End of training
Browse files- README.md +1 -1
- all_results.json +6 -6
- test_results.json +6 -6
- trainer_state.json +149 -261
README.md
CHANGED
|
@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
- Accuracy: 0.9290
|
| 23 |
|
| 24 |
## Model description
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.2267
|
| 22 |
- Accuracy: 0.9290
|
| 23 |
|
| 24 |
## Model description
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 7.
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 7.101351351351352,
|
| 3 |
+
"eval_accuracy": 0.9290322580645162,
|
| 4 |
+
"eval_loss": 0.2266511768102646,
|
| 5 |
+
"eval_runtime": 13.6046,
|
| 6 |
+
"eval_samples_per_second": 11.393,
|
| 7 |
+
"eval_steps_per_second": 0.735
|
| 8 |
}
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 7.
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 7.101351351351352,
|
| 3 |
+
"eval_accuracy": 0.9290322580645162,
|
| 4 |
+
"eval_loss": 0.2266511768102646,
|
| 5 |
+
"eval_runtime": 13.6046,
|
| 6 |
+
"eval_samples_per_second": 11.393,
|
| 7 |
+
"eval_steps_per_second": 0.735
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,325 +1,213 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-
|
| 4 |
-
"epoch": 7.
|
| 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": 6.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch":
|
| 27 |
-
"grad_norm": 7.
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch":
|
| 34 |
-
"eval_accuracy": 0.
|
| 35 |
-
"eval_loss":
|
| 36 |
-
"eval_runtime":
|
| 37 |
-
"eval_samples_per_second":
|
| 38 |
-
"eval_steps_per_second":
|
| 39 |
"step": 38
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch":
|
| 43 |
-
"grad_norm":
|
| 44 |
-
"learning_rate": 4.
|
| 45 |
-
"loss":
|
| 46 |
"step": 40
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"epoch":
|
| 50 |
-
"grad_norm":
|
| 51 |
-
"learning_rate":
|
| 52 |
-
"loss":
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch":
|
| 57 |
-
"
|
| 58 |
-
"
|
| 59 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
"step": 60
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch":
|
| 64 |
-
"grad_norm":
|
| 65 |
-
"learning_rate":
|
| 66 |
-
"loss": 0.
|
| 67 |
"step": 70
|
| 68 |
},
|
| 69 |
{
|
| 70 |
-
"epoch":
|
| 71 |
-
"eval_accuracy": 0.
|
| 72 |
-
"eval_loss": 0.
|
| 73 |
-
"eval_runtime":
|
| 74 |
-
"eval_samples_per_second":
|
| 75 |
-
"eval_steps_per_second":
|
| 76 |
"step": 76
|
| 77 |
},
|
| 78 |
{
|
| 79 |
-
"epoch":
|
| 80 |
-
"grad_norm": 7.
|
| 81 |
-
"learning_rate":
|
| 82 |
-
"loss": 0.
|
| 83 |
"step": 80
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch":
|
| 87 |
-
"grad_norm":
|
| 88 |
-
"learning_rate":
|
| 89 |
-
"loss": 0.
|
| 90 |
"step": 90
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch":
|
| 94 |
-
"
|
| 95 |
-
"
|
| 96 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 97 |
"step": 100
|
| 98 |
},
|
| 99 |
{
|
| 100 |
-
"epoch":
|
| 101 |
-
"grad_norm":
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss": 0.
|
| 104 |
"step": 110
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch":
|
| 108 |
-
"eval_accuracy": 0.
|
| 109 |
-
"eval_loss": 0.
|
| 110 |
-
"eval_runtime":
|
| 111 |
-
"eval_samples_per_second":
|
| 112 |
-
"eval_steps_per_second":
|
| 113 |
"step": 114
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch":
|
| 117 |
-
"grad_norm": 4.
|
| 118 |
-
"learning_rate":
|
| 119 |
-
"loss": 0.
|
| 120 |
"step": 120
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch":
|
| 124 |
-
"grad_norm":
|
| 125 |
-
"learning_rate":
|
| 126 |
-
"loss": 0.
|
| 127 |
"step": 130
|
| 128 |
},
|
| 129 |
{
|
| 130 |
-
"epoch":
|
| 131 |
-
"grad_norm": 5.644320964813232,
|
| 132 |
-
"learning_rate": 2.962962962962963e-05,
|
| 133 |
-
"loss": 0.2827,
|
| 134 |
-
"step": 140
|
| 135 |
-
},
|
| 136 |
-
{
|
| 137 |
-
"epoch": 3.12,
|
| 138 |
-
"grad_norm": 11.105957984924316,
|
| 139 |
-
"learning_rate": 2.777777777777778e-05,
|
| 140 |
-
"loss": 0.2102,
|
| 141 |
-
"step": 150
|
| 142 |
-
},
|
| 143 |
-
{
|
| 144 |
-
"epoch": 3.1266666666666665,
|
| 145 |
-
"eval_accuracy": 0.8857142857142857,
|
| 146 |
-
"eval_loss": 0.33364787697792053,
|
| 147 |
-
"eval_runtime": 6.7638,
|
| 148 |
-
"eval_samples_per_second": 10.349,
|
| 149 |
-
"eval_steps_per_second": 1.331,
|
| 150 |
-
"step": 152
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 4.026666666666666,
|
| 154 |
-
"grad_norm": 3.0925567150115967,
|
| 155 |
-
"learning_rate": 2.5925925925925925e-05,
|
| 156 |
-
"loss": 0.3526,
|
| 157 |
-
"step": 160
|
| 158 |
-
},
|
| 159 |
-
{
|
| 160 |
-
"epoch": 4.06,
|
| 161 |
-
"grad_norm": 2.3389172554016113,
|
| 162 |
-
"learning_rate": 2.4074074074074074e-05,
|
| 163 |
-
"loss": 0.3044,
|
| 164 |
-
"step": 170
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 4.093333333333334,
|
| 168 |
-
"grad_norm": 15.626021385192871,
|
| 169 |
-
"learning_rate": 2.2222222222222223e-05,
|
| 170 |
-
"loss": 0.1253,
|
| 171 |
-
"step": 180
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 4.126666666666667,
|
| 175 |
-
"grad_norm": 26.906152725219727,
|
| 176 |
-
"learning_rate": 2.037037037037037e-05,
|
| 177 |
-
"loss": 0.0994,
|
| 178 |
-
"step": 190
|
| 179 |
-
},
|
| 180 |
-
{
|
| 181 |
-
"epoch": 4.126666666666667,
|
| 182 |
-
"eval_accuracy": 0.9571428571428572,
|
| 183 |
-
"eval_loss": 0.13536687195301056,
|
| 184 |
-
"eval_runtime": 7.4727,
|
| 185 |
-
"eval_samples_per_second": 9.367,
|
| 186 |
-
"eval_steps_per_second": 1.204,
|
| 187 |
-
"step": 190
|
| 188 |
-
},
|
| 189 |
-
{
|
| 190 |
-
"epoch": 5.033333333333333,
|
| 191 |
-
"grad_norm": 0.45196783542633057,
|
| 192 |
-
"learning_rate": 1.8518518518518518e-05,
|
| 193 |
-
"loss": 0.1249,
|
| 194 |
-
"step": 200
|
| 195 |
-
},
|
| 196 |
-
{
|
| 197 |
-
"epoch": 5.066666666666666,
|
| 198 |
-
"grad_norm": 1.2230411767959595,
|
| 199 |
-
"learning_rate": 1.6666666666666667e-05,
|
| 200 |
-
"loss": 0.0302,
|
| 201 |
-
"step": 210
|
| 202 |
-
},
|
| 203 |
-
{
|
| 204 |
-
"epoch": 5.1,
|
| 205 |
-
"grad_norm": 0.2508547604084015,
|
| 206 |
-
"learning_rate": 1.4814814814814815e-05,
|
| 207 |
-
"loss": 0.1416,
|
| 208 |
-
"step": 220
|
| 209 |
-
},
|
| 210 |
-
{
|
| 211 |
-
"epoch": 5.126666666666667,
|
| 212 |
"eval_accuracy": 0.9,
|
| 213 |
-
"eval_loss": 0.
|
| 214 |
-
"eval_runtime":
|
| 215 |
-
"eval_samples_per_second":
|
| 216 |
-
"eval_steps_per_second":
|
| 217 |
-
"step":
|
| 218 |
-
},
|
| 219 |
-
{
|
| 220 |
-
"epoch": 6.006666666666667,
|
| 221 |
-
"grad_norm": 0.1509472280740738,
|
| 222 |
-
"learning_rate": 1.2962962962962962e-05,
|
| 223 |
-
"loss": 0.056,
|
| 224 |
-
"step": 230
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch":
|
| 228 |
-
"grad_norm":
|
| 229 |
-
"learning_rate":
|
| 230 |
-
"loss": 0.
|
| 231 |
-
"step":
|
| 232 |
-
},
|
| 233 |
-
{
|
| 234 |
-
"epoch": 6.073333333333333,
|
| 235 |
-
"grad_norm": 0.16968391835689545,
|
| 236 |
-
"learning_rate": 9.259259259259259e-06,
|
| 237 |
-
"loss": 0.0488,
|
| 238 |
-
"step": 250
|
| 239 |
-
},
|
| 240 |
-
{
|
| 241 |
-
"epoch": 6.1066666666666665,
|
| 242 |
-
"grad_norm": 0.1539120376110077,
|
| 243 |
-
"learning_rate": 7.4074074074074075e-06,
|
| 244 |
-
"loss": 0.0301,
|
| 245 |
-
"step": 260
|
| 246 |
-
},
|
| 247 |
-
{
|
| 248 |
-
"epoch": 6.126666666666667,
|
| 249 |
-
"eval_accuracy": 0.9428571428571428,
|
| 250 |
-
"eval_loss": 0.17422690987586975,
|
| 251 |
-
"eval_runtime": 8.163,
|
| 252 |
-
"eval_samples_per_second": 8.575,
|
| 253 |
-
"eval_steps_per_second": 1.103,
|
| 254 |
-
"step": 266
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 7.013333333333334,
|
| 258 |
-
"grad_norm": 0.12634336948394775,
|
| 259 |
-
"learning_rate": 5.555555555555556e-06,
|
| 260 |
-
"loss": 0.0144,
|
| 261 |
-
"step": 270
|
| 262 |
-
},
|
| 263 |
-
{
|
| 264 |
-
"epoch": 7.046666666666667,
|
| 265 |
-
"grad_norm": 0.1565423160791397,
|
| 266 |
-
"learning_rate": 3.7037037037037037e-06,
|
| 267 |
-
"loss": 0.0141,
|
| 268 |
-
"step": 280
|
| 269 |
-
},
|
| 270 |
-
{
|
| 271 |
-
"epoch": 7.08,
|
| 272 |
-
"grad_norm": 0.09596653282642365,
|
| 273 |
-
"learning_rate": 1.8518518518518519e-06,
|
| 274 |
-
"loss": 0.0144,
|
| 275 |
-
"step": 290
|
| 276 |
-
},
|
| 277 |
-
{
|
| 278 |
-
"epoch": 7.113333333333333,
|
| 279 |
-
"grad_norm": 0.09075827151536942,
|
| 280 |
-
"learning_rate": 0.0,
|
| 281 |
-
"loss": 0.023,
|
| 282 |
-
"step": 300
|
| 283 |
-
},
|
| 284 |
-
{
|
| 285 |
-
"epoch": 7.113333333333333,
|
| 286 |
-
"eval_accuracy": 0.9714285714285714,
|
| 287 |
-
"eval_loss": 0.0840597078204155,
|
| 288 |
-
"eval_runtime": 7.7829,
|
| 289 |
-
"eval_samples_per_second": 8.994,
|
| 290 |
-
"eval_steps_per_second": 1.156,
|
| 291 |
-
"step": 300
|
| 292 |
-
},
|
| 293 |
-
{
|
| 294 |
-
"epoch": 7.113333333333333,
|
| 295 |
-
"step": 300,
|
| 296 |
-
"total_flos": 2.955876078527447e+18,
|
| 297 |
-
"train_loss": 0.5999104872345924,
|
| 298 |
-
"train_runtime": 483.0932,
|
| 299 |
-
"train_samples_per_second": 4.968,
|
| 300 |
-
"train_steps_per_second": 0.621
|
| 301 |
-
},
|
| 302 |
-
{
|
| 303 |
-
"epoch": 7.113333333333333,
|
| 304 |
-
"eval_accuracy": 0.9548387096774194,
|
| 305 |
-
"eval_loss": 0.15970781445503235,
|
| 306 |
-
"eval_runtime": 18.7889,
|
| 307 |
-
"eval_samples_per_second": 8.25,
|
| 308 |
-
"eval_steps_per_second": 1.064,
|
| 309 |
-
"step": 300
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch": 7.
|
| 313 |
-
"eval_accuracy": 0.
|
| 314 |
-
"eval_loss": 0.
|
| 315 |
-
"eval_runtime":
|
| 316 |
-
"eval_samples_per_second":
|
| 317 |
-
"eval_steps_per_second":
|
| 318 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 319 |
}
|
| 320 |
],
|
| 321 |
"logging_steps": 10,
|
| 322 |
-
"max_steps":
|
| 323 |
"num_input_tokens_seen": 0,
|
| 324 |
"num_train_epochs": 9223372036854775807,
|
| 325 |
"save_steps": 500,
|
|
@@ -335,8 +223,8 @@
|
|
| 335 |
"attributes": {}
|
| 336 |
}
|
| 337 |
},
|
| 338 |
-
"total_flos": 2.
|
| 339 |
-
"train_batch_size":
|
| 340 |
"trial_name": null,
|
| 341 |
"trial_params": null
|
| 342 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9142857142857143,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-148",
|
| 4 |
+
"epoch": 7.101351351351352,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 148,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.06756756756756757,
|
| 13 |
+
"grad_norm": 6.34252405166626,
|
| 14 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 15 |
+
"loss": 2.0466,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.12837837837837837,
|
| 20 |
+
"eval_accuracy": 0.6142857142857143,
|
| 21 |
+
"eval_loss": 1.6348506212234497,
|
| 22 |
+
"eval_runtime": 6.1663,
|
| 23 |
+
"eval_samples_per_second": 11.352,
|
| 24 |
+
"eval_steps_per_second": 0.811,
|
| 25 |
+
"step": 19
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.0067567567567568,
|
| 29 |
+
"grad_norm": 7.780186176300049,
|
| 30 |
+
"learning_rate": 4.81203007518797e-05,
|
| 31 |
+
"loss": 1.8425,
|
| 32 |
"step": 20
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 1.0743243243243243,
|
| 36 |
+
"grad_norm": 7.907790184020996,
|
| 37 |
+
"learning_rate": 4.43609022556391e-05,
|
| 38 |
+
"loss": 1.348,
|
| 39 |
"step": 30
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 1.1283783783783783,
|
| 43 |
+
"eval_accuracy": 0.8428571428571429,
|
| 44 |
+
"eval_loss": 0.804053783416748,
|
| 45 |
+
"eval_runtime": 6.6922,
|
| 46 |
+
"eval_samples_per_second": 10.46,
|
| 47 |
+
"eval_steps_per_second": 0.747,
|
| 48 |
"step": 38
|
| 49 |
},
|
| 50 |
{
|
| 51 |
+
"epoch": 2.0135135135135136,
|
| 52 |
+
"grad_norm": 5.862756729125977,
|
| 53 |
+
"learning_rate": 4.0601503759398494e-05,
|
| 54 |
+
"loss": 0.8369,
|
| 55 |
"step": 40
|
| 56 |
},
|
| 57 |
{
|
| 58 |
+
"epoch": 2.081081081081081,
|
| 59 |
+
"grad_norm": 5.44201135635376,
|
| 60 |
+
"learning_rate": 3.6842105263157895e-05,
|
| 61 |
+
"loss": 0.6208,
|
| 62 |
"step": 50
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 2.1283783783783785,
|
| 66 |
+
"eval_accuracy": 0.7285714285714285,
|
| 67 |
+
"eval_loss": 0.7582912445068359,
|
| 68 |
+
"eval_runtime": 6.1646,
|
| 69 |
+
"eval_samples_per_second": 11.355,
|
| 70 |
+
"eval_steps_per_second": 0.811,
|
| 71 |
+
"step": 57
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 3.02027027027027,
|
| 75 |
+
"grad_norm": 12.206607818603516,
|
| 76 |
+
"learning_rate": 3.3082706766917295e-05,
|
| 77 |
+
"loss": 0.5199,
|
| 78 |
"step": 60
|
| 79 |
},
|
| 80 |
{
|
| 81 |
+
"epoch": 3.0878378378378377,
|
| 82 |
+
"grad_norm": 5.501271724700928,
|
| 83 |
+
"learning_rate": 2.9323308270676693e-05,
|
| 84 |
+
"loss": 0.332,
|
| 85 |
"step": 70
|
| 86 |
},
|
| 87 |
{
|
| 88 |
+
"epoch": 3.1283783783783785,
|
| 89 |
+
"eval_accuracy": 0.8285714285714286,
|
| 90 |
+
"eval_loss": 0.45565441250801086,
|
| 91 |
+
"eval_runtime": 6.0418,
|
| 92 |
+
"eval_samples_per_second": 11.586,
|
| 93 |
+
"eval_steps_per_second": 0.828,
|
| 94 |
"step": 76
|
| 95 |
},
|
| 96 |
{
|
| 97 |
+
"epoch": 4.027027027027027,
|
| 98 |
+
"grad_norm": 7.0609965324401855,
|
| 99 |
+
"learning_rate": 2.556390977443609e-05,
|
| 100 |
+
"loss": 0.2761,
|
| 101 |
"step": 80
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 4.094594594594595,
|
| 105 |
+
"grad_norm": 3.5807061195373535,
|
| 106 |
+
"learning_rate": 2.1804511278195487e-05,
|
| 107 |
+
"loss": 0.2229,
|
| 108 |
"step": 90
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 4.128378378378378,
|
| 112 |
+
"eval_accuracy": 0.8857142857142857,
|
| 113 |
+
"eval_loss": 0.3132798373699188,
|
| 114 |
+
"eval_runtime": 6.754,
|
| 115 |
+
"eval_samples_per_second": 10.364,
|
| 116 |
+
"eval_steps_per_second": 0.74,
|
| 117 |
+
"step": 95
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 5.033783783783784,
|
| 121 |
+
"grad_norm": 1.8226597309112549,
|
| 122 |
+
"learning_rate": 1.8045112781954888e-05,
|
| 123 |
+
"loss": 0.2292,
|
| 124 |
"step": 100
|
| 125 |
},
|
| 126 |
{
|
| 127 |
+
"epoch": 5.101351351351352,
|
| 128 |
+
"grad_norm": 6.401218414306641,
|
| 129 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 130 |
+
"loss": 0.1479,
|
| 131 |
"step": 110
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 5.128378378378378,
|
| 135 |
+
"eval_accuracy": 0.9,
|
| 136 |
+
"eval_loss": 0.2872016131877899,
|
| 137 |
+
"eval_runtime": 5.5369,
|
| 138 |
+
"eval_samples_per_second": 12.643,
|
| 139 |
+
"eval_steps_per_second": 0.903,
|
| 140 |
"step": 114
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 6.04054054054054,
|
| 144 |
+
"grad_norm": 4.436432361602783,
|
| 145 |
+
"learning_rate": 1.0526315789473684e-05,
|
| 146 |
+
"loss": 0.0848,
|
| 147 |
"step": 120
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 6.108108108108108,
|
| 151 |
+
"grad_norm": 1.9256197214126587,
|
| 152 |
+
"learning_rate": 6.766917293233083e-06,
|
| 153 |
+
"loss": 0.0761,
|
| 154 |
"step": 130
|
| 155 |
},
|
| 156 |
{
|
| 157 |
+
"epoch": 6.128378378378378,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 158 |
"eval_accuracy": 0.9,
|
| 159 |
+
"eval_loss": 0.28876611590385437,
|
| 160 |
+
"eval_runtime": 6.7507,
|
| 161 |
+
"eval_samples_per_second": 10.369,
|
| 162 |
+
"eval_steps_per_second": 0.741,
|
| 163 |
+
"step": 133
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 7.047297297297297,
|
| 167 |
+
"grad_norm": 7.098940372467041,
|
| 168 |
+
"learning_rate": 3.007518796992481e-06,
|
| 169 |
+
"loss": 0.0696,
|
| 170 |
+
"step": 140
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 7.101351351351352,
|
| 174 |
+
"eval_accuracy": 0.9142857142857143,
|
| 175 |
+
"eval_loss": 0.26643699407577515,
|
| 176 |
+
"eval_runtime": 6.7738,
|
| 177 |
+
"eval_samples_per_second": 10.334,
|
| 178 |
+
"eval_steps_per_second": 0.738,
|
| 179 |
+
"step": 148
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 7.101351351351352,
|
| 183 |
+
"step": 148,
|
| 184 |
+
"total_flos": 2.915999166844109e+18,
|
| 185 |
+
"train_loss": 0.5872005439690642,
|
| 186 |
+
"train_runtime": 676.4093,
|
| 187 |
+
"train_samples_per_second": 3.501,
|
| 188 |
+
"train_steps_per_second": 0.219
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 7.101351351351352,
|
| 192 |
+
"eval_accuracy": 0.9290322580645162,
|
| 193 |
+
"eval_loss": 0.2252088487148285,
|
| 194 |
+
"eval_runtime": 13.824,
|
| 195 |
+
"eval_samples_per_second": 11.212,
|
| 196 |
+
"eval_steps_per_second": 0.723,
|
| 197 |
+
"step": 148
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 7.101351351351352,
|
| 201 |
+
"eval_accuracy": 0.9290322580645162,
|
| 202 |
+
"eval_loss": 0.2266511768102646,
|
| 203 |
+
"eval_runtime": 13.6046,
|
| 204 |
+
"eval_samples_per_second": 11.393,
|
| 205 |
+
"eval_steps_per_second": 0.735,
|
| 206 |
+
"step": 148
|
| 207 |
}
|
| 208 |
],
|
| 209 |
"logging_steps": 10,
|
| 210 |
+
"max_steps": 148,
|
| 211 |
"num_input_tokens_seen": 0,
|
| 212 |
"num_train_epochs": 9223372036854775807,
|
| 213 |
"save_steps": 500,
|
|
|
|
| 223 |
"attributes": {}
|
| 224 |
}
|
| 225 |
},
|
| 226 |
+
"total_flos": 2.915999166844109e+18,
|
| 227 |
+
"train_batch_size": 16,
|
| 228 |
"trial_name": null,
|
| 229 |
"trial_params": null
|
| 230 |
}
|