| { | |
| "best_metric": 6.531481563650885, | |
| "best_model_checkpoint": "whisper-stakwork-crypto/checkpoint-150", | |
| "epoch": 38.0025, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 3.0000000000000004e-08, | |
| "loss": 0.8208, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 8e-08, | |
| "loss": 0.8146, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 1.3e-07, | |
| "loss": 0.7898, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.8e-07, | |
| "loss": 0.7761, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 2.3e-07, | |
| "loss": 0.7348, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.7377644777297974, | |
| "eval_runtime": 118.0826, | |
| "eval_samples_per_second": 2.32, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 9.30913608732307, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 2.8e-07, | |
| "loss": 0.7129, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 3.3e-07, | |
| "loss": 0.6506, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 3.8000000000000007e-07, | |
| "loss": 0.6193, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 4.3e-07, | |
| "loss": 0.574, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 4.800000000000001e-07, | |
| "loss": 0.5276, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_loss": 0.5290005207061768, | |
| "eval_runtime": 121.5574, | |
| "eval_samples_per_second": 2.254, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 8.612503882504326, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 5.3e-07, | |
| "loss": 0.4652, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 5.8e-07, | |
| "loss": 0.4235, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 6.3e-07, | |
| "loss": 0.3972, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 6.8e-07, | |
| "loss": 0.3675, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 7.300000000000001e-07, | |
| "loss": 0.3585, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.3874845504760742, | |
| "eval_runtime": 130.7132, | |
| "eval_samples_per_second": 2.096, | |
| "eval_steps_per_second": 0.138, | |
| "eval_wer": 6.531481563650885, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 7.8e-07, | |
| "loss": 0.3306, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 8.3e-07, | |
| "loss": 0.3255, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 8.8e-07, | |
| "loss": 0.307, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 9.3e-07, | |
| "loss": 0.3035, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 9.8e-07, | |
| "loss": 0.2924, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "eval_loss": 0.35476040840148926, | |
| "eval_runtime": 120.7053, | |
| "eval_samples_per_second": 2.27, | |
| "eval_steps_per_second": 0.149, | |
| "eval_wer": 6.677907441096862, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 1.03e-06, | |
| "loss": 0.2826, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 1.08e-06, | |
| "loss": 0.267, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 1.13e-06, | |
| "loss": 0.27, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 1.18e-06, | |
| "loss": 0.2492, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 1.2299999999999999e-06, | |
| "loss": 0.2506, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "eval_loss": 0.33637768030166626, | |
| "eval_runtime": 118.1569, | |
| "eval_samples_per_second": 2.319, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 6.788836136131694, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 1.28e-06, | |
| "loss": 0.2263, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 1.3300000000000002e-06, | |
| "loss": 0.23, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 1.3800000000000001e-06, | |
| "loss": 0.2102, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 1.43e-06, | |
| "loss": 0.2093, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 1.48e-06, | |
| "loss": 0.1946, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "eval_loss": 0.32616764307022095, | |
| "eval_runtime": 119.3365, | |
| "eval_samples_per_second": 2.296, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 7.148245108044549, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 1.53e-06, | |
| "loss": 0.1848, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 1.58e-06, | |
| "loss": 0.1752, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 15.02, | |
| "learning_rate": 1.63e-06, | |
| "loss": 0.1659, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 1.6800000000000002e-06, | |
| "loss": 0.1579, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "learning_rate": 1.73e-06, | |
| "loss": 0.1411, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 16.02, | |
| "eval_loss": 0.33289629220962524, | |
| "eval_runtime": 117.7955, | |
| "eval_samples_per_second": 2.326, | |
| "eval_steps_per_second": 0.153, | |
| "eval_wer": 7.210365177264055, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.7800000000000001e-06, | |
| "loss": 0.1374, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 1.83e-06, | |
| "loss": 0.1223, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.8800000000000002e-06, | |
| "loss": 0.1173, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 1.93e-06, | |
| "loss": 0.1034, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 1.98e-06, | |
| "loss": 0.1005, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.34216001629829407, | |
| "eval_runtime": 121.3922, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 7.556462705772729, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.01, | |
| "learning_rate": 2.03e-06, | |
| "loss": 0.0894, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 19.03, | |
| "learning_rate": 2.08e-06, | |
| "loss": 0.0786, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 20.01, | |
| "learning_rate": 2.13e-06, | |
| "loss": 0.0694, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 20.02, | |
| "learning_rate": 2.1800000000000003e-06, | |
| "loss": 0.066, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 2.23e-06, | |
| "loss": 0.0535, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "eval_loss": 0.35322892665863037, | |
| "eval_runtime": 120.2632, | |
| "eval_samples_per_second": 2.278, | |
| "eval_steps_per_second": 0.15, | |
| "eval_wer": 7.179305142654302, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 21.02, | |
| "learning_rate": 2.28e-06, | |
| "loss": 0.0494, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 22.01, | |
| "learning_rate": 2.3299999999999997e-06, | |
| "loss": 0.043, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "learning_rate": 2.38e-06, | |
| "loss": 0.0359, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "learning_rate": 2.43e-06, | |
| "loss": 0.0311, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "learning_rate": 2.48e-06, | |
| "loss": 0.0259, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "eval_loss": 0.34564533829689026, | |
| "eval_runtime": 119.3022, | |
| "eval_samples_per_second": 2.297, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 7.512091227758797, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 24.01, | |
| "learning_rate": 2.5300000000000003e-06, | |
| "loss": 0.0222, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "learning_rate": 2.58e-06, | |
| "loss": 0.0186, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 25.01, | |
| "learning_rate": 2.6300000000000002e-06, | |
| "loss": 0.0154, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 25.02, | |
| "learning_rate": 2.68e-06, | |
| "loss": 0.0154, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 2.73e-06, | |
| "loss": 0.0137, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.3586655259132385, | |
| "eval_runtime": 121.9747, | |
| "eval_samples_per_second": 2.246, | |
| "eval_steps_per_second": 0.148, | |
| "eval_wer": 7.654079957403381, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 26.02, | |
| "learning_rate": 2.78e-06, | |
| "loss": 0.0121, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 2.83e-06, | |
| "loss": 0.0113, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 27.02, | |
| "learning_rate": 2.88e-06, | |
| "loss": 0.0103, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 2.9300000000000003e-06, | |
| "loss": 0.0112, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 28.02, | |
| "learning_rate": 2.98e-06, | |
| "loss": 0.0078, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 28.02, | |
| "eval_loss": 0.3590793013572693, | |
| "eval_runtime": 125.6972, | |
| "eval_samples_per_second": 2.18, | |
| "eval_steps_per_second": 0.143, | |
| "eval_wer": 7.352353906908639, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 2.91e-06, | |
| "loss": 0.0074, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "learning_rate": 2.7600000000000003e-06, | |
| "loss": 0.0065, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "learning_rate": 2.61e-06, | |
| "loss": 0.0061, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 30.01, | |
| "learning_rate": 2.4599999999999997e-06, | |
| "loss": 0.0053, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 30.02, | |
| "learning_rate": 2.31e-06, | |
| "loss": 0.0041, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 30.02, | |
| "eval_loss": 0.3672121465206146, | |
| "eval_runtime": 119.005, | |
| "eval_samples_per_second": 2.302, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 7.303545281093313, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 31.01, | |
| "learning_rate": 2.16e-06, | |
| "loss": 0.0035, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 31.02, | |
| "learning_rate": 2.0100000000000002e-06, | |
| "loss": 0.0032, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 32.01, | |
| "learning_rate": 1.86e-06, | |
| "loss": 0.0029, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 32.02, | |
| "learning_rate": 1.71e-06, | |
| "loss": 0.0028, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 33.01, | |
| "learning_rate": 1.56e-06, | |
| "loss": 0.0026, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 33.01, | |
| "eval_loss": 0.3961912989616394, | |
| "eval_runtime": 131.1922, | |
| "eval_samples_per_second": 2.089, | |
| "eval_steps_per_second": 0.137, | |
| "eval_wer": 7.321293872298887, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "learning_rate": 1.41e-06, | |
| "loss": 0.0026, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 34.01, | |
| "learning_rate": 1.26e-06, | |
| "loss": 0.0024, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 34.02, | |
| "learning_rate": 1.11e-06, | |
| "loss": 0.0024, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 35.01, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 0.0026, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 35.02, | |
| "learning_rate": 8.100000000000001e-07, | |
| "loss": 0.0022, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 35.02, | |
| "eval_loss": 0.399670273065567, | |
| "eval_runtime": 119.4156, | |
| "eval_samples_per_second": 2.295, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 7.352353906908639, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 36.01, | |
| "learning_rate": 6.6e-07, | |
| "loss": 0.0023, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 36.02, | |
| "learning_rate": 5.100000000000001e-07, | |
| "loss": 0.0021, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "learning_rate": 3.6e-07, | |
| "loss": 0.0022, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 37.02, | |
| "learning_rate": 2.1000000000000003e-07, | |
| "loss": 0.0021, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 6.000000000000001e-08, | |
| "loss": 0.0022, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 0.4024691879749298, | |
| "eval_runtime": 131.5096, | |
| "eval_samples_per_second": 2.083, | |
| "eval_steps_per_second": 0.137, | |
| "eval_wer": 7.330168167901673, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "step": 800, | |
| "total_flos": 2.570087278116864e+19, | |
| "train_loss": 0.18918936033500358, | |
| "train_runtime": 13916.8243, | |
| "train_samples_per_second": 1.84, | |
| "train_steps_per_second": 0.057 | |
| } | |
| ], | |
| "max_steps": 800, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 2.570087278116864e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |