🍻 cheers
Browse files- README.md +6 -5
- all_results.json +12 -12
- eval_results.json +8 -8
- runs/Mar27_22-47-32_ryanserver/events.out.tfevents.1711599059.ryanserver.28085.1 +3 -0
- train_results.json +5 -5
- trainer_state.json +430 -390
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 |
model-index:
|
| 7 |
- name: ryan_model3272024
|
|
@@ -13,12 +14,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# ryan_model3272024
|
| 15 |
|
| 16 |
-
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
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
-
- Ordinal Mae: 0.
|
| 20 |
-
- Ordinal Accuracy: 0.
|
| 21 |
-
- Na Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
model-index:
|
| 8 |
- name: ryan_model3272024
|
|
|
|
| 14 |
|
| 15 |
# ryan_model3272024
|
| 16 |
|
| 17 |
+
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 properties dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.2636
|
| 20 |
+
- Ordinal Mae: 0.5544
|
| 21 |
+
- Ordinal Accuracy: 0.5810
|
| 22 |
+
- Na Accuracy: 0.7915
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_na_accuracy": 0.
|
| 5 |
-
"eval_ordinal_accuracy": 0.
|
| 6 |
-
"eval_ordinal_mae": 0.
|
| 7 |
-
"eval_runtime":
|
| 8 |
-
"eval_samples_per_second":
|
| 9 |
-
"eval_steps_per_second": 3.
|
| 10 |
-
"train_loss": 0.
|
| 11 |
-
"train_runtime":
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second":
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.65,
|
| 3 |
+
"eval_loss": 0.26356959342956543,
|
| 4 |
+
"eval_na_accuracy": 0.7915058135986328,
|
| 5 |
+
"eval_ordinal_accuracy": 0.581045925617218,
|
| 6 |
+
"eval_ordinal_mae": 0.5543876886367798,
|
| 7 |
+
"eval_runtime": 160.2802,
|
| 8 |
+
"eval_samples_per_second": 24.825,
|
| 9 |
+
"eval_steps_per_second": 3.107,
|
| 10 |
+
"train_loss": 0.29669314997536794,
|
| 11 |
+
"train_runtime": 4786.838,
|
| 12 |
+
"train_samples_per_second": 28.807,
|
| 13 |
+
"train_steps_per_second": 1.801
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_na_accuracy": 0.
|
| 5 |
-
"eval_ordinal_accuracy": 0.
|
| 6 |
-
"eval_ordinal_mae": 0.
|
| 7 |
-
"eval_runtime":
|
| 8 |
-
"eval_samples_per_second":
|
| 9 |
-
"eval_steps_per_second": 3.
|
| 10 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.65,
|
| 3 |
+
"eval_loss": 0.26356959342956543,
|
| 4 |
+
"eval_na_accuracy": 0.7915058135986328,
|
| 5 |
+
"eval_ordinal_accuracy": 0.581045925617218,
|
| 6 |
+
"eval_ordinal_mae": 0.5543876886367798,
|
| 7 |
+
"eval_runtime": 160.2802,
|
| 8 |
+
"eval_samples_per_second": 24.825,
|
| 9 |
+
"eval_steps_per_second": 3.107
|
| 10 |
}
|
runs/Mar27_22-47-32_ryanserver/events.out.tfevents.1711599059.ryanserver.28085.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bd8e9d9adf68ac410dc4be5a1c6abfd8c0da0c9c412c736aab1296102c953046
|
| 3 |
+
size 529
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second":
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.65,
|
| 3 |
+
"train_loss": 0.29669314997536794,
|
| 4 |
+
"train_runtime": 4786.838,
|
| 5 |
+
"train_samples_per_second": 28.807,
|
| 6 |
+
"train_steps_per_second": 1.801
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,535 +1,575 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./ryan_model3272024/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"eval_steps":
|
| 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": 0.
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss": 0.
|
| 16 |
-
"step":
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm": 0.
|
| 21 |
-
"learning_rate": 0.
|
| 22 |
-
"loss": 0.
|
| 23 |
-
"step":
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm": 0.
|
| 28 |
-
"learning_rate": 0.
|
| 29 |
-
"loss": 0.
|
| 30 |
-
"step":
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm": 0.
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss": 0.
|
| 37 |
-
"step":
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"
|
| 42 |
-
"
|
| 43 |
-
"
|
| 44 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"
|
| 49 |
-
"
|
| 50 |
-
"
|
| 51 |
-
"
|
| 52 |
-
"eval_runtime": 31.9197,
|
| 53 |
-
"eval_samples_per_second": 9.399,
|
| 54 |
-
"eval_steps_per_second": 1.19,
|
| 55 |
-
"step": 25
|
| 56 |
},
|
| 57 |
{
|
| 58 |
-
"epoch": 0.
|
| 59 |
-
"grad_norm": 0.
|
| 60 |
-
"learning_rate": 0.
|
| 61 |
-
"loss": 0.
|
| 62 |
-
"step":
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch": 0.
|
| 66 |
-
"grad_norm": 0.
|
| 67 |
-
"learning_rate": 0.
|
| 68 |
-
"loss": 0.
|
| 69 |
-
"step":
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
-
"grad_norm": 0.
|
| 74 |
-
"learning_rate": 0.
|
| 75 |
-
"loss": 0.
|
| 76 |
-
"step":
|
| 77 |
},
|
| 78 |
{
|
| 79 |
-
"epoch": 0.
|
| 80 |
-
"
|
| 81 |
-
"
|
| 82 |
-
"
|
| 83 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch": 0.
|
| 87 |
-
"grad_norm":
|
| 88 |
-
"learning_rate": 0.
|
| 89 |
-
"loss": 0.
|
| 90 |
-
"step":
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch": 0.
|
| 94 |
-
"
|
| 95 |
-
"
|
| 96 |
-
"
|
| 97 |
-
"
|
| 98 |
-
"eval_runtime": 12.0769,
|
| 99 |
-
"eval_samples_per_second": 24.841,
|
| 100 |
-
"eval_steps_per_second": 3.147,
|
| 101 |
-
"step": 50
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch": 0.
|
| 105 |
-
"grad_norm": 0.
|
| 106 |
-
"learning_rate": 0.
|
| 107 |
-
"loss": 0.
|
| 108 |
-
"step":
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch": 0.
|
| 112 |
-
"grad_norm": 0.
|
| 113 |
-
"learning_rate": 0.
|
| 114 |
-
"loss": 0.
|
| 115 |
-
"step":
|
| 116 |
},
|
| 117 |
{
|
| 118 |
-
"epoch": 0.
|
| 119 |
-
"
|
| 120 |
-
"
|
| 121 |
-
"
|
| 122 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch": 0.
|
| 126 |
-
"grad_norm": 0.
|
| 127 |
-
"learning_rate": 0.
|
| 128 |
-
"loss": 0.
|
| 129 |
-
"step":
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"grad_norm":
|
| 134 |
-
"learning_rate": 0.
|
| 135 |
-
"loss": 0.
|
| 136 |
-
"step":
|
| 137 |
},
|
| 138 |
{
|
| 139 |
-
"epoch": 0.
|
| 140 |
-
"
|
| 141 |
-
"
|
| 142 |
-
"
|
| 143 |
-
"
|
| 144 |
-
"eval_runtime": 11.5094,
|
| 145 |
-
"eval_samples_per_second": 26.066,
|
| 146 |
-
"eval_steps_per_second": 3.302,
|
| 147 |
-
"step": 75
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch": 0.
|
| 151 |
-
"grad_norm":
|
| 152 |
-
"learning_rate": 0.
|
| 153 |
-
"loss": 0.
|
| 154 |
-
"step":
|
| 155 |
},
|
| 156 |
{
|
| 157 |
-
"epoch": 0.
|
| 158 |
-
"
|
| 159 |
-
"
|
| 160 |
-
"
|
| 161 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch": 0.
|
| 165 |
-
"grad_norm": 0.
|
| 166 |
-
"learning_rate": 0.
|
| 167 |
-
"loss": 0.
|
| 168 |
-
"step":
|
| 169 |
},
|
| 170 |
{
|
| 171 |
-
"epoch": 0.
|
| 172 |
-
"grad_norm":
|
| 173 |
-
"learning_rate": 0.
|
| 174 |
-
"loss": 0.
|
| 175 |
-
"step":
|
| 176 |
},
|
| 177 |
{
|
| 178 |
-
"epoch": 0.
|
| 179 |
-
"grad_norm":
|
| 180 |
-
"learning_rate": 0.
|
| 181 |
-
"loss": 0.
|
| 182 |
-
"step":
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch": 0.
|
| 186 |
-
"
|
| 187 |
-
"
|
| 188 |
-
"
|
| 189 |
-
"
|
| 190 |
-
"eval_runtime": 12.2664,
|
| 191 |
-
"eval_samples_per_second": 24.457,
|
| 192 |
-
"eval_steps_per_second": 3.098,
|
| 193 |
-
"step": 100
|
| 194 |
},
|
| 195 |
{
|
| 196 |
-
"epoch": 0.
|
| 197 |
-
"
|
| 198 |
-
"
|
| 199 |
-
"
|
| 200 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch": 0.
|
| 204 |
-
"grad_norm": 0.
|
| 205 |
-
"learning_rate": 0.
|
| 206 |
-
"loss": 0.
|
| 207 |
-
"step":
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch": 0.
|
| 211 |
-
"grad_norm": 0.
|
| 212 |
-
"learning_rate": 0.
|
| 213 |
-
"loss": 0.
|
| 214 |
-
"step":
|
| 215 |
},
|
| 216 |
{
|
| 217 |
-
"epoch": 0.
|
| 218 |
-
"grad_norm":
|
| 219 |
-
"learning_rate": 0.
|
| 220 |
-
"loss": 0.
|
| 221 |
-
"step":
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch": 0.
|
| 225 |
-
"grad_norm": 0.
|
| 226 |
-
"learning_rate": 0.
|
| 227 |
-
"loss": 0.
|
| 228 |
-
"step":
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch": 0.
|
| 232 |
-
"eval_loss": 0.
|
| 233 |
-
"eval_na_accuracy": 0.
|
| 234 |
-
"eval_ordinal_accuracy": 0.
|
| 235 |
-
"eval_ordinal_mae": 0.
|
| 236 |
-
"eval_runtime":
|
| 237 |
-
"eval_samples_per_second":
|
| 238 |
-
"eval_steps_per_second": 3.
|
| 239 |
-
"step":
|
| 240 |
},
|
| 241 |
{
|
| 242 |
-
"epoch": 0.
|
| 243 |
-
"grad_norm": 0.
|
| 244 |
-
"learning_rate": 0.
|
| 245 |
-
"loss": 0.
|
| 246 |
-
"step":
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch": 0.
|
| 250 |
-
"grad_norm":
|
| 251 |
-
"learning_rate": 0.
|
| 252 |
-
"loss": 0.
|
| 253 |
-
"step":
|
| 254 |
},
|
| 255 |
{
|
| 256 |
-
"epoch": 0.
|
| 257 |
-
"grad_norm": 0.
|
| 258 |
-
"learning_rate": 0.
|
| 259 |
-
"loss": 0.
|
| 260 |
-
"step":
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch": 0.
|
| 264 |
-
"grad_norm":
|
| 265 |
-
"learning_rate": 0.
|
| 266 |
-
"loss": 0.
|
| 267 |
-
"step":
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch": 0.
|
| 271 |
-
"
|
| 272 |
-
"
|
| 273 |
-
"
|
| 274 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 275 |
},
|
| 276 |
{
|
| 277 |
-
"epoch": 0.
|
| 278 |
-
"
|
| 279 |
-
"
|
| 280 |
-
"
|
| 281 |
-
"
|
| 282 |
-
"eval_runtime": 12.509,
|
| 283 |
-
"eval_samples_per_second": 23.983,
|
| 284 |
-
"eval_steps_per_second": 3.038,
|
| 285 |
-
"step": 150
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch": 0.
|
| 289 |
-
"grad_norm":
|
| 290 |
-
"learning_rate": 0.
|
| 291 |
-
"loss": 0.
|
| 292 |
-
"step":
|
| 293 |
},
|
| 294 |
{
|
| 295 |
-
"epoch": 0.
|
| 296 |
-
"grad_norm": 1.
|
| 297 |
-
"learning_rate": 0.
|
| 298 |
-
"loss": 0.
|
| 299 |
-
"step":
|
| 300 |
},
|
| 301 |
{
|
| 302 |
-
"epoch": 0.
|
| 303 |
-
"grad_norm": 0.
|
| 304 |
-
"learning_rate": 0.
|
| 305 |
-
"loss": 0.
|
| 306 |
-
"step":
|
| 307 |
},
|
| 308 |
{
|
| 309 |
-
"epoch": 0.
|
| 310 |
-
"
|
| 311 |
-
"
|
| 312 |
-
"
|
| 313 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 314 |
},
|
| 315 |
{
|
| 316 |
-
"epoch": 0.
|
| 317 |
-
"grad_norm": 0.
|
| 318 |
-
"learning_rate": 0.
|
| 319 |
-
"loss": 0.
|
| 320 |
-
"step":
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"epoch": 0.
|
| 324 |
-
"
|
| 325 |
-
"
|
| 326 |
-
"
|
| 327 |
-
"
|
| 328 |
-
"eval_runtime": 12.3807,
|
| 329 |
-
"eval_samples_per_second": 24.231,
|
| 330 |
-
"eval_steps_per_second": 3.069,
|
| 331 |
-
"step": 175
|
| 332 |
},
|
| 333 |
{
|
| 334 |
-
"epoch": 0.
|
| 335 |
-
"grad_norm":
|
| 336 |
-
"learning_rate": 0.
|
| 337 |
-
"loss": 0.
|
| 338 |
-
"step":
|
| 339 |
},
|
| 340 |
{
|
| 341 |
-
"epoch": 0.
|
| 342 |
-
"grad_norm":
|
| 343 |
-
"learning_rate": 0.
|
| 344 |
-
"loss": 0.
|
| 345 |
-
"step":
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch":
|
| 349 |
-
"
|
| 350 |
-
"
|
| 351 |
-
"
|
| 352 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 353 |
},
|
| 354 |
{
|
| 355 |
-
"epoch":
|
| 356 |
-
"grad_norm": 0.
|
| 357 |
-
"learning_rate": 0.
|
| 358 |
-
"loss": 0.
|
| 359 |
-
"step":
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch":
|
| 363 |
-
"grad_norm":
|
| 364 |
-
"learning_rate": 0.
|
| 365 |
-
"loss": 0.
|
| 366 |
-
"step":
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch":
|
| 370 |
-
"
|
| 371 |
-
"
|
| 372 |
-
"
|
| 373 |
-
"
|
| 374 |
-
"eval_runtime": 11.8333,
|
| 375 |
-
"eval_samples_per_second": 25.352,
|
| 376 |
-
"eval_steps_per_second": 3.211,
|
| 377 |
-
"step": 200
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch":
|
| 381 |
-
"grad_norm":
|
| 382 |
-
"learning_rate": 0.
|
| 383 |
-
"loss": 0.
|
| 384 |
-
"step":
|
| 385 |
},
|
| 386 |
{
|
| 387 |
-
"epoch":
|
| 388 |
-
"
|
| 389 |
-
"
|
| 390 |
-
"
|
| 391 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 392 |
},
|
| 393 |
{
|
| 394 |
-
"epoch":
|
| 395 |
-
"grad_norm": 0.
|
| 396 |
-
"learning_rate": 0.
|
| 397 |
-
"loss": 0.
|
| 398 |
-
"step":
|
| 399 |
},
|
| 400 |
{
|
| 401 |
-
"epoch":
|
| 402 |
-
"grad_norm":
|
| 403 |
-
"learning_rate": 0.
|
| 404 |
-
"loss": 0.
|
| 405 |
-
"step":
|
| 406 |
},
|
| 407 |
{
|
| 408 |
-
"epoch":
|
| 409 |
-
"grad_norm": 2.
|
| 410 |
-
"learning_rate": 0.
|
| 411 |
-
"loss": 0.
|
| 412 |
-
"step":
|
| 413 |
},
|
| 414 |
{
|
| 415 |
-
"epoch":
|
| 416 |
-
"
|
| 417 |
-
"
|
| 418 |
-
"
|
| 419 |
-
"
|
| 420 |
-
"eval_runtime": 10.8667,
|
| 421 |
-
"eval_samples_per_second": 27.607,
|
| 422 |
-
"eval_steps_per_second": 3.497,
|
| 423 |
-
"step": 225
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch":
|
| 427 |
-
"
|
| 428 |
-
"
|
| 429 |
-
"
|
| 430 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"epoch":
|
| 434 |
-
"grad_norm": 0.
|
| 435 |
-
"learning_rate": 0.
|
| 436 |
-
"loss": 0.
|
| 437 |
-
"step":
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch":
|
| 441 |
-
"grad_norm":
|
| 442 |
-
"learning_rate": 0.
|
| 443 |
-
"loss": 0.
|
| 444 |
-
"step":
|
| 445 |
},
|
| 446 |
{
|
| 447 |
-
"epoch":
|
| 448 |
-
"grad_norm": 0.
|
| 449 |
-
"learning_rate": 0.
|
| 450 |
-
"loss": 0.
|
| 451 |
-
"step":
|
| 452 |
},
|
| 453 |
{
|
| 454 |
-
"epoch":
|
| 455 |
-
"grad_norm": 0.
|
| 456 |
-
"learning_rate": 0.
|
| 457 |
-
"loss": 0.
|
| 458 |
-
"step":
|
| 459 |
},
|
| 460 |
{
|
| 461 |
-
"epoch":
|
| 462 |
-
"eval_loss": 0.
|
| 463 |
-
"eval_na_accuracy": 0.
|
| 464 |
-
"eval_ordinal_accuracy": 0.
|
| 465 |
-
"eval_ordinal_mae": 0.
|
| 466 |
-
"eval_runtime":
|
| 467 |
-
"eval_samples_per_second": 24.
|
| 468 |
-
"eval_steps_per_second": 3.
|
| 469 |
-
"step":
|
| 470 |
},
|
| 471 |
{
|
| 472 |
-
"epoch":
|
| 473 |
-
"grad_norm":
|
| 474 |
-
"learning_rate": 0.
|
| 475 |
-
"loss": 0.
|
| 476 |
-
"step":
|
| 477 |
},
|
| 478 |
{
|
| 479 |
-
"epoch":
|
| 480 |
-
"grad_norm": 0.
|
| 481 |
-
"learning_rate": 0.
|
| 482 |
-
"loss": 0.
|
| 483 |
-
"step":
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch":
|
| 487 |
-
"grad_norm": 1.
|
| 488 |
-
"learning_rate": 0.
|
| 489 |
-
"loss": 0.
|
| 490 |
-
"step":
|
| 491 |
},
|
| 492 |
{
|
| 493 |
-
"epoch":
|
| 494 |
-
"grad_norm": 0.
|
| 495 |
-
"learning_rate": 0.
|
| 496 |
-
"loss": 0.
|
| 497 |
-
"step":
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch":
|
| 501 |
-
"
|
| 502 |
-
"
|
| 503 |
-
"
|
| 504 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 505 |
},
|
| 506 |
{
|
| 507 |
-
"epoch":
|
| 508 |
-
"
|
| 509 |
-
"
|
| 510 |
-
"
|
| 511 |
-
"
|
| 512 |
-
"eval_runtime": 11.4536,
|
| 513 |
-
"eval_samples_per_second": 26.193,
|
| 514 |
-
"eval_steps_per_second": 3.318,
|
| 515 |
-
"step": 275
|
| 516 |
},
|
| 517 |
{
|
| 518 |
-
"epoch":
|
| 519 |
-
"
|
| 520 |
-
"
|
| 521 |
-
"
|
| 522 |
-
"
|
| 523 |
-
|
| 524 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 525 |
}
|
| 526 |
],
|
| 527 |
-
"logging_steps":
|
| 528 |
-
"max_steps":
|
| 529 |
"num_input_tokens_seen": 0,
|
| 530 |
"num_train_epochs": 4,
|
| 531 |
-
"save_steps":
|
| 532 |
-
"total_flos":
|
| 533 |
"train_batch_size": 16,
|
| 534 |
"trial_name": null,
|
| 535 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.26356959342956543,
|
| 3 |
+
"best_model_checkpoint": "./ryan_model3272024/checkpoint-1000",
|
| 4 |
+
"epoch": 0.6496519721577726,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1400,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 0.38699468970298767,
|
| 14 |
+
"learning_rate": 0.0001994199535962877,
|
| 15 |
+
"loss": 0.4038,
|
| 16 |
+
"step": 25
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.02,
|
| 20 |
+
"grad_norm": 0.6787680387496948,
|
| 21 |
+
"learning_rate": 0.00019883990719257543,
|
| 22 |
+
"loss": 0.4003,
|
| 23 |
+
"step": 50
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.03,
|
| 27 |
+
"grad_norm": 0.5743306279182434,
|
| 28 |
+
"learning_rate": 0.00019825986078886312,
|
| 29 |
+
"loss": 0.3591,
|
| 30 |
+
"step": 75
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.05,
|
| 34 |
+
"grad_norm": 0.41705068945884705,
|
| 35 |
+
"learning_rate": 0.00019767981438515082,
|
| 36 |
+
"loss": 0.3524,
|
| 37 |
+
"step": 100
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.05,
|
| 41 |
+
"eval_loss": 0.339992493391037,
|
| 42 |
+
"eval_na_accuracy": 0.7586872577667236,
|
| 43 |
+
"eval_ordinal_accuracy": 0.38746026158332825,
|
| 44 |
+
"eval_ordinal_mae": 0.8904515504837036,
|
| 45 |
+
"eval_runtime": 335.205,
|
| 46 |
+
"eval_samples_per_second": 11.87,
|
| 47 |
+
"eval_steps_per_second": 1.486,
|
| 48 |
+
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
+
"epoch": 0.06,
|
| 52 |
+
"grad_norm": 0.36200761795043945,
|
| 53 |
+
"learning_rate": 0.0001970997679814385,
|
| 54 |
+
"loss": 0.3071,
|
| 55 |
+
"step": 125
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
},
|
| 57 |
{
|
| 58 |
+
"epoch": 0.07,
|
| 59 |
+
"grad_norm": 0.24589791893959045,
|
| 60 |
+
"learning_rate": 0.00019651972157772623,
|
| 61 |
+
"loss": 0.3475,
|
| 62 |
+
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.08,
|
| 66 |
+
"grad_norm": 0.6089735627174377,
|
| 67 |
+
"learning_rate": 0.00019593967517401393,
|
| 68 |
+
"loss": 0.3072,
|
| 69 |
+
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.09,
|
| 73 |
+
"grad_norm": 0.5671761631965637,
|
| 74 |
+
"learning_rate": 0.00019535962877030162,
|
| 75 |
+
"loss": 0.2683,
|
| 76 |
+
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
+
"epoch": 0.09,
|
| 80 |
+
"eval_loss": 0.36712726950645447,
|
| 81 |
+
"eval_na_accuracy": 0.623552143573761,
|
| 82 |
+
"eval_ordinal_accuracy": 0.48916497826576233,
|
| 83 |
+
"eval_ordinal_mae": 0.7306416630744934,
|
| 84 |
+
"eval_runtime": 155.9343,
|
| 85 |
+
"eval_samples_per_second": 25.517,
|
| 86 |
+
"eval_steps_per_second": 3.194,
|
| 87 |
+
"step": 200
|
| 88 |
},
|
| 89 |
{
|
| 90 |
+
"epoch": 0.1,
|
| 91 |
+
"grad_norm": 1.2764167785644531,
|
| 92 |
+
"learning_rate": 0.00019477958236658932,
|
| 93 |
+
"loss": 0.2953,
|
| 94 |
+
"step": 225
|
| 95 |
},
|
| 96 |
{
|
| 97 |
+
"epoch": 0.12,
|
| 98 |
+
"grad_norm": 1.9076497554779053,
|
| 99 |
+
"learning_rate": 0.00019419953596287704,
|
| 100 |
+
"loss": 0.3382,
|
| 101 |
+
"step": 250
|
|
|
|
|
|
|
|
|
|
|
|
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 0.13,
|
| 105 |
+
"grad_norm": 0.2747127115726471,
|
| 106 |
+
"learning_rate": 0.00019361948955916474,
|
| 107 |
+
"loss": 0.2752,
|
| 108 |
+
"step": 275
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 0.14,
|
| 112 |
+
"grad_norm": 0.9448749423027039,
|
| 113 |
+
"learning_rate": 0.00019303944315545243,
|
| 114 |
+
"loss": 0.3314,
|
| 115 |
+
"step": 300
|
| 116 |
},
|
| 117 |
{
|
| 118 |
+
"epoch": 0.14,
|
| 119 |
+
"eval_loss": 0.3450469672679901,
|
| 120 |
+
"eval_na_accuracy": 0.6969112157821655,
|
| 121 |
+
"eval_ordinal_accuracy": 0.4013291001319885,
|
| 122 |
+
"eval_ordinal_mae": 0.8077224493026733,
|
| 123 |
+
"eval_runtime": 156.2328,
|
| 124 |
+
"eval_samples_per_second": 25.468,
|
| 125 |
+
"eval_steps_per_second": 3.188,
|
| 126 |
+
"step": 300
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"epoch": 0.15,
|
| 130 |
+
"grad_norm": 0.2589721083641052,
|
| 131 |
+
"learning_rate": 0.00019245939675174015,
|
| 132 |
+
"loss": 0.3486,
|
| 133 |
+
"step": 325
|
| 134 |
},
|
| 135 |
{
|
| 136 |
+
"epoch": 0.16,
|
| 137 |
+
"grad_norm": 0.44286003708839417,
|
| 138 |
+
"learning_rate": 0.00019187935034802785,
|
| 139 |
+
"loss": 0.3386,
|
| 140 |
+
"step": 350
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 0.17,
|
| 144 |
+
"grad_norm": 0.3215602934360504,
|
| 145 |
+
"learning_rate": 0.00019129930394431554,
|
| 146 |
+
"loss": 0.3056,
|
| 147 |
+
"step": 375
|
|
|
|
|
|
|
|
|
|
|
|
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 0.19,
|
| 151 |
+
"grad_norm": 0.9510051012039185,
|
| 152 |
+
"learning_rate": 0.00019071925754060324,
|
| 153 |
+
"loss": 0.2747,
|
| 154 |
+
"step": 400
|
| 155 |
},
|
| 156 |
{
|
| 157 |
+
"epoch": 0.19,
|
| 158 |
+
"eval_loss": 0.28132036328315735,
|
| 159 |
+
"eval_na_accuracy": 0.7895752787590027,
|
| 160 |
+
"eval_ordinal_accuracy": 0.5423288345336914,
|
| 161 |
+
"eval_ordinal_mae": 0.6105712056159973,
|
| 162 |
+
"eval_runtime": 155.1965,
|
| 163 |
+
"eval_samples_per_second": 25.638,
|
| 164 |
+
"eval_steps_per_second": 3.209,
|
| 165 |
+
"step": 400
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"epoch": 0.2,
|
| 169 |
+
"grad_norm": 0.5417093634605408,
|
| 170 |
+
"learning_rate": 0.00019013921113689096,
|
| 171 |
+
"loss": 0.2522,
|
| 172 |
+
"step": 425
|
| 173 |
},
|
| 174 |
{
|
| 175 |
+
"epoch": 0.21,
|
| 176 |
+
"grad_norm": 1.405881643295288,
|
| 177 |
+
"learning_rate": 0.00018955916473317868,
|
| 178 |
+
"loss": 0.3589,
|
| 179 |
+
"step": 450
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 0.22,
|
| 183 |
+
"grad_norm": 0.8319898843765259,
|
| 184 |
+
"learning_rate": 0.00018897911832946638,
|
| 185 |
+
"loss": 0.2991,
|
| 186 |
+
"step": 475
|
| 187 |
},
|
| 188 |
{
|
| 189 |
+
"epoch": 0.23,
|
| 190 |
+
"grad_norm": 1.9455621242523193,
|
| 191 |
+
"learning_rate": 0.00018839907192575407,
|
| 192 |
+
"loss": 0.3247,
|
| 193 |
+
"step": 500
|
|
|
|
|
|
|
|
|
|
|
|
|
| 194 |
},
|
| 195 |
{
|
| 196 |
+
"epoch": 0.23,
|
| 197 |
+
"eval_loss": 0.3143959045410156,
|
| 198 |
+
"eval_na_accuracy": 0.7104247212409973,
|
| 199 |
+
"eval_ordinal_accuracy": 0.4524703919887543,
|
| 200 |
+
"eval_ordinal_mae": 0.7256373763084412,
|
| 201 |
+
"eval_runtime": 157.1141,
|
| 202 |
+
"eval_samples_per_second": 25.326,
|
| 203 |
+
"eval_steps_per_second": 3.17,
|
| 204 |
+
"step": 500
|
| 205 |
},
|
| 206 |
{
|
| 207 |
+
"epoch": 0.24,
|
| 208 |
+
"grad_norm": 0.6339251399040222,
|
| 209 |
+
"learning_rate": 0.00018781902552204177,
|
| 210 |
+
"loss": 0.303,
|
| 211 |
+
"step": 525
|
| 212 |
},
|
| 213 |
{
|
| 214 |
+
"epoch": 0.26,
|
| 215 |
+
"grad_norm": 0.3713740408420563,
|
| 216 |
+
"learning_rate": 0.0001872389791183295,
|
| 217 |
+
"loss": 0.3035,
|
| 218 |
+
"step": 550
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 0.27,
|
| 222 |
+
"grad_norm": 0.7050974369049072,
|
| 223 |
+
"learning_rate": 0.00018665893271461718,
|
| 224 |
+
"loss": 0.2609,
|
| 225 |
+
"step": 575
|
| 226 |
},
|
| 227 |
{
|
| 228 |
+
"epoch": 0.28,
|
| 229 |
+
"grad_norm": 0.791477620601654,
|
| 230 |
+
"learning_rate": 0.00018607888631090488,
|
| 231 |
+
"loss": 0.3612,
|
| 232 |
+
"step": 600
|
| 233 |
},
|
| 234 |
{
|
| 235 |
+
"epoch": 0.28,
|
| 236 |
+
"eval_loss": 0.3074879050254822,
|
| 237 |
+
"eval_na_accuracy": 0.7586872577667236,
|
| 238 |
+
"eval_ordinal_accuracy": 0.4984108507633209,
|
| 239 |
+
"eval_ordinal_mae": 0.6415887475013733,
|
| 240 |
+
"eval_runtime": 154.2538,
|
| 241 |
+
"eval_samples_per_second": 25.795,
|
| 242 |
+
"eval_steps_per_second": 3.228,
|
| 243 |
+
"step": 600
|
| 244 |
},
|
| 245 |
{
|
| 246 |
+
"epoch": 0.29,
|
| 247 |
+
"grad_norm": 0.39196524024009705,
|
| 248 |
+
"learning_rate": 0.0001854988399071926,
|
| 249 |
+
"loss": 0.31,
|
| 250 |
+
"step": 625
|
| 251 |
},
|
| 252 |
{
|
| 253 |
+
"epoch": 0.3,
|
| 254 |
+
"grad_norm": 1.0753191709518433,
|
| 255 |
+
"learning_rate": 0.0001849187935034803,
|
| 256 |
+
"loss": 0.2722,
|
| 257 |
+
"step": 650
|
| 258 |
},
|
| 259 |
{
|
| 260 |
+
"epoch": 0.31,
|
| 261 |
+
"grad_norm": 0.8922611474990845,
|
| 262 |
+
"learning_rate": 0.000184338747099768,
|
| 263 |
+
"loss": 0.3132,
|
| 264 |
+
"step": 675
|
| 265 |
},
|
| 266 |
{
|
| 267 |
+
"epoch": 0.32,
|
| 268 |
+
"grad_norm": 0.6866246461868286,
|
| 269 |
+
"learning_rate": 0.0001837587006960557,
|
| 270 |
+
"loss": 0.3031,
|
| 271 |
+
"step": 700
|
| 272 |
},
|
| 273 |
{
|
| 274 |
+
"epoch": 0.32,
|
| 275 |
+
"eval_loss": 0.2784635126590729,
|
| 276 |
+
"eval_na_accuracy": 0.7895752787590027,
|
| 277 |
+
"eval_ordinal_accuracy": 0.5556197762489319,
|
| 278 |
+
"eval_ordinal_mae": 0.5720168352127075,
|
| 279 |
+
"eval_runtime": 154.421,
|
| 280 |
+
"eval_samples_per_second": 25.767,
|
| 281 |
+
"eval_steps_per_second": 3.225,
|
| 282 |
+
"step": 700
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.34,
|
| 286 |
+
"grad_norm": 1.713051676750183,
|
| 287 |
+
"learning_rate": 0.0001831786542923434,
|
| 288 |
+
"loss": 0.337,
|
| 289 |
+
"step": 725
|
|
|
|
|
|
|
|
|
|
|
|
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.35,
|
| 293 |
+
"grad_norm": 1.0872548818588257,
|
| 294 |
+
"learning_rate": 0.0001825986078886311,
|
| 295 |
+
"loss": 0.2918,
|
| 296 |
+
"step": 750
|
| 297 |
},
|
| 298 |
{
|
| 299 |
+
"epoch": 0.36,
|
| 300 |
+
"grad_norm": 1.5099256038665771,
|
| 301 |
+
"learning_rate": 0.0001820185614849188,
|
| 302 |
+
"loss": 0.2509,
|
| 303 |
+
"step": 775
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.37,
|
| 307 |
+
"grad_norm": 0.5774210691452026,
|
| 308 |
+
"learning_rate": 0.0001814385150812065,
|
| 309 |
+
"loss": 0.2866,
|
| 310 |
+
"step": 800
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"epoch": 0.37,
|
| 314 |
+
"eval_loss": 0.28780511021614075,
|
| 315 |
+
"eval_na_accuracy": 0.7335907220840454,
|
| 316 |
+
"eval_ordinal_accuracy": 0.5775787234306335,
|
| 317 |
+
"eval_ordinal_mae": 0.5347856879234314,
|
| 318 |
+
"eval_runtime": 154.6062,
|
| 319 |
+
"eval_samples_per_second": 25.736,
|
| 320 |
+
"eval_steps_per_second": 3.221,
|
| 321 |
+
"step": 800
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 0.38,
|
| 325 |
+
"grad_norm": 0.33059367537498474,
|
| 326 |
+
"learning_rate": 0.00018085846867749422,
|
| 327 |
+
"loss": 0.2626,
|
| 328 |
+
"step": 825
|
| 329 |
},
|
| 330 |
{
|
| 331 |
+
"epoch": 0.39,
|
| 332 |
+
"grad_norm": 1.45087730884552,
|
| 333 |
+
"learning_rate": 0.0001802784222737819,
|
| 334 |
+
"loss": 0.3485,
|
| 335 |
+
"step": 850
|
|
|
|
|
|
|
|
|
|
|
|
|
| 336 |
},
|
| 337 |
{
|
| 338 |
+
"epoch": 0.41,
|
| 339 |
+
"grad_norm": 1.195901870727539,
|
| 340 |
+
"learning_rate": 0.0001796983758700696,
|
| 341 |
+
"loss": 0.3007,
|
| 342 |
+
"step": 875
|
| 343 |
},
|
| 344 |
{
|
| 345 |
+
"epoch": 0.42,
|
| 346 |
+
"grad_norm": 0.26779890060424805,
|
| 347 |
+
"learning_rate": 0.00017911832946635733,
|
| 348 |
+
"loss": 0.2927,
|
| 349 |
+
"step": 900
|
| 350 |
},
|
| 351 |
{
|
| 352 |
+
"epoch": 0.42,
|
| 353 |
+
"eval_loss": 0.2688673734664917,
|
| 354 |
+
"eval_na_accuracy": 0.7972972989082336,
|
| 355 |
+
"eval_ordinal_accuracy": 0.5573533773422241,
|
| 356 |
+
"eval_ordinal_mae": 0.5855077505111694,
|
| 357 |
+
"eval_runtime": 154.5178,
|
| 358 |
+
"eval_samples_per_second": 25.751,
|
| 359 |
+
"eval_steps_per_second": 3.223,
|
| 360 |
+
"step": 900
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 0.43,
|
| 364 |
+
"grad_norm": 0.5635965466499329,
|
| 365 |
+
"learning_rate": 0.00017853828306264502,
|
| 366 |
+
"loss": 0.269,
|
| 367 |
+
"step": 925
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 0.44,
|
| 371 |
+
"grad_norm": 2.8135786056518555,
|
| 372 |
+
"learning_rate": 0.00017795823665893272,
|
| 373 |
+
"loss": 0.2677,
|
| 374 |
+
"step": 950
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 0.45,
|
| 378 |
+
"grad_norm": 0.49396631121635437,
|
| 379 |
+
"learning_rate": 0.0001773781902552204,
|
| 380 |
+
"loss": 0.3069,
|
| 381 |
+
"step": 975
|
|
|
|
|
|
|
|
|
|
|
|
|
| 382 |
},
|
| 383 |
{
|
| 384 |
+
"epoch": 0.46,
|
| 385 |
+
"grad_norm": 1.3267723321914673,
|
| 386 |
+
"learning_rate": 0.00017679814385150814,
|
| 387 |
+
"loss": 0.3003,
|
| 388 |
+
"step": 1000
|
| 389 |
},
|
| 390 |
{
|
| 391 |
+
"epoch": 0.46,
|
| 392 |
+
"eval_loss": 0.26356959342956543,
|
| 393 |
+
"eval_na_accuracy": 0.7915058135986328,
|
| 394 |
+
"eval_ordinal_accuracy": 0.581045925617218,
|
| 395 |
+
"eval_ordinal_mae": 0.5543876886367798,
|
| 396 |
+
"eval_runtime": 157.946,
|
| 397 |
+
"eval_samples_per_second": 25.192,
|
| 398 |
+
"eval_steps_per_second": 3.153,
|
| 399 |
+
"step": 1000
|
| 400 |
},
|
| 401 |
{
|
| 402 |
+
"epoch": 0.48,
|
| 403 |
+
"grad_norm": 0.9938157200813293,
|
| 404 |
+
"learning_rate": 0.00017621809744779583,
|
| 405 |
+
"loss": 0.2521,
|
| 406 |
+
"step": 1025
|
| 407 |
},
|
| 408 |
{
|
| 409 |
+
"epoch": 0.49,
|
| 410 |
+
"grad_norm": 0.45715010166168213,
|
| 411 |
+
"learning_rate": 0.00017563805104408353,
|
| 412 |
+
"loss": 0.2926,
|
| 413 |
+
"step": 1050
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 0.5,
|
| 417 |
+
"grad_norm": 2.9666409492492676,
|
| 418 |
+
"learning_rate": 0.00017505800464037122,
|
| 419 |
+
"loss": 0.2581,
|
| 420 |
+
"step": 1075
|
| 421 |
},
|
| 422 |
{
|
| 423 |
+
"epoch": 0.51,
|
| 424 |
+
"grad_norm": 2.5301055908203125,
|
| 425 |
+
"learning_rate": 0.00017447795823665894,
|
| 426 |
+
"loss": 0.2522,
|
| 427 |
+
"step": 1100
|
|
|
|
|
|
|
|
|
|
|
|
|
| 428 |
},
|
| 429 |
{
|
| 430 |
+
"epoch": 0.51,
|
| 431 |
+
"eval_loss": 0.3009192943572998,
|
| 432 |
+
"eval_na_accuracy": 0.8571428656578064,
|
| 433 |
+
"eval_ordinal_accuracy": 0.54435133934021,
|
| 434 |
+
"eval_ordinal_mae": 0.5650931596755981,
|
| 435 |
+
"eval_runtime": 159.1216,
|
| 436 |
+
"eval_samples_per_second": 25.006,
|
| 437 |
+
"eval_steps_per_second": 3.13,
|
| 438 |
+
"step": 1100
|
| 439 |
},
|
| 440 |
{
|
| 441 |
+
"epoch": 0.52,
|
| 442 |
+
"grad_norm": 0.8192782998085022,
|
| 443 |
+
"learning_rate": 0.00017389791183294664,
|
| 444 |
+
"loss": 0.3584,
|
| 445 |
+
"step": 1125
|
| 446 |
},
|
| 447 |
{
|
| 448 |
+
"epoch": 0.53,
|
| 449 |
+
"grad_norm": 2.0657265186309814,
|
| 450 |
+
"learning_rate": 0.00017331786542923433,
|
| 451 |
+
"loss": 0.2547,
|
| 452 |
+
"step": 1150
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 0.55,
|
| 456 |
+
"grad_norm": 0.5887840390205383,
|
| 457 |
+
"learning_rate": 0.00017273781902552203,
|
| 458 |
+
"loss": 0.2335,
|
| 459 |
+
"step": 1175
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 0.56,
|
| 463 |
+
"grad_norm": 0.8169906735420227,
|
| 464 |
+
"learning_rate": 0.00017215777262180975,
|
| 465 |
+
"loss": 0.262,
|
| 466 |
+
"step": 1200
|
| 467 |
},
|
| 468 |
{
|
| 469 |
+
"epoch": 0.56,
|
| 470 |
+
"eval_loss": 0.279022216796875,
|
| 471 |
+
"eval_na_accuracy": 0.8301158547401428,
|
| 472 |
+
"eval_ordinal_accuracy": 0.5801791548728943,
|
| 473 |
+
"eval_ordinal_mae": 0.5203233361244202,
|
| 474 |
+
"eval_runtime": 159.9167,
|
| 475 |
+
"eval_samples_per_second": 24.882,
|
| 476 |
+
"eval_steps_per_second": 3.114,
|
| 477 |
+
"step": 1200
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"epoch": 0.57,
|
| 481 |
+
"grad_norm": 2.5461835861206055,
|
| 482 |
+
"learning_rate": 0.00017157772621809744,
|
| 483 |
+
"loss": 0.2387,
|
| 484 |
+
"step": 1225
|
| 485 |
},
|
| 486 |
{
|
| 487 |
+
"epoch": 0.58,
|
| 488 |
+
"grad_norm": 0.7304142117500305,
|
| 489 |
+
"learning_rate": 0.00017099767981438517,
|
| 490 |
+
"loss": 0.2366,
|
| 491 |
+
"step": 1250
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 0.59,
|
| 495 |
+
"grad_norm": 1.3845186233520508,
|
| 496 |
+
"learning_rate": 0.00017041763341067286,
|
| 497 |
+
"loss": 0.2309,
|
| 498 |
+
"step": 1275
|
| 499 |
},
|
| 500 |
{
|
| 501 |
+
"epoch": 0.6,
|
| 502 |
+
"grad_norm": 0.5202885270118713,
|
| 503 |
+
"learning_rate": 0.00016983758700696058,
|
| 504 |
+
"loss": 0.2139,
|
| 505 |
+
"step": 1300
|
| 506 |
},
|
| 507 |
{
|
| 508 |
+
"epoch": 0.6,
|
| 509 |
+
"eval_loss": 0.2653418481349945,
|
| 510 |
+
"eval_na_accuracy": 0.7509652376174927,
|
| 511 |
+
"eval_ordinal_accuracy": 0.5492632389068604,
|
| 512 |
+
"eval_ordinal_mae": 0.562603771686554,
|
| 513 |
+
"eval_runtime": 158.9921,
|
| 514 |
+
"eval_samples_per_second": 25.026,
|
| 515 |
+
"eval_steps_per_second": 3.132,
|
| 516 |
+
"step": 1300
|
| 517 |
},
|
| 518 |
{
|
| 519 |
+
"epoch": 0.61,
|
| 520 |
+
"grad_norm": 1.6506483554840088,
|
| 521 |
+
"learning_rate": 0.00016925754060324828,
|
| 522 |
+
"loss": 0.3071,
|
| 523 |
+
"step": 1325
|
|
|
|
|
|
|
|
|
|
|
|
|
| 524 |
},
|
| 525 |
{
|
| 526 |
+
"epoch": 0.63,
|
| 527 |
+
"grad_norm": 0.5789369940757751,
|
| 528 |
+
"learning_rate": 0.00016867749419953597,
|
| 529 |
+
"loss": 0.2689,
|
| 530 |
+
"step": 1350
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.64,
|
| 534 |
+
"grad_norm": 0.5665389895439148,
|
| 535 |
+
"learning_rate": 0.00016809744779582367,
|
| 536 |
+
"loss": 0.2598,
|
| 537 |
+
"step": 1375
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.65,
|
| 541 |
+
"grad_norm": 0.6937847137451172,
|
| 542 |
+
"learning_rate": 0.0001675174013921114,
|
| 543 |
+
"loss": 0.2655,
|
| 544 |
+
"step": 1400
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.65,
|
| 548 |
+
"eval_loss": 0.2760397493839264,
|
| 549 |
+
"eval_na_accuracy": 0.7123551964759827,
|
| 550 |
+
"eval_ordinal_accuracy": 0.5426177382469177,
|
| 551 |
+
"eval_ordinal_mae": 0.6106911897659302,
|
| 552 |
+
"eval_runtime": 160.1635,
|
| 553 |
+
"eval_samples_per_second": 24.843,
|
| 554 |
+
"eval_steps_per_second": 3.109,
|
| 555 |
+
"step": 1400
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.65,
|
| 559 |
+
"step": 1400,
|
| 560 |
+
"total_flos": 1.735882797809664e+18,
|
| 561 |
+
"train_loss": 0.29669314997536794,
|
| 562 |
+
"train_runtime": 4786.838,
|
| 563 |
+
"train_samples_per_second": 28.807,
|
| 564 |
+
"train_steps_per_second": 1.801
|
| 565 |
}
|
| 566 |
],
|
| 567 |
+
"logging_steps": 25,
|
| 568 |
+
"max_steps": 8620,
|
| 569 |
"num_input_tokens_seen": 0,
|
| 570 |
"num_train_epochs": 4,
|
| 571 |
+
"save_steps": 100,
|
| 572 |
+
"total_flos": 1.735882797809664e+18,
|
| 573 |
"train_batch_size": 16,
|
| 574 |
"trial_name": null,
|
| 575 |
"trial_params": null
|