Model save
Browse files- README.md +1 -1
- all_results.json +5 -5
- model.safetensors +1 -1
- runs/Feb14_06-44-19_yekyung-nah-0/events.out.tfevents.1739483078.yekyung-nah-0.3735356.0 +2 -2
- runs/Feb14_06-47-41_yekyung-nah-0/events.out.tfevents.1739483278.yekyung-nah-0.3747757.0 +3 -0
- runs/Feb14_06-47-47_yekyung-nah-0/events.out.tfevents.1739483287.yekyung-nah-0.3748246.0 +3 -0
- train_results.json +5 -5
- trainer_state.json +444 -493
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -27,7 +27,7 @@ print(output["generated_text"])
|
|
| 27 |
|
| 28 |
## Training procedure
|
| 29 |
|
| 30 |
-
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yknxxh-seoul-national-university/huggingface/runs/
|
| 31 |
|
| 32 |
This model was trained with SFT.
|
| 33 |
|
|
|
|
| 27 |
|
| 28 |
## Training procedure
|
| 29 |
|
| 30 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yknxxh-seoul-national-university/huggingface/runs/qk077co9)
|
| 31 |
|
| 32 |
This model was trained with SFT.
|
| 33 |
|
all_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 10000,
|
| 7 |
-
"train_samples_per_second": 38.
|
| 8 |
-
"train_steps_per_second": 2.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.5856227500097536e+16,
|
| 4 |
+
"train_loss": 1.017678025019103,
|
| 5 |
+
"train_runtime": 224.2528,
|
| 6 |
"train_samples": 10000,
|
| 7 |
+
"train_samples_per_second": 38.568,
|
| 8 |
+
"train_steps_per_second": 2.421
|
| 9 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2471645608
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e1d80a91a84e41f750f86c8e62de9533f001784a64e4505162111138a38af21
|
| 3 |
size 2471645608
|
runs/Feb14_06-44-19_yekyung-nah-0/events.out.tfevents.1739483078.yekyung-nah-0.3735356.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0619ad61ddd9702259f41551a76d2a8d20c1d373543c7fe2cfc85b3cc05f2882
|
| 3 |
+
size 28858
|
runs/Feb14_06-47-41_yekyung-nah-0/events.out.tfevents.1739483278.yekyung-nah-0.3747757.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1356a3b87766e684961517e4490bb72c4798f8f3eb9a3c37bd1f1e92a3fc45f6
|
| 3 |
+
size 10577
|
runs/Feb14_06-47-47_yekyung-nah-0/events.out.tfevents.1739483287.yekyung-nah-0.3748246.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f9a74a3bb1e50421899df919e0ee24f13c88d5b1c4097baaf2aed112704f7ef
|
| 3 |
+
size 9956
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 10000,
|
| 7 |
-
"train_samples_per_second": 38.
|
| 8 |
-
"train_steps_per_second": 2.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
+
"total_flos": 2.5856227500097536e+16,
|
| 4 |
+
"train_loss": 1.017678025019103,
|
| 5 |
+
"train_runtime": 224.2528,
|
| 6 |
"train_samples": 10000,
|
| 7 |
+
"train_samples_per_second": 38.568,
|
| 8 |
+
"train_steps_per_second": 2.421
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -3,835 +3,786 @@
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 3.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm": 7.
|
| 14 |
-
"learning_rate": 3.
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm": 7.
|
| 21 |
-
"learning_rate": 1.
|
| 22 |
-
"loss": 2.
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 7.
|
| 28 |
-
"learning_rate": 3.
|
| 29 |
-
"loss": 2.
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 7.
|
| 35 |
-
"learning_rate": 5.
|
| 36 |
-
"loss": 2.
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 2.
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm": 3.
|
| 56 |
-
"learning_rate": 1.
|
| 57 |
-
"loss":
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm": 3.
|
| 63 |
-
"learning_rate": 1.
|
| 64 |
-
"loss": 1.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm": 2.
|
| 70 |
-
"learning_rate": 1.
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm": 2.
|
| 77 |
-
"learning_rate": 1.
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm": 1.
|
| 84 |
-
"learning_rate": 1.
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 1.
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm": 1.
|
| 98 |
-
"learning_rate": 1.
|
| 99 |
-
"loss": 1.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
"grad_norm": 1.0078125,
|
| 105 |
-
"learning_rate": 1.
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
-
"learning_rate": 1.
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
-
"learning_rate": 1.
|
| 120 |
-
"loss": 1.
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
-
"learning_rate": 1.
|
| 127 |
-
"loss": 1.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0.
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
-
"learning_rate": 1.
|
| 134 |
-
"loss": 1.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
-
"learning_rate": 1.
|
| 141 |
-
"loss":
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch": 0.
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
-
"learning_rate": 1.
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
-
"learning_rate": 1.
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch": 0.
|
| 160 |
-
"grad_norm": 0.
|
| 161 |
-
"learning_rate": 1.
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch": 0.
|
| 167 |
-
"grad_norm": 0.
|
| 168 |
-
"learning_rate": 1.
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch": 0.
|
| 174 |
-
"grad_norm": 0.
|
| 175 |
-
"learning_rate": 1.
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
-
"learning_rate": 1.
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch": 0.
|
| 188 |
-
"grad_norm": 0.
|
| 189 |
-
"learning_rate": 1.
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch": 0.
|
| 195 |
-
"grad_norm": 0.
|
| 196 |
-
"learning_rate": 1.
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch": 0.
|
| 202 |
-
"grad_norm": 0.
|
| 203 |
-
"learning_rate": 1.
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch": 0.
|
| 209 |
-
"grad_norm": 0.
|
| 210 |
-
"learning_rate": 1.
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch": 0.
|
| 216 |
-
"grad_norm": 0.
|
| 217 |
-
"learning_rate": 1.
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch": 0.
|
| 223 |
-
"grad_norm": 0.
|
| 224 |
-
"learning_rate": 1.
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch": 0.
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
-
"learning_rate": 1.
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 0.
|
| 237 |
-
"grad_norm": 0.
|
| 238 |
-
"learning_rate": 1.
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"grad_norm": 0.
|
| 245 |
-
"learning_rate": 1.
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch": 0.
|
| 251 |
-
"grad_norm": 0.
|
| 252 |
-
"learning_rate": 1.
|
| 253 |
-
"loss": 0.
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 0.
|
| 258 |
-
"grad_norm": 0.
|
| 259 |
-
"learning_rate": 1.
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 0.
|
| 265 |
-
"grad_norm": 0.
|
| 266 |
-
"learning_rate": 1.
|
| 267 |
-
"loss": 0.
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
-
"epoch":
|
| 272 |
-
"grad_norm": 0.
|
| 273 |
-
"learning_rate": 1.
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
-
"grad_norm": 0.
|
| 280 |
-
"learning_rate": 1.
|
| 281 |
-
"loss": 0.
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
-
"epoch": 1.
|
| 286 |
-
"grad_norm": 0.
|
| 287 |
-
"learning_rate": 1.
|
| 288 |
-
"loss": 0.
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
-
"epoch": 1.
|
| 293 |
-
"grad_norm": 0.
|
| 294 |
-
"learning_rate": 1.
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 1.
|
| 300 |
-
"grad_norm": 0.
|
| 301 |
-
"learning_rate": 1.
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 1.
|
| 307 |
-
"grad_norm": 0.
|
| 308 |
-
"learning_rate": 1.
|
| 309 |
-
"loss": 0.
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"epoch": 1.
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
-
"learning_rate": 1.
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
-
"epoch": 1.
|
| 321 |
-
"grad_norm": 0.
|
| 322 |
-
"learning_rate": 1.
|
| 323 |
-
"loss": 0.
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"epoch": 1.
|
| 328 |
-
"grad_norm": 0.
|
| 329 |
-
"learning_rate": 1.
|
| 330 |
-
"loss": 0.
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 1.
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
-
"learning_rate": 1.
|
| 337 |
-
"loss": 0.
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 1.
|
| 342 |
-
"grad_norm": 0.
|
| 343 |
-
"learning_rate": 1.
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 1.
|
| 349 |
-
"grad_norm": 0.
|
| 350 |
-
"learning_rate": 1.
|
| 351 |
-
"loss": 0.
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch": 1.
|
| 356 |
-
"grad_norm": 0.
|
| 357 |
-
"learning_rate": 1.
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 1.
|
| 363 |
-
"grad_norm": 0.
|
| 364 |
-
"learning_rate": 1.
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 1.
|
| 370 |
-
"grad_norm": 0.
|
| 371 |
-
"learning_rate": 1.
|
| 372 |
-
"loss": 0.
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"epoch": 1.
|
| 377 |
-
"grad_norm": 0.
|
| 378 |
-
"learning_rate": 1.
|
| 379 |
-
"loss": 0.
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"epoch": 1.
|
| 384 |
-
"grad_norm": 0.
|
| 385 |
-
"learning_rate": 1.
|
| 386 |
-
"loss": 0.
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
-
"epoch": 1.
|
| 391 |
-
"grad_norm": 0.
|
| 392 |
-
"learning_rate": 1.
|
| 393 |
-
"loss": 0.
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
-
"epoch": 1.
|
| 398 |
-
"grad_norm": 0.
|
| 399 |
-
"learning_rate": 1.
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 1.
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
-
"learning_rate": 1.
|
| 407 |
-
"loss": 0.
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 1.
|
| 412 |
-
"grad_norm": 0.
|
| 413 |
-
"learning_rate": 1.
|
| 414 |
-
"loss": 0.
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
-
"epoch": 1.
|
| 419 |
-
"grad_norm": 0.
|
| 420 |
-
"learning_rate": 1.
|
| 421 |
-
"loss": 0.
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch": 1.
|
| 426 |
-
"grad_norm": 0.
|
| 427 |
-
"learning_rate": 1.
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch": 1.
|
| 433 |
-
"grad_norm": 0.
|
| 434 |
-
"learning_rate":
|
| 435 |
-
"loss": 0.
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
-
"epoch": 1.
|
| 440 |
-
"grad_norm": 0.
|
| 441 |
-
"learning_rate":
|
| 442 |
-
"loss": 0.
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch": 1.
|
| 447 |
-
"grad_norm": 0.
|
| 448 |
-
"learning_rate":
|
| 449 |
-
"loss": 0.
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch": 1.
|
| 454 |
-
"grad_norm": 0.
|
| 455 |
-
"learning_rate":
|
| 456 |
-
"loss": 0.
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch": 1.
|
| 461 |
-
"grad_norm": 0.
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss": 0.
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch": 1.
|
| 468 |
-
"grad_norm": 0.
|
| 469 |
-
"learning_rate":
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch": 1.
|
| 475 |
-
"grad_norm": 0.
|
| 476 |
-
"learning_rate":
|
| 477 |
-
"loss": 0.
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
-
"epoch": 1.
|
| 482 |
-
"grad_norm": 0.
|
| 483 |
-
"learning_rate":
|
| 484 |
-
"loss": 0.
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch": 1.
|
| 489 |
-
"grad_norm": 0.
|
| 490 |
-
"learning_rate":
|
| 491 |
-
"loss": 0.
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch": 1.
|
| 496 |
-
"grad_norm": 0.
|
| 497 |
-
"learning_rate":
|
| 498 |
-
"loss": 0.
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
-
"epoch": 1.
|
| 503 |
-
"grad_norm": 0.
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss": 0.
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch": 1.
|
| 510 |
-
"grad_norm": 0.
|
| 511 |
-
"learning_rate":
|
| 512 |
-
"loss": 0.
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"epoch": 1.
|
| 517 |
-
"grad_norm": 0.
|
| 518 |
-
"learning_rate":
|
| 519 |
-
"loss": 0.
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"epoch":
|
| 524 |
-
"grad_norm": 0.
|
| 525 |
-
"learning_rate":
|
| 526 |
-
"loss": 0.
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
-
"epoch":
|
| 531 |
-
"grad_norm": 0.
|
| 532 |
-
"learning_rate":
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch":
|
| 538 |
-
"grad_norm": 0.
|
| 539 |
-
"learning_rate":
|
| 540 |
-
"loss": 0.
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
-
"epoch":
|
| 545 |
-
"grad_norm": 0.
|
| 546 |
-
"learning_rate":
|
| 547 |
-
"loss": 0.
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
-
"epoch":
|
| 552 |
-
"grad_norm": 0.
|
| 553 |
-
"learning_rate":
|
| 554 |
-
"loss": 0.
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
-
"epoch": 2.
|
| 559 |
-
"grad_norm": 0.
|
| 560 |
-
"learning_rate":
|
| 561 |
-
"loss": 0.
|
| 562 |
"step": 390
|
| 563 |
},
|
| 564 |
{
|
| 565 |
-
"epoch": 2.
|
| 566 |
-
"grad_norm": 0.
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 395
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 2.
|
| 573 |
-
"grad_norm": 0.
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 400
|
| 577 |
},
|
| 578 |
{
|
| 579 |
-
"epoch": 2.
|
| 580 |
-
"grad_norm": 0.
|
| 581 |
-
"learning_rate":
|
| 582 |
-
"loss": 0.
|
| 583 |
"step": 405
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"epoch": 2.
|
| 587 |
-
"grad_norm": 0.
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 410
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"epoch": 2.
|
| 594 |
-
"grad_norm": 0.
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 415
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch": 2.
|
| 601 |
-
"grad_norm": 0.
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 420
|
| 605 |
},
|
| 606 |
{
|
| 607 |
-
"epoch": 2.
|
| 608 |
-
"grad_norm": 0.
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 425
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 2.
|
| 615 |
-
"grad_norm": 0.
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 430
|
| 619 |
},
|
| 620 |
{
|
| 621 |
-
"epoch": 2.
|
| 622 |
-
"grad_norm": 0.
|
| 623 |
-
"learning_rate":
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 435
|
| 626 |
},
|
| 627 |
{
|
| 628 |
-
"epoch": 2.
|
| 629 |
-
"grad_norm": 0.
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 440
|
| 633 |
},
|
| 634 |
{
|
| 635 |
-
"epoch": 2.
|
| 636 |
-
"grad_norm": 0.
|
| 637 |
-
"learning_rate":
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 445
|
| 640 |
},
|
| 641 |
{
|
| 642 |
-
"epoch": 2.
|
| 643 |
-
"grad_norm": 0.
|
| 644 |
-
"learning_rate":
|
| 645 |
-
"loss": 0.
|
| 646 |
"step": 450
|
| 647 |
},
|
| 648 |
{
|
| 649 |
-
"epoch": 2.
|
| 650 |
-
"grad_norm": 0.
|
| 651 |
-
"learning_rate":
|
| 652 |
-
"loss": 0.
|
| 653 |
"step": 455
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"epoch": 2.
|
| 657 |
-
"grad_norm": 0.
|
| 658 |
-
"learning_rate":
|
| 659 |
-
"loss": 0.
|
| 660 |
"step": 460
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch": 2.
|
| 664 |
-
"grad_norm": 0.
|
| 665 |
-
"learning_rate":
|
| 666 |
-
"loss": 0.
|
| 667 |
"step": 465
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch": 2.
|
| 671 |
-
"grad_norm": 0.
|
| 672 |
-
"learning_rate":
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 470
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch": 2.
|
| 678 |
-
"grad_norm": 0.
|
| 679 |
-
"learning_rate":
|
| 680 |
-
"loss": 0.
|
| 681 |
"step": 475
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch": 2.
|
| 685 |
-
"grad_norm": 0.
|
| 686 |
-
"learning_rate":
|
| 687 |
-
"loss": 0.
|
| 688 |
"step": 480
|
| 689 |
},
|
| 690 |
{
|
| 691 |
-
"epoch": 2.
|
| 692 |
-
"grad_norm": 0.
|
| 693 |
-
"learning_rate":
|
| 694 |
-
"loss": 0.
|
| 695 |
"step": 485
|
| 696 |
},
|
| 697 |
{
|
| 698 |
-
"epoch": 2.
|
| 699 |
-
"grad_norm": 0.
|
| 700 |
-
"learning_rate":
|
| 701 |
-
"loss": 0.
|
| 702 |
"step": 490
|
| 703 |
},
|
| 704 |
{
|
| 705 |
-
"epoch": 2.
|
| 706 |
-
"grad_norm": 0.
|
| 707 |
-
"learning_rate":
|
| 708 |
-
"loss": 0.
|
| 709 |
"step": 495
|
| 710 |
},
|
| 711 |
{
|
| 712 |
-
"epoch": 2.
|
| 713 |
-
"grad_norm": 0.
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 0.
|
| 716 |
"step": 500
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch": 2.
|
| 720 |
-
"grad_norm": 0.
|
| 721 |
-
"learning_rate":
|
| 722 |
-
"loss": 0.
|
| 723 |
"step": 505
|
| 724 |
},
|
| 725 |
{
|
| 726 |
-
"epoch": 2.
|
| 727 |
-
"grad_norm": 0.
|
| 728 |
-
"learning_rate":
|
| 729 |
-
"loss": 0.
|
| 730 |
"step": 510
|
| 731 |
},
|
| 732 |
{
|
| 733 |
-
"epoch": 2.
|
| 734 |
-
"grad_norm": 0.
|
| 735 |
-
"learning_rate":
|
| 736 |
-
"loss": 0.
|
| 737 |
"step": 515
|
| 738 |
},
|
| 739 |
{
|
| 740 |
-
"epoch": 2.
|
| 741 |
-
"grad_norm": 0.
|
| 742 |
-
"learning_rate":
|
| 743 |
-
"loss": 0.
|
| 744 |
"step": 520
|
| 745 |
},
|
| 746 |
{
|
| 747 |
-
"epoch": 2.
|
| 748 |
-
"grad_norm": 0.
|
| 749 |
-
"learning_rate":
|
| 750 |
-
"loss": 0.
|
| 751 |
"step": 525
|
| 752 |
},
|
| 753 |
{
|
| 754 |
-
"epoch": 2.
|
| 755 |
-
"grad_norm": 0.
|
| 756 |
-
"learning_rate":
|
| 757 |
-
"loss": 0.
|
| 758 |
"step": 530
|
| 759 |
},
|
| 760 |
{
|
| 761 |
-
"epoch": 2.
|
| 762 |
-
"grad_norm": 0.
|
| 763 |
-
"learning_rate":
|
| 764 |
-
"loss": 0.
|
| 765 |
"step": 535
|
| 766 |
},
|
| 767 |
{
|
| 768 |
-
"epoch": 2.
|
| 769 |
-
"grad_norm": 0.
|
| 770 |
-
"learning_rate":
|
| 771 |
-
"loss": 0.
|
| 772 |
"step": 540
|
| 773 |
},
|
| 774 |
-
{
|
| 775 |
-
"epoch": 2.823834196891192,
|
| 776 |
-
"grad_norm": 0.59375,
|
| 777 |
-
"learning_rate": 2.0942590842078503e-07,
|
| 778 |
-
"loss": 0.9117,
|
| 779 |
-
"step": 545
|
| 780 |
-
},
|
| 781 |
-
{
|
| 782 |
-
"epoch": 2.849740932642487,
|
| 783 |
-
"grad_norm": 0.6328125,
|
| 784 |
-
"learning_rate": 1.5250473693406486e-07,
|
| 785 |
-
"loss": 0.9154,
|
| 786 |
-
"step": 550
|
| 787 |
-
},
|
| 788 |
-
{
|
| 789 |
-
"epoch": 2.8756476683937824,
|
| 790 |
-
"grad_norm": 0.6328125,
|
| 791 |
-
"learning_rate": 1.0453426933830002e-07,
|
| 792 |
-
"loss": 0.942,
|
| 793 |
-
"step": 555
|
| 794 |
-
},
|
| 795 |
-
{
|
| 796 |
-
"epoch": 2.901554404145078,
|
| 797 |
-
"grad_norm": 0.609375,
|
| 798 |
-
"learning_rate": 6.555810752813308e-08,
|
| 799 |
-
"loss": 0.9441,
|
| 800 |
-
"step": 560
|
| 801 |
-
},
|
| 802 |
-
{
|
| 803 |
-
"epoch": 2.927461139896373,
|
| 804 |
-
"grad_norm": 0.59765625,
|
| 805 |
-
"learning_rate": 3.5611678185563106e-08,
|
| 806 |
-
"loss": 0.9353,
|
| 807 |
-
"step": 565
|
| 808 |
-
},
|
| 809 |
-
{
|
| 810 |
-
"epoch": 2.9533678756476682,
|
| 811 |
-
"grad_norm": 0.5859375,
|
| 812 |
-
"learning_rate": 1.4722200579497803e-08,
|
| 813 |
-
"loss": 0.9235,
|
| 814 |
-
"step": 570
|
| 815 |
-
},
|
| 816 |
-
{
|
| 817 |
-
"epoch": 2.9792746113989637,
|
| 818 |
-
"grad_norm": 0.59375,
|
| 819 |
-
"learning_rate": 2.9086618252893717e-09,
|
| 820 |
-
"loss": 0.9246,
|
| 821 |
-
"step": 575
|
| 822 |
-
},
|
| 823 |
{
|
| 824 |
"epoch": 3.0,
|
| 825 |
-
"step":
|
| 826 |
-
"total_flos": 2.
|
| 827 |
-
"train_loss": 1.
|
| 828 |
-
"train_runtime":
|
| 829 |
-
"train_samples_per_second": 38.
|
| 830 |
-
"train_steps_per_second": 2.
|
| 831 |
}
|
| 832 |
],
|
| 833 |
"logging_steps": 5,
|
| 834 |
-
"max_steps":
|
| 835 |
"num_input_tokens_seen": 0,
|
| 836 |
"num_train_epochs": 3,
|
| 837 |
"save_steps": 500,
|
|
@@ -847,7 +798,7 @@
|
|
| 847 |
"attributes": {}
|
| 848 |
}
|
| 849 |
},
|
| 850 |
-
"total_flos": 2.
|
| 851 |
"train_batch_size": 16,
|
| 852 |
"trial_name": null,
|
| 853 |
"trial_params": null
|
|
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 3.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 543,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.0055248618784530384,
|
| 13 |
+
"grad_norm": 7.4375,
|
| 14 |
+
"learning_rate": 3.6363636363636366e-07,
|
| 15 |
+
"loss": 2.4042,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.027624309392265192,
|
| 20 |
+
"grad_norm": 7.59375,
|
| 21 |
+
"learning_rate": 1.8181818181818183e-06,
|
| 22 |
+
"loss": 2.4209,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.055248618784530384,
|
| 27 |
+
"grad_norm": 7.1875,
|
| 28 |
+
"learning_rate": 3.6363636363636366e-06,
|
| 29 |
+
"loss": 2.4119,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.08287292817679558,
|
| 34 |
+
"grad_norm": 7.03125,
|
| 35 |
+
"learning_rate": 5.4545454545454545e-06,
|
| 36 |
+
"loss": 2.3657,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.11049723756906077,
|
| 41 |
+
"grad_norm": 7.28125,
|
| 42 |
+
"learning_rate": 7.272727272727273e-06,
|
| 43 |
+
"loss": 2.3123,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.13812154696132597,
|
| 48 |
+
"grad_norm": 5.65625,
|
| 49 |
+
"learning_rate": 9.090909090909091e-06,
|
| 50 |
+
"loss": 2.1773,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.16574585635359115,
|
| 55 |
+
"grad_norm": 3.84375,
|
| 56 |
+
"learning_rate": 1.0909090909090909e-05,
|
| 57 |
+
"loss": 2.0066,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.19337016574585636,
|
| 62 |
+
"grad_norm": 3.0,
|
| 63 |
+
"learning_rate": 1.2727272727272728e-05,
|
| 64 |
+
"loss": 1.9028,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.22099447513812154,
|
| 69 |
+
"grad_norm": 2.421875,
|
| 70 |
+
"learning_rate": 1.4545454545454546e-05,
|
| 71 |
+
"loss": 1.7204,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.24861878453038674,
|
| 76 |
+
"grad_norm": 2.109375,
|
| 77 |
+
"learning_rate": 1.6363636363636366e-05,
|
| 78 |
+
"loss": 1.5881,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.27624309392265195,
|
| 83 |
+
"grad_norm": 1.8125,
|
| 84 |
+
"learning_rate": 1.8181818181818182e-05,
|
| 85 |
+
"loss": 1.4278,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.30386740331491713,
|
| 90 |
+
"grad_norm": 1.71875,
|
| 91 |
+
"learning_rate": 2e-05,
|
| 92 |
+
"loss": 1.2964,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.3314917127071823,
|
| 97 |
+
"grad_norm": 1.2109375,
|
| 98 |
+
"learning_rate": 1.9994819965926346e-05,
|
| 99 |
+
"loss": 1.1979,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.35911602209944754,
|
| 104 |
"grad_norm": 1.0078125,
|
| 105 |
+
"learning_rate": 1.997928523025598e-05,
|
| 106 |
+
"loss": 1.1192,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.3867403314917127,
|
| 111 |
+
"grad_norm": 0.76171875,
|
| 112 |
+
"learning_rate": 1.9953411887080917e-05,
|
| 113 |
+
"loss": 1.0823,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.4143646408839779,
|
| 118 |
+
"grad_norm": 0.70703125,
|
| 119 |
+
"learning_rate": 1.9917226741361014e-05,
|
| 120 |
+
"loss": 1.0383,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.4419889502762431,
|
| 125 |
+
"grad_norm": 0.69921875,
|
| 126 |
+
"learning_rate": 1.987076728115383e-05,
|
| 127 |
+
"loss": 1.0168,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.4696132596685083,
|
| 132 |
+
"grad_norm": 0.68359375,
|
| 133 |
+
"learning_rate": 1.9814081638776743e-05,
|
| 134 |
+
"loss": 1.004,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.4972375690607735,
|
| 139 |
+
"grad_norm": 0.7109375,
|
| 140 |
+
"learning_rate": 1.9747228540941555e-05,
|
| 141 |
+
"loss": 0.993,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.5248618784530387,
|
| 146 |
+
"grad_norm": 0.765625,
|
| 147 |
+
"learning_rate": 1.9670277247913205e-05,
|
| 148 |
+
"loss": 0.9644,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.5524861878453039,
|
| 153 |
+
"grad_norm": 0.76953125,
|
| 154 |
+
"learning_rate": 1.958330748175568e-05,
|
| 155 |
+
"loss": 0.996,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.580110497237569,
|
| 160 |
+
"grad_norm": 0.734375,
|
| 161 |
+
"learning_rate": 1.948640934373939e-05,
|
| 162 |
+
"loss": 0.9704,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.6077348066298343,
|
| 167 |
+
"grad_norm": 0.76953125,
|
| 168 |
+
"learning_rate": 1.9379683220995657e-05,
|
| 169 |
+
"loss": 0.9572,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.6353591160220995,
|
| 174 |
+
"grad_norm": 0.86328125,
|
| 175 |
+
"learning_rate": 1.9263239682514953e-05,
|
| 176 |
+
"loss": 0.9553,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.6629834254143646,
|
| 181 |
+
"grad_norm": 0.75,
|
| 182 |
+
"learning_rate": 1.9137199364596673e-05,
|
| 183 |
+
"loss": 0.9454,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.6906077348066298,
|
| 188 |
+
"grad_norm": 0.75390625,
|
| 189 |
+
"learning_rate": 1.9001692845869113e-05,
|
| 190 |
+
"loss": 0.939,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
+
"epoch": 0.7182320441988951,
|
| 195 |
+
"grad_norm": 0.77734375,
|
| 196 |
+
"learning_rate": 1.8856860512009115e-05,
|
| 197 |
+
"loss": 0.9433,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
+
"epoch": 0.7458563535911602,
|
| 202 |
+
"grad_norm": 0.76953125,
|
| 203 |
+
"learning_rate": 1.8702852410301556e-05,
|
| 204 |
+
"loss": 0.9329,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
+
"epoch": 0.7734806629834254,
|
| 209 |
+
"grad_norm": 0.71484375,
|
| 210 |
+
"learning_rate": 1.853982809418932e-05,
|
| 211 |
+
"loss": 0.9416,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 0.8011049723756906,
|
| 216 |
+
"grad_norm": 0.7265625,
|
| 217 |
+
"learning_rate": 1.8367956457974872e-05,
|
| 218 |
+
"loss": 0.914,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.8287292817679558,
|
| 223 |
+
"grad_norm": 0.87890625,
|
| 224 |
+
"learning_rate": 1.8187415561844586e-05,
|
| 225 |
+
"loss": 0.9229,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
+
"epoch": 0.856353591160221,
|
| 230 |
+
"grad_norm": 0.828125,
|
| 231 |
+
"learning_rate": 1.7998392447397197e-05,
|
| 232 |
+
"loss": 0.9259,
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"epoch": 0.8839779005524862,
|
| 237 |
+
"grad_norm": 0.85546875,
|
| 238 |
+
"learning_rate": 1.7801082943867406e-05,
|
| 239 |
+
"loss": 0.9421,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.9116022099447514,
|
| 244 |
+
"grad_norm": 0.8046875,
|
| 245 |
+
"learning_rate": 1.7595691465245484e-05,
|
| 246 |
+
"loss": 0.9225,
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 0.9392265193370166,
|
| 251 |
+
"grad_norm": 0.7109375,
|
| 252 |
+
"learning_rate": 1.7382430798502977e-05,
|
| 253 |
+
"loss": 0.9066,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
+
"epoch": 0.9668508287292817,
|
| 258 |
+
"grad_norm": 0.83984375,
|
| 259 |
+
"learning_rate": 1.7161521883143936e-05,
|
| 260 |
+
"loss": 0.8903,
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 0.994475138121547,
|
| 265 |
+
"grad_norm": 0.875,
|
| 266 |
+
"learning_rate": 1.693319358231011e-05,
|
| 267 |
+
"loss": 0.9252,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
+
"epoch": 1.022099447513812,
|
| 272 |
+
"grad_norm": 0.8203125,
|
| 273 |
+
"learning_rate": 1.6697682445677158e-05,
|
| 274 |
+
"loss": 0.9035,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 1.0497237569060773,
|
| 279 |
+
"grad_norm": 0.69140625,
|
| 280 |
+
"learning_rate": 1.6455232464387587e-05,
|
| 281 |
+
"loss": 0.9036,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 1.0773480662983426,
|
| 286 |
+
"grad_norm": 0.71875,
|
| 287 |
+
"learning_rate": 1.6206094818274228e-05,
|
| 288 |
+
"loss": 0.8932,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 1.1049723756906078,
|
| 293 |
+
"grad_norm": 0.91796875,
|
| 294 |
+
"learning_rate": 1.595052761563627e-05,
|
| 295 |
+
"loss": 0.9065,
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 1.132596685082873,
|
| 300 |
+
"grad_norm": 0.859375,
|
| 301 |
+
"learning_rate": 1.5688795625837274e-05,
|
| 302 |
+
"loss": 0.8995,
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 1.160220994475138,
|
| 307 |
+
"grad_norm": 0.78125,
|
| 308 |
+
"learning_rate": 1.542117000500229e-05,
|
| 309 |
+
"loss": 0.8844,
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 1.1878453038674033,
|
| 314 |
+
"grad_norm": 0.765625,
|
| 315 |
+
"learning_rate": 1.5147928015098309e-05,
|
| 316 |
+
"loss": 0.8894,
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"epoch": 1.2154696132596685,
|
| 321 |
+
"grad_norm": 0.81640625,
|
| 322 |
+
"learning_rate": 1.4869352736688938e-05,
|
| 323 |
+
"loss": 0.894,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
+
"epoch": 1.2430939226519337,
|
| 328 |
+
"grad_norm": 0.76953125,
|
| 329 |
+
"learning_rate": 1.458573277566103e-05,
|
| 330 |
+
"loss": 0.9222,
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
+
"epoch": 1.270718232044199,
|
| 335 |
+
"grad_norm": 0.75390625,
|
| 336 |
+
"learning_rate": 1.4297361964227004e-05,
|
| 337 |
+
"loss": 0.9014,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 1.298342541436464,
|
| 342 |
+
"grad_norm": 0.8125,
|
| 343 |
+
"learning_rate": 1.4004539056512667e-05,
|
| 344 |
+
"loss": 0.9052,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 1.3259668508287292,
|
| 349 |
+
"grad_norm": 0.87109375,
|
| 350 |
+
"learning_rate": 1.3707567419045926e-05,
|
| 351 |
+
"loss": 0.894,
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
+
"epoch": 1.3535911602209945,
|
| 356 |
+
"grad_norm": 0.8125,
|
| 357 |
+
"learning_rate": 1.3406754716466978e-05,
|
| 358 |
+
"loss": 0.9045,
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 1.3812154696132597,
|
| 363 |
+
"grad_norm": 0.78515625,
|
| 364 |
+
"learning_rate": 1.3102412592785654e-05,
|
| 365 |
+
"loss": 0.8737,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 1.408839779005525,
|
| 370 |
+
"grad_norm": 0.7734375,
|
| 371 |
+
"learning_rate": 1.2794856348516095e-05,
|
| 372 |
+
"loss": 0.9029,
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 1.43646408839779,
|
| 377 |
+
"grad_norm": 0.83984375,
|
| 378 |
+
"learning_rate": 1.248440461402328e-05,
|
| 379 |
+
"loss": 0.8883,
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 1.4640883977900552,
|
| 384 |
+
"grad_norm": 0.96875,
|
| 385 |
+
"learning_rate": 1.2171379019419786e-05,
|
| 386 |
+
"loss": 0.8932,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
+
"epoch": 1.4917127071823204,
|
| 391 |
+
"grad_norm": 0.890625,
|
| 392 |
+
"learning_rate": 1.1856103861354809e-05,
|
| 393 |
+
"loss": 0.8917,
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
+
"epoch": 1.5193370165745856,
|
| 398 |
+
"grad_norm": 0.78515625,
|
| 399 |
+
"learning_rate": 1.153890576704062e-05,
|
| 400 |
+
"loss": 0.9033,
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
+
"epoch": 1.5469613259668509,
|
| 405 |
+
"grad_norm": 0.8046875,
|
| 406 |
+
"learning_rate": 1.1220113355864549e-05,
|
| 407 |
+
"loss": 0.8839,
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 1.5745856353591159,
|
| 412 |
+
"grad_norm": 0.8046875,
|
| 413 |
+
"learning_rate": 1.0900056898937055e-05,
|
| 414 |
+
"loss": 0.8887,
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
+
"epoch": 1.6022099447513813,
|
| 419 |
+
"grad_norm": 0.8046875,
|
| 420 |
+
"learning_rate": 1.0579067976928614e-05,
|
| 421 |
+
"loss": 0.8951,
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 1.6298342541436464,
|
| 426 |
+
"grad_norm": 0.78125,
|
| 427 |
+
"learning_rate": 1.0257479136549889e-05,
|
| 428 |
+
"loss": 0.8954,
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 1.6574585635359116,
|
| 433 |
+
"grad_norm": 0.8359375,
|
| 434 |
+
"learning_rate": 9.935623546031043e-06,
|
| 435 |
+
"loss": 0.9004,
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
+
"epoch": 1.6850828729281768,
|
| 440 |
+
"grad_norm": 0.85546875,
|
| 441 |
+
"learning_rate": 9.613834649957216e-06,
|
| 442 |
+
"loss": 0.9045,
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 1.7127071823204418,
|
| 447 |
+
"grad_norm": 0.765625,
|
| 448 |
+
"learning_rate": 9.292445823817647e-06,
|
| 449 |
+
"loss": 0.8737,
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"epoch": 1.7403314917127073,
|
| 454 |
+
"grad_norm": 0.8203125,
|
| 455 |
+
"learning_rate": 8.971790028626395e-06,
|
| 456 |
+
"loss": 0.8722,
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
+
"epoch": 1.7679558011049723,
|
| 461 |
+
"grad_norm": 0.921875,
|
| 462 |
+
"learning_rate": 8.652199465972462e-06,
|
| 463 |
+
"loss": 0.8995,
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 1.7955801104972375,
|
| 468 |
+
"grad_norm": 0.8203125,
|
| 469 |
+
"learning_rate": 8.334005233856681e-06,
|
| 470 |
+
"loss": 0.9114,
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 1.8232044198895028,
|
| 475 |
+
"grad_norm": 0.79296875,
|
| 476 |
+
"learning_rate": 8.017536983671929e-06,
|
| 477 |
+
"loss": 0.891,
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
+
"epoch": 1.850828729281768,
|
| 482 |
+
"grad_norm": 0.75,
|
| 483 |
+
"learning_rate": 7.703122578682047e-06,
|
| 484 |
+
"loss": 0.8875,
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
+
"epoch": 1.8784530386740332,
|
| 489 |
+
"grad_norm": 0.8125,
|
| 490 |
+
"learning_rate": 7.391087754353252e-06,
|
| 491 |
+
"loss": 0.8779,
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 1.9060773480662982,
|
| 496 |
+
"grad_norm": 0.76953125,
|
| 497 |
+
"learning_rate": 7.081755780889978e-06,
|
| 498 |
+
"loss": 0.885,
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 1.9337016574585635,
|
| 503 |
+
"grad_norm": 0.828125,
|
| 504 |
+
"learning_rate": 6.7754471283247594e-06,
|
| 505 |
+
"loss": 0.8875,
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
+
"epoch": 1.9613259668508287,
|
| 510 |
+
"grad_norm": 0.9140625,
|
| 511 |
+
"learning_rate": 6.472479134509052e-06,
|
| 512 |
+
"loss": 0.9037,
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"epoch": 1.988950276243094,
|
| 517 |
+
"grad_norm": 0.83203125,
|
| 518 |
+
"learning_rate": 6.173165676349103e-06,
|
| 519 |
+
"loss": 0.8817,
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"epoch": 2.016574585635359,
|
| 524 |
+
"grad_norm": 0.81640625,
|
| 525 |
+
"learning_rate": 5.8778168446273045e-06,
|
| 526 |
+
"loss": 0.8876,
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 2.044198895027624,
|
| 531 |
+
"grad_norm": 0.77734375,
|
| 532 |
+
"learning_rate": 5.586738622746042e-06,
|
| 533 |
+
"loss": 0.891,
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 2.0718232044198897,
|
| 538 |
+
"grad_norm": 0.80078125,
|
| 539 |
+
"learning_rate": 5.300232569726805e-06,
|
| 540 |
+
"loss": 0.8843,
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 2.0994475138121547,
|
| 545 |
+
"grad_norm": 0.9140625,
|
| 546 |
+
"learning_rate": 5.0185955077929774e-06,
|
| 547 |
+
"loss": 0.8696,
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
+
"epoch": 2.12707182320442,
|
| 552 |
+
"grad_norm": 0.7578125,
|
| 553 |
+
"learning_rate": 4.742119214860009e-06,
|
| 554 |
+
"loss": 0.8775,
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
+
"epoch": 2.154696132596685,
|
| 559 |
+
"grad_norm": 0.7578125,
|
| 560 |
+
"learning_rate": 4.471090122251496e-06,
|
| 561 |
+
"loss": 0.8797,
|
| 562 |
"step": 390
|
| 563 |
},
|
| 564 |
{
|
| 565 |
+
"epoch": 2.18232044198895,
|
| 566 |
+
"grad_norm": 0.78125,
|
| 567 |
+
"learning_rate": 4.205789017954364e-06,
|
| 568 |
+
"loss": 0.8832,
|
| 569 |
"step": 395
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 2.2099447513812156,
|
| 573 |
+
"grad_norm": 0.859375,
|
| 574 |
+
"learning_rate": 3.946490755720621e-06,
|
| 575 |
+
"loss": 0.884,
|
| 576 |
"step": 400
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 2.2375690607734806,
|
| 580 |
+
"grad_norm": 0.9453125,
|
| 581 |
+
"learning_rate": 3.6934639703169905e-06,
|
| 582 |
+
"loss": 0.8737,
|
| 583 |
"step": 405
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"epoch": 2.265193370165746,
|
| 587 |
+
"grad_norm": 0.86328125,
|
| 588 |
+
"learning_rate": 3.4469707992174607e-06,
|
| 589 |
+
"loss": 0.8981,
|
| 590 |
"step": 410
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"epoch": 2.292817679558011,
|
| 594 |
+
"grad_norm": 0.80078125,
|
| 595 |
+
"learning_rate": 3.207266611027069e-06,
|
| 596 |
+
"loss": 0.8736,
|
| 597 |
"step": 415
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 2.320441988950276,
|
| 601 |
+
"grad_norm": 0.78125,
|
| 602 |
+
"learning_rate": 2.97459974091831e-06,
|
| 603 |
+
"loss": 0.8757,
|
| 604 |
"step": 420
|
| 605 |
},
|
| 606 |
{
|
| 607 |
+
"epoch": 2.3480662983425415,
|
| 608 |
+
"grad_norm": 0.8203125,
|
| 609 |
+
"learning_rate": 2.7492112333541744e-06,
|
| 610 |
+
"loss": 0.902,
|
| 611 |
"step": 425
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 2.3756906077348066,
|
| 615 |
+
"grad_norm": 0.90625,
|
| 616 |
+
"learning_rate": 2.531334592364457e-06,
|
| 617 |
+
"loss": 0.8766,
|
| 618 |
"step": 430
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 2.403314917127072,
|
| 622 |
+
"grad_norm": 0.7734375,
|
| 623 |
+
"learning_rate": 2.3211955396340003e-06,
|
| 624 |
+
"loss": 0.8982,
|
| 625 |
"step": 435
|
| 626 |
},
|
| 627 |
{
|
| 628 |
+
"epoch": 2.430939226519337,
|
| 629 |
+
"grad_norm": 0.74609375,
|
| 630 |
+
"learning_rate": 2.1190117806534714e-06,
|
| 631 |
+
"loss": 0.8801,
|
| 632 |
"step": 440
|
| 633 |
},
|
| 634 |
{
|
| 635 |
+
"epoch": 2.458563535911602,
|
| 636 |
+
"grad_norm": 0.828125,
|
| 637 |
+
"learning_rate": 1.924992779174999e-06,
|
| 638 |
+
"loss": 0.8707,
|
| 639 |
"step": 445
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 2.4861878453038675,
|
| 643 |
+
"grad_norm": 0.859375,
|
| 644 |
+
"learning_rate": 1.7393395402063085e-06,
|
| 645 |
+
"loss": 0.8939,
|
| 646 |
"step": 450
|
| 647 |
},
|
| 648 |
{
|
| 649 |
+
"epoch": 2.5138121546961325,
|
| 650 |
+
"grad_norm": 0.859375,
|
| 651 |
+
"learning_rate": 1.5622444017681438e-06,
|
| 652 |
+
"loss": 0.8707,
|
| 653 |
"step": 455
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 2.541436464088398,
|
| 657 |
+
"grad_norm": 0.78515625,
|
| 658 |
+
"learning_rate": 1.3938908356307846e-06,
|
| 659 |
+
"loss": 0.8771,
|
| 660 |
"step": 460
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 2.569060773480663,
|
| 664 |
+
"grad_norm": 0.76171875,
|
| 665 |
+
"learning_rate": 1.2344532572360325e-06,
|
| 666 |
+
"loss": 0.857,
|
| 667 |
"step": 465
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 2.596685082872928,
|
| 671 |
+
"grad_norm": 0.83984375,
|
| 672 |
+
"learning_rate": 1.0840968450016276e-06,
|
| 673 |
+
"loss": 0.885,
|
| 674 |
"step": 470
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 2.6243093922651934,
|
| 678 |
+
"grad_norm": 0.78125,
|
| 679 |
+
"learning_rate": 9.42977369195286e-07,
|
| 680 |
+
"loss": 0.9007,
|
| 681 |
"step": 475
|
| 682 |
},
|
| 683 |
{
|
| 684 |
+
"epoch": 2.6519337016574585,
|
| 685 |
+
"grad_norm": 0.90234375,
|
| 686 |
+
"learning_rate": 8.112410305556307e-07,
|
| 687 |
+
"loss": 0.8988,
|
| 688 |
"step": 480
|
| 689 |
},
|
| 690 |
{
|
| 691 |
+
"epoch": 2.679558011049724,
|
| 692 |
+
"grad_norm": 0.78125,
|
| 693 |
+
"learning_rate": 6.890243088272453e-07,
|
| 694 |
+
"loss": 0.8702,
|
| 695 |
"step": 485
|
| 696 |
},
|
| 697 |
{
|
| 698 |
+
"epoch": 2.707182320441989,
|
| 699 |
+
"grad_norm": 0.875,
|
| 700 |
+
"learning_rate": 5.764538213667103e-07,
|
| 701 |
+
"loss": 0.8981,
|
| 702 |
"step": 490
|
| 703 |
},
|
| 704 |
{
|
| 705 |
+
"epoch": 2.734806629834254,
|
| 706 |
+
"grad_norm": 0.89453125,
|
| 707 |
+
"learning_rate": 4.73646191966175e-07,
|
| 708 |
+
"loss": 0.8912,
|
| 709 |
"step": 495
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 2.7624309392265194,
|
| 713 |
+
"grad_norm": 0.875,
|
| 714 |
+
"learning_rate": 3.8070793003030296e-07,
|
| 715 |
+
"loss": 0.8967,
|
| 716 |
"step": 500
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 2.7900552486187844,
|
| 720 |
+
"grad_norm": 0.828125,
|
| 721 |
+
"learning_rate": 2.9773532023180897e-07,
|
| 722 |
+
"loss": 0.8933,
|
| 723 |
"step": 505
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 2.81767955801105,
|
| 727 |
+
"grad_norm": 0.7578125,
|
| 728 |
+
"learning_rate": 2.248143227598809e-07,
|
| 729 |
+
"loss": 0.9039,
|
| 730 |
"step": 510
|
| 731 |
},
|
| 732 |
{
|
| 733 |
+
"epoch": 2.845303867403315,
|
| 734 |
+
"grad_norm": 0.7578125,
|
| 735 |
+
"learning_rate": 1.6202048426483652e-07,
|
| 736 |
+
"loss": 0.8864,
|
| 737 |
"step": 515
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 2.87292817679558,
|
| 741 |
+
"grad_norm": 0.8515625,
|
| 742 |
+
"learning_rate": 1.094188595912804e-07,
|
| 743 |
+
"loss": 0.8997,
|
| 744 |
"step": 520
|
| 745 |
},
|
| 746 |
{
|
| 747 |
+
"epoch": 2.9005524861878453,
|
| 748 |
+
"grad_norm": 0.87109375,
|
| 749 |
+
"learning_rate": 6.706394438083962e-08,
|
| 750 |
+
"loss": 0.8805,
|
| 751 |
"step": 525
|
| 752 |
},
|
| 753 |
{
|
| 754 |
+
"epoch": 2.9281767955801103,
|
| 755 |
+
"grad_norm": 0.7890625,
|
| 756 |
+
"learning_rate": 3.4999618614309784e-08,
|
| 757 |
+
"loss": 0.8959,
|
| 758 |
"step": 530
|
| 759 |
},
|
| 760 |
{
|
| 761 |
+
"epoch": 2.955801104972376,
|
| 762 |
+
"grad_norm": 0.7265625,
|
| 763 |
+
"learning_rate": 1.325910115169471e-08,
|
| 764 |
+
"loss": 0.8552,
|
| 765 |
"step": 535
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 2.983425414364641,
|
| 769 |
+
"grad_norm": 0.82421875,
|
| 770 |
+
"learning_rate": 1.8649153172423106e-09,
|
| 771 |
+
"loss": 0.8753,
|
| 772 |
"step": 540
|
| 773 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 774 |
{
|
| 775 |
"epoch": 3.0,
|
| 776 |
+
"step": 543,
|
| 777 |
+
"total_flos": 2.5856227500097536e+16,
|
| 778 |
+
"train_loss": 1.017678025019103,
|
| 779 |
+
"train_runtime": 224.2528,
|
| 780 |
+
"train_samples_per_second": 38.568,
|
| 781 |
+
"train_steps_per_second": 2.421
|
| 782 |
}
|
| 783 |
],
|
| 784 |
"logging_steps": 5,
|
| 785 |
+
"max_steps": 543,
|
| 786 |
"num_input_tokens_seen": 0,
|
| 787 |
"num_train_epochs": 3,
|
| 788 |
"save_steps": 500,
|
|
|
|
| 798 |
"attributes": {}
|
| 799 |
}
|
| 800 |
},
|
| 801 |
+
"total_flos": 2.5856227500097536e+16,
|
| 802 |
"train_batch_size": 16,
|
| 803 |
"trial_name": null,
|
| 804 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5560
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:96799ddaa0622d20b44b5fd1f13fd4ee51acc21d18c3376a5108d307ef6d1ed1
|
| 3 |
size 5560
|