🍻 cheers
Browse files- README.md +6 -5
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Apr30_15-11-45_55836b80922a/events.out.tfevents.1714491534.55836b80922a.34.1 +3 -0
- train_results.json +6 -6
- trainer_state.json +1657 -516
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- imagefolder
|
|
@@ -14,7 +15,7 @@ model-index:
|
|
| 14 |
name: Image Classification
|
| 15 |
type: image-classification
|
| 16 |
dataset:
|
| 17 |
-
name:
|
| 18 |
type: imagefolder
|
| 19 |
config: default
|
| 20 |
split: train
|
|
@@ -22,7 +23,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -30,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
# Action_all_10_class
|
| 32 |
|
| 33 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
- imagefolder
|
|
|
|
| 15 |
name: Image Classification
|
| 16 |
type: image-classification
|
| 17 |
dataset:
|
| 18 |
+
name: Action_small_dataset
|
| 19 |
type: imagefolder
|
| 20 |
config: default
|
| 21 |
split: train
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.8699386503067484
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 31 |
|
| 32 |
# Action_all_10_class
|
| 33 |
|
| 34 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the Action_small_dataset dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.4618
|
| 37 |
+
- Accuracy: 0.8699
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 11.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 8.
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 1.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8699386503067484,
|
| 4 |
+
"eval_loss": 0.46183693408966064,
|
| 5 |
+
"eval_runtime": 11.7258,
|
| 6 |
+
"eval_samples_per_second": 69.505,
|
| 7 |
+
"eval_steps_per_second": 8.699,
|
| 8 |
+
"total_flos": 3.57806169439359e+18,
|
| 9 |
+
"train_loss": 0.4476649462557994,
|
| 10 |
+
"train_runtime": 1596.8664,
|
| 11 |
+
"train_samples_per_second": 28.913,
|
| 12 |
+
"train_steps_per_second": 1.81
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 11.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 8.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8699386503067484,
|
| 4 |
+
"eval_loss": 0.46183693408966064,
|
| 5 |
+
"eval_runtime": 11.7258,
|
| 6 |
+
"eval_samples_per_second": 69.505,
|
| 7 |
+
"eval_steps_per_second": 8.699
|
| 8 |
}
|
runs/Apr30_15-11-45_55836b80922a/events.out.tfevents.1714491534.55836b80922a.34.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1807a6e0f0eb8673728b302c6bbf5d790eb0842deae50421a53363a9aef40686
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 3.57806169439359e+18,
|
| 4 |
+
"train_loss": 0.4476649462557994,
|
| 5 |
+
"train_runtime": 1596.8664,
|
| 6 |
+
"train_samples_per_second": 28.913,
|
| 7 |
+
"train_steps_per_second": 1.81
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1163 +1,2304 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "Action_all_10_class/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.03,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.07,
|
| 20 |
-
"grad_norm": 1.
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 2.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.1,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.14,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.17,
|
| 41 |
-
"grad_norm": 2.
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.21,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 1.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.24,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 1.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.28,
|
| 62 |
-
"grad_norm": 3.
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 1.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.31,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.35,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.35,
|
| 83 |
-
"eval_accuracy": 0.
|
| 84 |
-
"eval_loss":
|
| 85 |
-
"eval_runtime":
|
| 86 |
-
"eval_samples_per_second":
|
| 87 |
-
"eval_steps_per_second":
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.38,
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss": 1.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.42,
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate":
|
| 101 |
-
"loss": 1.
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 0.45,
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate":
|
| 108 |
-
"loss": 1.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 0.48,
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss": 1.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.52,
|
| 120 |
-
"grad_norm":
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss": 1.
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.55,
|
| 127 |
-
"grad_norm": 4.
|
| 128 |
-
"learning_rate":
|
| 129 |
-
"loss": 1.
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.59,
|
| 134 |
-
"grad_norm":
|
| 135 |
-
"learning_rate":
|
| 136 |
-
"loss":
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.62,
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"learning_rate":
|
| 143 |
-
"loss": 1.
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 0.66,
|
| 148 |
-
"grad_norm": 4.
|
| 149 |
-
"learning_rate":
|
| 150 |
-
"loss":
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 0.69,
|
| 155 |
-
"grad_norm":
|
| 156 |
-
"learning_rate":
|
| 157 |
-
"loss":
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 0.69,
|
| 162 |
-
"eval_accuracy": 0.
|
| 163 |
-
"eval_loss": 0.
|
| 164 |
-
"eval_runtime":
|
| 165 |
-
"eval_samples_per_second":
|
| 166 |
-
"eval_steps_per_second":
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.73,
|
| 171 |
-
"grad_norm":
|
| 172 |
-
"learning_rate":
|
| 173 |
-
"loss":
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 0.76,
|
| 178 |
-
"grad_norm": 4.
|
| 179 |
-
"learning_rate":
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 0.8,
|
| 185 |
-
"grad_norm":
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss":
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 0.83,
|
| 192 |
-
"grad_norm":
|
| 193 |
-
"learning_rate":
|
| 194 |
-
"loss":
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 0.87,
|
| 199 |
-
"grad_norm":
|
| 200 |
-
"learning_rate":
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 0.9,
|
| 206 |
-
"grad_norm":
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 0.93,
|
| 213 |
-
"grad_norm":
|
| 214 |
-
"learning_rate":
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 0.97,
|
| 220 |
-
"grad_norm":
|
| 221 |
-
"learning_rate":
|
| 222 |
-
"loss":
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 1.0,
|
| 227 |
-
"grad_norm":
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 1.04,
|
| 234 |
-
"grad_norm":
|
| 235 |
-
"learning_rate":
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 1.04,
|
| 241 |
-
"eval_accuracy": 0.
|
| 242 |
-
"eval_loss": 0.
|
| 243 |
-
"eval_runtime":
|
| 244 |
-
"eval_samples_per_second":
|
| 245 |
-
"eval_steps_per_second":
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 1.07,
|
| 250 |
-
"grad_norm":
|
| 251 |
-
"learning_rate":
|
| 252 |
-
"loss": 0.
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 1.11,
|
| 257 |
-
"grad_norm":
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 1.14,
|
| 264 |
-
"grad_norm": 4.
|
| 265 |
-
"learning_rate":
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 1.18,
|
| 271 |
-
"grad_norm":
|
| 272 |
-
"learning_rate":
|
| 273 |
-
"loss": 0.
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 1.21,
|
| 278 |
-
"grad_norm":
|
| 279 |
-
"learning_rate":
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 1.25,
|
| 285 |
-
"grad_norm":
|
| 286 |
-
"learning_rate":
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 1.28,
|
| 292 |
-
"grad_norm": 4.
|
| 293 |
-
"learning_rate":
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 1.31,
|
| 299 |
-
"grad_norm": 3.
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 1.35,
|
| 306 |
-
"grad_norm":
|
| 307 |
-
"learning_rate":
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 1.38,
|
| 313 |
-
"grad_norm":
|
| 314 |
-
"learning_rate":
|
| 315 |
-
"loss": 0.
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 1.38,
|
| 320 |
-
"eval_accuracy": 0.
|
| 321 |
-
"eval_loss": 0.
|
| 322 |
-
"eval_runtime": 11.
|
| 323 |
-
"eval_samples_per_second":
|
| 324 |
-
"eval_steps_per_second":
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 1.42,
|
| 329 |
-
"grad_norm":
|
| 330 |
-
"learning_rate":
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 1.45,
|
| 336 |
-
"grad_norm":
|
| 337 |
-
"learning_rate":
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 1.49,
|
| 343 |
-
"grad_norm":
|
| 344 |
-
"learning_rate":
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 1.52,
|
| 350 |
-
"grad_norm":
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 1.56,
|
| 357 |
-
"grad_norm":
|
| 358 |
-
"learning_rate":
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 1.59,
|
| 364 |
-
"grad_norm": 5.
|
| 365 |
-
"learning_rate":
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 1.63,
|
| 371 |
-
"grad_norm":
|
| 372 |
-
"learning_rate":
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 1.66,
|
| 378 |
-
"grad_norm":
|
| 379 |
-
"learning_rate":
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 1.7,
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate":
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 1.73,
|
| 392 |
-
"grad_norm":
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 1.73,
|
| 399 |
-
"eval_accuracy": 0.
|
| 400 |
-
"eval_loss": 0.
|
| 401 |
-
"eval_runtime": 11.
|
| 402 |
-
"eval_samples_per_second":
|
| 403 |
-
"eval_steps_per_second":
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 1.76,
|
| 408 |
-
"grad_norm": 5.
|
| 409 |
-
"learning_rate":
|
| 410 |
-
"loss": 0.
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 1.8,
|
| 415 |
-
"grad_norm":
|
| 416 |
-
"learning_rate":
|
| 417 |
-
"loss": 0.
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 1.83,
|
| 422 |
-
"grad_norm": 4.
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 1.87,
|
| 429 |
-
"grad_norm":
|
| 430 |
-
"learning_rate":
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 1.9,
|
| 436 |
-
"grad_norm":
|
| 437 |
-
"learning_rate":
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 1.94,
|
| 443 |
-
"grad_norm":
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 1.97,
|
| 450 |
-
"grad_norm":
|
| 451 |
-
"learning_rate":
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 2.01,
|
| 457 |
-
"grad_norm":
|
| 458 |
-
"learning_rate":
|
| 459 |
-
"loss": 0.
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 2.04,
|
| 464 |
-
"grad_norm":
|
| 465 |
-
"learning_rate":
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 2.08,
|
| 471 |
-
"grad_norm":
|
| 472 |
-
"learning_rate":
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 2.08,
|
| 478 |
-
"eval_accuracy": 0.
|
| 479 |
-
"eval_loss": 0.
|
| 480 |
-
"eval_runtime":
|
| 481 |
-
"eval_samples_per_second":
|
| 482 |
-
"eval_steps_per_second":
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 2.11,
|
| 487 |
-
"grad_norm":
|
| 488 |
-
"learning_rate":
|
| 489 |
-
"loss": 0.
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 2.15,
|
| 494 |
-
"grad_norm":
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 2.18,
|
| 501 |
-
"grad_norm":
|
| 502 |
-
"learning_rate":
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 2.21,
|
| 508 |
-
"grad_norm": 6.
|
| 509 |
-
"learning_rate":
|
| 510 |
-
"loss": 0.
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 2.25,
|
| 515 |
-
"grad_norm":
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 2.28,
|
| 522 |
-
"grad_norm":
|
| 523 |
-
"learning_rate":
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 2.32,
|
| 529 |
-
"grad_norm":
|
| 530 |
-
"learning_rate":
|
| 531 |
-
"loss": 0.
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 2.35,
|
| 536 |
-
"grad_norm":
|
| 537 |
-
"learning_rate":
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 2.39,
|
| 543 |
-
"grad_norm":
|
| 544 |
-
"learning_rate":
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 2.42,
|
| 550 |
-
"grad_norm":
|
| 551 |
-
"learning_rate":
|
| 552 |
-
"loss": 0.
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 2.42,
|
| 557 |
-
"eval_accuracy": 0.
|
| 558 |
-
"eval_loss": 0.
|
| 559 |
-
"eval_runtime": 11.
|
| 560 |
-
"eval_samples_per_second":
|
| 561 |
-
"eval_steps_per_second": 8.
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 2.46,
|
| 566 |
-
"grad_norm":
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 2.49,
|
| 573 |
-
"grad_norm":
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 2.53,
|
| 580 |
-
"grad_norm":
|
| 581 |
-
"learning_rate":
|
| 582 |
-
"loss": 0.
|
| 583 |
"step": 730
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 2.56,
|
| 587 |
-
"grad_norm":
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 2.6,
|
| 594 |
-
"grad_norm":
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 2.63,
|
| 601 |
-
"grad_norm":
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 2.66,
|
| 608 |
-
"grad_norm":
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 2.7,
|
| 615 |
-
"grad_norm":
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 2.73,
|
| 622 |
-
"grad_norm":
|
| 623 |
-
"learning_rate":
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 2.77,
|
| 629 |
-
"grad_norm":
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 2.77,
|
| 636 |
-
"eval_accuracy": 0.
|
| 637 |
-
"eval_loss": 0.
|
| 638 |
-
"eval_runtime":
|
| 639 |
-
"eval_samples_per_second":
|
| 640 |
-
"eval_steps_per_second":
|
| 641 |
"step": 800
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 2.8,
|
| 645 |
-
"grad_norm":
|
| 646 |
-
"learning_rate":
|
| 647 |
-
"loss": 0.
|
| 648 |
"step": 810
|
| 649 |
},
|
| 650 |
{
|
| 651 |
"epoch": 2.84,
|
| 652 |
-
"grad_norm": 5.
|
| 653 |
-
"learning_rate":
|
| 654 |
-
"loss": 0.
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"epoch": 2.87,
|
| 659 |
-
"grad_norm": 2.
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 2.91,
|
| 666 |
-
"grad_norm":
|
| 667 |
-
"learning_rate":
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 2.94,
|
| 673 |
-
"grad_norm":
|
| 674 |
-
"learning_rate":
|
| 675 |
-
"loss": 0.
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 2.98,
|
| 680 |
-
"grad_norm":
|
| 681 |
-
"learning_rate":
|
| 682 |
-
"loss": 0.
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 3.01,
|
| 687 |
-
"grad_norm":
|
| 688 |
-
"learning_rate":
|
| 689 |
-
"loss": 0.
|
| 690 |
"step": 870
|
| 691 |
},
|
| 692 |
{
|
| 693 |
"epoch": 3.04,
|
| 694 |
-
"grad_norm": 4.
|
| 695 |
-
"learning_rate":
|
| 696 |
-
"loss": 0.
|
| 697 |
"step": 880
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 3.08,
|
| 701 |
-
"grad_norm":
|
| 702 |
-
"learning_rate":
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 890
|
| 705 |
},
|
| 706 |
{
|
| 707 |
"epoch": 3.11,
|
| 708 |
-
"grad_norm": 3.
|
| 709 |
-
"learning_rate":
|
| 710 |
-
"loss": 0.
|
| 711 |
"step": 900
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 3.11,
|
| 715 |
-
"eval_accuracy": 0.
|
| 716 |
-
"eval_loss": 0.
|
| 717 |
-
"eval_runtime": 11.
|
| 718 |
-
"eval_samples_per_second":
|
| 719 |
-
"eval_steps_per_second": 8.
|
| 720 |
"step": 900
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 3.15,
|
| 724 |
-
"grad_norm": 4.
|
| 725 |
-
"learning_rate":
|
| 726 |
-
"loss": 0.
|
| 727 |
"step": 910
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 3.18,
|
| 731 |
-
"grad_norm": 6.
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 920
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 3.22,
|
| 738 |
-
"grad_norm":
|
| 739 |
-
"learning_rate":
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 930
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 3.25,
|
| 745 |
-
"grad_norm":
|
| 746 |
-
"learning_rate":
|
| 747 |
-
"loss": 0.
|
| 748 |
"step": 940
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 3.29,
|
| 752 |
-
"grad_norm":
|
| 753 |
-
"learning_rate": 6.
|
| 754 |
-
"loss": 0.
|
| 755 |
"step": 950
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 3.32,
|
| 759 |
-
"grad_norm":
|
| 760 |
-
"learning_rate": 6.
|
| 761 |
-
"loss": 0.
|
| 762 |
"step": 960
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 3.36,
|
| 766 |
-
"grad_norm":
|
| 767 |
-
"learning_rate": 6.
|
| 768 |
-
"loss": 0.
|
| 769 |
"step": 970
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 3.39,
|
| 773 |
-
"grad_norm":
|
| 774 |
-
"learning_rate": 6.
|
| 775 |
-
"loss": 0.
|
| 776 |
"step": 980
|
| 777 |
},
|
| 778 |
{
|
| 779 |
"epoch": 3.43,
|
| 780 |
-
"grad_norm":
|
| 781 |
-
"learning_rate": 6.
|
| 782 |
-
"loss": 0.
|
| 783 |
"step": 990
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 3.46,
|
| 787 |
-
"grad_norm":
|
| 788 |
-
"learning_rate": 6.
|
| 789 |
-
"loss": 0.
|
| 790 |
"step": 1000
|
| 791 |
},
|
| 792 |
{
|
| 793 |
"epoch": 3.46,
|
| 794 |
-
"eval_accuracy": 0.
|
| 795 |
-
"eval_loss": 0.
|
| 796 |
-
"eval_runtime": 11.
|
| 797 |
-
"eval_samples_per_second":
|
| 798 |
-
"eval_steps_per_second":
|
| 799 |
"step": 1000
|
| 800 |
},
|
| 801 |
{
|
| 802 |
"epoch": 3.49,
|
| 803 |
-
"grad_norm":
|
| 804 |
-
"learning_rate": 6.
|
| 805 |
-
"loss": 0.
|
| 806 |
"step": 1010
|
| 807 |
},
|
| 808 |
{
|
| 809 |
"epoch": 3.53,
|
| 810 |
-
"grad_norm":
|
| 811 |
-
"learning_rate":
|
| 812 |
-
"loss": 0.
|
| 813 |
"step": 1020
|
| 814 |
},
|
| 815 |
{
|
| 816 |
"epoch": 3.56,
|
| 817 |
-
"grad_norm":
|
| 818 |
-
"learning_rate":
|
| 819 |
-
"loss": 0.
|
| 820 |
"step": 1030
|
| 821 |
},
|
| 822 |
{
|
| 823 |
"epoch": 3.6,
|
| 824 |
-
"grad_norm":
|
| 825 |
-
"learning_rate":
|
| 826 |
-
"loss": 0.
|
| 827 |
"step": 1040
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 3.63,
|
| 831 |
-
"grad_norm":
|
| 832 |
-
"learning_rate":
|
| 833 |
-
"loss": 0.
|
| 834 |
"step": 1050
|
| 835 |
},
|
| 836 |
{
|
| 837 |
"epoch": 3.67,
|
| 838 |
-
"grad_norm":
|
| 839 |
-
"learning_rate":
|
| 840 |
-
"loss": 0.
|
| 841 |
"step": 1060
|
| 842 |
},
|
| 843 |
{
|
| 844 |
"epoch": 3.7,
|
| 845 |
-
"grad_norm":
|
| 846 |
-
"learning_rate":
|
| 847 |
-
"loss": 0.
|
| 848 |
"step": 1070
|
| 849 |
},
|
| 850 |
{
|
| 851 |
"epoch": 3.74,
|
| 852 |
-
"grad_norm": 2.
|
| 853 |
-
"learning_rate":
|
| 854 |
-
"loss": 0.
|
| 855 |
"step": 1080
|
| 856 |
},
|
| 857 |
{
|
| 858 |
"epoch": 3.77,
|
| 859 |
-
"grad_norm":
|
| 860 |
-
"learning_rate":
|
| 861 |
-
"loss": 0.
|
| 862 |
"step": 1090
|
| 863 |
},
|
| 864 |
{
|
| 865 |
"epoch": 3.81,
|
| 866 |
-
"grad_norm":
|
| 867 |
-
"learning_rate":
|
| 868 |
-
"loss": 0.
|
| 869 |
"step": 1100
|
| 870 |
},
|
| 871 |
{
|
| 872 |
"epoch": 3.81,
|
| 873 |
-
"eval_accuracy": 0.
|
| 874 |
-
"eval_loss": 0.
|
| 875 |
-
"eval_runtime": 11.
|
| 876 |
-
"eval_samples_per_second":
|
| 877 |
-
"eval_steps_per_second":
|
| 878 |
"step": 1100
|
| 879 |
},
|
| 880 |
{
|
| 881 |
"epoch": 3.84,
|
| 882 |
-
"grad_norm": 2.
|
| 883 |
-
"learning_rate":
|
| 884 |
-
"loss": 0.
|
| 885 |
"step": 1110
|
| 886 |
},
|
| 887 |
{
|
| 888 |
"epoch": 3.88,
|
| 889 |
-
"grad_norm":
|
| 890 |
-
"learning_rate":
|
| 891 |
-
"loss": 0.
|
| 892 |
"step": 1120
|
| 893 |
},
|
| 894 |
{
|
| 895 |
"epoch": 3.91,
|
| 896 |
-
"grad_norm":
|
| 897 |
-
"learning_rate":
|
| 898 |
-
"loss": 0.
|
| 899 |
"step": 1130
|
| 900 |
},
|
| 901 |
{
|
| 902 |
"epoch": 3.94,
|
| 903 |
-
"grad_norm": 2.
|
| 904 |
-
"learning_rate":
|
| 905 |
-
"loss": 0.
|
| 906 |
"step": 1140
|
| 907 |
},
|
| 908 |
{
|
| 909 |
"epoch": 3.98,
|
| 910 |
-
"grad_norm": 6.
|
| 911 |
-
"learning_rate":
|
| 912 |
-
"loss": 0.
|
| 913 |
"step": 1150
|
| 914 |
},
|
| 915 |
{
|
| 916 |
"epoch": 4.01,
|
| 917 |
-
"grad_norm":
|
| 918 |
-
"learning_rate":
|
| 919 |
-
"loss": 0.
|
| 920 |
"step": 1160
|
| 921 |
},
|
| 922 |
{
|
| 923 |
"epoch": 4.05,
|
| 924 |
-
"grad_norm":
|
| 925 |
-
"learning_rate":
|
| 926 |
-
"loss": 0.
|
| 927 |
"step": 1170
|
| 928 |
},
|
| 929 |
{
|
| 930 |
"epoch": 4.08,
|
| 931 |
-
"grad_norm":
|
| 932 |
-
"learning_rate":
|
| 933 |
-
"loss": 0.
|
| 934 |
"step": 1180
|
| 935 |
},
|
| 936 |
{
|
| 937 |
"epoch": 4.12,
|
| 938 |
-
"grad_norm":
|
| 939 |
-
"learning_rate":
|
| 940 |
-
"loss": 0.
|
| 941 |
"step": 1190
|
| 942 |
},
|
| 943 |
{
|
| 944 |
"epoch": 4.15,
|
| 945 |
-
"grad_norm":
|
| 946 |
-
"learning_rate":
|
| 947 |
-
"loss": 0.
|
| 948 |
"step": 1200
|
| 949 |
},
|
| 950 |
{
|
| 951 |
"epoch": 4.15,
|
| 952 |
-
"eval_accuracy": 0.
|
| 953 |
-
"eval_loss": 0.
|
| 954 |
-
"eval_runtime": 11.
|
| 955 |
-
"eval_samples_per_second":
|
| 956 |
-
"eval_steps_per_second":
|
| 957 |
"step": 1200
|
| 958 |
},
|
| 959 |
{
|
| 960 |
"epoch": 4.19,
|
| 961 |
-
"grad_norm":
|
| 962 |
-
"learning_rate":
|
| 963 |
-
"loss": 0.
|
| 964 |
"step": 1210
|
| 965 |
},
|
| 966 |
{
|
| 967 |
"epoch": 4.22,
|
| 968 |
-
"grad_norm":
|
| 969 |
-
"learning_rate":
|
| 970 |
-
"loss": 0.
|
| 971 |
"step": 1220
|
| 972 |
},
|
| 973 |
{
|
| 974 |
"epoch": 4.26,
|
| 975 |
-
"grad_norm":
|
| 976 |
-
"learning_rate":
|
| 977 |
-
"loss": 0.
|
| 978 |
"step": 1230
|
| 979 |
},
|
| 980 |
{
|
| 981 |
"epoch": 4.29,
|
| 982 |
-
"grad_norm":
|
| 983 |
-
"learning_rate":
|
| 984 |
-
"loss": 0.
|
| 985 |
"step": 1240
|
| 986 |
},
|
| 987 |
{
|
| 988 |
"epoch": 4.33,
|
| 989 |
-
"grad_norm": 3.
|
| 990 |
-
"learning_rate":
|
| 991 |
-
"loss": 0.
|
| 992 |
"step": 1250
|
| 993 |
},
|
| 994 |
{
|
| 995 |
"epoch": 4.36,
|
| 996 |
-
"grad_norm":
|
| 997 |
-
"learning_rate":
|
| 998 |
-
"loss": 0.
|
| 999 |
"step": 1260
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
"epoch": 4.39,
|
| 1003 |
-
"grad_norm":
|
| 1004 |
-
"learning_rate":
|
| 1005 |
-
"loss": 0.
|
| 1006 |
"step": 1270
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
"epoch": 4.43,
|
| 1010 |
-
"grad_norm":
|
| 1011 |
-
"learning_rate":
|
| 1012 |
-
"loss": 0.
|
| 1013 |
"step": 1280
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
"epoch": 4.46,
|
| 1017 |
-
"grad_norm":
|
| 1018 |
-
"learning_rate":
|
| 1019 |
-
"loss": 0.
|
| 1020 |
"step": 1290
|
| 1021 |
},
|
| 1022 |
{
|
| 1023 |
"epoch": 4.5,
|
| 1024 |
-
"grad_norm":
|
| 1025 |
-
"learning_rate":
|
| 1026 |
-
"loss": 0.
|
| 1027 |
"step": 1300
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
"epoch": 4.5,
|
| 1031 |
-
"eval_accuracy": 0.
|
| 1032 |
-
"eval_loss": 0.
|
| 1033 |
-
"eval_runtime": 11.
|
| 1034 |
-
"eval_samples_per_second":
|
| 1035 |
-
"eval_steps_per_second":
|
| 1036 |
"step": 1300
|
| 1037 |
},
|
| 1038 |
{
|
| 1039 |
"epoch": 4.53,
|
| 1040 |
-
"grad_norm":
|
| 1041 |
-
"learning_rate":
|
| 1042 |
-
"loss": 0.
|
| 1043 |
"step": 1310
|
| 1044 |
},
|
| 1045 |
{
|
| 1046 |
"epoch": 4.57,
|
| 1047 |
-
"grad_norm":
|
| 1048 |
-
"learning_rate":
|
| 1049 |
-
"loss": 0.
|
| 1050 |
"step": 1320
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
"epoch": 4.6,
|
| 1054 |
-
"grad_norm": 0.
|
| 1055 |
-
"learning_rate":
|
| 1056 |
-
"loss": 0.
|
| 1057 |
"step": 1330
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
"epoch": 4.64,
|
| 1061 |
-
"grad_norm":
|
| 1062 |
-
"learning_rate":
|
| 1063 |
-
"loss": 0.
|
| 1064 |
"step": 1340
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
"epoch": 4.67,
|
| 1068 |
-
"grad_norm":
|
| 1069 |
-
"learning_rate":
|
| 1070 |
-
"loss": 0.
|
| 1071 |
"step": 1350
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
"epoch": 4.71,
|
| 1075 |
-
"grad_norm":
|
| 1076 |
-
"learning_rate":
|
| 1077 |
-
"loss": 0.
|
| 1078 |
"step": 1360
|
| 1079 |
},
|
| 1080 |
{
|
| 1081 |
"epoch": 4.74,
|
| 1082 |
-
"grad_norm":
|
| 1083 |
-
"learning_rate":
|
| 1084 |
-
"loss": 0.
|
| 1085 |
"step": 1370
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
"epoch": 4.78,
|
| 1089 |
-
"grad_norm":
|
| 1090 |
-
"learning_rate":
|
| 1091 |
-
"loss": 0.
|
| 1092 |
"step": 1380
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
"epoch": 4.81,
|
| 1096 |
-
"grad_norm":
|
| 1097 |
-
"learning_rate":
|
| 1098 |
-
"loss": 0.
|
| 1099 |
"step": 1390
|
| 1100 |
},
|
| 1101 |
{
|
| 1102 |
"epoch": 4.84,
|
| 1103 |
-
"grad_norm":
|
| 1104 |
-
"learning_rate":
|
| 1105 |
-
"loss": 0.
|
| 1106 |
"step": 1400
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
"epoch": 4.84,
|
| 1110 |
-
"eval_accuracy": 0.
|
| 1111 |
-
"eval_loss": 0.
|
| 1112 |
-
"eval_runtime": 11.
|
| 1113 |
-
"eval_samples_per_second": 70.
|
| 1114 |
-
"eval_steps_per_second": 8.
|
| 1115 |
"step": 1400
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
"epoch": 4.88,
|
| 1119 |
-
"grad_norm":
|
| 1120 |
-
"learning_rate": 5.
|
| 1121 |
-
"loss": 0.
|
| 1122 |
"step": 1410
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
"epoch": 4.91,
|
| 1126 |
-
"grad_norm":
|
| 1127 |
-
"learning_rate":
|
| 1128 |
-
"loss": 0.
|
| 1129 |
"step": 1420
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
"epoch": 4.95,
|
| 1133 |
-
"grad_norm":
|
| 1134 |
-
"learning_rate":
|
| 1135 |
-
"loss": 0.
|
| 1136 |
"step": 1430
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
"epoch": 4.98,
|
| 1140 |
-
"grad_norm":
|
| 1141 |
-
"learning_rate":
|
| 1142 |
-
"loss": 0.
|
| 1143 |
"step": 1440
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
-
"epoch": 5.
|
| 1147 |
-
"
|
| 1148 |
-
"
|
| 1149 |
-
"
|
| 1150 |
-
"
|
| 1151 |
-
|
| 1152 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1153 |
}
|
| 1154 |
],
|
| 1155 |
"logging_steps": 10,
|
| 1156 |
-
"max_steps":
|
| 1157 |
"num_input_tokens_seen": 0,
|
| 1158 |
-
"num_train_epochs":
|
| 1159 |
"save_steps": 100,
|
| 1160 |
-
"total_flos":
|
| 1161 |
"train_batch_size": 16,
|
| 1162 |
"trial_name": null,
|
| 1163 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.46183693408966064,
|
| 3 |
+
"best_model_checkpoint": "Action_all_10_class/checkpoint-1600",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 2890,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.03,
|
| 13 |
+
"grad_norm": 1.9832468032836914,
|
| 14 |
+
"learning_rate": 9.965397923875432e-05,
|
| 15 |
+
"loss": 2.2618,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.07,
|
| 20 |
+
"grad_norm": 1.9726794958114624,
|
| 21 |
+
"learning_rate": 9.930795847750865e-05,
|
| 22 |
+
"loss": 2.1513,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.1,
|
| 27 |
+
"grad_norm": 1.8990365266799927,
|
| 28 |
+
"learning_rate": 9.896193771626297e-05,
|
| 29 |
+
"loss": 2.0562,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.14,
|
| 34 |
+
"grad_norm": 1.9971903562545776,
|
| 35 |
+
"learning_rate": 9.86159169550173e-05,
|
| 36 |
+
"loss": 1.9375,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.17,
|
| 41 |
+
"grad_norm": 2.781921148300171,
|
| 42 |
+
"learning_rate": 9.826989619377162e-05,
|
| 43 |
+
"loss": 1.7386,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.21,
|
| 48 |
+
"grad_norm": 2.6308515071868896,
|
| 49 |
+
"learning_rate": 9.792387543252595e-05,
|
| 50 |
+
"loss": 1.5669,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.24,
|
| 55 |
+
"grad_norm": 3.349245309829712,
|
| 56 |
+
"learning_rate": 9.757785467128027e-05,
|
| 57 |
+
"loss": 1.4557,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.28,
|
| 62 |
+
"grad_norm": 3.6303634643554688,
|
| 63 |
+
"learning_rate": 9.72318339100346e-05,
|
| 64 |
+
"loss": 1.3707,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.31,
|
| 69 |
+
"grad_norm": 4.0813469886779785,
|
| 70 |
+
"learning_rate": 9.688581314878892e-05,
|
| 71 |
+
"loss": 1.3667,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.35,
|
| 76 |
+
"grad_norm": 4.506680488586426,
|
| 77 |
+
"learning_rate": 9.653979238754325e-05,
|
| 78 |
+
"loss": 1.1996,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.35,
|
| 83 |
+
"eval_accuracy": 0.7730061349693251,
|
| 84 |
+
"eval_loss": 1.0634636878967285,
|
| 85 |
+
"eval_runtime": 16.9399,
|
| 86 |
+
"eval_samples_per_second": 48.111,
|
| 87 |
+
"eval_steps_per_second": 6.021,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.38,
|
| 92 |
+
"grad_norm": 2.7484147548675537,
|
| 93 |
+
"learning_rate": 9.619377162629759e-05,
|
| 94 |
+
"loss": 1.1942,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.42,
|
| 99 |
+
"grad_norm": 3.1658833026885986,
|
| 100 |
+
"learning_rate": 9.584775086505192e-05,
|
| 101 |
+
"loss": 1.2623,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 0.45,
|
| 106 |
+
"grad_norm": 3.366626024246216,
|
| 107 |
+
"learning_rate": 9.550173010380624e-05,
|
| 108 |
+
"loss": 1.0979,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 0.48,
|
| 113 |
+
"grad_norm": 4.570044994354248,
|
| 114 |
+
"learning_rate": 9.515570934256057e-05,
|
| 115 |
+
"loss": 1.0803,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.52,
|
| 120 |
+
"grad_norm": 3.996645927429199,
|
| 121 |
+
"learning_rate": 9.480968858131488e-05,
|
| 122 |
+
"loss": 1.069,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.55,
|
| 127 |
+
"grad_norm": 4.534286975860596,
|
| 128 |
+
"learning_rate": 9.446366782006922e-05,
|
| 129 |
+
"loss": 1.0673,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.59,
|
| 134 |
+
"grad_norm": 4.0312581062316895,
|
| 135 |
+
"learning_rate": 9.411764705882353e-05,
|
| 136 |
+
"loss": 1.0314,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 0.62,
|
| 141 |
+
"grad_norm": 3.871572494506836,
|
| 142 |
+
"learning_rate": 9.380622837370243e-05,
|
| 143 |
+
"loss": 1.056,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 0.66,
|
| 148 |
+
"grad_norm": 4.05521821975708,
|
| 149 |
+
"learning_rate": 9.346020761245676e-05,
|
| 150 |
+
"loss": 0.9677,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 0.69,
|
| 155 |
+
"grad_norm": 5.704447269439697,
|
| 156 |
+
"learning_rate": 9.311418685121109e-05,
|
| 157 |
+
"loss": 1.0335,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 0.69,
|
| 162 |
+
"eval_accuracy": 0.7717791411042945,
|
| 163 |
+
"eval_loss": 0.8391550779342651,
|
| 164 |
+
"eval_runtime": 11.6297,
|
| 165 |
+
"eval_samples_per_second": 70.079,
|
| 166 |
+
"eval_steps_per_second": 8.771,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.73,
|
| 171 |
+
"grad_norm": 6.939559459686279,
|
| 172 |
+
"learning_rate": 9.27681660899654e-05,
|
| 173 |
+
"loss": 0.9873,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 0.76,
|
| 178 |
+
"grad_norm": 4.459825038909912,
|
| 179 |
+
"learning_rate": 9.242214532871972e-05,
|
| 180 |
+
"loss": 0.9642,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 0.8,
|
| 185 |
+
"grad_norm": 5.480250835418701,
|
| 186 |
+
"learning_rate": 9.207612456747406e-05,
|
| 187 |
+
"loss": 0.9012,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 0.83,
|
| 192 |
+
"grad_norm": 3.0692737102508545,
|
| 193 |
+
"learning_rate": 9.173010380622837e-05,
|
| 194 |
+
"loss": 0.9618,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 0.87,
|
| 199 |
+
"grad_norm": 2.740853786468506,
|
| 200 |
+
"learning_rate": 9.13840830449827e-05,
|
| 201 |
+
"loss": 0.8414,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 0.9,
|
| 206 |
+
"grad_norm": 3.637920379638672,
|
| 207 |
+
"learning_rate": 9.103806228373702e-05,
|
| 208 |
+
"loss": 0.908,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 0.93,
|
| 213 |
+
"grad_norm": 3.488283634185791,
|
| 214 |
+
"learning_rate": 9.069204152249136e-05,
|
| 215 |
+
"loss": 0.9001,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 0.97,
|
| 220 |
+
"grad_norm": 4.264199256896973,
|
| 221 |
+
"learning_rate": 9.034602076124567e-05,
|
| 222 |
+
"loss": 0.8283,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 1.0,
|
| 227 |
+
"grad_norm": 3.3874425888061523,
|
| 228 |
+
"learning_rate": 9e-05,
|
| 229 |
+
"loss": 0.7862,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 1.04,
|
| 234 |
+
"grad_norm": 3.339252233505249,
|
| 235 |
+
"learning_rate": 8.965397923875432e-05,
|
| 236 |
+
"loss": 0.6279,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 1.04,
|
| 241 |
+
"eval_accuracy": 0.8294478527607362,
|
| 242 |
+
"eval_loss": 0.6463050842285156,
|
| 243 |
+
"eval_runtime": 12.9239,
|
| 244 |
+
"eval_samples_per_second": 63.061,
|
| 245 |
+
"eval_steps_per_second": 7.892,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 1.07,
|
| 250 |
+
"grad_norm": 5.215779781341553,
|
| 251 |
+
"learning_rate": 8.930795847750866e-05,
|
| 252 |
+
"loss": 0.7314,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 1.11,
|
| 257 |
+
"grad_norm": 5.00214958190918,
|
| 258 |
+
"learning_rate": 8.896193771626297e-05,
|
| 259 |
+
"loss": 0.6857,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 1.14,
|
| 264 |
+
"grad_norm": 4.8398542404174805,
|
| 265 |
+
"learning_rate": 8.86159169550173e-05,
|
| 266 |
+
"loss": 0.767,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 1.18,
|
| 271 |
+
"grad_norm": 4.89760160446167,
|
| 272 |
+
"learning_rate": 8.826989619377162e-05,
|
| 273 |
+
"loss": 0.6085,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 1.21,
|
| 278 |
+
"grad_norm": 3.459028720855713,
|
| 279 |
+
"learning_rate": 8.792387543252596e-05,
|
| 280 |
+
"loss": 0.6735,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 1.25,
|
| 285 |
+
"grad_norm": 5.60736608505249,
|
| 286 |
+
"learning_rate": 8.757785467128027e-05,
|
| 287 |
+
"loss": 0.8843,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 1.28,
|
| 292 |
+
"grad_norm": 4.042627811431885,
|
| 293 |
+
"learning_rate": 8.72318339100346e-05,
|
| 294 |
+
"loss": 0.7819,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 1.31,
|
| 299 |
+
"grad_norm": 3.9558098316192627,
|
| 300 |
+
"learning_rate": 8.688581314878894e-05,
|
| 301 |
+
"loss": 0.8151,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 1.35,
|
| 306 |
+
"grad_norm": 2.589348554611206,
|
| 307 |
+
"learning_rate": 8.653979238754327e-05,
|
| 308 |
+
"loss": 0.7008,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 1.38,
|
| 313 |
+
"grad_norm": 5.536679744720459,
|
| 314 |
+
"learning_rate": 8.619377162629759e-05,
|
| 315 |
+
"loss": 0.8633,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 1.38,
|
| 320 |
+
"eval_accuracy": 0.7926380368098159,
|
| 321 |
+
"eval_loss": 0.7172089219093323,
|
| 322 |
+
"eval_runtime": 11.6484,
|
| 323 |
+
"eval_samples_per_second": 69.967,
|
| 324 |
+
"eval_steps_per_second": 8.757,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 1.42,
|
| 329 |
+
"grad_norm": 3.1925909519195557,
|
| 330 |
+
"learning_rate": 8.584775086505192e-05,
|
| 331 |
+
"loss": 0.7146,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 1.45,
|
| 336 |
+
"grad_norm": 1.3583850860595703,
|
| 337 |
+
"learning_rate": 8.550173010380624e-05,
|
| 338 |
+
"loss": 0.6679,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 1.49,
|
| 343 |
+
"grad_norm": 2.476952075958252,
|
| 344 |
+
"learning_rate": 8.515570934256055e-05,
|
| 345 |
+
"loss": 0.7093,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 1.52,
|
| 350 |
+
"grad_norm": 4.362217426300049,
|
| 351 |
+
"learning_rate": 8.480968858131489e-05,
|
| 352 |
+
"loss": 0.5813,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 1.56,
|
| 357 |
+
"grad_norm": 6.683523178100586,
|
| 358 |
+
"learning_rate": 8.44636678200692e-05,
|
| 359 |
+
"loss": 0.6022,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 1.59,
|
| 364 |
+
"grad_norm": 5.650263786315918,
|
| 365 |
+
"learning_rate": 8.411764705882354e-05,
|
| 366 |
+
"loss": 0.6538,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 1.63,
|
| 371 |
+
"grad_norm": 2.9296507835388184,
|
| 372 |
+
"learning_rate": 8.377162629757785e-05,
|
| 373 |
+
"loss": 0.7312,
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 1.66,
|
| 378 |
+
"grad_norm": 4.2093048095703125,
|
| 379 |
+
"learning_rate": 8.342560553633219e-05,
|
| 380 |
+
"loss": 0.5509,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 1.7,
|
| 385 |
+
"grad_norm": 4.481240749359131,
|
| 386 |
+
"learning_rate": 8.30795847750865e-05,
|
| 387 |
+
"loss": 0.7108,
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 1.73,
|
| 392 |
+
"grad_norm": 6.15458869934082,
|
| 393 |
+
"learning_rate": 8.273356401384084e-05,
|
| 394 |
+
"loss": 0.5851,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 1.73,
|
| 399 |
+
"eval_accuracy": 0.838036809815951,
|
| 400 |
+
"eval_loss": 0.5857788324356079,
|
| 401 |
+
"eval_runtime": 11.1755,
|
| 402 |
+
"eval_samples_per_second": 72.927,
|
| 403 |
+
"eval_steps_per_second": 9.127,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 1.76,
|
| 408 |
+
"grad_norm": 5.6646409034729,
|
| 409 |
+
"learning_rate": 8.238754325259515e-05,
|
| 410 |
+
"loss": 0.7466,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 1.8,
|
| 415 |
+
"grad_norm": 3.708621025085449,
|
| 416 |
+
"learning_rate": 8.204152249134948e-05,
|
| 417 |
+
"loss": 0.6417,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 1.83,
|
| 422 |
+
"grad_norm": 4.252438545227051,
|
| 423 |
+
"learning_rate": 8.16955017301038e-05,
|
| 424 |
+
"loss": 0.5843,
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 1.87,
|
| 429 |
+
"grad_norm": 6.532431602478027,
|
| 430 |
+
"learning_rate": 8.134948096885813e-05,
|
| 431 |
+
"loss": 0.7773,
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 1.9,
|
| 436 |
+
"grad_norm": 2.0311672687530518,
|
| 437 |
+
"learning_rate": 8.100346020761245e-05,
|
| 438 |
+
"loss": 0.6273,
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 1.94,
|
| 443 |
+
"grad_norm": 4.033968448638916,
|
| 444 |
+
"learning_rate": 8.065743944636678e-05,
|
| 445 |
+
"loss": 0.6618,
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 1.97,
|
| 450 |
+
"grad_norm": 3.268759250640869,
|
| 451 |
+
"learning_rate": 8.03114186851211e-05,
|
| 452 |
+
"loss": 0.5404,
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 2.01,
|
| 457 |
+
"grad_norm": 5.866196632385254,
|
| 458 |
+
"learning_rate": 7.996539792387543e-05,
|
| 459 |
+
"loss": 0.6579,
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 2.04,
|
| 464 |
+
"grad_norm": 1.0091450214385986,
|
| 465 |
+
"learning_rate": 7.961937716262977e-05,
|
| 466 |
+
"loss": 0.4575,
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 2.08,
|
| 471 |
+
"grad_norm": 4.265220642089844,
|
| 472 |
+
"learning_rate": 7.92733564013841e-05,
|
| 473 |
+
"loss": 0.5305,
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 2.08,
|
| 478 |
+
"eval_accuracy": 0.8355828220858895,
|
| 479 |
+
"eval_loss": 0.5779993534088135,
|
| 480 |
+
"eval_runtime": 12.1337,
|
| 481 |
+
"eval_samples_per_second": 67.168,
|
| 482 |
+
"eval_steps_per_second": 8.406,
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 2.11,
|
| 487 |
+
"grad_norm": 3.7668848037719727,
|
| 488 |
+
"learning_rate": 7.892733564013842e-05,
|
| 489 |
+
"loss": 0.5928,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 2.15,
|
| 494 |
+
"grad_norm": 3.6394691467285156,
|
| 495 |
+
"learning_rate": 7.858131487889275e-05,
|
| 496 |
+
"loss": 0.533,
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 2.18,
|
| 501 |
+
"grad_norm": 4.226925373077393,
|
| 502 |
+
"learning_rate": 7.823529411764707e-05,
|
| 503 |
+
"loss": 0.4776,
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 2.21,
|
| 508 |
+
"grad_norm": 6.159184455871582,
|
| 509 |
+
"learning_rate": 7.78892733564014e-05,
|
| 510 |
+
"loss": 0.5517,
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 2.25,
|
| 515 |
+
"grad_norm": 7.021705150604248,
|
| 516 |
+
"learning_rate": 7.754325259515572e-05,
|
| 517 |
+
"loss": 0.4969,
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 2.28,
|
| 522 |
+
"grad_norm": 5.880521297454834,
|
| 523 |
+
"learning_rate": 7.719723183391003e-05,
|
| 524 |
+
"loss": 0.5172,
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 2.32,
|
| 529 |
+
"grad_norm": 4.049355506896973,
|
| 530 |
+
"learning_rate": 7.685121107266437e-05,
|
| 531 |
+
"loss": 0.5188,
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 2.35,
|
| 536 |
+
"grad_norm": 6.603725433349609,
|
| 537 |
+
"learning_rate": 7.650519031141868e-05,
|
| 538 |
+
"loss": 0.5329,
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 2.39,
|
| 543 |
+
"grad_norm": 6.908223628997803,
|
| 544 |
+
"learning_rate": 7.615916955017301e-05,
|
| 545 |
+
"loss": 0.5938,
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 2.42,
|
| 550 |
+
"grad_norm": 5.728999614715576,
|
| 551 |
+
"learning_rate": 7.581314878892733e-05,
|
| 552 |
+
"loss": 0.5511,
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 2.42,
|
| 557 |
+
"eval_accuracy": 0.8392638036809816,
|
| 558 |
+
"eval_loss": 0.5313203930854797,
|
| 559 |
+
"eval_runtime": 11.9712,
|
| 560 |
+
"eval_samples_per_second": 68.08,
|
| 561 |
+
"eval_steps_per_second": 8.52,
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 2.46,
|
| 566 |
+
"grad_norm": 3.414422035217285,
|
| 567 |
+
"learning_rate": 7.546712802768166e-05,
|
| 568 |
+
"loss": 0.4702,
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 2.49,
|
| 573 |
+
"grad_norm": 7.404133319854736,
|
| 574 |
+
"learning_rate": 7.512110726643598e-05,
|
| 575 |
+
"loss": 0.5877,
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 2.53,
|
| 580 |
+
"grad_norm": 4.851989269256592,
|
| 581 |
+
"learning_rate": 7.477508650519031e-05,
|
| 582 |
+
"loss": 0.5855,
|
| 583 |
"step": 730
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 2.56,
|
| 587 |
+
"grad_norm": 6.651949882507324,
|
| 588 |
+
"learning_rate": 7.442906574394463e-05,
|
| 589 |
+
"loss": 0.5618,
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 2.6,
|
| 594 |
+
"grad_norm": 9.513969421386719,
|
| 595 |
+
"learning_rate": 7.408304498269896e-05,
|
| 596 |
+
"loss": 0.5871,
|
| 597 |
"step": 750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 2.63,
|
| 601 |
+
"grad_norm": 11.049631118774414,
|
| 602 |
+
"learning_rate": 7.373702422145328e-05,
|
| 603 |
+
"loss": 0.4098,
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 2.66,
|
| 608 |
+
"grad_norm": 4.604189872741699,
|
| 609 |
+
"learning_rate": 7.339100346020761e-05,
|
| 610 |
+
"loss": 0.5752,
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 2.7,
|
| 615 |
+
"grad_norm": 5.650027275085449,
|
| 616 |
+
"learning_rate": 7.304498269896195e-05,
|
| 617 |
+
"loss": 0.6265,
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 2.73,
|
| 622 |
+
"grad_norm": 7.081886291503906,
|
| 623 |
+
"learning_rate": 7.269896193771626e-05,
|
| 624 |
+
"loss": 0.4814,
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 2.77,
|
| 629 |
+
"grad_norm": 9.59929084777832,
|
| 630 |
+
"learning_rate": 7.23529411764706e-05,
|
| 631 |
+
"loss": 0.4657,
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 2.77,
|
| 636 |
+
"eval_accuracy": 0.8368098159509203,
|
| 637 |
+
"eval_loss": 0.5443019866943359,
|
| 638 |
+
"eval_runtime": 14.7103,
|
| 639 |
+
"eval_samples_per_second": 55.403,
|
| 640 |
+
"eval_steps_per_second": 6.934,
|
| 641 |
"step": 800
|
| 642 |
},
|
| 643 |
{
|
| 644 |
"epoch": 2.8,
|
| 645 |
+
"grad_norm": 3.963488817214966,
|
| 646 |
+
"learning_rate": 7.200692041522493e-05,
|
| 647 |
+
"loss": 0.4586,
|
| 648 |
"step": 810
|
| 649 |
},
|
| 650 |
{
|
| 651 |
"epoch": 2.84,
|
| 652 |
+
"grad_norm": 5.124209403991699,
|
| 653 |
+
"learning_rate": 7.166089965397925e-05,
|
| 654 |
+
"loss": 0.4099,
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
"epoch": 2.87,
|
| 659 |
+
"grad_norm": 2.2323999404907227,
|
| 660 |
+
"learning_rate": 7.131487889273358e-05,
|
| 661 |
+
"loss": 0.5683,
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 2.91,
|
| 666 |
+
"grad_norm": 6.542498588562012,
|
| 667 |
+
"learning_rate": 7.09688581314879e-05,
|
| 668 |
+
"loss": 0.4908,
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 2.94,
|
| 673 |
+
"grad_norm": 5.042673110961914,
|
| 674 |
+
"learning_rate": 7.062283737024223e-05,
|
| 675 |
+
"loss": 0.5441,
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 2.98,
|
| 680 |
+
"grad_norm": 7.417181491851807,
|
| 681 |
+
"learning_rate": 7.027681660899654e-05,
|
| 682 |
+
"loss": 0.4688,
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 3.01,
|
| 687 |
+
"grad_norm": 3.395101547241211,
|
| 688 |
+
"learning_rate": 6.993079584775086e-05,
|
| 689 |
+
"loss": 0.6159,
|
| 690 |
"step": 870
|
| 691 |
},
|
| 692 |
{
|
| 693 |
"epoch": 3.04,
|
| 694 |
+
"grad_norm": 4.752573013305664,
|
| 695 |
+
"learning_rate": 6.95847750865052e-05,
|
| 696 |
+
"loss": 0.4304,
|
| 697 |
"step": 880
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 3.08,
|
| 701 |
+
"grad_norm": 6.356733322143555,
|
| 702 |
+
"learning_rate": 6.923875432525951e-05,
|
| 703 |
+
"loss": 0.4116,
|
| 704 |
"step": 890
|
| 705 |
},
|
| 706 |
{
|
| 707 |
"epoch": 3.11,
|
| 708 |
+
"grad_norm": 3.9821412563323975,
|
| 709 |
+
"learning_rate": 6.889273356401384e-05,
|
| 710 |
+
"loss": 0.3615,
|
| 711 |
"step": 900
|
| 712 |
},
|
| 713 |
{
|
| 714 |
"epoch": 3.11,
|
| 715 |
+
"eval_accuracy": 0.8429447852760736,
|
| 716 |
+
"eval_loss": 0.5038174986839294,
|
| 717 |
+
"eval_runtime": 11.7579,
|
| 718 |
+
"eval_samples_per_second": 69.315,
|
| 719 |
+
"eval_steps_per_second": 8.675,
|
| 720 |
"step": 900
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 3.15,
|
| 724 |
+
"grad_norm": 4.009409427642822,
|
| 725 |
+
"learning_rate": 6.854671280276816e-05,
|
| 726 |
+
"loss": 0.3844,
|
| 727 |
"step": 910
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 3.18,
|
| 731 |
+
"grad_norm": 6.364121437072754,
|
| 732 |
+
"learning_rate": 6.82006920415225e-05,
|
| 733 |
+
"loss": 0.3863,
|
| 734 |
"step": 920
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 3.22,
|
| 738 |
+
"grad_norm": 5.382229804992676,
|
| 739 |
+
"learning_rate": 6.785467128027681e-05,
|
| 740 |
+
"loss": 0.4159,
|
| 741 |
"step": 930
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 3.25,
|
| 745 |
+
"grad_norm": 3.8384909629821777,
|
| 746 |
+
"learning_rate": 6.750865051903114e-05,
|
| 747 |
+
"loss": 0.4783,
|
| 748 |
"step": 940
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 3.29,
|
| 752 |
+
"grad_norm": 0.2637701630592346,
|
| 753 |
+
"learning_rate": 6.716262975778546e-05,
|
| 754 |
+
"loss": 0.4072,
|
| 755 |
"step": 950
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 3.32,
|
| 759 |
+
"grad_norm": 3.8876795768737793,
|
| 760 |
+
"learning_rate": 6.68166089965398e-05,
|
| 761 |
+
"loss": 0.3698,
|
| 762 |
"step": 960
|
| 763 |
},
|
| 764 |
{
|
| 765 |
"epoch": 3.36,
|
| 766 |
+
"grad_norm": 2.9388973712921143,
|
| 767 |
+
"learning_rate": 6.647058823529411e-05,
|
| 768 |
+
"loss": 0.4197,
|
| 769 |
"step": 970
|
| 770 |
},
|
| 771 |
{
|
| 772 |
"epoch": 3.39,
|
| 773 |
+
"grad_norm": 1.423226237297058,
|
| 774 |
+
"learning_rate": 6.612456747404844e-05,
|
| 775 |
+
"loss": 0.344,
|
| 776 |
"step": 980
|
| 777 |
},
|
| 778 |
{
|
| 779 |
"epoch": 3.43,
|
| 780 |
+
"grad_norm": 5.080854892730713,
|
| 781 |
+
"learning_rate": 6.577854671280278e-05,
|
| 782 |
+
"loss": 0.3596,
|
| 783 |
"step": 990
|
| 784 |
},
|
| 785 |
{
|
| 786 |
"epoch": 3.46,
|
| 787 |
+
"grad_norm": 4.027952671051025,
|
| 788 |
+
"learning_rate": 6.54325259515571e-05,
|
| 789 |
+
"loss": 0.5301,
|
| 790 |
"step": 1000
|
| 791 |
},
|
| 792 |
{
|
| 793 |
"epoch": 3.46,
|
| 794 |
+
"eval_accuracy": 0.8503067484662576,
|
| 795 |
+
"eval_loss": 0.5100666284561157,
|
| 796 |
+
"eval_runtime": 11.1502,
|
| 797 |
+
"eval_samples_per_second": 73.093,
|
| 798 |
+
"eval_steps_per_second": 9.148,
|
| 799 |
"step": 1000
|
| 800 |
},
|
| 801 |
{
|
| 802 |
"epoch": 3.49,
|
| 803 |
+
"grad_norm": 7.976325988769531,
|
| 804 |
+
"learning_rate": 6.508650519031142e-05,
|
| 805 |
+
"loss": 0.3365,
|
| 806 |
"step": 1010
|
| 807 |
},
|
| 808 |
{
|
| 809 |
"epoch": 3.53,
|
| 810 |
+
"grad_norm": 5.674125671386719,
|
| 811 |
+
"learning_rate": 6.474048442906576e-05,
|
| 812 |
+
"loss": 0.3745,
|
| 813 |
"step": 1020
|
| 814 |
},
|
| 815 |
{
|
| 816 |
"epoch": 3.56,
|
| 817 |
+
"grad_norm": 2.311941623687744,
|
| 818 |
+
"learning_rate": 6.439446366782007e-05,
|
| 819 |
+
"loss": 0.4887,
|
| 820 |
"step": 1030
|
| 821 |
},
|
| 822 |
{
|
| 823 |
"epoch": 3.6,
|
| 824 |
+
"grad_norm": 2.7413113117218018,
|
| 825 |
+
"learning_rate": 6.40484429065744e-05,
|
| 826 |
+
"loss": 0.4548,
|
| 827 |
"step": 1040
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 3.63,
|
| 831 |
+
"grad_norm": 4.1170830726623535,
|
| 832 |
+
"learning_rate": 6.370242214532872e-05,
|
| 833 |
+
"loss": 0.5191,
|
| 834 |
"step": 1050
|
| 835 |
},
|
| 836 |
{
|
| 837 |
"epoch": 3.67,
|
| 838 |
+
"grad_norm": 3.560255765914917,
|
| 839 |
+
"learning_rate": 6.335640138408306e-05,
|
| 840 |
+
"loss": 0.496,
|
| 841 |
"step": 1060
|
| 842 |
},
|
| 843 |
{
|
| 844 |
"epoch": 3.7,
|
| 845 |
+
"grad_norm": 7.074652671813965,
|
| 846 |
+
"learning_rate": 6.301038062283737e-05,
|
| 847 |
+
"loss": 0.5216,
|
| 848 |
"step": 1070
|
| 849 |
},
|
| 850 |
{
|
| 851 |
"epoch": 3.74,
|
| 852 |
+
"grad_norm": 2.202094793319702,
|
| 853 |
+
"learning_rate": 6.26643598615917e-05,
|
| 854 |
+
"loss": 0.3693,
|
| 855 |
"step": 1080
|
| 856 |
},
|
| 857 |
{
|
| 858 |
"epoch": 3.77,
|
| 859 |
+
"grad_norm": 5.584536075592041,
|
| 860 |
+
"learning_rate": 6.231833910034602e-05,
|
| 861 |
+
"loss": 0.3557,
|
| 862 |
"step": 1090
|
| 863 |
},
|
| 864 |
{
|
| 865 |
"epoch": 3.81,
|
| 866 |
+
"grad_norm": 4.547154903411865,
|
| 867 |
+
"learning_rate": 6.197231833910034e-05,
|
| 868 |
+
"loss": 0.4108,
|
| 869 |
"step": 1100
|
| 870 |
},
|
| 871 |
{
|
| 872 |
"epoch": 3.81,
|
| 873 |
+
"eval_accuracy": 0.8478527607361963,
|
| 874 |
+
"eval_loss": 0.5211547613143921,
|
| 875 |
+
"eval_runtime": 11.576,
|
| 876 |
+
"eval_samples_per_second": 70.404,
|
| 877 |
+
"eval_steps_per_second": 8.811,
|
| 878 |
"step": 1100
|
| 879 |
},
|
| 880 |
{
|
| 881 |
"epoch": 3.84,
|
| 882 |
+
"grad_norm": 2.4568958282470703,
|
| 883 |
+
"learning_rate": 6.162629757785467e-05,
|
| 884 |
+
"loss": 0.4367,
|
| 885 |
"step": 1110
|
| 886 |
},
|
| 887 |
{
|
| 888 |
"epoch": 3.88,
|
| 889 |
+
"grad_norm": 3.632456064224243,
|
| 890 |
+
"learning_rate": 6.128027681660899e-05,
|
| 891 |
+
"loss": 0.4811,
|
| 892 |
"step": 1120
|
| 893 |
},
|
| 894 |
{
|
| 895 |
"epoch": 3.91,
|
| 896 |
+
"grad_norm": 5.1356329917907715,
|
| 897 |
+
"learning_rate": 6.0934256055363324e-05,
|
| 898 |
+
"loss": 0.3121,
|
| 899 |
"step": 1130
|
| 900 |
},
|
| 901 |
{
|
| 902 |
"epoch": 3.94,
|
| 903 |
+
"grad_norm": 2.870601177215576,
|
| 904 |
+
"learning_rate": 6.058823529411765e-05,
|
| 905 |
+
"loss": 0.42,
|
| 906 |
"step": 1140
|
| 907 |
},
|
| 908 |
{
|
| 909 |
"epoch": 3.98,
|
| 910 |
+
"grad_norm": 6.305639743804932,
|
| 911 |
+
"learning_rate": 6.0242214532871973e-05,
|
| 912 |
+
"loss": 0.4733,
|
| 913 |
"step": 1150
|
| 914 |
},
|
| 915 |
{
|
| 916 |
"epoch": 4.01,
|
| 917 |
+
"grad_norm": 5.95077657699585,
|
| 918 |
+
"learning_rate": 5.98961937716263e-05,
|
| 919 |
+
"loss": 0.3604,
|
| 920 |
"step": 1160
|
| 921 |
},
|
| 922 |
{
|
| 923 |
"epoch": 4.05,
|
| 924 |
+
"grad_norm": 4.88993501663208,
|
| 925 |
+
"learning_rate": 5.955017301038063e-05,
|
| 926 |
+
"loss": 0.3475,
|
| 927 |
"step": 1170
|
| 928 |
},
|
| 929 |
{
|
| 930 |
"epoch": 4.08,
|
| 931 |
+
"grad_norm": 2.9184603691101074,
|
| 932 |
+
"learning_rate": 5.920415224913495e-05,
|
| 933 |
+
"loss": 0.395,
|
| 934 |
"step": 1180
|
| 935 |
},
|
| 936 |
{
|
| 937 |
"epoch": 4.12,
|
| 938 |
+
"grad_norm": 3.779115915298462,
|
| 939 |
+
"learning_rate": 5.885813148788928e-05,
|
| 940 |
+
"loss": 0.4199,
|
| 941 |
"step": 1190
|
| 942 |
},
|
| 943 |
{
|
| 944 |
"epoch": 4.15,
|
| 945 |
+
"grad_norm": 4.5922322273254395,
|
| 946 |
+
"learning_rate": 5.854671280276816e-05,
|
| 947 |
+
"loss": 0.4223,
|
| 948 |
"step": 1200
|
| 949 |
},
|
| 950 |
{
|
| 951 |
"epoch": 4.15,
|
| 952 |
+
"eval_accuracy": 0.8429447852760736,
|
| 953 |
+
"eval_loss": 0.5328340530395508,
|
| 954 |
+
"eval_runtime": 11.5883,
|
| 955 |
+
"eval_samples_per_second": 70.33,
|
| 956 |
+
"eval_steps_per_second": 8.802,
|
| 957 |
"step": 1200
|
| 958 |
},
|
| 959 |
{
|
| 960 |
"epoch": 4.19,
|
| 961 |
+
"grad_norm": 5.457287311553955,
|
| 962 |
+
"learning_rate": 5.8200692041522495e-05,
|
| 963 |
+
"loss": 0.2775,
|
| 964 |
"step": 1210
|
| 965 |
},
|
| 966 |
{
|
| 967 |
"epoch": 4.22,
|
| 968 |
+
"grad_norm": 5.398082733154297,
|
| 969 |
+
"learning_rate": 5.785467128027682e-05,
|
| 970 |
+
"loss": 0.3019,
|
| 971 |
"step": 1220
|
| 972 |
},
|
| 973 |
{
|
| 974 |
"epoch": 4.26,
|
| 975 |
+
"grad_norm": 5.261845111846924,
|
| 976 |
+
"learning_rate": 5.750865051903115e-05,
|
| 977 |
+
"loss": 0.4365,
|
| 978 |
"step": 1230
|
| 979 |
},
|
| 980 |
{
|
| 981 |
"epoch": 4.29,
|
| 982 |
+
"grad_norm": 5.616456508636475,
|
| 983 |
+
"learning_rate": 5.716262975778547e-05,
|
| 984 |
+
"loss": 0.412,
|
| 985 |
"step": 1240
|
| 986 |
},
|
| 987 |
{
|
| 988 |
"epoch": 4.33,
|
| 989 |
+
"grad_norm": 3.715630292892456,
|
| 990 |
+
"learning_rate": 5.68166089965398e-05,
|
| 991 |
+
"loss": 0.3644,
|
| 992 |
"step": 1250
|
| 993 |
},
|
| 994 |
{
|
| 995 |
"epoch": 4.36,
|
| 996 |
+
"grad_norm": 5.539583683013916,
|
| 997 |
+
"learning_rate": 5.647058823529412e-05,
|
| 998 |
+
"loss": 0.2519,
|
| 999 |
"step": 1260
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
"epoch": 4.39,
|
| 1003 |
+
"grad_norm": 7.531044960021973,
|
| 1004 |
+
"learning_rate": 5.612456747404845e-05,
|
| 1005 |
+
"loss": 0.4028,
|
| 1006 |
"step": 1270
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
"epoch": 4.43,
|
| 1010 |
+
"grad_norm": 4.245101451873779,
|
| 1011 |
+
"learning_rate": 5.577854671280277e-05,
|
| 1012 |
+
"loss": 0.483,
|
| 1013 |
"step": 1280
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
"epoch": 4.46,
|
| 1017 |
+
"grad_norm": 4.674008846282959,
|
| 1018 |
+
"learning_rate": 5.54325259515571e-05,
|
| 1019 |
+
"loss": 0.4101,
|
| 1020 |
"step": 1290
|
| 1021 |
},
|
| 1022 |
{
|
| 1023 |
"epoch": 4.5,
|
| 1024 |
+
"grad_norm": 5.6747331619262695,
|
| 1025 |
+
"learning_rate": 5.508650519031142e-05,
|
| 1026 |
+
"loss": 0.3877,
|
| 1027 |
"step": 1300
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
"epoch": 4.5,
|
| 1031 |
+
"eval_accuracy": 0.8294478527607362,
|
| 1032 |
+
"eval_loss": 0.5815134048461914,
|
| 1033 |
+
"eval_runtime": 11.4501,
|
| 1034 |
+
"eval_samples_per_second": 71.178,
|
| 1035 |
+
"eval_steps_per_second": 8.908,
|
| 1036 |
"step": 1300
|
| 1037 |
},
|
| 1038 |
{
|
| 1039 |
"epoch": 4.53,
|
| 1040 |
+
"grad_norm": 4.670008659362793,
|
| 1041 |
+
"learning_rate": 5.4740484429065744e-05,
|
| 1042 |
+
"loss": 0.4169,
|
| 1043 |
"step": 1310
|
| 1044 |
},
|
| 1045 |
{
|
| 1046 |
"epoch": 4.57,
|
| 1047 |
+
"grad_norm": 6.947169780731201,
|
| 1048 |
+
"learning_rate": 5.4394463667820076e-05,
|
| 1049 |
+
"loss": 0.3955,
|
| 1050 |
"step": 1320
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
"epoch": 4.6,
|
| 1054 |
+
"grad_norm": 0.6032439470291138,
|
| 1055 |
+
"learning_rate": 5.4048442906574394e-05,
|
| 1056 |
+
"loss": 0.3438,
|
| 1057 |
"step": 1330
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
"epoch": 4.64,
|
| 1061 |
+
"grad_norm": 2.0391006469726562,
|
| 1062 |
+
"learning_rate": 5.3702422145328725e-05,
|
| 1063 |
+
"loss": 0.3032,
|
| 1064 |
"step": 1340
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
"epoch": 4.67,
|
| 1068 |
+
"grad_norm": 1.2689828872680664,
|
| 1069 |
+
"learning_rate": 5.3356401384083043e-05,
|
| 1070 |
+
"loss": 0.3091,
|
| 1071 |
"step": 1350
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
"epoch": 4.71,
|
| 1075 |
+
"grad_norm": 9.588872909545898,
|
| 1076 |
+
"learning_rate": 5.3010380622837375e-05,
|
| 1077 |
+
"loss": 0.3301,
|
| 1078 |
"step": 1360
|
| 1079 |
},
|
| 1080 |
{
|
| 1081 |
"epoch": 4.74,
|
| 1082 |
+
"grad_norm": 9.207768440246582,
|
| 1083 |
+
"learning_rate": 5.266435986159169e-05,
|
| 1084 |
+
"loss": 0.4184,
|
| 1085 |
"step": 1370
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
"epoch": 4.78,
|
| 1089 |
+
"grad_norm": 3.842557191848755,
|
| 1090 |
+
"learning_rate": 5.2318339100346025e-05,
|
| 1091 |
+
"loss": 0.3254,
|
| 1092 |
"step": 1380
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
"epoch": 4.81,
|
| 1096 |
+
"grad_norm": 5.358198642730713,
|
| 1097 |
+
"learning_rate": 5.197231833910034e-05,
|
| 1098 |
+
"loss": 0.2876,
|
| 1099 |
"step": 1390
|
| 1100 |
},
|
| 1101 |
{
|
| 1102 |
"epoch": 4.84,
|
| 1103 |
+
"grad_norm": 4.02155065536499,
|
| 1104 |
+
"learning_rate": 5.1626297577854675e-05,
|
| 1105 |
+
"loss": 0.3879,
|
| 1106 |
"step": 1400
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
"epoch": 4.84,
|
| 1110 |
+
"eval_accuracy": 0.8503067484662576,
|
| 1111 |
+
"eval_loss": 0.5150614380836487,
|
| 1112 |
+
"eval_runtime": 11.4927,
|
| 1113 |
+
"eval_samples_per_second": 70.914,
|
| 1114 |
+
"eval_steps_per_second": 8.875,
|
| 1115 |
"step": 1400
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
"epoch": 4.88,
|
| 1119 |
+
"grad_norm": 4.721061706542969,
|
| 1120 |
+
"learning_rate": 5.1280276816609e-05,
|
| 1121 |
+
"loss": 0.4051,
|
| 1122 |
"step": 1410
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
"epoch": 4.91,
|
| 1126 |
+
"grad_norm": 6.063509941101074,
|
| 1127 |
+
"learning_rate": 5.0934256055363325e-05,
|
| 1128 |
+
"loss": 0.436,
|
| 1129 |
"step": 1420
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
"epoch": 4.95,
|
| 1133 |
+
"grad_norm": 4.278838634490967,
|
| 1134 |
+
"learning_rate": 5.058823529411765e-05,
|
| 1135 |
+
"loss": 0.3533,
|
| 1136 |
"step": 1430
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
"epoch": 4.98,
|
| 1140 |
+
"grad_norm": 5.166705131530762,
|
| 1141 |
+
"learning_rate": 5.024221453287198e-05,
|
| 1142 |
+
"loss": 0.3281,
|
| 1143 |
"step": 1440
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
+
"epoch": 5.02,
|
| 1147 |
+
"grad_norm": 4.790309906005859,
|
| 1148 |
+
"learning_rate": 4.98961937716263e-05,
|
| 1149 |
+
"loss": 0.3324,
|
| 1150 |
+
"step": 1450
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 5.05,
|
| 1154 |
+
"grad_norm": 6.665502548217773,
|
| 1155 |
+
"learning_rate": 4.9550173010380624e-05,
|
| 1156 |
+
"loss": 0.4309,
|
| 1157 |
+
"step": 1460
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 5.09,
|
| 1161 |
+
"grad_norm": 2.508721113204956,
|
| 1162 |
+
"learning_rate": 4.920415224913495e-05,
|
| 1163 |
+
"loss": 0.4029,
|
| 1164 |
+
"step": 1470
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 5.12,
|
| 1168 |
+
"grad_norm": 3.7062671184539795,
|
| 1169 |
+
"learning_rate": 4.8858131487889274e-05,
|
| 1170 |
+
"loss": 0.3016,
|
| 1171 |
+
"step": 1480
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 5.16,
|
| 1175 |
+
"grad_norm": 1.6657217741012573,
|
| 1176 |
+
"learning_rate": 4.85121107266436e-05,
|
| 1177 |
+
"loss": 0.2906,
|
| 1178 |
+
"step": 1490
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 5.19,
|
| 1182 |
+
"grad_norm": 5.122851848602295,
|
| 1183 |
+
"learning_rate": 4.8166089965397924e-05,
|
| 1184 |
+
"loss": 0.2797,
|
| 1185 |
+
"step": 1500
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 5.19,
|
| 1189 |
+
"eval_accuracy": 0.8564417177914111,
|
| 1190 |
+
"eval_loss": 0.5160326361656189,
|
| 1191 |
+
"eval_runtime": 11.658,
|
| 1192 |
+
"eval_samples_per_second": 69.909,
|
| 1193 |
+
"eval_steps_per_second": 8.749,
|
| 1194 |
+
"step": 1500
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 5.22,
|
| 1198 |
+
"grad_norm": 1.0947901010513306,
|
| 1199 |
+
"learning_rate": 4.782006920415225e-05,
|
| 1200 |
+
"loss": 0.2272,
|
| 1201 |
+
"step": 1510
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 5.26,
|
| 1205 |
+
"grad_norm": 6.223177909851074,
|
| 1206 |
+
"learning_rate": 4.747404844290658e-05,
|
| 1207 |
+
"loss": 0.4336,
|
| 1208 |
+
"step": 1520
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 5.29,
|
| 1212 |
+
"grad_norm": 4.786419868469238,
|
| 1213 |
+
"learning_rate": 4.7128027681660905e-05,
|
| 1214 |
+
"loss": 0.3644,
|
| 1215 |
+
"step": 1530
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 5.33,
|
| 1219 |
+
"grad_norm": 2.7506916522979736,
|
| 1220 |
+
"learning_rate": 4.678200692041523e-05,
|
| 1221 |
+
"loss": 0.3686,
|
| 1222 |
+
"step": 1540
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 5.36,
|
| 1226 |
+
"grad_norm": 5.9919352531433105,
|
| 1227 |
+
"learning_rate": 4.6435986159169555e-05,
|
| 1228 |
+
"loss": 0.2422,
|
| 1229 |
+
"step": 1550
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 5.4,
|
| 1233 |
+
"grad_norm": 2.702258825302124,
|
| 1234 |
+
"learning_rate": 4.608996539792388e-05,
|
| 1235 |
+
"loss": 0.4158,
|
| 1236 |
+
"step": 1560
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 5.43,
|
| 1240 |
+
"grad_norm": 0.17089949548244476,
|
| 1241 |
+
"learning_rate": 4.5743944636678205e-05,
|
| 1242 |
+
"loss": 0.2326,
|
| 1243 |
+
"step": 1570
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 5.47,
|
| 1247 |
+
"grad_norm": 1.4725955724716187,
|
| 1248 |
+
"learning_rate": 4.539792387543253e-05,
|
| 1249 |
+
"loss": 0.2891,
|
| 1250 |
+
"step": 1580
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 5.5,
|
| 1254 |
+
"grad_norm": 3.754875659942627,
|
| 1255 |
+
"learning_rate": 4.5051903114186854e-05,
|
| 1256 |
+
"loss": 0.3169,
|
| 1257 |
+
"step": 1590
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 5.54,
|
| 1261 |
+
"grad_norm": 1.3562734127044678,
|
| 1262 |
+
"learning_rate": 4.470588235294118e-05,
|
| 1263 |
+
"loss": 0.2628,
|
| 1264 |
+
"step": 1600
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 5.54,
|
| 1268 |
+
"eval_accuracy": 0.8699386503067484,
|
| 1269 |
+
"eval_loss": 0.46183693408966064,
|
| 1270 |
+
"eval_runtime": 12.3863,
|
| 1271 |
+
"eval_samples_per_second": 65.799,
|
| 1272 |
+
"eval_steps_per_second": 8.235,
|
| 1273 |
+
"step": 1600
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 5.57,
|
| 1277 |
+
"grad_norm": 1.3551547527313232,
|
| 1278 |
+
"learning_rate": 4.4359861591695504e-05,
|
| 1279 |
+
"loss": 0.2924,
|
| 1280 |
+
"step": 1610
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 5.61,
|
| 1284 |
+
"grad_norm": 2.6148874759674072,
|
| 1285 |
+
"learning_rate": 4.401384083044983e-05,
|
| 1286 |
+
"loss": 0.2792,
|
| 1287 |
+
"step": 1620
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 5.64,
|
| 1291 |
+
"grad_norm": 3.0203983783721924,
|
| 1292 |
+
"learning_rate": 4.3667820069204154e-05,
|
| 1293 |
+
"loss": 0.2955,
|
| 1294 |
+
"step": 1630
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 5.67,
|
| 1298 |
+
"grad_norm": 10.279276847839355,
|
| 1299 |
+
"learning_rate": 4.332179930795848e-05,
|
| 1300 |
+
"loss": 0.4448,
|
| 1301 |
+
"step": 1640
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 5.71,
|
| 1305 |
+
"grad_norm": 0.37972888350486755,
|
| 1306 |
+
"learning_rate": 4.2975778546712804e-05,
|
| 1307 |
+
"loss": 0.2687,
|
| 1308 |
+
"step": 1650
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 5.74,
|
| 1312 |
+
"grad_norm": 5.29146671295166,
|
| 1313 |
+
"learning_rate": 4.262975778546713e-05,
|
| 1314 |
+
"loss": 0.2522,
|
| 1315 |
+
"step": 1660
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 5.78,
|
| 1319 |
+
"grad_norm": 1.9747039079666138,
|
| 1320 |
+
"learning_rate": 4.2283737024221454e-05,
|
| 1321 |
+
"loss": 0.2579,
|
| 1322 |
+
"step": 1670
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 5.81,
|
| 1326 |
+
"grad_norm": 1.1432113647460938,
|
| 1327 |
+
"learning_rate": 4.193771626297578e-05,
|
| 1328 |
+
"loss": 0.2522,
|
| 1329 |
+
"step": 1680
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 5.85,
|
| 1333 |
+
"grad_norm": 0.5862625241279602,
|
| 1334 |
+
"learning_rate": 4.15916955017301e-05,
|
| 1335 |
+
"loss": 0.265,
|
| 1336 |
+
"step": 1690
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 5.88,
|
| 1340 |
+
"grad_norm": 5.743435859680176,
|
| 1341 |
+
"learning_rate": 4.124567474048443e-05,
|
| 1342 |
+
"loss": 0.3404,
|
| 1343 |
+
"step": 1700
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 5.88,
|
| 1347 |
+
"eval_accuracy": 0.8674846625766871,
|
| 1348 |
+
"eval_loss": 0.4903254508972168,
|
| 1349 |
+
"eval_runtime": 11.4951,
|
| 1350 |
+
"eval_samples_per_second": 70.9,
|
| 1351 |
+
"eval_steps_per_second": 8.873,
|
| 1352 |
+
"step": 1700
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 5.92,
|
| 1356 |
+
"grad_norm": 2.558229446411133,
|
| 1357 |
+
"learning_rate": 4.089965397923875e-05,
|
| 1358 |
+
"loss": 0.3082,
|
| 1359 |
+
"step": 1710
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 5.95,
|
| 1363 |
+
"grad_norm": 1.225205659866333,
|
| 1364 |
+
"learning_rate": 4.055363321799308e-05,
|
| 1365 |
+
"loss": 0.2875,
|
| 1366 |
+
"step": 1720
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 5.99,
|
| 1370 |
+
"grad_norm": 0.44406113028526306,
|
| 1371 |
+
"learning_rate": 4.020761245674741e-05,
|
| 1372 |
+
"loss": 0.2987,
|
| 1373 |
+
"step": 1730
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 6.02,
|
| 1377 |
+
"grad_norm": 6.770223617553711,
|
| 1378 |
+
"learning_rate": 3.9861591695501735e-05,
|
| 1379 |
+
"loss": 0.381,
|
| 1380 |
+
"step": 1740
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 6.06,
|
| 1384 |
+
"grad_norm": 5.378862380981445,
|
| 1385 |
+
"learning_rate": 3.951557093425606e-05,
|
| 1386 |
+
"loss": 0.3109,
|
| 1387 |
+
"step": 1750
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 6.09,
|
| 1391 |
+
"grad_norm": 5.086582183837891,
|
| 1392 |
+
"learning_rate": 3.9169550173010384e-05,
|
| 1393 |
+
"loss": 0.3031,
|
| 1394 |
+
"step": 1760
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 6.12,
|
| 1398 |
+
"grad_norm": 3.3848397731781006,
|
| 1399 |
+
"learning_rate": 3.882352941176471e-05,
|
| 1400 |
+
"loss": 0.3092,
|
| 1401 |
+
"step": 1770
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 6.16,
|
| 1405 |
+
"grad_norm": 1.4123233556747437,
|
| 1406 |
+
"learning_rate": 3.8477508650519034e-05,
|
| 1407 |
+
"loss": 0.352,
|
| 1408 |
+
"step": 1780
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 6.19,
|
| 1412 |
+
"grad_norm": 1.3674020767211914,
|
| 1413 |
+
"learning_rate": 3.813148788927336e-05,
|
| 1414 |
+
"loss": 0.1128,
|
| 1415 |
+
"step": 1790
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 6.23,
|
| 1419 |
+
"grad_norm": 3.944838047027588,
|
| 1420 |
+
"learning_rate": 3.7785467128027684e-05,
|
| 1421 |
+
"loss": 0.3033,
|
| 1422 |
+
"step": 1800
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 6.23,
|
| 1426 |
+
"eval_accuracy": 0.8662576687116564,
|
| 1427 |
+
"eval_loss": 0.4860667288303375,
|
| 1428 |
+
"eval_runtime": 11.4194,
|
| 1429 |
+
"eval_samples_per_second": 71.37,
|
| 1430 |
+
"eval_steps_per_second": 8.932,
|
| 1431 |
+
"step": 1800
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 6.26,
|
| 1435 |
+
"grad_norm": 4.455410957336426,
|
| 1436 |
+
"learning_rate": 3.743944636678201e-05,
|
| 1437 |
+
"loss": 0.2315,
|
| 1438 |
+
"step": 1810
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 6.3,
|
| 1442 |
+
"grad_norm": 3.5710418224334717,
|
| 1443 |
+
"learning_rate": 3.7093425605536334e-05,
|
| 1444 |
+
"loss": 0.2175,
|
| 1445 |
+
"step": 1820
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 6.33,
|
| 1449 |
+
"grad_norm": 6.50364875793457,
|
| 1450 |
+
"learning_rate": 3.674740484429066e-05,
|
| 1451 |
+
"loss": 0.2477,
|
| 1452 |
+
"step": 1830
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 6.37,
|
| 1456 |
+
"grad_norm": 2.710531711578369,
|
| 1457 |
+
"learning_rate": 3.640138408304499e-05,
|
| 1458 |
+
"loss": 0.221,
|
| 1459 |
+
"step": 1840
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 6.4,
|
| 1463 |
+
"grad_norm": 2.1464736461639404,
|
| 1464 |
+
"learning_rate": 3.605536332179931e-05,
|
| 1465 |
+
"loss": 0.2387,
|
| 1466 |
+
"step": 1850
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 6.44,
|
| 1470 |
+
"grad_norm": 7.897326469421387,
|
| 1471 |
+
"learning_rate": 3.570934256055363e-05,
|
| 1472 |
+
"loss": 0.2723,
|
| 1473 |
+
"step": 1860
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 6.47,
|
| 1477 |
+
"grad_norm": 2.9161367416381836,
|
| 1478 |
+
"learning_rate": 3.536332179930796e-05,
|
| 1479 |
+
"loss": 0.2621,
|
| 1480 |
+
"step": 1870
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 6.51,
|
| 1484 |
+
"grad_norm": 7.048449516296387,
|
| 1485 |
+
"learning_rate": 3.501730103806228e-05,
|
| 1486 |
+
"loss": 0.2955,
|
| 1487 |
+
"step": 1880
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 6.54,
|
| 1491 |
+
"grad_norm": 5.2737250328063965,
|
| 1492 |
+
"learning_rate": 3.467128027681661e-05,
|
| 1493 |
+
"loss": 0.3076,
|
| 1494 |
+
"step": 1890
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 6.57,
|
| 1498 |
+
"grad_norm": 4.985352039337158,
|
| 1499 |
+
"learning_rate": 3.432525951557093e-05,
|
| 1500 |
+
"loss": 0.214,
|
| 1501 |
+
"step": 1900
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 6.57,
|
| 1505 |
+
"eval_accuracy": 0.8687116564417178,
|
| 1506 |
+
"eval_loss": 0.4853067100048065,
|
| 1507 |
+
"eval_runtime": 11.47,
|
| 1508 |
+
"eval_samples_per_second": 71.055,
|
| 1509 |
+
"eval_steps_per_second": 8.893,
|
| 1510 |
+
"step": 1900
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 6.61,
|
| 1514 |
+
"grad_norm": 6.846111297607422,
|
| 1515 |
+
"learning_rate": 3.397923875432526e-05,
|
| 1516 |
+
"loss": 0.2693,
|
| 1517 |
+
"step": 1910
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 6.64,
|
| 1521 |
+
"grad_norm": 2.8362717628479004,
|
| 1522 |
+
"learning_rate": 3.363321799307958e-05,
|
| 1523 |
+
"loss": 0.2582,
|
| 1524 |
+
"step": 1920
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 6.68,
|
| 1528 |
+
"grad_norm": 1.0130655765533447,
|
| 1529 |
+
"learning_rate": 3.3287197231833914e-05,
|
| 1530 |
+
"loss": 0.278,
|
| 1531 |
+
"step": 1930
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 6.71,
|
| 1535 |
+
"grad_norm": 1.3184999227523804,
|
| 1536 |
+
"learning_rate": 3.294117647058824e-05,
|
| 1537 |
+
"loss": 0.3776,
|
| 1538 |
+
"step": 1940
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 6.75,
|
| 1542 |
+
"grad_norm": 3.637080430984497,
|
| 1543 |
+
"learning_rate": 3.2595155709342564e-05,
|
| 1544 |
+
"loss": 0.1598,
|
| 1545 |
+
"step": 1950
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 6.78,
|
| 1549 |
+
"grad_norm": 4.093498706817627,
|
| 1550 |
+
"learning_rate": 3.224913494809689e-05,
|
| 1551 |
+
"loss": 0.2604,
|
| 1552 |
+
"step": 1960
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 6.82,
|
| 1556 |
+
"grad_norm": 6.110915184020996,
|
| 1557 |
+
"learning_rate": 3.1903114186851214e-05,
|
| 1558 |
+
"loss": 0.3471,
|
| 1559 |
+
"step": 1970
|
| 1560 |
+
},
|
| 1561 |
+
{
|
| 1562 |
+
"epoch": 6.85,
|
| 1563 |
+
"grad_norm": 6.045133590698242,
|
| 1564 |
+
"learning_rate": 3.155709342560554e-05,
|
| 1565 |
+
"loss": 0.1886,
|
| 1566 |
+
"step": 1980
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 6.89,
|
| 1570 |
+
"grad_norm": 0.4137258231639862,
|
| 1571 |
+
"learning_rate": 3.1211072664359864e-05,
|
| 1572 |
+
"loss": 0.183,
|
| 1573 |
+
"step": 1990
|
| 1574 |
+
},
|
| 1575 |
+
{
|
| 1576 |
+
"epoch": 6.92,
|
| 1577 |
+
"grad_norm": 3.405696392059326,
|
| 1578 |
+
"learning_rate": 3.086505190311419e-05,
|
| 1579 |
+
"loss": 0.2763,
|
| 1580 |
+
"step": 2000
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 6.92,
|
| 1584 |
+
"eval_accuracy": 0.8736196319018404,
|
| 1585 |
+
"eval_loss": 0.4705219566822052,
|
| 1586 |
+
"eval_runtime": 11.4342,
|
| 1587 |
+
"eval_samples_per_second": 71.277,
|
| 1588 |
+
"eval_steps_per_second": 8.921,
|
| 1589 |
+
"step": 2000
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 6.96,
|
| 1593 |
+
"grad_norm": 6.817298889160156,
|
| 1594 |
+
"learning_rate": 3.0519031141868513e-05,
|
| 1595 |
+
"loss": 0.2825,
|
| 1596 |
+
"step": 2010
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 6.99,
|
| 1600 |
+
"grad_norm": 5.460371494293213,
|
| 1601 |
+
"learning_rate": 3.0173010380622842e-05,
|
| 1602 |
+
"loss": 0.2853,
|
| 1603 |
+
"step": 2020
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 7.02,
|
| 1607 |
+
"grad_norm": 6.517573833465576,
|
| 1608 |
+
"learning_rate": 2.9826989619377167e-05,
|
| 1609 |
+
"loss": 0.2913,
|
| 1610 |
+
"step": 2030
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 7.06,
|
| 1614 |
+
"grad_norm": 2.988943099975586,
|
| 1615 |
+
"learning_rate": 2.948096885813149e-05,
|
| 1616 |
+
"loss": 0.3647,
|
| 1617 |
+
"step": 2040
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 7.09,
|
| 1621 |
+
"grad_norm": 3.221747636795044,
|
| 1622 |
+
"learning_rate": 2.9134948096885816e-05,
|
| 1623 |
+
"loss": 0.2344,
|
| 1624 |
+
"step": 2050
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 7.13,
|
| 1628 |
+
"grad_norm": 3.342573404312134,
|
| 1629 |
+
"learning_rate": 2.878892733564014e-05,
|
| 1630 |
+
"loss": 0.2741,
|
| 1631 |
+
"step": 2060
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 7.16,
|
| 1635 |
+
"grad_norm": 4.326902866363525,
|
| 1636 |
+
"learning_rate": 2.8442906574394463e-05,
|
| 1637 |
+
"loss": 0.2503,
|
| 1638 |
+
"step": 2070
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 7.2,
|
| 1642 |
+
"grad_norm": 7.965173244476318,
|
| 1643 |
+
"learning_rate": 2.8096885813148788e-05,
|
| 1644 |
+
"loss": 0.2061,
|
| 1645 |
+
"step": 2080
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 7.23,
|
| 1649 |
+
"grad_norm": 0.314100056886673,
|
| 1650 |
+
"learning_rate": 2.7750865051903113e-05,
|
| 1651 |
+
"loss": 0.2205,
|
| 1652 |
+
"step": 2090
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 7.27,
|
| 1656 |
+
"grad_norm": 4.415227890014648,
|
| 1657 |
+
"learning_rate": 2.7404844290657437e-05,
|
| 1658 |
+
"loss": 0.3009,
|
| 1659 |
+
"step": 2100
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 7.27,
|
| 1663 |
+
"eval_accuracy": 0.8625766871165644,
|
| 1664 |
+
"eval_loss": 0.47227197885513306,
|
| 1665 |
+
"eval_runtime": 11.4055,
|
| 1666 |
+
"eval_samples_per_second": 71.457,
|
| 1667 |
+
"eval_steps_per_second": 8.943,
|
| 1668 |
+
"step": 2100
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 7.3,
|
| 1672 |
+
"grad_norm": 4.705826759338379,
|
| 1673 |
+
"learning_rate": 2.7058823529411766e-05,
|
| 1674 |
+
"loss": 0.2255,
|
| 1675 |
+
"step": 2110
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 7.34,
|
| 1679 |
+
"grad_norm": 1.2122201919555664,
|
| 1680 |
+
"learning_rate": 2.671280276816609e-05,
|
| 1681 |
+
"loss": 0.2103,
|
| 1682 |
+
"step": 2120
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 7.37,
|
| 1686 |
+
"grad_norm": 11.807892799377441,
|
| 1687 |
+
"learning_rate": 2.6366782006920416e-05,
|
| 1688 |
+
"loss": 0.2001,
|
| 1689 |
+
"step": 2130
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 7.4,
|
| 1693 |
+
"grad_norm": 5.092467784881592,
|
| 1694 |
+
"learning_rate": 2.602076124567474e-05,
|
| 1695 |
+
"loss": 0.2208,
|
| 1696 |
+
"step": 2140
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 7.44,
|
| 1700 |
+
"grad_norm": 2.7508227825164795,
|
| 1701 |
+
"learning_rate": 2.5674740484429065e-05,
|
| 1702 |
+
"loss": 0.2397,
|
| 1703 |
+
"step": 2150
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 7.47,
|
| 1707 |
+
"grad_norm": 5.095257759094238,
|
| 1708 |
+
"learning_rate": 2.532871972318339e-05,
|
| 1709 |
+
"loss": 0.1861,
|
| 1710 |
+
"step": 2160
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 7.51,
|
| 1714 |
+
"grad_norm": 2.8208508491516113,
|
| 1715 |
+
"learning_rate": 2.498269896193772e-05,
|
| 1716 |
+
"loss": 0.1776,
|
| 1717 |
+
"step": 2170
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 7.54,
|
| 1721 |
+
"grad_norm": 16.025461196899414,
|
| 1722 |
+
"learning_rate": 2.4636678200692043e-05,
|
| 1723 |
+
"loss": 0.2899,
|
| 1724 |
+
"step": 2180
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 7.58,
|
| 1728 |
+
"grad_norm": 8.661275863647461,
|
| 1729 |
+
"learning_rate": 2.4290657439446368e-05,
|
| 1730 |
+
"loss": 0.3018,
|
| 1731 |
+
"step": 2190
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 7.61,
|
| 1735 |
+
"grad_norm": 4.188193321228027,
|
| 1736 |
+
"learning_rate": 2.3944636678200693e-05,
|
| 1737 |
+
"loss": 0.1543,
|
| 1738 |
+
"step": 2200
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 7.61,
|
| 1742 |
+
"eval_accuracy": 0.8638036809815951,
|
| 1743 |
+
"eval_loss": 0.4983120858669281,
|
| 1744 |
+
"eval_runtime": 11.4805,
|
| 1745 |
+
"eval_samples_per_second": 70.99,
|
| 1746 |
+
"eval_steps_per_second": 8.885,
|
| 1747 |
+
"step": 2200
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 7.65,
|
| 1751 |
+
"grad_norm": 2.8907430171966553,
|
| 1752 |
+
"learning_rate": 2.3598615916955018e-05,
|
| 1753 |
+
"loss": 0.2289,
|
| 1754 |
+
"step": 2210
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 7.68,
|
| 1758 |
+
"grad_norm": 6.132763385772705,
|
| 1759 |
+
"learning_rate": 2.3252595155709346e-05,
|
| 1760 |
+
"loss": 0.1687,
|
| 1761 |
+
"step": 2220
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 7.72,
|
| 1765 |
+
"grad_norm": 8.132895469665527,
|
| 1766 |
+
"learning_rate": 2.290657439446367e-05,
|
| 1767 |
+
"loss": 0.1936,
|
| 1768 |
+
"step": 2230
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"epoch": 7.75,
|
| 1772 |
+
"grad_norm": 2.7748658657073975,
|
| 1773 |
+
"learning_rate": 2.2560553633217993e-05,
|
| 1774 |
+
"loss": 0.2018,
|
| 1775 |
+
"step": 2240
|
| 1776 |
+
},
|
| 1777 |
+
{
|
| 1778 |
+
"epoch": 7.79,
|
| 1779 |
+
"grad_norm": 3.723444938659668,
|
| 1780 |
+
"learning_rate": 2.2214532871972318e-05,
|
| 1781 |
+
"loss": 0.331,
|
| 1782 |
+
"step": 2250
|
| 1783 |
+
},
|
| 1784 |
+
{
|
| 1785 |
+
"epoch": 7.82,
|
| 1786 |
+
"grad_norm": 0.9698171615600586,
|
| 1787 |
+
"learning_rate": 2.1868512110726642e-05,
|
| 1788 |
+
"loss": 0.2071,
|
| 1789 |
+
"step": 2260
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 7.85,
|
| 1793 |
+
"grad_norm": 5.589043140411377,
|
| 1794 |
+
"learning_rate": 2.152249134948097e-05,
|
| 1795 |
+
"loss": 0.1308,
|
| 1796 |
+
"step": 2270
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 7.89,
|
| 1800 |
+
"grad_norm": 1.0597747564315796,
|
| 1801 |
+
"learning_rate": 2.1176470588235296e-05,
|
| 1802 |
+
"loss": 0.2911,
|
| 1803 |
+
"step": 2280
|
| 1804 |
+
},
|
| 1805 |
+
{
|
| 1806 |
+
"epoch": 7.92,
|
| 1807 |
+
"grad_norm": 0.17304980754852295,
|
| 1808 |
+
"learning_rate": 2.083044982698962e-05,
|
| 1809 |
+
"loss": 0.3127,
|
| 1810 |
+
"step": 2290
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 7.96,
|
| 1814 |
+
"grad_norm": 2.3576455116271973,
|
| 1815 |
+
"learning_rate": 2.0484429065743945e-05,
|
| 1816 |
+
"loss": 0.2407,
|
| 1817 |
+
"step": 2300
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 7.96,
|
| 1821 |
+
"eval_accuracy": 0.8650306748466258,
|
| 1822 |
+
"eval_loss": 0.47417759895324707,
|
| 1823 |
+
"eval_runtime": 11.7355,
|
| 1824 |
+
"eval_samples_per_second": 69.447,
|
| 1825 |
+
"eval_steps_per_second": 8.692,
|
| 1826 |
+
"step": 2300
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 7.99,
|
| 1830 |
+
"grad_norm": 6.587134838104248,
|
| 1831 |
+
"learning_rate": 2.013840830449827e-05,
|
| 1832 |
+
"loss": 0.1954,
|
| 1833 |
+
"step": 2310
|
| 1834 |
+
},
|
| 1835 |
+
{
|
| 1836 |
+
"epoch": 8.03,
|
| 1837 |
+
"grad_norm": 5.971607685089111,
|
| 1838 |
+
"learning_rate": 1.9792387543252595e-05,
|
| 1839 |
+
"loss": 0.139,
|
| 1840 |
+
"step": 2320
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 8.06,
|
| 1844 |
+
"grad_norm": 2.352544069290161,
|
| 1845 |
+
"learning_rate": 1.9446366782006923e-05,
|
| 1846 |
+
"loss": 0.1648,
|
| 1847 |
+
"step": 2330
|
| 1848 |
+
},
|
| 1849 |
+
{
|
| 1850 |
+
"epoch": 8.1,
|
| 1851 |
+
"grad_norm": 1.0423003435134888,
|
| 1852 |
+
"learning_rate": 1.910034602076125e-05,
|
| 1853 |
+
"loss": 0.3657,
|
| 1854 |
+
"step": 2340
|
| 1855 |
+
},
|
| 1856 |
+
{
|
| 1857 |
+
"epoch": 8.13,
|
| 1858 |
+
"grad_norm": 3.612190008163452,
|
| 1859 |
+
"learning_rate": 1.875432525951557e-05,
|
| 1860 |
+
"loss": 0.1846,
|
| 1861 |
+
"step": 2350
|
| 1862 |
+
},
|
| 1863 |
+
{
|
| 1864 |
+
"epoch": 8.17,
|
| 1865 |
+
"grad_norm": 3.1284472942352295,
|
| 1866 |
+
"learning_rate": 1.8408304498269895e-05,
|
| 1867 |
+
"loss": 0.2876,
|
| 1868 |
+
"step": 2360
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 8.2,
|
| 1872 |
+
"grad_norm": 6.02339506149292,
|
| 1873 |
+
"learning_rate": 1.8062283737024223e-05,
|
| 1874 |
+
"loss": 0.3388,
|
| 1875 |
+
"step": 2370
|
| 1876 |
+
},
|
| 1877 |
+
{
|
| 1878 |
+
"epoch": 8.24,
|
| 1879 |
+
"grad_norm": 7.388739109039307,
|
| 1880 |
+
"learning_rate": 1.7716262975778548e-05,
|
| 1881 |
+
"loss": 0.3095,
|
| 1882 |
+
"step": 2380
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"epoch": 8.27,
|
| 1886 |
+
"grad_norm": 6.4854912757873535,
|
| 1887 |
+
"learning_rate": 1.7370242214532873e-05,
|
| 1888 |
+
"loss": 0.1786,
|
| 1889 |
+
"step": 2390
|
| 1890 |
+
},
|
| 1891 |
+
{
|
| 1892 |
+
"epoch": 8.3,
|
| 1893 |
+
"grad_norm": 3.4759864807128906,
|
| 1894 |
+
"learning_rate": 1.7024221453287198e-05,
|
| 1895 |
+
"loss": 0.2679,
|
| 1896 |
+
"step": 2400
|
| 1897 |
+
},
|
| 1898 |
+
{
|
| 1899 |
+
"epoch": 8.3,
|
| 1900 |
+
"eval_accuracy": 0.8723926380368098,
|
| 1901 |
+
"eval_loss": 0.4934893846511841,
|
| 1902 |
+
"eval_runtime": 11.4149,
|
| 1903 |
+
"eval_samples_per_second": 71.398,
|
| 1904 |
+
"eval_steps_per_second": 8.936,
|
| 1905 |
+
"step": 2400
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 8.34,
|
| 1909 |
+
"grad_norm": 0.9941917657852173,
|
| 1910 |
+
"learning_rate": 1.6678200692041523e-05,
|
| 1911 |
+
"loss": 0.1666,
|
| 1912 |
+
"step": 2410
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 8.37,
|
| 1916 |
+
"grad_norm": 2.615389108657837,
|
| 1917 |
+
"learning_rate": 1.6332179930795848e-05,
|
| 1918 |
+
"loss": 0.2039,
|
| 1919 |
+
"step": 2420
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 8.41,
|
| 1923 |
+
"grad_norm": 0.12697245180606842,
|
| 1924 |
+
"learning_rate": 1.5986159169550176e-05,
|
| 1925 |
+
"loss": 0.2026,
|
| 1926 |
+
"step": 2430
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 8.44,
|
| 1930 |
+
"grad_norm": 2.1617543697357178,
|
| 1931 |
+
"learning_rate": 1.56401384083045e-05,
|
| 1932 |
+
"loss": 0.2141,
|
| 1933 |
+
"step": 2440
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 8.48,
|
| 1937 |
+
"grad_norm": 6.248876094818115,
|
| 1938 |
+
"learning_rate": 1.5294117647058826e-05,
|
| 1939 |
+
"loss": 0.1039,
|
| 1940 |
+
"step": 2450
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 8.51,
|
| 1944 |
+
"grad_norm": 3.2735066413879395,
|
| 1945 |
+
"learning_rate": 1.4948096885813149e-05,
|
| 1946 |
+
"loss": 0.269,
|
| 1947 |
+
"step": 2460
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 8.55,
|
| 1951 |
+
"grad_norm": 0.2865382730960846,
|
| 1952 |
+
"learning_rate": 1.4602076124567474e-05,
|
| 1953 |
+
"loss": 0.1724,
|
| 1954 |
+
"step": 2470
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 8.58,
|
| 1958 |
+
"grad_norm": 0.17031227052211761,
|
| 1959 |
+
"learning_rate": 1.4256055363321799e-05,
|
| 1960 |
+
"loss": 0.1477,
|
| 1961 |
+
"step": 2480
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 8.62,
|
| 1965 |
+
"grad_norm": 5.686491966247559,
|
| 1966 |
+
"learning_rate": 1.3910034602076125e-05,
|
| 1967 |
+
"loss": 0.2774,
|
| 1968 |
+
"step": 2490
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 8.65,
|
| 1972 |
+
"grad_norm": 0.2951521873474121,
|
| 1973 |
+
"learning_rate": 1.356401384083045e-05,
|
| 1974 |
+
"loss": 0.1508,
|
| 1975 |
+
"step": 2500
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 8.65,
|
| 1979 |
+
"eval_accuracy": 0.8674846625766871,
|
| 1980 |
+
"eval_loss": 0.4825909435749054,
|
| 1981 |
+
"eval_runtime": 11.5579,
|
| 1982 |
+
"eval_samples_per_second": 70.515,
|
| 1983 |
+
"eval_steps_per_second": 8.825,
|
| 1984 |
+
"step": 2500
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 8.69,
|
| 1988 |
+
"grad_norm": 4.706648826599121,
|
| 1989 |
+
"learning_rate": 1.3217993079584775e-05,
|
| 1990 |
+
"loss": 0.1022,
|
| 1991 |
+
"step": 2510
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 8.72,
|
| 1995 |
+
"grad_norm": 7.546770095825195,
|
| 1996 |
+
"learning_rate": 1.2871972318339102e-05,
|
| 1997 |
+
"loss": 0.1778,
|
| 1998 |
+
"step": 2520
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 8.75,
|
| 2002 |
+
"grad_norm": 8.35574722290039,
|
| 2003 |
+
"learning_rate": 1.2525951557093426e-05,
|
| 2004 |
+
"loss": 0.242,
|
| 2005 |
+
"step": 2530
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 8.79,
|
| 2009 |
+
"grad_norm": 2.115328788757324,
|
| 2010 |
+
"learning_rate": 1.2179930795847751e-05,
|
| 2011 |
+
"loss": 0.1634,
|
| 2012 |
+
"step": 2540
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 8.82,
|
| 2016 |
+
"grad_norm": 7.35970401763916,
|
| 2017 |
+
"learning_rate": 1.1833910034602076e-05,
|
| 2018 |
+
"loss": 0.1618,
|
| 2019 |
+
"step": 2550
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 8.86,
|
| 2023 |
+
"grad_norm": 0.3500753939151764,
|
| 2024 |
+
"learning_rate": 1.1487889273356401e-05,
|
| 2025 |
+
"loss": 0.2685,
|
| 2026 |
+
"step": 2560
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 8.89,
|
| 2030 |
+
"grad_norm": 4.974752426147461,
|
| 2031 |
+
"learning_rate": 1.1141868512110728e-05,
|
| 2032 |
+
"loss": 0.1975,
|
| 2033 |
+
"step": 2570
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 8.93,
|
| 2037 |
+
"grad_norm": 2.2218804359436035,
|
| 2038 |
+
"learning_rate": 1.0795847750865053e-05,
|
| 2039 |
+
"loss": 0.1887,
|
| 2040 |
+
"step": 2580
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 8.96,
|
| 2044 |
+
"grad_norm": 2.4438514709472656,
|
| 2045 |
+
"learning_rate": 1.0449826989619377e-05,
|
| 2046 |
+
"loss": 0.2348,
|
| 2047 |
+
"step": 2590
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 9.0,
|
| 2051 |
+
"grad_norm": 7.3225998878479,
|
| 2052 |
+
"learning_rate": 1.0103806228373702e-05,
|
| 2053 |
+
"loss": 0.2129,
|
| 2054 |
+
"step": 2600
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 9.0,
|
| 2058 |
+
"eval_accuracy": 0.8711656441717791,
|
| 2059 |
+
"eval_loss": 0.49807000160217285,
|
| 2060 |
+
"eval_runtime": 11.7207,
|
| 2061 |
+
"eval_samples_per_second": 69.535,
|
| 2062 |
+
"eval_steps_per_second": 8.703,
|
| 2063 |
+
"step": 2600
|
| 2064 |
+
},
|
| 2065 |
+
{
|
| 2066 |
+
"epoch": 9.03,
|
| 2067 |
+
"grad_norm": 3.7790167331695557,
|
| 2068 |
+
"learning_rate": 9.757785467128027e-06,
|
| 2069 |
+
"loss": 0.1668,
|
| 2070 |
+
"step": 2610
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"epoch": 9.07,
|
| 2074 |
+
"grad_norm": 0.14809288084506989,
|
| 2075 |
+
"learning_rate": 9.411764705882354e-06,
|
| 2076 |
+
"loss": 0.1686,
|
| 2077 |
+
"step": 2620
|
| 2078 |
+
},
|
| 2079 |
+
{
|
| 2080 |
+
"epoch": 9.1,
|
| 2081 |
+
"grad_norm": 0.21380960941314697,
|
| 2082 |
+
"learning_rate": 9.065743944636679e-06,
|
| 2083 |
+
"loss": 0.2266,
|
| 2084 |
+
"step": 2630
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 9.13,
|
| 2088 |
+
"grad_norm": 1.157197117805481,
|
| 2089 |
+
"learning_rate": 8.719723183391004e-06,
|
| 2090 |
+
"loss": 0.1608,
|
| 2091 |
+
"step": 2640
|
| 2092 |
+
},
|
| 2093 |
+
{
|
| 2094 |
+
"epoch": 9.17,
|
| 2095 |
+
"grad_norm": 1.3275336027145386,
|
| 2096 |
+
"learning_rate": 8.37370242214533e-06,
|
| 2097 |
+
"loss": 0.1909,
|
| 2098 |
+
"step": 2650
|
| 2099 |
+
},
|
| 2100 |
+
{
|
| 2101 |
+
"epoch": 9.2,
|
| 2102 |
+
"grad_norm": 0.963979959487915,
|
| 2103 |
+
"learning_rate": 8.027681660899653e-06,
|
| 2104 |
+
"loss": 0.1803,
|
| 2105 |
+
"step": 2660
|
| 2106 |
+
},
|
| 2107 |
+
{
|
| 2108 |
+
"epoch": 9.24,
|
| 2109 |
+
"grad_norm": NaN,
|
| 2110 |
+
"learning_rate": 7.716262975778547e-06,
|
| 2111 |
+
"loss": 0.1407,
|
| 2112 |
+
"step": 2670
|
| 2113 |
+
},
|
| 2114 |
+
{
|
| 2115 |
+
"epoch": 9.27,
|
| 2116 |
+
"grad_norm": 2.7852437496185303,
|
| 2117 |
+
"learning_rate": 7.370242214532873e-06,
|
| 2118 |
+
"loss": 0.2366,
|
| 2119 |
+
"step": 2680
|
| 2120 |
+
},
|
| 2121 |
+
{
|
| 2122 |
+
"epoch": 9.31,
|
| 2123 |
+
"grad_norm": 0.2830531895160675,
|
| 2124 |
+
"learning_rate": 7.024221453287197e-06,
|
| 2125 |
+
"loss": 0.2925,
|
| 2126 |
+
"step": 2690
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 9.34,
|
| 2130 |
+
"grad_norm": 2.379260301589966,
|
| 2131 |
+
"learning_rate": 6.678200692041523e-06,
|
| 2132 |
+
"loss": 0.1131,
|
| 2133 |
+
"step": 2700
|
| 2134 |
+
},
|
| 2135 |
+
{
|
| 2136 |
+
"epoch": 9.34,
|
| 2137 |
+
"eval_accuracy": 0.8711656441717791,
|
| 2138 |
+
"eval_loss": 0.47180497646331787,
|
| 2139 |
+
"eval_runtime": 11.6646,
|
| 2140 |
+
"eval_samples_per_second": 69.87,
|
| 2141 |
+
"eval_steps_per_second": 8.744,
|
| 2142 |
+
"step": 2700
|
| 2143 |
+
},
|
| 2144 |
+
{
|
| 2145 |
+
"epoch": 9.38,
|
| 2146 |
+
"grad_norm": 7.354953289031982,
|
| 2147 |
+
"learning_rate": 6.3321799307958475e-06,
|
| 2148 |
+
"loss": 0.1997,
|
| 2149 |
+
"step": 2710
|
| 2150 |
+
},
|
| 2151 |
+
{
|
| 2152 |
+
"epoch": 9.41,
|
| 2153 |
+
"grad_norm": 0.7862864136695862,
|
| 2154 |
+
"learning_rate": 5.986159169550173e-06,
|
| 2155 |
+
"loss": 0.1147,
|
| 2156 |
+
"step": 2720
|
| 2157 |
+
},
|
| 2158 |
+
{
|
| 2159 |
+
"epoch": 9.45,
|
| 2160 |
+
"grad_norm": 2.8488690853118896,
|
| 2161 |
+
"learning_rate": 5.640138408304498e-06,
|
| 2162 |
+
"loss": 0.1345,
|
| 2163 |
+
"step": 2730
|
| 2164 |
+
},
|
| 2165 |
+
{
|
| 2166 |
+
"epoch": 9.48,
|
| 2167 |
+
"grad_norm": 4.022710800170898,
|
| 2168 |
+
"learning_rate": 5.294117647058824e-06,
|
| 2169 |
+
"loss": 0.1466,
|
| 2170 |
+
"step": 2740
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 9.52,
|
| 2174 |
+
"grad_norm": 1.0064830780029297,
|
| 2175 |
+
"learning_rate": 4.948096885813149e-06,
|
| 2176 |
+
"loss": 0.2983,
|
| 2177 |
+
"step": 2750
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 9.55,
|
| 2181 |
+
"grad_norm": 2.4876391887664795,
|
| 2182 |
+
"learning_rate": 4.602076124567474e-06,
|
| 2183 |
+
"loss": 0.2516,
|
| 2184 |
+
"step": 2760
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 9.58,
|
| 2188 |
+
"grad_norm": 8.971895217895508,
|
| 2189 |
+
"learning_rate": 4.2560553633217994e-06,
|
| 2190 |
+
"loss": 0.2621,
|
| 2191 |
+
"step": 2770
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 9.62,
|
| 2195 |
+
"grad_norm": 6.835991859436035,
|
| 2196 |
+
"learning_rate": 3.910034602076125e-06,
|
| 2197 |
+
"loss": 0.1548,
|
| 2198 |
+
"step": 2780
|
| 2199 |
+
},
|
| 2200 |
+
{
|
| 2201 |
+
"epoch": 9.65,
|
| 2202 |
+
"grad_norm": 3.8968513011932373,
|
| 2203 |
+
"learning_rate": 3.5640138408304496e-06,
|
| 2204 |
+
"loss": 0.2505,
|
| 2205 |
+
"step": 2790
|
| 2206 |
+
},
|
| 2207 |
+
{
|
| 2208 |
+
"epoch": 9.69,
|
| 2209 |
+
"grad_norm": 2.595637083053589,
|
| 2210 |
+
"learning_rate": 3.2179930795847754e-06,
|
| 2211 |
+
"loss": 0.2144,
|
| 2212 |
+
"step": 2800
|
| 2213 |
+
},
|
| 2214 |
+
{
|
| 2215 |
+
"epoch": 9.69,
|
| 2216 |
+
"eval_accuracy": 0.8711656441717791,
|
| 2217 |
+
"eval_loss": 0.4744942784309387,
|
| 2218 |
+
"eval_runtime": 11.7899,
|
| 2219 |
+
"eval_samples_per_second": 69.127,
|
| 2220 |
+
"eval_steps_per_second": 8.651,
|
| 2221 |
+
"step": 2800
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 9.72,
|
| 2225 |
+
"grad_norm": 8.595541954040527,
|
| 2226 |
+
"learning_rate": 2.8719723183391003e-06,
|
| 2227 |
+
"loss": 0.1468,
|
| 2228 |
+
"step": 2810
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 9.76,
|
| 2232 |
+
"grad_norm": 3.5496714115142822,
|
| 2233 |
+
"learning_rate": 2.5259515570934256e-06,
|
| 2234 |
+
"loss": 0.2275,
|
| 2235 |
+
"step": 2820
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 9.79,
|
| 2239 |
+
"grad_norm": 6.138185977935791,
|
| 2240 |
+
"learning_rate": 2.179930795847751e-06,
|
| 2241 |
+
"loss": 0.3254,
|
| 2242 |
+
"step": 2830
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 9.83,
|
| 2246 |
+
"grad_norm": 1.1723095178604126,
|
| 2247 |
+
"learning_rate": 1.8339100346020762e-06,
|
| 2248 |
+
"loss": 0.109,
|
| 2249 |
+
"step": 2840
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 9.86,
|
| 2253 |
+
"grad_norm": 5.896970272064209,
|
| 2254 |
+
"learning_rate": 1.4878892733564015e-06,
|
| 2255 |
+
"loss": 0.2334,
|
| 2256 |
+
"step": 2850
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 9.9,
|
| 2260 |
+
"grad_norm": 2.945650577545166,
|
| 2261 |
+
"learning_rate": 1.1764705882352942e-06,
|
| 2262 |
+
"loss": 0.3131,
|
| 2263 |
+
"step": 2860
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 9.93,
|
| 2267 |
+
"grad_norm": 5.7923583984375,
|
| 2268 |
+
"learning_rate": 8.304498269896194e-07,
|
| 2269 |
+
"loss": 0.3139,
|
| 2270 |
+
"step": 2870
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 9.97,
|
| 2274 |
+
"grad_norm": 2.053051471710205,
|
| 2275 |
+
"learning_rate": 4.844290657439446e-07,
|
| 2276 |
+
"loss": 0.2643,
|
| 2277 |
+
"step": 2880
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 10.0,
|
| 2281 |
+
"grad_norm": 0.1167406514286995,
|
| 2282 |
+
"learning_rate": 1.384083044982699e-07,
|
| 2283 |
+
"loss": 0.1631,
|
| 2284 |
+
"step": 2890
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 10.0,
|
| 2288 |
+
"step": 2890,
|
| 2289 |
+
"total_flos": 3.57806169439359e+18,
|
| 2290 |
+
"train_loss": 0.4476649462557994,
|
| 2291 |
+
"train_runtime": 1596.8664,
|
| 2292 |
+
"train_samples_per_second": 28.913,
|
| 2293 |
+
"train_steps_per_second": 1.81
|
| 2294 |
}
|
| 2295 |
],
|
| 2296 |
"logging_steps": 10,
|
| 2297 |
+
"max_steps": 2890,
|
| 2298 |
"num_input_tokens_seen": 0,
|
| 2299 |
+
"num_train_epochs": 10,
|
| 2300 |
"save_steps": 100,
|
| 2301 |
+
"total_flos": 3.57806169439359e+18,
|
| 2302 |
"train_batch_size": 16,
|
| 2303 |
"trial_name": null,
|
| 2304 |
"trial_params": null
|