Upload 9 files
Browse files- optimizer.pt +1 -1
- pytorch_model.bin +1 -1
- rng_state.pth +1 -1
- scaler.pt +1 -1
- scheduler.pt +1 -1
- trainer_state.json +10 -295
- training_args.bin +1 -1
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2490593669
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cdf2c9a2b141d61ae97e6f49cd706632b3b9db6675f7fb64adc99a2beffed4a6
|
| 3 |
size 2490593669
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262168365
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f1bc7bd2434b9ee5cf00b9927285575898d67205317086fa93bb0ea75f822ef
|
| 3 |
size 1262168365
|
rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14639
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9be6e271ede82a9bafb9710b6f305bb3591279ca8f3cffac6f33b9b4d8093cc7
|
| 3 |
size 14639
|
scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 557
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2891524b7f9cad24992893a2dc3e9813564138d608b7f65cca009a6af698600
|
| 3 |
size 557
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:934b8e9fba487b03faa18f476924d39420f50ac287656b5ab10fb96e2162d17f
|
| 3 |
size 627
|
trainer_state.json
CHANGED
|
@@ -1,316 +1,31 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 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": 0.08,
|
| 12 |
-
"learning_rate": 1e-05,
|
| 13 |
-
"loss": 0.0212,
|
| 14 |
-
"step": 10
|
| 15 |
-
},
|
| 16 |
-
{
|
| 17 |
-
"epoch": 0.08,
|
| 18 |
-
"eval_loss": 0.32025715708732605,
|
| 19 |
-
"eval_runtime": 212.3535,
|
| 20 |
-
"eval_samples_per_second": 16.929,
|
| 21 |
-
"eval_steps_per_second": 2.119,
|
| 22 |
-
"eval_wer": 0.21324007807417047,
|
| 23 |
-
"step": 10
|
| 24 |
-
},
|
| 25 |
-
{
|
| 26 |
-
"epoch": 0.16,
|
| 27 |
-
"learning_rate": 2e-05,
|
| 28 |
-
"loss": 0.0188,
|
| 29 |
-
"step": 20
|
| 30 |
-
},
|
| 31 |
-
{
|
| 32 |
-
"epoch": 0.16,
|
| 33 |
-
"eval_loss": 0.3204110860824585,
|
| 34 |
-
"eval_runtime": 140.687,
|
| 35 |
-
"eval_samples_per_second": 25.553,
|
| 36 |
-
"eval_steps_per_second": 3.199,
|
| 37 |
-
"eval_wer": 0.21329429624810237,
|
| 38 |
-
"step": 20
|
| 39 |
-
},
|
| 40 |
{
|
| 41 |
"epoch": 0.24,
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 30
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.24,
|
| 48 |
-
"eval_loss": 0.
|
| 49 |
-
"eval_runtime":
|
| 50 |
-
"eval_samples_per_second":
|
| 51 |
-
"eval_steps_per_second":
|
| 52 |
-
"eval_wer": 0.
|
| 53 |
"step": 30
|
| 54 |
-
},
|
| 55 |
-
{
|
| 56 |
-
"epoch": 0.32,
|
| 57 |
-
"learning_rate": 4e-05,
|
| 58 |
-
"loss": 0.0187,
|
| 59 |
-
"step": 40
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.32,
|
| 63 |
-
"eval_loss": 0.3250657021999359,
|
| 64 |
-
"eval_runtime": 144.0633,
|
| 65 |
-
"eval_samples_per_second": 24.954,
|
| 66 |
-
"eval_steps_per_second": 3.124,
|
| 67 |
-
"eval_wer": 0.21242680546519194,
|
| 68 |
-
"step": 40
|
| 69 |
-
},
|
| 70 |
-
{
|
| 71 |
-
"epoch": 0.4,
|
| 72 |
-
"learning_rate": 5e-05,
|
| 73 |
-
"loss": 0.0199,
|
| 74 |
-
"step": 50
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 0.4,
|
| 78 |
-
"eval_loss": 0.3246767818927765,
|
| 79 |
-
"eval_runtime": 144.6093,
|
| 80 |
-
"eval_samples_per_second": 24.86,
|
| 81 |
-
"eval_steps_per_second": 3.112,
|
| 82 |
-
"eval_wer": 0.21258945998698764,
|
| 83 |
-
"step": 50
|
| 84 |
-
},
|
| 85 |
-
{
|
| 86 |
-
"epoch": 0.48,
|
| 87 |
-
"learning_rate": 6e-05,
|
| 88 |
-
"loss": 0.0203,
|
| 89 |
-
"step": 60
|
| 90 |
-
},
|
| 91 |
-
{
|
| 92 |
-
"epoch": 0.48,
|
| 93 |
-
"eval_loss": 0.31979820132255554,
|
| 94 |
-
"eval_runtime": 152.9805,
|
| 95 |
-
"eval_samples_per_second": 23.5,
|
| 96 |
-
"eval_steps_per_second": 2.942,
|
| 97 |
-
"eval_wer": 0.21242680546519194,
|
| 98 |
-
"step": 60
|
| 99 |
-
},
|
| 100 |
-
{
|
| 101 |
-
"epoch": 0.56,
|
| 102 |
-
"learning_rate": 7e-05,
|
| 103 |
-
"loss": 0.0199,
|
| 104 |
-
"step": 70
|
| 105 |
-
},
|
| 106 |
-
{
|
| 107 |
-
"epoch": 0.56,
|
| 108 |
-
"eval_loss": 0.31615573167800903,
|
| 109 |
-
"eval_runtime": 152.4511,
|
| 110 |
-
"eval_samples_per_second": 23.581,
|
| 111 |
-
"eval_steps_per_second": 2.952,
|
| 112 |
-
"eval_wer": 0.21378225981348947,
|
| 113 |
-
"step": 70
|
| 114 |
-
},
|
| 115 |
-
{
|
| 116 |
-
"epoch": 0.64,
|
| 117 |
-
"learning_rate": 8e-05,
|
| 118 |
-
"loss": 0.0203,
|
| 119 |
-
"step": 80
|
| 120 |
-
},
|
| 121 |
-
{
|
| 122 |
-
"epoch": 0.64,
|
| 123 |
-
"eval_loss": 0.3140092194080353,
|
| 124 |
-
"eval_runtime": 146.0972,
|
| 125 |
-
"eval_samples_per_second": 24.607,
|
| 126 |
-
"eval_steps_per_second": 3.08,
|
| 127 |
-
"eval_wer": 0.2144328779006723,
|
| 128 |
-
"step": 80
|
| 129 |
-
},
|
| 130 |
-
{
|
| 131 |
-
"epoch": 0.71,
|
| 132 |
-
"learning_rate": 9e-05,
|
| 133 |
-
"loss": 0.0189,
|
| 134 |
-
"step": 90
|
| 135 |
-
},
|
| 136 |
-
{
|
| 137 |
-
"epoch": 0.71,
|
| 138 |
-
"eval_loss": 0.3206515908241272,
|
| 139 |
-
"eval_runtime": 151.0056,
|
| 140 |
-
"eval_samples_per_second": 23.807,
|
| 141 |
-
"eval_steps_per_second": 2.98,
|
| 142 |
-
"eval_wer": 0.21307742355237475,
|
| 143 |
-
"step": 90
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.79,
|
| 147 |
-
"learning_rate": 0.0001,
|
| 148 |
-
"loss": 0.0203,
|
| 149 |
-
"step": 100
|
| 150 |
-
},
|
| 151 |
-
{
|
| 152 |
-
"epoch": 0.79,
|
| 153 |
-
"eval_loss": 0.3265531361103058,
|
| 154 |
-
"eval_runtime": 163.535,
|
| 155 |
-
"eval_samples_per_second": 21.983,
|
| 156 |
-
"eval_steps_per_second": 2.752,
|
| 157 |
-
"eval_wer": 0.21296898720451096,
|
| 158 |
-
"step": 100
|
| 159 |
-
},
|
| 160 |
-
{
|
| 161 |
-
"epoch": 0.87,
|
| 162 |
-
"learning_rate": 9.999199359487592e-05,
|
| 163 |
-
"loss": 0.0221,
|
| 164 |
-
"step": 110
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 0.87,
|
| 168 |
-
"eval_loss": 0.3243820071220398,
|
| 169 |
-
"eval_runtime": 159.6061,
|
| 170 |
-
"eval_samples_per_second": 22.524,
|
| 171 |
-
"eval_steps_per_second": 2.819,
|
| 172 |
-
"eval_wer": 0.2141075688570809,
|
| 173 |
-
"step": 110
|
| 174 |
-
},
|
| 175 |
-
{
|
| 176 |
-
"epoch": 0.95,
|
| 177 |
-
"learning_rate": 9.99839871897518e-05,
|
| 178 |
-
"loss": 0.0219,
|
| 179 |
-
"step": 120
|
| 180 |
-
},
|
| 181 |
-
{
|
| 182 |
-
"epoch": 0.95,
|
| 183 |
-
"eval_loss": 0.3229624629020691,
|
| 184 |
-
"eval_runtime": 154.2088,
|
| 185 |
-
"eval_samples_per_second": 23.313,
|
| 186 |
-
"eval_steps_per_second": 2.918,
|
| 187 |
-
"eval_wer": 0.21486662329212752,
|
| 188 |
-
"step": 120
|
| 189 |
-
},
|
| 190 |
-
{
|
| 191 |
-
"epoch": 1.04,
|
| 192 |
-
"learning_rate": 9.997598078462771e-05,
|
| 193 |
-
"loss": 0.0245,
|
| 194 |
-
"step": 130
|
| 195 |
-
},
|
| 196 |
-
{
|
| 197 |
-
"epoch": 1.04,
|
| 198 |
-
"eval_loss": 0.31454530358314514,
|
| 199 |
-
"eval_runtime": 154.6481,
|
| 200 |
-
"eval_samples_per_second": 23.246,
|
| 201 |
-
"eval_steps_per_second": 2.91,
|
| 202 |
-
"eval_wer": 0.2169811320754717,
|
| 203 |
-
"step": 130
|
| 204 |
-
},
|
| 205 |
-
{
|
| 206 |
-
"epoch": 1.12,
|
| 207 |
-
"learning_rate": 9.996797437950361e-05,
|
| 208 |
-
"loss": 0.0231,
|
| 209 |
-
"step": 140
|
| 210 |
-
},
|
| 211 |
-
{
|
| 212 |
-
"epoch": 1.12,
|
| 213 |
-
"eval_loss": 0.3071616590023041,
|
| 214 |
-
"eval_runtime": 157.0332,
|
| 215 |
-
"eval_samples_per_second": 22.893,
|
| 216 |
-
"eval_steps_per_second": 2.866,
|
| 217 |
-
"eval_wer": 0.21508349598785512,
|
| 218 |
-
"step": 140
|
| 219 |
-
},
|
| 220 |
-
{
|
| 221 |
-
"epoch": 1.2,
|
| 222 |
-
"learning_rate": 9.995996797437951e-05,
|
| 223 |
-
"loss": 0.0194,
|
| 224 |
-
"step": 150
|
| 225 |
-
},
|
| 226 |
-
{
|
| 227 |
-
"epoch": 1.2,
|
| 228 |
-
"eval_loss": 0.323686808347702,
|
| 229 |
-
"eval_runtime": 158.852,
|
| 230 |
-
"eval_samples_per_second": 22.631,
|
| 231 |
-
"eval_steps_per_second": 2.833,
|
| 232 |
-
"eval_wer": 0.21611364129256128,
|
| 233 |
-
"step": 150
|
| 234 |
-
},
|
| 235 |
-
{
|
| 236 |
-
"epoch": 1.28,
|
| 237 |
-
"learning_rate": 9.995196156925542e-05,
|
| 238 |
-
"loss": 0.0158,
|
| 239 |
-
"step": 160
|
| 240 |
-
},
|
| 241 |
-
{
|
| 242 |
-
"epoch": 1.28,
|
| 243 |
-
"eval_loss": 0.35077646374702454,
|
| 244 |
-
"eval_runtime": 143.668,
|
| 245 |
-
"eval_samples_per_second": 25.023,
|
| 246 |
-
"eval_steps_per_second": 3.132,
|
| 247 |
-
"eval_wer": 0.21752331381479073,
|
| 248 |
-
"step": 160
|
| 249 |
-
},
|
| 250 |
-
{
|
| 251 |
-
"epoch": 1.36,
|
| 252 |
-
"learning_rate": 9.99439551641313e-05,
|
| 253 |
-
"loss": 0.0233,
|
| 254 |
-
"step": 170
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 1.36,
|
| 258 |
-
"eval_loss": 0.34845927357673645,
|
| 259 |
-
"eval_runtime": 146.6957,
|
| 260 |
-
"eval_samples_per_second": 24.507,
|
| 261 |
-
"eval_steps_per_second": 3.068,
|
| 262 |
-
"eval_wer": 0.2139991325092171,
|
| 263 |
-
"step": 170
|
| 264 |
-
},
|
| 265 |
-
{
|
| 266 |
-
"epoch": 1.44,
|
| 267 |
-
"learning_rate": 9.993594875900721e-05,
|
| 268 |
-
"loss": 0.0232,
|
| 269 |
-
"step": 180
|
| 270 |
-
},
|
| 271 |
-
{
|
| 272 |
-
"epoch": 1.44,
|
| 273 |
-
"eval_loss": 0.330231636762619,
|
| 274 |
-
"eval_runtime": 160.7938,
|
| 275 |
-
"eval_samples_per_second": 22.358,
|
| 276 |
-
"eval_steps_per_second": 2.799,
|
| 277 |
-
"eval_wer": 0.2168726957276079,
|
| 278 |
-
"step": 180
|
| 279 |
-
},
|
| 280 |
-
{
|
| 281 |
-
"epoch": 1.52,
|
| 282 |
-
"learning_rate": 9.992794235388311e-05,
|
| 283 |
-
"loss": 0.0164,
|
| 284 |
-
"step": 190
|
| 285 |
-
},
|
| 286 |
-
{
|
| 287 |
-
"epoch": 1.52,
|
| 288 |
-
"eval_loss": 0.32822760939598083,
|
| 289 |
-
"eval_runtime": 163.3169,
|
| 290 |
-
"eval_samples_per_second": 22.012,
|
| 291 |
-
"eval_steps_per_second": 2.755,
|
| 292 |
-
"eval_wer": 0.21356538711776188,
|
| 293 |
-
"step": 190
|
| 294 |
-
},
|
| 295 |
-
{
|
| 296 |
-
"epoch": 1.6,
|
| 297 |
-
"learning_rate": 9.991993594875901e-05,
|
| 298 |
-
"loss": 0.0218,
|
| 299 |
-
"step": 200
|
| 300 |
-
},
|
| 301 |
-
{
|
| 302 |
-
"epoch": 1.6,
|
| 303 |
-
"eval_loss": 0.33395659923553467,
|
| 304 |
-
"eval_runtime": 163.325,
|
| 305 |
-
"eval_samples_per_second": 22.011,
|
| 306 |
-
"eval_steps_per_second": 2.755,
|
| 307 |
-
"eval_wer": 0.21085447842116678,
|
| 308 |
-
"step": 200
|
| 309 |
}
|
| 310 |
],
|
| 311 |
"max_steps": 125000,
|
| 312 |
"num_train_epochs": 1000,
|
| 313 |
-
"total_flos":
|
| 314 |
"trial_name": null,
|
| 315 |
"trial_params": null
|
| 316 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.23833167825223436,
|
| 5 |
+
"global_step": 30,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
{
|
| 11 |
"epoch": 0.24,
|
| 12 |
+
"learning_rate": 6e-05,
|
| 13 |
+
"loss": 0.019,
|
| 14 |
"step": 30
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.24,
|
| 18 |
+
"eval_loss": 0.333274781703949,
|
| 19 |
+
"eval_runtime": 217.3871,
|
| 20 |
+
"eval_samples_per_second": 16.537,
|
| 21 |
+
"eval_steps_per_second": 2.07,
|
| 22 |
+
"eval_wer": 0.20998698763825635,
|
| 23 |
"step": 30
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
}
|
| 25 |
],
|
| 26 |
"max_steps": 125000,
|
| 27 |
"num_train_epochs": 1000,
|
| 28 |
+
"total_flos": 6.914475006165965e+17,
|
| 29 |
"trial_name": null,
|
| 30 |
"trial_params": null
|
| 31 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3323
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1bc78f10706cccf49dff72ccdd19bc3f1b28bbe47c4e276e818cce22f2537b6b
|
| 3 |
size 3323
|