| 550 |
+
"grad_norm": 405533.625,
|
| 551 |
+
"learning_rate": 0.00015041381792011514,
|
| 552 |
+
"loss": 0.8143,
|
| 553 |
"step": 690
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 1.7632241813602016,
|
| 557 |
+
"grad_norm": 357302.3125,
|
| 558 |
+
"learning_rate": 0.00014969413458078447,
|
| 559 |
+
"loss": 0.6885,
|
| 560 |
"step": 700
|
| 561 |
},
|
| 562 |
{
|
| 563 |
"epoch": 1.7632241813602016,
|
| 564 |
+
"eval_accuracy": 0.3701171875,
|
| 565 |
+
"eval_f1_macro": 0.2926079223907959,
|
| 566 |
+
"eval_loss": 3.1089859008789062,
|
| 567 |
+
"eval_runtime": 11.4817,
|
| 568 |
+
"eval_samples_per_second": 89.186,
|
| 569 |
+
"eval_steps_per_second": 5.574,
|
| 570 |
"step": 700
|
| 571 |
},
|
| 572 |
{
|
| 573 |
"epoch": 1.7884130982367759,
|
| 574 |
+
"grad_norm": 448783.3125,
|
| 575 |
+
"learning_rate": 0.00014897445124145378,
|
| 576 |
+
"loss": 0.8981,
|
| 577 |
"step": 710
|
| 578 |
},
|
| 579 |
{
|
| 580 |
"epoch": 1.81360201511335,
|
| 581 |
+
"grad_norm": 663808.1875,
|
| 582 |
+
"learning_rate": 0.00014825476790212308,
|
| 583 |
+
"loss": 0.6493,
|
| 584 |
"step": 720
|
| 585 |
},
|
| 586 |
{
|
| 587 |
"epoch": 1.8387909319899243,
|
| 588 |
+
"grad_norm": 1166550.75,
|
| 589 |
+
"learning_rate": 0.00014753508456279236,
|
| 590 |
+
"loss": 0.6288,
|
| 591 |
"step": 730
|
| 592 |
},
|
| 593 |
{
|
| 594 |
"epoch": 1.8639798488664987,
|
| 595 |
+
"grad_norm": 774232.6875,
|
| 596 |
+
"learning_rate": 0.0001468154012234617,
|
| 597 |
+
"loss": 0.8717,
|
| 598 |
"step": 740
|
| 599 |
},
|
| 600 |
{
|
| 601 |
"epoch": 1.8891687657430731,
|
| 602 |
+
"grad_norm": 565763.0,
|
| 603 |
+
"learning_rate": 0.000146095717884131,
|
| 604 |
+
"loss": 0.7199,
|
| 605 |
"step": 750
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 1.9143576826196473,
|
| 609 |
+
"grad_norm": 365545.625,
|
| 610 |
+
"learning_rate": 0.0001453760345448003,
|
| 611 |
+
"loss": 0.721,
|
| 612 |
"step": 760
|
| 613 |
},
|
| 614 |
{
|
| 615 |
"epoch": 1.9395465994962215,
|
| 616 |
+
"grad_norm": 616748.25,
|
| 617 |
+
"learning_rate": 0.0001446563512054696,
|
| 618 |
+
"loss": 0.6793,
|
| 619 |
"step": 770
|
| 620 |
},
|
| 621 |
{
|
| 622 |
"epoch": 1.964735516372796,
|
| 623 |
+
"grad_norm": 503789.65625,
|
| 624 |
+
"learning_rate": 0.0001439366678661389,
|
| 625 |
+
"loss": 0.5703,
|
| 626 |
"step": 780
|
| 627 |
},
|
| 628 |
{
|
| 629 |
"epoch": 1.9899244332493704,
|
| 630 |
+
"grad_norm": 619143.4375,
|
| 631 |
+
"learning_rate": 0.00014321698452680822,
|
| 632 |
+
"loss": 0.679,
|
| 633 |
"step": 790
|
| 634 |
},
|
| 635 |
{
|
| 636 |
"epoch": 2.0151133501259446,
|
| 637 |
+
"grad_norm": 661451.3125,
|
| 638 |
+
"learning_rate": 0.00014249730118747752,
|
| 639 |
+
"loss": 0.839,
|
| 640 |
"step": 800
|
| 641 |
},
|
| 642 |
{
|
| 643 |
"epoch": 2.0151133501259446,
|
| 644 |
"eval_accuracy": 0.484375,
|
| 645 |
+
"eval_f1_macro": 0.414907768165119,
|
| 646 |
+
"eval_loss": 2.584502935409546,
|
| 647 |
+
"eval_runtime": 11.5153,
|
| 648 |
+
"eval_samples_per_second": 88.925,
|
| 649 |
+
"eval_steps_per_second": 5.558,
|
| 650 |
"step": 800
|
| 651 |
},
|
| 652 |
{
|
| 653 |
"epoch": 2.040302267002519,
|
| 654 |
+
"grad_norm": 316634.78125,
|
| 655 |
+
"learning_rate": 0.00014177761784814683,
|
| 656 |
+
"loss": 0.4317,
|
| 657 |
"step": 810
|
| 658 |
},
|
| 659 |
{
|
| 660 |
"epoch": 2.065491183879093,
|
| 661 |
+
"grad_norm": 347729.625,
|
| 662 |
+
"learning_rate": 0.00014105793450881613,
|
| 663 |
+
"loss": 0.3792,
|
| 664 |
"step": 820
|
| 665 |
},
|
| 666 |
{
|
| 667 |
"epoch": 2.0906801007556677,
|
| 668 |
+
"grad_norm": 659445.4375,
|
| 669 |
+
"learning_rate": 0.00014033825116948544,
|
| 670 |
+
"loss": 0.452,
|
| 671 |
"step": 830
|
| 672 |
},
|
| 673 |
{
|
| 674 |
"epoch": 2.115869017632242,
|
| 675 |
+
"grad_norm": 735795.1875,
|
| 676 |
+
"learning_rate": 0.00013961856783015474,
|
| 677 |
+
"loss": 0.5447,
|
| 678 |
"step": 840
|
| 679 |
},
|
| 680 |
{
|
| 681 |
"epoch": 2.141057934508816,
|
| 682 |
+
"grad_norm": 107777.5390625,
|
| 683 |
+
"learning_rate": 0.00013889888449082404,
|
| 684 |
+
"loss": 0.5247,
|
| 685 |
"step": 850
|
| 686 |
},
|
| 687 |
{
|
| 688 |
"epoch": 2.1662468513853903,
|
| 689 |
+
"grad_norm": 577874.5,
|
| 690 |
+
"learning_rate": 0.00013817920115149335,
|
| 691 |
+
"loss": 0.5108,
|
| 692 |
"step": 860
|
| 693 |
},
|
| 694 |
{
|
| 695 |
"epoch": 2.1914357682619645,
|
| 696 |
+
"grad_norm": 441668.625,
|
| 697 |
+
"learning_rate": 0.00013745951781216265,
|
| 698 |
+
"loss": 0.4324,
|
| 699 |
"step": 870
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 2.216624685138539,
|
| 703 |
+
"grad_norm": 489653.59375,
|
| 704 |
+
"learning_rate": 0.00013673983447283196,
|
| 705 |
+
"loss": 0.4427,
|
| 706 |
"step": 880
|
| 707 |
},
|
| 708 |
{
|
| 709 |
"epoch": 2.2418136020151134,
|
| 710 |
+
"grad_norm": 296346.84375,
|
| 711 |
+
"learning_rate": 0.00013602015113350126,
|
| 712 |
+
"loss": 0.5105,
|
| 713 |
"step": 890
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 2.2670025188916876,
|
| 717 |
+
"grad_norm": 187145.796875,
|
| 718 |
+
"learning_rate": 0.00013530046779417057,
|
| 719 |
+
"loss": 0.325,
|
| 720 |
"step": 900
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 2.2670025188916876,
|
| 724 |
+
"eval_accuracy": 0.5068359375,
|
| 725 |
+
"eval_f1_macro": 0.41278416064519763,
|
| 726 |
+
"eval_loss": 2.5142855644226074,
|
| 727 |
+
"eval_runtime": 12.173,
|
| 728 |
+
"eval_samples_per_second": 84.12,
|
| 729 |
+
"eval_steps_per_second": 5.258,
|
| 730 |
"step": 900
|
| 731 |
},
|
| 732 |
{
|
| 733 |
"epoch": 2.292191435768262,
|
| 734 |
+
"grad_norm": 406714.6875,
|
| 735 |
+
"learning_rate": 0.00013458078445483987,
|
| 736 |
+
"loss": 0.4576,
|
| 737 |
"step": 910
|
| 738 |
},
|
| 739 |
{
|
| 740 |
"epoch": 2.3173803526448364,
|
| 741 |
+
"grad_norm": 379892.40625,
|
| 742 |
+
"learning_rate": 0.00013386110111550918,
|
| 743 |
+
"loss": 0.4179,
|
| 744 |
"step": 920
|
| 745 |
},
|
| 746 |
{
|
| 747 |
"epoch": 2.3425692695214106,
|
| 748 |
+
"grad_norm": 540661.9375,
|
| 749 |
+
"learning_rate": 0.00013314141777617848,
|
| 750 |
+
"loss": 0.3386,
|
| 751 |
"step": 930
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 2.367758186397985,
|
| 755 |
+
"grad_norm": 850949.25,
|
| 756 |
+
"learning_rate": 0.0001324217344368478,
|
| 757 |
+
"loss": 0.5695,
|
| 758 |
"step": 940
|
| 759 |
},
|
| 760 |
{
|
| 761 |
"epoch": 2.392947103274559,
|
| 762 |
+
"grad_norm": 363627.53125,
|
| 763 |
+
"learning_rate": 0.0001317020510975171,
|
| 764 |
+
"loss": 0.3362,
|
| 765 |
"step": 950
|
| 766 |
},
|
| 767 |
{
|
| 768 |
"epoch": 2.4181360201511337,
|
| 769 |
+
"grad_norm": 644468.0625,
|
| 770 |
+
"learning_rate": 0.0001309823677581864,
|
| 771 |
+
"loss": 0.444,
|
| 772 |
"step": 960
|
| 773 |
},
|
| 774 |
{
|
| 775 |
"epoch": 2.443324937027708,
|
| 776 |
+
"grad_norm": 164236.78125,
|
| 777 |
+
"learning_rate": 0.00013026268441885573,
|
| 778 |
+
"loss": 0.285,
|
| 779 |
"step": 970
|
| 780 |
},
|
| 781 |
{
|
| 782 |
"epoch": 2.468513853904282,
|
| 783 |
+
"grad_norm": 688494.375,
|
| 784 |
+
"learning_rate": 0.000129543001079525,
|
| 785 |
+
"loss": 0.4024,
|
| 786 |
"step": 980
|
| 787 |
},
|
| 788 |
{
|
| 789 |
"epoch": 2.4937027707808563,
|
| 790 |
+
"grad_norm": 486211.0625,
|
| 791 |
+
"learning_rate": 0.0001288233177401943,
|
| 792 |
+
"loss": 0.4544,
|
| 793 |
"step": 990
|
| 794 |
},
|
| 795 |
{
|
| 796 |
"epoch": 2.5188916876574305,
|
| 797 |
+
"grad_norm": 273390.03125,
|
| 798 |
+
"learning_rate": 0.00012810363440086365,
|
| 799 |
+
"loss": 0.4501,
|
| 800 |
"step": 1000
|
| 801 |
},
|
| 802 |
{
|
| 803 |
"epoch": 2.5188916876574305,
|
| 804 |
+
"eval_accuracy": 0.4482421875,
|
| 805 |
+
"eval_f1_macro": 0.40562802574511003,
|
| 806 |
+
"eval_loss": 2.7683629989624023,
|
| 807 |
+
"eval_runtime": 11.8873,
|
| 808 |
+
"eval_samples_per_second": 86.142,
|
| 809 |
+
"eval_steps_per_second": 5.384,
|
| 810 |
"step": 1000
|
| 811 |
},
|
| 812 |
{
|
| 813 |
"epoch": 2.544080604534005,
|
| 814 |
+
"grad_norm": 730712.125,
|
| 815 |
+
"learning_rate": 0.00012738395106153292,
|
| 816 |
+
"loss": 0.5234,
|
| 817 |
"step": 1010
|
| 818 |
},
|
| 819 |
{
|
| 820 |
"epoch": 2.5692695214105794,
|
| 821 |
+
"grad_norm": 369909.34375,
|
| 822 |
+
"learning_rate": 0.00012666426772220223,
|
| 823 |
+
"loss": 0.4051,
|
| 824 |
"step": 1020
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 2.5944584382871536,
|
| 828 |
+
"grad_norm": 507635.78125,
|
| 829 |
+
"learning_rate": 0.00012594458438287153,
|
| 830 |
+
"loss": 0.2967,
|
| 831 |
"step": 1030
|
| 832 |
},
|
| 833 |
{
|
| 834 |
"epoch": 2.619647355163728,
|
| 835 |
+
"grad_norm": 688013.125,
|
| 836 |
+
"learning_rate": 0.00012522490104354087,
|
| 837 |
+
"loss": 0.5321,
|
| 838 |
"step": 1040
|
| 839 |
},
|
| 840 |
{
|
| 841 |
"epoch": 2.644836272040302,
|
| 842 |
+
"grad_norm": 505216.40625,
|
| 843 |
+
"learning_rate": 0.00012450521770421014,
|
| 844 |
+
"loss": 0.4807,
|
| 845 |
"step": 1050
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 2.6700251889168767,
|
| 849 |
+
"grad_norm": 252679.53125,
|
| 850 |
+
"learning_rate": 0.00012378553436487945,
|
| 851 |
+
"loss": 0.4088,
|
| 852 |
"step": 1060
|
| 853 |
},
|
| 854 |
{
|
| 855 |
"epoch": 2.695214105793451,
|
| 856 |
+
"grad_norm": 235546.9375,
|
| 857 |
+
"learning_rate": 0.00012306585102554878,
|
| 858 |
+
"loss": 0.393,
|
| 859 |
"step": 1070
|
| 860 |
},
|
| 861 |
{
|
| 862 |
"epoch": 2.720403022670025,
|
| 863 |
+
"grad_norm": 398018.84375,
|
| 864 |
+
"learning_rate": 0.00012234616768621808,
|
| 865 |
+
"loss": 0.3694,
|
| 866 |
"step": 1080
|
| 867 |
},
|
| 868 |
{
|
| 869 |
"epoch": 2.7455919395465997,
|
| 870 |
+
"grad_norm": 374467.3125,
|
| 871 |
+
"learning_rate": 0.00012162648434688738,
|
| 872 |
+
"loss": 0.3599,
|
| 873 |
"step": 1090
|
| 874 |
},
|
| 875 |
{
|
| 876 |
"epoch": 2.770780856423174,
|
| 877 |
+
"grad_norm": 533788.9375,
|
| 878 |
+
"learning_rate": 0.0001209068010075567,
|
| 879 |
+
"loss": 0.3191,
|
| 880 |
"step": 1100
|
| 881 |
},
|
| 882 |
{
|
| 883 |
"epoch": 2.770780856423174,
|
| 884 |
+
"eval_accuracy": 0.5146484375,
|
| 885 |
+
"eval_f1_macro": 0.432831730682114,
|
| 886 |
+
"eval_loss": 2.467733144760132,
|
| 887 |
+
"eval_runtime": 11.9402,
|
| 888 |
+
"eval_samples_per_second": 85.761,
|
| 889 |
+
"eval_steps_per_second": 5.36,
|
| 890 |
"step": 1100
|
| 891 |
},
|
| 892 |
{
|
| 893 |
"epoch": 2.795969773299748,
|
| 894 |
+
"grad_norm": 27070.849609375,
|
| 895 |
+
"learning_rate": 0.00012018711766822599,
|
| 896 |
+
"loss": 0.2222,
|
| 897 |
"step": 1110
|
| 898 |
},
|
| 899 |
{
|
| 900 |
"epoch": 2.8211586901763224,
|
| 901 |
+
"grad_norm": 742011.0625,
|
| 902 |
+
"learning_rate": 0.00011946743432889529,
|
| 903 |
+
"loss": 0.3194,
|
| 904 |
"step": 1120
|
| 905 |
},
|
| 906 |
{
|
| 907 |
"epoch": 2.8463476070528966,
|
| 908 |
+
"grad_norm": 508084.53125,
|
| 909 |
+
"learning_rate": 0.00011874775098956458,
|
| 910 |
+
"loss": 0.5068,
|
| 911 |
"step": 1130
|
| 912 |
},
|
| 913 |
{
|
| 914 |
"epoch": 2.8715365239294712,
|
| 915 |
+
"grad_norm": 472090.21875,
|
| 916 |
+
"learning_rate": 0.00011802806765023391,
|
| 917 |
+
"loss": 0.3457,
|
| 918 |
"step": 1140
|
| 919 |
},
|
| 920 |
{
|
| 921 |
"epoch": 2.8967254408060454,
|
| 922 |
+
"grad_norm": 345946.625,
|
| 923 |
+
"learning_rate": 0.0001173083843109032,
|
| 924 |
+
"loss": 0.4024,
|
| 925 |
"step": 1150
|
| 926 |
},
|
| 927 |
{
|
| 928 |
"epoch": 2.9219143576826196,
|
| 929 |
+
"grad_norm": 116557.78125,
|
| 930 |
+
"learning_rate": 0.00011658870097157251,
|
| 931 |
+
"loss": 0.3248,
|
| 932 |
"step": 1160
|
| 933 |
},
|
| 934 |
{
|
| 935 |
"epoch": 2.947103274559194,
|
| 936 |
+
"grad_norm": 544007.125,
|
| 937 |
+
"learning_rate": 0.00011586901763224183,
|
| 938 |
+
"loss": 0.2957,
|
| 939 |
"step": 1170
|
| 940 |
},
|
| 941 |
{
|
| 942 |
"epoch": 2.972292191435768,
|
| 943 |
+
"grad_norm": 558989.0,
|
| 944 |
+
"learning_rate": 0.00011514933429291112,
|
| 945 |
+
"loss": 0.3386,
|
| 946 |
"step": 1180
|
| 947 |
},
|
| 948 |
{
|
| 949 |
"epoch": 2.9974811083123427,
|
| 950 |
+
"grad_norm": 509623.65625,
|
| 951 |
+
"learning_rate": 0.00011442965095358043,
|
| 952 |
+
"loss": 0.4525,
|
| 953 |
"step": 1190
|
| 954 |
},
|
| 955 |
{
|
| 956 |
"epoch": 3.022670025188917,
|
| 957 |
+
"grad_norm": 374462.28125,
|
| 958 |
+
"learning_rate": 0.00011370996761424974,
|
| 959 |
+
"loss": 0.1664,
|
| 960 |
"step": 1200
|
| 961 |
},
|
| 962 |
{
|
| 963 |
"epoch": 3.022670025188917,
|
| 964 |
+
"eval_accuracy": 0.5361328125,
|
| 965 |
+
"eval_f1_macro": 0.4597358092759388,
|
| 966 |
+
"eval_loss": 2.477670192718506,
|
| 967 |
+
"eval_runtime": 11.8885,
|
| 968 |
+
"eval_samples_per_second": 86.134,
|
| 969 |
+
"eval_steps_per_second": 5.383,
|
| 970 |
"step": 1200
|
| 971 |
},
|
| 972 |
{
|
| 973 |
"epoch": 3.047858942065491,
|
| 974 |
+
"grad_norm": 417895.8125,
|
| 975 |
+
"learning_rate": 0.00011299028427491905,
|
| 976 |
+
"loss": 0.2631,
|
| 977 |
"step": 1210
|
| 978 |
},
|
| 979 |
{
|
| 980 |
"epoch": 3.0730478589420653,
|
| 981 |
+
"grad_norm": 307081.21875,
|
| 982 |
+
"learning_rate": 0.00011227060093558834,
|
| 983 |
+
"loss": 0.1544,
|
| 984 |
"step": 1220
|
| 985 |
},
|
| 986 |
{
|
| 987 |
"epoch": 3.09823677581864,
|
| 988 |
+
"grad_norm": 1034528.125,
|
| 989 |
+
"learning_rate": 0.00011155091759625764,
|
| 990 |
+
"loss": 0.2107,
|
| 991 |
"step": 1230
|
| 992 |
},
|
| 993 |
{
|
| 994 |
"epoch": 3.123425692695214,
|
| 995 |
+
"grad_norm": 659800.0625,
|
| 996 |
+
"learning_rate": 0.00011083123425692696,
|
| 997 |
+
"loss": 0.3405,
|
| 998 |
"step": 1240
|
| 999 |
},
|
| 1000 |
{
|
| 1001 |
"epoch": 3.1486146095717884,
|
| 1002 |
+
"grad_norm": 129759.8671875,
|
| 1003 |
+
"learning_rate": 0.00011011155091759627,
|
| 1004 |
+
"loss": 0.2225,
|
| 1005 |
"step": 1250
|
| 1006 |
},
|
| 1007 |
{
|
| 1008 |
"epoch": 3.1738035264483626,
|
| 1009 |
+
"grad_norm": 560157.25,
|
| 1010 |
+
"learning_rate": 0.00010939186757826556,
|
| 1011 |
+
"loss": 0.1545,
|
| 1012 |
"step": 1260
|
| 1013 |
},
|
| 1014 |
{
|
| 1015 |
"epoch": 3.1989924433249373,
|
| 1016 |
+
"grad_norm": 253132.671875,
|
| 1017 |
+
"learning_rate": 0.00010867218423893488,
|
| 1018 |
+
"loss": 0.3073,
|
| 1019 |
"step": 1270
|
| 1020 |
},
|
| 1021 |
{
|
| 1022 |
"epoch": 3.2241813602015115,
|
| 1023 |
+
"grad_norm": 407181.90625,
|
| 1024 |
+
"learning_rate": 0.00010795250089960418,
|
| 1025 |
+
"loss": 0.3247,
|
| 1026 |
"step": 1280
|
| 1027 |
},
|
| 1028 |
{
|
| 1029 |
"epoch": 3.2493702770780857,
|
| 1030 |
+
"grad_norm": 433259.09375,
|
| 1031 |
+
"learning_rate": 0.00010723281756027347,
|
| 1032 |
+
"loss": 0.127,
|
| 1033 |
"step": 1290
|
| 1034 |
},
|
| 1035 |
{
|
| 1036 |
"epoch": 3.27455919395466,
|
| 1037 |
+
"grad_norm": 187479.984375,
|
| 1038 |
+
"learning_rate": 0.0001065131342209428,
|
| 1039 |
+
"loss": 0.1469,
|
| 1040 |
"step": 1300
|
| 1041 |
},
|
| 1042 |
{
|
| 1043 |
"epoch": 3.27455919395466,
|
| 1044 |
+
"eval_accuracy": 0.5205078125,
|
| 1045 |
+
"eval_f1_macro": 0.4495261806770087,
|
| 1046 |
+
"eval_loss": 2.6402528285980225,
|
| 1047 |
+
"eval_runtime": 11.5773,
|
| 1048 |
+
"eval_samples_per_second": 88.449,
|
| 1049 |
+
"eval_steps_per_second": 5.528,
|
| 1050 |
"step": 1300
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
"epoch": 3.299748110831234,
|
| 1054 |
+
"grad_norm": 736520.0625,
|
| 1055 |
+
"learning_rate": 0.0001057934508816121,
|
| 1056 |
+
"loss": 0.2567,
|
| 1057 |
"step": 1310
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
"epoch": 3.3249370277078087,
|
| 1061 |
+
"grad_norm": 376107.03125,
|
| 1062 |
+
"learning_rate": 0.0001050737675422814,
|
| 1063 |
+
"loss": 0.1442,
|
| 1064 |
"step": 1320
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
"epoch": 3.350125944584383,
|
| 1068 |
+
"grad_norm": 723020.3125,
|
| 1069 |
+
"learning_rate": 0.0001043540842029507,
|
| 1070 |
+
"loss": 0.1422,
|
| 1071 |
"step": 1330
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
"epoch": 3.375314861460957,
|
| 1075 |
+
"grad_norm": 60796.48046875,
|
| 1076 |
+
"learning_rate": 0.00010363440086362001,
|
| 1077 |
+
"loss": 0.2077,
|
| 1078 |
"step": 1340
|
| 1079 |
},
|
| 1080 |
{
|
| 1081 |
"epoch": 3.4005037783375314,
|
| 1082 |
+
"grad_norm": 624072.0,
|
| 1083 |
+
"learning_rate": 0.00010291471752428932,
|
| 1084 |
+
"loss": 0.1422,
|
| 1085 |
"step": 1350
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
"epoch": 3.4256926952141056,
|
| 1089 |
+
"grad_norm": 191143.5625,
|
| 1090 |
+
"learning_rate": 0.00010219503418495862,
|
| 1091 |
+
"loss": 0.2425,
|
| 1092 |
"step": 1360
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
"epoch": 3.4508816120906802,
|
| 1096 |
+
"grad_norm": 291693.75,
|
| 1097 |
+
"learning_rate": 0.00010147535084562794,
|
| 1098 |
+
"loss": 0.2693,
|
| 1099 |
"step": 1370
|
| 1100 |
},
|
| 1101 |
{
|
| 1102 |
"epoch": 3.4760705289672544,
|
| 1103 |
+
"grad_norm": 516212.4375,
|
| 1104 |
+
"learning_rate": 0.00010075566750629723,
|
| 1105 |
+
"loss": 0.2397,
|
| 1106 |
"step": 1380
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
"epoch": 3.5012594458438286,
|
| 1110 |
+
"grad_norm": 273365.9375,
|
| 1111 |
+
"learning_rate": 0.00010003598416696654,
|
| 1112 |
+
"loss": 0.1774,
|
| 1113 |
"step": 1390
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
"epoch": 3.5264483627204033,
|
| 1117 |
+
"grad_norm": 518260.21875,
|
| 1118 |
+
"learning_rate": 9.931630082763584e-05,
|
| 1119 |
+
"loss": 0.3063,
|
| 1120 |
"step": 1400
|
| 1121 |
},
|
| 1122 |
{
|
| 1123 |
"epoch": 3.5264483627204033,
|
| 1124 |
+
"eval_accuracy": 0.5009765625,
|
| 1125 |
+
"eval_f1_macro": 0.441499104615912,
|
| 1126 |
+
"eval_loss": 2.7999606132507324,
|
| 1127 |
+
"eval_runtime": 11.7898,
|
| 1128 |
+
"eval_samples_per_second": 86.855,
|
| 1129 |
+
"eval_steps_per_second": 5.428,
|
| 1130 |
"step": 1400
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
"epoch": 3.551637279596977,
|
| 1134 |
+
"grad_norm": 552209.5,
|
| 1135 |
+
"learning_rate": 9.859661748830516e-05,
|
| 1136 |
+
"loss": 0.2254,
|
| 1137 |
"step": 1410
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
"epoch": 3.5768261964735517,
|
| 1141 |
+
"grad_norm": 7195.90625,
|
| 1142 |
+
"learning_rate": 9.787693414897445e-05,
|
| 1143 |
+
"loss": 0.1378,
|
| 1144 |
"step": 1420
|
| 1145 |
},
|
| 1146 |
{
|
| 1147 |
"epoch": 3.602015113350126,
|
| 1148 |
+
"grad_norm": 879434.0625,
|
| 1149 |
+
"learning_rate": 9.715725080964376e-05,
|
| 1150 |
+
"loss": 0.2466,
|
| 1151 |
"step": 1430
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
"epoch": 3.6272040302267,
|
| 1155 |
+
"grad_norm": 116491.9609375,
|
| 1156 |
+
"learning_rate": 9.643756747031306e-05,
|
| 1157 |
+
"loss": 0.2094,
|
| 1158 |
"step": 1440
|
| 1159 |
},
|
| 1160 |
{
|
| 1161 |
"epoch": 3.652392947103275,
|
| 1162 |
+
"grad_norm": 524990.0625,
|
| 1163 |
+
"learning_rate": 9.571788413098237e-05,
|
| 1164 |
+
"loss": 0.2522,
|
| 1165 |
"step": 1450
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
"epoch": 3.677581863979849,
|
| 1169 |
+
"grad_norm": 939990.8125,
|
| 1170 |
+
"learning_rate": 9.499820079165168e-05,
|
| 1171 |
+
"loss": 0.2382,
|
| 1172 |
"step": 1460
|
| 1173 |
},
|
| 1174 |
{
|
| 1175 |
"epoch": 3.702770780856423,
|
| 1176 |
+
"grad_norm": 337334.625,
|
| 1177 |
+
"learning_rate": 9.427851745232098e-05,
|
| 1178 |
+
"loss": 0.299,
|
| 1179 |
"step": 1470
|
| 1180 |
},
|
| 1181 |
{
|
| 1182 |
"epoch": 3.7279596977329974,
|
| 1183 |
+
"grad_norm": 322309.28125,
|
| 1184 |
+
"learning_rate": 9.35588341129903e-05,
|
| 1185 |
+
"loss": 0.2307,
|
| 1186 |
"step": 1480
|
| 1187 |
},
|
| 1188 |
{
|
| 1189 |
"epoch": 3.7531486146095716,
|
| 1190 |
+
"grad_norm": 80700.328125,
|
| 1191 |
+
"learning_rate": 9.283915077365959e-05,
|
| 1192 |
+
"loss": 0.1816,
|
| 1193 |
"step": 1490
|
| 1194 |
},
|
| 1195 |
{
|
| 1196 |
"epoch": 3.7783375314861463,
|
| 1197 |
+
"grad_norm": 667783.125,
|
| 1198 |
+
"learning_rate": 9.21194674343289e-05,
|
| 1199 |
+
"loss": 0.1786,
|
| 1200 |
"step": 1500
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
"epoch": 3.7783375314861463,
|
| 1204 |
+
"eval_accuracy": 0.533203125,
|
| 1205 |
+
"eval_f1_macro": 0.45250014527358284,
|
| 1206 |
+
"eval_loss": 2.8164846897125244,
|
| 1207 |
+
"eval_runtime": 11.728,
|
| 1208 |
+
"eval_samples_per_second": 87.312,
|
| 1209 |
+
"eval_steps_per_second": 5.457,
|
| 1210 |
"step": 1500
|
| 1211 |
},
|
| 1212 |
{
|
| 1213 |
"epoch": 3.8035264483627205,
|
| 1214 |
+
"grad_norm": 340034.53125,
|
| 1215 |
+
"learning_rate": 9.139978409499821e-05,
|
| 1216 |
+
"loss": 0.1934,
|
| 1217 |
"step": 1510
|
| 1218 |
},
|
| 1219 |
{
|
| 1220 |
"epoch": 3.8287153652392947,
|
| 1221 |
+
"grad_norm": 569343.8125,
|
| 1222 |
+
"learning_rate": 9.068010075566751e-05,
|
| 1223 |
+
"loss": 0.3571,
|
| 1224 |
"step": 1520
|
| 1225 |
},
|
| 1226 |
{
|
| 1227 |
"epoch": 3.853904282115869,
|
| 1228 |
+
"grad_norm": 73828.2578125,
|
| 1229 |
+
"learning_rate": 8.996041741633682e-05,
|
| 1230 |
+
"loss": 0.2595,
|
| 1231 |
"step": 1530
|
| 1232 |
},
|
| 1233 |
{
|
| 1234 |
"epoch": 3.879093198992443,
|
| 1235 |
+
"grad_norm": 610885.5625,
|
| 1236 |
+
"learning_rate": 8.924073407700611e-05,
|
| 1237 |
+
"loss": 0.2454,
|
| 1238 |
"step": 1540
|
| 1239 |
},
|
| 1240 |
{
|
| 1241 |
"epoch": 3.9042821158690177,
|
| 1242 |
+
"grad_norm": 549705.5,
|
| 1243 |
+
"learning_rate": 8.852105073767543e-05,
|
| 1244 |
+
"loss": 0.2246,
|
| 1245 |
"step": 1550
|
| 1246 |
},
|
| 1247 |
{
|
| 1248 |
"epoch": 3.929471032745592,
|
| 1249 |
+
"grad_norm": 946495.1875,
|
| 1250 |
+
"learning_rate": 8.780136739834473e-05,
|
| 1251 |
+
"loss": 0.3156,
|
| 1252 |
"step": 1560
|
| 1253 |
},
|
| 1254 |
{
|
| 1255 |
"epoch": 3.954659949622166,
|
| 1256 |
+
"grad_norm": 89126.0546875,
|
| 1257 |
+
"learning_rate": 8.708168405901404e-05,
|
| 1258 |
+
"loss": 0.127,
|
| 1259 |
"step": 1570
|
| 1260 |
},
|
| 1261 |
{
|
| 1262 |
"epoch": 3.979848866498741,
|
| 1263 |
+
"grad_norm": 364322.40625,
|
| 1264 |
+
"learning_rate": 8.636200071968334e-05,
|
| 1265 |
+
"loss": 0.2114,
|
| 1266 |
"step": 1580
|
| 1267 |
},
|
| 1268 |
{
|
| 1269 |
"epoch": 4.005037783375315,
|
| 1270 |
+
"grad_norm": 646444.625,
|
| 1271 |
+
"learning_rate": 8.564231738035265e-05,
|
| 1272 |
+
"loss": 0.1633,
|
| 1273 |
"step": 1590
|
| 1274 |
},
|
| 1275 |
{
|
| 1276 |
"epoch": 4.030226700251889,
|
| 1277 |
+
"grad_norm": 281335.90625,
|
| 1278 |
+
"learning_rate": 8.492263404102195e-05,
|
| 1279 |
+
"loss": 0.0687,
|
| 1280 |
"step": 1600
|
| 1281 |
},
|
| 1282 |
{
|
| 1283 |
"epoch": 4.030226700251889,
|
| 1284 |
+
"eval_accuracy": 0.568359375,
|
| 1285 |
+
"eval_f1_macro": 0.49420184108051124,
|
| 1286 |
+
"eval_loss": 2.9026849269866943,
|
| 1287 |
+
"eval_runtime": 11.365,
|
| 1288 |
+
"eval_samples_per_second": 90.101,
|
| 1289 |
+
"eval_steps_per_second": 5.631,
|
| 1290 |
"step": 1600
|
| 1291 |
},
|
| 1292 |
{
|
| 1293 |
"epoch": 4.055415617128464,
|
| 1294 |
+
"grad_norm": 12626.9794921875,
|
| 1295 |
+
"learning_rate": 8.420295070169126e-05,
|
| 1296 |
+
"loss": 0.0641,
|
| 1297 |
"step": 1610
|
| 1298 |
},
|
| 1299 |
{
|
| 1300 |
"epoch": 4.080604534005038,
|
| 1301 |
+
"grad_norm": 55986.578125,
|
| 1302 |
+
"learning_rate": 8.348326736236056e-05,
|
| 1303 |
+
"loss": 0.0647,
|
| 1304 |
"step": 1620
|
| 1305 |
},
|
| 1306 |
{
|
| 1307 |
"epoch": 4.105793450881612,
|
| 1308 |
+
"grad_norm": 509303.3125,
|
| 1309 |
+
"learning_rate": 8.276358402302987e-05,
|
| 1310 |
+
"loss": 0.0746,
|
| 1311 |
"step": 1630
|
| 1312 |
},
|
| 1313 |
{
|
| 1314 |
"epoch": 4.130982367758186,
|
| 1315 |
+
"grad_norm": 21481.740234375,
|
| 1316 |
+
"learning_rate": 8.204390068369917e-05,
|
| 1317 |
+
"loss": 0.0746,
|
| 1318 |
"step": 1640
|
| 1319 |
},
|
| 1320 |
{
|
| 1321 |
"epoch": 4.156171284634761,
|
| 1322 |
+
"grad_norm": 11360.7412109375,
|
| 1323 |
+
"learning_rate": 8.132421734436848e-05,
|
| 1324 |
+
"loss": 0.0555,
|
| 1325 |
"step": 1650
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
"epoch": 4.181360201511335,
|
| 1329 |
+
"grad_norm": 396739.0,
|
| 1330 |
+
"learning_rate": 8.06045340050378e-05,
|
| 1331 |
+
"loss": 0.0595,
|
| 1332 |
"step": 1660
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
"epoch": 4.206549118387909,
|
| 1336 |
+
"grad_norm": 8099.65478515625,
|
| 1337 |
+
"learning_rate": 7.988485066570709e-05,
|
| 1338 |
+
"loss": 0.1391,
|
| 1339 |
"step": 1670
|
| 1340 |
},
|
| 1341 |
{
|
| 1342 |
"epoch": 4.231738035264484,
|
| 1343 |
+
"grad_norm": 208596.65625,
|
| 1344 |
+
"learning_rate": 7.916516732637639e-05,
|
| 1345 |
+
"loss": 0.1038,
|
| 1346 |
"step": 1680
|
| 1347 |
},
|
| 1348 |
{
|
| 1349 |
"epoch": 4.2569269521410575,
|
| 1350 |
+
"grad_norm": 359488.34375,
|
| 1351 |
+
"learning_rate": 7.84454839870457e-05,
|
| 1352 |
+
"loss": 0.025,
|
| 1353 |
"step": 1690
|
| 1354 |
},
|
| 1355 |
{
|
| 1356 |
"epoch": 4.282115869017632,
|
| 1357 |
+
"grad_norm": 247004.875,
|
| 1358 |
+
"learning_rate": 7.7725800647715e-05,
|
| 1359 |
+
"loss": 0.0427,
|
| 1360 |
"step": 1700
|
| 1361 |
},
|
| 1362 |
{
|
| 1363 |
"epoch": 4.282115869017632,
|
| 1364 |
+
"eval_accuracy": 0.4912109375,
|
| 1365 |
+
"eval_f1_macro": 0.4362345681944286,
|
| 1366 |
+
"eval_loss": 3.321627616882324,
|
| 1367 |
+
"eval_runtime": 11.8095,
|
| 1368 |
+
"eval_samples_per_second": 86.71,
|
| 1369 |
+
"eval_steps_per_second": 5.419,
|
| 1370 |
"step": 1700
|
| 1371 |
},
|
| 1372 |
{
|
| 1373 |
"epoch": 4.307304785894207,
|
| 1374 |
+
"grad_norm": 22134.611328125,
|
| 1375 |
+
"learning_rate": 7.700611730838432e-05,
|
| 1376 |
+
"loss": 0.1015,
|
| 1377 |
"step": 1710
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
"epoch": 4.332493702770781,
|
| 1381 |
+
"grad_norm": 163005.84375,
|
| 1382 |
+
"learning_rate": 7.628643396905361e-05,
|
| 1383 |
+
"loss": 0.0524,
|
| 1384 |
"step": 1720
|
| 1385 |
},
|
| 1386 |
{
|
| 1387 |
"epoch": 4.357682619647355,
|
| 1388 |
+
"grad_norm": 25011.78515625,
|
| 1389 |
+
"learning_rate": 7.556675062972293e-05,
|
| 1390 |
+
"loss": 0.1,
|
| 1391 |
"step": 1730
|
| 1392 |
},
|
| 1393 |
{
|
| 1394 |
"epoch": 4.382871536523929,
|
| 1395 |
+
"grad_norm": 606683.75,
|
| 1396 |
+
"learning_rate": 7.484706729039224e-05,
|
| 1397 |
+
"loss": 0.1503,
|
| 1398 |
"step": 1740
|
| 1399 |
},
|
| 1400 |
{
|
| 1401 |
"epoch": 4.408060453400504,
|
| 1402 |
+
"grad_norm": 411167.28125,
|
| 1403 |
+
"learning_rate": 7.412738395106154e-05,
|
| 1404 |
+
"loss": 0.1075,
|
| 1405 |
"step": 1750
|
| 1406 |
},
|
| 1407 |
{
|
| 1408 |
"epoch": 4.433249370277078,
|
| 1409 |
+
"grad_norm": 20846.5390625,
|
| 1410 |
+
"learning_rate": 7.340770061173085e-05,
|
| 1411 |
+
"loss": 0.0793,
|
| 1412 |
"step": 1760
|
| 1413 |
},
|
| 1414 |
{
|
| 1415 |
"epoch": 4.458438287153652,
|
| 1416 |
+
"grad_norm": 35871.06640625,
|
| 1417 |
+
"learning_rate": 7.268801727240015e-05,
|
| 1418 |
+
"loss": 0.0607,
|
| 1419 |
"step": 1770
|
| 1420 |
},
|
| 1421 |
{
|
| 1422 |
"epoch": 4.483627204030227,
|
| 1423 |
+
"grad_norm": 58382.01171875,
|
| 1424 |
+
"learning_rate": 7.196833393306946e-05,
|
| 1425 |
+
"loss": 0.082,
|
| 1426 |
"step": 1780
|
| 1427 |
},
|
| 1428 |
{
|
| 1429 |
+
"epoch": 4.508816120906801,
|
| 1430 |
+
"grad_norm": 13096.5205078125,
|
| 1431 |
+
"learning_rate": 7.124865059373876e-05,
|
| 1432 |
+
"loss": 0.0754,
|
| 1433 |
+
"step": 1790
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 4.534005037783375,
|
| 1437 |
+
"grad_norm": 21618.96875,
|
| 1438 |
+
"learning_rate": 7.052896725440807e-05,
|
| 1439 |
+
"loss": 0.1825,
|
| 1440 |
+
"step": 1800
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 4.534005037783375,
|
| 1444 |
+
"eval_accuracy": 0.53125,
|
| 1445 |
+
"eval_f1_macro": 0.4663662196170286,
|
| 1446 |
+
"eval_loss": 3.1456074714660645,
|
| 1447 |
+
"eval_runtime": 11.9035,
|
| 1448 |
+
"eval_samples_per_second": 86.025,
|
| 1449 |
+
"eval_steps_per_second": 5.377,
|
| 1450 |
+
"step": 1800
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 4.55919395465995,
|
| 1454 |
+
"grad_norm": 97136.0546875,
|
| 1455 |
+
"learning_rate": 6.980928391507737e-05,
|
| 1456 |
+
"loss": 0.0821,
|
| 1457 |
+
"step": 1810
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 4.584382871536524,
|
| 1461 |
+
"grad_norm": 14197.41796875,
|
| 1462 |
+
"learning_rate": 6.908960057574667e-05,
|
| 1463 |
+
"loss": 0.1245,
|
| 1464 |
+
"step": 1820
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 4.609571788413098,
|
| 1468 |
+
"grad_norm": 792662.5,
|
| 1469 |
+
"learning_rate": 6.836991723641598e-05,
|
| 1470 |
+
"loss": 0.0678,
|
| 1471 |
+
"step": 1830
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 4.634760705289673,
|
| 1475 |
+
"grad_norm": 70302.1484375,
|
| 1476 |
+
"learning_rate": 6.765023389708528e-05,
|
| 1477 |
+
"loss": 0.0369,
|
| 1478 |
+
"step": 1840
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 4.659949622166247,
|
| 1482 |
+
"grad_norm": 315541.75,
|
| 1483 |
+
"learning_rate": 6.693055055775459e-05,
|
| 1484 |
+
"loss": 0.086,
|
| 1485 |
+
"step": 1850
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 4.685138539042821,
|
| 1489 |
+
"grad_norm": 426322.5625,
|
| 1490 |
+
"learning_rate": 6.62108672184239e-05,
|
| 1491 |
+
"loss": 0.1367,
|
| 1492 |
+
"step": 1860
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 4.710327455919396,
|
| 1496 |
+
"grad_norm": 5201.7265625,
|
| 1497 |
+
"learning_rate": 6.54911838790932e-05,
|
| 1498 |
+
"loss": 0.0518,
|
| 1499 |
+
"step": 1870
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 4.73551637279597,
|
| 1503 |
+
"grad_norm": 16552.916015625,
|
| 1504 |
+
"learning_rate": 6.47715005397625e-05,
|
| 1505 |
+
"loss": 0.0857,
|
| 1506 |
+
"step": 1880
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 4.760705289672544,
|
| 1510 |
+
"grad_norm": 74746.0234375,
|
| 1511 |
+
"learning_rate": 6.405181720043182e-05,
|
| 1512 |
+
"loss": 0.178,
|
| 1513 |
+
"step": 1890
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 4.785894206549118,
|
| 1517 |
+
"grad_norm": 30216.1328125,
|
| 1518 |
+
"learning_rate": 6.333213386110111e-05,
|
| 1519 |
+
"loss": 0.0758,
|
| 1520 |
+
"step": 1900
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 4.785894206549118,
|
| 1524 |
+
"eval_accuracy": 0.5546875,
|
| 1525 |
+
"eval_f1_macro": 0.4578418057053371,
|
| 1526 |
+
"eval_loss": 3.2782468795776367,
|
| 1527 |
+
"eval_runtime": 11.4304,
|
| 1528 |
+
"eval_samples_per_second": 89.586,
|
| 1529 |
+
"eval_steps_per_second": 5.599,
|
| 1530 |
+
"step": 1900
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 4.811083123425693,
|
| 1534 |
+
"grad_norm": 122862.0703125,
|
| 1535 |
+
"learning_rate": 6.261245052177043e-05,
|
| 1536 |
+
"loss": 0.1008,
|
| 1537 |
+
"step": 1910
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 4.836272040302267,
|
| 1541 |
+
"grad_norm": 355039.96875,
|
| 1542 |
+
"learning_rate": 6.189276718243972e-05,
|
| 1543 |
+
"loss": 0.0619,
|
| 1544 |
+
"step": 1920
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 4.861460957178841,
|
| 1548 |
+
"grad_norm": 180546.546875,
|
| 1549 |
+
"learning_rate": 6.117308384310904e-05,
|
| 1550 |
+
"loss": 0.212,
|
| 1551 |
+
"step": 1930
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 4.886649874055416,
|
| 1555 |
+
"grad_norm": 1149145.875,
|
| 1556 |
+
"learning_rate": 6.045340050377835e-05,
|
| 1557 |
+
"loss": 0.1821,
|
| 1558 |
+
"step": 1940
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 4.91183879093199,
|
| 1562 |
+
"grad_norm": 197082.046875,
|
| 1563 |
+
"learning_rate": 5.9733717164447645e-05,
|
| 1564 |
+
"loss": 0.0539,
|
| 1565 |
+
"step": 1950
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 4.937027707808564,
|
| 1569 |
+
"grad_norm": 361.24676513671875,
|
| 1570 |
+
"learning_rate": 5.901403382511696e-05,
|
| 1571 |
+
"loss": 0.1526,
|
| 1572 |
+
"step": 1960
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 4.962216624685139,
|
| 1576 |
+
"grad_norm": 362799.34375,
|
| 1577 |
+
"learning_rate": 5.8294350485786255e-05,
|
| 1578 |
+
"loss": 0.0838,
|
| 1579 |
+
"step": 1970
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 4.987405541561713,
|
| 1583 |
+
"grad_norm": 4683.50732421875,
|
| 1584 |
+
"learning_rate": 5.757466714645556e-05,
|
| 1585 |
+
"loss": 0.0995,
|
| 1586 |
+
"step": 1980
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 5.012594458438287,
|
| 1590 |
+
"grad_norm": 5118.85986328125,
|
| 1591 |
+
"learning_rate": 5.685498380712487e-05,
|
| 1592 |
+
"loss": 0.0488,
|
| 1593 |
+
"step": 1990
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 5.037783375314861,
|
| 1597 |
+
"grad_norm": 6831.2880859375,
|
| 1598 |
+
"learning_rate": 5.613530046779417e-05,
|
| 1599 |
+
"loss": 0.0471,
|
| 1600 |
+
"step": 2000
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 5.037783375314861,
|
| 1604 |
+
"eval_accuracy": 0.5517578125,
|
| 1605 |
+
"eval_f1_macro": 0.4725336026660133,
|
| 1606 |
+
"eval_loss": 3.334784507751465,
|
| 1607 |
+
"eval_runtime": 11.4554,
|
| 1608 |
+
"eval_samples_per_second": 89.39,
|
| 1609 |
+
"eval_steps_per_second": 5.587,
|
| 1610 |
+
"step": 2000
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 5.062972292191436,
|
| 1614 |
+
"grad_norm": 88275.8359375,
|
| 1615 |
+
"learning_rate": 5.541561712846348e-05,
|
| 1616 |
+
"loss": 0.0503,
|
| 1617 |
+
"step": 2010
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 5.08816120906801,
|
| 1621 |
+
"grad_norm": 1161589.25,
|
| 1622 |
+
"learning_rate": 5.469593378913278e-05,
|
| 1623 |
+
"loss": 0.0326,
|
| 1624 |
+
"step": 2020
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 5.113350125944584,
|
| 1628 |
+
"grad_norm": 263008.125,
|
| 1629 |
+
"learning_rate": 5.397625044980209e-05,
|
| 1630 |
+
"loss": 0.103,
|
| 1631 |
+
"step": 2030
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 5.138539042821159,
|
| 1635 |
+
"grad_norm": 2268.762939453125,
|
| 1636 |
+
"learning_rate": 5.32565671104714e-05,
|
| 1637 |
+
"loss": 0.0093,
|
| 1638 |
+
"step": 2040
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 5.163727959697733,
|
| 1642 |
+
"grad_norm": 530844.6875,
|
| 1643 |
+
"learning_rate": 5.25368837711407e-05,
|
| 1644 |
+
"loss": 0.0919,
|
| 1645 |
+
"step": 2050
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 5.188916876574307,
|
| 1649 |
+
"grad_norm": 181022.703125,
|
| 1650 |
+
"learning_rate": 5.1817200431810006e-05,
|
| 1651 |
+
"loss": 0.049,
|
| 1652 |
+
"step": 2060
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 5.214105793450882,
|
| 1656 |
+
"grad_norm": 1360.029541015625,
|
| 1657 |
+
"learning_rate": 5.109751709247931e-05,
|
| 1658 |
+
"loss": 0.0656,
|
| 1659 |
+
"step": 2070
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 5.239294710327456,
|
| 1663 |
+
"grad_norm": 541578.0625,
|
| 1664 |
+
"learning_rate": 5.0377833753148616e-05,
|
| 1665 |
+
"loss": 0.0287,
|
| 1666 |
+
"step": 2080
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 5.26448362720403,
|
| 1670 |
+
"grad_norm": 8820.35546875,
|
| 1671 |
+
"learning_rate": 4.965815041381792e-05,
|
| 1672 |
+
"loss": 0.0252,
|
| 1673 |
+
"step": 2090
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 5.289672544080605,
|
| 1677 |
+
"grad_norm": 50239.13671875,
|
| 1678 |
+
"learning_rate": 4.8938467074487226e-05,
|
| 1679 |
+
"loss": 0.0512,
|
| 1680 |
+
"step": 2100
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 5.289672544080605,
|
| 1684 |
+
"eval_accuracy": 0.5283203125,
|
| 1685 |
+
"eval_f1_macro": 0.45143425422893263,
|
| 1686 |
+
"eval_loss": 3.718236207962036,
|
| 1687 |
+
"eval_runtime": 11.3333,
|
| 1688 |
+
"eval_samples_per_second": 90.353,
|
| 1689 |
+
"eval_steps_per_second": 5.647,
|
| 1690 |
+
"step": 2100
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 5.314861460957179,
|
| 1694 |
+
"grad_norm": 1060627.125,
|
| 1695 |
+
"learning_rate": 4.821878373515653e-05,
|
| 1696 |
+
"loss": 0.0412,
|
| 1697 |
+
"step": 2110
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 5.340050377833753,
|
| 1701 |
+
"grad_norm": 258190.28125,
|
| 1702 |
+
"learning_rate": 4.749910039582584e-05,
|
| 1703 |
+
"loss": 0.1177,
|
| 1704 |
+
"step": 2120
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 5.365239294710327,
|
| 1708 |
+
"grad_norm": 126000.2265625,
|
| 1709 |
+
"learning_rate": 4.677941705649515e-05,
|
| 1710 |
+
"loss": 0.0049,
|
| 1711 |
+
"step": 2130
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 5.390428211586902,
|
| 1715 |
+
"grad_norm": 2938.684814453125,
|
| 1716 |
+
"learning_rate": 4.605973371716445e-05,
|
| 1717 |
+
"loss": 0.1016,
|
| 1718 |
+
"step": 2140
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 5.415617128463476,
|
| 1722 |
+
"grad_norm": 191.32302856445312,
|
| 1723 |
+
"learning_rate": 4.534005037783376e-05,
|
| 1724 |
+
"loss": 0.0051,
|
| 1725 |
+
"step": 2150
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 5.44080604534005,
|
| 1729 |
+
"grad_norm": 70.82958984375,
|
| 1730 |
+
"learning_rate": 4.4620367038503055e-05,
|
| 1731 |
+
"loss": 0.0335,
|
| 1732 |
+
"step": 2160
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 5.465994962216625,
|
| 1736 |
+
"grad_norm": 2224.07568359375,
|
| 1737 |
+
"learning_rate": 4.390068369917237e-05,
|
| 1738 |
+
"loss": 0.0019,
|
| 1739 |
+
"step": 2170
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 5.491183879093199,
|
| 1743 |
+
"grad_norm": 7065.70703125,
|
| 1744 |
+
"learning_rate": 4.318100035984167e-05,
|
| 1745 |
+
"loss": 0.0489,
|
| 1746 |
+
"step": 2180
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 5.516372795969773,
|
| 1750 |
+
"grad_norm": 218.8759765625,
|
| 1751 |
+
"learning_rate": 4.246131702051098e-05,
|
| 1752 |
+
"loss": 0.0761,
|
| 1753 |
+
"step": 2190
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 5.541561712846348,
|
| 1757 |
+
"grad_norm": 2789.82763671875,
|
| 1758 |
+
"learning_rate": 4.174163368118028e-05,
|
| 1759 |
+
"loss": 0.0095,
|
| 1760 |
+
"step": 2200
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 5.541561712846348,
|
| 1764 |
+
"eval_accuracy": 0.5341796875,
|
| 1765 |
+
"eval_f1_macro": 0.478510457034419,
|
| 1766 |
+
"eval_loss": 3.902801036834717,
|
| 1767 |
+
"eval_runtime": 11.7659,
|
| 1768 |
+
"eval_samples_per_second": 87.031,
|
| 1769 |
+
"eval_steps_per_second": 5.439,
|
| 1770 |
+
"step": 2200
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 5.566750629722922,
|
| 1774 |
+
"grad_norm": 33521.52734375,
|
| 1775 |
+
"learning_rate": 4.1021950341849587e-05,
|
| 1776 |
+
"loss": 0.0445,
|
| 1777 |
+
"step": 2210
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 5.591939546599496,
|
| 1781 |
+
"grad_norm": 339091.78125,
|
| 1782 |
+
"learning_rate": 4.03022670025189e-05,
|
| 1783 |
+
"loss": 0.0717,
|
| 1784 |
+
"step": 2220
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 5.617128463476071,
|
| 1788 |
+
"grad_norm": 570797.6875,
|
| 1789 |
+
"learning_rate": 3.9582583663188196e-05,
|
| 1790 |
+
"loss": 0.0664,
|
| 1791 |
+
"step": 2230
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 5.642317380352645,
|
| 1795 |
+
"grad_norm": 638298.25,
|
| 1796 |
+
"learning_rate": 3.88629003238575e-05,
|
| 1797 |
+
"loss": 0.0152,
|
| 1798 |
+
"step": 2240
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"epoch": 5.667506297229219,
|
| 1802 |
+
"grad_norm": 1308.215087890625,
|
| 1803 |
+
"learning_rate": 3.8143216984526806e-05,
|
| 1804 |
+
"loss": 0.0029,
|
| 1805 |
+
"step": 2250
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 5.692695214105793,
|
| 1809 |
+
"grad_norm": 47.38713836669922,
|
| 1810 |
+
"learning_rate": 3.742353364519612e-05,
|
| 1811 |
+
"loss": 0.0199,
|
| 1812 |
+
"step": 2260
|
| 1813 |
+
},
|
| 1814 |
+
{
|
| 1815 |
+
"epoch": 5.717884130982368,
|
| 1816 |
+
"grad_norm": 56138.89453125,
|
| 1817 |
+
"learning_rate": 3.670385030586542e-05,
|
| 1818 |
+
"loss": 0.0748,
|
| 1819 |
+
"step": 2270
|
| 1820 |
+
},
|
| 1821 |
+
{
|
| 1822 |
+
"epoch": 5.7430730478589425,
|
| 1823 |
+
"grad_norm": 49.50386047363281,
|
| 1824 |
+
"learning_rate": 3.598416696653473e-05,
|
| 1825 |
+
"loss": 0.0174,
|
| 1826 |
+
"step": 2280
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 5.768261964735516,
|
| 1830 |
+
"grad_norm": 7627.2783203125,
|
| 1831 |
+
"learning_rate": 3.526448362720403e-05,
|
| 1832 |
+
"loss": 0.002,
|
| 1833 |
+
"step": 2290
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 5.793450881612091,
|
| 1837 |
+
"grad_norm": 123.52435302734375,
|
| 1838 |
+
"learning_rate": 3.454480028787334e-05,
|
| 1839 |
+
"loss": 0.0247,
|
| 1840 |
+
"step": 2300
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 5.793450881612091,
|
| 1844 |
+
"eval_accuracy": 0.5712890625,
|
| 1845 |
+
"eval_f1_macro": 0.48791698609028533,
|
| 1846 |
+
"eval_loss": 3.960580348968506,
|
| 1847 |
+
"eval_runtime": 11.8204,
|
| 1848 |
+
"eval_samples_per_second": 86.63,
|
| 1849 |
+
"eval_steps_per_second": 5.414,
|
| 1850 |
+
"step": 2300
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 5.818639798488665,
|
| 1854 |
+
"grad_norm": 534.7195434570312,
|
| 1855 |
+
"learning_rate": 3.382511694854264e-05,
|
| 1856 |
+
"loss": 0.0496,
|
| 1857 |
+
"step": 2310
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 5.843828715365239,
|
| 1861 |
+
"grad_norm": 17011.57421875,
|
| 1862 |
+
"learning_rate": 3.310543360921195e-05,
|
| 1863 |
+
"loss": 0.1476,
|
| 1864 |
+
"step": 2320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 5.869017632241814,
|
| 1868 |
+
"grad_norm": 386354.28125,
|
| 1869 |
+
"learning_rate": 3.238575026988125e-05,
|
| 1870 |
+
"loss": 0.0593,
|
| 1871 |
+
"step": 2330
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 5.894206549118388,
|
| 1875 |
+
"grad_norm": 102287.3828125,
|
| 1876 |
+
"learning_rate": 3.166606693055056e-05,
|
| 1877 |
+
"loss": 0.0117,
|
| 1878 |
+
"step": 2340
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 5.919395465994962,
|
| 1882 |
+
"grad_norm": 29.17066764831543,
|
| 1883 |
+
"learning_rate": 3.094638359121986e-05,
|
| 1884 |
+
"loss": 0.0702,
|
| 1885 |
+
"step": 2350
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 5.944584382871536,
|
| 1889 |
+
"grad_norm": 1611.7230224609375,
|
| 1890 |
+
"learning_rate": 3.0226700251889174e-05,
|
| 1891 |
+
"loss": 0.0687,
|
| 1892 |
+
"step": 2360
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 5.969773299748111,
|
| 1896 |
+
"grad_norm": 168.02320861816406,
|
| 1897 |
+
"learning_rate": 2.950701691255848e-05,
|
| 1898 |
+
"loss": 0.0872,
|
| 1899 |
+
"step": 2370
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 5.994962216624685,
|
| 1903 |
+
"grad_norm": 1058.51904296875,
|
| 1904 |
+
"learning_rate": 2.878733357322778e-05,
|
| 1905 |
+
"loss": 0.0008,
|
| 1906 |
+
"step": 2380
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 6.020151133501259,
|
| 1910 |
+
"grad_norm": 23.99247932434082,
|
| 1911 |
+
"learning_rate": 2.8067650233897085e-05,
|
| 1912 |
+
"loss": 0.0526,
|
| 1913 |
+
"step": 2390
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 6.045340050377834,
|
| 1917 |
+
"grad_norm": 2630.893798828125,
|
| 1918 |
+
"learning_rate": 2.734796689456639e-05,
|
| 1919 |
+
"loss": 0.0008,
|
| 1920 |
+
"step": 2400
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 6.045340050377834,
|
| 1924 |
+
"eval_accuracy": 0.5654296875,
|
| 1925 |
+
"eval_f1_macro": 0.49182304765519874,
|
| 1926 |
+
"eval_loss": 4.129029750823975,
|
| 1927 |
+
"eval_runtime": 11.7471,
|
| 1928 |
+
"eval_samples_per_second": 87.17,
|
| 1929 |
+
"eval_steps_per_second": 5.448,
|
| 1930 |
+
"step": 2400
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 6.0705289672544085,
|
| 1934 |
+
"grad_norm": 3671.589111328125,
|
| 1935 |
+
"learning_rate": 2.66282835552357e-05,
|
| 1936 |
+
"loss": 0.0009,
|
| 1937 |
+
"step": 2410
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 6.095717884130982,
|
| 1941 |
+
"grad_norm": 246.3783721923828,
|
| 1942 |
+
"learning_rate": 2.5908600215905003e-05,
|
| 1943 |
+
"loss": 0.0279,
|
| 1944 |
+
"step": 2420
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 6.120906801007557,
|
| 1948 |
+
"grad_norm": 250481.90625,
|
| 1949 |
+
"learning_rate": 2.5188916876574308e-05,
|
| 1950 |
+
"loss": 0.0367,
|
| 1951 |
+
"step": 2430
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 6.146095717884131,
|
| 1955 |
+
"grad_norm": 5997.31396484375,
|
| 1956 |
+
"learning_rate": 2.4469233537243613e-05,
|
| 1957 |
+
"loss": 0.0002,
|
| 1958 |
+
"step": 2440
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 6.171284634760705,
|
| 1962 |
+
"grad_norm": 294020.6875,
|
| 1963 |
+
"learning_rate": 2.374955019791292e-05,
|
| 1964 |
+
"loss": 0.0447,
|
| 1965 |
+
"step": 2450
|
| 1966 |
+
},
|
| 1967 |
+
{
|
| 1968 |
+
"epoch": 6.19647355163728,
|
| 1969 |
+
"grad_norm": 5173.7607421875,
|
| 1970 |
+
"learning_rate": 2.3029866858582226e-05,
|
| 1971 |
+
"loss": 0.0006,
|
| 1972 |
+
"step": 2460
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"epoch": 6.221662468513854,
|
| 1976 |
+
"grad_norm": 4.966667175292969,
|
| 1977 |
+
"learning_rate": 2.2310183519251528e-05,
|
| 1978 |
+
"loss": 0.0394,
|
| 1979 |
+
"step": 2470
|
| 1980 |
+
},
|
| 1981 |
+
{
|
| 1982 |
+
"epoch": 6.246851385390428,
|
| 1983 |
+
"grad_norm": 328896.6875,
|
| 1984 |
+
"learning_rate": 2.1590500179920836e-05,
|
| 1985 |
+
"loss": 0.0725,
|
| 1986 |
+
"step": 2480
|
| 1987 |
+
},
|
| 1988 |
+
{
|
| 1989 |
+
"epoch": 6.272040302267002,
|
| 1990 |
+
"grad_norm": 855.643798828125,
|
| 1991 |
+
"learning_rate": 2.087081684059014e-05,
|
| 1992 |
+
"loss": 0.0001,
|
| 1993 |
+
"step": 2490
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 6.297229219143577,
|
| 1997 |
+
"grad_norm": 18.617643356323242,
|
| 1998 |
+
"learning_rate": 2.015113350125945e-05,
|
| 1999 |
+
"loss": 0.0024,
|
| 2000 |
+
"step": 2500
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 6.297229219143577,
|
| 2004 |
+
"eval_accuracy": 0.5654296875,
|
| 2005 |
+
"eval_f1_macro": 0.4862593826724424,
|
| 2006 |
+
"eval_loss": 4.414713382720947,
|
| 2007 |
+
"eval_runtime": 11.3987,
|
| 2008 |
+
"eval_samples_per_second": 89.835,
|
| 2009 |
+
"eval_steps_per_second": 5.615,
|
| 2010 |
+
"step": 2500
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 6.3224181360201515,
|
| 2014 |
+
"grad_norm": 43.19185256958008,
|
| 2015 |
+
"learning_rate": 1.943145016192875e-05,
|
| 2016 |
+
"loss": 0.005,
|
| 2017 |
+
"step": 2510
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 6.347607052896725,
|
| 2021 |
+
"grad_norm": 636742.8125,
|
| 2022 |
+
"learning_rate": 1.871176682259806e-05,
|
| 2023 |
+
"loss": 0.0187,
|
| 2024 |
+
"step": 2520
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 6.3727959697733,
|
| 2028 |
+
"grad_norm": 14.576433181762695,
|
| 2029 |
+
"learning_rate": 1.7992083483267364e-05,
|
| 2030 |
+
"loss": 0.0008,
|
| 2031 |
+
"step": 2530
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 6.3979848866498745,
|
| 2035 |
+
"grad_norm": 22.115917205810547,
|
| 2036 |
+
"learning_rate": 1.727240014393667e-05,
|
| 2037 |
+
"loss": 0.0038,
|
| 2038 |
+
"step": 2540
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 6.423173803526448,
|
| 2042 |
+
"grad_norm": 618.1704711914062,
|
| 2043 |
+
"learning_rate": 1.6552716804605974e-05,
|
| 2044 |
+
"loss": 0.0685,
|
| 2045 |
+
"step": 2550
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 6.448362720403023,
|
| 2049 |
+
"grad_norm": 279531.71875,
|
| 2050 |
+
"learning_rate": 1.583303346527528e-05,
|
| 2051 |
+
"loss": 0.0035,
|
| 2052 |
+
"step": 2560
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"epoch": 6.473551637279597,
|
| 2056 |
+
"grad_norm": 265.0247802734375,
|
| 2057 |
+
"learning_rate": 1.5113350125944587e-05,
|
| 2058 |
+
"loss": 0.0245,
|
| 2059 |
+
"step": 2570
|
| 2060 |
+
},
|
| 2061 |
+
{
|
| 2062 |
+
"epoch": 6.498740554156171,
|
| 2063 |
+
"grad_norm": 635805.875,
|
| 2064 |
+
"learning_rate": 1.439366678661389e-05,
|
| 2065 |
+
"loss": 0.0052,
|
| 2066 |
+
"step": 2580
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 6.523929471032746,
|
| 2070 |
+
"grad_norm": 8.843326568603516,
|
| 2071 |
+
"learning_rate": 1.3673983447283195e-05,
|
| 2072 |
+
"loss": 0.0,
|
| 2073 |
+
"step": 2590
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 6.54911838790932,
|
| 2077 |
+
"grad_norm": 87.60523986816406,
|
| 2078 |
+
"learning_rate": 1.2954300107952502e-05,
|
| 2079 |
+
"loss": 0.0002,
|
| 2080 |
+
"step": 2600
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 6.54911838790932,
|
| 2084 |
+
"eval_accuracy": 0.5654296875,
|
| 2085 |
+
"eval_f1_macro": 0.4913330578351924,
|
| 2086 |
+
"eval_loss": 4.520939826965332,
|
| 2087 |
+
"eval_runtime": 11.8845,
|
| 2088 |
+
"eval_samples_per_second": 86.163,
|
| 2089 |
+
"eval_steps_per_second": 5.385,
|
| 2090 |
+
"step": 2600
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 6.574307304785894,
|
| 2094 |
+
"grad_norm": 8.957763671875,
|
| 2095 |
+
"learning_rate": 1.2234616768621806e-05,
|
| 2096 |
+
"loss": 0.0,
|
| 2097 |
+
"step": 2610
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 6.599496221662468,
|
| 2101 |
+
"grad_norm": 6.1030144691467285,
|
| 2102 |
+
"learning_rate": 1.1514933429291113e-05,
|
| 2103 |
+
"loss": 0.0009,
|
| 2104 |
+
"step": 2620
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 6.624685138539043,
|
| 2108 |
+
"grad_norm": 4.862893581390381,
|
| 2109 |
+
"learning_rate": 1.0795250089960418e-05,
|
| 2110 |
+
"loss": 0.1231,
|
| 2111 |
+
"step": 2630
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 6.6498740554156175,
|
| 2115 |
+
"grad_norm": 598.0420532226562,
|
| 2116 |
+
"learning_rate": 1.0075566750629725e-05,
|
| 2117 |
+
"loss": 0.0001,
|
| 2118 |
+
"step": 2640
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 6.675062972292191,
|
| 2122 |
+
"grad_norm": 5.682479381561279,
|
| 2123 |
+
"learning_rate": 9.35588341129903e-06,
|
| 2124 |
+
"loss": 0.0365,
|
| 2125 |
+
"step": 2650
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 6.700251889168766,
|
| 2129 |
+
"grad_norm": 1.800890564918518,
|
| 2130 |
+
"learning_rate": 8.636200071968334e-06,
|
| 2131 |
+
"loss": 0.133,
|
| 2132 |
+
"step": 2660
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 6.72544080604534,
|
| 2136 |
+
"grad_norm": 718.6954345703125,
|
| 2137 |
+
"learning_rate": 7.91651673263764e-06,
|
| 2138 |
+
"loss": 0.0,
|
| 2139 |
+
"step": 2670
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 6.750629722921914,
|
| 2143 |
+
"grad_norm": 338166.46875,
|
| 2144 |
+
"learning_rate": 7.196833393306945e-06,
|
| 2145 |
+
"loss": 0.034,
|
| 2146 |
+
"step": 2680
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 6.775818639798489,
|
| 2150 |
+
"grad_norm": 471.62005615234375,
|
| 2151 |
+
"learning_rate": 6.477150053976251e-06,
|
| 2152 |
+
"loss": 0.0599,
|
| 2153 |
+
"step": 2690
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 6.801007556675063,
|
| 2157 |
+
"grad_norm": 308.09417724609375,
|
| 2158 |
+
"learning_rate": 5.7574667146455565e-06,
|
| 2159 |
+
"loss": 0.0055,
|
| 2160 |
+
"step": 2700
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 6.801007556675063,
|
| 2164 |
+
"eval_accuracy": 0.58203125,
|
| 2165 |
+
"eval_f1_macro": 0.5067161880167751,
|
| 2166 |
+
"eval_loss": 4.515384197235107,
|
| 2167 |
+
"eval_runtime": 11.8883,
|
| 2168 |
+
"eval_samples_per_second": 86.135,
|
| 2169 |
+
"eval_steps_per_second": 5.383,
|
| 2170 |
+
"step": 2700
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 6.826196473551637,
|
| 2174 |
+
"grad_norm": 407.3280334472656,
|
| 2175 |
+
"learning_rate": 5.037783375314862e-06,
|
| 2176 |
+
"loss": 0.0,
|
| 2177 |
+
"step": 2710
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 6.851385390428211,
|
| 2181 |
+
"grad_norm": 396.4019470214844,
|
| 2182 |
+
"learning_rate": 4.318100035984167e-06,
|
| 2183 |
+
"loss": 0.0015,
|
| 2184 |
+
"step": 2720
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 6.876574307304786,
|
| 2188 |
+
"grad_norm": 2.2462317943573,
|
| 2189 |
+
"learning_rate": 3.5984166966534725e-06,
|
| 2190 |
+
"loss": 0.0007,
|
| 2191 |
+
"step": 2730
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 6.9017632241813605,
|
| 2195 |
+
"grad_norm": 40.78224182128906,
|
| 2196 |
+
"learning_rate": 2.8787333573227783e-06,
|
| 2197 |
+
"loss": 0.0001,
|
| 2198 |
+
"step": 2740
|
| 2199 |
+
},
|
| 2200 |
+
{
|
| 2201 |
+
"epoch": 6.926952141057934,
|
| 2202 |
+
"grad_norm": 11996.0771484375,
|
| 2203 |
+
"learning_rate": 2.1590500179920836e-06,
|
| 2204 |
+
"loss": 0.0002,
|
| 2205 |
+
"step": 2750
|
| 2206 |
+
},
|
| 2207 |
+
{
|
| 2208 |
+
"epoch": 6.952141057934509,
|
| 2209 |
+
"grad_norm": 2.2435834407806396,
|
| 2210 |
+
"learning_rate": 1.4393666786613891e-06,
|
| 2211 |
+
"loss": 0.0001,
|
| 2212 |
+
"step": 2760
|
| 2213 |
+
},
|
| 2214 |
+
{
|
| 2215 |
+
"epoch": 6.977329974811083,
|
| 2216 |
+
"grad_norm": 2109.901123046875,
|
| 2217 |
+
"learning_rate": 7.196833393306946e-07,
|
| 2218 |
+
"loss": 0.0832,
|
| 2219 |
+
"step": 2770
|
| 2220 |
+
},
|
| 2221 |
+
{
|
| 2222 |
+
"epoch": 7.0,
|
| 2223 |
+
"step": 2779,
|
| 2224 |
+
"total_flos": 2.5585840915697664e+18,
|
| 2225 |
+
"train_loss": 0.4836694428009911,
|
| 2226 |
+
"train_runtime": 1538.4023,
|
| 2227 |
+
"train_samples_per_second": 28.894,
|
| 2228 |
+
"train_steps_per_second": 1.806
|
| 2229 |
}
|
| 2230 |
],
|
| 2231 |
"logging_steps": 10,
|
| 2232 |
+
"max_steps": 2779,
|
| 2233 |
"num_input_tokens_seen": 0,
|
| 2234 |
+
"num_train_epochs": 7,
|
| 2235 |
"save_steps": 100,
|
| 2236 |
"stateful_callbacks": {
|
| 2237 |
"TrainerControl": {
|
|
|
|
| 2245 |
"attributes": {}
|
| 2246 |
}
|
| 2247 |
},
|
| 2248 |
+
"total_flos": 2.5585840915697664e+18,
|
| 2249 |
"train_batch_size": 16,
|
| 2250 |
"trial_name": null,
|
| 2251 |
"trial_params": null
|