End of training
Browse files- README.md +4 -3
- all_results.json +10 -10
- eval_results.json +5 -5
- runs/Sep18_12-52-12_user/events.out.tfevents.1726645427.user.2655605.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +511 -315
README.md
CHANGED
|
@@ -3,6 +3,7 @@ library_name: transformers
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: facebook/wav2vec2-base
|
| 5 |
tags:
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
- fleurs
|
|
@@ -15,7 +16,7 @@ model-index:
|
|
| 15 |
name: Audio Classification
|
| 16 |
type: audio-classification
|
| 17 |
dataset:
|
| 18 |
-
name: fleurs
|
| 19 |
type: fleurs
|
| 20 |
config: bn_in
|
| 21 |
split: validation
|
|
@@ -31,9 +32,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 31 |
|
| 32 |
# wav2vec2-base-lang-id
|
| 33 |
|
| 34 |
-
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the fleurs dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 37 |
- Accuracy: 1.0
|
| 38 |
|
| 39 |
## Model description
|
|
|
|
| 3 |
license: apache-2.0
|
| 4 |
base_model: facebook/wav2vec2-base
|
| 5 |
tags:
|
| 6 |
+
- audio-classification
|
| 7 |
- generated_from_trainer
|
| 8 |
datasets:
|
| 9 |
- fleurs
|
|
|
|
| 16 |
name: Audio Classification
|
| 17 |
type: audio-classification
|
| 18 |
dataset:
|
| 19 |
+
name: google/fleurs
|
| 20 |
type: fleurs
|
| 21 |
config: bn_in
|
| 22 |
split: validation
|
|
|
|
| 32 |
|
| 33 |
# wav2vec2-base-lang-id
|
| 34 |
|
| 35 |
+
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the google/fleurs dataset.
|
| 36 |
It achieves the following results on the evaluation set:
|
| 37 |
+
- Loss: 0.0001
|
| 38 |
- Accuracy: 1.0
|
| 39 |
|
| 40 |
## Model description
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"eval_accuracy": 1.0,
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second": 27.
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 5.837103526573628e-05,
|
| 5 |
+
"eval_runtime": 9.8113,
|
| 6 |
+
"eval_samples_per_second": 40.973,
|
| 7 |
+
"eval_steps_per_second": 40.973,
|
| 8 |
+
"total_flos": 4.315241031363276e+18,
|
| 9 |
+
"train_loss": 0.13065080859353445,
|
| 10 |
+
"train_runtime": 1086.1024,
|
| 11 |
+
"train_samples_per_second": 27.677,
|
| 12 |
+
"train_steps_per_second": 0.865
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"eval_accuracy": 1.0,
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 5.837103526573628e-05,
|
| 5 |
+
"eval_runtime": 9.8113,
|
| 6 |
+
"eval_samples_per_second": 40.973,
|
| 7 |
+
"eval_steps_per_second": 40.973
|
| 8 |
}
|
runs/Sep18_12-52-12_user/events.out.tfevents.1726645427.user.2655605.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d3b22c9014b7f71890514c01e5ca7a5172f2eb8d2d9d0bb88d8a9fa1faaec2d7
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second": 27.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 4.315241031363276e+18,
|
| 4 |
+
"train_loss": 0.13065080859353445,
|
| 5 |
+
"train_runtime": 1086.1024,
|
| 6 |
+
"train_samples_per_second": 27.677,
|
| 7 |
+
"train_steps_per_second": 0.865
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,577 +1,773 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": 1.0,
|
| 3 |
-
"best_model_checkpoint": "wav2vec2-base-lang-id/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 4.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 3.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 8.
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 2.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss":
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"
|
| 56 |
-
"
|
| 57 |
-
"
|
| 58 |
-
"eval_samples_per_second": 30.202,
|
| 59 |
-
"eval_steps_per_second": 30.202,
|
| 60 |
-
"step": 66
|
| 61 |
-
},
|
| 62 |
-
{
|
| 63 |
-
"epoch": 1.0566037735849056,
|
| 64 |
-
"grad_norm": 0.0018947708886116743,
|
| 65 |
-
"learning_rate": 0.00029797979797979794,
|
| 66 |
-
"loss": 0.0001,
|
| 67 |
"step": 70
|
| 68 |
},
|
| 69 |
{
|
| 70 |
-
"epoch":
|
| 71 |
-
"grad_norm": 0.
|
| 72 |
-
"learning_rate": 0.
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 80
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch":
|
| 78 |
-
"grad_norm": 0.
|
| 79 |
-
"learning_rate": 0.
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch": 1.
|
| 85 |
-
"
|
| 86 |
-
"
|
| 87 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
-
"epoch": 1.
|
| 92 |
-
"grad_norm": 0.
|
| 93 |
-
"learning_rate": 0.
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch": 1.
|
| 99 |
-
"grad_norm": 0.
|
| 100 |
-
"learning_rate": 0.
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch": 1.
|
| 106 |
-
"grad_norm": 0.
|
| 107 |
-
"learning_rate": 0.
|
| 108 |
-
"loss": 0.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
-
"epoch": 1.
|
| 113 |
-
"
|
| 114 |
-
"
|
| 115 |
-
"
|
| 116 |
-
"eval_samples_per_second": 33.293,
|
| 117 |
-
"eval_steps_per_second": 33.293,
|
| 118 |
-
"step": 132
|
| 119 |
-
},
|
| 120 |
-
{
|
| 121 |
-
"epoch": 2.1132075471698113,
|
| 122 |
-
"grad_norm": 0.0005885771824978292,
|
| 123 |
-
"learning_rate": 0.0002626262626262626,
|
| 124 |
-
"loss": 0.0,
|
| 125 |
"step": 140
|
| 126 |
},
|
| 127 |
{
|
| 128 |
-
"epoch":
|
| 129 |
-
"grad_norm": 0.
|
| 130 |
-
"learning_rate": 0.
|
| 131 |
-
"loss": 0.
|
| 132 |
"step": 150
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch":
|
| 136 |
-
"grad_norm": 0.
|
| 137 |
-
"learning_rate": 0.
|
| 138 |
-
"loss": 0.
|
| 139 |
"step": 160
|
| 140 |
},
|
| 141 |
{
|
| 142 |
-
"epoch":
|
| 143 |
-
"grad_norm": 0.
|
| 144 |
-
"learning_rate": 0.
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 170
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch":
|
| 150 |
-
"grad_norm": 0.
|
| 151 |
-
"learning_rate": 0.
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch": 2.
|
| 157 |
-
"
|
| 158 |
-
"
|
| 159 |
-
"
|
| 160 |
-
"
|
|
|
|
|
|
|
| 161 |
},
|
| 162 |
{
|
| 163 |
-
"epoch": 2.
|
| 164 |
-
"
|
| 165 |
-
"
|
| 166 |
-
"
|
| 167 |
-
"
|
| 168 |
-
"eval_steps_per_second": 33.764,
|
| 169 |
-
"step": 198
|
| 170 |
},
|
| 171 |
{
|
| 172 |
-
"epoch":
|
| 173 |
-
"grad_norm": 0.
|
| 174 |
-
"learning_rate": 0.
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 200
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch":
|
| 180 |
-
"grad_norm": 0.
|
| 181 |
-
"learning_rate": 0.
|
| 182 |
"loss": 0.0,
|
| 183 |
"step": 210
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch":
|
| 187 |
-
"grad_norm": 0.
|
| 188 |
-
"learning_rate": 0.
|
| 189 |
"loss": 0.0,
|
| 190 |
"step": 220
|
| 191 |
},
|
| 192 |
{
|
| 193 |
-
"epoch":
|
| 194 |
-
"grad_norm": 0.
|
| 195 |
-
"learning_rate": 0.
|
| 196 |
"loss": 0.0,
|
| 197 |
"step": 230
|
| 198 |
},
|
| 199 |
{
|
| 200 |
-
"epoch":
|
| 201 |
-
"grad_norm": 0.
|
| 202 |
-
"learning_rate": 0.
|
| 203 |
"loss": 0.0,
|
| 204 |
"step": 240
|
| 205 |
},
|
| 206 |
{
|
| 207 |
-
"epoch":
|
| 208 |
-
"grad_norm": 0.
|
| 209 |
-
"learning_rate": 0.
|
| 210 |
"loss": 0.0,
|
| 211 |
"step": 250
|
| 212 |
},
|
| 213 |
{
|
| 214 |
-
"epoch":
|
| 215 |
-
"grad_norm": 0.
|
| 216 |
-
"learning_rate": 0.
|
| 217 |
"loss": 0.0,
|
| 218 |
"step": 260
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch":
|
| 222 |
-
"
|
| 223 |
-
"
|
| 224 |
-
"eval_runtime": 7.0836,
|
| 225 |
-
"eval_samples_per_second": 33.74,
|
| 226 |
-
"eval_steps_per_second": 33.74,
|
| 227 |
-
"step": 265
|
| 228 |
-
},
|
| 229 |
-
{
|
| 230 |
-
"epoch": 4.0754716981132075,
|
| 231 |
-
"grad_norm": 0.0005262196063995361,
|
| 232 |
-
"learning_rate": 0.00019696969696969695,
|
| 233 |
"loss": 0.0,
|
| 234 |
"step": 270
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch":
|
| 238 |
-
"grad_norm": 0.
|
| 239 |
-
"learning_rate": 0.
|
| 240 |
"loss": 0.0,
|
| 241 |
"step": 280
|
| 242 |
},
|
| 243 |
{
|
| 244 |
-
"epoch":
|
| 245 |
-
"
|
| 246 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 247 |
"loss": 0.0,
|
| 248 |
"step": 290
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"grad_norm": 0.
|
| 253 |
-
"learning_rate": 0.
|
| 254 |
"loss": 0.0,
|
| 255 |
"step": 300
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch":
|
| 259 |
-
"grad_norm": 0.
|
| 260 |
-
"learning_rate": 0.
|
| 261 |
"loss": 0.0,
|
| 262 |
"step": 310
|
| 263 |
},
|
| 264 |
{
|
| 265 |
-
"epoch":
|
| 266 |
-
"grad_norm": 0.
|
| 267 |
-
"learning_rate": 0.
|
| 268 |
"loss": 0.0,
|
| 269 |
"step": 320
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch":
|
| 273 |
-
"grad_norm": 0.
|
| 274 |
-
"learning_rate": 0.
|
| 275 |
"loss": 0.0,
|
| 276 |
"step": 330
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch":
|
| 280 |
-
"
|
| 281 |
-
"
|
| 282 |
-
"eval_runtime": 7.0645,
|
| 283 |
-
"eval_samples_per_second": 33.831,
|
| 284 |
-
"eval_steps_per_second": 33.831,
|
| 285 |
-
"step": 331
|
| 286 |
-
},
|
| 287 |
-
{
|
| 288 |
-
"epoch": 5.132075471698113,
|
| 289 |
-
"grad_norm": 0.0004946214030496776,
|
| 290 |
-
"learning_rate": 0.0001616161616161616,
|
| 291 |
"loss": 0.0,
|
| 292 |
"step": 340
|
| 293 |
},
|
| 294 |
{
|
| 295 |
-
"epoch":
|
| 296 |
-
"grad_norm": 0.
|
| 297 |
-
"learning_rate": 0.
|
| 298 |
"loss": 0.0,
|
| 299 |
"step": 350
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch":
|
| 303 |
-
"grad_norm": 0.
|
| 304 |
-
"learning_rate": 0.
|
| 305 |
"loss": 0.0,
|
| 306 |
"step": 360
|
| 307 |
},
|
| 308 |
{
|
| 309 |
-
"epoch":
|
| 310 |
-
"grad_norm": 0.
|
| 311 |
-
"learning_rate": 0.
|
| 312 |
"loss": 0.0,
|
| 313 |
"step": 370
|
| 314 |
},
|
| 315 |
{
|
| 316 |
-
"epoch":
|
| 317 |
-
"
|
| 318 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 319 |
"loss": 0.0,
|
| 320 |
"step": 380
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"epoch":
|
| 324 |
-
"grad_norm": 0.
|
| 325 |
-
"learning_rate": 0.
|
| 326 |
"loss": 0.0,
|
| 327 |
"step": 390
|
| 328 |
},
|
| 329 |
{
|
| 330 |
-
"epoch":
|
| 331 |
-
"
|
| 332 |
-
"
|
| 333 |
-
"eval_runtime": 7.2087,
|
| 334 |
-
"eval_samples_per_second": 33.154,
|
| 335 |
-
"eval_steps_per_second": 33.154,
|
| 336 |
-
"step": 397
|
| 337 |
-
},
|
| 338 |
-
{
|
| 339 |
-
"epoch": 6.037735849056604,
|
| 340 |
-
"grad_norm": 0.00047443213406950235,
|
| 341 |
-
"learning_rate": 0.0001313131313131313,
|
| 342 |
"loss": 0.0,
|
| 343 |
"step": 400
|
| 344 |
},
|
| 345 |
{
|
| 346 |
-
"epoch":
|
| 347 |
-
"grad_norm": 0.
|
| 348 |
-
"learning_rate": 0.
|
| 349 |
"loss": 0.0,
|
| 350 |
"step": 410
|
| 351 |
},
|
| 352 |
{
|
| 353 |
-
"epoch":
|
| 354 |
-
"grad_norm": 0.
|
| 355 |
-
"learning_rate": 0.
|
| 356 |
"loss": 0.0,
|
| 357 |
"step": 420
|
| 358 |
},
|
| 359 |
{
|
| 360 |
-
"epoch":
|
| 361 |
-
"grad_norm": 0.
|
| 362 |
-
"learning_rate": 0.
|
| 363 |
"loss": 0.0,
|
| 364 |
"step": 430
|
| 365 |
},
|
| 366 |
{
|
| 367 |
-
"epoch":
|
| 368 |
-
"grad_norm": 0.
|
| 369 |
-
"learning_rate": 0.
|
| 370 |
"loss": 0.0,
|
| 371 |
"step": 440
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch":
|
| 375 |
-
"grad_norm": 0.
|
| 376 |
-
"learning_rate": 0.
|
| 377 |
"loss": 0.0,
|
| 378 |
"step": 450
|
| 379 |
},
|
| 380 |
{
|
| 381 |
-
"epoch":
|
| 382 |
-
"grad_norm": 0.
|
| 383 |
-
"learning_rate": 0.
|
| 384 |
"loss": 0.0,
|
| 385 |
"step": 460
|
| 386 |
},
|
| 387 |
{
|
| 388 |
-
"epoch":
|
| 389 |
-
"
|
| 390 |
-
"
|
| 391 |
-
"
|
| 392 |
-
"
|
| 393 |
-
"eval_steps_per_second": 33.857,
|
| 394 |
-
"step": 463
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch":
|
| 398 |
-
"
|
| 399 |
-
"
|
| 400 |
-
"
|
|
|
|
|
|
|
| 401 |
"step": 470
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch":
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
-
"learning_rate":
|
| 407 |
"loss": 0.0,
|
| 408 |
"step": 480
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch":
|
| 412 |
-
"grad_norm": 0.
|
| 413 |
-
"learning_rate":
|
| 414 |
"loss": 0.0,
|
| 415 |
"step": 490
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch":
|
| 419 |
-
"grad_norm": 0.
|
| 420 |
-
"learning_rate":
|
| 421 |
"loss": 0.0,
|
| 422 |
"step": 500
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch":
|
| 426 |
-
"grad_norm": 0.
|
| 427 |
-
"learning_rate":
|
| 428 |
"loss": 0.0,
|
| 429 |
"step": 510
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch":
|
| 433 |
-
"grad_norm": 0.
|
| 434 |
-
"learning_rate":
|
| 435 |
"loss": 0.0,
|
| 436 |
"step": 520
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch":
|
| 440 |
-
"grad_norm": 0.
|
| 441 |
-
"learning_rate":
|
| 442 |
"loss": 0.0,
|
| 443 |
"step": 530
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch":
|
| 447 |
-
"
|
| 448 |
-
"
|
| 449 |
-
"eval_runtime": 7.273,
|
| 450 |
-
"eval_samples_per_second": 32.861,
|
| 451 |
-
"eval_steps_per_second": 32.861,
|
| 452 |
-
"step": 530
|
| 453 |
-
},
|
| 454 |
-
{
|
| 455 |
-
"epoch": 8.150943396226415,
|
| 456 |
-
"grad_norm": 0.00044037611223757267,
|
| 457 |
-
"learning_rate": 6.06060606060606e-05,
|
| 458 |
"loss": 0.0,
|
| 459 |
"step": 540
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch":
|
| 463 |
-
"grad_norm": 0.
|
| 464 |
-
"learning_rate":
|
| 465 |
"loss": 0.0,
|
| 466 |
"step": 550
|
| 467 |
},
|
| 468 |
{
|
| 469 |
-
"epoch":
|
| 470 |
-
"grad_norm": 0.
|
| 471 |
-
"learning_rate":
|
| 472 |
"loss": 0.0,
|
| 473 |
"step": 560
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch":
|
| 477 |
-
"
|
| 478 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 479 |
"loss": 0.0,
|
| 480 |
"step": 570
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"epoch":
|
| 484 |
-
"grad_norm": 0.
|
| 485 |
-
"learning_rate":
|
| 486 |
"loss": 0.0,
|
| 487 |
"step": 580
|
| 488 |
},
|
| 489 |
{
|
| 490 |
-
"epoch":
|
| 491 |
-
"grad_norm": 0.
|
| 492 |
-
"learning_rate":
|
| 493 |
"loss": 0.0,
|
| 494 |
"step": 590
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"
|
| 499 |
-
"
|
| 500 |
-
"eval_runtime": 7.0835,
|
| 501 |
-
"eval_samples_per_second": 33.74,
|
| 502 |
-
"eval_steps_per_second": 33.74,
|
| 503 |
-
"step": 596
|
| 504 |
-
},
|
| 505 |
-
{
|
| 506 |
-
"epoch": 9.056603773584905,
|
| 507 |
-
"grad_norm": 0.0004325976187828928,
|
| 508 |
-
"learning_rate": 3.03030303030303e-05,
|
| 509 |
"loss": 0.0,
|
| 510 |
"step": 600
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch":
|
| 514 |
-
"grad_norm": 0.
|
| 515 |
-
"learning_rate":
|
| 516 |
"loss": 0.0,
|
| 517 |
"step": 610
|
| 518 |
},
|
| 519 |
{
|
| 520 |
-
"epoch":
|
| 521 |
-
"grad_norm": 0.
|
| 522 |
-
"learning_rate":
|
| 523 |
"loss": 0.0,
|
| 524 |
"step": 620
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch":
|
| 528 |
-
"grad_norm": 0.
|
| 529 |
-
"learning_rate":
|
| 530 |
"loss": 0.0,
|
| 531 |
"step": 630
|
| 532 |
},
|
| 533 |
{
|
| 534 |
-
"epoch":
|
| 535 |
-
"grad_norm": 0.
|
| 536 |
-
"learning_rate":
|
| 537 |
"loss": 0.0,
|
| 538 |
"step": 640
|
| 539 |
},
|
| 540 |
{
|
| 541 |
-
"epoch":
|
| 542 |
-
"grad_norm": 0.
|
| 543 |
-
"learning_rate":
|
| 544 |
"loss": 0.0,
|
| 545 |
"step": 650
|
| 546 |
},
|
| 547 |
{
|
| 548 |
-
"epoch":
|
| 549 |
-
"
|
| 550 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 551 |
"loss": 0.0,
|
| 552 |
"step": 660
|
| 553 |
},
|
| 554 |
{
|
| 555 |
-
"epoch":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 556 |
"eval_accuracy": 1.0,
|
| 557 |
-
"eval_loss":
|
| 558 |
-
"eval_runtime":
|
| 559 |
-
"eval_samples_per_second":
|
| 560 |
-
"eval_steps_per_second":
|
| 561 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 562 |
},
|
| 563 |
{
|
| 564 |
-
"epoch":
|
| 565 |
-
"
|
| 566 |
-
"
|
| 567 |
-
"
|
| 568 |
-
"
|
| 569 |
-
|
| 570 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 571 |
}
|
| 572 |
],
|
| 573 |
"logging_steps": 10,
|
| 574 |
-
"max_steps":
|
| 575 |
"num_input_tokens_seen": 0,
|
| 576 |
"num_train_epochs": 10,
|
| 577 |
"save_steps": 500,
|
|
@@ -587,7 +783,7 @@
|
|
| 587 |
"attributes": {}
|
| 588 |
}
|
| 589 |
},
|
| 590 |
-
"total_flos":
|
| 591 |
"train_batch_size": 8,
|
| 592 |
"trial_name": null,
|
| 593 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": 1.0,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-base-lang-id/checkpoint-94",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 940,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.10638297872340426,
|
| 13 |
+
"grad_norm": 3.8006672859191895,
|
| 14 |
+
"learning_rate": 3.1914893617021275e-05,
|
| 15 |
+
"loss": 4.4798,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.2127659574468085,
|
| 20 |
+
"grad_norm": 6.771675109863281,
|
| 21 |
+
"learning_rate": 6.382978723404255e-05,
|
| 22 |
+
"loss": 3.4754,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.3191489361702128,
|
| 27 |
+
"grad_norm": 8.21432113647461,
|
| 28 |
+
"learning_rate": 9.574468085106382e-05,
|
| 29 |
+
"loss": 2.4467,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.425531914893617,
|
| 34 |
+
"grad_norm": 7.680928707122803,
|
| 35 |
+
"learning_rate": 0.0001276595744680851,
|
| 36 |
+
"loss": 1.4232,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.5319148936170213,
|
| 41 |
+
"grad_norm": 2.195053815841675,
|
| 42 |
+
"learning_rate": 0.00015957446808510637,
|
| 43 |
+
"loss": 0.4151,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.6382978723404256,
|
| 48 |
+
"grad_norm": 0.13116297125816345,
|
| 49 |
+
"learning_rate": 0.00019148936170212765,
|
| 50 |
+
"loss": 0.0361,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.7446808510638298,
|
| 55 |
+
"grad_norm": 0.014657862484455109,
|
| 56 |
+
"learning_rate": 0.0002234042553191489,
|
| 57 |
+
"loss": 0.0017,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.851063829787234,
|
| 62 |
+
"grad_norm": 0.004712587222456932,
|
| 63 |
+
"learning_rate": 0.0002553191489361702,
|
| 64 |
+
"loss": 0.0003,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.9574468085106383,
|
| 69 |
+
"grad_norm": 0.0027682166546583176,
|
| 70 |
+
"learning_rate": 0.0002872340425531915,
|
| 71 |
+
"loss": 0.0001,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 1.0,
|
| 76 |
+
"eval_accuracy": 1.0,
|
| 77 |
+
"eval_loss": 5.837103526573628e-05,
|
| 78 |
+
"eval_runtime": 11.542,
|
| 79 |
+
"eval_samples_per_second": 34.829,
|
| 80 |
+
"eval_steps_per_second": 34.829,
|
| 81 |
+
"step": 94
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.0638297872340425,
|
| 85 |
+
"grad_norm": 0.0021660495549440384,
|
| 86 |
+
"learning_rate": 0.00029787234042553186,
|
| 87 |
+
"loss": 0.0001,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"epoch": 1.1702127659574468,
|
| 92 |
+
"grad_norm": 0.001946191769093275,
|
| 93 |
+
"learning_rate": 0.00029432624113475173,
|
| 94 |
+
"loss": 0.0001,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 1.2765957446808511,
|
| 99 |
+
"grad_norm": 0.001837807591073215,
|
| 100 |
+
"learning_rate": 0.0002907801418439716,
|
| 101 |
+
"loss": 0.0001,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 1.3829787234042552,
|
| 106 |
+
"grad_norm": 0.0017718354938551784,
|
| 107 |
+
"learning_rate": 0.0002872340425531915,
|
| 108 |
+
"loss": 0.0001,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 1.4893617021276595,
|
| 113 |
+
"grad_norm": 0.0017246523639187217,
|
| 114 |
+
"learning_rate": 0.00028368794326241134,
|
| 115 |
+
"loss": 0.0001,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 1.5957446808510638,
|
| 120 |
+
"grad_norm": 0.0016802914906293154,
|
| 121 |
+
"learning_rate": 0.00028014184397163116,
|
| 122 |
+
"loss": 0.0001,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 1.702127659574468,
|
| 127 |
+
"grad_norm": 0.0016365655465051532,
|
| 128 |
+
"learning_rate": 0.00027659574468085103,
|
| 129 |
+
"loss": 0.0001,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"epoch": 1.8085106382978724,
|
| 134 |
+
"grad_norm": 0.001596157904714346,
|
| 135 |
+
"learning_rate": 0.0002730496453900709,
|
| 136 |
+
"loss": 0.0001,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 1.9148936170212765,
|
| 141 |
+
"grad_norm": 0.0015629915287718177,
|
| 142 |
+
"learning_rate": 0.00026950354609929077,
|
| 143 |
+
"loss": 0.0001,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 2.0,
|
| 148 |
+
"eval_accuracy": 1.0,
|
| 149 |
+
"eval_loss": 3.635883331298828e-05,
|
| 150 |
+
"eval_runtime": 10.0577,
|
| 151 |
+
"eval_samples_per_second": 39.969,
|
| 152 |
+
"eval_steps_per_second": 39.969,
|
| 153 |
+
"step": 188
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 2.021276595744681,
|
| 157 |
+
"grad_norm": 0.001525467843748629,
|
| 158 |
+
"learning_rate": 0.0002659574468085106,
|
| 159 |
+
"loss": 0.0001,
|
| 160 |
+
"step": 190
|
|
|
|
|
|
|
| 161 |
},
|
| 162 |
{
|
| 163 |
+
"epoch": 2.127659574468085,
|
| 164 |
+
"grad_norm": 0.0014883955009281635,
|
| 165 |
+
"learning_rate": 0.00026241134751773046,
|
| 166 |
+
"loss": 0.0001,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 2.2340425531914896,
|
| 171 |
+
"grad_norm": 0.0014570483472198248,
|
| 172 |
+
"learning_rate": 0.00025886524822695033,
|
| 173 |
"loss": 0.0,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 2.3404255319148937,
|
| 178 |
+
"grad_norm": 0.0014193649403750896,
|
| 179 |
+
"learning_rate": 0.0002553191489361702,
|
| 180 |
"loss": 0.0,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 2.4468085106382977,
|
| 185 |
+
"grad_norm": 0.0013909874251112342,
|
| 186 |
+
"learning_rate": 0.00025177304964539007,
|
| 187 |
"loss": 0.0,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 2.5531914893617023,
|
| 192 |
+
"grad_norm": 0.0013578328071162105,
|
| 193 |
+
"learning_rate": 0.0002482269503546099,
|
| 194 |
"loss": 0.0,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 2.6595744680851063,
|
| 199 |
+
"grad_norm": 0.0013242242857813835,
|
| 200 |
+
"learning_rate": 0.00024468085106382976,
|
| 201 |
"loss": 0.0,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 2.7659574468085104,
|
| 206 |
+
"grad_norm": 0.0013041673228144646,
|
| 207 |
+
"learning_rate": 0.00024113475177304963,
|
| 208 |
"loss": 0.0,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 2.872340425531915,
|
| 213 |
+
"grad_norm": 0.0012726597487926483,
|
| 214 |
+
"learning_rate": 0.0002375886524822695,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 215 |
"loss": 0.0,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 2.978723404255319,
|
| 220 |
+
"grad_norm": 0.0012477930868044496,
|
| 221 |
+
"learning_rate": 0.00023404255319148934,
|
| 222 |
"loss": 0.0,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 3.0,
|
| 227 |
+
"eval_accuracy": 1.0,
|
| 228 |
+
"eval_loss": 2.9017082852078602e-05,
|
| 229 |
+
"eval_runtime": 9.9982,
|
| 230 |
+
"eval_samples_per_second": 40.207,
|
| 231 |
+
"eval_steps_per_second": 40.207,
|
| 232 |
+
"step": 282
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.0851063829787235,
|
| 236 |
+
"grad_norm": 0.0012191747082397342,
|
| 237 |
+
"learning_rate": 0.0002304964539007092,
|
| 238 |
"loss": 0.0,
|
| 239 |
"step": 290
|
| 240 |
},
|
| 241 |
{
|
| 242 |
+
"epoch": 3.1914893617021276,
|
| 243 |
+
"grad_norm": 0.0011935862712562084,
|
| 244 |
+
"learning_rate": 0.00022695035460992905,
|
| 245 |
"loss": 0.0,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 3.297872340425532,
|
| 250 |
+
"grad_norm": 0.0011742267524823546,
|
| 251 |
+
"learning_rate": 0.0002234042553191489,
|
| 252 |
"loss": 0.0,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 3.404255319148936,
|
| 257 |
+
"grad_norm": 0.0011481853434816003,
|
| 258 |
+
"learning_rate": 0.00021985815602836877,
|
| 259 |
"loss": 0.0,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 3.5106382978723403,
|
| 264 |
+
"grad_norm": 0.0011253234697505832,
|
| 265 |
+
"learning_rate": 0.00021631205673758864,
|
| 266 |
"loss": 0.0,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 3.617021276595745,
|
| 271 |
+
"grad_norm": 0.0011064092395827174,
|
| 272 |
+
"learning_rate": 0.0002127659574468085,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 273 |
"loss": 0.0,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 3.723404255319149,
|
| 278 |
+
"grad_norm": 0.0010885618394240737,
|
| 279 |
+
"learning_rate": 0.00020921985815602835,
|
| 280 |
"loss": 0.0,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 3.829787234042553,
|
| 285 |
+
"grad_norm": 0.001064595184288919,
|
| 286 |
+
"learning_rate": 0.0002056737588652482,
|
| 287 |
"loss": 0.0,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 3.9361702127659575,
|
| 292 |
+
"grad_norm": 0.0010447927052155137,
|
| 293 |
+
"learning_rate": 0.00020212765957446807,
|
| 294 |
"loss": 0.0,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
+
"epoch": 4.0,
|
| 299 |
+
"eval_accuracy": 1.0,
|
| 300 |
+
"eval_loss": 2.372264862060547e-05,
|
| 301 |
+
"eval_runtime": 9.9011,
|
| 302 |
+
"eval_samples_per_second": 40.602,
|
| 303 |
+
"eval_steps_per_second": 40.602,
|
| 304 |
+
"step": 376
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.042553191489362,
|
| 308 |
+
"grad_norm": 0.001030342886224389,
|
| 309 |
+
"learning_rate": 0.0001985815602836879,
|
| 310 |
"loss": 0.0,
|
| 311 |
"step": 380
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"epoch": 4.148936170212766,
|
| 315 |
+
"grad_norm": 0.0010120035149157047,
|
| 316 |
+
"learning_rate": 0.0001950354609929078,
|
| 317 |
"loss": 0.0,
|
| 318 |
"step": 390
|
| 319 |
},
|
| 320 |
{
|
| 321 |
+
"epoch": 4.25531914893617,
|
| 322 |
+
"grad_norm": 0.0009937717113643885,
|
| 323 |
+
"learning_rate": 0.00019148936170212765,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 324 |
"loss": 0.0,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 4.361702127659575,
|
| 329 |
+
"grad_norm": 0.0009758667438291013,
|
| 330 |
+
"learning_rate": 0.0001879432624113475,
|
| 331 |
"loss": 0.0,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 4.468085106382979,
|
| 336 |
+
"grad_norm": 0.0009558630990795791,
|
| 337 |
+
"learning_rate": 0.00018439716312056736,
|
| 338 |
"loss": 0.0,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 4.574468085106383,
|
| 343 |
+
"grad_norm": 0.0009456143015995622,
|
| 344 |
+
"learning_rate": 0.0001808510638297872,
|
| 345 |
"loss": 0.0,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
+
"epoch": 4.680851063829787,
|
| 350 |
+
"grad_norm": 0.0009262987296096981,
|
| 351 |
+
"learning_rate": 0.00017730496453900708,
|
| 352 |
"loss": 0.0,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 4.787234042553192,
|
| 357 |
+
"grad_norm": 0.0009140170877799392,
|
| 358 |
+
"learning_rate": 0.00017375886524822692,
|
| 359 |
"loss": 0.0,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 4.8936170212765955,
|
| 364 |
+
"grad_norm": 0.000904095999430865,
|
| 365 |
+
"learning_rate": 0.00017021276595744682,
|
| 366 |
"loss": 0.0,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 5.0,
|
| 371 |
+
"grad_norm": 0.0008834420586936176,
|
| 372 |
+
"learning_rate": 0.00016666666666666666,
|
| 373 |
+
"loss": 0.0,
|
| 374 |
+
"step": 470
|
|
|
|
|
|
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 5.0,
|
| 378 |
+
"eval_accuracy": 1.0,
|
| 379 |
+
"eval_loss": 2.014636993408203e-05,
|
| 380 |
+
"eval_runtime": 9.9558,
|
| 381 |
+
"eval_samples_per_second": 40.379,
|
| 382 |
+
"eval_steps_per_second": 40.379,
|
| 383 |
"step": 470
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"epoch": 5.1063829787234045,
|
| 387 |
+
"grad_norm": 0.0008752320427447557,
|
| 388 |
+
"learning_rate": 0.0001631205673758865,
|
| 389 |
"loss": 0.0,
|
| 390 |
"step": 480
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"epoch": 5.212765957446808,
|
| 394 |
+
"grad_norm": 0.0008663799380883574,
|
| 395 |
+
"learning_rate": 0.00015957446808510637,
|
| 396 |
"loss": 0.0,
|
| 397 |
"step": 490
|
| 398 |
},
|
| 399 |
{
|
| 400 |
+
"epoch": 5.319148936170213,
|
| 401 |
+
"grad_norm": 0.0008535313536413014,
|
| 402 |
+
"learning_rate": 0.00015602836879432622,
|
| 403 |
"loss": 0.0,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 5.425531914893617,
|
| 408 |
+
"grad_norm": 0.0008452454931102693,
|
| 409 |
+
"learning_rate": 0.00015248226950354606,
|
| 410 |
"loss": 0.0,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 5.531914893617021,
|
| 415 |
+
"grad_norm": 0.0008268958772532642,
|
| 416 |
+
"learning_rate": 0.00014893617021276593,
|
| 417 |
"loss": 0.0,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
+
"epoch": 5.638297872340425,
|
| 422 |
+
"grad_norm": 0.0008181555895134807,
|
| 423 |
+
"learning_rate": 0.0001453900709219858,
|
| 424 |
"loss": 0.0,
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 5.74468085106383,
|
| 429 |
+
"grad_norm": 0.0008063354762271047,
|
| 430 |
+
"learning_rate": 0.00014184397163120567,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 431 |
"loss": 0.0,
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
+
"epoch": 5.851063829787234,
|
| 436 |
+
"grad_norm": 0.0007958101341500878,
|
| 437 |
+
"learning_rate": 0.00013829787234042552,
|
| 438 |
"loss": 0.0,
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
+
"epoch": 5.957446808510638,
|
| 443 |
+
"grad_norm": 0.0007865344523452222,
|
| 444 |
+
"learning_rate": 0.00013475177304964539,
|
| 445 |
"loss": 0.0,
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 6.0,
|
| 450 |
+
"eval_accuracy": 1.0,
|
| 451 |
+
"eval_loss": 1.7642974853515625e-05,
|
| 452 |
+
"eval_runtime": 9.9723,
|
| 453 |
+
"eval_samples_per_second": 40.312,
|
| 454 |
+
"eval_steps_per_second": 40.312,
|
| 455 |
+
"step": 564
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 6.0638297872340425,
|
| 459 |
+
"grad_norm": 0.0007770307711325586,
|
| 460 |
+
"learning_rate": 0.00013120567375886523,
|
| 461 |
"loss": 0.0,
|
| 462 |
"step": 570
|
| 463 |
},
|
| 464 |
{
|
| 465 |
+
"epoch": 6.170212765957447,
|
| 466 |
+
"grad_norm": 0.0007697618566453457,
|
| 467 |
+
"learning_rate": 0.0001276595744680851,
|
| 468 |
"loss": 0.0,
|
| 469 |
"step": 580
|
| 470 |
},
|
| 471 |
{
|
| 472 |
+
"epoch": 6.276595744680851,
|
| 473 |
+
"grad_norm": 0.0007645227597095072,
|
| 474 |
+
"learning_rate": 0.00012411347517730494,
|
| 475 |
"loss": 0.0,
|
| 476 |
"step": 590
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 6.382978723404255,
|
| 480 |
+
"grad_norm": 0.0007574139162898064,
|
| 481 |
+
"learning_rate": 0.00012056737588652481,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 482 |
"loss": 0.0,
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 6.48936170212766,
|
| 487 |
+
"grad_norm": 0.000741077761631459,
|
| 488 |
+
"learning_rate": 0.00011702127659574467,
|
| 489 |
"loss": 0.0,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"epoch": 6.595744680851064,
|
| 494 |
+
"grad_norm": 0.000734071247279644,
|
| 495 |
+
"learning_rate": 0.00011347517730496453,
|
| 496 |
"loss": 0.0,
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 6.702127659574468,
|
| 501 |
+
"grad_norm": 0.000728779355995357,
|
| 502 |
+
"learning_rate": 0.00010992907801418438,
|
| 503 |
"loss": 0.0,
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 6.808510638297872,
|
| 508 |
+
"grad_norm": 0.0007212815107777715,
|
| 509 |
+
"learning_rate": 0.00010638297872340425,
|
| 510 |
"loss": 0.0,
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
+
"epoch": 6.914893617021277,
|
| 515 |
+
"grad_norm": 0.0007169453892856836,
|
| 516 |
+
"learning_rate": 0.0001028368794326241,
|
| 517 |
"loss": 0.0,
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 7.0,
|
| 522 |
+
"eval_accuracy": 1.0,
|
| 523 |
+
"eval_loss": 1.596455513208639e-05,
|
| 524 |
+
"eval_runtime": 9.9126,
|
| 525 |
+
"eval_samples_per_second": 40.554,
|
| 526 |
+
"eval_steps_per_second": 40.554,
|
| 527 |
+
"step": 658
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.0212765957446805,
|
| 531 |
+
"grad_norm": 0.000711097614839673,
|
| 532 |
+
"learning_rate": 9.929078014184395e-05,
|
| 533 |
"loss": 0.0,
|
| 534 |
"step": 660
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 7.127659574468085,
|
| 538 |
+
"grad_norm": 0.0007094301981851459,
|
| 539 |
+
"learning_rate": 9.574468085106382e-05,
|
| 540 |
+
"loss": 0.0,
|
| 541 |
+
"step": 670
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 7.23404255319149,
|
| 545 |
+
"grad_norm": 0.0006968958768993616,
|
| 546 |
+
"learning_rate": 9.219858156028368e-05,
|
| 547 |
+
"loss": 0.0,
|
| 548 |
+
"step": 680
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 7.340425531914893,
|
| 552 |
+
"grad_norm": 0.0006909930380061269,
|
| 553 |
+
"learning_rate": 8.865248226950354e-05,
|
| 554 |
+
"loss": 0.0,
|
| 555 |
+
"step": 690
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 7.446808510638298,
|
| 559 |
+
"grad_norm": 0.0006865290924906731,
|
| 560 |
+
"learning_rate": 8.510638297872341e-05,
|
| 561 |
+
"loss": 0.0,
|
| 562 |
+
"step": 700
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 7.553191489361702,
|
| 566 |
+
"grad_norm": 0.0006844609742984176,
|
| 567 |
+
"learning_rate": 8.156028368794325e-05,
|
| 568 |
+
"loss": 0.0,
|
| 569 |
+
"step": 710
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 7.659574468085106,
|
| 573 |
+
"grad_norm": 0.0006792008061893284,
|
| 574 |
+
"learning_rate": 7.801418439716311e-05,
|
| 575 |
+
"loss": 0.0,
|
| 576 |
+
"step": 720
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 7.76595744680851,
|
| 580 |
+
"grad_norm": 0.0006731408648192883,
|
| 581 |
+
"learning_rate": 7.446808510638297e-05,
|
| 582 |
+
"loss": 0.0,
|
| 583 |
+
"step": 730
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 7.872340425531915,
|
| 587 |
+
"grad_norm": 0.0006701324600726366,
|
| 588 |
+
"learning_rate": 7.092198581560284e-05,
|
| 589 |
+
"loss": 0.0,
|
| 590 |
+
"step": 740
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 7.9787234042553195,
|
| 594 |
+
"grad_norm": 0.0006633326993323863,
|
| 595 |
+
"learning_rate": 6.737588652482269e-05,
|
| 596 |
+
"loss": 0.0,
|
| 597 |
+
"step": 750
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 8.0,
|
| 601 |
"eval_accuracy": 1.0,
|
| 602 |
+
"eval_loss": 1.4901161193847656e-05,
|
| 603 |
+
"eval_runtime": 9.9494,
|
| 604 |
+
"eval_samples_per_second": 40.405,
|
| 605 |
+
"eval_steps_per_second": 40.405,
|
| 606 |
+
"step": 752
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 8.085106382978724,
|
| 610 |
+
"grad_norm": 0.0006632324075326324,
|
| 611 |
+
"learning_rate": 6.382978723404255e-05,
|
| 612 |
+
"loss": 0.0,
|
| 613 |
+
"step": 760
|
| 614 |
},
|
| 615 |
{
|
| 616 |
+
"epoch": 8.191489361702128,
|
| 617 |
+
"grad_norm": 0.000655403477139771,
|
| 618 |
+
"learning_rate": 6.028368794326241e-05,
|
| 619 |
+
"loss": 0.0,
|
| 620 |
+
"step": 770
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 8.297872340425531,
|
| 624 |
+
"grad_norm": 0.0006541645270772278,
|
| 625 |
+
"learning_rate": 5.6737588652482264e-05,
|
| 626 |
+
"loss": 0.0,
|
| 627 |
+
"step": 780
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 8.404255319148936,
|
| 631 |
+
"grad_norm": 0.0006483749020844698,
|
| 632 |
+
"learning_rate": 5.319148936170213e-05,
|
| 633 |
+
"loss": 0.0,
|
| 634 |
+
"step": 790
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 8.51063829787234,
|
| 638 |
+
"grad_norm": 0.0006471078377217054,
|
| 639 |
+
"learning_rate": 4.964539007092198e-05,
|
| 640 |
+
"loss": 0.0,
|
| 641 |
+
"step": 800
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 8.617021276595745,
|
| 645 |
+
"grad_norm": 0.0006466888007707894,
|
| 646 |
+
"learning_rate": 4.609929078014184e-05,
|
| 647 |
+
"loss": 0.0,
|
| 648 |
+
"step": 810
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 8.72340425531915,
|
| 652 |
+
"grad_norm": 0.0006407785695046186,
|
| 653 |
+
"learning_rate": 4.2553191489361704e-05,
|
| 654 |
+
"loss": 0.0,
|
| 655 |
+
"step": 820
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 8.829787234042554,
|
| 659 |
+
"grad_norm": 0.0006384547450579703,
|
| 660 |
+
"learning_rate": 3.9007092198581555e-05,
|
| 661 |
+
"loss": 0.0,
|
| 662 |
+
"step": 830
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 8.936170212765958,
|
| 666 |
+
"grad_norm": 0.0006379844271577895,
|
| 667 |
+
"learning_rate": 3.546099290780142e-05,
|
| 668 |
+
"loss": 0.0,
|
| 669 |
+
"step": 840
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 9.0,
|
| 673 |
+
"eval_accuracy": 1.0,
|
| 674 |
+
"eval_loss": 1.4424324035644531e-05,
|
| 675 |
+
"eval_runtime": 9.9315,
|
| 676 |
+
"eval_samples_per_second": 40.477,
|
| 677 |
+
"eval_steps_per_second": 40.477,
|
| 678 |
+
"step": 846
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 9.042553191489361,
|
| 682 |
+
"grad_norm": 0.0006334384088404477,
|
| 683 |
+
"learning_rate": 3.1914893617021275e-05,
|
| 684 |
+
"loss": 0.0,
|
| 685 |
+
"step": 850
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 9.148936170212766,
|
| 689 |
+
"grad_norm": 0.0006309397285804152,
|
| 690 |
+
"learning_rate": 2.8368794326241132e-05,
|
| 691 |
+
"loss": 0.0,
|
| 692 |
+
"step": 860
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 9.25531914893617,
|
| 696 |
+
"grad_norm": 0.0006310406024567783,
|
| 697 |
+
"learning_rate": 2.482269503546099e-05,
|
| 698 |
+
"loss": 0.0,
|
| 699 |
+
"step": 870
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 9.361702127659575,
|
| 703 |
+
"grad_norm": 0.0006308447918854654,
|
| 704 |
+
"learning_rate": 2.1276595744680852e-05,
|
| 705 |
+
"loss": 0.0,
|
| 706 |
+
"step": 880
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 9.46808510638298,
|
| 710 |
+
"grad_norm": 0.0006286040297709405,
|
| 711 |
+
"learning_rate": 1.773049645390071e-05,
|
| 712 |
+
"loss": 0.0,
|
| 713 |
+
"step": 890
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 9.574468085106384,
|
| 717 |
+
"grad_norm": 0.0006258686189539731,
|
| 718 |
+
"learning_rate": 1.4184397163120566e-05,
|
| 719 |
+
"loss": 0.0,
|
| 720 |
+
"step": 900
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 9.680851063829786,
|
| 724 |
+
"grad_norm": 0.0006251951563172042,
|
| 725 |
+
"learning_rate": 1.0638297872340426e-05,
|
| 726 |
+
"loss": 0.0,
|
| 727 |
+
"step": 910
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 9.787234042553191,
|
| 731 |
+
"grad_norm": 0.0006240535294637084,
|
| 732 |
+
"learning_rate": 7.092198581560283e-06,
|
| 733 |
+
"loss": 0.0,
|
| 734 |
+
"step": 920
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 9.893617021276595,
|
| 738 |
+
"grad_norm": 0.0006240674993023276,
|
| 739 |
+
"learning_rate": 3.5460992907801415e-06,
|
| 740 |
+
"loss": 0.0,
|
| 741 |
+
"step": 930
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 10.0,
|
| 745 |
+
"grad_norm": 0.0006238200003281236,
|
| 746 |
+
"learning_rate": 0.0,
|
| 747 |
+
"loss": 0.0,
|
| 748 |
+
"step": 940
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 10.0,
|
| 752 |
+
"eval_accuracy": 1.0,
|
| 753 |
+
"eval_loss": 1.4065806681173854e-05,
|
| 754 |
+
"eval_runtime": 9.9903,
|
| 755 |
+
"eval_samples_per_second": 40.239,
|
| 756 |
+
"eval_steps_per_second": 40.239,
|
| 757 |
+
"step": 940
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 10.0,
|
| 761 |
+
"step": 940,
|
| 762 |
+
"total_flos": 4.315241031363276e+18,
|
| 763 |
+
"train_loss": 0.13065080859353445,
|
| 764 |
+
"train_runtime": 1086.1024,
|
| 765 |
+
"train_samples_per_second": 27.677,
|
| 766 |
+
"train_steps_per_second": 0.865
|
| 767 |
}
|
| 768 |
],
|
| 769 |
"logging_steps": 10,
|
| 770 |
+
"max_steps": 940,
|
| 771 |
"num_input_tokens_seen": 0,
|
| 772 |
"num_train_epochs": 10,
|
| 773 |
"save_steps": 500,
|
|
|
|
| 783 |
"attributes": {}
|
| 784 |
}
|
| 785 |
},
|
| 786 |
+
"total_flos": 4.315241031363276e+18,
|
| 787 |
"train_batch_size": 8,
|
| 788 |
"trial_name": null,
|
| 789 |
"trial_params": null
|