🍻 cheers
Browse files- README.md +6 -5
- all_results.json +10 -10
- eval_results.json +5 -5
- runs/May05_19-40-36_9c8f6c45512e/events.out.tfevents.1714939591.9c8f6c45512e.34.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +1303 -1254
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: 0.8752556237218814
|
| 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: 0.4228
|
| 37 |
+
- Accuracy: 0.8753
|
| 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": 9.
|
| 8 |
-
"total_flos": 3.
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 1.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8752556237218814,
|
| 4 |
+
"eval_loss": 0.42281925678253174,
|
| 5 |
+
"eval_runtime": 13.2413,
|
| 6 |
+
"eval_samples_per_second": 73.86,
|
| 7 |
+
"eval_steps_per_second": 9.289,
|
| 8 |
+
"total_flos": 3.451740694569861e+18,
|
| 9 |
+
"train_loss": 0.45333294201922675,
|
| 10 |
+
"train_runtime": 1524.4216,
|
| 11 |
+
"train_samples_per_second": 29.218,
|
| 12 |
+
"train_steps_per_second": 1.83
|
| 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": 9.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8752556237218814,
|
| 4 |
+
"eval_loss": 0.42281925678253174,
|
| 5 |
+
"eval_runtime": 13.2413,
|
| 6 |
+
"eval_samples_per_second": 73.86,
|
| 7 |
+
"eval_steps_per_second": 9.289
|
| 8 |
}
|
runs/May05_19-40-36_9c8f6c45512e/events.out.tfevents.1714939591.9c8f6c45512e.34.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:549c368dc602d66ed7ad02f0dd83c42be596fe51a5088662e3162ec45d735d25
|
| 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":
|
| 7 |
-
"train_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 10.0,
|
| 3 |
+
"total_flos": 3.451740694569861e+18,
|
| 4 |
+
"train_loss": 0.45333294201922675,
|
| 5 |
+
"train_runtime": 1524.4216,
|
| 6 |
+
"train_samples_per_second": 29.218,
|
| 7 |
+
"train_steps_per_second": 1.83
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,2176 +1,2225 @@
|
|
| 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": 0.04,
|
| 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": 2.1701,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.11,
|
| 27 |
-
"grad_norm": 1.
|
| 28 |
-
"learning_rate": 9.
|
| 29 |
-
"loss": 2.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 9.
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.18,
|
| 41 |
-
"grad_norm": 2.
|
| 42 |
-
"learning_rate": 9.
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.22,
|
| 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": 0.29,
|
| 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": 7.
|
| 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": 2.
|
| 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": 2.
|
| 121 |
-
"learning_rate": 9.
|
| 122 |
-
"loss": 1.
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 0.
|
| 127 |
-
"grad_norm": 3.
|
| 128 |
-
"learning_rate": 9.
|
| 129 |
-
"loss": 1.
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
-
"epoch": 0.
|
| 134 |
-
"grad_norm":
|
| 135 |
-
"learning_rate": 9.
|
| 136 |
-
"loss":
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch": 0.
|
| 141 |
-
"grad_norm": 2.
|
| 142 |
-
"learning_rate": 9.
|
| 143 |
-
"loss":
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
-
"epoch": 0.
|
| 148 |
-
"grad_norm":
|
| 149 |
-
"learning_rate": 9.
|
| 150 |
-
"loss": 1.
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
-
"epoch": 0.
|
| 155 |
-
"grad_norm":
|
| 156 |
-
"learning_rate": 9.
|
| 157 |
-
"loss": 0.
|
| 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": 9.
|
| 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": 3.
|
| 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":
|
| 193 |
-
"learning_rate": 9.
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch": 0.
|
| 199 |
-
"grad_norm":
|
| 200 |
-
"learning_rate": 9.
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"epoch": 0.
|
| 206 |
-
"grad_norm": 5.
|
| 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": 1.
|
| 220 |
-
"grad_norm":
|
| 221 |
-
"learning_rate":
|
| 222 |
-
"loss": 0.
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
-
"epoch": 1.
|
| 227 |
-
"grad_norm": 3.
|
| 228 |
-
"learning_rate": 8.
|
| 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": 9.
|
| 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": 4.
|
| 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": 74.
|
| 324 |
-
"eval_steps_per_second": 9.
|
| 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": 3.
|
| 337 |
-
"learning_rate": 8.
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch": 1.
|
| 343 |
-
"grad_norm": 5.
|
| 344 |
-
"learning_rate": 8.
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch": 1.
|
| 350 |
-
"grad_norm":
|
| 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": 73.
|
| 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": 4.
|
| 416 |
-
"learning_rate": 8.
|
| 417 |
-
"loss": 0.
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
-
"epoch": 1.
|
| 422 |
-
"grad_norm": 2.
|
| 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":
|
| 437 |
-
"learning_rate":
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
-
"epoch": 2.
|
| 443 |
-
"grad_norm":
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
-
"epoch": 2.
|
| 450 |
-
"grad_norm":
|
| 451 |
-
"learning_rate": 7.
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch": 2.
|
| 457 |
-
"grad_norm":
|
| 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": 74.
|
| 482 |
-
"eval_steps_per_second": 9.
|
| 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": 3.
|
| 523 |
-
"learning_rate": 7.
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
-
"epoch": 2.
|
| 529 |
-
"grad_norm":
|
| 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":
|
| 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": 74.
|
| 561 |
-
"eval_steps_per_second": 9.
|
| 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": 6.
|
| 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":
|
| 609 |
-
"learning_rate": 7.
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 2.
|
| 615 |
-
"grad_norm":
|
| 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":
|
| 639 |
-
"eval_samples_per_second": 74.
|
| 640 |
-
"eval_steps_per_second": 9.
|
| 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":
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
-
"epoch": 3.
|
| 666 |
-
"grad_norm":
|
| 667 |
-
"learning_rate":
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
-
"epoch": 3.
|
| 673 |
-
"grad_norm":
|
| 674 |
-
"learning_rate": 6.
|
| 675 |
-
"loss": 0.
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
-
"epoch": 3.
|
| 680 |
-
"grad_norm":
|
| 681 |
-
"learning_rate": 6.
|
| 682 |
-
"loss": 0.
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
-
"epoch": 3.
|
| 687 |
-
"grad_norm": 7.
|
| 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":
|
| 702 |
-
"learning_rate": 6.
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 890
|
| 705 |
},
|
| 706 |
{
|
| 707 |
-
"epoch": 3.
|
| 708 |
-
"grad_norm":
|
| 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": 9.
|
| 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":
|
| 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": 6.
|
| 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": 4.
|
| 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":
|
| 798 |
-
"eval_steps_per_second": 9.
|
| 799 |
"step": 1000
|
| 800 |
},
|
| 801 |
{
|
| 802 |
-
"epoch": 3.
|
| 803 |
-
"grad_norm": 3.
|
| 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": 5.
|
| 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": 6.
|
| 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": 9.
|
| 878 |
"step": 1100
|
| 879 |
},
|
| 880 |
{
|
| 881 |
-
"epoch":
|
| 882 |
-
"grad_norm":
|
| 883 |
-
"learning_rate":
|
| 884 |
-
"loss": 0.
|
| 885 |
"step": 1110
|
| 886 |
},
|
| 887 |
{
|
| 888 |
-
"epoch": 4.
|
| 889 |
-
"grad_norm": 3.
|
| 890 |
-
"learning_rate": 5.
|
| 891 |
-
"loss": 0.
|
| 892 |
"step": 1120
|
| 893 |
},
|
| 894 |
{
|
| 895 |
-
"epoch": 4.
|
| 896 |
-
"grad_norm":
|
| 897 |
-
"learning_rate": 5.
|
| 898 |
-
"loss": 0.
|
| 899 |
"step": 1130
|
| 900 |
},
|
| 901 |
{
|
| 902 |
-
"epoch": 4.
|
| 903 |
-
"grad_norm":
|
| 904 |
-
"learning_rate": 5.
|
| 905 |
-
"loss": 0.
|
| 906 |
"step": 1140
|
| 907 |
},
|
| 908 |
{
|
| 909 |
-
"epoch": 4.
|
| 910 |
-
"grad_norm":
|
| 911 |
-
"learning_rate": 5.
|
| 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": 6.
|
| 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": 74.
|
| 956 |
-
"eval_steps_per_second": 9.
|
| 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": 9.
|
| 1036 |
"step": 1300
|
| 1037 |
},
|
| 1038 |
{
|
| 1039 |
-
"epoch": 4.
|
| 1040 |
-
"grad_norm": 3.
|
| 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": 3.
|
| 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":
|
| 1076 |
-
"learning_rate": 5.
|
| 1077 |
-
"loss": 0.
|
| 1078 |
"step": 1360
|
| 1079 |
},
|
| 1080 |
{
|
| 1081 |
-
"epoch":
|
| 1082 |
-
"grad_norm": 4.
|
| 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": 5.
|
| 1103 |
-
"grad_norm":
|
| 1104 |
-
"learning_rate": 4.
|
| 1105 |
-
"loss": 0.
|
| 1106 |
"step": 1400
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
-
"epoch": 5.
|
| 1110 |
-
"eval_accuracy": 0.
|
| 1111 |
-
"eval_loss": 0.
|
| 1112 |
-
"eval_runtime": 14.
|
| 1113 |
-
"eval_samples_per_second":
|
| 1114 |
-
"eval_steps_per_second":
|
| 1115 |
"step": 1400
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
-
"epoch": 5.
|
| 1119 |
-
"grad_norm":
|
| 1120 |
-
"learning_rate": 4.
|
| 1121 |
-
"loss": 0.
|
| 1122 |
"step": 1410
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
-
"epoch": 5.
|
| 1126 |
-
"grad_norm":
|
| 1127 |
-
"learning_rate": 4.
|
| 1128 |
-
"loss": 0.
|
| 1129 |
"step": 1420
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
-
"epoch": 5.
|
| 1133 |
-
"grad_norm":
|
| 1134 |
-
"learning_rate": 4.
|
| 1135 |
-
"loss": 0.
|
| 1136 |
"step": 1430
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
-
"epoch": 5.
|
| 1140 |
-
"grad_norm":
|
| 1141 |
-
"learning_rate": 4.
|
| 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":
|
| 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": 74.
|
| 1193 |
-
"eval_steps_per_second": 9.
|
| 1194 |
"step": 1500
|
| 1195 |
},
|
| 1196 |
{
|
| 1197 |
-
"epoch": 5.
|
| 1198 |
-
"grad_norm":
|
| 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":
|
| 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": 0.
|
| 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": 3.
|
| 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": 74.
|
| 1272 |
-
"eval_steps_per_second": 9.
|
| 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": 4.
|
| 1320 |
-
"learning_rate":
|
| 1321 |
-
"loss": 0.
|
| 1322 |
"step": 1670
|
| 1323 |
},
|
| 1324 |
{
|
| 1325 |
-
"epoch": 6.
|
| 1326 |
-
"grad_norm":
|
| 1327 |
-
"learning_rate": 3.
|
| 1328 |
-
"loss": 0.
|
| 1329 |
"step": 1680
|
| 1330 |
},
|
| 1331 |
{
|
| 1332 |
-
"epoch": 6.
|
| 1333 |
-
"grad_norm":
|
| 1334 |
-
"learning_rate": 3.
|
| 1335 |
-
"loss": 0.
|
| 1336 |
"step": 1690
|
| 1337 |
},
|
| 1338 |
{
|
| 1339 |
-
"epoch": 6.
|
| 1340 |
-
"grad_norm":
|
| 1341 |
-
"learning_rate": 3.
|
| 1342 |
-
"loss": 0.
|
| 1343 |
"step": 1700
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
-
"epoch": 6.
|
| 1347 |
-
"eval_accuracy": 0.
|
| 1348 |
-
"eval_loss": 0.
|
| 1349 |
-
"eval_runtime":
|
| 1350 |
-
"eval_samples_per_second":
|
| 1351 |
-
"eval_steps_per_second": 9.
|
| 1352 |
"step": 1700
|
| 1353 |
},
|
| 1354 |
{
|
| 1355 |
-
"epoch": 6.
|
| 1356 |
-
"grad_norm":
|
| 1357 |
-
"learning_rate": 3.
|
| 1358 |
-
"loss": 0.
|
| 1359 |
"step": 1710
|
| 1360 |
},
|
| 1361 |
{
|
| 1362 |
-
"epoch": 6.
|
| 1363 |
-
"grad_norm":
|
| 1364 |
-
"learning_rate": 3.
|
| 1365 |
-
"loss": 0.
|
| 1366 |
"step": 1720
|
| 1367 |
},
|
| 1368 |
{
|
| 1369 |
-
"epoch": 6.
|
| 1370 |
-
"grad_norm":
|
| 1371 |
-
"learning_rate": 3.
|
| 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": 2.
|
| 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": 73.
|
| 1430 |
-
"eval_steps_per_second": 9.
|
| 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":
|
| 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":
|
| 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": 74.
|
| 1509 |
-
"eval_steps_per_second": 9.
|
| 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":
|
| 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":
|
| 1543 |
-
"learning_rate":
|
| 1544 |
-
"loss": 0.
|
| 1545 |
"step": 1950
|
| 1546 |
},
|
| 1547 |
{
|
| 1548 |
-
"epoch": 7.
|
| 1549 |
-
"grad_norm":
|
| 1550 |
-
"learning_rate": 2.
|
| 1551 |
-
"loss": 0.
|
| 1552 |
"step": 1960
|
| 1553 |
},
|
| 1554 |
{
|
| 1555 |
-
"epoch": 7.
|
| 1556 |
-
"grad_norm":
|
| 1557 |
-
"learning_rate": 2.
|
| 1558 |
-
"loss": 0.
|
| 1559 |
"step": 1970
|
| 1560 |
},
|
| 1561 |
{
|
| 1562 |
-
"epoch": 7.
|
| 1563 |
-
"grad_norm":
|
| 1564 |
-
"learning_rate": 2.
|
| 1565 |
-
"loss": 0.
|
| 1566 |
"step": 1980
|
| 1567 |
},
|
| 1568 |
{
|
| 1569 |
-
"epoch": 7.
|
| 1570 |
-
"grad_norm":
|
| 1571 |
-
"learning_rate": 2.
|
| 1572 |
-
"loss": 0.
|
| 1573 |
"step": 1990
|
| 1574 |
},
|
| 1575 |
{
|
| 1576 |
-
"epoch": 7.
|
| 1577 |
-
"grad_norm":
|
| 1578 |
-
"learning_rate": 2.
|
| 1579 |
-
"loss": 0.
|
| 1580 |
"step": 2000
|
| 1581 |
},
|
| 1582 |
{
|
| 1583 |
-
"epoch": 7.
|
| 1584 |
-
"eval_accuracy": 0.
|
| 1585 |
-
"eval_loss": 0.
|
| 1586 |
-
"eval_runtime":
|
| 1587 |
-
"eval_samples_per_second":
|
| 1588 |
-
"eval_steps_per_second": 9.
|
| 1589 |
"step": 2000
|
| 1590 |
},
|
| 1591 |
{
|
| 1592 |
-
"epoch": 7.
|
| 1593 |
-
"grad_norm":
|
| 1594 |
-
"learning_rate": 2.
|
| 1595 |
-
"loss": 0.
|
| 1596 |
"step": 2010
|
| 1597 |
},
|
| 1598 |
{
|
| 1599 |
-
"epoch": 7.
|
| 1600 |
-
"grad_norm":
|
| 1601 |
-
"learning_rate": 2.
|
| 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":
|
| 1643 |
-
"learning_rate": 2.
|
| 1644 |
-
"loss": 0.
|
| 1645 |
"step": 2080
|
| 1646 |
},
|
| 1647 |
{
|
| 1648 |
-
"epoch": 7.
|
| 1649 |
-
"grad_norm":
|
| 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": 9.
|
| 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": 6.
|
| 1694 |
-
"learning_rate": 2.
|
| 1695 |
-
"loss": 0.
|
| 1696 |
"step": 2140
|
| 1697 |
},
|
| 1698 |
{
|
| 1699 |
-
"epoch": 7.
|
| 1700 |
-
"grad_norm":
|
| 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": 9.
|
| 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":
|
| 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": 8.
|
| 1772 |
-
"grad_norm":
|
| 1773 |
-
"learning_rate": 1.
|
| 1774 |
-
"loss": 0.
|
| 1775 |
"step": 2240
|
| 1776 |
},
|
| 1777 |
{
|
| 1778 |
-
"epoch": 8.
|
| 1779 |
-
"grad_norm":
|
| 1780 |
-
"learning_rate": 1.
|
| 1781 |
-
"loss": 0.
|
| 1782 |
"step": 2250
|
| 1783 |
},
|
| 1784 |
{
|
| 1785 |
-
"epoch": 8.
|
| 1786 |
-
"grad_norm":
|
| 1787 |
-
"learning_rate": 1.
|
| 1788 |
-
"loss": 0.
|
| 1789 |
"step": 2260
|
| 1790 |
},
|
| 1791 |
{
|
| 1792 |
-
"epoch": 8.
|
| 1793 |
-
"grad_norm":
|
| 1794 |
-
"learning_rate": 1.
|
| 1795 |
-
"loss": 0.
|
| 1796 |
"step": 2270
|
| 1797 |
},
|
| 1798 |
{
|
| 1799 |
-
"epoch": 8.
|
| 1800 |
-
"grad_norm":
|
| 1801 |
-
"learning_rate": 1.
|
| 1802 |
-
"loss": 0.
|
| 1803 |
"step": 2280
|
| 1804 |
},
|
| 1805 |
{
|
| 1806 |
-
"epoch": 8.
|
| 1807 |
-
"grad_norm":
|
| 1808 |
-
"learning_rate": 1.
|
| 1809 |
-
"loss": 0.
|
| 1810 |
"step": 2290
|
| 1811 |
},
|
| 1812 |
{
|
| 1813 |
-
"epoch": 8.
|
| 1814 |
-
"grad_norm":
|
| 1815 |
-
"learning_rate": 1.
|
| 1816 |
-
"loss": 0.
|
| 1817 |
"step": 2300
|
| 1818 |
},
|
| 1819 |
{
|
| 1820 |
-
"epoch": 8.
|
| 1821 |
-
"eval_accuracy": 0.
|
| 1822 |
-
"eval_loss": 0.
|
| 1823 |
-
"eval_runtime":
|
| 1824 |
-
"eval_samples_per_second":
|
| 1825 |
-
"eval_steps_per_second": 9.
|
| 1826 |
"step": 2300
|
| 1827 |
},
|
| 1828 |
{
|
| 1829 |
-
"epoch": 8.
|
| 1830 |
-
"grad_norm":
|
| 1831 |
-
"learning_rate": 1.
|
| 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":
|
| 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": 9.
|
| 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": 73.
|
| 1983 |
-
"eval_steps_per_second": 9.
|
| 1984 |
"step": 2500
|
| 1985 |
},
|
| 1986 |
{
|
| 1987 |
-
"epoch": 9.
|
| 1988 |
-
"grad_norm":
|
| 1989 |
-
"learning_rate":
|
| 1990 |
-
"loss": 0.
|
| 1991 |
"step": 2510
|
| 1992 |
},
|
| 1993 |
{
|
| 1994 |
-
"epoch": 9.
|
| 1995 |
-
"grad_norm":
|
| 1996 |
-
"learning_rate":
|
| 1997 |
-
"loss": 0.
|
| 1998 |
"step": 2520
|
| 1999 |
},
|
| 2000 |
{
|
| 2001 |
-
"epoch": 9.
|
| 2002 |
-
"grad_norm":
|
| 2003 |
-
"learning_rate":
|
| 2004 |
-
"loss": 0.
|
| 2005 |
"step": 2530
|
| 2006 |
},
|
| 2007 |
{
|
| 2008 |
-
"epoch": 9.
|
| 2009 |
-
"grad_norm":
|
| 2010 |
-
"learning_rate":
|
| 2011 |
-
"loss": 0.
|
| 2012 |
"step": 2540
|
| 2013 |
},
|
| 2014 |
{
|
| 2015 |
-
"epoch": 9.
|
| 2016 |
-
"grad_norm":
|
| 2017 |
-
"learning_rate":
|
| 2018 |
-
"loss": 0.
|
| 2019 |
"step": 2550
|
| 2020 |
},
|
| 2021 |
{
|
| 2022 |
-
"epoch": 9.
|
| 2023 |
-
"grad_norm":
|
| 2024 |
-
"learning_rate":
|
| 2025 |
-
"loss": 0.
|
| 2026 |
"step": 2560
|
| 2027 |
},
|
| 2028 |
{
|
| 2029 |
-
"epoch": 9.
|
| 2030 |
-
"grad_norm":
|
| 2031 |
-
"learning_rate":
|
| 2032 |
-
"loss": 0.
|
| 2033 |
"step": 2570
|
| 2034 |
},
|
| 2035 |
{
|
| 2036 |
-
"epoch": 9.
|
| 2037 |
-
"grad_norm":
|
| 2038 |
-
"learning_rate":
|
| 2039 |
-
"loss": 0.
|
| 2040 |
"step": 2580
|
| 2041 |
},
|
| 2042 |
{
|
| 2043 |
-
"epoch": 9.
|
| 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": 73.
|
| 2062 |
-
"eval_steps_per_second": 9.
|
| 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":
|
| 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": 9.
|
| 2142 |
"step": 2700
|
| 2143 |
},
|
| 2144 |
{
|
| 2145 |
-
"epoch": 9.
|
| 2146 |
-
"grad_norm":
|
| 2147 |
-
"learning_rate":
|
| 2148 |
-
"loss": 0.
|
| 2149 |
"step": 2710
|
| 2150 |
},
|
| 2151 |
{
|
| 2152 |
-
"epoch":
|
| 2153 |
-
"grad_norm":
|
| 2154 |
-
"learning_rate":
|
| 2155 |
-
"loss": 0.
|
| 2156 |
"step": 2720
|
| 2157 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2158 |
{
|
| 2159 |
"epoch": 10.0,
|
| 2160 |
-
"step":
|
| 2161 |
-
"total_flos": 3.
|
| 2162 |
-
"train_loss": 0.
|
| 2163 |
-
"train_runtime":
|
| 2164 |
-
"train_samples_per_second":
|
| 2165 |
-
"train_steps_per_second": 1.
|
| 2166 |
}
|
| 2167 |
],
|
| 2168 |
"logging_steps": 10,
|
| 2169 |
-
"max_steps":
|
| 2170 |
"num_input_tokens_seen": 0,
|
| 2171 |
"num_train_epochs": 10,
|
| 2172 |
"save_steps": 100,
|
| 2173 |
-
"total_flos": 3.
|
| 2174 |
"train_batch_size": 16,
|
| 2175 |
"trial_name": null,
|
| 2176 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.42281925678253174,
|
| 3 |
+
"best_model_checkpoint": "Action_all_10_class/checkpoint-1400",
|
| 4 |
"epoch": 10.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 2790,
|
| 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": 1.7864842414855957,
|
| 14 |
+
"learning_rate": 9.96415770609319e-05,
|
| 15 |
+
"loss": 2.2581,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.07,
|
| 20 |
+
"grad_norm": 1.855455756187439,
|
| 21 |
+
"learning_rate": 9.928315412186381e-05,
|
| 22 |
"loss": 2.1701,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.11,
|
| 27 |
+
"grad_norm": 1.9831140041351318,
|
| 28 |
+
"learning_rate": 9.892473118279571e-05,
|
| 29 |
+
"loss": 2.0292,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.14,
|
| 34 |
+
"grad_norm": 2.3516452312469482,
|
| 35 |
+
"learning_rate": 9.85663082437276e-05,
|
| 36 |
+
"loss": 1.9225,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.18,
|
| 41 |
+
"grad_norm": 2.3717079162597656,
|
| 42 |
+
"learning_rate": 9.820788530465951e-05,
|
| 43 |
+
"loss": 1.7377,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.22,
|
| 48 |
+
"grad_norm": 2.2911033630371094,
|
| 49 |
+
"learning_rate": 9.78494623655914e-05,
|
| 50 |
+
"loss": 1.6861,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.25,
|
| 55 |
+
"grad_norm": 3.3348388671875,
|
| 56 |
+
"learning_rate": 9.74910394265233e-05,
|
| 57 |
+
"loss": 1.5431,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.29,
|
| 62 |
+
"grad_norm": 3.6830499172210693,
|
| 63 |
+
"learning_rate": 9.716845878136201e-05,
|
| 64 |
+
"loss": 1.5209,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.32,
|
| 69 |
+
"grad_norm": 4.707543849945068,
|
| 70 |
+
"learning_rate": 9.681003584229391e-05,
|
| 71 |
+
"loss": 1.4562,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.36,
|
| 76 |
+
"grad_norm": 4.678582191467285,
|
| 77 |
+
"learning_rate": 9.645161290322581e-05,
|
| 78 |
+
"loss": 1.3076,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.36,
|
| 83 |
+
"eval_accuracy": 0.7259713701431493,
|
| 84 |
+
"eval_loss": 1.150874137878418,
|
| 85 |
+
"eval_runtime": 16.6221,
|
| 86 |
+
"eval_samples_per_second": 58.837,
|
| 87 |
+
"eval_steps_per_second": 7.4,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
+
"epoch": 0.39,
|
| 92 |
+
"grad_norm": 3.086634397506714,
|
| 93 |
+
"learning_rate": 9.609318996415772e-05,
|
| 94 |
+
"loss": 1.2684,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 0.43,
|
| 99 |
+
"grad_norm": 2.6688361167907715,
|
| 100 |
+
"learning_rate": 9.573476702508962e-05,
|
| 101 |
+
"loss": 1.1876,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 0.47,
|
| 106 |
+
"grad_norm": 3.8015213012695312,
|
| 107 |
+
"learning_rate": 9.53763440860215e-05,
|
| 108 |
+
"loss": 1.135,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
+
"epoch": 0.5,
|
| 113 |
+
"grad_norm": 5.519741535186768,
|
| 114 |
+
"learning_rate": 9.501792114695342e-05,
|
| 115 |
+
"loss": 1.1882,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 0.54,
|
| 120 |
+
"grad_norm": 2.872791290283203,
|
| 121 |
+
"learning_rate": 9.465949820788531e-05,
|
| 122 |
+
"loss": 1.0513,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 0.57,
|
| 127 |
+
"grad_norm": 3.185737371444702,
|
| 128 |
+
"learning_rate": 9.430107526881721e-05,
|
| 129 |
+
"loss": 1.1026,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"epoch": 0.61,
|
| 134 |
+
"grad_norm": 5.020754814147949,
|
| 135 |
+
"learning_rate": 9.39426523297491e-05,
|
| 136 |
+
"loss": 0.9906,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 0.65,
|
| 141 |
+
"grad_norm": 2.217747688293457,
|
| 142 |
+
"learning_rate": 9.358422939068101e-05,
|
| 143 |
+
"loss": 0.9737,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 0.68,
|
| 148 |
+
"grad_norm": 3.152509927749634,
|
| 149 |
+
"learning_rate": 9.32258064516129e-05,
|
| 150 |
+
"loss": 1.0249,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
+
"epoch": 0.72,
|
| 155 |
+
"grad_norm": 5.835267066955566,
|
| 156 |
+
"learning_rate": 9.28673835125448e-05,
|
| 157 |
+
"loss": 0.9605,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.72,
|
| 162 |
+
"eval_accuracy": 0.8057259713701431,
|
| 163 |
+
"eval_loss": 0.7632270455360413,
|
| 164 |
+
"eval_runtime": 13.1226,
|
| 165 |
+
"eval_samples_per_second": 74.528,
|
| 166 |
+
"eval_steps_per_second": 9.373,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 0.75,
|
| 171 |
+
"grad_norm": 3.458073616027832,
|
| 172 |
+
"learning_rate": 9.25089605734767e-05,
|
| 173 |
+
"loss": 0.8623,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
+
"epoch": 0.79,
|
| 178 |
+
"grad_norm": 3.9055118560791016,
|
| 179 |
+
"learning_rate": 9.215053763440861e-05,
|
| 180 |
+
"loss": 0.8075,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
+
"epoch": 0.82,
|
| 185 |
+
"grad_norm": 5.982291221618652,
|
| 186 |
+
"learning_rate": 9.17921146953405e-05,
|
| 187 |
+
"loss": 0.8726,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 0.86,
|
| 192 |
+
"grad_norm": 6.739437103271484,
|
| 193 |
+
"learning_rate": 9.143369175627241e-05,
|
| 194 |
+
"loss": 0.9887,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 0.9,
|
| 199 |
+
"grad_norm": 3.962754249572754,
|
| 200 |
+
"learning_rate": 9.10752688172043e-05,
|
| 201 |
+
"loss": 0.8384,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 0.93,
|
| 206 |
+
"grad_norm": 5.224504470825195,
|
| 207 |
+
"learning_rate": 9.07168458781362e-05,
|
| 208 |
+
"loss": 0.8309,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
+
"epoch": 0.97,
|
| 213 |
+
"grad_norm": 6.783730983734131,
|
| 214 |
+
"learning_rate": 9.03584229390681e-05,
|
| 215 |
+
"loss": 0.756,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 1.0,
|
| 220 |
+
"grad_norm": 1.871099829673767,
|
| 221 |
+
"learning_rate": 9e-05,
|
| 222 |
+
"loss": 0.7874,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 1.04,
|
| 227 |
+
"grad_norm": 3.4176278114318848,
|
| 228 |
+
"learning_rate": 8.964157706093191e-05,
|
| 229 |
+
"loss": 0.7627,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"epoch": 1.08,
|
| 234 |
+
"grad_norm": 2.1947131156921387,
|
| 235 |
+
"learning_rate": 8.92831541218638e-05,
|
| 236 |
+
"loss": 0.7508,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 1.08,
|
| 241 |
+
"eval_accuracy": 0.8006134969325154,
|
| 242 |
+
"eval_loss": 0.6995136141777039,
|
| 243 |
+
"eval_runtime": 13.1501,
|
| 244 |
+
"eval_samples_per_second": 74.372,
|
| 245 |
+
"eval_steps_per_second": 9.354,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 1.11,
|
| 250 |
+
"grad_norm": 4.5141119956970215,
|
| 251 |
+
"learning_rate": 8.892473118279571e-05,
|
| 252 |
+
"loss": 0.9078,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 1.15,
|
| 257 |
+
"grad_norm": 3.0271410942077637,
|
| 258 |
+
"learning_rate": 8.85663082437276e-05,
|
| 259 |
+
"loss": 0.8313,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"epoch": 1.18,
|
| 264 |
+
"grad_norm": 4.396864891052246,
|
| 265 |
+
"learning_rate": 8.82078853046595e-05,
|
| 266 |
+
"loss": 0.735,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
+
"epoch": 1.22,
|
| 271 |
+
"grad_norm": 6.498298168182373,
|
| 272 |
+
"learning_rate": 8.78494623655914e-05,
|
| 273 |
+
"loss": 0.6945,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
+
"epoch": 1.25,
|
| 278 |
+
"grad_norm": 3.187629461288452,
|
| 279 |
+
"learning_rate": 8.74910394265233e-05,
|
| 280 |
+
"loss": 0.6215,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 1.29,
|
| 285 |
+
"grad_norm": 5.515172004699707,
|
| 286 |
+
"learning_rate": 8.713261648745519e-05,
|
| 287 |
+
"loss": 0.7575,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
+
"epoch": 1.33,
|
| 292 |
+
"grad_norm": 4.944267749786377,
|
| 293 |
+
"learning_rate": 8.677419354838711e-05,
|
| 294 |
+
"loss": 0.7753,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
+
"epoch": 1.36,
|
| 299 |
+
"grad_norm": 4.4278388023376465,
|
| 300 |
+
"learning_rate": 8.6415770609319e-05,
|
| 301 |
+
"loss": 0.7316,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
+
"epoch": 1.4,
|
| 306 |
+
"grad_norm": 4.232585906982422,
|
| 307 |
+
"learning_rate": 8.60573476702509e-05,
|
| 308 |
+
"loss": 0.7011,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 1.43,
|
| 313 |
+
"grad_norm": 2.1777145862579346,
|
| 314 |
+
"learning_rate": 8.56989247311828e-05,
|
| 315 |
+
"loss": 0.7542,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
+
"epoch": 1.43,
|
| 320 |
+
"eval_accuracy": 0.83640081799591,
|
| 321 |
+
"eval_loss": 0.5710071921348572,
|
| 322 |
+
"eval_runtime": 13.0952,
|
| 323 |
+
"eval_samples_per_second": 74.684,
|
| 324 |
+
"eval_steps_per_second": 9.393,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
+
"epoch": 1.47,
|
| 329 |
+
"grad_norm": 3.8884618282318115,
|
| 330 |
+
"learning_rate": 8.53405017921147e-05,
|
| 331 |
+
"loss": 0.6824,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
+
"epoch": 1.51,
|
| 336 |
+
"grad_norm": 3.020177125930786,
|
| 337 |
+
"learning_rate": 8.49820788530466e-05,
|
| 338 |
+
"loss": 0.6009,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 1.54,
|
| 343 |
+
"grad_norm": 5.7138519287109375,
|
| 344 |
+
"learning_rate": 8.462365591397849e-05,
|
| 345 |
+
"loss": 0.5374,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
+
"epoch": 1.58,
|
| 350 |
+
"grad_norm": 6.61539888381958,
|
| 351 |
+
"learning_rate": 8.426523297491041e-05,
|
| 352 |
+
"loss": 0.7105,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"epoch": 1.61,
|
| 357 |
+
"grad_norm": 3.700871706008911,
|
| 358 |
+
"learning_rate": 8.39068100358423e-05,
|
| 359 |
+
"loss": 0.6331,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 1.65,
|
| 364 |
+
"grad_norm": 6.238260269165039,
|
| 365 |
+
"learning_rate": 8.35483870967742e-05,
|
| 366 |
+
"loss": 0.7435,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 1.68,
|
| 371 |
+
"grad_norm": 1.4030262231826782,
|
| 372 |
+
"learning_rate": 8.32258064516129e-05,
|
| 373 |
+
"loss": 0.6601,
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 1.72,
|
| 378 |
+
"grad_norm": 9.980137825012207,
|
| 379 |
+
"learning_rate": 8.28673835125448e-05,
|
| 380 |
+
"loss": 0.649,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 1.76,
|
| 385 |
+
"grad_norm": 11.748491287231445,
|
| 386 |
+
"learning_rate": 8.250896057347671e-05,
|
| 387 |
+
"loss": 0.7075,
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
+
"epoch": 1.79,
|
| 392 |
+
"grad_norm": 7.283904552459717,
|
| 393 |
+
"learning_rate": 8.215053763440861e-05,
|
| 394 |
+
"loss": 0.6945,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
+
"epoch": 1.79,
|
| 399 |
+
"eval_accuracy": 0.8588957055214724,
|
| 400 |
+
"eval_loss": 0.5103773474693298,
|
| 401 |
+
"eval_runtime": 13.2386,
|
| 402 |
+
"eval_samples_per_second": 73.875,
|
| 403 |
+
"eval_steps_per_second": 9.291,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 1.83,
|
| 408 |
+
"grad_norm": 3.227905035018921,
|
| 409 |
+
"learning_rate": 8.179211469534051e-05,
|
| 410 |
+
"loss": 0.5835,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 1.86,
|
| 415 |
+
"grad_norm": 4.389718532562256,
|
| 416 |
+
"learning_rate": 8.14336917562724e-05,
|
| 417 |
+
"loss": 0.6946,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
+
"epoch": 1.9,
|
| 422 |
+
"grad_norm": 2.76204514503479,
|
| 423 |
+
"learning_rate": 8.107526881720431e-05,
|
| 424 |
+
"loss": 0.6389,
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 1.94,
|
| 429 |
+
"grad_norm": 3.973482370376587,
|
| 430 |
+
"learning_rate": 8.07168458781362e-05,
|
| 431 |
+
"loss": 0.5478,
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
+
"epoch": 1.97,
|
| 436 |
+
"grad_norm": 7.488864898681641,
|
| 437 |
+
"learning_rate": 8.03584229390681e-05,
|
| 438 |
+
"loss": 0.5571,
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
+
"epoch": 2.01,
|
| 443 |
+
"grad_norm": 4.2505059242248535,
|
| 444 |
+
"learning_rate": 8e-05,
|
| 445 |
+
"loss": 0.7184,
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 2.04,
|
| 450 |
+
"grad_norm": 1.9729084968566895,
|
| 451 |
+
"learning_rate": 7.964157706093191e-05,
|
| 452 |
+
"loss": 0.5643,
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"epoch": 2.08,
|
| 457 |
+
"grad_norm": 2.607386589050293,
|
| 458 |
+
"learning_rate": 7.92831541218638e-05,
|
| 459 |
+
"loss": 0.4484,
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
+
"epoch": 2.11,
|
| 464 |
+
"grad_norm": 4.689638614654541,
|
| 465 |
+
"learning_rate": 7.892473118279571e-05,
|
| 466 |
+
"loss": 0.5449,
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 2.15,
|
| 471 |
+
"grad_norm": 6.3727707862854,
|
| 472 |
+
"learning_rate": 7.85663082437276e-05,
|
| 473 |
+
"loss": 0.5218,
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 2.15,
|
| 478 |
+
"eval_accuracy": 0.8343558282208589,
|
| 479 |
+
"eval_loss": 0.5389257073402405,
|
| 480 |
+
"eval_runtime": 13.1796,
|
| 481 |
+
"eval_samples_per_second": 74.205,
|
| 482 |
+
"eval_steps_per_second": 9.333,
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 2.19,
|
| 487 |
+
"grad_norm": 6.6133832931518555,
|
| 488 |
+
"learning_rate": 7.82078853046595e-05,
|
| 489 |
+
"loss": 0.5242,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"epoch": 2.22,
|
| 494 |
+
"grad_norm": 6.967311382293701,
|
| 495 |
+
"learning_rate": 7.784946236559139e-05,
|
| 496 |
+
"loss": 0.5191,
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 2.26,
|
| 501 |
+
"grad_norm": 6.818421363830566,
|
| 502 |
+
"learning_rate": 7.74910394265233e-05,
|
| 503 |
+
"loss": 0.4675,
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 2.29,
|
| 508 |
+
"grad_norm": 3.4351966381073,
|
| 509 |
+
"learning_rate": 7.71326164874552e-05,
|
| 510 |
+
"loss": 0.538,
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
+
"epoch": 2.33,
|
| 515 |
+
"grad_norm": 5.848361968994141,
|
| 516 |
+
"learning_rate": 7.67741935483871e-05,
|
| 517 |
+
"loss": 0.513,
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
+
"epoch": 2.37,
|
| 522 |
+
"grad_norm": 3.990499496459961,
|
| 523 |
+
"learning_rate": 7.6415770609319e-05,
|
| 524 |
+
"loss": 0.5287,
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
+
"epoch": 2.4,
|
| 529 |
+
"grad_norm": 3.2983336448669434,
|
| 530 |
+
"learning_rate": 7.60573476702509e-05,
|
| 531 |
+
"loss": 0.5794,
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
+
"epoch": 2.44,
|
| 536 |
+
"grad_norm": 2.981754779815674,
|
| 537 |
+
"learning_rate": 7.56989247311828e-05,
|
| 538 |
+
"loss": 0.3556,
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 2.47,
|
| 543 |
+
"grad_norm": 7.759634494781494,
|
| 544 |
+
"learning_rate": 7.53405017921147e-05,
|
| 545 |
+
"loss": 0.5055,
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
+
"epoch": 2.51,
|
| 550 |
+
"grad_norm": 7.100767135620117,
|
| 551 |
+
"learning_rate": 7.49820788530466e-05,
|
| 552 |
+
"loss": 0.5365,
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
+
"epoch": 2.51,
|
| 557 |
+
"eval_accuracy": 0.8466257668711656,
|
| 558 |
+
"eval_loss": 0.5034566521644592,
|
| 559 |
+
"eval_runtime": 13.0487,
|
| 560 |
+
"eval_samples_per_second": 74.95,
|
| 561 |
+
"eval_steps_per_second": 9.426,
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 2.54,
|
| 566 |
+
"grad_norm": 3.7873051166534424,
|
| 567 |
+
"learning_rate": 7.462365591397849e-05,
|
| 568 |
+
"loss": 0.7049,
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 2.58,
|
| 573 |
+
"grad_norm": 4.464657783508301,
|
| 574 |
+
"learning_rate": 7.42652329749104e-05,
|
| 575 |
+
"loss": 0.6012,
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 2.62,
|
| 580 |
+
"grad_norm": 6.518985271453857,
|
| 581 |
+
"learning_rate": 7.39426523297491e-05,
|
| 582 |
+
"loss": 0.6301,
|
| 583 |
"step": 730
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 2.65,
|
| 587 |
+
"grad_norm": 4.162339210510254,
|
| 588 |
+
"learning_rate": 7.3584229390681e-05,
|
| 589 |
+
"loss": 0.4366,
|
| 590 |
"step": 740
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 2.69,
|
| 594 |
+
"grad_norm": 5.646885395050049,
|
| 595 |
+
"learning_rate": 7.32258064516129e-05,
|
| 596 |
+
"loss": 0.6134,
|
| 597 |
"step": 750
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 2.72,
|
| 601 |
+
"grad_norm": 8.28348445892334,
|
| 602 |
+
"learning_rate": 7.28673835125448e-05,
|
| 603 |
+
"loss": 0.4007,
|
| 604 |
"step": 760
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 2.76,
|
| 608 |
+
"grad_norm": 2.606187105178833,
|
| 609 |
+
"learning_rate": 7.250896057347671e-05,
|
| 610 |
+
"loss": 0.5909,
|
| 611 |
"step": 770
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 2.8,
|
| 615 |
+
"grad_norm": 8.181116104125977,
|
| 616 |
+
"learning_rate": 7.215053763440861e-05,
|
| 617 |
+
"loss": 0.5286,
|
| 618 |
"step": 780
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 2.83,
|
| 622 |
+
"grad_norm": 5.969821929931641,
|
| 623 |
+
"learning_rate": 7.179211469534051e-05,
|
| 624 |
+
"loss": 0.4924,
|
| 625 |
"step": 790
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 2.87,
|
| 629 |
+
"grad_norm": 5.566159248352051,
|
| 630 |
+
"learning_rate": 7.14336917562724e-05,
|
| 631 |
+
"loss": 0.5564,
|
| 632 |
"step": 800
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 2.87,
|
| 636 |
+
"eval_accuracy": 0.8476482617586912,
|
| 637 |
+
"eval_loss": 0.48173415660858154,
|
| 638 |
+
"eval_runtime": 13.1664,
|
| 639 |
+
"eval_samples_per_second": 74.28,
|
| 640 |
+
"eval_steps_per_second": 9.342,
|
| 641 |
"step": 800
|
| 642 |
},
|
| 643 |
{
|
| 644 |
+
"epoch": 2.9,
|
| 645 |
+
"grad_norm": 6.260211944580078,
|
| 646 |
+
"learning_rate": 7.10752688172043e-05,
|
| 647 |
+
"loss": 0.3901,
|
| 648 |
"step": 810
|
| 649 |
},
|
| 650 |
{
|
| 651 |
+
"epoch": 2.94,
|
| 652 |
+
"grad_norm": 5.212798118591309,
|
| 653 |
+
"learning_rate": 7.07168458781362e-05,
|
| 654 |
+
"loss": 0.3447,
|
| 655 |
"step": 820
|
| 656 |
},
|
| 657 |
{
|
| 658 |
+
"epoch": 2.97,
|
| 659 |
+
"grad_norm": 3.084397554397583,
|
| 660 |
+
"learning_rate": 7.03584229390681e-05,
|
| 661 |
+
"loss": 0.5139,
|
| 662 |
"step": 830
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 3.01,
|
| 666 |
+
"grad_norm": 4.144620418548584,
|
| 667 |
+
"learning_rate": 7e-05,
|
| 668 |
+
"loss": 0.5502,
|
| 669 |
"step": 840
|
| 670 |
},
|
| 671 |
{
|
| 672 |
+
"epoch": 3.05,
|
| 673 |
+
"grad_norm": 4.02711296081543,
|
| 674 |
+
"learning_rate": 6.964157706093191e-05,
|
| 675 |
+
"loss": 0.3868,
|
| 676 |
"step": 850
|
| 677 |
},
|
| 678 |
{
|
| 679 |
+
"epoch": 3.08,
|
| 680 |
+
"grad_norm": 4.601646900177002,
|
| 681 |
+
"learning_rate": 6.92831541218638e-05,
|
| 682 |
+
"loss": 0.5119,
|
| 683 |
"step": 860
|
| 684 |
},
|
| 685 |
{
|
| 686 |
+
"epoch": 3.12,
|
| 687 |
+
"grad_norm": 7.631004810333252,
|
| 688 |
+
"learning_rate": 6.89247311827957e-05,
|
| 689 |
+
"loss": 0.4146,
|
| 690 |
"step": 870
|
| 691 |
},
|
| 692 |
{
|
| 693 |
+
"epoch": 3.15,
|
| 694 |
+
"grad_norm": 3.341120719909668,
|
| 695 |
+
"learning_rate": 6.85663082437276e-05,
|
| 696 |
+
"loss": 0.4403,
|
| 697 |
"step": 880
|
| 698 |
},
|
| 699 |
{
|
| 700 |
+
"epoch": 3.19,
|
| 701 |
+
"grad_norm": 3.3383867740631104,
|
| 702 |
+
"learning_rate": 6.82078853046595e-05,
|
| 703 |
+
"loss": 0.4715,
|
| 704 |
"step": 890
|
| 705 |
},
|
| 706 |
{
|
| 707 |
+
"epoch": 3.23,
|
| 708 |
+
"grad_norm": 5.930158615112305,
|
| 709 |
+
"learning_rate": 6.78494623655914e-05,
|
| 710 |
+
"loss": 0.4597,
|
| 711 |
"step": 900
|
| 712 |
},
|
| 713 |
{
|
| 714 |
+
"epoch": 3.23,
|
| 715 |
+
"eval_accuracy": 0.8599182004089979,
|
| 716 |
+
"eval_loss": 0.46898409724235535,
|
| 717 |
+
"eval_runtime": 13.1156,
|
| 718 |
+
"eval_samples_per_second": 74.568,
|
| 719 |
+
"eval_steps_per_second": 9.378,
|
| 720 |
"step": 900
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"epoch": 3.26,
|
| 724 |
+
"grad_norm": 8.335565567016602,
|
| 725 |
+
"learning_rate": 6.74910394265233e-05,
|
| 726 |
+
"loss": 0.3672,
|
| 727 |
"step": 910
|
| 728 |
},
|
| 729 |
{
|
| 730 |
+
"epoch": 3.3,
|
| 731 |
+
"grad_norm": 4.635217189788818,
|
| 732 |
+
"learning_rate": 6.713261648745521e-05,
|
| 733 |
+
"loss": 0.4008,
|
| 734 |
"step": 920
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 3.33,
|
| 738 |
+
"grad_norm": 6.6547322273254395,
|
| 739 |
+
"learning_rate": 6.67741935483871e-05,
|
| 740 |
+
"loss": 0.4973,
|
| 741 |
"step": 930
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 3.37,
|
| 745 |
+
"grad_norm": 2.2684273719787598,
|
| 746 |
+
"learning_rate": 6.6415770609319e-05,
|
| 747 |
+
"loss": 0.3125,
|
| 748 |
"step": 940
|
| 749 |
},
|
| 750 |
{
|
| 751 |
+
"epoch": 3.41,
|
| 752 |
+
"grad_norm": 6.180244445800781,
|
| 753 |
+
"learning_rate": 6.60573476702509e-05,
|
| 754 |
+
"loss": 0.4356,
|
| 755 |
"step": 950
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 3.44,
|
| 759 |
+
"grad_norm": 3.5774285793304443,
|
| 760 |
+
"learning_rate": 6.56989247311828e-05,
|
| 761 |
+
"loss": 0.4316,
|
| 762 |
"step": 960
|
| 763 |
},
|
| 764 |
{
|
| 765 |
+
"epoch": 3.48,
|
| 766 |
+
"grad_norm": 4.347971439361572,
|
| 767 |
+
"learning_rate": 6.534050179211469e-05,
|
| 768 |
+
"loss": 0.4942,
|
| 769 |
"step": 970
|
| 770 |
},
|
| 771 |
{
|
| 772 |
+
"epoch": 3.51,
|
| 773 |
+
"grad_norm": 1.7022749185562134,
|
| 774 |
+
"learning_rate": 6.49820788530466e-05,
|
| 775 |
+
"loss": 0.5551,
|
| 776 |
"step": 980
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"epoch": 3.55,
|
| 780 |
+
"grad_norm": 5.636997699737549,
|
| 781 |
+
"learning_rate": 6.46236559139785e-05,
|
| 782 |
+
"loss": 0.4212,
|
| 783 |
"step": 990
|
| 784 |
},
|
| 785 |
{
|
| 786 |
+
"epoch": 3.58,
|
| 787 |
+
"grad_norm": 1.4300520420074463,
|
| 788 |
+
"learning_rate": 6.42652329749104e-05,
|
| 789 |
+
"loss": 0.3516,
|
| 790 |
"step": 1000
|
| 791 |
},
|
| 792 |
{
|
| 793 |
+
"epoch": 3.58,
|
| 794 |
+
"eval_accuracy": 0.8507157464212679,
|
| 795 |
+
"eval_loss": 0.4920569360256195,
|
| 796 |
+
"eval_runtime": 13.0448,
|
| 797 |
+
"eval_samples_per_second": 74.972,
|
| 798 |
+
"eval_steps_per_second": 9.429,
|
| 799 |
"step": 1000
|
| 800 |
},
|
| 801 |
{
|
| 802 |
+
"epoch": 3.62,
|
| 803 |
+
"grad_norm": 3.524423837661743,
|
| 804 |
+
"learning_rate": 6.39068100358423e-05,
|
| 805 |
+
"loss": 0.4023,
|
| 806 |
"step": 1010
|
| 807 |
},
|
| 808 |
{
|
| 809 |
+
"epoch": 3.66,
|
| 810 |
+
"grad_norm": 7.19805383682251,
|
| 811 |
+
"learning_rate": 6.35483870967742e-05,
|
| 812 |
+
"loss": 0.3299,
|
| 813 |
"step": 1020
|
| 814 |
},
|
| 815 |
{
|
| 816 |
+
"epoch": 3.69,
|
| 817 |
+
"grad_norm": 5.69304084777832,
|
| 818 |
+
"learning_rate": 6.318996415770609e-05,
|
| 819 |
+
"loss": 0.4332,
|
| 820 |
"step": 1030
|
| 821 |
},
|
| 822 |
{
|
| 823 |
+
"epoch": 3.73,
|
| 824 |
+
"grad_norm": 5.353736400604248,
|
| 825 |
+
"learning_rate": 6.283154121863799e-05,
|
| 826 |
+
"loss": 0.4798,
|
| 827 |
"step": 1040
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 3.76,
|
| 831 |
+
"grad_norm": 5.782208442687988,
|
| 832 |
+
"learning_rate": 6.247311827956989e-05,
|
| 833 |
+
"loss": 0.4301,
|
| 834 |
"step": 1050
|
| 835 |
},
|
| 836 |
{
|
| 837 |
+
"epoch": 3.8,
|
| 838 |
+
"grad_norm": 5.975741386413574,
|
| 839 |
+
"learning_rate": 6.211469534050179e-05,
|
| 840 |
+
"loss": 0.3599,
|
| 841 |
"step": 1060
|
| 842 |
},
|
| 843 |
{
|
| 844 |
+
"epoch": 3.84,
|
| 845 |
+
"grad_norm": 6.242609977722168,
|
| 846 |
+
"learning_rate": 6.17562724014337e-05,
|
| 847 |
+
"loss": 0.4275,
|
| 848 |
"step": 1070
|
| 849 |
},
|
| 850 |
{
|
| 851 |
+
"epoch": 3.87,
|
| 852 |
+
"grad_norm": 5.965251922607422,
|
| 853 |
+
"learning_rate": 6.13978494623656e-05,
|
| 854 |
+
"loss": 0.5298,
|
| 855 |
"step": 1080
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"epoch": 3.91,
|
| 859 |
+
"grad_norm": 9.964472770690918,
|
| 860 |
+
"learning_rate": 6.10394265232975e-05,
|
| 861 |
+
"loss": 0.479,
|
| 862 |
"step": 1090
|
| 863 |
},
|
| 864 |
{
|
| 865 |
+
"epoch": 3.94,
|
| 866 |
+
"grad_norm": 2.9509897232055664,
|
| 867 |
+
"learning_rate": 6.068100358422939e-05,
|
| 868 |
+
"loss": 0.3741,
|
| 869 |
"step": 1100
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 3.94,
|
| 873 |
+
"eval_accuracy": 0.8568507157464212,
|
| 874 |
+
"eval_loss": 0.4820023775100708,
|
| 875 |
+
"eval_runtime": 12.9964,
|
| 876 |
+
"eval_samples_per_second": 75.252,
|
| 877 |
+
"eval_steps_per_second": 9.464,
|
| 878 |
"step": 1100
|
| 879 |
},
|
| 880 |
{
|
| 881 |
+
"epoch": 3.98,
|
| 882 |
+
"grad_norm": 4.46811056137085,
|
| 883 |
+
"learning_rate": 6.0322580645161295e-05,
|
| 884 |
+
"loss": 0.2701,
|
| 885 |
"step": 1110
|
| 886 |
},
|
| 887 |
{
|
| 888 |
+
"epoch": 4.01,
|
| 889 |
+
"grad_norm": 3.02339506149292,
|
| 890 |
+
"learning_rate": 5.996415770609319e-05,
|
| 891 |
+
"loss": 0.3165,
|
| 892 |
"step": 1120
|
| 893 |
},
|
| 894 |
{
|
| 895 |
+
"epoch": 4.05,
|
| 896 |
+
"grad_norm": 4.451502323150635,
|
| 897 |
+
"learning_rate": 5.960573476702509e-05,
|
| 898 |
+
"loss": 0.4528,
|
| 899 |
"step": 1130
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 4.09,
|
| 903 |
+
"grad_norm": 3.656177520751953,
|
| 904 |
+
"learning_rate": 5.9247311827956994e-05,
|
| 905 |
+
"loss": 0.3612,
|
| 906 |
"step": 1140
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"epoch": 4.12,
|
| 910 |
+
"grad_norm": 5.621472358703613,
|
| 911 |
+
"learning_rate": 5.8888888888888896e-05,
|
| 912 |
+
"loss": 0.3641,
|
| 913 |
"step": 1150
|
| 914 |
},
|
| 915 |
{
|
| 916 |
+
"epoch": 4.16,
|
| 917 |
+
"grad_norm": 10.402291297912598,
|
| 918 |
+
"learning_rate": 5.853046594982079e-05,
|
| 919 |
+
"loss": 0.4598,
|
| 920 |
"step": 1160
|
| 921 |
},
|
| 922 |
{
|
| 923 |
+
"epoch": 4.19,
|
| 924 |
+
"grad_norm": 5.005283355712891,
|
| 925 |
+
"learning_rate": 5.8172043010752686e-05,
|
| 926 |
+
"loss": 0.3746,
|
| 927 |
"step": 1170
|
| 928 |
},
|
| 929 |
{
|
| 930 |
+
"epoch": 4.23,
|
| 931 |
+
"grad_norm": 6.668752670288086,
|
| 932 |
+
"learning_rate": 5.7813620071684594e-05,
|
| 933 |
+
"loss": 0.416,
|
| 934 |
"step": 1180
|
| 935 |
},
|
| 936 |
{
|
| 937 |
+
"epoch": 4.27,
|
| 938 |
+
"grad_norm": 6.447160720825195,
|
| 939 |
+
"learning_rate": 5.745519713261649e-05,
|
| 940 |
+
"loss": 0.3504,
|
| 941 |
"step": 1190
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"epoch": 4.3,
|
| 945 |
+
"grad_norm": 8.023221969604492,
|
| 946 |
+
"learning_rate": 5.7096774193548384e-05,
|
| 947 |
+
"loss": 0.4169,
|
| 948 |
"step": 1200
|
| 949 |
},
|
| 950 |
{
|
| 951 |
+
"epoch": 4.3,
|
| 952 |
+
"eval_accuracy": 0.852760736196319,
|
| 953 |
+
"eval_loss": 0.47300344705581665,
|
| 954 |
+
"eval_runtime": 13.1292,
|
| 955 |
+
"eval_samples_per_second": 74.491,
|
| 956 |
+
"eval_steps_per_second": 9.368,
|
| 957 |
"step": 1200
|
| 958 |
},
|
| 959 |
{
|
| 960 |
+
"epoch": 4.34,
|
| 961 |
+
"grad_norm": 2.819065570831299,
|
| 962 |
+
"learning_rate": 5.673835125448029e-05,
|
| 963 |
+
"loss": 0.3808,
|
| 964 |
"step": 1210
|
| 965 |
},
|
| 966 |
{
|
| 967 |
+
"epoch": 4.37,
|
| 968 |
+
"grad_norm": 10.216522216796875,
|
| 969 |
+
"learning_rate": 5.637992831541219e-05,
|
| 970 |
+
"loss": 0.3039,
|
| 971 |
"step": 1220
|
| 972 |
},
|
| 973 |
{
|
| 974 |
+
"epoch": 4.41,
|
| 975 |
+
"grad_norm": 4.38183069229126,
|
| 976 |
+
"learning_rate": 5.602150537634408e-05,
|
| 977 |
+
"loss": 0.3508,
|
| 978 |
"step": 1230
|
| 979 |
},
|
| 980 |
{
|
| 981 |
+
"epoch": 4.44,
|
| 982 |
+
"grad_norm": 3.4877617359161377,
|
| 983 |
+
"learning_rate": 5.566308243727599e-05,
|
| 984 |
+
"loss": 0.2875,
|
| 985 |
"step": 1240
|
| 986 |
},
|
| 987 |
{
|
| 988 |
+
"epoch": 4.48,
|
| 989 |
+
"grad_norm": 5.170544624328613,
|
| 990 |
+
"learning_rate": 5.530465949820789e-05,
|
| 991 |
+
"loss": 0.2425,
|
| 992 |
"step": 1250
|
| 993 |
},
|
| 994 |
{
|
| 995 |
+
"epoch": 4.52,
|
| 996 |
+
"grad_norm": 1.037807583808899,
|
| 997 |
+
"learning_rate": 5.494623655913979e-05,
|
| 998 |
+
"loss": 0.2618,
|
| 999 |
"step": 1260
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
+
"epoch": 4.55,
|
| 1003 |
+
"grad_norm": 1.1233432292938232,
|
| 1004 |
+
"learning_rate": 5.458781362007169e-05,
|
| 1005 |
+
"loss": 0.4105,
|
| 1006 |
"step": 1270
|
| 1007 |
},
|
| 1008 |
{
|
| 1009 |
+
"epoch": 4.59,
|
| 1010 |
+
"grad_norm": 6.105014801025391,
|
| 1011 |
+
"learning_rate": 5.422939068100359e-05,
|
| 1012 |
+
"loss": 0.3639,
|
| 1013 |
"step": 1280
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
+
"epoch": 4.62,
|
| 1017 |
+
"grad_norm": 5.100095272064209,
|
| 1018 |
+
"learning_rate": 5.387096774193549e-05,
|
| 1019 |
+
"loss": 0.4289,
|
| 1020 |
"step": 1290
|
| 1021 |
},
|
| 1022 |
{
|
| 1023 |
+
"epoch": 4.66,
|
| 1024 |
+
"grad_norm": 4.874768257141113,
|
| 1025 |
+
"learning_rate": 5.351254480286738e-05,
|
| 1026 |
+
"loss": 0.5131,
|
| 1027 |
"step": 1300
|
| 1028 |
},
|
| 1029 |
{
|
| 1030 |
+
"epoch": 4.66,
|
| 1031 |
+
"eval_accuracy": 0.8680981595092024,
|
| 1032 |
+
"eval_loss": 0.4481976628303528,
|
| 1033 |
+
"eval_runtime": 12.8876,
|
| 1034 |
+
"eval_samples_per_second": 75.887,
|
| 1035 |
+
"eval_steps_per_second": 9.544,
|
| 1036 |
"step": 1300
|
| 1037 |
},
|
| 1038 |
{
|
| 1039 |
+
"epoch": 4.7,
|
| 1040 |
+
"grad_norm": 3.7920286655426025,
|
| 1041 |
+
"learning_rate": 5.315412186379929e-05,
|
| 1042 |
+
"loss": 0.3811,
|
| 1043 |
"step": 1310
|
| 1044 |
},
|
| 1045 |
{
|
| 1046 |
+
"epoch": 4.73,
|
| 1047 |
+
"grad_norm": 4.628345489501953,
|
| 1048 |
+
"learning_rate": 5.2795698924731186e-05,
|
| 1049 |
+
"loss": 0.3271,
|
| 1050 |
"step": 1320
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
+
"epoch": 4.77,
|
| 1054 |
+
"grad_norm": 7.546787261962891,
|
| 1055 |
+
"learning_rate": 5.243727598566308e-05,
|
| 1056 |
+
"loss": 0.4059,
|
| 1057 |
"step": 1330
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
+
"epoch": 4.8,
|
| 1061 |
+
"grad_norm": 3.106943130493164,
|
| 1062 |
+
"learning_rate": 5.207885304659499e-05,
|
| 1063 |
+
"loss": 0.2255,
|
| 1064 |
"step": 1340
|
| 1065 |
},
|
| 1066 |
{
|
| 1067 |
+
"epoch": 4.84,
|
| 1068 |
+
"grad_norm": 7.233306407928467,
|
| 1069 |
+
"learning_rate": 5.1720430107526885e-05,
|
| 1070 |
+
"loss": 0.299,
|
| 1071 |
"step": 1350
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
+
"epoch": 4.87,
|
| 1075 |
+
"grad_norm": 8.275094032287598,
|
| 1076 |
+
"learning_rate": 5.136200716845878e-05,
|
| 1077 |
+
"loss": 0.387,
|
| 1078 |
"step": 1360
|
| 1079 |
},
|
| 1080 |
{
|
| 1081 |
+
"epoch": 4.91,
|
| 1082 |
+
"grad_norm": 4.1648640632629395,
|
| 1083 |
+
"learning_rate": 5.100358422939069e-05,
|
| 1084 |
+
"loss": 0.2891,
|
| 1085 |
"step": 1370
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
+
"epoch": 4.95,
|
| 1089 |
+
"grad_norm": 2.7225232124328613,
|
| 1090 |
+
"learning_rate": 5.064516129032258e-05,
|
| 1091 |
+
"loss": 0.3828,
|
| 1092 |
"step": 1380
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
+
"epoch": 4.98,
|
| 1096 |
+
"grad_norm": 4.712428092956543,
|
| 1097 |
+
"learning_rate": 5.028673835125448e-05,
|
| 1098 |
+
"loss": 0.3843,
|
| 1099 |
"step": 1390
|
| 1100 |
},
|
| 1101 |
{
|
| 1102 |
+
"epoch": 5.02,
|
| 1103 |
+
"grad_norm": 7.5443644523620605,
|
| 1104 |
+
"learning_rate": 4.992831541218638e-05,
|
| 1105 |
+
"loss": 0.3673,
|
| 1106 |
"step": 1400
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
+
"epoch": 5.02,
|
| 1110 |
+
"eval_accuracy": 0.8752556237218814,
|
| 1111 |
+
"eval_loss": 0.42281925678253174,
|
| 1112 |
+
"eval_runtime": 14.3907,
|
| 1113 |
+
"eval_samples_per_second": 67.961,
|
| 1114 |
+
"eval_steps_per_second": 8.547,
|
| 1115 |
"step": 1400
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
+
"epoch": 5.05,
|
| 1119 |
+
"grad_norm": 2.7451975345611572,
|
| 1120 |
+
"learning_rate": 4.956989247311828e-05,
|
| 1121 |
+
"loss": 0.3115,
|
| 1122 |
"step": 1410
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
+
"epoch": 5.09,
|
| 1126 |
+
"grad_norm": 0.5151819586753845,
|
| 1127 |
+
"learning_rate": 4.9211469534050184e-05,
|
| 1128 |
+
"loss": 0.3281,
|
| 1129 |
"step": 1420
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
+
"epoch": 5.13,
|
| 1133 |
+
"grad_norm": 7.1587419509887695,
|
| 1134 |
+
"learning_rate": 4.8853046594982085e-05,
|
| 1135 |
+
"loss": 0.2715,
|
| 1136 |
"step": 1430
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
+
"epoch": 5.16,
|
| 1140 |
+
"grad_norm": 6.464181423187256,
|
| 1141 |
+
"learning_rate": 4.849462365591398e-05,
|
| 1142 |
+
"loss": 0.2884,
|
| 1143 |
"step": 1440
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
+
"epoch": 5.2,
|
| 1147 |
+
"grad_norm": 6.002290725708008,
|
| 1148 |
+
"learning_rate": 4.813620071684588e-05,
|
| 1149 |
+
"loss": 0.3214,
|
| 1150 |
"step": 1450
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
+
"epoch": 5.23,
|
| 1154 |
+
"grad_norm": 7.911471366882324,
|
| 1155 |
+
"learning_rate": 4.7777777777777784e-05,
|
| 1156 |
+
"loss": 0.3749,
|
| 1157 |
"step": 1460
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
+
"epoch": 5.27,
|
| 1161 |
+
"grad_norm": 6.183435440063477,
|
| 1162 |
+
"learning_rate": 4.741935483870968e-05,
|
| 1163 |
+
"loss": 0.4746,
|
| 1164 |
"step": 1470
|
| 1165 |
},
|
| 1166 |
{
|
| 1167 |
+
"epoch": 5.3,
|
| 1168 |
+
"grad_norm": 2.4367499351501465,
|
| 1169 |
+
"learning_rate": 4.706093189964158e-05,
|
| 1170 |
+
"loss": 0.2526,
|
| 1171 |
"step": 1480
|
| 1172 |
},
|
| 1173 |
{
|
| 1174 |
+
"epoch": 5.34,
|
| 1175 |
+
"grad_norm": 3.299232006072998,
|
| 1176 |
+
"learning_rate": 4.6702508960573476e-05,
|
| 1177 |
+
"loss": 0.3413,
|
| 1178 |
"step": 1490
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
+
"epoch": 5.38,
|
| 1182 |
+
"grad_norm": 5.005585193634033,
|
| 1183 |
+
"learning_rate": 4.634408602150538e-05,
|
| 1184 |
+
"loss": 0.3722,
|
| 1185 |
"step": 1500
|
| 1186 |
},
|
| 1187 |
{
|
| 1188 |
+
"epoch": 5.38,
|
| 1189 |
+
"eval_accuracy": 0.8599182004089979,
|
| 1190 |
+
"eval_loss": 0.4820214807987213,
|
| 1191 |
+
"eval_runtime": 13.0602,
|
| 1192 |
+
"eval_samples_per_second": 74.884,
|
| 1193 |
+
"eval_steps_per_second": 9.418,
|
| 1194 |
"step": 1500
|
| 1195 |
},
|
| 1196 |
{
|
| 1197 |
+
"epoch": 5.41,
|
| 1198 |
+
"grad_norm": 4.578291416168213,
|
| 1199 |
+
"learning_rate": 4.598566308243728e-05,
|
| 1200 |
+
"loss": 0.2471,
|
| 1201 |
"step": 1510
|
| 1202 |
},
|
| 1203 |
{
|
| 1204 |
+
"epoch": 5.45,
|
| 1205 |
+
"grad_norm": 4.164848327636719,
|
| 1206 |
+
"learning_rate": 4.5627240143369175e-05,
|
| 1207 |
+
"loss": 0.2803,
|
| 1208 |
"step": 1520
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
+
"epoch": 5.48,
|
| 1212 |
+
"grad_norm": 3.4090826511383057,
|
| 1213 |
+
"learning_rate": 4.5268817204301076e-05,
|
| 1214 |
+
"loss": 0.3843,
|
| 1215 |
"step": 1530
|
| 1216 |
},
|
| 1217 |
{
|
| 1218 |
+
"epoch": 5.52,
|
| 1219 |
+
"grad_norm": 3.8364851474761963,
|
| 1220 |
+
"learning_rate": 4.491039426523297e-05,
|
| 1221 |
+
"loss": 0.348,
|
| 1222 |
"step": 1540
|
| 1223 |
},
|
| 1224 |
{
|
| 1225 |
+
"epoch": 5.56,
|
| 1226 |
+
"grad_norm": 0.9995656609535217,
|
| 1227 |
+
"learning_rate": 4.455197132616487e-05,
|
| 1228 |
+
"loss": 0.2812,
|
| 1229 |
"step": 1550
|
| 1230 |
},
|
| 1231 |
{
|
| 1232 |
+
"epoch": 5.59,
|
| 1233 |
+
"grad_norm": 0.11745692044496536,
|
| 1234 |
+
"learning_rate": 4.4193548387096775e-05,
|
| 1235 |
+
"loss": 0.2775,
|
| 1236 |
"step": 1560
|
| 1237 |
},
|
| 1238 |
{
|
| 1239 |
+
"epoch": 5.63,
|
| 1240 |
+
"grad_norm": 8.940442085266113,
|
| 1241 |
+
"learning_rate": 4.383512544802868e-05,
|
| 1242 |
+
"loss": 0.3364,
|
| 1243 |
"step": 1570
|
| 1244 |
},
|
| 1245 |
{
|
| 1246 |
+
"epoch": 5.66,
|
| 1247 |
+
"grad_norm": 3.694096088409424,
|
| 1248 |
+
"learning_rate": 4.347670250896058e-05,
|
| 1249 |
+
"loss": 0.3487,
|
| 1250 |
"step": 1580
|
| 1251 |
},
|
| 1252 |
{
|
| 1253 |
+
"epoch": 5.7,
|
| 1254 |
+
"grad_norm": 5.201242446899414,
|
| 1255 |
+
"learning_rate": 4.3118279569892474e-05,
|
| 1256 |
+
"loss": 0.4002,
|
| 1257 |
"step": 1590
|
| 1258 |
},
|
| 1259 |
{
|
| 1260 |
+
"epoch": 5.73,
|
| 1261 |
+
"grad_norm": 1.3319661617279053,
|
| 1262 |
+
"learning_rate": 4.2759856630824376e-05,
|
| 1263 |
+
"loss": 0.3005,
|
| 1264 |
"step": 1600
|
| 1265 |
},
|
| 1266 |
{
|
| 1267 |
+
"epoch": 5.73,
|
| 1268 |
+
"eval_accuracy": 0.8732106339468303,
|
| 1269 |
+
"eval_loss": 0.4456250071525574,
|
| 1270 |
+
"eval_runtime": 13.0438,
|
| 1271 |
+
"eval_samples_per_second": 74.978,
|
| 1272 |
+
"eval_steps_per_second": 9.43,
|
| 1273 |
"step": 1600
|
| 1274 |
},
|
| 1275 |
{
|
| 1276 |
+
"epoch": 5.77,
|
| 1277 |
+
"grad_norm": 1.4663386344909668,
|
| 1278 |
+
"learning_rate": 4.240143369175628e-05,
|
| 1279 |
+
"loss": 0.3404,
|
| 1280 |
"step": 1610
|
| 1281 |
},
|
| 1282 |
{
|
| 1283 |
+
"epoch": 5.81,
|
| 1284 |
+
"grad_norm": 2.7660508155822754,
|
| 1285 |
+
"learning_rate": 4.204301075268817e-05,
|
| 1286 |
+
"loss": 0.3851,
|
| 1287 |
"step": 1620
|
| 1288 |
},
|
| 1289 |
{
|
| 1290 |
+
"epoch": 5.84,
|
| 1291 |
+
"grad_norm": 3.7340736389160156,
|
| 1292 |
+
"learning_rate": 4.1684587813620074e-05,
|
| 1293 |
+
"loss": 0.4027,
|
| 1294 |
"step": 1630
|
| 1295 |
},
|
| 1296 |
{
|
| 1297 |
+
"epoch": 5.88,
|
| 1298 |
+
"grad_norm": 3.9350473880767822,
|
| 1299 |
+
"learning_rate": 4.1326164874551976e-05,
|
| 1300 |
+
"loss": 0.2706,
|
| 1301 |
"step": 1640
|
| 1302 |
},
|
| 1303 |
{
|
| 1304 |
+
"epoch": 5.91,
|
| 1305 |
+
"grad_norm": 6.938910961151123,
|
| 1306 |
+
"learning_rate": 4.096774193548387e-05,
|
| 1307 |
+
"loss": 0.3173,
|
| 1308 |
"step": 1650
|
| 1309 |
},
|
| 1310 |
{
|
| 1311 |
+
"epoch": 5.95,
|
| 1312 |
+
"grad_norm": 2.325490713119507,
|
| 1313 |
+
"learning_rate": 4.060931899641577e-05,
|
| 1314 |
+
"loss": 0.2399,
|
| 1315 |
"step": 1660
|
| 1316 |
},
|
| 1317 |
{
|
| 1318 |
+
"epoch": 5.99,
|
| 1319 |
+
"grad_norm": 4.819274425506592,
|
| 1320 |
+
"learning_rate": 4.025089605734767e-05,
|
| 1321 |
+
"loss": 0.2409,
|
| 1322 |
"step": 1670
|
| 1323 |
},
|
| 1324 |
{
|
| 1325 |
+
"epoch": 6.02,
|
| 1326 |
+
"grad_norm": 1.799713134765625,
|
| 1327 |
+
"learning_rate": 3.989247311827957e-05,
|
| 1328 |
+
"loss": 0.3328,
|
| 1329 |
"step": 1680
|
| 1330 |
},
|
| 1331 |
{
|
| 1332 |
+
"epoch": 6.06,
|
| 1333 |
+
"grad_norm": 3.0742380619049072,
|
| 1334 |
+
"learning_rate": 3.953405017921147e-05,
|
| 1335 |
+
"loss": 0.3429,
|
| 1336 |
"step": 1690
|
| 1337 |
},
|
| 1338 |
{
|
| 1339 |
+
"epoch": 6.09,
|
| 1340 |
+
"grad_norm": 6.019571304321289,
|
| 1341 |
+
"learning_rate": 3.9175627240143367e-05,
|
| 1342 |
+
"loss": 0.2183,
|
| 1343 |
"step": 1700
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
+
"epoch": 6.09,
|
| 1347 |
+
"eval_accuracy": 0.8752556237218814,
|
| 1348 |
+
"eval_loss": 0.42549625039100647,
|
| 1349 |
+
"eval_runtime": 13.4069,
|
| 1350 |
+
"eval_samples_per_second": 72.947,
|
| 1351 |
+
"eval_steps_per_second": 9.174,
|
| 1352 |
"step": 1700
|
| 1353 |
},
|
| 1354 |
{
|
| 1355 |
+
"epoch": 6.13,
|
| 1356 |
+
"grad_norm": 3.634247303009033,
|
| 1357 |
+
"learning_rate": 3.881720430107527e-05,
|
| 1358 |
+
"loss": 0.2822,
|
| 1359 |
"step": 1710
|
| 1360 |
},
|
| 1361 |
{
|
| 1362 |
+
"epoch": 6.16,
|
| 1363 |
+
"grad_norm": 1.7334699630737305,
|
| 1364 |
+
"learning_rate": 3.845878136200717e-05,
|
| 1365 |
+
"loss": 0.342,
|
| 1366 |
"step": 1720
|
| 1367 |
},
|
| 1368 |
{
|
| 1369 |
+
"epoch": 6.2,
|
| 1370 |
+
"grad_norm": 2.8848531246185303,
|
| 1371 |
+
"learning_rate": 3.810035842293907e-05,
|
| 1372 |
+
"loss": 0.2375,
|
| 1373 |
"step": 1730
|
| 1374 |
},
|
| 1375 |
{
|
| 1376 |
+
"epoch": 6.24,
|
| 1377 |
+
"grad_norm": 8.433989524841309,
|
| 1378 |
+
"learning_rate": 3.7741935483870974e-05,
|
| 1379 |
+
"loss": 0.3862,
|
| 1380 |
"step": 1740
|
| 1381 |
},
|
| 1382 |
{
|
| 1383 |
+
"epoch": 6.27,
|
| 1384 |
+
"grad_norm": 5.4030022621154785,
|
| 1385 |
+
"learning_rate": 3.738351254480287e-05,
|
| 1386 |
+
"loss": 0.2464,
|
| 1387 |
"step": 1750
|
| 1388 |
},
|
| 1389 |
{
|
| 1390 |
+
"epoch": 6.31,
|
| 1391 |
+
"grad_norm": 0.6388465762138367,
|
| 1392 |
+
"learning_rate": 3.702508960573477e-05,
|
| 1393 |
+
"loss": 0.2487,
|
| 1394 |
"step": 1760
|
| 1395 |
},
|
| 1396 |
{
|
| 1397 |
+
"epoch": 6.34,
|
| 1398 |
+
"grad_norm": 2.5588126182556152,
|
| 1399 |
+
"learning_rate": 3.6666666666666666e-05,
|
| 1400 |
+
"loss": 0.315,
|
| 1401 |
"step": 1770
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
+
"epoch": 6.38,
|
| 1405 |
+
"grad_norm": 2.1331589221954346,
|
| 1406 |
+
"learning_rate": 3.630824372759857e-05,
|
| 1407 |
+
"loss": 0.3208,
|
| 1408 |
"step": 1780
|
| 1409 |
},
|
| 1410 |
{
|
| 1411 |
+
"epoch": 6.42,
|
| 1412 |
+
"grad_norm": 5.283192157745361,
|
| 1413 |
+
"learning_rate": 3.594982078853047e-05,
|
| 1414 |
+
"loss": 0.231,
|
| 1415 |
"step": 1790
|
| 1416 |
},
|
| 1417 |
{
|
| 1418 |
+
"epoch": 6.45,
|
| 1419 |
+
"grad_norm": 1.1380605697631836,
|
| 1420 |
+
"learning_rate": 3.5591397849462364e-05,
|
| 1421 |
+
"loss": 0.2546,
|
| 1422 |
"step": 1800
|
| 1423 |
},
|
| 1424 |
{
|
| 1425 |
+
"epoch": 6.45,
|
| 1426 |
+
"eval_accuracy": 0.8680981595092024,
|
| 1427 |
+
"eval_loss": 0.4643884301185608,
|
| 1428 |
+
"eval_runtime": 13.3117,
|
| 1429 |
+
"eval_samples_per_second": 73.469,
|
| 1430 |
+
"eval_steps_per_second": 9.24,
|
| 1431 |
"step": 1800
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
+
"epoch": 6.49,
|
| 1435 |
+
"grad_norm": 6.120726585388184,
|
| 1436 |
+
"learning_rate": 3.5232974910394266e-05,
|
| 1437 |
+
"loss": 0.3022,
|
| 1438 |
"step": 1810
|
| 1439 |
},
|
| 1440 |
{
|
| 1441 |
+
"epoch": 6.52,
|
| 1442 |
+
"grad_norm": 4.132528305053711,
|
| 1443 |
+
"learning_rate": 3.487455197132617e-05,
|
| 1444 |
+
"loss": 0.2711,
|
| 1445 |
"step": 1820
|
| 1446 |
},
|
| 1447 |
{
|
| 1448 |
+
"epoch": 6.56,
|
| 1449 |
+
"grad_norm": 0.2836366295814514,
|
| 1450 |
+
"learning_rate": 3.451612903225806e-05,
|
| 1451 |
+
"loss": 0.2759,
|
| 1452 |
"step": 1830
|
| 1453 |
},
|
| 1454 |
{
|
| 1455 |
+
"epoch": 6.59,
|
| 1456 |
+
"grad_norm": 4.501558303833008,
|
| 1457 |
+
"learning_rate": 3.4157706093189965e-05,
|
| 1458 |
+
"loss": 0.2287,
|
| 1459 |
"step": 1840
|
| 1460 |
},
|
| 1461 |
{
|
| 1462 |
+
"epoch": 6.63,
|
| 1463 |
+
"grad_norm": 7.157699108123779,
|
| 1464 |
+
"learning_rate": 3.379928315412187e-05,
|
| 1465 |
+
"loss": 0.1862,
|
| 1466 |
"step": 1850
|
| 1467 |
},
|
| 1468 |
{
|
| 1469 |
+
"epoch": 6.67,
|
| 1470 |
+
"grad_norm": 6.248540878295898,
|
| 1471 |
+
"learning_rate": 3.344086021505377e-05,
|
| 1472 |
+
"loss": 0.3247,
|
| 1473 |
"step": 1860
|
| 1474 |
},
|
| 1475 |
{
|
| 1476 |
+
"epoch": 6.7,
|
| 1477 |
+
"grad_norm": 4.108403205871582,
|
| 1478 |
+
"learning_rate": 3.308243727598567e-05,
|
| 1479 |
+
"loss": 0.3535,
|
| 1480 |
"step": 1870
|
| 1481 |
},
|
| 1482 |
{
|
| 1483 |
+
"epoch": 6.74,
|
| 1484 |
+
"grad_norm": 4.134276866912842,
|
| 1485 |
+
"learning_rate": 3.2724014336917565e-05,
|
| 1486 |
+
"loss": 0.217,
|
| 1487 |
"step": 1880
|
| 1488 |
},
|
| 1489 |
{
|
| 1490 |
+
"epoch": 6.77,
|
| 1491 |
+
"grad_norm": 2.2216591835021973,
|
| 1492 |
+
"learning_rate": 3.236559139784947e-05,
|
| 1493 |
+
"loss": 0.1901,
|
| 1494 |
"step": 1890
|
| 1495 |
},
|
| 1496 |
{
|
| 1497 |
+
"epoch": 6.81,
|
| 1498 |
+
"grad_norm": 3.7293379306793213,
|
| 1499 |
+
"learning_rate": 3.200716845878136e-05,
|
| 1500 |
+
"loss": 0.3798,
|
| 1501 |
"step": 1900
|
| 1502 |
},
|
| 1503 |
{
|
| 1504 |
+
"epoch": 6.81,
|
| 1505 |
+
"eval_accuracy": 0.852760736196319,
|
| 1506 |
+
"eval_loss": 0.4917815029621124,
|
| 1507 |
+
"eval_runtime": 13.2098,
|
| 1508 |
+
"eval_samples_per_second": 74.036,
|
| 1509 |
+
"eval_steps_per_second": 9.311,
|
| 1510 |
"step": 1900
|
| 1511 |
},
|
| 1512 |
{
|
| 1513 |
+
"epoch": 6.85,
|
| 1514 |
+
"grad_norm": 9.140816688537598,
|
| 1515 |
+
"learning_rate": 3.1648745519713264e-05,
|
| 1516 |
+
"loss": 0.3056,
|
| 1517 |
"step": 1910
|
| 1518 |
},
|
| 1519 |
{
|
| 1520 |
+
"epoch": 6.88,
|
| 1521 |
+
"grad_norm": 5.005599498748779,
|
| 1522 |
+
"learning_rate": 3.1290322580645166e-05,
|
| 1523 |
+
"loss": 0.144,
|
| 1524 |
"step": 1920
|
| 1525 |
},
|
| 1526 |
{
|
| 1527 |
+
"epoch": 6.92,
|
| 1528 |
+
"grad_norm": 8.45984172821045,
|
| 1529 |
+
"learning_rate": 3.093189964157706e-05,
|
| 1530 |
+
"loss": 0.2561,
|
| 1531 |
"step": 1930
|
| 1532 |
},
|
| 1533 |
{
|
| 1534 |
+
"epoch": 6.95,
|
| 1535 |
+
"grad_norm": 9.944540023803711,
|
| 1536 |
+
"learning_rate": 3.057347670250896e-05,
|
| 1537 |
+
"loss": 0.3116,
|
| 1538 |
"step": 1940
|
| 1539 |
},
|
| 1540 |
{
|
| 1541 |
+
"epoch": 6.99,
|
| 1542 |
+
"grad_norm": 5.759355545043945,
|
| 1543 |
+
"learning_rate": 3.0215053763440858e-05,
|
| 1544 |
+
"loss": 0.3236,
|
| 1545 |
"step": 1950
|
| 1546 |
},
|
| 1547 |
{
|
| 1548 |
+
"epoch": 7.03,
|
| 1549 |
+
"grad_norm": 3.6583213806152344,
|
| 1550 |
+
"learning_rate": 2.985663082437276e-05,
|
| 1551 |
+
"loss": 0.2651,
|
| 1552 |
"step": 1960
|
| 1553 |
},
|
| 1554 |
{
|
| 1555 |
+
"epoch": 7.06,
|
| 1556 |
+
"grad_norm": 3.7694921493530273,
|
| 1557 |
+
"learning_rate": 2.949820788530466e-05,
|
| 1558 |
+
"loss": 0.273,
|
| 1559 |
"step": 1970
|
| 1560 |
},
|
| 1561 |
{
|
| 1562 |
+
"epoch": 7.1,
|
| 1563 |
+
"grad_norm": 7.003334045410156,
|
| 1564 |
+
"learning_rate": 2.913978494623656e-05,
|
| 1565 |
+
"loss": 0.2574,
|
| 1566 |
"step": 1980
|
| 1567 |
},
|
| 1568 |
{
|
| 1569 |
+
"epoch": 7.13,
|
| 1570 |
+
"grad_norm": 5.475174427032471,
|
| 1571 |
+
"learning_rate": 2.878136200716846e-05,
|
| 1572 |
+
"loss": 0.1923,
|
| 1573 |
"step": 1990
|
| 1574 |
},
|
| 1575 |
{
|
| 1576 |
+
"epoch": 7.17,
|
| 1577 |
+
"grad_norm": 4.163592338562012,
|
| 1578 |
+
"learning_rate": 2.8422939068100357e-05,
|
| 1579 |
+
"loss": 0.2851,
|
| 1580 |
"step": 2000
|
| 1581 |
},
|
| 1582 |
{
|
| 1583 |
+
"epoch": 7.17,
|
| 1584 |
+
"eval_accuracy": 0.8660531697341514,
|
| 1585 |
+
"eval_loss": 0.45738592743873596,
|
| 1586 |
+
"eval_runtime": 13.0257,
|
| 1587 |
+
"eval_samples_per_second": 75.082,
|
| 1588 |
+
"eval_steps_per_second": 9.443,
|
| 1589 |
"step": 2000
|
| 1590 |
},
|
| 1591 |
{
|
| 1592 |
+
"epoch": 7.2,
|
| 1593 |
+
"grad_norm": 1.4400478601455688,
|
| 1594 |
+
"learning_rate": 2.806451612903226e-05,
|
| 1595 |
+
"loss": 0.157,
|
| 1596 |
"step": 2010
|
| 1597 |
},
|
| 1598 |
{
|
| 1599 |
+
"epoch": 7.24,
|
| 1600 |
+
"grad_norm": 3.958745241165161,
|
| 1601 |
+
"learning_rate": 2.770609318996416e-05,
|
| 1602 |
+
"loss": 0.2556,
|
| 1603 |
"step": 2020
|
| 1604 |
},
|
| 1605 |
{
|
| 1606 |
+
"epoch": 7.28,
|
| 1607 |
+
"grad_norm": 2.083286762237549,
|
| 1608 |
+
"learning_rate": 2.734767025089606e-05,
|
| 1609 |
+
"loss": 0.2341,
|
| 1610 |
"step": 2030
|
| 1611 |
},
|
| 1612 |
{
|
| 1613 |
+
"epoch": 7.31,
|
| 1614 |
+
"grad_norm": 9.330533027648926,
|
| 1615 |
+
"learning_rate": 2.698924731182796e-05,
|
| 1616 |
+
"loss": 0.2742,
|
| 1617 |
"step": 2040
|
| 1618 |
},
|
| 1619 |
{
|
| 1620 |
+
"epoch": 7.35,
|
| 1621 |
+
"grad_norm": 1.6954762935638428,
|
| 1622 |
+
"learning_rate": 2.6630824372759862e-05,
|
| 1623 |
+
"loss": 0.1567,
|
| 1624 |
"step": 2050
|
| 1625 |
},
|
| 1626 |
{
|
| 1627 |
+
"epoch": 7.38,
|
| 1628 |
+
"grad_norm": 12.555899620056152,
|
| 1629 |
+
"learning_rate": 2.6272401433691757e-05,
|
| 1630 |
+
"loss": 0.1988,
|
| 1631 |
"step": 2060
|
| 1632 |
},
|
| 1633 |
{
|
| 1634 |
+
"epoch": 7.42,
|
| 1635 |
+
"grad_norm": 2.7985002994537354,
|
| 1636 |
+
"learning_rate": 2.591397849462366e-05,
|
| 1637 |
+
"loss": 0.2954,
|
| 1638 |
"step": 2070
|
| 1639 |
},
|
| 1640 |
{
|
| 1641 |
+
"epoch": 7.46,
|
| 1642 |
+
"grad_norm": 3.0381124019622803,
|
| 1643 |
+
"learning_rate": 2.5555555555555554e-05,
|
| 1644 |
+
"loss": 0.2345,
|
| 1645 |
"step": 2080
|
| 1646 |
},
|
| 1647 |
{
|
| 1648 |
+
"epoch": 7.49,
|
| 1649 |
+
"grad_norm": 4.35617208480835,
|
| 1650 |
+
"learning_rate": 2.5197132616487456e-05,
|
| 1651 |
+
"loss": 0.2736,
|
| 1652 |
"step": 2090
|
| 1653 |
},
|
| 1654 |
{
|
| 1655 |
+
"epoch": 7.53,
|
| 1656 |
+
"grad_norm": 7.855186939239502,
|
| 1657 |
+
"learning_rate": 2.4838709677419354e-05,
|
| 1658 |
+
"loss": 0.3897,
|
| 1659 |
"step": 2100
|
| 1660 |
},
|
| 1661 |
{
|
| 1662 |
+
"epoch": 7.53,
|
| 1663 |
+
"eval_accuracy": 0.8650306748466258,
|
| 1664 |
+
"eval_loss": 0.47503402829170227,
|
| 1665 |
+
"eval_runtime": 13.2431,
|
| 1666 |
+
"eval_samples_per_second": 73.85,
|
| 1667 |
+
"eval_steps_per_second": 9.288,
|
| 1668 |
"step": 2100
|
| 1669 |
},
|
| 1670 |
{
|
| 1671 |
+
"epoch": 7.56,
|
| 1672 |
+
"grad_norm": 8.723631858825684,
|
| 1673 |
+
"learning_rate": 2.4480286738351256e-05,
|
| 1674 |
+
"loss": 0.2814,
|
| 1675 |
"step": 2110
|
| 1676 |
},
|
| 1677 |
{
|
| 1678 |
+
"epoch": 7.6,
|
| 1679 |
+
"grad_norm": 8.104424476623535,
|
| 1680 |
+
"learning_rate": 2.4121863799283158e-05,
|
| 1681 |
+
"loss": 0.345,
|
| 1682 |
"step": 2120
|
| 1683 |
},
|
| 1684 |
{
|
| 1685 |
+
"epoch": 7.63,
|
| 1686 |
+
"grad_norm": 2.0964772701263428,
|
| 1687 |
+
"learning_rate": 2.3763440860215056e-05,
|
| 1688 |
+
"loss": 0.1643,
|
| 1689 |
"step": 2130
|
| 1690 |
},
|
| 1691 |
{
|
| 1692 |
+
"epoch": 7.67,
|
| 1693 |
+
"grad_norm": 6.627722263336182,
|
| 1694 |
+
"learning_rate": 2.3405017921146955e-05,
|
| 1695 |
+
"loss": 0.2667,
|
| 1696 |
"step": 2140
|
| 1697 |
},
|
| 1698 |
{
|
| 1699 |
+
"epoch": 7.71,
|
| 1700 |
+
"grad_norm": 8.275358200073242,
|
| 1701 |
+
"learning_rate": 2.3046594982078853e-05,
|
| 1702 |
+
"loss": 0.1747,
|
| 1703 |
"step": 2150
|
| 1704 |
},
|
| 1705 |
{
|
| 1706 |
+
"epoch": 7.74,
|
| 1707 |
+
"grad_norm": 2.7480711936950684,
|
| 1708 |
+
"learning_rate": 2.268817204301075e-05,
|
| 1709 |
+
"loss": 0.2789,
|
| 1710 |
"step": 2160
|
| 1711 |
},
|
| 1712 |
{
|
| 1713 |
+
"epoch": 7.78,
|
| 1714 |
+
"grad_norm": 4.440505504608154,
|
| 1715 |
+
"learning_rate": 2.2329749103942653e-05,
|
| 1716 |
+
"loss": 0.244,
|
| 1717 |
"step": 2170
|
| 1718 |
},
|
| 1719 |
{
|
| 1720 |
+
"epoch": 7.81,
|
| 1721 |
+
"grad_norm": 3.1027097702026367,
|
| 1722 |
+
"learning_rate": 2.1971326164874552e-05,
|
| 1723 |
+
"loss": 0.1924,
|
| 1724 |
"step": 2180
|
| 1725 |
},
|
| 1726 |
{
|
| 1727 |
+
"epoch": 7.85,
|
| 1728 |
+
"grad_norm": 0.49080890417099,
|
| 1729 |
+
"learning_rate": 2.1612903225806454e-05,
|
| 1730 |
+
"loss": 0.1648,
|
| 1731 |
"step": 2190
|
| 1732 |
},
|
| 1733 |
{
|
| 1734 |
+
"epoch": 7.89,
|
| 1735 |
+
"grad_norm": 7.749617099761963,
|
| 1736 |
+
"learning_rate": 2.1254480286738352e-05,
|
| 1737 |
+
"loss": 0.2879,
|
| 1738 |
"step": 2200
|
| 1739 |
},
|
| 1740 |
{
|
| 1741 |
+
"epoch": 7.89,
|
| 1742 |
+
"eval_accuracy": 0.8640081799591002,
|
| 1743 |
+
"eval_loss": 0.4739380478858948,
|
| 1744 |
+
"eval_runtime": 13.2662,
|
| 1745 |
+
"eval_samples_per_second": 73.721,
|
| 1746 |
+
"eval_steps_per_second": 9.272,
|
| 1747 |
"step": 2200
|
| 1748 |
},
|
| 1749 |
{
|
| 1750 |
+
"epoch": 7.92,
|
| 1751 |
+
"grad_norm": 4.843421459197998,
|
| 1752 |
+
"learning_rate": 2.0896057347670254e-05,
|
| 1753 |
+
"loss": 0.1315,
|
| 1754 |
"step": 2210
|
| 1755 |
},
|
| 1756 |
{
|
| 1757 |
+
"epoch": 7.96,
|
| 1758 |
+
"grad_norm": 11.301743507385254,
|
| 1759 |
+
"learning_rate": 2.0537634408602152e-05,
|
| 1760 |
+
"loss": 0.2476,
|
| 1761 |
"step": 2220
|
| 1762 |
},
|
| 1763 |
{
|
| 1764 |
+
"epoch": 7.99,
|
| 1765 |
+
"grad_norm": 5.654230117797852,
|
| 1766 |
+
"learning_rate": 2.017921146953405e-05,
|
| 1767 |
+
"loss": 0.2347,
|
| 1768 |
"step": 2230
|
| 1769 |
},
|
| 1770 |
{
|
| 1771 |
+
"epoch": 8.03,
|
| 1772 |
+
"grad_norm": 1.8168110847473145,
|
| 1773 |
+
"learning_rate": 1.982078853046595e-05,
|
| 1774 |
+
"loss": 0.2233,
|
| 1775 |
"step": 2240
|
| 1776 |
},
|
| 1777 |
{
|
| 1778 |
+
"epoch": 8.06,
|
| 1779 |
+
"grad_norm": 6.872961521148682,
|
| 1780 |
+
"learning_rate": 1.9462365591397848e-05,
|
| 1781 |
+
"loss": 0.295,
|
| 1782 |
"step": 2250
|
| 1783 |
},
|
| 1784 |
{
|
| 1785 |
+
"epoch": 8.1,
|
| 1786 |
+
"grad_norm": 1.988438606262207,
|
| 1787 |
+
"learning_rate": 1.910394265232975e-05,
|
| 1788 |
+
"loss": 0.2193,
|
| 1789 |
"step": 2260
|
| 1790 |
},
|
| 1791 |
{
|
| 1792 |
+
"epoch": 8.14,
|
| 1793 |
+
"grad_norm": 2.3927271366119385,
|
| 1794 |
+
"learning_rate": 1.874551971326165e-05,
|
| 1795 |
+
"loss": 0.1755,
|
| 1796 |
"step": 2270
|
| 1797 |
},
|
| 1798 |
{
|
| 1799 |
+
"epoch": 8.17,
|
| 1800 |
+
"grad_norm": 9.944202423095703,
|
| 1801 |
+
"learning_rate": 1.838709677419355e-05,
|
| 1802 |
+
"loss": 0.3156,
|
| 1803 |
"step": 2280
|
| 1804 |
},
|
| 1805 |
{
|
| 1806 |
+
"epoch": 8.21,
|
| 1807 |
+
"grad_norm": 1.3722007274627686,
|
| 1808 |
+
"learning_rate": 1.8028673835125448e-05,
|
| 1809 |
+
"loss": 0.1877,
|
| 1810 |
"step": 2290
|
| 1811 |
},
|
| 1812 |
{
|
| 1813 |
+
"epoch": 8.24,
|
| 1814 |
+
"grad_norm": 2.698289394378662,
|
| 1815 |
+
"learning_rate": 1.767025089605735e-05,
|
| 1816 |
+
"loss": 0.1619,
|
| 1817 |
"step": 2300
|
| 1818 |
},
|
| 1819 |
{
|
| 1820 |
+
"epoch": 8.24,
|
| 1821 |
+
"eval_accuracy": 0.8701431492842536,
|
| 1822 |
+
"eval_loss": 0.46446332335472107,
|
| 1823 |
+
"eval_runtime": 13.2992,
|
| 1824 |
+
"eval_samples_per_second": 73.538,
|
| 1825 |
+
"eval_steps_per_second": 9.249,
|
| 1826 |
"step": 2300
|
| 1827 |
},
|
| 1828 |
{
|
| 1829 |
+
"epoch": 8.28,
|
| 1830 |
+
"grad_norm": 4.195601463317871,
|
| 1831 |
+
"learning_rate": 1.7311827956989248e-05,
|
| 1832 |
+
"loss": 0.1302,
|
| 1833 |
"step": 2310
|
| 1834 |
},
|
| 1835 |
{
|
| 1836 |
+
"epoch": 8.32,
|
| 1837 |
+
"grad_norm": 8.367218971252441,
|
| 1838 |
+
"learning_rate": 1.6953405017921147e-05,
|
| 1839 |
+
"loss": 0.2366,
|
| 1840 |
"step": 2320
|
| 1841 |
},
|
| 1842 |
{
|
| 1843 |
+
"epoch": 8.35,
|
| 1844 |
+
"grad_norm": 1.637813925743103,
|
| 1845 |
+
"learning_rate": 1.659498207885305e-05,
|
| 1846 |
+
"loss": 0.1395,
|
| 1847 |
"step": 2330
|
| 1848 |
},
|
| 1849 |
{
|
| 1850 |
+
"epoch": 8.39,
|
| 1851 |
+
"grad_norm": 7.9814677238464355,
|
| 1852 |
+
"learning_rate": 1.6236559139784947e-05,
|
| 1853 |
+
"loss": 0.2804,
|
| 1854 |
"step": 2340
|
| 1855 |
},
|
| 1856 |
{
|
| 1857 |
+
"epoch": 8.42,
|
| 1858 |
+
"grad_norm": 0.23666299879550934,
|
| 1859 |
+
"learning_rate": 1.587813620071685e-05,
|
| 1860 |
+
"loss": 0.2186,
|
| 1861 |
"step": 2350
|
| 1862 |
},
|
| 1863 |
{
|
| 1864 |
+
"epoch": 8.46,
|
| 1865 |
+
"grad_norm": 7.456205368041992,
|
| 1866 |
+
"learning_rate": 1.5519713261648747e-05,
|
| 1867 |
+
"loss": 0.2733,
|
| 1868 |
"step": 2360
|
| 1869 |
},
|
| 1870 |
{
|
| 1871 |
+
"epoch": 8.49,
|
| 1872 |
+
"grad_norm": 3.73573899269104,
|
| 1873 |
+
"learning_rate": 1.5161290322580646e-05,
|
| 1874 |
+
"loss": 0.2891,
|
| 1875 |
"step": 2370
|
| 1876 |
},
|
| 1877 |
{
|
| 1878 |
+
"epoch": 8.53,
|
| 1879 |
+
"grad_norm": 1.723080039024353,
|
| 1880 |
+
"learning_rate": 1.4802867383512544e-05,
|
| 1881 |
+
"loss": 0.2959,
|
| 1882 |
"step": 2380
|
| 1883 |
},
|
| 1884 |
{
|
| 1885 |
+
"epoch": 8.57,
|
| 1886 |
+
"grad_norm": 2.3597002029418945,
|
| 1887 |
+
"learning_rate": 1.4444444444444444e-05,
|
| 1888 |
+
"loss": 0.2784,
|
| 1889 |
"step": 2390
|
| 1890 |
},
|
| 1891 |
{
|
| 1892 |
+
"epoch": 8.6,
|
| 1893 |
+
"grad_norm": 0.19159385561943054,
|
| 1894 |
+
"learning_rate": 1.4086021505376346e-05,
|
| 1895 |
+
"loss": 0.1791,
|
| 1896 |
"step": 2400
|
| 1897 |
},
|
| 1898 |
{
|
| 1899 |
+
"epoch": 8.6,
|
| 1900 |
+
"eval_accuracy": 0.8670756646216768,
|
| 1901 |
+
"eval_loss": 0.46364837884902954,
|
| 1902 |
+
"eval_runtime": 13.0224,
|
| 1903 |
+
"eval_samples_per_second": 75.101,
|
| 1904 |
+
"eval_steps_per_second": 9.445,
|
| 1905 |
"step": 2400
|
| 1906 |
},
|
| 1907 |
{
|
| 1908 |
+
"epoch": 8.64,
|
| 1909 |
+
"grad_norm": 8.25283145904541,
|
| 1910 |
+
"learning_rate": 1.3727598566308244e-05,
|
| 1911 |
+
"loss": 0.2331,
|
| 1912 |
"step": 2410
|
| 1913 |
},
|
| 1914 |
{
|
| 1915 |
+
"epoch": 8.67,
|
| 1916 |
+
"grad_norm": 4.816408157348633,
|
| 1917 |
+
"learning_rate": 1.3369175627240143e-05,
|
| 1918 |
+
"loss": 0.1704,
|
| 1919 |
"step": 2420
|
| 1920 |
},
|
| 1921 |
{
|
| 1922 |
+
"epoch": 8.71,
|
| 1923 |
+
"grad_norm": 10.364704132080078,
|
| 1924 |
+
"learning_rate": 1.3010752688172043e-05,
|
| 1925 |
+
"loss": 0.2854,
|
| 1926 |
"step": 2430
|
| 1927 |
},
|
| 1928 |
{
|
| 1929 |
+
"epoch": 8.75,
|
| 1930 |
+
"grad_norm": 6.165342807769775,
|
| 1931 |
+
"learning_rate": 1.2652329749103945e-05,
|
| 1932 |
+
"loss": 0.1635,
|
| 1933 |
"step": 2440
|
| 1934 |
},
|
| 1935 |
{
|
| 1936 |
+
"epoch": 8.78,
|
| 1937 |
+
"grad_norm": 10.325529098510742,
|
| 1938 |
+
"learning_rate": 1.2293906810035843e-05,
|
| 1939 |
+
"loss": 0.2685,
|
| 1940 |
"step": 2450
|
| 1941 |
},
|
| 1942 |
{
|
| 1943 |
+
"epoch": 8.82,
|
| 1944 |
+
"grad_norm": 1.4126335382461548,
|
| 1945 |
+
"learning_rate": 1.1935483870967743e-05,
|
| 1946 |
+
"loss": 0.1575,
|
| 1947 |
"step": 2460
|
| 1948 |
},
|
| 1949 |
{
|
| 1950 |
+
"epoch": 8.85,
|
| 1951 |
+
"grad_norm": 8.191924095153809,
|
| 1952 |
+
"learning_rate": 1.1577060931899642e-05,
|
| 1953 |
+
"loss": 0.2585,
|
| 1954 |
"step": 2470
|
| 1955 |
},
|
| 1956 |
{
|
| 1957 |
+
"epoch": 8.89,
|
| 1958 |
+
"grad_norm": 6.928045272827148,
|
| 1959 |
+
"learning_rate": 1.1218637992831542e-05,
|
| 1960 |
+
"loss": 0.2129,
|
| 1961 |
"step": 2480
|
| 1962 |
},
|
| 1963 |
{
|
| 1964 |
+
"epoch": 8.92,
|
| 1965 |
+
"grad_norm": 2.3231565952301025,
|
| 1966 |
+
"learning_rate": 1.086021505376344e-05,
|
| 1967 |
+
"loss": 0.3085,
|
| 1968 |
"step": 2490
|
| 1969 |
},
|
| 1970 |
{
|
| 1971 |
+
"epoch": 8.96,
|
| 1972 |
+
"grad_norm": 2.7480709552764893,
|
| 1973 |
+
"learning_rate": 1.0501792114695342e-05,
|
| 1974 |
+
"loss": 0.1697,
|
| 1975 |
"step": 2500
|
| 1976 |
},
|
| 1977 |
{
|
| 1978 |
+
"epoch": 8.96,
|
| 1979 |
+
"eval_accuracy": 0.8640081799591002,
|
| 1980 |
+
"eval_loss": 0.4716458022594452,
|
| 1981 |
+
"eval_runtime": 13.2964,
|
| 1982 |
+
"eval_samples_per_second": 73.554,
|
| 1983 |
+
"eval_steps_per_second": 9.251,
|
| 1984 |
"step": 2500
|
| 1985 |
},
|
| 1986 |
{
|
| 1987 |
+
"epoch": 9.0,
|
| 1988 |
+
"grad_norm": 6.852436542510986,
|
| 1989 |
+
"learning_rate": 1.014336917562724e-05,
|
| 1990 |
+
"loss": 0.1856,
|
| 1991 |
"step": 2510
|
| 1992 |
},
|
| 1993 |
{
|
| 1994 |
+
"epoch": 9.03,
|
| 1995 |
+
"grad_norm": 5.506394863128662,
|
| 1996 |
+
"learning_rate": 9.78494623655914e-06,
|
| 1997 |
+
"loss": 0.1951,
|
| 1998 |
"step": 2520
|
| 1999 |
},
|
| 2000 |
{
|
| 2001 |
+
"epoch": 9.07,
|
| 2002 |
+
"grad_norm": 3.1270835399627686,
|
| 2003 |
+
"learning_rate": 9.426523297491039e-06,
|
| 2004 |
+
"loss": 0.1896,
|
| 2005 |
"step": 2530
|
| 2006 |
},
|
| 2007 |
{
|
| 2008 |
+
"epoch": 9.1,
|
| 2009 |
+
"grad_norm": 3.797590732574463,
|
| 2010 |
+
"learning_rate": 9.06810035842294e-06,
|
| 2011 |
+
"loss": 0.2276,
|
| 2012 |
"step": 2540
|
| 2013 |
},
|
| 2014 |
{
|
| 2015 |
+
"epoch": 9.14,
|
| 2016 |
+
"grad_norm": 4.3236985206604,
|
| 2017 |
+
"learning_rate": 8.70967741935484e-06,
|
| 2018 |
+
"loss": 0.2272,
|
| 2019 |
"step": 2550
|
| 2020 |
},
|
| 2021 |
{
|
| 2022 |
+
"epoch": 9.18,
|
| 2023 |
+
"grad_norm": 4.975371837615967,
|
| 2024 |
+
"learning_rate": 8.351254480286738e-06,
|
| 2025 |
+
"loss": 0.2491,
|
| 2026 |
"step": 2560
|
| 2027 |
},
|
| 2028 |
{
|
| 2029 |
+
"epoch": 9.21,
|
| 2030 |
+
"grad_norm": 1.405785083770752,
|
| 2031 |
+
"learning_rate": 7.992831541218638e-06,
|
| 2032 |
+
"loss": 0.1985,
|
| 2033 |
"step": 2570
|
| 2034 |
},
|
| 2035 |
{
|
| 2036 |
+
"epoch": 9.25,
|
| 2037 |
+
"grad_norm": 0.13112181425094604,
|
| 2038 |
+
"learning_rate": 7.634408602150538e-06,
|
| 2039 |
+
"loss": 0.1134,
|
| 2040 |
"step": 2580
|
| 2041 |
},
|
| 2042 |
{
|
| 2043 |
+
"epoch": 9.28,
|
| 2044 |
+
"grad_norm": 2.597059965133667,
|
| 2045 |
+
"learning_rate": 7.275985663082438e-06,
|
| 2046 |
+
"loss": 0.229,
|
| 2047 |
"step": 2590
|
| 2048 |
},
|
| 2049 |
{
|
| 2050 |
+
"epoch": 9.32,
|
| 2051 |
+
"grad_norm": 0.1491578221321106,
|
| 2052 |
+
"learning_rate": 6.917562724014337e-06,
|
| 2053 |
+
"loss": 0.1608,
|
| 2054 |
"step": 2600
|
| 2055 |
},
|
| 2056 |
{
|
| 2057 |
+
"epoch": 9.32,
|
| 2058 |
+
"eval_accuracy": 0.8742331288343558,
|
| 2059 |
+
"eval_loss": 0.4680761694908142,
|
| 2060 |
+
"eval_runtime": 13.2277,
|
| 2061 |
+
"eval_samples_per_second": 73.936,
|
| 2062 |
+
"eval_steps_per_second": 9.299,
|
| 2063 |
"step": 2600
|
| 2064 |
},
|
| 2065 |
{
|
| 2066 |
+
"epoch": 9.35,
|
| 2067 |
+
"grad_norm": 3.264333724975586,
|
| 2068 |
+
"learning_rate": 6.559139784946237e-06,
|
| 2069 |
+
"loss": 0.2228,
|
| 2070 |
"step": 2610
|
| 2071 |
},
|
| 2072 |
{
|
| 2073 |
+
"epoch": 9.39,
|
| 2074 |
+
"grad_norm": 3.251711130142212,
|
| 2075 |
+
"learning_rate": 6.200716845878137e-06,
|
| 2076 |
+
"loss": 0.1504,
|
| 2077 |
"step": 2620
|
| 2078 |
},
|
| 2079 |
{
|
| 2080 |
+
"epoch": 9.43,
|
| 2081 |
+
"grad_norm": 0.05231141671538353,
|
| 2082 |
+
"learning_rate": 5.842293906810036e-06,
|
| 2083 |
+
"loss": 0.1097,
|
| 2084 |
"step": 2630
|
| 2085 |
},
|
| 2086 |
{
|
| 2087 |
+
"epoch": 9.46,
|
| 2088 |
+
"grad_norm": 6.941389083862305,
|
| 2089 |
+
"learning_rate": 5.483870967741936e-06,
|
| 2090 |
+
"loss": 0.2046,
|
| 2091 |
"step": 2640
|
| 2092 |
},
|
| 2093 |
{
|
| 2094 |
+
"epoch": 9.5,
|
| 2095 |
+
"grad_norm": 1.9816231727600098,
|
| 2096 |
+
"learning_rate": 5.125448028673835e-06,
|
| 2097 |
+
"loss": 0.2879,
|
| 2098 |
"step": 2650
|
| 2099 |
},
|
| 2100 |
{
|
| 2101 |
+
"epoch": 9.53,
|
| 2102 |
+
"grad_norm": 1.5215052366256714,
|
| 2103 |
+
"learning_rate": 4.767025089605735e-06,
|
| 2104 |
+
"loss": 0.2352,
|
| 2105 |
"step": 2660
|
| 2106 |
},
|
| 2107 |
{
|
| 2108 |
+
"epoch": 9.57,
|
| 2109 |
+
"grad_norm": 0.055237527936697006,
|
| 2110 |
+
"learning_rate": 4.408602150537635e-06,
|
| 2111 |
+
"loss": 0.1954,
|
| 2112 |
"step": 2670
|
| 2113 |
},
|
| 2114 |
{
|
| 2115 |
+
"epoch": 9.61,
|
| 2116 |
+
"grad_norm": 4.320139408111572,
|
| 2117 |
+
"learning_rate": 4.050179211469534e-06,
|
| 2118 |
+
"loss": 0.3845,
|
| 2119 |
"step": 2680
|
| 2120 |
},
|
| 2121 |
{
|
| 2122 |
+
"epoch": 9.64,
|
| 2123 |
+
"grad_norm": 3.767547607421875,
|
| 2124 |
+
"learning_rate": 3.6917562724014336e-06,
|
| 2125 |
+
"loss": 0.3031,
|
| 2126 |
"step": 2690
|
| 2127 |
},
|
| 2128 |
{
|
| 2129 |
+
"epoch": 9.68,
|
| 2130 |
+
"grad_norm": 0.48145973682403564,
|
| 2131 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 2132 |
+
"loss": 0.2105,
|
| 2133 |
"step": 2700
|
| 2134 |
},
|
| 2135 |
{
|
| 2136 |
+
"epoch": 9.68,
|
| 2137 |
+
"eval_accuracy": 0.8721881390593047,
|
| 2138 |
+
"eval_loss": 0.47190210223197937,
|
| 2139 |
+
"eval_runtime": 13.2161,
|
| 2140 |
+
"eval_samples_per_second": 74.001,
|
| 2141 |
+
"eval_steps_per_second": 9.307,
|
| 2142 |
"step": 2700
|
| 2143 |
},
|
| 2144 |
{
|
| 2145 |
+
"epoch": 9.71,
|
| 2146 |
+
"grad_norm": 1.6004977226257324,
|
| 2147 |
+
"learning_rate": 2.974910394265233e-06,
|
| 2148 |
+
"loss": 0.2186,
|
| 2149 |
"step": 2710
|
| 2150 |
},
|
| 2151 |
{
|
| 2152 |
+
"epoch": 9.75,
|
| 2153 |
+
"grad_norm": 0.4472026228904724,
|
| 2154 |
+
"learning_rate": 2.6164874551971327e-06,
|
| 2155 |
+
"loss": 0.1528,
|
| 2156 |
"step": 2720
|
| 2157 |
},
|
| 2158 |
+
{
|
| 2159 |
+
"epoch": 9.78,
|
| 2160 |
+
"grad_norm": 0.12501764297485352,
|
| 2161 |
+
"learning_rate": 2.2580645161290324e-06,
|
| 2162 |
+
"loss": 0.1218,
|
| 2163 |
+
"step": 2730
|
| 2164 |
+
},
|
| 2165 |
+
{
|
| 2166 |
+
"epoch": 9.82,
|
| 2167 |
+
"grad_norm": 2.206127405166626,
|
| 2168 |
+
"learning_rate": 1.8996415770609319e-06,
|
| 2169 |
+
"loss": 0.1324,
|
| 2170 |
+
"step": 2740
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 9.86,
|
| 2174 |
+
"grad_norm": 0.9404085278511047,
|
| 2175 |
+
"learning_rate": 1.5412186379928316e-06,
|
| 2176 |
+
"loss": 0.2832,
|
| 2177 |
+
"step": 2750
|
| 2178 |
+
},
|
| 2179 |
+
{
|
| 2180 |
+
"epoch": 9.89,
|
| 2181 |
+
"grad_norm": 5.883946418762207,
|
| 2182 |
+
"learning_rate": 1.1827956989247313e-06,
|
| 2183 |
+
"loss": 0.2113,
|
| 2184 |
+
"step": 2760
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 9.93,
|
| 2188 |
+
"grad_norm": 5.288418292999268,
|
| 2189 |
+
"learning_rate": 8.243727598566309e-07,
|
| 2190 |
+
"loss": 0.1662,
|
| 2191 |
+
"step": 2770
|
| 2192 |
+
},
|
| 2193 |
+
{
|
| 2194 |
+
"epoch": 9.96,
|
| 2195 |
+
"grad_norm": 8.01339054107666,
|
| 2196 |
+
"learning_rate": 4.6594982078853055e-07,
|
| 2197 |
+
"loss": 0.1672,
|
| 2198 |
+
"step": 2780
|
| 2199 |
+
},
|
| 2200 |
+
{
|
| 2201 |
+
"epoch": 10.0,
|
| 2202 |
+
"grad_norm": 7.813081741333008,
|
| 2203 |
+
"learning_rate": 1.0752688172043011e-07,
|
| 2204 |
+
"loss": 0.1825,
|
| 2205 |
+
"step": 2790
|
| 2206 |
+
},
|
| 2207 |
{
|
| 2208 |
"epoch": 10.0,
|
| 2209 |
+
"step": 2790,
|
| 2210 |
+
"total_flos": 3.451740694569861e+18,
|
| 2211 |
+
"train_loss": 0.45333294201922675,
|
| 2212 |
+
"train_runtime": 1524.4216,
|
| 2213 |
+
"train_samples_per_second": 29.218,
|
| 2214 |
+
"train_steps_per_second": 1.83
|
| 2215 |
}
|
| 2216 |
],
|
| 2217 |
"logging_steps": 10,
|
| 2218 |
+
"max_steps": 2790,
|
| 2219 |
"num_input_tokens_seen": 0,
|
| 2220 |
"num_train_epochs": 10,
|
| 2221 |
"save_steps": 100,
|
| 2222 |
+
"total_flos": 3.451740694569861e+18,
|
| 2223 |
"train_batch_size": 16,
|
| 2224 |
"trial_name": null,
|
| 2225 |
"trial_params": null
|