🍻 cheers
Browse files- README.md +6 -5
- all_results.json +10 -10
- eval_results.json +5 -5
- runs/May05_18-32-14_a7a783b06297/events.out.tfevents.1714935493.a7a783b06297.35.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +1260 -1388
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: 0.8785648574057038
|
| 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: 0.4375
|
| 37 |
+
- Accuracy: 0.8786
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos": 3.
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second": 28.
|
| 12 |
-
"train_steps_per_second": 1.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8785648574057038,
|
| 4 |
+
"eval_loss": 0.4375220239162445,
|
| 5 |
+
"eval_runtime": 14.5705,
|
| 6 |
+
"eval_samples_per_second": 74.603,
|
| 7 |
+
"eval_steps_per_second": 9.334,
|
| 8 |
+
"total_flos": 3.3672683695343616e+18,
|
| 9 |
+
"train_loss": 0.4566012972418,
|
| 10 |
+
"train_runtime": 1529.7234,
|
| 11 |
+
"train_samples_per_second": 28.404,
|
| 12 |
+
"train_steps_per_second": 1.778
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8785648574057038,
|
| 4 |
+
"eval_loss": 0.4375220239162445,
|
| 5 |
+
"eval_runtime": 14.5705,
|
| 6 |
+
"eval_samples_per_second": 74.603,
|
| 7 |
+
"eval_steps_per_second": 9.334
|
| 8 |
}
|
runs/May05_18-32-14_a7a783b06297/events.out.tfevents.1714935493.a7a783b06297.35.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:323a6c0ce1b17dae00c9c69e48b0a72afdac406dc754bc4bd8b0f350291bcf4e
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
-
"total_flos": 3.
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second": 28.
|
| 7 |
-
"train_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"total_flos": 3.3672683695343616e+18,
|
| 4 |
+
"train_loss": 0.4566012972418,
|
| 5 |
+
"train_runtime": 1529.7234,
|
| 6 |
+
"train_samples_per_second": 28.404,
|
| 7 |
+
"train_steps_per_second": 1.778
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,2304 +1,2176 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "Action_all_10_class/checkpoint-
|
| 4 |
"epoch": 10.0,
|
| 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.
|
| 13 |
-
"grad_norm": 1.
|
| 14 |
-
"learning_rate": 9.
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.07,
|
| 20 |
-
"grad_norm": 1.
|
| 21 |
-
"learning_rate": 9.
|
| 22 |
-
"loss": 2.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 1.
|
| 28 |
-
"learning_rate": 9.
|
| 29 |
-
"loss": 2.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 1.
|
| 35 |
-
"learning_rate": 9.
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm": 2.
|
| 42 |
-
"learning_rate": 9.
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm": 2.
|
| 49 |
-
"learning_rate": 9.
|
| 50 |
-
"loss": 1.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 9.
|
| 57 |
-
"loss": 1.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm": 3.
|
| 63 |
-
"learning_rate": 9.
|
| 64 |
-
"loss": 1.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 9.
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate": 9.
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"eval_accuracy": 0.
|
| 84 |
-
"eval_loss": 1.
|
| 85 |
-
"eval_runtime":
|
| 86 |
-
"eval_samples_per_second":
|
| 87 |
-
"eval_steps_per_second":
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
-
"epoch": 0.
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate": 9.
|
| 94 |
-
"loss": 1.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch": 0.
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate": 9.
|
| 101 |
-
"loss": 1.
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch": 0.
|
| 106 |
-
"grad_norm": 3.
|
| 107 |
-
"learning_rate": 9.
|
| 108 |
-
"loss": 1.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
-
"epoch": 0.
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate": 9.
|
| 115 |
-
"loss": 1.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 0.
|
| 120 |
-
"grad_norm":
|
| 121 |
-
"learning_rate": 9.
|
| 122 |
-
"loss": 1.
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 0.
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate": 9.
|
| 129 |
-
"loss": 1.
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
-
"epoch": 0.
|
| 134 |
-
"grad_norm":
|
| 135 |
-
"learning_rate": 9.
|
| 136 |
-
"loss": 1.
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch": 0.
|
| 141 |
-
"grad_norm":
|
| 142 |
-
"learning_rate": 9.
|
| 143 |
-
"loss": 1.
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch": 0.
|
| 148 |
-
"grad_norm":
|
| 149 |
-
"learning_rate": 9.
|
| 150 |
-
"loss":
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
-
"epoch": 0.
|
| 155 |
-
"grad_norm":
|
| 156 |
-
"learning_rate": 9.
|
| 157 |
-
"loss":
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch": 0.
|
| 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.
|
| 171 |
-
"grad_norm":
|
| 172 |
-
"learning_rate": 9.
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch": 0.
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate": 9.
|
| 180 |
-
"loss":
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
-
"epoch": 0.
|
| 185 |
-
"grad_norm":
|
| 186 |
-
"learning_rate": 9.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch": 0.
|
| 192 |
-
"grad_norm": 3.
|
| 193 |
-
"learning_rate": 9.
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch": 0.
|
| 199 |
-
"grad_norm": 2.
|
| 200 |
-
"learning_rate": 9.
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"epoch": 0.
|
| 206 |
-
"grad_norm":
|
| 207 |
-
"learning_rate": 9.
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch": 0.
|
| 213 |
-
"grad_norm":
|
| 214 |
-
"learning_rate": 9.
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch":
|
| 220 |
-
"grad_norm":
|
| 221 |
-
"learning_rate":
|
| 222 |
-
"loss": 0.
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
-
"epoch": 1.
|
| 227 |
-
"grad_norm": 3.
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch": 1.
|
| 234 |
-
"grad_norm":
|
| 235 |
-
"learning_rate": 8.
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch": 1.
|
| 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.
|
| 250 |
-
"grad_norm":
|
| 251 |
-
"learning_rate": 8.
|
| 252 |
-
"loss": 0.
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
-
"epoch": 1.
|
| 257 |
-
"grad_norm":
|
| 258 |
-
"learning_rate": 8.
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch": 1.
|
| 264 |
-
"grad_norm": 4.
|
| 265 |
-
"learning_rate": 8.
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch": 1.
|
| 271 |
-
"grad_norm":
|
| 272 |
-
"learning_rate": 8.
|
| 273 |
-
"loss": 0.
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
-
"epoch": 1.
|
| 278 |
-
"grad_norm":
|
| 279 |
-
"learning_rate": 8.
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
-
"epoch": 1.
|
| 285 |
-
"grad_norm": 5.
|
| 286 |
-
"learning_rate": 8.
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
-
"epoch": 1.
|
| 292 |
-
"grad_norm": 4.
|
| 293 |
-
"learning_rate": 8.
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
-
"epoch": 1.
|
| 299 |
-
"grad_norm":
|
| 300 |
-
"learning_rate": 8.
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
-
"epoch": 1.
|
| 306 |
-
"grad_norm":
|
| 307 |
-
"learning_rate": 8.
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch": 1.
|
| 313 |
-
"grad_norm":
|
| 314 |
-
"learning_rate": 8.
|
| 315 |
-
"loss": 0.
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
-
"epoch": 1.
|
| 320 |
-
"eval_accuracy": 0.
|
| 321 |
-
"eval_loss": 0.
|
| 322 |
-
"eval_runtime":
|
| 323 |
-
"eval_samples_per_second":
|
| 324 |
-
"eval_steps_per_second":
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
-
"epoch": 1.
|
| 329 |
-
"grad_norm":
|
| 330 |
-
"learning_rate": 8.
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch": 1.
|
| 336 |
-
"grad_norm":
|
| 337 |
-
"learning_rate": 8.
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch": 1.
|
| 343 |
-
"grad_norm":
|
| 344 |
-
"learning_rate": 8.
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch": 1.
|
| 350 |
-
"grad_norm": 4.
|
| 351 |
-
"learning_rate": 8.
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch": 1.
|
| 357 |
-
"grad_norm":
|
| 358 |
-
"learning_rate": 8.
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch": 1.
|
| 364 |
-
"grad_norm":
|
| 365 |
-
"learning_rate": 8.
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch": 1.
|
| 371 |
-
"grad_norm":
|
| 372 |
-
"learning_rate": 8.
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch": 1.
|
| 378 |
-
"grad_norm":
|
| 379 |
-
"learning_rate": 8.
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch": 1.
|
| 385 |
-
"grad_norm":
|
| 386 |
-
"learning_rate": 8.
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch": 1.
|
| 392 |
-
"grad_norm":
|
| 393 |
-
"learning_rate": 8.
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch": 1.
|
| 399 |
-
"eval_accuracy": 0.
|
| 400 |
-
"eval_loss": 0.
|
| 401 |
-
"eval_runtime":
|
| 402 |
-
"eval_samples_per_second":
|
| 403 |
-
"eval_steps_per_second": 9.
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
-
"epoch": 1.
|
| 408 |
-
"grad_norm":
|
| 409 |
-
"learning_rate": 8.
|
| 410 |
-
"loss": 0.
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
-
"epoch": 1.
|
| 415 |
-
"grad_norm":
|
| 416 |
-
"learning_rate": 8.
|
| 417 |
-
"loss": 0.
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
-
"epoch": 1.
|
| 422 |
-
"grad_norm":
|
| 423 |
-
"learning_rate": 8.
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch": 1.
|
| 429 |
-
"grad_norm":
|
| 430 |
-
"learning_rate": 8.
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
-
"epoch":
|
| 436 |
-
"grad_norm": 2.
|
| 437 |
-
"learning_rate":
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
-
"epoch":
|
| 443 |
-
"grad_norm":
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
-
"epoch": 1
|
| 450 |
-
"grad_norm":
|
| 451 |
-
"learning_rate":
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch": 2.
|
| 457 |
-
"grad_norm": 5.
|
| 458 |
-
"learning_rate": 7.
|
| 459 |
-
"loss": 0.
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
-
"epoch": 2.
|
| 464 |
-
"grad_norm":
|
| 465 |
-
"learning_rate": 7.
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
-
"epoch": 2.
|
| 471 |
-
"grad_norm":
|
| 472 |
-
"learning_rate": 7.
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
-
"epoch": 2.
|
| 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.
|
| 487 |
-
"grad_norm":
|
| 488 |
-
"learning_rate": 7.
|
| 489 |
-
"loss": 0.
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
-
"epoch": 2.
|
| 494 |
-
"grad_norm":
|
| 495 |
-
"learning_rate": 7.
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch": 2.
|
| 501 |
-
"grad_norm":
|
| 502 |
-
"learning_rate": 7.
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
-
"epoch": 2.
|
| 508 |
-
"grad_norm":
|
| 509 |
-
"learning_rate": 7.
|
| 510 |
-
"loss": 0.
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
-
"epoch": 2.
|
| 515 |
-
"grad_norm":
|
| 516 |
-
"learning_rate": 7.
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
-
"epoch": 2.
|
| 522 |
-
"grad_norm":
|
| 523 |
-
"learning_rate": 7.
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
-
"epoch": 2.
|
| 529 |
-
"grad_norm": 4.
|
| 530 |
-
"learning_rate": 7.
|
| 531 |
-
"loss": 0.
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
-
"epoch": 2.
|
| 536 |
-
"grad_norm":
|
| 537 |
-
"learning_rate": 7.
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
-
"epoch": 2.
|
| 543 |
-
"grad_norm":
|
| 544 |
-
"learning_rate": 7.
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
-
"epoch": 2.
|
| 550 |
-
"grad_norm": 5.
|
| 551 |
-
"learning_rate": 7.
|
| 552 |
-
"loss": 0.
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
-
"epoch": 2.
|
| 557 |
-
"eval_accuracy": 0.
|
| 558 |
-
"eval_loss": 0.
|
| 559 |
-
"eval_runtime":
|
| 560 |
-
"eval_samples_per_second":
|
| 561 |
-
"eval_steps_per_second":
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch": 2.
|
| 566 |
-
"grad_norm": 3.
|
| 567 |
-
"learning_rate": 7.
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 2.
|
| 573 |
-
"grad_norm":
|
| 574 |
-
"learning_rate": 7.
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch": 2.
|
| 580 |
-
"grad_norm":
|
| 581 |
-
"learning_rate": 7.
|
| 582 |
-
"loss": 0.
|
| 583 |
"step": 730
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 2.
|
| 587 |
-
"grad_norm":
|
| 588 |
-
"learning_rate": 7.
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 2.
|
| 594 |
-
"grad_norm":
|
| 595 |
-
"learning_rate": 7.
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 2.
|
| 601 |
-
"grad_norm":
|
| 602 |
-
"learning_rate": 7.
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch": 2.
|
| 608 |
-
"grad_norm": 4.
|
| 609 |
-
"learning_rate": 7.
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 2.
|
| 615 |
-
"grad_norm": 5.
|
| 616 |
-
"learning_rate": 7.
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch": 2.
|
| 622 |
-
"grad_norm":
|
| 623 |
-
"learning_rate": 7.
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch": 2.
|
| 629 |
-
"grad_norm":
|
| 630 |
-
"learning_rate": 7.
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 2.
|
| 636 |
-
"eval_accuracy": 0.
|
| 637 |
-
"eval_loss": 0.
|
| 638 |
-
"eval_runtime": 14.
|
| 639 |
-
"eval_samples_per_second":
|
| 640 |
-
"eval_steps_per_second":
|
| 641 |
"step": 800
|
| 642 |
},
|
| 643 |
{
|
| 644 |
-
"epoch": 2.
|
| 645 |
-
"grad_norm":
|
| 646 |
-
"learning_rate": 7.
|
| 647 |
-
"loss": 0.
|
| 648 |
"step": 810
|
| 649 |
},
|
| 650 |
{
|
| 651 |
-
"epoch":
|
| 652 |
-
"grad_norm":
|
| 653 |
-
"learning_rate":
|
| 654 |
-
"loss": 0.
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
-
"epoch":
|
| 659 |
-
"grad_norm": 2.
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
-
"epoch":
|
| 666 |
-
"grad_norm":
|
| 667 |
-
"learning_rate":
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
-
"epoch":
|
| 673 |
-
"grad_norm": 5.
|
| 674 |
-
"learning_rate":
|
| 675 |
-
"loss": 0.
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
-
"epoch":
|
| 680 |
-
"grad_norm":
|
| 681 |
-
"learning_rate":
|
| 682 |
-
"loss": 0.
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
-
"epoch": 3.
|
| 687 |
-
"grad_norm":
|
| 688 |
-
"learning_rate": 6.
|
| 689 |
-
"loss": 0.
|
| 690 |
"step": 870
|
| 691 |
},
|
| 692 |
{
|
| 693 |
-
"epoch": 3.
|
| 694 |
-
"grad_norm":
|
| 695 |
-
"learning_rate": 6.
|
| 696 |
-
"loss": 0.
|
| 697 |
"step": 880
|
| 698 |
},
|
| 699 |
{
|
| 700 |
-
"epoch": 3.
|
| 701 |
-
"grad_norm": 6.
|
| 702 |
-
"learning_rate": 6.
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 890
|
| 705 |
},
|
| 706 |
{
|
| 707 |
-
"epoch": 3.
|
| 708 |
-
"grad_norm": 3.
|
| 709 |
-
"learning_rate": 6.
|
| 710 |
-
"loss": 0.
|
| 711 |
"step": 900
|
| 712 |
},
|
| 713 |
{
|
| 714 |
-
"epoch": 3.
|
| 715 |
-
"eval_accuracy": 0.
|
| 716 |
-
"eval_loss": 0.
|
| 717 |
-
"eval_runtime":
|
| 718 |
-
"eval_samples_per_second":
|
| 719 |
-
"eval_steps_per_second":
|
| 720 |
"step": 900
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"epoch": 3.
|
| 724 |
-
"grad_norm":
|
| 725 |
-
"learning_rate": 6.
|
| 726 |
-
"loss": 0.
|
| 727 |
"step": 910
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"epoch": 3.
|
| 731 |
-
"grad_norm":
|
| 732 |
-
"learning_rate": 6.
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 920
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch": 3.
|
| 738 |
-
"grad_norm": 5.
|
| 739 |
-
"learning_rate": 6.
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 930
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch": 3.
|
| 745 |
-
"grad_norm":
|
| 746 |
-
"learning_rate": 6.
|
| 747 |
-
"loss": 0.
|
| 748 |
"step": 940
|
| 749 |
},
|
| 750 |
{
|
| 751 |
-
"epoch": 3.
|
| 752 |
-
"grad_norm":
|
| 753 |
-
"learning_rate": 6.
|
| 754 |
-
"loss": 0.
|
| 755 |
"step": 950
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch": 3.
|
| 759 |
-
"grad_norm":
|
| 760 |
-
"learning_rate": 6.
|
| 761 |
-
"loss": 0.
|
| 762 |
"step": 960
|
| 763 |
},
|
| 764 |
{
|
| 765 |
-
"epoch": 3.
|
| 766 |
-
"grad_norm":
|
| 767 |
-
"learning_rate": 6.
|
| 768 |
-
"loss": 0.
|
| 769 |
"step": 970
|
| 770 |
},
|
| 771 |
{
|
| 772 |
-
"epoch": 3.
|
| 773 |
-
"grad_norm": 1.
|
| 774 |
-
"learning_rate": 6.
|
| 775 |
-
"loss": 0.
|
| 776 |
"step": 980
|
| 777 |
},
|
| 778 |
{
|
| 779 |
-
"epoch": 3.
|
| 780 |
-
"grad_norm":
|
| 781 |
-
"learning_rate": 6.
|
| 782 |
-
"loss": 0.
|
| 783 |
"step": 990
|
| 784 |
},
|
| 785 |
{
|
| 786 |
-
"epoch": 3.
|
| 787 |
-
"grad_norm":
|
| 788 |
-
"learning_rate": 6.
|
| 789 |
-
"loss": 0.
|
| 790 |
"step": 1000
|
| 791 |
},
|
| 792 |
{
|
| 793 |
-
"epoch": 3.
|
| 794 |
-
"eval_accuracy": 0.
|
| 795 |
-
"eval_loss": 0.
|
| 796 |
-
"eval_runtime":
|
| 797 |
-
"eval_samples_per_second": 73.
|
| 798 |
-
"eval_steps_per_second": 9.
|
| 799 |
"step": 1000
|
| 800 |
},
|
| 801 |
{
|
| 802 |
-
"epoch": 3.
|
| 803 |
-
"grad_norm":
|
| 804 |
-
"learning_rate": 6.
|
| 805 |
-
"loss": 0.
|
| 806 |
"step": 1010
|
| 807 |
},
|
| 808 |
{
|
| 809 |
-
"epoch": 3.
|
| 810 |
-
"grad_norm":
|
| 811 |
-
"learning_rate": 6.
|
| 812 |
-
"loss": 0.
|
| 813 |
"step": 1020
|
| 814 |
},
|
| 815 |
{
|
| 816 |
-
"epoch": 3.
|
| 817 |
-
"grad_norm":
|
| 818 |
-
"learning_rate": 6.
|
| 819 |
-
"loss": 0.
|
| 820 |
"step": 1030
|
| 821 |
},
|
| 822 |
{
|
| 823 |
-
"epoch": 3.
|
| 824 |
-
"grad_norm":
|
| 825 |
-
"learning_rate": 6.
|
| 826 |
-
"loss": 0.
|
| 827 |
"step": 1040
|
| 828 |
},
|
| 829 |
{
|
| 830 |
-
"epoch": 3.
|
| 831 |
-
"grad_norm":
|
| 832 |
-
"learning_rate": 6.
|
| 833 |
-
"loss": 0.
|
| 834 |
"step": 1050
|
| 835 |
},
|
| 836 |
{
|
| 837 |
-
"epoch": 3.
|
| 838 |
-
"grad_norm":
|
| 839 |
-
"learning_rate": 6.
|
| 840 |
-
"loss": 0.
|
| 841 |
"step": 1060
|
| 842 |
},
|
| 843 |
{
|
| 844 |
-
"epoch": 3.
|
| 845 |
-
"grad_norm":
|
| 846 |
-
"learning_rate": 6.
|
| 847 |
-
"loss": 0.
|
| 848 |
"step": 1070
|
| 849 |
},
|
| 850 |
{
|
| 851 |
-
"epoch": 3.
|
| 852 |
-
"grad_norm":
|
| 853 |
-
"learning_rate": 6.
|
| 854 |
-
"loss": 0.
|
| 855 |
"step": 1080
|
| 856 |
},
|
| 857 |
{
|
| 858 |
-
"epoch":
|
| 859 |
-
"grad_norm":
|
| 860 |
-
"learning_rate":
|
| 861 |
-
"loss": 0.
|
| 862 |
"step": 1090
|
| 863 |
},
|
| 864 |
{
|
| 865 |
-
"epoch":
|
| 866 |
-
"grad_norm":
|
| 867 |
-
"learning_rate":
|
| 868 |
-
"loss": 0.
|
| 869 |
"step": 1100
|
| 870 |
},
|
| 871 |
{
|
| 872 |
-
"epoch":
|
| 873 |
-
"eval_accuracy": 0.
|
| 874 |
-
"eval_loss": 0.
|
| 875 |
-
"eval_runtime":
|
| 876 |
-
"eval_samples_per_second":
|
| 877 |
-
"eval_steps_per_second":
|
| 878 |
"step": 1100
|
| 879 |
},
|
| 880 |
{
|
| 881 |
-
"epoch":
|
| 882 |
-
"grad_norm":
|
| 883 |
-
"learning_rate":
|
| 884 |
-
"loss": 0.
|
| 885 |
"step": 1110
|
| 886 |
},
|
| 887 |
{
|
| 888 |
-
"epoch":
|
| 889 |
-
"grad_norm": 3.
|
| 890 |
-
"learning_rate":
|
| 891 |
-
"loss": 0.
|
| 892 |
"step": 1120
|
| 893 |
},
|
| 894 |
{
|
| 895 |
-
"epoch":
|
| 896 |
-
"grad_norm":
|
| 897 |
-
"learning_rate":
|
| 898 |
-
"loss": 0.
|
| 899 |
"step": 1130
|
| 900 |
},
|
| 901 |
{
|
| 902 |
-
"epoch":
|
| 903 |
-
"grad_norm":
|
| 904 |
-
"learning_rate":
|
| 905 |
-
"loss": 0.
|
| 906 |
"step": 1140
|
| 907 |
},
|
| 908 |
{
|
| 909 |
-
"epoch":
|
| 910 |
-
"grad_norm":
|
| 911 |
-
"learning_rate":
|
| 912 |
-
"loss": 0.
|
| 913 |
"step": 1150
|
| 914 |
},
|
| 915 |
{
|
| 916 |
-
"epoch": 4.
|
| 917 |
-
"grad_norm":
|
| 918 |
-
"learning_rate": 5.
|
| 919 |
-
"loss": 0.
|
| 920 |
"step": 1160
|
| 921 |
},
|
| 922 |
{
|
| 923 |
-
"epoch": 4.
|
| 924 |
-
"grad_norm":
|
| 925 |
-
"learning_rate": 5.
|
| 926 |
-
"loss": 0.
|
| 927 |
"step": 1170
|
| 928 |
},
|
| 929 |
{
|
| 930 |
-
"epoch": 4.
|
| 931 |
-
"grad_norm":
|
| 932 |
-
"learning_rate": 5.
|
| 933 |
-
"loss": 0.
|
| 934 |
"step": 1180
|
| 935 |
},
|
| 936 |
{
|
| 937 |
-
"epoch": 4.
|
| 938 |
-
"grad_norm":
|
| 939 |
-
"learning_rate": 5.
|
| 940 |
-
"loss": 0.
|
| 941 |
"step": 1190
|
| 942 |
},
|
| 943 |
{
|
| 944 |
-
"epoch": 4.
|
| 945 |
-
"grad_norm":
|
| 946 |
-
"learning_rate": 5.
|
| 947 |
-
"loss": 0.
|
| 948 |
"step": 1200
|
| 949 |
},
|
| 950 |
{
|
| 951 |
-
"epoch": 4.
|
| 952 |
-
"eval_accuracy": 0.
|
| 953 |
-
"eval_loss": 0.
|
| 954 |
-
"eval_runtime":
|
| 955 |
-
"eval_samples_per_second":
|
| 956 |
-
"eval_steps_per_second":
|
| 957 |
"step": 1200
|
| 958 |
},
|
| 959 |
{
|
| 960 |
-
"epoch": 4.
|
| 961 |
-
"grad_norm":
|
| 962 |
-
"learning_rate": 5.
|
| 963 |
-
"loss": 0.
|
| 964 |
"step": 1210
|
| 965 |
},
|
| 966 |
{
|
| 967 |
-
"epoch": 4.
|
| 968 |
-
"grad_norm":
|
| 969 |
-
"learning_rate": 5.
|
| 970 |
-
"loss": 0.
|
| 971 |
"step": 1220
|
| 972 |
},
|
| 973 |
{
|
| 974 |
-
"epoch": 4.
|
| 975 |
-
"grad_norm":
|
| 976 |
-
"learning_rate": 5.
|
| 977 |
-
"loss": 0.
|
| 978 |
"step": 1230
|
| 979 |
},
|
| 980 |
{
|
| 981 |
-
"epoch": 4.
|
| 982 |
-
"grad_norm":
|
| 983 |
-
"learning_rate": 5.
|
| 984 |
-
"loss": 0.
|
| 985 |
"step": 1240
|
| 986 |
},
|
| 987 |
{
|
| 988 |
-
"epoch": 4.
|
| 989 |
-
"grad_norm":
|
| 990 |
-
"learning_rate": 5.
|
| 991 |
-
"loss": 0.
|
| 992 |
"step": 1250
|
| 993 |
},
|
| 994 |
{
|
| 995 |
-
"epoch": 4.
|
| 996 |
-
"grad_norm":
|
| 997 |
-
"learning_rate": 5.
|
| 998 |
-
"loss": 0.
|
| 999 |
"step": 1260
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
-
"epoch": 4.
|
| 1003 |
-
"grad_norm":
|
| 1004 |
-
"learning_rate": 5.
|
| 1005 |
-
"loss": 0.
|
| 1006 |
"step": 1270
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
-
"epoch": 4.
|
| 1010 |
-
"grad_norm":
|
| 1011 |
-
"learning_rate": 5.
|
| 1012 |
-
"loss": 0.
|
| 1013 |
"step": 1280
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
-
"epoch": 4.
|
| 1017 |
-
"grad_norm":
|
| 1018 |
-
"learning_rate": 5.
|
| 1019 |
-
"loss": 0.
|
| 1020 |
"step": 1290
|
| 1021 |
},
|
| 1022 |
{
|
| 1023 |
-
"epoch": 4.
|
| 1024 |
-
"grad_norm":
|
| 1025 |
-
"learning_rate": 5.
|
| 1026 |
-
"loss": 0.
|
| 1027 |
"step": 1300
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
-
"epoch": 4.
|
| 1031 |
-
"eval_accuracy": 0.
|
| 1032 |
-
"eval_loss": 0.
|
| 1033 |
-
"eval_runtime":
|
| 1034 |
-
"eval_samples_per_second":
|
| 1035 |
-
"eval_steps_per_second":
|
| 1036 |
"step": 1300
|
| 1037 |
},
|
| 1038 |
{
|
| 1039 |
-
"epoch": 4.
|
| 1040 |
-
"grad_norm":
|
| 1041 |
-
"learning_rate": 5.
|
| 1042 |
-
"loss": 0.
|
| 1043 |
"step": 1310
|
| 1044 |
},
|
| 1045 |
{
|
| 1046 |
-
"epoch": 4.
|
| 1047 |
-
"grad_norm":
|
| 1048 |
-
"learning_rate": 5.
|
| 1049 |
-
"loss": 0.
|
| 1050 |
"step": 1320
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
-
"epoch": 4.
|
| 1054 |
-
"grad_norm":
|
| 1055 |
-
"learning_rate": 5.
|
| 1056 |
-
"loss": 0.
|
| 1057 |
"step": 1330
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
-
"epoch": 4.
|
| 1061 |
-
"grad_norm":
|
| 1062 |
-
"learning_rate": 5.
|
| 1063 |
-
"loss": 0.
|
| 1064 |
"step": 1340
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
-
"epoch": 4.
|
| 1068 |
-
"grad_norm":
|
| 1069 |
-
"learning_rate": 5.
|
| 1070 |
-
"loss": 0.
|
| 1071 |
"step": 1350
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
-
"epoch":
|
| 1075 |
-
"grad_norm": 9.
|
| 1076 |
-
"learning_rate": 5.
|
| 1077 |
-
"loss": 0.
|
| 1078 |
"step": 1360
|
| 1079 |
},
|
| 1080 |
{
|
| 1081 |
-
"epoch":
|
| 1082 |
-
"grad_norm":
|
| 1083 |
-
"learning_rate":
|
| 1084 |
-
"loss": 0.
|
| 1085 |
"step": 1370
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
-
"epoch":
|
| 1089 |
-
"grad_norm":
|
| 1090 |
-
"learning_rate":
|
| 1091 |
-
"loss": 0.
|
| 1092 |
"step": 1380
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
-
"epoch":
|
| 1096 |
-
"grad_norm":
|
| 1097 |
-
"learning_rate":
|
| 1098 |
-
"loss": 0.
|
| 1099 |
"step": 1390
|
| 1100 |
},
|
| 1101 |
{
|
| 1102 |
-
"epoch":
|
| 1103 |
-
"grad_norm":
|
| 1104 |
-
"learning_rate":
|
| 1105 |
-
"loss": 0.
|
| 1106 |
"step": 1400
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
-
"epoch":
|
| 1110 |
-
"eval_accuracy": 0.
|
| 1111 |
-
"eval_loss": 0.
|
| 1112 |
-
"eval_runtime":
|
| 1113 |
-
"eval_samples_per_second":
|
| 1114 |
-
"eval_steps_per_second":
|
| 1115 |
"step": 1400
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
-
"epoch":
|
| 1119 |
-
"grad_norm":
|
| 1120 |
-
"learning_rate":
|
| 1121 |
-
"loss": 0.
|
| 1122 |
"step": 1410
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
-
"epoch":
|
| 1126 |
-
"grad_norm": 6.
|
| 1127 |
-
"learning_rate":
|
| 1128 |
-
"loss": 0.
|
| 1129 |
"step": 1420
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
-
"epoch":
|
| 1133 |
-
"grad_norm":
|
| 1134 |
-
"learning_rate":
|
| 1135 |
-
"loss": 0.
|
| 1136 |
"step": 1430
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
-
"epoch":
|
| 1140 |
-
"grad_norm":
|
| 1141 |
-
"learning_rate":
|
| 1142 |
-
"loss": 0.
|
| 1143 |
"step": 1440
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
-
"epoch": 5.
|
| 1147 |
-
"grad_norm":
|
| 1148 |
-
"learning_rate": 4.
|
| 1149 |
-
"loss": 0.
|
| 1150 |
"step": 1450
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
-
"epoch": 5.
|
| 1154 |
-
"grad_norm": 6.
|
| 1155 |
-
"learning_rate": 4.
|
| 1156 |
-
"loss": 0.
|
| 1157 |
"step": 1460
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
-
"epoch": 5.
|
| 1161 |
-
"grad_norm":
|
| 1162 |
-
"learning_rate": 4.
|
| 1163 |
-
"loss": 0.
|
| 1164 |
"step": 1470
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
-
"epoch": 5.
|
| 1168 |
-
"grad_norm":
|
| 1169 |
-
"learning_rate": 4.
|
| 1170 |
-
"loss": 0.
|
| 1171 |
"step": 1480
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
-
"epoch": 5.
|
| 1175 |
-
"grad_norm":
|
| 1176 |
-
"learning_rate": 4.
|
| 1177 |
-
"loss": 0.
|
| 1178 |
"step": 1490
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
-
"epoch": 5.
|
| 1182 |
-
"grad_norm":
|
| 1183 |
-
"learning_rate": 4.
|
| 1184 |
-
"loss": 0.
|
| 1185 |
"step": 1500
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
-
"epoch": 5.
|
| 1189 |
-
"eval_accuracy": 0.
|
| 1190 |
-
"eval_loss": 0.
|
| 1191 |
-
"eval_runtime":
|
| 1192 |
-
"eval_samples_per_second":
|
| 1193 |
-
"eval_steps_per_second":
|
| 1194 |
"step": 1500
|
| 1195 |
},
|
| 1196 |
{
|
| 1197 |
-
"epoch": 5.
|
| 1198 |
-
"grad_norm": 1.
|
| 1199 |
-
"learning_rate": 4.
|
| 1200 |
-
"loss": 0.
|
| 1201 |
"step": 1510
|
| 1202 |
},
|
| 1203 |
{
|
| 1204 |
-
"epoch": 5.
|
| 1205 |
-
"grad_norm":
|
| 1206 |
-
"learning_rate": 4.
|
| 1207 |
-
"loss": 0.
|
| 1208 |
"step": 1520
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
-
"epoch": 5.
|
| 1212 |
-
"grad_norm": 4.
|
| 1213 |
-
"learning_rate": 4.
|
| 1214 |
-
"loss": 0.
|
| 1215 |
"step": 1530
|
| 1216 |
},
|
| 1217 |
{
|
| 1218 |
-
"epoch": 5.
|
| 1219 |
-
"grad_norm":
|
| 1220 |
-
"learning_rate": 4.
|
| 1221 |
-
"loss": 0.
|
| 1222 |
"step": 1540
|
| 1223 |
},
|
| 1224 |
{
|
| 1225 |
-
"epoch": 5.
|
| 1226 |
-
"grad_norm":
|
| 1227 |
-
"learning_rate": 4.
|
| 1228 |
-
"loss": 0.
|
| 1229 |
"step": 1550
|
| 1230 |
},
|
| 1231 |
{
|
| 1232 |
-
"epoch": 5.
|
| 1233 |
-
"grad_norm":
|
| 1234 |
-
"learning_rate": 4.
|
| 1235 |
-
"loss": 0.
|
| 1236 |
"step": 1560
|
| 1237 |
},
|
| 1238 |
{
|
| 1239 |
-
"epoch": 5.
|
| 1240 |
-
"grad_norm":
|
| 1241 |
-
"learning_rate": 4.
|
| 1242 |
-
"loss": 0.
|
| 1243 |
"step": 1570
|
| 1244 |
},
|
| 1245 |
{
|
| 1246 |
-
"epoch": 5.
|
| 1247 |
-
"grad_norm":
|
| 1248 |
-
"learning_rate": 4.
|
| 1249 |
-
"loss": 0.
|
| 1250 |
"step": 1580
|
| 1251 |
},
|
| 1252 |
{
|
| 1253 |
-
"epoch": 5.
|
| 1254 |
-
"grad_norm":
|
| 1255 |
-
"learning_rate": 4.
|
| 1256 |
-
"loss": 0.
|
| 1257 |
"step": 1590
|
| 1258 |
},
|
| 1259 |
{
|
| 1260 |
-
"epoch": 5.
|
| 1261 |
-
"grad_norm":
|
| 1262 |
-
"learning_rate": 4.
|
| 1263 |
-
"loss": 0.
|
| 1264 |
"step": 1600
|
| 1265 |
},
|
| 1266 |
{
|
| 1267 |
-
"epoch": 5.
|
| 1268 |
-
"eval_accuracy": 0.
|
| 1269 |
-
"eval_loss": 0.
|
| 1270 |
-
"eval_runtime":
|
| 1271 |
-
"eval_samples_per_second":
|
| 1272 |
-
"eval_steps_per_second":
|
| 1273 |
"step": 1600
|
| 1274 |
},
|
| 1275 |
{
|
| 1276 |
-
"epoch": 5.
|
| 1277 |
-
"grad_norm":
|
| 1278 |
-
"learning_rate": 4.
|
| 1279 |
-
"loss": 0.
|
| 1280 |
"step": 1610
|
| 1281 |
},
|
| 1282 |
{
|
| 1283 |
-
"epoch": 5.
|
| 1284 |
-
"grad_norm":
|
| 1285 |
-
"learning_rate": 4.
|
| 1286 |
-
"loss": 0.
|
| 1287 |
"step": 1620
|
| 1288 |
},
|
| 1289 |
{
|
| 1290 |
-
"epoch": 5.
|
| 1291 |
-
"grad_norm":
|
| 1292 |
-
"learning_rate": 4.
|
| 1293 |
-
"loss": 0.
|
| 1294 |
"step": 1630
|
| 1295 |
},
|
| 1296 |
{
|
| 1297 |
-
"epoch":
|
| 1298 |
-
"grad_norm":
|
| 1299 |
-
"learning_rate":
|
| 1300 |
-
"loss": 0.
|
| 1301 |
"step": 1640
|
| 1302 |
},
|
| 1303 |
{
|
| 1304 |
-
"epoch":
|
| 1305 |
-
"grad_norm":
|
| 1306 |
-
"learning_rate":
|
| 1307 |
-
"loss": 0.
|
| 1308 |
"step": 1650
|
| 1309 |
},
|
| 1310 |
{
|
| 1311 |
-
"epoch":
|
| 1312 |
-
"grad_norm":
|
| 1313 |
-
"learning_rate":
|
| 1314 |
-
"loss": 0.
|
| 1315 |
"step": 1660
|
| 1316 |
},
|
| 1317 |
{
|
| 1318 |
-
"epoch":
|
| 1319 |
-
"grad_norm":
|
| 1320 |
-
"learning_rate":
|
| 1321 |
-
"loss": 0.
|
| 1322 |
"step": 1670
|
| 1323 |
},
|
| 1324 |
{
|
| 1325 |
-
"epoch":
|
| 1326 |
-
"grad_norm":
|
| 1327 |
-
"learning_rate":
|
| 1328 |
-
"loss": 0.
|
| 1329 |
"step": 1680
|
| 1330 |
},
|
| 1331 |
{
|
| 1332 |
-
"epoch":
|
| 1333 |
-
"grad_norm":
|
| 1334 |
-
"learning_rate":
|
| 1335 |
-
"loss": 0.
|
| 1336 |
"step": 1690
|
| 1337 |
},
|
| 1338 |
{
|
| 1339 |
-
"epoch":
|
| 1340 |
-
"grad_norm":
|
| 1341 |
-
"learning_rate":
|
| 1342 |
-
"loss": 0.
|
| 1343 |
"step": 1700
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
-
"epoch":
|
| 1347 |
-
"eval_accuracy": 0.
|
| 1348 |
-
"eval_loss": 0.
|
| 1349 |
-
"eval_runtime":
|
| 1350 |
-
"eval_samples_per_second":
|
| 1351 |
-
"eval_steps_per_second":
|
| 1352 |
"step": 1700
|
| 1353 |
},
|
| 1354 |
{
|
| 1355 |
-
"epoch":
|
| 1356 |
-
"grad_norm":
|
| 1357 |
-
"learning_rate":
|
| 1358 |
-
"loss": 0.
|
| 1359 |
"step": 1710
|
| 1360 |
},
|
| 1361 |
{
|
| 1362 |
-
"epoch":
|
| 1363 |
-
"grad_norm":
|
| 1364 |
-
"learning_rate":
|
| 1365 |
-
"loss": 0.
|
| 1366 |
"step": 1720
|
| 1367 |
},
|
| 1368 |
{
|
| 1369 |
-
"epoch":
|
| 1370 |
-
"grad_norm":
|
| 1371 |
-
"learning_rate":
|
| 1372 |
-
"loss": 0.
|
| 1373 |
"step": 1730
|
| 1374 |
},
|
| 1375 |
{
|
| 1376 |
-
"epoch": 6.
|
| 1377 |
-
"grad_norm":
|
| 1378 |
-
"learning_rate": 3.
|
| 1379 |
-
"loss": 0.
|
| 1380 |
"step": 1740
|
| 1381 |
},
|
| 1382 |
{
|
| 1383 |
-
"epoch": 6.
|
| 1384 |
-
"grad_norm":
|
| 1385 |
-
"learning_rate": 3.
|
| 1386 |
-
"loss": 0.
|
| 1387 |
"step": 1750
|
| 1388 |
},
|
| 1389 |
{
|
| 1390 |
-
"epoch": 6.
|
| 1391 |
-
"grad_norm":
|
| 1392 |
-
"learning_rate": 3.
|
| 1393 |
-
"loss": 0.
|
| 1394 |
"step": 1760
|
| 1395 |
},
|
| 1396 |
{
|
| 1397 |
-
"epoch": 6.
|
| 1398 |
-
"grad_norm":
|
| 1399 |
-
"learning_rate": 3.
|
| 1400 |
-
"loss": 0.
|
| 1401 |
"step": 1770
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
-
"epoch": 6.
|
| 1405 |
-
"grad_norm":
|
| 1406 |
-
"learning_rate": 3.
|
| 1407 |
-
"loss": 0.
|
| 1408 |
"step": 1780
|
| 1409 |
},
|
| 1410 |
{
|
| 1411 |
-
"epoch": 6.
|
| 1412 |
-
"grad_norm":
|
| 1413 |
-
"learning_rate": 3.
|
| 1414 |
-
"loss": 0.
|
| 1415 |
"step": 1790
|
| 1416 |
},
|
| 1417 |
{
|
| 1418 |
-
"epoch": 6.
|
| 1419 |
-
"grad_norm":
|
| 1420 |
-
"learning_rate": 3.
|
| 1421 |
-
"loss": 0.
|
| 1422 |
"step": 1800
|
| 1423 |
},
|
| 1424 |
{
|
| 1425 |
-
"epoch": 6.
|
| 1426 |
-
"eval_accuracy": 0.
|
| 1427 |
-
"eval_loss": 0.
|
| 1428 |
-
"eval_runtime":
|
| 1429 |
-
"eval_samples_per_second":
|
| 1430 |
-
"eval_steps_per_second":
|
| 1431 |
"step": 1800
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
-
"epoch": 6.
|
| 1435 |
-
"grad_norm":
|
| 1436 |
-
"learning_rate": 3.
|
| 1437 |
-
"loss": 0.
|
| 1438 |
"step": 1810
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
-
"epoch": 6.
|
| 1442 |
-
"grad_norm":
|
| 1443 |
-
"learning_rate": 3.
|
| 1444 |
-
"loss": 0.
|
| 1445 |
"step": 1820
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
-
"epoch": 6.
|
| 1449 |
-
"grad_norm":
|
| 1450 |
-
"learning_rate": 3.
|
| 1451 |
-
"loss": 0.
|
| 1452 |
"step": 1830
|
| 1453 |
},
|
| 1454 |
{
|
| 1455 |
-
"epoch": 6.
|
| 1456 |
-
"grad_norm":
|
| 1457 |
-
"learning_rate": 3.
|
| 1458 |
-
"loss": 0.
|
| 1459 |
"step": 1840
|
| 1460 |
},
|
| 1461 |
{
|
| 1462 |
-
"epoch": 6.
|
| 1463 |
-
"grad_norm":
|
| 1464 |
-
"learning_rate": 3.
|
| 1465 |
-
"loss": 0.
|
| 1466 |
"step": 1850
|
| 1467 |
},
|
| 1468 |
{
|
| 1469 |
-
"epoch": 6.
|
| 1470 |
-
"grad_norm":
|
| 1471 |
-
"learning_rate": 3.
|
| 1472 |
-
"loss": 0.
|
| 1473 |
"step": 1860
|
| 1474 |
},
|
| 1475 |
{
|
| 1476 |
-
"epoch": 6.
|
| 1477 |
-
"grad_norm": 2.
|
| 1478 |
-
"learning_rate": 3.
|
| 1479 |
-
"loss": 0.
|
| 1480 |
"step": 1870
|
| 1481 |
},
|
| 1482 |
{
|
| 1483 |
-
"epoch": 6.
|
| 1484 |
-
"grad_norm":
|
| 1485 |
-
"learning_rate": 3.
|
| 1486 |
-
"loss": 0.
|
| 1487 |
"step": 1880
|
| 1488 |
},
|
| 1489 |
{
|
| 1490 |
-
"epoch": 6.
|
| 1491 |
-
"grad_norm":
|
| 1492 |
-
"learning_rate": 3.
|
| 1493 |
-
"loss": 0.
|
| 1494 |
"step": 1890
|
| 1495 |
},
|
| 1496 |
{
|
| 1497 |
-
"epoch": 6.
|
| 1498 |
-
"grad_norm": 4.
|
| 1499 |
-
"learning_rate": 3.
|
| 1500 |
-
"loss": 0.
|
| 1501 |
"step": 1900
|
| 1502 |
},
|
| 1503 |
{
|
| 1504 |
-
"epoch": 6.
|
| 1505 |
-
"eval_accuracy": 0.
|
| 1506 |
-
"eval_loss": 0.
|
| 1507 |
-
"eval_runtime":
|
| 1508 |
-
"eval_samples_per_second":
|
| 1509 |
-
"eval_steps_per_second":
|
| 1510 |
"step": 1900
|
| 1511 |
},
|
| 1512 |
{
|
| 1513 |
-
"epoch":
|
| 1514 |
-
"grad_norm":
|
| 1515 |
-
"learning_rate":
|
| 1516 |
-
"loss": 0.
|
| 1517 |
"step": 1910
|
| 1518 |
},
|
| 1519 |
{
|
| 1520 |
-
"epoch":
|
| 1521 |
-
"grad_norm": 2.
|
| 1522 |
-
"learning_rate":
|
| 1523 |
-
"loss": 0.
|
| 1524 |
"step": 1920
|
| 1525 |
},
|
| 1526 |
{
|
| 1527 |
-
"epoch":
|
| 1528 |
-
"grad_norm":
|
| 1529 |
-
"learning_rate":
|
| 1530 |
-
"loss": 0.
|
| 1531 |
"step": 1930
|
| 1532 |
},
|
| 1533 |
{
|
| 1534 |
-
"epoch":
|
| 1535 |
-
"grad_norm":
|
| 1536 |
-
"learning_rate":
|
| 1537 |
-
"loss": 0.
|
| 1538 |
"step": 1940
|
| 1539 |
},
|
| 1540 |
{
|
| 1541 |
-
"epoch":
|
| 1542 |
-
"grad_norm": 3.
|
| 1543 |
-
"learning_rate":
|
| 1544 |
-
"loss": 0.
|
| 1545 |
"step": 1950
|
| 1546 |
},
|
| 1547 |
{
|
| 1548 |
-
"epoch":
|
| 1549 |
-
"grad_norm":
|
| 1550 |
-
"learning_rate":
|
| 1551 |
-
"loss": 0.
|
| 1552 |
"step": 1960
|
| 1553 |
},
|
| 1554 |
{
|
| 1555 |
-
"epoch":
|
| 1556 |
-
"grad_norm":
|
| 1557 |
-
"learning_rate":
|
| 1558 |
-
"loss": 0.
|
| 1559 |
"step": 1970
|
| 1560 |
},
|
| 1561 |
{
|
| 1562 |
-
"epoch":
|
| 1563 |
-
"grad_norm":
|
| 1564 |
-
"learning_rate":
|
| 1565 |
-
"loss": 0.
|
| 1566 |
"step": 1980
|
| 1567 |
},
|
| 1568 |
{
|
| 1569 |
-
"epoch":
|
| 1570 |
-
"grad_norm":
|
| 1571 |
-
"learning_rate":
|
| 1572 |
-
"loss": 0.
|
| 1573 |
"step": 1990
|
| 1574 |
},
|
| 1575 |
{
|
| 1576 |
-
"epoch":
|
| 1577 |
-
"grad_norm":
|
| 1578 |
-
"learning_rate":
|
| 1579 |
-
"loss": 0.
|
| 1580 |
"step": 2000
|
| 1581 |
},
|
| 1582 |
{
|
| 1583 |
-
"epoch":
|
| 1584 |
-
"eval_accuracy": 0.
|
| 1585 |
-
"eval_loss": 0.
|
| 1586 |
-
"eval_runtime":
|
| 1587 |
-
"eval_samples_per_second":
|
| 1588 |
-
"eval_steps_per_second":
|
| 1589 |
"step": 2000
|
| 1590 |
},
|
| 1591 |
{
|
| 1592 |
-
"epoch":
|
| 1593 |
-
"grad_norm":
|
| 1594 |
-
"learning_rate":
|
| 1595 |
-
"loss": 0.
|
| 1596 |
"step": 2010
|
| 1597 |
},
|
| 1598 |
{
|
| 1599 |
-
"epoch":
|
| 1600 |
-
"grad_norm":
|
| 1601 |
-
"learning_rate":
|
| 1602 |
-
"loss": 0.
|
| 1603 |
"step": 2020
|
| 1604 |
},
|
| 1605 |
{
|
| 1606 |
-
"epoch": 7.
|
| 1607 |
-
"grad_norm":
|
| 1608 |
-
"learning_rate": 2.
|
| 1609 |
-
"loss": 0.
|
| 1610 |
"step": 2030
|
| 1611 |
},
|
| 1612 |
{
|
| 1613 |
-
"epoch": 7.
|
| 1614 |
-
"grad_norm":
|
| 1615 |
-
"learning_rate": 2.
|
| 1616 |
-
"loss": 0.
|
| 1617 |
"step": 2040
|
| 1618 |
},
|
| 1619 |
{
|
| 1620 |
-
"epoch": 7.
|
| 1621 |
-
"grad_norm":
|
| 1622 |
-
"learning_rate": 2.
|
| 1623 |
-
"loss": 0.
|
| 1624 |
"step": 2050
|
| 1625 |
},
|
| 1626 |
{
|
| 1627 |
-
"epoch": 7.
|
| 1628 |
-
"grad_norm":
|
| 1629 |
-
"learning_rate": 2.
|
| 1630 |
-
"loss": 0.
|
| 1631 |
"step": 2060
|
| 1632 |
},
|
| 1633 |
{
|
| 1634 |
-
"epoch": 7.
|
| 1635 |
-
"grad_norm":
|
| 1636 |
-
"learning_rate": 2.
|
| 1637 |
-
"loss": 0.
|
| 1638 |
"step": 2070
|
| 1639 |
},
|
| 1640 |
{
|
| 1641 |
-
"epoch": 7.
|
| 1642 |
-
"grad_norm": 7.
|
| 1643 |
-
"learning_rate": 2.
|
| 1644 |
-
"loss": 0.
|
| 1645 |
"step": 2080
|
| 1646 |
},
|
| 1647 |
{
|
| 1648 |
-
"epoch": 7.
|
| 1649 |
-
"grad_norm": 0.
|
| 1650 |
-
"learning_rate": 2.
|
| 1651 |
-
"loss": 0.
|
| 1652 |
"step": 2090
|
| 1653 |
},
|
| 1654 |
{
|
| 1655 |
-
"epoch": 7.
|
| 1656 |
-
"grad_norm":
|
| 1657 |
-
"learning_rate": 2.
|
| 1658 |
-
"loss": 0.
|
| 1659 |
"step": 2100
|
| 1660 |
},
|
| 1661 |
{
|
| 1662 |
-
"epoch": 7.
|
| 1663 |
-
"eval_accuracy": 0.
|
| 1664 |
-
"eval_loss": 0.
|
| 1665 |
-
"eval_runtime":
|
| 1666 |
-
"eval_samples_per_second":
|
| 1667 |
-
"eval_steps_per_second":
|
| 1668 |
"step": 2100
|
| 1669 |
},
|
| 1670 |
{
|
| 1671 |
-
"epoch": 7.
|
| 1672 |
-
"grad_norm":
|
| 1673 |
-
"learning_rate": 2.
|
| 1674 |
-
"loss": 0.
|
| 1675 |
"step": 2110
|
| 1676 |
},
|
| 1677 |
{
|
| 1678 |
-
"epoch": 7.
|
| 1679 |
-
"grad_norm":
|
| 1680 |
-
"learning_rate": 2.
|
| 1681 |
-
"loss": 0.
|
| 1682 |
"step": 2120
|
| 1683 |
},
|
| 1684 |
{
|
| 1685 |
-
"epoch": 7.
|
| 1686 |
-
"grad_norm":
|
| 1687 |
-
"learning_rate": 2.
|
| 1688 |
-
"loss": 0.
|
| 1689 |
"step": 2130
|
| 1690 |
},
|
| 1691 |
{
|
| 1692 |
-
"epoch": 7.
|
| 1693 |
-
"grad_norm":
|
| 1694 |
-
"learning_rate": 2.
|
| 1695 |
-
"loss": 0.
|
| 1696 |
"step": 2140
|
| 1697 |
},
|
| 1698 |
{
|
| 1699 |
-
"epoch": 7.
|
| 1700 |
-
"grad_norm": 2.
|
| 1701 |
-
"learning_rate": 2.
|
| 1702 |
-
"loss": 0.
|
| 1703 |
"step": 2150
|
| 1704 |
},
|
| 1705 |
{
|
| 1706 |
-
"epoch": 7.
|
| 1707 |
-
"grad_norm":
|
| 1708 |
-
"learning_rate": 2.
|
| 1709 |
-
"loss": 0.
|
| 1710 |
"step": 2160
|
| 1711 |
},
|
| 1712 |
{
|
| 1713 |
-
"epoch": 7.
|
| 1714 |
-
"grad_norm":
|
| 1715 |
-
"learning_rate": 2.
|
| 1716 |
-
"loss": 0.
|
| 1717 |
"step": 2170
|
| 1718 |
},
|
| 1719 |
{
|
| 1720 |
-
"epoch":
|
| 1721 |
-
"grad_norm":
|
| 1722 |
-
"learning_rate":
|
| 1723 |
-
"loss": 0.
|
| 1724 |
"step": 2180
|
| 1725 |
},
|
| 1726 |
{
|
| 1727 |
-
"epoch":
|
| 1728 |
-
"grad_norm":
|
| 1729 |
-
"learning_rate":
|
| 1730 |
-
"loss": 0.
|
| 1731 |
"step": 2190
|
| 1732 |
},
|
| 1733 |
{
|
| 1734 |
-
"epoch":
|
| 1735 |
-
"grad_norm":
|
| 1736 |
-
"learning_rate":
|
| 1737 |
-
"loss": 0.
|
| 1738 |
"step": 2200
|
| 1739 |
},
|
| 1740 |
{
|
| 1741 |
-
"epoch":
|
| 1742 |
-
"eval_accuracy": 0.
|
| 1743 |
-
"eval_loss": 0.
|
| 1744 |
-
"eval_runtime":
|
| 1745 |
-
"eval_samples_per_second":
|
| 1746 |
-
"eval_steps_per_second":
|
| 1747 |
"step": 2200
|
| 1748 |
},
|
| 1749 |
{
|
| 1750 |
-
"epoch":
|
| 1751 |
-
"grad_norm":
|
| 1752 |
-
"learning_rate":
|
| 1753 |
-
"loss": 0.
|
| 1754 |
"step": 2210
|
| 1755 |
},
|
| 1756 |
{
|
| 1757 |
-
"epoch":
|
| 1758 |
-
"grad_norm": 6.
|
| 1759 |
-
"learning_rate":
|
| 1760 |
-
"loss": 0.
|
| 1761 |
"step": 2220
|
| 1762 |
},
|
| 1763 |
{
|
| 1764 |
-
"epoch":
|
| 1765 |
-
"grad_norm":
|
| 1766 |
-
"learning_rate":
|
| 1767 |
-
"loss": 0.
|
| 1768 |
"step": 2230
|
| 1769 |
},
|
| 1770 |
{
|
| 1771 |
-
"epoch":
|
| 1772 |
-
"grad_norm":
|
| 1773 |
-
"learning_rate":
|
| 1774 |
-
"loss": 0.
|
| 1775 |
"step": 2240
|
| 1776 |
},
|
| 1777 |
{
|
| 1778 |
-
"epoch":
|
| 1779 |
-
"grad_norm": 3.
|
| 1780 |
-
"learning_rate":
|
| 1781 |
-
"loss": 0.
|
| 1782 |
"step": 2250
|
| 1783 |
},
|
| 1784 |
{
|
| 1785 |
-
"epoch":
|
| 1786 |
-
"grad_norm":
|
| 1787 |
-
"learning_rate":
|
| 1788 |
-
"loss": 0.
|
| 1789 |
"step": 2260
|
| 1790 |
},
|
| 1791 |
{
|
| 1792 |
-
"epoch":
|
| 1793 |
-
"grad_norm":
|
| 1794 |
-
"learning_rate":
|
| 1795 |
-
"loss": 0.
|
| 1796 |
"step": 2270
|
| 1797 |
},
|
| 1798 |
{
|
| 1799 |
-
"epoch":
|
| 1800 |
-
"grad_norm":
|
| 1801 |
-
"learning_rate":
|
| 1802 |
-
"loss": 0.
|
| 1803 |
"step": 2280
|
| 1804 |
},
|
| 1805 |
{
|
| 1806 |
-
"epoch":
|
| 1807 |
-
"grad_norm":
|
| 1808 |
-
"learning_rate":
|
| 1809 |
-
"loss": 0.
|
| 1810 |
"step": 2290
|
| 1811 |
},
|
| 1812 |
{
|
| 1813 |
-
"epoch":
|
| 1814 |
-
"grad_norm":
|
| 1815 |
-
"learning_rate":
|
| 1816 |
-
"loss": 0.
|
| 1817 |
"step": 2300
|
| 1818 |
},
|
| 1819 |
{
|
| 1820 |
-
"epoch":
|
| 1821 |
-
"eval_accuracy": 0.
|
| 1822 |
-
"eval_loss": 0.
|
| 1823 |
-
"eval_runtime":
|
| 1824 |
-
"eval_samples_per_second":
|
| 1825 |
-
"eval_steps_per_second":
|
| 1826 |
"step": 2300
|
| 1827 |
},
|
| 1828 |
{
|
| 1829 |
-
"epoch":
|
| 1830 |
-
"grad_norm":
|
| 1831 |
-
"learning_rate":
|
| 1832 |
-
"loss": 0.
|
| 1833 |
"step": 2310
|
| 1834 |
},
|
| 1835 |
{
|
| 1836 |
-
"epoch": 8.
|
| 1837 |
-
"grad_norm":
|
| 1838 |
-
"learning_rate": 1.
|
| 1839 |
-
"loss": 0.
|
| 1840 |
"step": 2320
|
| 1841 |
},
|
| 1842 |
{
|
| 1843 |
-
"epoch": 8.
|
| 1844 |
-
"grad_norm":
|
| 1845 |
-
"learning_rate": 1.
|
| 1846 |
-
"loss": 0.
|
| 1847 |
"step": 2330
|
| 1848 |
},
|
| 1849 |
{
|
| 1850 |
-
"epoch": 8.
|
| 1851 |
-
"grad_norm":
|
| 1852 |
-
"learning_rate": 1.
|
| 1853 |
-
"loss": 0.
|
| 1854 |
"step": 2340
|
| 1855 |
},
|
| 1856 |
{
|
| 1857 |
-
"epoch": 8.
|
| 1858 |
-
"grad_norm":
|
| 1859 |
-
"learning_rate": 1.
|
| 1860 |
-
"loss": 0.
|
| 1861 |
"step": 2350
|
| 1862 |
},
|
| 1863 |
{
|
| 1864 |
-
"epoch": 8.
|
| 1865 |
-
"grad_norm":
|
| 1866 |
-
"learning_rate": 1.
|
| 1867 |
-
"loss": 0.
|
| 1868 |
"step": 2360
|
| 1869 |
},
|
| 1870 |
{
|
| 1871 |
-
"epoch": 8.
|
| 1872 |
-
"grad_norm": 6.
|
| 1873 |
-
"learning_rate": 1.
|
| 1874 |
-
"loss": 0.
|
| 1875 |
"step": 2370
|
| 1876 |
},
|
| 1877 |
{
|
| 1878 |
-
"epoch": 8.
|
| 1879 |
-
"grad_norm":
|
| 1880 |
-
"learning_rate": 1.
|
| 1881 |
-
"loss": 0.
|
| 1882 |
"step": 2380
|
| 1883 |
},
|
| 1884 |
{
|
| 1885 |
-
"epoch": 8.
|
| 1886 |
-
"grad_norm":
|
| 1887 |
-
"learning_rate": 1.
|
| 1888 |
-
"loss": 0.
|
| 1889 |
"step": 2390
|
| 1890 |
},
|
| 1891 |
{
|
| 1892 |
-
"epoch": 8.
|
| 1893 |
-
"grad_norm":
|
| 1894 |
-
"learning_rate": 1.
|
| 1895 |
-
"loss": 0.
|
| 1896 |
"step": 2400
|
| 1897 |
},
|
| 1898 |
{
|
| 1899 |
-
"epoch": 8.
|
| 1900 |
-
"eval_accuracy": 0.
|
| 1901 |
-
"eval_loss": 0.
|
| 1902 |
-
"eval_runtime":
|
| 1903 |
-
"eval_samples_per_second":
|
| 1904 |
-
"eval_steps_per_second":
|
| 1905 |
"step": 2400
|
| 1906 |
},
|
| 1907 |
{
|
| 1908 |
-
"epoch": 8.
|
| 1909 |
-
"grad_norm":
|
| 1910 |
-
"learning_rate": 1.
|
| 1911 |
-
"loss": 0.
|
| 1912 |
"step": 2410
|
| 1913 |
},
|
| 1914 |
{
|
| 1915 |
-
"epoch": 8.
|
| 1916 |
-
"grad_norm":
|
| 1917 |
-
"learning_rate": 1.
|
| 1918 |
-
"loss": 0.
|
| 1919 |
"step": 2420
|
| 1920 |
},
|
| 1921 |
{
|
| 1922 |
-
"epoch": 8.
|
| 1923 |
-
"grad_norm":
|
| 1924 |
-
"learning_rate": 1.
|
| 1925 |
-
"loss": 0.
|
| 1926 |
"step": 2430
|
| 1927 |
},
|
| 1928 |
{
|
| 1929 |
-
"epoch": 8.
|
| 1930 |
-
"grad_norm":
|
| 1931 |
-
"learning_rate": 1.
|
| 1932 |
-
"loss": 0.
|
| 1933 |
"step": 2440
|
| 1934 |
},
|
| 1935 |
{
|
| 1936 |
-
"epoch":
|
| 1937 |
-
"grad_norm":
|
| 1938 |
-
"learning_rate": 1.
|
| 1939 |
-
"loss": 0.
|
| 1940 |
"step": 2450
|
| 1941 |
},
|
| 1942 |
{
|
| 1943 |
-
"epoch":
|
| 1944 |
-
"grad_norm":
|
| 1945 |
-
"learning_rate":
|
| 1946 |
-
"loss": 0.
|
| 1947 |
"step": 2460
|
| 1948 |
},
|
| 1949 |
{
|
| 1950 |
-
"epoch":
|
| 1951 |
-
"grad_norm":
|
| 1952 |
-
"learning_rate":
|
| 1953 |
-
"loss": 0.
|
| 1954 |
"step": 2470
|
| 1955 |
},
|
| 1956 |
{
|
| 1957 |
-
"epoch":
|
| 1958 |
-
"grad_norm":
|
| 1959 |
-
"learning_rate":
|
| 1960 |
-
"loss": 0.
|
| 1961 |
"step": 2480
|
| 1962 |
},
|
| 1963 |
{
|
| 1964 |
-
"epoch":
|
| 1965 |
-
"grad_norm":
|
| 1966 |
-
"learning_rate":
|
| 1967 |
-
"loss": 0.
|
| 1968 |
"step": 2490
|
| 1969 |
},
|
| 1970 |
{
|
| 1971 |
-
"epoch":
|
| 1972 |
-
"grad_norm":
|
| 1973 |
-
"learning_rate":
|
| 1974 |
-
"loss": 0.
|
| 1975 |
"step": 2500
|
| 1976 |
},
|
| 1977 |
{
|
| 1978 |
-
"epoch":
|
| 1979 |
-
"eval_accuracy": 0.
|
| 1980 |
-
"eval_loss": 0.
|
| 1981 |
-
"eval_runtime":
|
| 1982 |
-
"eval_samples_per_second":
|
| 1983 |
-
"eval_steps_per_second":
|
| 1984 |
"step": 2500
|
| 1985 |
},
|
| 1986 |
{
|
| 1987 |
-
"epoch":
|
| 1988 |
-
"grad_norm":
|
| 1989 |
-
"learning_rate":
|
| 1990 |
-
"loss": 0.
|
| 1991 |
"step": 2510
|
| 1992 |
},
|
| 1993 |
{
|
| 1994 |
-
"epoch":
|
| 1995 |
-
"grad_norm":
|
| 1996 |
-
"learning_rate":
|
| 1997 |
-
"loss": 0.
|
| 1998 |
"step": 2520
|
| 1999 |
},
|
| 2000 |
{
|
| 2001 |
-
"epoch":
|
| 2002 |
-
"grad_norm":
|
| 2003 |
-
"learning_rate":
|
| 2004 |
-
"loss": 0.
|
| 2005 |
"step": 2530
|
| 2006 |
},
|
| 2007 |
{
|
| 2008 |
-
"epoch":
|
| 2009 |
-
"grad_norm":
|
| 2010 |
-
"learning_rate":
|
| 2011 |
-
"loss": 0.
|
| 2012 |
"step": 2540
|
| 2013 |
},
|
| 2014 |
{
|
| 2015 |
-
"epoch":
|
| 2016 |
-
"grad_norm":
|
| 2017 |
-
"learning_rate":
|
| 2018 |
-
"loss": 0.
|
| 2019 |
"step": 2550
|
| 2020 |
},
|
| 2021 |
{
|
| 2022 |
-
"epoch":
|
| 2023 |
-
"grad_norm":
|
| 2024 |
-
"learning_rate":
|
| 2025 |
-
"loss": 0.
|
| 2026 |
"step": 2560
|
| 2027 |
},
|
| 2028 |
{
|
| 2029 |
-
"epoch":
|
| 2030 |
-
"grad_norm": 4.
|
| 2031 |
-
"learning_rate":
|
| 2032 |
-
"loss": 0.
|
| 2033 |
"step": 2570
|
| 2034 |
},
|
| 2035 |
{
|
| 2036 |
-
"epoch":
|
| 2037 |
-
"grad_norm":
|
| 2038 |
-
"learning_rate":
|
| 2039 |
-
"loss": 0.
|
| 2040 |
"step": 2580
|
| 2041 |
},
|
| 2042 |
{
|
| 2043 |
-
"epoch":
|
| 2044 |
-
"grad_norm":
|
| 2045 |
-
"learning_rate":
|
| 2046 |
-
"loss": 0.
|
| 2047 |
"step": 2590
|
| 2048 |
},
|
| 2049 |
{
|
| 2050 |
-
"epoch": 9.
|
| 2051 |
-
"grad_norm":
|
| 2052 |
-
"learning_rate":
|
| 2053 |
-
"loss": 0.
|
| 2054 |
"step": 2600
|
| 2055 |
},
|
| 2056 |
{
|
| 2057 |
-
"epoch": 9.
|
| 2058 |
-
"eval_accuracy": 0.
|
| 2059 |
-
"eval_loss": 0.
|
| 2060 |
-
"eval_runtime":
|
| 2061 |
-
"eval_samples_per_second":
|
| 2062 |
-
"eval_steps_per_second":
|
| 2063 |
"step": 2600
|
| 2064 |
},
|
| 2065 |
{
|
| 2066 |
-
"epoch": 9.
|
| 2067 |
-
"grad_norm":
|
| 2068 |
-
"learning_rate":
|
| 2069 |
-
"loss": 0.
|
| 2070 |
"step": 2610
|
| 2071 |
},
|
| 2072 |
{
|
| 2073 |
-
"epoch": 9.
|
| 2074 |
-
"grad_norm":
|
| 2075 |
-
"learning_rate":
|
| 2076 |
-
"loss": 0.
|
| 2077 |
"step": 2620
|
| 2078 |
},
|
| 2079 |
{
|
| 2080 |
-
"epoch": 9.
|
| 2081 |
-
"grad_norm": 0.
|
| 2082 |
-
"learning_rate":
|
| 2083 |
-
"loss": 0.
|
| 2084 |
"step": 2630
|
| 2085 |
},
|
| 2086 |
{
|
| 2087 |
-
"epoch": 9.
|
| 2088 |
-
"grad_norm":
|
| 2089 |
-
"learning_rate":
|
| 2090 |
-
"loss": 0.
|
| 2091 |
"step": 2640
|
| 2092 |
},
|
| 2093 |
{
|
| 2094 |
-
"epoch": 9.
|
| 2095 |
-
"grad_norm": 1.
|
| 2096 |
-
"learning_rate":
|
| 2097 |
-
"loss": 0.
|
| 2098 |
"step": 2650
|
| 2099 |
},
|
| 2100 |
{
|
| 2101 |
-
"epoch": 9.
|
| 2102 |
-
"grad_norm":
|
| 2103 |
-
"learning_rate":
|
| 2104 |
-
"loss": 0.
|
| 2105 |
"step": 2660
|
| 2106 |
},
|
| 2107 |
{
|
| 2108 |
-
"epoch": 9.
|
| 2109 |
-
"grad_norm":
|
| 2110 |
-
"learning_rate":
|
| 2111 |
-
"loss": 0.
|
| 2112 |
"step": 2670
|
| 2113 |
},
|
| 2114 |
{
|
| 2115 |
-
"epoch": 9.
|
| 2116 |
-
"grad_norm":
|
| 2117 |
-
"learning_rate":
|
| 2118 |
-
"loss": 0.
|
| 2119 |
"step": 2680
|
| 2120 |
},
|
| 2121 |
{
|
| 2122 |
-
"epoch": 9.
|
| 2123 |
-
"grad_norm": 0.
|
| 2124 |
-
"learning_rate":
|
| 2125 |
-
"loss": 0.
|
| 2126 |
"step": 2690
|
| 2127 |
},
|
| 2128 |
{
|
| 2129 |
-
"epoch": 9.
|
| 2130 |
-
"grad_norm":
|
| 2131 |
-
"learning_rate":
|
| 2132 |
-
"loss": 0.
|
| 2133 |
"step": 2700
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
-
"epoch": 9.
|
| 2137 |
-
"eval_accuracy": 0.
|
| 2138 |
-
"eval_loss": 0.
|
| 2139 |
-
"eval_runtime":
|
| 2140 |
-
"eval_samples_per_second":
|
| 2141 |
-
"eval_steps_per_second":
|
| 2142 |
"step": 2700
|
| 2143 |
},
|
| 2144 |
{
|
| 2145 |
-
"epoch": 9.
|
| 2146 |
-
"grad_norm":
|
| 2147 |
-
"learning_rate":
|
| 2148 |
-
"loss": 0.
|
| 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":
|
| 2282 |
-
"learning_rate": 1.
|
| 2283 |
-
"loss": 0.
|
| 2284 |
-
"step":
|
| 2285 |
},
|
| 2286 |
{
|
| 2287 |
"epoch": 10.0,
|
| 2288 |
-
"step":
|
| 2289 |
-
"total_flos": 3.
|
| 2290 |
-
"train_loss": 0.
|
| 2291 |
-
"train_runtime":
|
| 2292 |
-
"train_samples_per_second": 28.
|
| 2293 |
-
"train_steps_per_second": 1.
|
| 2294 |
}
|
| 2295 |
],
|
| 2296 |
"logging_steps": 10,
|
| 2297 |
-
"max_steps":
|
| 2298 |
"num_input_tokens_seen": 0,
|
| 2299 |
"num_train_epochs": 10,
|
| 2300 |
"save_steps": 100,
|
| 2301 |
-
"total_flos": 3.
|
| 2302 |
"train_batch_size": 16,
|
| 2303 |
"trial_name": null,
|
| 2304 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.4375220239162445,
|
| 3 |
+
"best_model_checkpoint": "Action_all_10_class/checkpoint-1800",
|
| 4 |
"epoch": 10.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 2720,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"grad_norm": 1.6838732957839966,
|
| 14 |
+
"learning_rate": 9.963235294117647e-05,
|
| 15 |
+
"loss": 2.2569,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.07,
|
| 20 |
+
"grad_norm": 1.8766860961914062,
|
| 21 |
+
"learning_rate": 9.926470588235295e-05,
|
| 22 |
+
"loss": 2.1701,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.11,
|
| 27 |
+
"grad_norm": 1.974852204322815,
|
| 28 |
+
"learning_rate": 9.889705882352942e-05,
|
| 29 |
+
"loss": 2.0385,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.15,
|
| 34 |
+
"grad_norm": 1.9496526718139648,
|
| 35 |
+
"learning_rate": 9.852941176470589e-05,
|
| 36 |
+
"loss": 1.8707,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.18,
|
| 41 |
+
"grad_norm": 2.7034614086151123,
|
| 42 |
+
"learning_rate": 9.816176470588235e-05,
|
| 43 |
+
"loss": 1.7473,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.22,
|
| 48 |
+
"grad_norm": 2.373265266418457,
|
| 49 |
+
"learning_rate": 9.779411764705882e-05,
|
| 50 |
+
"loss": 1.5826,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.26,
|
| 55 |
+
"grad_norm": 2.903897285461426,
|
| 56 |
+
"learning_rate": 9.74264705882353e-05,
|
| 57 |
+
"loss": 1.4941,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.29,
|
| 62 |
+
"grad_norm": 3.3323962688446045,
|
| 63 |
+
"learning_rate": 9.705882352941177e-05,
|
| 64 |
+
"loss": 1.4106,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.33,
|
| 69 |
+
"grad_norm": 2.146582841873169,
|
| 70 |
+
"learning_rate": 9.669117647058824e-05,
|
| 71 |
+
"loss": 1.2607,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.37,
|
| 76 |
+
"grad_norm": 3.2986233234405518,
|
| 77 |
+
"learning_rate": 9.632352941176472e-05,
|
| 78 |
+
"loss": 1.3206,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.37,
|
| 83 |
+
"eval_accuracy": 0.7387304507819688,
|
| 84 |
+
"eval_loss": 1.1437705755233765,
|
| 85 |
+
"eval_runtime": 17.686,
|
| 86 |
+
"eval_samples_per_second": 61.461,
|
| 87 |
+
"eval_steps_per_second": 7.69,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"epoch": 0.4,
|
| 92 |
+
"grad_norm": 4.9779276847839355,
|
| 93 |
+
"learning_rate": 9.595588235294119e-05,
|
| 94 |
+
"loss": 1.2119,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 0.44,
|
| 99 |
+
"grad_norm": 2.838693141937256,
|
| 100 |
+
"learning_rate": 9.558823529411765e-05,
|
| 101 |
+
"loss": 1.1979,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 0.48,
|
| 106 |
+
"grad_norm": 3.1596128940582275,
|
| 107 |
+
"learning_rate": 9.522058823529412e-05,
|
| 108 |
+
"loss": 1.1736,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 0.51,
|
| 113 |
+
"grad_norm": 3.491159439086914,
|
| 114 |
+
"learning_rate": 9.485294117647059e-05,
|
| 115 |
+
"loss": 1.1305,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.55,
|
| 120 |
+
"grad_norm": 2.965158224105835,
|
| 121 |
+
"learning_rate": 9.448529411764707e-05,
|
| 122 |
+
"loss": 1.1087,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 0.59,
|
| 127 |
+
"grad_norm": 3.669126033782959,
|
| 128 |
+
"learning_rate": 9.411764705882353e-05,
|
| 129 |
+
"loss": 1.1046,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"epoch": 0.62,
|
| 134 |
+
"grad_norm": 3.2335517406463623,
|
| 135 |
+
"learning_rate": 9.375e-05,
|
| 136 |
+
"loss": 1.0291,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 0.66,
|
| 141 |
+
"grad_norm": 2.1661763191223145,
|
| 142 |
+
"learning_rate": 9.338235294117648e-05,
|
| 143 |
+
"loss": 1.0999,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 0.7,
|
| 148 |
+
"grad_norm": 2.994340419769287,
|
| 149 |
+
"learning_rate": 9.30514705882353e-05,
|
| 150 |
+
"loss": 1.0534,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
+
"epoch": 0.74,
|
| 155 |
+
"grad_norm": 3.946995258331299,
|
| 156 |
+
"learning_rate": 9.268382352941177e-05,
|
| 157 |
+
"loss": 0.9631,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.74,
|
| 162 |
+
"eval_accuracy": 0.7893284268629255,
|
| 163 |
+
"eval_loss": 0.8203188180923462,
|
| 164 |
+
"eval_runtime": 14.7459,
|
| 165 |
+
"eval_samples_per_second": 73.715,
|
| 166 |
+
"eval_steps_per_second": 9.223,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 0.77,
|
| 171 |
+
"grad_norm": 7.194287300109863,
|
| 172 |
+
"learning_rate": 9.231617647058824e-05,
|
| 173 |
+
"loss": 0.9835,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 0.81,
|
| 178 |
+
"grad_norm": 3.7055001258850098,
|
| 179 |
+
"learning_rate": 9.19485294117647e-05,
|
| 180 |
+
"loss": 1.0033,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 0.85,
|
| 185 |
+
"grad_norm": 7.64039945602417,
|
| 186 |
+
"learning_rate": 9.158088235294117e-05,
|
| 187 |
+
"loss": 0.9074,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.88,
|
| 192 |
+
"grad_norm": 3.5172407627105713,
|
| 193 |
+
"learning_rate": 9.121323529411765e-05,
|
| 194 |
+
"loss": 0.9709,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 0.92,
|
| 199 |
+
"grad_norm": 2.5609283447265625,
|
| 200 |
+
"learning_rate": 9.084558823529412e-05,
|
| 201 |
+
"loss": 0.7347,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 0.96,
|
| 206 |
+
"grad_norm": 5.291449069976807,
|
| 207 |
+
"learning_rate": 9.047794117647059e-05,
|
| 208 |
+
"loss": 0.77,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 0.99,
|
| 213 |
+
"grad_norm": 2.085615396499634,
|
| 214 |
+
"learning_rate": 9.011029411764707e-05,
|
| 215 |
+
"loss": 0.8655,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 1.03,
|
| 220 |
+
"grad_norm": 5.507621765136719,
|
| 221 |
+
"learning_rate": 8.974264705882354e-05,
|
| 222 |
+
"loss": 0.7275,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 1.07,
|
| 227 |
+
"grad_norm": 3.2387466430664062,
|
| 228 |
+
"learning_rate": 8.9375e-05,
|
| 229 |
+
"loss": 0.6008,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 1.1,
|
| 234 |
+
"grad_norm": 4.184783935546875,
|
| 235 |
+
"learning_rate": 8.900735294117647e-05,
|
| 236 |
+
"loss": 0.6572,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 1.1,
|
| 241 |
+
"eval_accuracy": 0.8104875804967802,
|
| 242 |
+
"eval_loss": 0.6729719638824463,
|
| 243 |
+
"eval_runtime": 14.7051,
|
| 244 |
+
"eval_samples_per_second": 73.92,
|
| 245 |
+
"eval_steps_per_second": 9.249,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 1.14,
|
| 250 |
+
"grad_norm": 3.7986650466918945,
|
| 251 |
+
"learning_rate": 8.863970588235294e-05,
|
| 252 |
+
"loss": 0.8916,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 1.18,
|
| 257 |
+
"grad_norm": 4.8309221267700195,
|
| 258 |
+
"learning_rate": 8.827205882352942e-05,
|
| 259 |
+
"loss": 0.8594,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 1.21,
|
| 264 |
+
"grad_norm": 4.573574542999268,
|
| 265 |
+
"learning_rate": 8.790441176470589e-05,
|
| 266 |
+
"loss": 0.8141,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 1.25,
|
| 271 |
+
"grad_norm": 10.174393653869629,
|
| 272 |
+
"learning_rate": 8.753676470588235e-05,
|
| 273 |
+
"loss": 0.7751,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 1.29,
|
| 278 |
+
"grad_norm": 5.100869655609131,
|
| 279 |
+
"learning_rate": 8.716911764705883e-05,
|
| 280 |
+
"loss": 0.7263,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 1.32,
|
| 285 |
+
"grad_norm": 5.836120128631592,
|
| 286 |
+
"learning_rate": 8.68014705882353e-05,
|
| 287 |
+
"loss": 0.7111,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 1.36,
|
| 292 |
+
"grad_norm": 4.5023956298828125,
|
| 293 |
+
"learning_rate": 8.643382352941177e-05,
|
| 294 |
+
"loss": 0.7607,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
+
"epoch": 1.4,
|
| 299 |
+
"grad_norm": 7.608530521392822,
|
| 300 |
+
"learning_rate": 8.606617647058825e-05,
|
| 301 |
+
"loss": 0.6117,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 1.43,
|
| 306 |
+
"grad_norm": 4.743827819824219,
|
| 307 |
+
"learning_rate": 8.56985294117647e-05,
|
| 308 |
+
"loss": 0.7964,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 1.47,
|
| 313 |
+
"grad_norm": 1.817112684249878,
|
| 314 |
+
"learning_rate": 8.533088235294117e-05,
|
| 315 |
+
"loss": 0.629,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
+
"epoch": 1.47,
|
| 320 |
+
"eval_accuracy": 0.8334866605335787,
|
| 321 |
+
"eval_loss": 0.5896809101104736,
|
| 322 |
+
"eval_runtime": 14.6784,
|
| 323 |
+
"eval_samples_per_second": 74.054,
|
| 324 |
+
"eval_steps_per_second": 9.265,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 1.51,
|
| 329 |
+
"grad_norm": 4.885681629180908,
|
| 330 |
+
"learning_rate": 8.496323529411765e-05,
|
| 331 |
+
"loss": 0.7329,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 1.54,
|
| 336 |
+
"grad_norm": 3.4335737228393555,
|
| 337 |
+
"learning_rate": 8.459558823529412e-05,
|
| 338 |
+
"loss": 0.5917,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 1.58,
|
| 343 |
+
"grad_norm": 5.689188480377197,
|
| 344 |
+
"learning_rate": 8.42279411764706e-05,
|
| 345 |
+
"loss": 0.6292,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
+
"epoch": 1.62,
|
| 350 |
+
"grad_norm": 4.069520473480225,
|
| 351 |
+
"learning_rate": 8.386029411764707e-05,
|
| 352 |
+
"loss": 0.679,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 1.65,
|
| 357 |
+
"grad_norm": 4.512510776519775,
|
| 358 |
+
"learning_rate": 8.349264705882353e-05,
|
| 359 |
+
"loss": 0.74,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 1.69,
|
| 364 |
+
"grad_norm": 4.557438373565674,
|
| 365 |
+
"learning_rate": 8.312500000000001e-05,
|
| 366 |
+
"loss": 0.7421,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 1.73,
|
| 371 |
+
"grad_norm": 5.320840835571289,
|
| 372 |
+
"learning_rate": 8.275735294117647e-05,
|
| 373 |
+
"loss": 0.75,
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 1.76,
|
| 378 |
+
"grad_norm": 3.9637961387634277,
|
| 379 |
+
"learning_rate": 8.238970588235294e-05,
|
| 380 |
+
"loss": 0.6704,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 1.8,
|
| 385 |
+
"grad_norm": 2.7711901664733887,
|
| 386 |
+
"learning_rate": 8.202205882352942e-05,
|
| 387 |
+
"loss": 0.6684,
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
+
"epoch": 1.84,
|
| 392 |
+
"grad_norm": 4.7916579246521,
|
| 393 |
+
"learning_rate": 8.165441176470588e-05,
|
| 394 |
+
"loss": 0.722,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 1.84,
|
| 399 |
+
"eval_accuracy": 0.8491260349586016,
|
| 400 |
+
"eval_loss": 0.5344882607460022,
|
| 401 |
+
"eval_runtime": 14.8731,
|
| 402 |
+
"eval_samples_per_second": 73.085,
|
| 403 |
+
"eval_steps_per_second": 9.144,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 1.88,
|
| 408 |
+
"grad_norm": 4.628171920776367,
|
| 409 |
+
"learning_rate": 8.128676470588235e-05,
|
| 410 |
+
"loss": 0.6098,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 1.91,
|
| 415 |
+
"grad_norm": 4.231683254241943,
|
| 416 |
+
"learning_rate": 8.091911764705883e-05,
|
| 417 |
+
"loss": 0.5521,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
+
"epoch": 1.95,
|
| 422 |
+
"grad_norm": 2.9706532955169678,
|
| 423 |
+
"learning_rate": 8.058823529411765e-05,
|
| 424 |
+
"loss": 0.7859,
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 1.99,
|
| 429 |
+
"grad_norm": 10.902141571044922,
|
| 430 |
+
"learning_rate": 8.022058823529412e-05,
|
| 431 |
+
"loss": 0.7419,
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
+
"epoch": 2.02,
|
| 436 |
+
"grad_norm": 2.9177191257476807,
|
| 437 |
+
"learning_rate": 7.98529411764706e-05,
|
| 438 |
+
"loss": 0.5061,
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
+
"epoch": 2.06,
|
| 443 |
+
"grad_norm": 3.3445937633514404,
|
| 444 |
+
"learning_rate": 7.948529411764706e-05,
|
| 445 |
+
"loss": 0.5721,
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 2.1,
|
| 450 |
+
"grad_norm": 7.377439498901367,
|
| 451 |
+
"learning_rate": 7.911764705882354e-05,
|
| 452 |
+
"loss": 0.581,
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"epoch": 2.13,
|
| 457 |
+
"grad_norm": 5.139944076538086,
|
| 458 |
+
"learning_rate": 7.875e-05,
|
| 459 |
+
"loss": 0.559,
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
+
"epoch": 2.17,
|
| 464 |
+
"grad_norm": 5.972865104675293,
|
| 465 |
+
"learning_rate": 7.838235294117647e-05,
|
| 466 |
+
"loss": 0.5372,
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 2.21,
|
| 471 |
+
"grad_norm": 3.0057387351989746,
|
| 472 |
+
"learning_rate": 7.801470588235295e-05,
|
| 473 |
+
"loss": 0.5069,
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 2.21,
|
| 478 |
+
"eval_accuracy": 0.8472861085556578,
|
| 479 |
+
"eval_loss": 0.5243943929672241,
|
| 480 |
+
"eval_runtime": 14.5207,
|
| 481 |
+
"eval_samples_per_second": 74.858,
|
| 482 |
+
"eval_steps_per_second": 9.366,
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 2.24,
|
| 487 |
+
"grad_norm": 4.579606056213379,
|
| 488 |
+
"learning_rate": 7.764705882352942e-05,
|
| 489 |
+
"loss": 0.5327,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"epoch": 2.28,
|
| 494 |
+
"grad_norm": 2.765660285949707,
|
| 495 |
+
"learning_rate": 7.727941176470589e-05,
|
| 496 |
+
"loss": 0.5983,
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 2.32,
|
| 501 |
+
"grad_norm": 5.36717414855957,
|
| 502 |
+
"learning_rate": 7.691176470588237e-05,
|
| 503 |
+
"loss": 0.5637,
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 2.35,
|
| 508 |
+
"grad_norm": 4.482629776000977,
|
| 509 |
+
"learning_rate": 7.654411764705882e-05,
|
| 510 |
+
"loss": 0.54,
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
+
"epoch": 2.39,
|
| 515 |
+
"grad_norm": 3.6941115856170654,
|
| 516 |
+
"learning_rate": 7.617647058823529e-05,
|
| 517 |
+
"loss": 0.5142,
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 2.43,
|
| 522 |
+
"grad_norm": 3.7526047229766846,
|
| 523 |
+
"learning_rate": 7.580882352941177e-05,
|
| 524 |
+
"loss": 0.5058,
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
+
"epoch": 2.46,
|
| 529 |
+
"grad_norm": 4.210299491882324,
|
| 530 |
+
"learning_rate": 7.544117647058824e-05,
|
| 531 |
+
"loss": 0.5504,
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
+
"epoch": 2.5,
|
| 536 |
+
"grad_norm": 5.4189276695251465,
|
| 537 |
+
"learning_rate": 7.50735294117647e-05,
|
| 538 |
+
"loss": 0.5287,
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 2.54,
|
| 543 |
+
"grad_norm": 4.421909809112549,
|
| 544 |
+
"learning_rate": 7.470588235294118e-05,
|
| 545 |
+
"loss": 0.4464,
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
+
"epoch": 2.57,
|
| 550 |
+
"grad_norm": 5.694902420043945,
|
| 551 |
+
"learning_rate": 7.433823529411765e-05,
|
| 552 |
+
"loss": 0.5711,
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
+
"epoch": 2.57,
|
| 557 |
+
"eval_accuracy": 0.8574057037718491,
|
| 558 |
+
"eval_loss": 0.49967309832572937,
|
| 559 |
+
"eval_runtime": 14.5091,
|
| 560 |
+
"eval_samples_per_second": 74.919,
|
| 561 |
+
"eval_steps_per_second": 9.373,
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 2.61,
|
| 566 |
+
"grad_norm": 3.0326895713806152,
|
| 567 |
+
"learning_rate": 7.397058823529413e-05,
|
| 568 |
+
"loss": 0.5711,
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 2.65,
|
| 573 |
+
"grad_norm": 6.757785320281982,
|
| 574 |
+
"learning_rate": 7.36029411764706e-05,
|
| 575 |
+
"loss": 0.4469,
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 2.68,
|
| 580 |
+
"grad_norm": 6.860453128814697,
|
| 581 |
+
"learning_rate": 7.323529411764705e-05,
|
| 582 |
+
"loss": 0.6197,
|
| 583 |
"step": 730
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 2.72,
|
| 587 |
+
"grad_norm": 1.575364351272583,
|
| 588 |
+
"learning_rate": 7.286764705882353e-05,
|
| 589 |
+
"loss": 0.5648,
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 2.76,
|
| 594 |
+
"grad_norm": 8.490889549255371,
|
| 595 |
+
"learning_rate": 7.25e-05,
|
| 596 |
+
"loss": 0.5884,
|
| 597 |
"step": 750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 2.79,
|
| 601 |
+
"grad_norm": 5.70561408996582,
|
| 602 |
+
"learning_rate": 7.213235294117647e-05,
|
| 603 |
+
"loss": 0.5584,
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 2.83,
|
| 608 |
+
"grad_norm": 4.843407154083252,
|
| 609 |
+
"learning_rate": 7.176470588235295e-05,
|
| 610 |
+
"loss": 0.5252,
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 2.87,
|
| 615 |
+
"grad_norm": 5.284633159637451,
|
| 616 |
+
"learning_rate": 7.139705882352942e-05,
|
| 617 |
+
"loss": 0.5514,
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 2.9,
|
| 622 |
+
"grad_norm": 4.284489631652832,
|
| 623 |
+
"learning_rate": 7.102941176470588e-05,
|
| 624 |
+
"loss": 0.4549,
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 2.94,
|
| 629 |
+
"grad_norm": 4.714504718780518,
|
| 630 |
+
"learning_rate": 7.066176470588236e-05,
|
| 631 |
+
"loss": 0.5162,
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 2.94,
|
| 636 |
+
"eval_accuracy": 0.8472861085556578,
|
| 637 |
+
"eval_loss": 0.5145568251609802,
|
| 638 |
+
"eval_runtime": 14.6797,
|
| 639 |
+
"eval_samples_per_second": 74.048,
|
| 640 |
+
"eval_steps_per_second": 9.264,
|
| 641 |
"step": 800
|
| 642 |
},
|
| 643 |
{
|
| 644 |
+
"epoch": 2.98,
|
| 645 |
+
"grad_norm": 4.9226460456848145,
|
| 646 |
+
"learning_rate": 7.029411764705882e-05,
|
| 647 |
+
"loss": 0.5992,
|
| 648 |
"step": 810
|
| 649 |
},
|
| 650 |
{
|
| 651 |
+
"epoch": 3.01,
|
| 652 |
+
"grad_norm": 3.4406116008758545,
|
| 653 |
+
"learning_rate": 6.99264705882353e-05,
|
| 654 |
+
"loss": 0.3953,
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
+
"epoch": 3.05,
|
| 659 |
+
"grad_norm": 2.7723779678344727,
|
| 660 |
+
"learning_rate": 6.955882352941177e-05,
|
| 661 |
+
"loss": 0.46,
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 3.09,
|
| 666 |
+
"grad_norm": 2.4229788780212402,
|
| 667 |
+
"learning_rate": 6.919117647058823e-05,
|
| 668 |
+
"loss": 0.3365,
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
+
"epoch": 3.12,
|
| 673 |
+
"grad_norm": 5.223120212554932,
|
| 674 |
+
"learning_rate": 6.882352941176471e-05,
|
| 675 |
+
"loss": 0.4607,
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
+
"epoch": 3.16,
|
| 680 |
+
"grad_norm": 6.825634956359863,
|
| 681 |
+
"learning_rate": 6.845588235294118e-05,
|
| 682 |
+
"loss": 0.5047,
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
+
"epoch": 3.2,
|
| 687 |
+
"grad_norm": 7.074869632720947,
|
| 688 |
+
"learning_rate": 6.808823529411765e-05,
|
| 689 |
+
"loss": 0.4205,
|
| 690 |
"step": 870
|
| 691 |
},
|
| 692 |
{
|
| 693 |
+
"epoch": 3.24,
|
| 694 |
+
"grad_norm": 8.000801086425781,
|
| 695 |
+
"learning_rate": 6.772058823529413e-05,
|
| 696 |
+
"loss": 0.4157,
|
| 697 |
"step": 880
|
| 698 |
},
|
| 699 |
{
|
| 700 |
+
"epoch": 3.27,
|
| 701 |
+
"grad_norm": 6.804930210113525,
|
| 702 |
+
"learning_rate": 6.73529411764706e-05,
|
| 703 |
+
"loss": 0.563,
|
| 704 |
"step": 890
|
| 705 |
},
|
| 706 |
{
|
| 707 |
+
"epoch": 3.31,
|
| 708 |
+
"grad_norm": 3.697648048400879,
|
| 709 |
+
"learning_rate": 6.698529411764706e-05,
|
| 710 |
+
"loss": 0.444,
|
| 711 |
"step": 900
|
| 712 |
},
|
| 713 |
{
|
| 714 |
+
"epoch": 3.31,
|
| 715 |
+
"eval_accuracy": 0.8528058877644894,
|
| 716 |
+
"eval_loss": 0.45834270119667053,
|
| 717 |
+
"eval_runtime": 14.7852,
|
| 718 |
+
"eval_samples_per_second": 73.52,
|
| 719 |
+
"eval_steps_per_second": 9.198,
|
| 720 |
"step": 900
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 3.35,
|
| 724 |
+
"grad_norm": 1.375609278678894,
|
| 725 |
+
"learning_rate": 6.661764705882353e-05,
|
| 726 |
+
"loss": 0.399,
|
| 727 |
"step": 910
|
| 728 |
},
|
| 729 |
{
|
| 730 |
+
"epoch": 3.38,
|
| 731 |
+
"grad_norm": 2.9514737129211426,
|
| 732 |
+
"learning_rate": 6.625e-05,
|
| 733 |
+
"loss": 0.5142,
|
| 734 |
"step": 920
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 3.42,
|
| 738 |
+
"grad_norm": 5.124215126037598,
|
| 739 |
+
"learning_rate": 6.588235294117648e-05,
|
| 740 |
+
"loss": 0.4603,
|
| 741 |
"step": 930
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 3.46,
|
| 745 |
+
"grad_norm": 9.651627540588379,
|
| 746 |
+
"learning_rate": 6.551470588235295e-05,
|
| 747 |
+
"loss": 0.3788,
|
| 748 |
"step": 940
|
| 749 |
},
|
| 750 |
{
|
| 751 |
+
"epoch": 3.49,
|
| 752 |
+
"grad_norm": 6.899481296539307,
|
| 753 |
+
"learning_rate": 6.514705882352941e-05,
|
| 754 |
+
"loss": 0.4569,
|
| 755 |
"step": 950
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 3.53,
|
| 759 |
+
"grad_norm": 0.4424718916416168,
|
| 760 |
+
"learning_rate": 6.47794117647059e-05,
|
| 761 |
+
"loss": 0.3358,
|
| 762 |
"step": 960
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 3.57,
|
| 766 |
+
"grad_norm": 4.599928855895996,
|
| 767 |
+
"learning_rate": 6.441176470588236e-05,
|
| 768 |
+
"loss": 0.4226,
|
| 769 |
"step": 970
|
| 770 |
},
|
| 771 |
{
|
| 772 |
+
"epoch": 3.6,
|
| 773 |
+
"grad_norm": 1.4953221082687378,
|
| 774 |
+
"learning_rate": 6.404411764705881e-05,
|
| 775 |
+
"loss": 0.2849,
|
| 776 |
"step": 980
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"epoch": 3.64,
|
| 780 |
+
"grad_norm": 4.287179946899414,
|
| 781 |
+
"learning_rate": 6.36764705882353e-05,
|
| 782 |
+
"loss": 0.416,
|
| 783 |
"step": 990
|
| 784 |
},
|
| 785 |
{
|
| 786 |
+
"epoch": 3.68,
|
| 787 |
+
"grad_norm": 3.138073205947876,
|
| 788 |
+
"learning_rate": 6.330882352941176e-05,
|
| 789 |
+
"loss": 0.3708,
|
| 790 |
"step": 1000
|
| 791 |
},
|
| 792 |
{
|
| 793 |
+
"epoch": 3.68,
|
| 794 |
+
"eval_accuracy": 0.8371665133394665,
|
| 795 |
+
"eval_loss": 0.5439407229423523,
|
| 796 |
+
"eval_runtime": 14.8081,
|
| 797 |
+
"eval_samples_per_second": 73.406,
|
| 798 |
+
"eval_steps_per_second": 9.184,
|
| 799 |
"step": 1000
|
| 800 |
},
|
| 801 |
{
|
| 802 |
+
"epoch": 3.71,
|
| 803 |
+
"grad_norm": 3.8845012187957764,
|
| 804 |
+
"learning_rate": 6.294117647058824e-05,
|
| 805 |
+
"loss": 0.4332,
|
| 806 |
"step": 1010
|
| 807 |
},
|
| 808 |
{
|
| 809 |
+
"epoch": 3.75,
|
| 810 |
+
"grad_norm": 9.119804382324219,
|
| 811 |
+
"learning_rate": 6.257352941176471e-05,
|
| 812 |
+
"loss": 0.341,
|
| 813 |
"step": 1020
|
| 814 |
},
|
| 815 |
{
|
| 816 |
+
"epoch": 3.79,
|
| 817 |
+
"grad_norm": 5.062876224517822,
|
| 818 |
+
"learning_rate": 6.220588235294118e-05,
|
| 819 |
+
"loss": 0.42,
|
| 820 |
"step": 1030
|
| 821 |
},
|
| 822 |
{
|
| 823 |
+
"epoch": 3.82,
|
| 824 |
+
"grad_norm": 3.2720649242401123,
|
| 825 |
+
"learning_rate": 6.183823529411766e-05,
|
| 826 |
+
"loss": 0.4504,
|
| 827 |
"step": 1040
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 3.86,
|
| 831 |
+
"grad_norm": 12.041254043579102,
|
| 832 |
+
"learning_rate": 6.147058823529413e-05,
|
| 833 |
+
"loss": 0.5898,
|
| 834 |
"step": 1050
|
| 835 |
},
|
| 836 |
{
|
| 837 |
+
"epoch": 3.9,
|
| 838 |
+
"grad_norm": 1.3308639526367188,
|
| 839 |
+
"learning_rate": 6.110294117647059e-05,
|
| 840 |
+
"loss": 0.4639,
|
| 841 |
"step": 1060
|
| 842 |
},
|
| 843 |
{
|
| 844 |
+
"epoch": 3.93,
|
| 845 |
+
"grad_norm": 6.7824015617370605,
|
| 846 |
+
"learning_rate": 6.073529411764706e-05,
|
| 847 |
+
"loss": 0.4416,
|
| 848 |
"step": 1070
|
| 849 |
},
|
| 850 |
{
|
| 851 |
+
"epoch": 3.97,
|
| 852 |
+
"grad_norm": 3.1110615730285645,
|
| 853 |
+
"learning_rate": 6.036764705882353e-05,
|
| 854 |
+
"loss": 0.5546,
|
| 855 |
"step": 1080
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"epoch": 4.01,
|
| 859 |
+
"grad_norm": 7.720857620239258,
|
| 860 |
+
"learning_rate": 6e-05,
|
| 861 |
+
"loss": 0.4896,
|
| 862 |
"step": 1090
|
| 863 |
},
|
| 864 |
{
|
| 865 |
+
"epoch": 4.04,
|
| 866 |
+
"grad_norm": 7.00632905960083,
|
| 867 |
+
"learning_rate": 5.9632352941176476e-05,
|
| 868 |
+
"loss": 0.4273,
|
| 869 |
"step": 1100
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 4.04,
|
| 873 |
+
"eval_accuracy": 0.860165593376265,
|
| 874 |
+
"eval_loss": 0.4629393517971039,
|
| 875 |
+
"eval_runtime": 14.7268,
|
| 876 |
+
"eval_samples_per_second": 73.811,
|
| 877 |
+
"eval_steps_per_second": 9.235,
|
| 878 |
"step": 1100
|
| 879 |
},
|
| 880 |
{
|
| 881 |
+
"epoch": 4.08,
|
| 882 |
+
"grad_norm": 3.2143478393554688,
|
| 883 |
+
"learning_rate": 5.926470588235294e-05,
|
| 884 |
+
"loss": 0.3569,
|
| 885 |
"step": 1110
|
| 886 |
},
|
| 887 |
{
|
| 888 |
+
"epoch": 4.12,
|
| 889 |
+
"grad_norm": 3.429755449295044,
|
| 890 |
+
"learning_rate": 5.889705882352942e-05,
|
| 891 |
+
"loss": 0.364,
|
| 892 |
"step": 1120
|
| 893 |
},
|
| 894 |
{
|
| 895 |
+
"epoch": 4.15,
|
| 896 |
+
"grad_norm": 1.637468695640564,
|
| 897 |
+
"learning_rate": 5.852941176470589e-05,
|
| 898 |
+
"loss": 0.2794,
|
| 899 |
"step": 1130
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 4.19,
|
| 903 |
+
"grad_norm": 1.0666115283966064,
|
| 904 |
+
"learning_rate": 5.816176470588236e-05,
|
| 905 |
+
"loss": 0.3449,
|
| 906 |
"step": 1140
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"epoch": 4.23,
|
| 910 |
+
"grad_norm": 10.332019805908203,
|
| 911 |
+
"learning_rate": 5.779411764705882e-05,
|
| 912 |
+
"loss": 0.3412,
|
| 913 |
"step": 1150
|
| 914 |
},
|
| 915 |
{
|
| 916 |
+
"epoch": 4.26,
|
| 917 |
+
"grad_norm": 7.264928817749023,
|
| 918 |
+
"learning_rate": 5.742647058823529e-05,
|
| 919 |
+
"loss": 0.4725,
|
| 920 |
"step": 1160
|
| 921 |
},
|
| 922 |
{
|
| 923 |
+
"epoch": 4.3,
|
| 924 |
+
"grad_norm": 2.231255531311035,
|
| 925 |
+
"learning_rate": 5.7058823529411766e-05,
|
| 926 |
+
"loss": 0.4627,
|
| 927 |
"step": 1170
|
| 928 |
},
|
| 929 |
{
|
| 930 |
+
"epoch": 4.34,
|
| 931 |
+
"grad_norm": 6.666164875030518,
|
| 932 |
+
"learning_rate": 5.6691176470588234e-05,
|
| 933 |
+
"loss": 0.2595,
|
| 934 |
"step": 1180
|
| 935 |
},
|
| 936 |
{
|
| 937 |
+
"epoch": 4.38,
|
| 938 |
+
"grad_norm": 2.6702773571014404,
|
| 939 |
+
"learning_rate": 5.632352941176471e-05,
|
| 940 |
+
"loss": 0.3103,
|
| 941 |
"step": 1190
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"epoch": 4.41,
|
| 945 |
+
"grad_norm": 3.820765733718872,
|
| 946 |
+
"learning_rate": 5.595588235294118e-05,
|
| 947 |
+
"loss": 0.3491,
|
| 948 |
"step": 1200
|
| 949 |
},
|
| 950 |
{
|
| 951 |
+
"epoch": 4.41,
|
| 952 |
+
"eval_accuracy": 0.859245630174793,
|
| 953 |
+
"eval_loss": 0.4810539484024048,
|
| 954 |
+
"eval_runtime": 14.5136,
|
| 955 |
+
"eval_samples_per_second": 74.895,
|
| 956 |
+
"eval_steps_per_second": 9.371,
|
| 957 |
"step": 1200
|
| 958 |
},
|
| 959 |
{
|
| 960 |
+
"epoch": 4.45,
|
| 961 |
+
"grad_norm": 7.34071683883667,
|
| 962 |
+
"learning_rate": 5.558823529411765e-05,
|
| 963 |
+
"loss": 0.3211,
|
| 964 |
"step": 1210
|
| 965 |
},
|
| 966 |
{
|
| 967 |
+
"epoch": 4.49,
|
| 968 |
+
"grad_norm": 8.903788566589355,
|
| 969 |
+
"learning_rate": 5.522058823529412e-05,
|
| 970 |
+
"loss": 0.3483,
|
| 971 |
"step": 1220
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"epoch": 4.52,
|
| 975 |
+
"grad_norm": 0.47331002354621887,
|
| 976 |
+
"learning_rate": 5.48529411764706e-05,
|
| 977 |
+
"loss": 0.3932,
|
| 978 |
"step": 1230
|
| 979 |
},
|
| 980 |
{
|
| 981 |
+
"epoch": 4.56,
|
| 982 |
+
"grad_norm": 7.376969814300537,
|
| 983 |
+
"learning_rate": 5.448529411764706e-05,
|
| 984 |
+
"loss": 0.5006,
|
| 985 |
"step": 1240
|
| 986 |
},
|
| 987 |
{
|
| 988 |
+
"epoch": 4.6,
|
| 989 |
+
"grad_norm": 4.141767978668213,
|
| 990 |
+
"learning_rate": 5.411764705882353e-05,
|
| 991 |
+
"loss": 0.2654,
|
| 992 |
"step": 1250
|
| 993 |
},
|
| 994 |
{
|
| 995 |
+
"epoch": 4.63,
|
| 996 |
+
"grad_norm": 6.543585777282715,
|
| 997 |
+
"learning_rate": 5.375e-05,
|
| 998 |
+
"loss": 0.4202,
|
| 999 |
"step": 1260
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
+
"epoch": 4.67,
|
| 1003 |
+
"grad_norm": 6.273320198059082,
|
| 1004 |
+
"learning_rate": 5.338235294117647e-05,
|
| 1005 |
+
"loss": 0.4166,
|
| 1006 |
"step": 1270
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
+
"epoch": 4.71,
|
| 1010 |
+
"grad_norm": 3.7427868843078613,
|
| 1011 |
+
"learning_rate": 5.301470588235295e-05,
|
| 1012 |
+
"loss": 0.3726,
|
| 1013 |
"step": 1280
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
+
"epoch": 4.74,
|
| 1017 |
+
"grad_norm": 10.277591705322266,
|
| 1018 |
+
"learning_rate": 5.2647058823529414e-05,
|
| 1019 |
+
"loss": 0.4394,
|
| 1020 |
"step": 1290
|
| 1021 |
},
|
| 1022 |
{
|
| 1023 |
+
"epoch": 4.78,
|
| 1024 |
+
"grad_norm": 2.809276580810547,
|
| 1025 |
+
"learning_rate": 5.227941176470589e-05,
|
| 1026 |
+
"loss": 0.3507,
|
| 1027 |
"step": 1300
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
+
"epoch": 4.78,
|
| 1031 |
+
"eval_accuracy": 0.8518859245630175,
|
| 1032 |
+
"eval_loss": 0.4860253632068634,
|
| 1033 |
+
"eval_runtime": 14.6249,
|
| 1034 |
+
"eval_samples_per_second": 74.325,
|
| 1035 |
+
"eval_steps_per_second": 9.299,
|
| 1036 |
"step": 1300
|
| 1037 |
},
|
| 1038 |
{
|
| 1039 |
+
"epoch": 4.82,
|
| 1040 |
+
"grad_norm": 3.2382688522338867,
|
| 1041 |
+
"learning_rate": 5.191176470588236e-05,
|
| 1042 |
+
"loss": 0.2858,
|
| 1043 |
"step": 1310
|
| 1044 |
},
|
| 1045 |
{
|
| 1046 |
+
"epoch": 4.85,
|
| 1047 |
+
"grad_norm": 2.7890524864196777,
|
| 1048 |
+
"learning_rate": 5.154411764705882e-05,
|
| 1049 |
+
"loss": 0.4912,
|
| 1050 |
"step": 1320
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
+
"epoch": 4.89,
|
| 1054 |
+
"grad_norm": 3.5275235176086426,
|
| 1055 |
+
"learning_rate": 5.117647058823529e-05,
|
| 1056 |
+
"loss": 0.3221,
|
| 1057 |
"step": 1330
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
+
"epoch": 4.93,
|
| 1061 |
+
"grad_norm": 3.107577085494995,
|
| 1062 |
+
"learning_rate": 5.0808823529411764e-05,
|
| 1063 |
+
"loss": 0.2965,
|
| 1064 |
"step": 1340
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
+
"epoch": 4.96,
|
| 1068 |
+
"grad_norm": 2.5357701778411865,
|
| 1069 |
+
"learning_rate": 5.044117647058824e-05,
|
| 1070 |
+
"loss": 0.2669,
|
| 1071 |
"step": 1350
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
+
"epoch": 5.0,
|
| 1075 |
+
"grad_norm": 9.606520652770996,
|
| 1076 |
+
"learning_rate": 5.0073529411764705e-05,
|
| 1077 |
+
"loss": 0.3497,
|
| 1078 |
"step": 1360
|
| 1079 |
},
|
| 1080 |
{
|
| 1081 |
+
"epoch": 5.04,
|
| 1082 |
+
"grad_norm": 4.934056282043457,
|
| 1083 |
+
"learning_rate": 4.970588235294118e-05,
|
| 1084 |
+
"loss": 0.3759,
|
| 1085 |
"step": 1370
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
+
"epoch": 5.07,
|
| 1089 |
+
"grad_norm": 1.4073066711425781,
|
| 1090 |
+
"learning_rate": 4.933823529411765e-05,
|
| 1091 |
+
"loss": 0.2889,
|
| 1092 |
"step": 1380
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
+
"epoch": 5.11,
|
| 1096 |
+
"grad_norm": 8.593390464782715,
|
| 1097 |
+
"learning_rate": 4.897058823529412e-05,
|
| 1098 |
+
"loss": 0.2558,
|
| 1099 |
"step": 1390
|
| 1100 |
},
|
| 1101 |
{
|
| 1102 |
+
"epoch": 5.15,
|
| 1103 |
+
"grad_norm": 3.6950008869171143,
|
| 1104 |
+
"learning_rate": 4.860294117647059e-05,
|
| 1105 |
+
"loss": 0.2684,
|
| 1106 |
"step": 1400
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
+
"epoch": 5.15,
|
| 1110 |
+
"eval_accuracy": 0.8629254829806807,
|
| 1111 |
+
"eval_loss": 0.46262994408607483,
|
| 1112 |
+
"eval_runtime": 14.9522,
|
| 1113 |
+
"eval_samples_per_second": 72.698,
|
| 1114 |
+
"eval_steps_per_second": 9.096,
|
| 1115 |
"step": 1400
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
+
"epoch": 5.18,
|
| 1119 |
+
"grad_norm": 6.775074005126953,
|
| 1120 |
+
"learning_rate": 4.823529411764706e-05,
|
| 1121 |
+
"loss": 0.2233,
|
| 1122 |
"step": 1410
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
+
"epoch": 5.22,
|
| 1126 |
+
"grad_norm": 6.125876426696777,
|
| 1127 |
+
"learning_rate": 4.7867647058823535e-05,
|
| 1128 |
+
"loss": 0.4279,
|
| 1129 |
"step": 1420
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
+
"epoch": 5.26,
|
| 1133 |
+
"grad_norm": 2.7117116451263428,
|
| 1134 |
+
"learning_rate": 4.75e-05,
|
| 1135 |
+
"loss": 0.3462,
|
| 1136 |
"step": 1430
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
+
"epoch": 5.29,
|
| 1140 |
+
"grad_norm": 0.45755088329315186,
|
| 1141 |
+
"learning_rate": 4.713235294117647e-05,
|
| 1142 |
+
"loss": 0.2898,
|
| 1143 |
"step": 1440
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
+
"epoch": 5.33,
|
| 1147 |
+
"grad_norm": 1.8368974924087524,
|
| 1148 |
+
"learning_rate": 4.6764705882352944e-05,
|
| 1149 |
+
"loss": 0.3559,
|
| 1150 |
"step": 1450
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
+
"epoch": 5.37,
|
| 1154 |
+
"grad_norm": 6.555761814117432,
|
| 1155 |
+
"learning_rate": 4.639705882352942e-05,
|
| 1156 |
+
"loss": 0.5559,
|
| 1157 |
"step": 1460
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
+
"epoch": 5.4,
|
| 1161 |
+
"grad_norm": 3.9688079357147217,
|
| 1162 |
+
"learning_rate": 4.6029411764705885e-05,
|
| 1163 |
+
"loss": 0.3366,
|
| 1164 |
"step": 1470
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
+
"epoch": 5.44,
|
| 1168 |
+
"grad_norm": 4.957942485809326,
|
| 1169 |
+
"learning_rate": 4.566176470588235e-05,
|
| 1170 |
+
"loss": 0.34,
|
| 1171 |
"step": 1480
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
+
"epoch": 5.48,
|
| 1175 |
+
"grad_norm": 2.670506238937378,
|
| 1176 |
+
"learning_rate": 4.5294117647058826e-05,
|
| 1177 |
+
"loss": 0.2709,
|
| 1178 |
"step": 1490
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
+
"epoch": 5.51,
|
| 1182 |
+
"grad_norm": 2.5184476375579834,
|
| 1183 |
+
"learning_rate": 4.49264705882353e-05,
|
| 1184 |
+
"loss": 0.294,
|
| 1185 |
"step": 1500
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
+
"epoch": 5.51,
|
| 1189 |
+
"eval_accuracy": 0.859245630174793,
|
| 1190 |
+
"eval_loss": 0.46787214279174805,
|
| 1191 |
+
"eval_runtime": 14.635,
|
| 1192 |
+
"eval_samples_per_second": 74.274,
|
| 1193 |
+
"eval_steps_per_second": 9.293,
|
| 1194 |
"step": 1500
|
| 1195 |
},
|
| 1196 |
{
|
| 1197 |
+
"epoch": 5.55,
|
| 1198 |
+
"grad_norm": 1.6012226343154907,
|
| 1199 |
+
"learning_rate": 4.455882352941177e-05,
|
| 1200 |
+
"loss": 0.215,
|
| 1201 |
"step": 1510
|
| 1202 |
},
|
| 1203 |
{
|
| 1204 |
+
"epoch": 5.59,
|
| 1205 |
+
"grad_norm": 7.58560848236084,
|
| 1206 |
+
"learning_rate": 4.4191176470588235e-05,
|
| 1207 |
+
"loss": 0.3916,
|
| 1208 |
"step": 1520
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
+
"epoch": 5.62,
|
| 1212 |
+
"grad_norm": 4.081338405609131,
|
| 1213 |
+
"learning_rate": 4.382352941176471e-05,
|
| 1214 |
+
"loss": 0.3002,
|
| 1215 |
"step": 1530
|
| 1216 |
},
|
| 1217 |
{
|
| 1218 |
+
"epoch": 5.66,
|
| 1219 |
+
"grad_norm": 6.238924026489258,
|
| 1220 |
+
"learning_rate": 4.345588235294118e-05,
|
| 1221 |
+
"loss": 0.2925,
|
| 1222 |
"step": 1540
|
| 1223 |
},
|
| 1224 |
{
|
| 1225 |
+
"epoch": 5.7,
|
| 1226 |
+
"grad_norm": 2.2608234882354736,
|
| 1227 |
+
"learning_rate": 4.308823529411765e-05,
|
| 1228 |
+
"loss": 0.3005,
|
| 1229 |
"step": 1550
|
| 1230 |
},
|
| 1231 |
{
|
| 1232 |
+
"epoch": 5.74,
|
| 1233 |
+
"grad_norm": 0.5036042928695679,
|
| 1234 |
+
"learning_rate": 4.272058823529412e-05,
|
| 1235 |
+
"loss": 0.2476,
|
| 1236 |
"step": 1560
|
| 1237 |
},
|
| 1238 |
{
|
| 1239 |
+
"epoch": 5.77,
|
| 1240 |
+
"grad_norm": 4.822847843170166,
|
| 1241 |
+
"learning_rate": 4.235294117647059e-05,
|
| 1242 |
+
"loss": 0.2672,
|
| 1243 |
"step": 1570
|
| 1244 |
},
|
| 1245 |
{
|
| 1246 |
+
"epoch": 5.81,
|
| 1247 |
+
"grad_norm": 3.3567371368408203,
|
| 1248 |
+
"learning_rate": 4.198529411764706e-05,
|
| 1249 |
+
"loss": 0.1596,
|
| 1250 |
"step": 1580
|
| 1251 |
},
|
| 1252 |
{
|
| 1253 |
+
"epoch": 5.85,
|
| 1254 |
+
"grad_norm": 6.035266399383545,
|
| 1255 |
+
"learning_rate": 4.161764705882353e-05,
|
| 1256 |
+
"loss": 0.2478,
|
| 1257 |
"step": 1590
|
| 1258 |
},
|
| 1259 |
{
|
| 1260 |
+
"epoch": 5.88,
|
| 1261 |
+
"grad_norm": 2.863812208175659,
|
| 1262 |
+
"learning_rate": 4.125e-05,
|
| 1263 |
+
"loss": 0.3686,
|
| 1264 |
"step": 1600
|
| 1265 |
},
|
| 1266 |
{
|
| 1267 |
+
"epoch": 5.88,
|
| 1268 |
+
"eval_accuracy": 0.859245630174793,
|
| 1269 |
+
"eval_loss": 0.49989455938339233,
|
| 1270 |
+
"eval_runtime": 14.6545,
|
| 1271 |
+
"eval_samples_per_second": 74.175,
|
| 1272 |
+
"eval_steps_per_second": 9.28,
|
| 1273 |
"step": 1600
|
| 1274 |
},
|
| 1275 |
{
|
| 1276 |
+
"epoch": 5.92,
|
| 1277 |
+
"grad_norm": 6.965085983276367,
|
| 1278 |
+
"learning_rate": 4.0882352941176474e-05,
|
| 1279 |
+
"loss": 0.4171,
|
| 1280 |
"step": 1610
|
| 1281 |
},
|
| 1282 |
{
|
| 1283 |
+
"epoch": 5.96,
|
| 1284 |
+
"grad_norm": 8.275744438171387,
|
| 1285 |
+
"learning_rate": 4.051470588235294e-05,
|
| 1286 |
+
"loss": 0.3547,
|
| 1287 |
"step": 1620
|
| 1288 |
},
|
| 1289 |
{
|
| 1290 |
+
"epoch": 5.99,
|
| 1291 |
+
"grad_norm": 12.617408752441406,
|
| 1292 |
+
"learning_rate": 4.0147058823529415e-05,
|
| 1293 |
+
"loss": 0.2986,
|
| 1294 |
"step": 1630
|
| 1295 |
},
|
| 1296 |
{
|
| 1297 |
+
"epoch": 6.03,
|
| 1298 |
+
"grad_norm": 4.8493332862854,
|
| 1299 |
+
"learning_rate": 3.977941176470588e-05,
|
| 1300 |
+
"loss": 0.2851,
|
| 1301 |
"step": 1640
|
| 1302 |
},
|
| 1303 |
{
|
| 1304 |
+
"epoch": 6.07,
|
| 1305 |
+
"grad_norm": 4.861184120178223,
|
| 1306 |
+
"learning_rate": 3.9411764705882356e-05,
|
| 1307 |
+
"loss": 0.2588,
|
| 1308 |
"step": 1650
|
| 1309 |
},
|
| 1310 |
{
|
| 1311 |
+
"epoch": 6.1,
|
| 1312 |
+
"grad_norm": 0.13496477901935577,
|
| 1313 |
+
"learning_rate": 3.9044117647058823e-05,
|
| 1314 |
+
"loss": 0.2329,
|
| 1315 |
"step": 1660
|
| 1316 |
},
|
| 1317 |
{
|
| 1318 |
+
"epoch": 6.14,
|
| 1319 |
+
"grad_norm": 4.618465423583984,
|
| 1320 |
+
"learning_rate": 3.86764705882353e-05,
|
| 1321 |
+
"loss": 0.2163,
|
| 1322 |
"step": 1670
|
| 1323 |
},
|
| 1324 |
{
|
| 1325 |
+
"epoch": 6.18,
|
| 1326 |
+
"grad_norm": 4.234817981719971,
|
| 1327 |
+
"learning_rate": 3.830882352941177e-05,
|
| 1328 |
+
"loss": 0.2164,
|
| 1329 |
"step": 1680
|
| 1330 |
},
|
| 1331 |
{
|
| 1332 |
+
"epoch": 6.21,
|
| 1333 |
+
"grad_norm": 8.013242721557617,
|
| 1334 |
+
"learning_rate": 3.794117647058824e-05,
|
| 1335 |
+
"loss": 0.2066,
|
| 1336 |
"step": 1690
|
| 1337 |
},
|
| 1338 |
{
|
| 1339 |
+
"epoch": 6.25,
|
| 1340 |
+
"grad_norm": 1.7797431945800781,
|
| 1341 |
+
"learning_rate": 3.7573529411764706e-05,
|
| 1342 |
+
"loss": 0.3195,
|
| 1343 |
"step": 1700
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
+
"epoch": 6.25,
|
| 1347 |
+
"eval_accuracy": 0.8702851885924563,
|
| 1348 |
+
"eval_loss": 0.44596514105796814,
|
| 1349 |
+
"eval_runtime": 14.5693,
|
| 1350 |
+
"eval_samples_per_second": 74.609,
|
| 1351 |
+
"eval_steps_per_second": 9.335,
|
| 1352 |
"step": 1700
|
| 1353 |
},
|
| 1354 |
{
|
| 1355 |
+
"epoch": 6.29,
|
| 1356 |
+
"grad_norm": 1.100888967514038,
|
| 1357 |
+
"learning_rate": 3.720588235294118e-05,
|
| 1358 |
+
"loss": 0.3209,
|
| 1359 |
"step": 1710
|
| 1360 |
},
|
| 1361 |
{
|
| 1362 |
+
"epoch": 6.32,
|
| 1363 |
+
"grad_norm": 0.9825335741043091,
|
| 1364 |
+
"learning_rate": 3.6838235294117654e-05,
|
| 1365 |
+
"loss": 0.252,
|
| 1366 |
"step": 1720
|
| 1367 |
},
|
| 1368 |
{
|
| 1369 |
+
"epoch": 6.36,
|
| 1370 |
+
"grad_norm": 1.056939959526062,
|
| 1371 |
+
"learning_rate": 3.6470588235294114e-05,
|
| 1372 |
+
"loss": 0.283,
|
| 1373 |
"step": 1730
|
| 1374 |
},
|
| 1375 |
{
|
| 1376 |
+
"epoch": 6.4,
|
| 1377 |
+
"grad_norm": 4.441556453704834,
|
| 1378 |
+
"learning_rate": 3.610294117647059e-05,
|
| 1379 |
+
"loss": 0.3047,
|
| 1380 |
"step": 1740
|
| 1381 |
},
|
| 1382 |
{
|
| 1383 |
+
"epoch": 6.43,
|
| 1384 |
+
"grad_norm": 4.250430107116699,
|
| 1385 |
+
"learning_rate": 3.573529411764706e-05,
|
| 1386 |
+
"loss": 0.2678,
|
| 1387 |
"step": 1750
|
| 1388 |
},
|
| 1389 |
{
|
| 1390 |
+
"epoch": 6.47,
|
| 1391 |
+
"grad_norm": 2.4963274002075195,
|
| 1392 |
+
"learning_rate": 3.5367647058823536e-05,
|
| 1393 |
+
"loss": 0.3009,
|
| 1394 |
"step": 1760
|
| 1395 |
},
|
| 1396 |
{
|
| 1397 |
+
"epoch": 6.51,
|
| 1398 |
+
"grad_norm": 4.849697589874268,
|
| 1399 |
+
"learning_rate": 3.5e-05,
|
| 1400 |
+
"loss": 0.2309,
|
| 1401 |
"step": 1770
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
+
"epoch": 6.54,
|
| 1405 |
+
"grad_norm": 2.5027718544006348,
|
| 1406 |
+
"learning_rate": 3.463235294117647e-05,
|
| 1407 |
+
"loss": 0.2398,
|
| 1408 |
"step": 1780
|
| 1409 |
},
|
| 1410 |
{
|
| 1411 |
+
"epoch": 6.58,
|
| 1412 |
+
"grad_norm": 6.149186134338379,
|
| 1413 |
+
"learning_rate": 3.4264705882352945e-05,
|
| 1414 |
+
"loss": 0.239,
|
| 1415 |
"step": 1790
|
| 1416 |
},
|
| 1417 |
{
|
| 1418 |
+
"epoch": 6.62,
|
| 1419 |
+
"grad_norm": 7.9045023918151855,
|
| 1420 |
+
"learning_rate": 3.389705882352941e-05,
|
| 1421 |
+
"loss": 0.1949,
|
| 1422 |
"step": 1800
|
| 1423 |
},
|
| 1424 |
{
|
| 1425 |
+
"epoch": 6.62,
|
| 1426 |
+
"eval_accuracy": 0.8785648574057038,
|
| 1427 |
+
"eval_loss": 0.4375220239162445,
|
| 1428 |
+
"eval_runtime": 14.7139,
|
| 1429 |
+
"eval_samples_per_second": 73.876,
|
| 1430 |
+
"eval_steps_per_second": 9.243,
|
| 1431 |
"step": 1800
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
+
"epoch": 6.65,
|
| 1435 |
+
"grad_norm": 3.0206644535064697,
|
| 1436 |
+
"learning_rate": 3.352941176470588e-05,
|
| 1437 |
+
"loss": 0.185,
|
| 1438 |
"step": 1810
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
+
"epoch": 6.69,
|
| 1442 |
+
"grad_norm": 10.069550514221191,
|
| 1443 |
+
"learning_rate": 3.3161764705882353e-05,
|
| 1444 |
+
"loss": 0.1867,
|
| 1445 |
"step": 1820
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
+
"epoch": 6.73,
|
| 1449 |
+
"grad_norm": 3.1332738399505615,
|
| 1450 |
+
"learning_rate": 3.279411764705883e-05,
|
| 1451 |
+
"loss": 0.2411,
|
| 1452 |
"step": 1830
|
| 1453 |
},
|
| 1454 |
{
|
| 1455 |
+
"epoch": 6.76,
|
| 1456 |
+
"grad_norm": 10.362940788269043,
|
| 1457 |
+
"learning_rate": 3.2426470588235295e-05,
|
| 1458 |
+
"loss": 0.2853,
|
| 1459 |
"step": 1840
|
| 1460 |
},
|
| 1461 |
{
|
| 1462 |
+
"epoch": 6.8,
|
| 1463 |
+
"grad_norm": 1.3874270915985107,
|
| 1464 |
+
"learning_rate": 3.205882352941177e-05,
|
| 1465 |
+
"loss": 0.2228,
|
| 1466 |
"step": 1850
|
| 1467 |
},
|
| 1468 |
{
|
| 1469 |
+
"epoch": 6.84,
|
| 1470 |
+
"grad_norm": 4.133223056793213,
|
| 1471 |
+
"learning_rate": 3.1691176470588236e-05,
|
| 1472 |
+
"loss": 0.2583,
|
| 1473 |
"step": 1860
|
| 1474 |
},
|
| 1475 |
{
|
| 1476 |
+
"epoch": 6.88,
|
| 1477 |
+
"grad_norm": 2.4844319820404053,
|
| 1478 |
+
"learning_rate": 3.132352941176471e-05,
|
| 1479 |
+
"loss": 0.2522,
|
| 1480 |
"step": 1870
|
| 1481 |
},
|
| 1482 |
{
|
| 1483 |
+
"epoch": 6.91,
|
| 1484 |
+
"grad_norm": 1.4133566617965698,
|
| 1485 |
+
"learning_rate": 3.095588235294118e-05,
|
| 1486 |
+
"loss": 0.1675,
|
| 1487 |
"step": 1880
|
| 1488 |
},
|
| 1489 |
{
|
| 1490 |
+
"epoch": 6.95,
|
| 1491 |
+
"grad_norm": 4.407350540161133,
|
| 1492 |
+
"learning_rate": 3.058823529411765e-05,
|
| 1493 |
+
"loss": 0.4815,
|
| 1494 |
"step": 1890
|
| 1495 |
},
|
| 1496 |
{
|
| 1497 |
+
"epoch": 6.99,
|
| 1498 |
+
"grad_norm": 4.365607261657715,
|
| 1499 |
+
"learning_rate": 3.022058823529412e-05,
|
| 1500 |
+
"loss": 0.2524,
|
| 1501 |
"step": 1900
|
| 1502 |
},
|
| 1503 |
{
|
| 1504 |
+
"epoch": 6.99,
|
| 1505 |
+
"eval_accuracy": 0.8794848206071757,
|
| 1506 |
+
"eval_loss": 0.43937617540359497,
|
| 1507 |
+
"eval_runtime": 14.6572,
|
| 1508 |
+
"eval_samples_per_second": 74.161,
|
| 1509 |
+
"eval_steps_per_second": 9.279,
|
| 1510 |
"step": 1900
|
| 1511 |
},
|
| 1512 |
{
|
| 1513 |
+
"epoch": 7.02,
|
| 1514 |
+
"grad_norm": 2.426751136779785,
|
| 1515 |
+
"learning_rate": 2.985294117647059e-05,
|
| 1516 |
+
"loss": 0.3648,
|
| 1517 |
"step": 1910
|
| 1518 |
},
|
| 1519 |
{
|
| 1520 |
+
"epoch": 7.06,
|
| 1521 |
+
"grad_norm": 2.1712534427642822,
|
| 1522 |
+
"learning_rate": 2.948529411764706e-05,
|
| 1523 |
+
"loss": 0.2766,
|
| 1524 |
"step": 1920
|
| 1525 |
},
|
| 1526 |
{
|
| 1527 |
+
"epoch": 7.1,
|
| 1528 |
+
"grad_norm": 5.872859477996826,
|
| 1529 |
+
"learning_rate": 2.9117647058823534e-05,
|
| 1530 |
+
"loss": 0.2266,
|
| 1531 |
"step": 1930
|
| 1532 |
},
|
| 1533 |
{
|
| 1534 |
+
"epoch": 7.13,
|
| 1535 |
+
"grad_norm": 5.788787841796875,
|
| 1536 |
+
"learning_rate": 2.8749999999999997e-05,
|
| 1537 |
+
"loss": 0.2683,
|
| 1538 |
"step": 1940
|
| 1539 |
},
|
| 1540 |
{
|
| 1541 |
+
"epoch": 7.17,
|
| 1542 |
+
"grad_norm": 3.856825590133667,
|
| 1543 |
+
"learning_rate": 2.838235294117647e-05,
|
| 1544 |
+
"loss": 0.2663,
|
| 1545 |
"step": 1950
|
| 1546 |
},
|
| 1547 |
{
|
| 1548 |
+
"epoch": 7.21,
|
| 1549 |
+
"grad_norm": 5.601237773895264,
|
| 1550 |
+
"learning_rate": 2.8014705882352942e-05,
|
| 1551 |
+
"loss": 0.3422,
|
| 1552 |
"step": 1960
|
| 1553 |
},
|
| 1554 |
{
|
| 1555 |
+
"epoch": 7.24,
|
| 1556 |
+
"grad_norm": 4.638755798339844,
|
| 1557 |
+
"learning_rate": 2.7647058823529416e-05,
|
| 1558 |
+
"loss": 0.2962,
|
| 1559 |
"step": 1970
|
| 1560 |
},
|
| 1561 |
{
|
| 1562 |
+
"epoch": 7.28,
|
| 1563 |
+
"grad_norm": 3.0878632068634033,
|
| 1564 |
+
"learning_rate": 2.727941176470588e-05,
|
| 1565 |
+
"loss": 0.195,
|
| 1566 |
"step": 1980
|
| 1567 |
},
|
| 1568 |
{
|
| 1569 |
+
"epoch": 7.32,
|
| 1570 |
+
"grad_norm": 3.4959683418273926,
|
| 1571 |
+
"learning_rate": 2.6911764705882354e-05,
|
| 1572 |
+
"loss": 0.2453,
|
| 1573 |
"step": 1990
|
| 1574 |
},
|
| 1575 |
{
|
| 1576 |
+
"epoch": 7.35,
|
| 1577 |
+
"grad_norm": 7.705379009246826,
|
| 1578 |
+
"learning_rate": 2.6544117647058825e-05,
|
| 1579 |
+
"loss": 0.1885,
|
| 1580 |
"step": 2000
|
| 1581 |
},
|
| 1582 |
{
|
| 1583 |
+
"epoch": 7.35,
|
| 1584 |
+
"eval_accuracy": 0.8647654093836247,
|
| 1585 |
+
"eval_loss": 0.48834648728370667,
|
| 1586 |
+
"eval_runtime": 14.5234,
|
| 1587 |
+
"eval_samples_per_second": 74.845,
|
| 1588 |
+
"eval_steps_per_second": 9.364,
|
| 1589 |
"step": 2000
|
| 1590 |
},
|
| 1591 |
{
|
| 1592 |
+
"epoch": 7.39,
|
| 1593 |
+
"grad_norm": 3.521489381790161,
|
| 1594 |
+
"learning_rate": 2.6176470588235295e-05,
|
| 1595 |
+
"loss": 0.1937,
|
| 1596 |
"step": 2010
|
| 1597 |
},
|
| 1598 |
{
|
| 1599 |
+
"epoch": 7.43,
|
| 1600 |
+
"grad_norm": 0.4547409117221832,
|
| 1601 |
+
"learning_rate": 2.580882352941177e-05,
|
| 1602 |
+
"loss": 0.2121,
|
| 1603 |
"step": 2020
|
| 1604 |
},
|
| 1605 |
{
|
| 1606 |
+
"epoch": 7.46,
|
| 1607 |
+
"grad_norm": 3.6252083778381348,
|
| 1608 |
+
"learning_rate": 2.5441176470588236e-05,
|
| 1609 |
+
"loss": 0.2637,
|
| 1610 |
"step": 2030
|
| 1611 |
},
|
| 1612 |
{
|
| 1613 |
+
"epoch": 7.5,
|
| 1614 |
+
"grad_norm": 3.650339365005493,
|
| 1615 |
+
"learning_rate": 2.5073529411764707e-05,
|
| 1616 |
+
"loss": 0.2771,
|
| 1617 |
"step": 2040
|
| 1618 |
},
|
| 1619 |
{
|
| 1620 |
+
"epoch": 7.54,
|
| 1621 |
+
"grad_norm": 7.28909969329834,
|
| 1622 |
+
"learning_rate": 2.4705882352941178e-05,
|
| 1623 |
+
"loss": 0.2559,
|
| 1624 |
"step": 2050
|
| 1625 |
},
|
| 1626 |
{
|
| 1627 |
+
"epoch": 7.57,
|
| 1628 |
+
"grad_norm": 11.111797332763672,
|
| 1629 |
+
"learning_rate": 2.433823529411765e-05,
|
| 1630 |
+
"loss": 0.1597,
|
| 1631 |
"step": 2060
|
| 1632 |
},
|
| 1633 |
{
|
| 1634 |
+
"epoch": 7.61,
|
| 1635 |
+
"grad_norm": 6.980329990386963,
|
| 1636 |
+
"learning_rate": 2.397058823529412e-05,
|
| 1637 |
+
"loss": 0.2273,
|
| 1638 |
"step": 2070
|
| 1639 |
},
|
| 1640 |
{
|
| 1641 |
+
"epoch": 7.65,
|
| 1642 |
+
"grad_norm": 7.767569065093994,
|
| 1643 |
+
"learning_rate": 2.360294117647059e-05,
|
| 1644 |
+
"loss": 0.2912,
|
| 1645 |
"step": 2080
|
| 1646 |
},
|
| 1647 |
{
|
| 1648 |
+
"epoch": 7.68,
|
| 1649 |
+
"grad_norm": 0.5350354313850403,
|
| 1650 |
+
"learning_rate": 2.323529411764706e-05,
|
| 1651 |
+
"loss": 0.1747,
|
| 1652 |
"step": 2090
|
| 1653 |
},
|
| 1654 |
{
|
| 1655 |
+
"epoch": 7.72,
|
| 1656 |
+
"grad_norm": 3.1379992961883545,
|
| 1657 |
+
"learning_rate": 2.2904411764705882e-05,
|
| 1658 |
+
"loss": 0.2804,
|
| 1659 |
"step": 2100
|
| 1660 |
},
|
| 1661 |
{
|
| 1662 |
+
"epoch": 7.72,
|
| 1663 |
+
"eval_accuracy": 0.8739650413983441,
|
| 1664 |
+
"eval_loss": 0.45086538791656494,
|
| 1665 |
+
"eval_runtime": 14.5726,
|
| 1666 |
+
"eval_samples_per_second": 74.592,
|
| 1667 |
+
"eval_steps_per_second": 9.333,
|
| 1668 |
"step": 2100
|
| 1669 |
},
|
| 1670 |
{
|
| 1671 |
+
"epoch": 7.76,
|
| 1672 |
+
"grad_norm": 3.015038013458252,
|
| 1673 |
+
"learning_rate": 2.2536764705882356e-05,
|
| 1674 |
+
"loss": 0.2127,
|
| 1675 |
"step": 2110
|
| 1676 |
},
|
| 1677 |
{
|
| 1678 |
+
"epoch": 7.79,
|
| 1679 |
+
"grad_norm": 6.477613925933838,
|
| 1680 |
+
"learning_rate": 2.2169117647058823e-05,
|
| 1681 |
+
"loss": 0.3081,
|
| 1682 |
"step": 2120
|
| 1683 |
},
|
| 1684 |
{
|
| 1685 |
+
"epoch": 7.83,
|
| 1686 |
+
"grad_norm": 5.475728988647461,
|
| 1687 |
+
"learning_rate": 2.1801470588235294e-05,
|
| 1688 |
+
"loss": 0.2435,
|
| 1689 |
"step": 2130
|
| 1690 |
},
|
| 1691 |
{
|
| 1692 |
+
"epoch": 7.87,
|
| 1693 |
+
"grad_norm": 6.454700946807861,
|
| 1694 |
+
"learning_rate": 2.1433823529411765e-05,
|
| 1695 |
+
"loss": 0.2963,
|
| 1696 |
"step": 2140
|
| 1697 |
},
|
| 1698 |
{
|
| 1699 |
+
"epoch": 7.9,
|
| 1700 |
+
"grad_norm": 2.8447799682617188,
|
| 1701 |
+
"learning_rate": 2.1066176470588235e-05,
|
| 1702 |
+
"loss": 0.1675,
|
| 1703 |
"step": 2150
|
| 1704 |
},
|
| 1705 |
{
|
| 1706 |
+
"epoch": 7.94,
|
| 1707 |
+
"grad_norm": 6.482878684997559,
|
| 1708 |
+
"learning_rate": 2.0698529411764706e-05,
|
| 1709 |
+
"loss": 0.2058,
|
| 1710 |
"step": 2160
|
| 1711 |
},
|
| 1712 |
{
|
| 1713 |
+
"epoch": 7.98,
|
| 1714 |
+
"grad_norm": 0.8823295831680298,
|
| 1715 |
+
"learning_rate": 2.0330882352941176e-05,
|
| 1716 |
+
"loss": 0.2346,
|
| 1717 |
"step": 2170
|
| 1718 |
},
|
| 1719 |
{
|
| 1720 |
+
"epoch": 8.01,
|
| 1721 |
+
"grad_norm": 10.87021541595459,
|
| 1722 |
+
"learning_rate": 2e-05,
|
| 1723 |
+
"loss": 0.2055,
|
| 1724 |
"step": 2180
|
| 1725 |
},
|
| 1726 |
{
|
| 1727 |
+
"epoch": 8.05,
|
| 1728 |
+
"grad_norm": 4.414636135101318,
|
| 1729 |
+
"learning_rate": 1.9632352941176472e-05,
|
| 1730 |
+
"loss": 0.2275,
|
| 1731 |
"step": 2190
|
| 1732 |
},
|
| 1733 |
{
|
| 1734 |
+
"epoch": 8.09,
|
| 1735 |
+
"grad_norm": 0.8171437382698059,
|
| 1736 |
+
"learning_rate": 1.9264705882352943e-05,
|
| 1737 |
+
"loss": 0.2532,
|
| 1738 |
"step": 2200
|
| 1739 |
},
|
| 1740 |
{
|
| 1741 |
+
"epoch": 8.09,
|
| 1742 |
+
"eval_accuracy": 0.8629254829806807,
|
| 1743 |
+
"eval_loss": 0.4744163453578949,
|
| 1744 |
+
"eval_runtime": 14.6235,
|
| 1745 |
+
"eval_samples_per_second": 74.332,
|
| 1746 |
+
"eval_steps_per_second": 9.3,
|
| 1747 |
"step": 2200
|
| 1748 |
},
|
| 1749 |
{
|
| 1750 |
+
"epoch": 8.12,
|
| 1751 |
+
"grad_norm": 3.7062883377075195,
|
| 1752 |
+
"learning_rate": 1.8897058823529413e-05,
|
| 1753 |
+
"loss": 0.2292,
|
| 1754 |
"step": 2210
|
| 1755 |
},
|
| 1756 |
{
|
| 1757 |
+
"epoch": 8.16,
|
| 1758 |
+
"grad_norm": 6.667297840118408,
|
| 1759 |
+
"learning_rate": 1.8529411764705884e-05,
|
| 1760 |
+
"loss": 0.266,
|
| 1761 |
"step": 2220
|
| 1762 |
},
|
| 1763 |
{
|
| 1764 |
+
"epoch": 8.2,
|
| 1765 |
+
"grad_norm": 3.517827033996582,
|
| 1766 |
+
"learning_rate": 1.8161764705882355e-05,
|
| 1767 |
+
"loss": 0.1754,
|
| 1768 |
"step": 2230
|
| 1769 |
},
|
| 1770 |
{
|
| 1771 |
+
"epoch": 8.24,
|
| 1772 |
+
"grad_norm": 4.565939903259277,
|
| 1773 |
+
"learning_rate": 1.7794117647058825e-05,
|
| 1774 |
+
"loss": 0.2276,
|
| 1775 |
"step": 2240
|
| 1776 |
},
|
| 1777 |
{
|
| 1778 |
+
"epoch": 8.27,
|
| 1779 |
+
"grad_norm": 3.808417320251465,
|
| 1780 |
+
"learning_rate": 1.7426470588235293e-05,
|
| 1781 |
+
"loss": 0.2818,
|
| 1782 |
"step": 2250
|
| 1783 |
},
|
| 1784 |
{
|
| 1785 |
+
"epoch": 8.31,
|
| 1786 |
+
"grad_norm": 8.612093925476074,
|
| 1787 |
+
"learning_rate": 1.7058823529411767e-05,
|
| 1788 |
+
"loss": 0.2267,
|
| 1789 |
"step": 2260
|
| 1790 |
},
|
| 1791 |
{
|
| 1792 |
+
"epoch": 8.35,
|
| 1793 |
+
"grad_norm": 1.0861879587173462,
|
| 1794 |
+
"learning_rate": 1.6691176470588234e-05,
|
| 1795 |
+
"loss": 0.1542,
|
| 1796 |
"step": 2270
|
| 1797 |
},
|
| 1798 |
{
|
| 1799 |
+
"epoch": 8.38,
|
| 1800 |
+
"grad_norm": 0.5087774395942688,
|
| 1801 |
+
"learning_rate": 1.6323529411764708e-05,
|
| 1802 |
+
"loss": 0.138,
|
| 1803 |
"step": 2280
|
| 1804 |
},
|
| 1805 |
{
|
| 1806 |
+
"epoch": 8.42,
|
| 1807 |
+
"grad_norm": 3.2882983684539795,
|
| 1808 |
+
"learning_rate": 1.5955882352941175e-05,
|
| 1809 |
+
"loss": 0.3559,
|
| 1810 |
"step": 2290
|
| 1811 |
},
|
| 1812 |
{
|
| 1813 |
+
"epoch": 8.46,
|
| 1814 |
+
"grad_norm": 12.479865074157715,
|
| 1815 |
+
"learning_rate": 1.558823529411765e-05,
|
| 1816 |
+
"loss": 0.2038,
|
| 1817 |
"step": 2300
|
| 1818 |
},
|
| 1819 |
{
|
| 1820 |
+
"epoch": 8.46,
|
| 1821 |
+
"eval_accuracy": 0.8666053357865685,
|
| 1822 |
+
"eval_loss": 0.4631020724773407,
|
| 1823 |
+
"eval_runtime": 14.4937,
|
| 1824 |
+
"eval_samples_per_second": 74.998,
|
| 1825 |
+
"eval_steps_per_second": 9.383,
|
| 1826 |
"step": 2300
|
| 1827 |
},
|
| 1828 |
{
|
| 1829 |
+
"epoch": 8.49,
|
| 1830 |
+
"grad_norm": 8.007225036621094,
|
| 1831 |
+
"learning_rate": 1.5220588235294116e-05,
|
| 1832 |
+
"loss": 0.1528,
|
| 1833 |
"step": 2310
|
| 1834 |
},
|
| 1835 |
{
|
| 1836 |
+
"epoch": 8.53,
|
| 1837 |
+
"grad_norm": 2.102635622024536,
|
| 1838 |
+
"learning_rate": 1.4852941176470589e-05,
|
| 1839 |
+
"loss": 0.213,
|
| 1840 |
"step": 2320
|
| 1841 |
},
|
| 1842 |
{
|
| 1843 |
+
"epoch": 8.57,
|
| 1844 |
+
"grad_norm": 4.558967590332031,
|
| 1845 |
+
"learning_rate": 1.4485294117647061e-05,
|
| 1846 |
+
"loss": 0.2238,
|
| 1847 |
"step": 2330
|
| 1848 |
},
|
| 1849 |
{
|
| 1850 |
+
"epoch": 8.6,
|
| 1851 |
+
"grad_norm": 9.647042274475098,
|
| 1852 |
+
"learning_rate": 1.411764705882353e-05,
|
| 1853 |
+
"loss": 0.3028,
|
| 1854 |
"step": 2340
|
| 1855 |
},
|
| 1856 |
{
|
| 1857 |
+
"epoch": 8.64,
|
| 1858 |
+
"grad_norm": 1.0331591367721558,
|
| 1859 |
+
"learning_rate": 1.3750000000000002e-05,
|
| 1860 |
+
"loss": 0.1826,
|
| 1861 |
"step": 2350
|
| 1862 |
},
|
| 1863 |
{
|
| 1864 |
+
"epoch": 8.68,
|
| 1865 |
+
"grad_norm": 5.348261833190918,
|
| 1866 |
+
"learning_rate": 1.3382352941176471e-05,
|
| 1867 |
+
"loss": 0.2772,
|
| 1868 |
"step": 2360
|
| 1869 |
},
|
| 1870 |
{
|
| 1871 |
+
"epoch": 8.71,
|
| 1872 |
+
"grad_norm": 6.000495433807373,
|
| 1873 |
+
"learning_rate": 1.3014705882352943e-05,
|
| 1874 |
+
"loss": 0.298,
|
| 1875 |
"step": 2370
|
| 1876 |
},
|
| 1877 |
{
|
| 1878 |
+
"epoch": 8.75,
|
| 1879 |
+
"grad_norm": 4.316232204437256,
|
| 1880 |
+
"learning_rate": 1.2647058823529412e-05,
|
| 1881 |
+
"loss": 0.1962,
|
| 1882 |
"step": 2380
|
| 1883 |
},
|
| 1884 |
{
|
| 1885 |
+
"epoch": 8.79,
|
| 1886 |
+
"grad_norm": 4.092772483825684,
|
| 1887 |
+
"learning_rate": 1.2279411764705883e-05,
|
| 1888 |
+
"loss": 0.2675,
|
| 1889 |
"step": 2390
|
| 1890 |
},
|
| 1891 |
{
|
| 1892 |
+
"epoch": 8.82,
|
| 1893 |
+
"grad_norm": 1.2714035511016846,
|
| 1894 |
+
"learning_rate": 1.1911764705882354e-05,
|
| 1895 |
+
"loss": 0.3046,
|
| 1896 |
"step": 2400
|
| 1897 |
},
|
| 1898 |
{
|
| 1899 |
+
"epoch": 8.82,
|
| 1900 |
+
"eval_accuracy": 0.8712051517939282,
|
| 1901 |
+
"eval_loss": 0.4486115574836731,
|
| 1902 |
+
"eval_runtime": 14.6395,
|
| 1903 |
+
"eval_samples_per_second": 74.251,
|
| 1904 |
+
"eval_steps_per_second": 9.29,
|
| 1905 |
"step": 2400
|
| 1906 |
},
|
| 1907 |
{
|
| 1908 |
+
"epoch": 8.86,
|
| 1909 |
+
"grad_norm": 1.3313093185424805,
|
| 1910 |
+
"learning_rate": 1.1544117647058824e-05,
|
| 1911 |
+
"loss": 0.1964,
|
| 1912 |
"step": 2410
|
| 1913 |
},
|
| 1914 |
{
|
| 1915 |
+
"epoch": 8.9,
|
| 1916 |
+
"grad_norm": 1.0101081132888794,
|
| 1917 |
+
"learning_rate": 1.1176470588235295e-05,
|
| 1918 |
+
"loss": 0.1616,
|
| 1919 |
"step": 2420
|
| 1920 |
},
|
| 1921 |
{
|
| 1922 |
+
"epoch": 8.93,
|
| 1923 |
+
"grad_norm": 3.4032843112945557,
|
| 1924 |
+
"learning_rate": 1.0808823529411765e-05,
|
| 1925 |
+
"loss": 0.2101,
|
| 1926 |
"step": 2430
|
| 1927 |
},
|
| 1928 |
{
|
| 1929 |
+
"epoch": 8.97,
|
| 1930 |
+
"grad_norm": 5.762178897857666,
|
| 1931 |
+
"learning_rate": 1.0441176470588236e-05,
|
| 1932 |
+
"loss": 0.221,
|
| 1933 |
"step": 2440
|
| 1934 |
},
|
| 1935 |
{
|
| 1936 |
+
"epoch": 9.01,
|
| 1937 |
+
"grad_norm": 7.056640625,
|
| 1938 |
+
"learning_rate": 1.0073529411764707e-05,
|
| 1939 |
+
"loss": 0.2376,
|
| 1940 |
"step": 2450
|
| 1941 |
},
|
| 1942 |
{
|
| 1943 |
+
"epoch": 9.04,
|
| 1944 |
+
"grad_norm": 2.540907859802246,
|
| 1945 |
+
"learning_rate": 9.705882352941177e-06,
|
| 1946 |
+
"loss": 0.1109,
|
| 1947 |
"step": 2460
|
| 1948 |
},
|
| 1949 |
{
|
| 1950 |
+
"epoch": 9.08,
|
| 1951 |
+
"grad_norm": 4.972515106201172,
|
| 1952 |
+
"learning_rate": 9.338235294117648e-06,
|
| 1953 |
+
"loss": 0.1603,
|
| 1954 |
"step": 2470
|
| 1955 |
},
|
| 1956 |
{
|
| 1957 |
+
"epoch": 9.12,
|
| 1958 |
+
"grad_norm": 5.1937150955200195,
|
| 1959 |
+
"learning_rate": 8.970588235294119e-06,
|
| 1960 |
+
"loss": 0.2862,
|
| 1961 |
"step": 2480
|
| 1962 |
},
|
| 1963 |
{
|
| 1964 |
+
"epoch": 9.15,
|
| 1965 |
+
"grad_norm": 8.317940711975098,
|
| 1966 |
+
"learning_rate": 8.60294117647059e-06,
|
| 1967 |
+
"loss": 0.2001,
|
| 1968 |
"step": 2490
|
| 1969 |
},
|
| 1970 |
{
|
| 1971 |
+
"epoch": 9.19,
|
| 1972 |
+
"grad_norm": 5.177839756011963,
|
| 1973 |
+
"learning_rate": 8.23529411764706e-06,
|
| 1974 |
+
"loss": 0.2271,
|
| 1975 |
"step": 2500
|
| 1976 |
},
|
| 1977 |
{
|
| 1978 |
+
"epoch": 9.19,
|
| 1979 |
+
"eval_accuracy": 0.8739650413983441,
|
| 1980 |
+
"eval_loss": 0.4654560089111328,
|
| 1981 |
+
"eval_runtime": 14.713,
|
| 1982 |
+
"eval_samples_per_second": 73.88,
|
| 1983 |
+
"eval_steps_per_second": 9.244,
|
| 1984 |
"step": 2500
|
| 1985 |
},
|
| 1986 |
{
|
| 1987 |
+
"epoch": 9.23,
|
| 1988 |
+
"grad_norm": 13.221484184265137,
|
| 1989 |
+
"learning_rate": 7.86764705882353e-06,
|
| 1990 |
+
"loss": 0.1775,
|
| 1991 |
"step": 2510
|
| 1992 |
},
|
| 1993 |
{
|
| 1994 |
+
"epoch": 9.26,
|
| 1995 |
+
"grad_norm": 1.31935715675354,
|
| 1996 |
+
"learning_rate": 7.5e-06,
|
| 1997 |
+
"loss": 0.1663,
|
| 1998 |
"step": 2520
|
| 1999 |
},
|
| 2000 |
{
|
| 2001 |
+
"epoch": 9.3,
|
| 2002 |
+
"grad_norm": 2.4796762466430664,
|
| 2003 |
+
"learning_rate": 7.132352941176471e-06,
|
| 2004 |
+
"loss": 0.2448,
|
| 2005 |
"step": 2530
|
| 2006 |
},
|
| 2007 |
{
|
| 2008 |
+
"epoch": 9.34,
|
| 2009 |
+
"grad_norm": 6.287135601043701,
|
| 2010 |
+
"learning_rate": 6.7647058823529414e-06,
|
| 2011 |
+
"loss": 0.2089,
|
| 2012 |
"step": 2540
|
| 2013 |
},
|
| 2014 |
{
|
| 2015 |
+
"epoch": 9.38,
|
| 2016 |
+
"grad_norm": 3.7011852264404297,
|
| 2017 |
+
"learning_rate": 6.397058823529411e-06,
|
| 2018 |
+
"loss": 0.2437,
|
| 2019 |
"step": 2550
|
| 2020 |
},
|
| 2021 |
{
|
| 2022 |
+
"epoch": 9.41,
|
| 2023 |
+
"grad_norm": 2.945911407470703,
|
| 2024 |
+
"learning_rate": 6.029411764705883e-06,
|
| 2025 |
+
"loss": 0.279,
|
| 2026 |
"step": 2560
|
| 2027 |
},
|
| 2028 |
{
|
| 2029 |
+
"epoch": 9.45,
|
| 2030 |
+
"grad_norm": 4.291398525238037,
|
| 2031 |
+
"learning_rate": 5.661764705882353e-06,
|
| 2032 |
+
"loss": 0.1282,
|
| 2033 |
"step": 2570
|
| 2034 |
},
|
| 2035 |
{
|
| 2036 |
+
"epoch": 9.49,
|
| 2037 |
+
"grad_norm": 1.904885172843933,
|
| 2038 |
+
"learning_rate": 5.294117647058824e-06,
|
| 2039 |
+
"loss": 0.2205,
|
| 2040 |
"step": 2580
|
| 2041 |
},
|
| 2042 |
{
|
| 2043 |
+
"epoch": 9.52,
|
| 2044 |
+
"grad_norm": 3.374128818511963,
|
| 2045 |
+
"learning_rate": 4.9264705882352945e-06,
|
| 2046 |
+
"loss": 0.1051,
|
| 2047 |
"step": 2590
|
| 2048 |
},
|
| 2049 |
{
|
| 2050 |
+
"epoch": 9.56,
|
| 2051 |
+
"grad_norm": 2.0110669136047363,
|
| 2052 |
+
"learning_rate": 4.558823529411764e-06,
|
| 2053 |
+
"loss": 0.2719,
|
| 2054 |
"step": 2600
|
| 2055 |
},
|
| 2056 |
{
|
| 2057 |
+
"epoch": 9.56,
|
| 2058 |
+
"eval_accuracy": 0.8758049678012879,
|
| 2059 |
+
"eval_loss": 0.4518091082572937,
|
| 2060 |
+
"eval_runtime": 14.8455,
|
| 2061 |
+
"eval_samples_per_second": 73.221,
|
| 2062 |
+
"eval_steps_per_second": 9.161,
|
| 2063 |
"step": 2600
|
| 2064 |
},
|
| 2065 |
{
|
| 2066 |
+
"epoch": 9.6,
|
| 2067 |
+
"grad_norm": 0.11158547550439835,
|
| 2068 |
+
"learning_rate": 4.191176470588235e-06,
|
| 2069 |
+
"loss": 0.2106,
|
| 2070 |
"step": 2610
|
| 2071 |
},
|
| 2072 |
{
|
| 2073 |
+
"epoch": 9.63,
|
| 2074 |
+
"grad_norm": 10.953994750976562,
|
| 2075 |
+
"learning_rate": 3.823529411764706e-06,
|
| 2076 |
+
"loss": 0.2553,
|
| 2077 |
"step": 2620
|
| 2078 |
},
|
| 2079 |
{
|
| 2080 |
+
"epoch": 9.67,
|
| 2081 |
+
"grad_norm": 0.048644546419382095,
|
| 2082 |
+
"learning_rate": 3.455882352941177e-06,
|
| 2083 |
+
"loss": 0.0943,
|
| 2084 |
"step": 2630
|
| 2085 |
},
|
| 2086 |
{
|
| 2087 |
+
"epoch": 9.71,
|
| 2088 |
+
"grad_norm": 3.4947116374969482,
|
| 2089 |
+
"learning_rate": 3.0882352941176472e-06,
|
| 2090 |
+
"loss": 0.1888,
|
| 2091 |
"step": 2640
|
| 2092 |
},
|
| 2093 |
{
|
| 2094 |
+
"epoch": 9.74,
|
| 2095 |
+
"grad_norm": 1.8250272274017334,
|
| 2096 |
+
"learning_rate": 2.720588235294118e-06,
|
| 2097 |
+
"loss": 0.2192,
|
| 2098 |
"step": 2650
|
| 2099 |
},
|
| 2100 |
{
|
| 2101 |
+
"epoch": 9.78,
|
| 2102 |
+
"grad_norm": 10.61952018737793,
|
| 2103 |
+
"learning_rate": 2.3529411764705885e-06,
|
| 2104 |
+
"loss": 0.169,
|
| 2105 |
"step": 2660
|
| 2106 |
},
|
| 2107 |
{
|
| 2108 |
+
"epoch": 9.82,
|
| 2109 |
+
"grad_norm": 15.31037425994873,
|
| 2110 |
+
"learning_rate": 1.9852941176470586e-06,
|
| 2111 |
+
"loss": 0.257,
|
| 2112 |
"step": 2670
|
| 2113 |
},
|
| 2114 |
{
|
| 2115 |
+
"epoch": 9.85,
|
| 2116 |
+
"grad_norm": 7.124182224273682,
|
| 2117 |
+
"learning_rate": 1.6176470588235297e-06,
|
| 2118 |
+
"loss": 0.3049,
|
| 2119 |
"step": 2680
|
| 2120 |
},
|
| 2121 |
{
|
| 2122 |
+
"epoch": 9.89,
|
| 2123 |
+
"grad_norm": 0.369510680437088,
|
| 2124 |
+
"learning_rate": 1.25e-06,
|
| 2125 |
+
"loss": 0.1468,
|
| 2126 |
"step": 2690
|
| 2127 |
},
|
| 2128 |
{
|
| 2129 |
+
"epoch": 9.93,
|
| 2130 |
+
"grad_norm": 6.830472469329834,
|
| 2131 |
+
"learning_rate": 8.823529411764706e-07,
|
| 2132 |
+
"loss": 0.1563,
|
| 2133 |
"step": 2700
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
+
"epoch": 9.93,
|
| 2137 |
+
"eval_accuracy": 0.8767249310027599,
|
| 2138 |
+
"eval_loss": 0.4466676115989685,
|
| 2139 |
+
"eval_runtime": 14.9125,
|
| 2140 |
+
"eval_samples_per_second": 72.892,
|
| 2141 |
+
"eval_steps_per_second": 9.12,
|
| 2142 |
"step": 2700
|
| 2143 |
},
|
| 2144 |
{
|
| 2145 |
+
"epoch": 9.96,
|
| 2146 |
+
"grad_norm": 2.819904088973999,
|
| 2147 |
+
"learning_rate": 5.147058823529412e-07,
|
| 2148 |
+
"loss": 0.2226,
|
| 2149 |
"step": 2710
|
| 2150 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2151 |
{
|
| 2152 |
"epoch": 10.0,
|
| 2153 |
+
"grad_norm": 3.1663293838500977,
|
| 2154 |
+
"learning_rate": 1.4705882352941178e-07,
|
| 2155 |
+
"loss": 0.205,
|
| 2156 |
+
"step": 2720
|
| 2157 |
},
|
| 2158 |
{
|
| 2159 |
"epoch": 10.0,
|
| 2160 |
+
"step": 2720,
|
| 2161 |
+
"total_flos": 3.3672683695343616e+18,
|
| 2162 |
+
"train_loss": 0.4566012972418,
|
| 2163 |
+
"train_runtime": 1529.7234,
|
| 2164 |
+
"train_samples_per_second": 28.404,
|
| 2165 |
+
"train_steps_per_second": 1.778
|
| 2166 |
}
|
| 2167 |
],
|
| 2168 |
"logging_steps": 10,
|
| 2169 |
+
"max_steps": 2720,
|
| 2170 |
"num_input_tokens_seen": 0,
|
| 2171 |
"num_train_epochs": 10,
|
| 2172 |
"save_steps": 100,
|
| 2173 |
+
"total_flos": 3.3672683695343616e+18,
|
| 2174 |
"train_batch_size": 16,
|
| 2175 |
"trial_name": null,
|
| 2176 |
"trial_params": null
|