Training in progress, epoch 1
Browse files- all_results.json +11 -11
- eval_results.json +6 -6
- pytorch_model.bin +1 -1
- train_results.json +6 -6
- trainer_state.json +332 -191
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 1.
|
| 5 |
-
"eval_runtime": 0.
|
| 6 |
"eval_samples": 500,
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second":
|
| 9 |
-
"total_flos":
|
| 10 |
-
"train_loss": 1.
|
| 11 |
-
"train_runtime":
|
| 12 |
"train_samples": 4000,
|
| 13 |
-
"train_samples_per_second":
|
| 14 |
-
"train_steps_per_second":
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.2919999957084656,
|
| 4 |
+
"eval_loss": 1.569107174873352,
|
| 5 |
+
"eval_runtime": 0.7522,
|
| 6 |
"eval_samples": 500,
|
| 7 |
+
"eval_samples_per_second": 664.689,
|
| 8 |
+
"eval_steps_per_second": 42.54,
|
| 9 |
+
"total_flos": 10492735676640.0,
|
| 10 |
+
"train_loss": 1.5705763679504394,
|
| 11 |
+
"train_runtime": 78.9966,
|
| 12 |
"train_samples": 4000,
|
| 13 |
+
"train_samples_per_second": 253.176,
|
| 14 |
+
"train_steps_per_second": 15.823
|
| 15 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 1.
|
| 5 |
-
"eval_runtime": 0.
|
| 6 |
"eval_samples": 500,
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 0.2919999957084656,
|
| 4 |
+
"eval_loss": 1.569107174873352,
|
| 5 |
+
"eval_runtime": 0.7522,
|
| 6 |
"eval_samples": 500,
|
| 7 |
+
"eval_samples_per_second": 664.689,
|
| 8 |
+
"eval_steps_per_second": 42.54
|
| 9 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 12755881
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fb9208d3ee68d75842fcb7a1f0a04d765ee4bf462a4876219b125988db0d436c
|
| 3 |
size 12755881
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 4000,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 10492735676640.0,
|
| 4 |
+
"train_loss": 1.5705763679504394,
|
| 5 |
+
"train_runtime": 78.9966,
|
| 6 |
"train_samples": 4000,
|
| 7 |
+
"train_samples_per_second": 253.176,
|
| 8 |
+
"train_steps_per_second": 15.823
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,301 +1,442 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 1.
|
| 3 |
-
"best_model_checkpoint": "chinese_roberta_L-2_H-128-finetuned-MC-hyper/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
-
"epoch": 0.
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss": 1.
|
| 14 |
"step": 20
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss": 1.
|
| 20 |
"step": 40
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 0.
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss": 1.
|
| 26 |
"step": 60
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch":
|
| 30 |
-
"
|
| 31 |
-
"
|
| 32 |
-
"eval_runtime": 0.3306,
|
| 33 |
-
"eval_samples_per_second": 1512.408,
|
| 34 |
-
"eval_steps_per_second": 24.199,
|
| 35 |
-
"step": 63
|
| 36 |
-
},
|
| 37 |
-
{
|
| 38 |
-
"epoch": 1.27,
|
| 39 |
-
"learning_rate": 8.730158730158731e-06,
|
| 40 |
-
"loss": 1.6092,
|
| 41 |
"step": 80
|
| 42 |
},
|
| 43 |
{
|
| 44 |
-
"epoch":
|
| 45 |
-
"learning_rate":
|
| 46 |
-
"loss": 1.
|
| 47 |
"step": 100
|
| 48 |
},
|
| 49 |
{
|
| 50 |
-
"epoch":
|
| 51 |
-
"learning_rate":
|
| 52 |
-
"loss": 1.
|
| 53 |
"step": 120
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch":
|
| 57 |
-
"
|
| 58 |
-
"
|
| 59 |
-
"eval_runtime": 0.3325,
|
| 60 |
-
"eval_samples_per_second": 1503.597,
|
| 61 |
-
"eval_steps_per_second": 24.058,
|
| 62 |
-
"step": 126
|
| 63 |
-
},
|
| 64 |
-
{
|
| 65 |
-
"epoch": 2.22,
|
| 66 |
-
"learning_rate": 7.77777777777778e-06,
|
| 67 |
-
"loss": 1.6073,
|
| 68 |
"step": 140
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch":
|
| 72 |
-
"learning_rate":
|
| 73 |
-
"loss": 1.
|
| 74 |
"step": 160
|
| 75 |
},
|
| 76 |
{
|
| 77 |
-
"epoch":
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 1.
|
| 80 |
"step": 180
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch":
|
| 84 |
-
"
|
| 85 |
-
"
|
| 86 |
-
"eval_runtime": 0.3267,
|
| 87 |
-
"eval_samples_per_second": 1530.439,
|
| 88 |
-
"eval_steps_per_second": 24.487,
|
| 89 |
-
"step": 189
|
| 90 |
-
},
|
| 91 |
-
{
|
| 92 |
-
"epoch": 3.17,
|
| 93 |
-
"learning_rate": 6.825396825396826e-06,
|
| 94 |
-
"loss": 1.6061,
|
| 95 |
"step": 200
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch":
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 1.
|
| 101 |
"step": 220
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch":
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 240
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch":
|
| 111 |
-
"eval_accuracy": 0.
|
| 112 |
-
"eval_loss": 1.
|
| 113 |
-
"eval_runtime": 0.
|
| 114 |
-
"eval_samples_per_second":
|
| 115 |
-
"eval_steps_per_second":
|
| 116 |
-
"step":
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch":
|
| 120 |
-
"learning_rate":
|
| 121 |
-
"loss": 1.
|
| 122 |
"step": 260
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch":
|
| 126 |
-
"learning_rate":
|
| 127 |
-
"loss": 1.
|
| 128 |
"step": 280
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch":
|
| 132 |
-
"learning_rate":
|
| 133 |
-
"loss": 1.
|
| 134 |
"step": 300
|
| 135 |
},
|
| 136 |
{
|
| 137 |
-
"epoch":
|
| 138 |
-
"
|
| 139 |
-
"
|
| 140 |
-
"eval_runtime": 0.3348,
|
| 141 |
-
"eval_samples_per_second": 1493.284,
|
| 142 |
-
"eval_steps_per_second": 23.893,
|
| 143 |
-
"step": 315
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 5.08,
|
| 147 |
-
"learning_rate": 4.920634920634921e-06,
|
| 148 |
-
"loss": 1.6059,
|
| 149 |
"step": 320
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch":
|
| 153 |
-
"learning_rate":
|
| 154 |
-
"loss": 1.
|
| 155 |
"step": 340
|
| 156 |
},
|
| 157 |
{
|
| 158 |
-
"epoch":
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"loss": 1.
|
| 161 |
"step": 360
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch":
|
| 165 |
-
"
|
| 166 |
-
"
|
| 167 |
-
"eval_runtime": 0.3262,
|
| 168 |
-
"eval_samples_per_second": 1532.953,
|
| 169 |
-
"eval_steps_per_second": 24.527,
|
| 170 |
-
"step": 378
|
| 171 |
-
},
|
| 172 |
-
{
|
| 173 |
-
"epoch": 6.03,
|
| 174 |
-
"learning_rate": 3.968253968253968e-06,
|
| 175 |
-
"loss": 1.6031,
|
| 176 |
"step": 380
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch": 6
|
| 180 |
-
"learning_rate": 3.
|
| 181 |
-
"loss": 1.
|
| 182 |
"step": 400
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"learning_rate": 3.
|
| 187 |
-
"loss": 1.
|
| 188 |
"step": 420
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"learning_rate": 3.
|
| 193 |
-
"loss": 1.
|
| 194 |
"step": 440
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"
|
| 199 |
-
"
|
| 200 |
-
"eval_runtime": 0.3276,
|
| 201 |
-
"eval_samples_per_second": 1526.145,
|
| 202 |
-
"eval_steps_per_second": 24.418,
|
| 203 |
-
"step": 441
|
| 204 |
-
},
|
| 205 |
-
{
|
| 206 |
-
"epoch": 7.3,
|
| 207 |
-
"learning_rate": 2.6984126984126986e-06,
|
| 208 |
-
"loss": 1.6022,
|
| 209 |
"step": 460
|
| 210 |
},
|
| 211 |
{
|
| 212 |
-
"epoch":
|
| 213 |
-
"learning_rate":
|
| 214 |
-
"loss": 1.
|
| 215 |
"step": 480
|
| 216 |
},
|
| 217 |
{
|
| 218 |
-
"epoch":
|
| 219 |
-
"learning_rate":
|
| 220 |
-
"loss": 1.
|
| 221 |
"step": 500
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch":
|
| 225 |
-
"eval_accuracy": 0.
|
| 226 |
-
"eval_loss": 1.
|
| 227 |
-
"eval_runtime": 0.
|
| 228 |
-
"eval_samples_per_second":
|
| 229 |
-
"eval_steps_per_second":
|
| 230 |
-
"step":
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch":
|
| 234 |
-
"learning_rate":
|
| 235 |
-
"loss": 1.
|
| 236 |
"step": 520
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch":
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 1.
|
| 242 |
"step": 540
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch":
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss": 1.
|
| 248 |
"step": 560
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"
|
| 253 |
-
"
|
| 254 |
-
"eval_runtime": 0.3269,
|
| 255 |
-
"eval_samples_per_second": 1529.374,
|
| 256 |
-
"eval_steps_per_second": 24.47,
|
| 257 |
-
"step": 567
|
| 258 |
-
},
|
| 259 |
-
{
|
| 260 |
-
"epoch": 9.21,
|
| 261 |
-
"learning_rate": 7.936507936507937e-07,
|
| 262 |
-
"loss": 1.6016,
|
| 263 |
"step": 580
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"learning_rate":
|
| 268 |
-
"loss": 1.
|
| 269 |
"step": 600
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch":
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 1.
|
| 275 |
"step": 620
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
-
"
|
| 280 |
-
"
|
| 281 |
-
"
|
| 282 |
-
"eval_samples_per_second": 1535.263,
|
| 283 |
-
"eval_steps_per_second": 24.564,
|
| 284 |
-
"step": 630
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch":
|
| 288 |
-
"
|
| 289 |
-
"
|
| 290 |
-
"
|
| 291 |
-
|
| 292 |
-
|
| 293 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 294 |
}
|
| 295 |
],
|
| 296 |
-
"max_steps":
|
| 297 |
-
"num_train_epochs":
|
| 298 |
-
"total_flos":
|
| 299 |
"trial_name": null,
|
| 300 |
"trial_params": null
|
| 301 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 1.569107174873352,
|
| 3 |
+
"best_model_checkpoint": "chinese_roberta_L-2_H-128-finetuned-MC-hyper/checkpoint-1250",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 1250,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.08,
|
| 12 |
+
"learning_rate": 4.92e-05,
|
| 13 |
+
"loss": 1.6109,
|
| 14 |
"step": 20
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.16,
|
| 18 |
+
"learning_rate": 4.8400000000000004e-05,
|
| 19 |
+
"loss": 1.6093,
|
| 20 |
"step": 40
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.24,
|
| 24 |
+
"learning_rate": 4.76e-05,
|
| 25 |
+
"loss": 1.6059,
|
| 26 |
"step": 60
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.32,
|
| 30 |
+
"learning_rate": 4.6800000000000006e-05,
|
| 31 |
+
"loss": 1.6083,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
"step": 80
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.4,
|
| 36 |
+
"learning_rate": 4.600000000000001e-05,
|
| 37 |
+
"loss": 1.6063,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.48,
|
| 42 |
+
"learning_rate": 4.52e-05,
|
| 43 |
+
"loss": 1.606,
|
| 44 |
"step": 120
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.56,
|
| 48 |
+
"learning_rate": 4.44e-05,
|
| 49 |
+
"loss": 1.6085,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
"step": 140
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"epoch": 0.64,
|
| 54 |
+
"learning_rate": 4.36e-05,
|
| 55 |
+
"loss": 1.609,
|
| 56 |
"step": 160
|
| 57 |
},
|
| 58 |
{
|
| 59 |
+
"epoch": 0.72,
|
| 60 |
+
"learning_rate": 4.2800000000000004e-05,
|
| 61 |
+
"loss": 1.6032,
|
| 62 |
"step": 180
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.8,
|
| 66 |
+
"learning_rate": 4.2e-05,
|
| 67 |
+
"loss": 1.609,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 68 |
"step": 200
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 0.88,
|
| 72 |
+
"learning_rate": 4.12e-05,
|
| 73 |
+
"loss": 1.6022,
|
| 74 |
"step": 220
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 0.96,
|
| 78 |
+
"learning_rate": 4.0400000000000006e-05,
|
| 79 |
+
"loss": 1.5993,
|
| 80 |
"step": 240
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 1.0,
|
| 84 |
+
"eval_accuracy": 0.27799999713897705,
|
| 85 |
+
"eval_loss": 1.596663236618042,
|
| 86 |
+
"eval_runtime": 0.7533,
|
| 87 |
+
"eval_samples_per_second": 663.789,
|
| 88 |
+
"eval_steps_per_second": 42.483,
|
| 89 |
+
"step": 250
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 1.04,
|
| 93 |
+
"learning_rate": 3.960000000000001e-05,
|
| 94 |
+
"loss": 1.6047,
|
| 95 |
"step": 260
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 1.12,
|
| 99 |
+
"learning_rate": 3.88e-05,
|
| 100 |
+
"loss": 1.6005,
|
| 101 |
"step": 280
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 1.2,
|
| 105 |
+
"learning_rate": 3.8e-05,
|
| 106 |
+
"loss": 1.5939,
|
| 107 |
"step": 300
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 1.28,
|
| 111 |
+
"learning_rate": 3.72e-05,
|
| 112 |
+
"loss": 1.5916,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 113 |
"step": 320
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 1.36,
|
| 117 |
+
"learning_rate": 3.6400000000000004e-05,
|
| 118 |
+
"loss": 1.5908,
|
| 119 |
"step": 340
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 1.44,
|
| 123 |
+
"learning_rate": 3.56e-05,
|
| 124 |
+
"loss": 1.59,
|
| 125 |
"step": 360
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 1.52,
|
| 129 |
+
"learning_rate": 3.48e-05,
|
| 130 |
+
"loss": 1.5996,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 131 |
"step": 380
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 1.6,
|
| 135 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 136 |
+
"loss": 1.586,
|
| 137 |
"step": 400
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 1.68,
|
| 141 |
+
"learning_rate": 3.32e-05,
|
| 142 |
+
"loss": 1.5712,
|
| 143 |
"step": 420
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 1.76,
|
| 147 |
+
"learning_rate": 3.24e-05,
|
| 148 |
+
"loss": 1.5726,
|
| 149 |
"step": 440
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 1.84,
|
| 153 |
+
"learning_rate": 3.16e-05,
|
| 154 |
+
"loss": 1.5727,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 155 |
"step": 460
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 1.92,
|
| 159 |
+
"learning_rate": 3.08e-05,
|
| 160 |
+
"loss": 1.5822,
|
| 161 |
"step": 480
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 2.0,
|
| 165 |
+
"learning_rate": 3e-05,
|
| 166 |
+
"loss": 1.5731,
|
| 167 |
"step": 500
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 2.0,
|
| 171 |
+
"eval_accuracy": 0.3019999861717224,
|
| 172 |
+
"eval_loss": 1.5877653360366821,
|
| 173 |
+
"eval_runtime": 0.647,
|
| 174 |
+
"eval_samples_per_second": 772.82,
|
| 175 |
+
"eval_steps_per_second": 49.46,
|
| 176 |
+
"step": 500
|
| 177 |
},
|
| 178 |
{
|
| 179 |
+
"epoch": 2.08,
|
| 180 |
+
"learning_rate": 2.9199999999999998e-05,
|
| 181 |
+
"loss": 1.5914,
|
| 182 |
"step": 520
|
| 183 |
},
|
| 184 |
{
|
| 185 |
+
"epoch": 2.16,
|
| 186 |
+
"learning_rate": 2.84e-05,
|
| 187 |
+
"loss": 1.5627,
|
| 188 |
"step": 540
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 2.24,
|
| 192 |
+
"learning_rate": 2.7600000000000003e-05,
|
| 193 |
+
"loss": 1.5672,
|
| 194 |
"step": 560
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 2.32,
|
| 198 |
+
"learning_rate": 2.6800000000000004e-05,
|
| 199 |
+
"loss": 1.564,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 200 |
"step": 580
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 2.4,
|
| 204 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 205 |
+
"loss": 1.577,
|
| 206 |
"step": 600
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 2.48,
|
| 210 |
+
"learning_rate": 2.5200000000000003e-05,
|
| 211 |
+
"loss": 1.5712,
|
| 212 |
"step": 620
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 2.56,
|
| 216 |
+
"learning_rate": 2.44e-05,
|
| 217 |
+
"loss": 1.5725,
|
| 218 |
+
"step": 640
|
|
|
|
|
|
|
|
|
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 2.64,
|
| 222 |
+
"learning_rate": 2.36e-05,
|
| 223 |
+
"loss": 1.5532,
|
| 224 |
+
"step": 660
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.72,
|
| 228 |
+
"learning_rate": 2.2800000000000002e-05,
|
| 229 |
+
"loss": 1.5703,
|
| 230 |
+
"step": 680
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.8,
|
| 234 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 235 |
+
"loss": 1.5803,
|
| 236 |
+
"step": 700
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.88,
|
| 240 |
+
"learning_rate": 2.12e-05,
|
| 241 |
+
"loss": 1.5728,
|
| 242 |
+
"step": 720
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.96,
|
| 246 |
+
"learning_rate": 2.04e-05,
|
| 247 |
+
"loss": 1.5434,
|
| 248 |
+
"step": 740
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 3.0,
|
| 252 |
+
"eval_accuracy": 0.29600000381469727,
|
| 253 |
+
"eval_loss": 1.5791035890579224,
|
| 254 |
+
"eval_runtime": 0.7356,
|
| 255 |
+
"eval_samples_per_second": 679.717,
|
| 256 |
+
"eval_steps_per_second": 43.502,
|
| 257 |
+
"step": 750
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.04,
|
| 261 |
+
"learning_rate": 1.9600000000000002e-05,
|
| 262 |
+
"loss": 1.5569,
|
| 263 |
+
"step": 760
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 3.12,
|
| 267 |
+
"learning_rate": 1.88e-05,
|
| 268 |
+
"loss": 1.5382,
|
| 269 |
+
"step": 780
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.2,
|
| 273 |
+
"learning_rate": 1.8e-05,
|
| 274 |
+
"loss": 1.5478,
|
| 275 |
+
"step": 800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 3.28,
|
| 279 |
+
"learning_rate": 1.7199999999999998e-05,
|
| 280 |
+
"loss": 1.5665,
|
| 281 |
+
"step": 820
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.36,
|
| 285 |
+
"learning_rate": 1.6400000000000002e-05,
|
| 286 |
+
"loss": 1.5494,
|
| 287 |
+
"step": 840
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 3.44,
|
| 291 |
+
"learning_rate": 1.56e-05,
|
| 292 |
+
"loss": 1.5489,
|
| 293 |
+
"step": 860
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.52,
|
| 297 |
+
"learning_rate": 1.48e-05,
|
| 298 |
+
"loss": 1.5483,
|
| 299 |
+
"step": 880
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.6,
|
| 303 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 304 |
+
"loss": 1.5388,
|
| 305 |
+
"step": 900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 3.68,
|
| 309 |
+
"learning_rate": 1.32e-05,
|
| 310 |
+
"loss": 1.5563,
|
| 311 |
+
"step": 920
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 3.76,
|
| 315 |
+
"learning_rate": 1.24e-05,
|
| 316 |
+
"loss": 1.5616,
|
| 317 |
+
"step": 940
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 3.84,
|
| 321 |
+
"learning_rate": 1.16e-05,
|
| 322 |
+
"loss": 1.5512,
|
| 323 |
+
"step": 960
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.92,
|
| 327 |
+
"learning_rate": 1.08e-05,
|
| 328 |
+
"loss": 1.5288,
|
| 329 |
+
"step": 980
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 4.0,
|
| 333 |
+
"learning_rate": 1e-05,
|
| 334 |
+
"loss": 1.5473,
|
| 335 |
+
"step": 1000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 4.0,
|
| 339 |
+
"eval_accuracy": 0.2939999997615814,
|
| 340 |
+
"eval_loss": 1.5701889991760254,
|
| 341 |
+
"eval_runtime": 0.7589,
|
| 342 |
+
"eval_samples_per_second": 658.833,
|
| 343 |
+
"eval_steps_per_second": 42.165,
|
| 344 |
+
"step": 1000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 4.08,
|
| 348 |
+
"learning_rate": 9.2e-06,
|
| 349 |
+
"loss": 1.5642,
|
| 350 |
+
"step": 1020
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 4.16,
|
| 354 |
+
"learning_rate": 8.400000000000001e-06,
|
| 355 |
+
"loss": 1.5334,
|
| 356 |
+
"step": 1040
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 4.24,
|
| 360 |
+
"learning_rate": 7.6e-06,
|
| 361 |
+
"loss": 1.526,
|
| 362 |
+
"step": 1060
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 4.32,
|
| 366 |
+
"learning_rate": 6.800000000000001e-06,
|
| 367 |
+
"loss": 1.5361,
|
| 368 |
+
"step": 1080
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 4.4,
|
| 372 |
+
"learning_rate": 6e-06,
|
| 373 |
+
"loss": 1.573,
|
| 374 |
+
"step": 1100
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.48,
|
| 378 |
+
"learning_rate": 5.2e-06,
|
| 379 |
+
"loss": 1.5543,
|
| 380 |
+
"step": 1120
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 4.56,
|
| 384 |
+
"learning_rate": 4.4e-06,
|
| 385 |
+
"loss": 1.5193,
|
| 386 |
+
"step": 1140
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 4.64,
|
| 390 |
+
"learning_rate": 3.6e-06,
|
| 391 |
+
"loss": 1.5401,
|
| 392 |
+
"step": 1160
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 4.72,
|
| 396 |
+
"learning_rate": 2.8000000000000003e-06,
|
| 397 |
+
"loss": 1.5373,
|
| 398 |
+
"step": 1180
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.8,
|
| 402 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 403 |
+
"loss": 1.5279,
|
| 404 |
+
"step": 1200
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.88,
|
| 408 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 409 |
+
"loss": 1.5503,
|
| 410 |
+
"step": 1220
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 4.96,
|
| 414 |
+
"learning_rate": 4.0000000000000003e-07,
|
| 415 |
+
"loss": 1.5422,
|
| 416 |
+
"step": 1240
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 5.0,
|
| 420 |
+
"eval_accuracy": 0.2919999957084656,
|
| 421 |
+
"eval_loss": 1.569107174873352,
|
| 422 |
+
"eval_runtime": 0.7454,
|
| 423 |
+
"eval_samples_per_second": 670.819,
|
| 424 |
+
"eval_steps_per_second": 42.932,
|
| 425 |
+
"step": 1250
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 5.0,
|
| 429 |
+
"step": 1250,
|
| 430 |
+
"total_flos": 10492735676640.0,
|
| 431 |
+
"train_loss": 1.5705763679504394,
|
| 432 |
+
"train_runtime": 78.9966,
|
| 433 |
+
"train_samples_per_second": 253.176,
|
| 434 |
+
"train_steps_per_second": 15.823
|
| 435 |
}
|
| 436 |
],
|
| 437 |
+
"max_steps": 1250,
|
| 438 |
+
"num_train_epochs": 5,
|
| 439 |
+
"total_flos": 10492735676640.0,
|
| 440 |
"trial_name": null,
|
| 441 |
"trial_params": null
|
| 442 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2799
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f1eb0324ba0a73ee29fb2559c05fad4e6c64a97f34e7bb4de884b1c0c9cc415f
|
| 3 |
size 2799
|