| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.093366093366093, | |
| "global_step": 407, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0, | |
| "loss": 1.8118, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 3.6585365853658536e-07, | |
| "loss": 1.8122, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 8.53658536585366e-07, | |
| "loss": 1.8174, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.3414634146341465e-06, | |
| "loss": 1.7616, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.8292682926829268e-06, | |
| "loss": 1.6875, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.317073170731708e-06, | |
| "loss": 1.5201, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 2.8048780487804884e-06, | |
| "loss": 1.3982, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 3.292682926829269e-06, | |
| "loss": 1.3541, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 3.780487804878049e-06, | |
| "loss": 1.2092, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.268292682926829e-06, | |
| "loss": 1.1599, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.142654299736023, | |
| "eval_runtime": 101.9854, | |
| "eval_samples_per_second": 3.863, | |
| "eval_steps_per_second": 0.127, | |
| "eval_wer": 15.213946117274169, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 4.75609756097561e-06, | |
| "loss": 1.0124, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 5.243902439024391e-06, | |
| "loss": 0.9171, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 5.731707317073171e-06, | |
| "loss": 0.8027, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 6.219512195121951e-06, | |
| "loss": 0.7284, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 6.707317073170733e-06, | |
| "loss": 0.6185, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 7.1951219512195125e-06, | |
| "loss": 0.57, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 7.682926829268293e-06, | |
| "loss": 0.4985, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 8.170731707317073e-06, | |
| "loss": 0.488, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 8.658536585365854e-06, | |
| "loss": 0.4569, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 9.146341463414635e-06, | |
| "loss": 0.4655, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.5613037943840027, | |
| "eval_runtime": 91.9697, | |
| "eval_samples_per_second": 4.284, | |
| "eval_steps_per_second": 0.141, | |
| "eval_wer": 17.591125198098258, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 9.634146341463415e-06, | |
| "loss": 0.425, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.96923076923077e-06, | |
| "loss": 0.4162, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.3809, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.723076923076924e-06, | |
| "loss": 0.3533, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.3511, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 9.476923076923079e-06, | |
| "loss": 0.3475, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 9.353846153846155e-06, | |
| "loss": 0.321, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.2859, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.107692307692308e-06, | |
| "loss": 0.3191, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 8.984615384615386e-06, | |
| "loss": 0.2753, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "eval_loss": 0.5241264700889587, | |
| "eval_runtime": 88.0526, | |
| "eval_samples_per_second": 4.475, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 17.21321467755699, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.861538461538463e-06, | |
| "loss": 0.3104, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 8.73846153846154e-06, | |
| "loss": 0.2734, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.2608, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.492307692307693e-06, | |
| "loss": 0.2509, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 8.36923076923077e-06, | |
| "loss": 0.2548, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 8.246153846153848e-06, | |
| "loss": 0.2469, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 8.123076923076924e-06, | |
| "loss": 0.2231, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.2138, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 7.876923076923077e-06, | |
| "loss": 0.2349, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 7.753846153846155e-06, | |
| "loss": 0.2077, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "eval_loss": 0.5241798758506775, | |
| "eval_runtime": 88.5317, | |
| "eval_samples_per_second": 4.45, | |
| "eval_steps_per_second": 0.147, | |
| "eval_wer": 17.26197732536877, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 7.630769230769232e-06, | |
| "loss": 0.2322, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 7.507692307692308e-06, | |
| "loss": 0.2036, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.2058, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 7.261538461538462e-06, | |
| "loss": 0.1797, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 7.1384615384615385e-06, | |
| "loss": 0.186, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 7.015384615384616e-06, | |
| "loss": 0.2035, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 6.892307692307693e-06, | |
| "loss": 0.1794, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.1589, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 6.646153846153846e-06, | |
| "loss": 0.1879, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 6.523076923076923e-06, | |
| "loss": 0.1636, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "eval_loss": 0.5289868712425232, | |
| "eval_runtime": 95.5188, | |
| "eval_samples_per_second": 4.125, | |
| "eval_steps_per_second": 0.136, | |
| "eval_wer": 17.66426916981592, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 0.1767, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 6.276923076923077e-06, | |
| "loss": 0.1657, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.1607, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 6.030769230769231e-06, | |
| "loss": 0.1458, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 5.907692307692308e-06, | |
| "loss": 0.1541, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 5.784615384615385e-06, | |
| "loss": 0.1494, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 5.661538461538462e-06, | |
| "loss": 0.144, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.1311, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 5.415384615384615e-06, | |
| "loss": 0.1411, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 5.292307692307693e-06, | |
| "loss": 0.1322, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "eval_loss": 0.5350630283355713, | |
| "eval_runtime": 92.5111, | |
| "eval_samples_per_second": 4.259, | |
| "eval_steps_per_second": 0.141, | |
| "eval_wer": 18.2128489576984, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 5.16923076923077e-06, | |
| "loss": 0.1436, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 5.046153846153846e-06, | |
| "loss": 0.1375, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.1361, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 0.129, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 4.676923076923077e-06, | |
| "loss": 0.1127, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 4.553846153846154e-06, | |
| "loss": 0.1266, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 4.430769230769232e-06, | |
| "loss": 0.1193, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.1127, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 4.184615384615385e-06, | |
| "loss": 0.1064, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 4.061538461538462e-06, | |
| "loss": 0.123, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "eval_loss": 0.5429388284683228, | |
| "eval_runtime": 91.5818, | |
| "eval_samples_per_second": 4.302, | |
| "eval_steps_per_second": 0.142, | |
| "eval_wer": 18.907716689016212, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 3.938461538461539e-06, | |
| "loss": 0.1057, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.815384615384616e-06, | |
| "loss": 0.1258, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.1108, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 3.5692307692307692e-06, | |
| "loss": 0.1115, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 3.4461538461538464e-06, | |
| "loss": 0.0998, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 3.323076923076923e-06, | |
| "loss": 0.1106, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 0.1045, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.0908, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 2.953846153846154e-06, | |
| "loss": 0.0931, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 2.830769230769231e-06, | |
| "loss": 0.1074, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "eval_loss": 0.5500437021255493, | |
| "eval_runtime": 104.0907, | |
| "eval_samples_per_second": 3.785, | |
| "eval_steps_per_second": 0.125, | |
| "eval_wer": 19.054004632451544, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 2.7076923076923076e-06, | |
| "loss": 0.0937, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 2.584615384615385e-06, | |
| "loss": 0.1091, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.0951, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 2.3384615384615387e-06, | |
| "loss": 0.1003, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 2.215384615384616e-06, | |
| "loss": 0.0836, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 2.0923076923076926e-06, | |
| "loss": 0.0907, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 1.9692307692307693e-06, | |
| "loss": 0.1013, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.0891, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 1.7230769230769232e-06, | |
| "loss": 0.077, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 0.1007, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "eval_loss": 0.5552565455436707, | |
| "eval_runtime": 88.458, | |
| "eval_samples_per_second": 4.454, | |
| "eval_steps_per_second": 0.147, | |
| "eval_wer": 19.310008533463368, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "learning_rate": 1.476923076923077e-06, | |
| "loss": 0.0849, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 1.3538461538461538e-06, | |
| "loss": 0.0971, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.0876, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 1.107692307692308e-06, | |
| "loss": 0.0879, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 9.846153846153847e-07, | |
| "loss": 0.0805, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 8.615384615384616e-07, | |
| "loss": 0.0888, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 7.384615384615385e-07, | |
| "loss": 0.0858, | |
| "step": 388 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.0825, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 4.923076923076923e-07, | |
| "loss": 0.0748, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 3.6923076923076927e-07, | |
| "loss": 0.0876, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_loss": 0.5568162202835083, | |
| "eval_runtime": 89.7223, | |
| "eval_samples_per_second": 4.391, | |
| "eval_steps_per_second": 0.145, | |
| "eval_wer": 19.3465805193222, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 2.4615384615384616e-07, | |
| "loss": 0.0802, | |
| "step": 404 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "step": 407, | |
| "total_flos": 6.36398180352e+17, | |
| "train_loss": 0.35074408769753995, | |
| "train_runtime": 2707.3827, | |
| "train_samples_per_second": 9.621, | |
| "train_steps_per_second": 0.15 | |
| } | |
| ], | |
| "max_steps": 407, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 6.36398180352e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |