| { | |
| "best_metric": 18.863785917964464, | |
| "best_model_checkpoint": "./checkpoint-400", | |
| "epoch": 87.00571428571429, | |
| "global_step": 1400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 4.4e-07, | |
| "loss": 1.9798, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.400000000000001e-07, | |
| "loss": 1.2092, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 1.44e-06, | |
| "loss": 0.4287, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 1.94e-06, | |
| "loss": 0.1862, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 2.4400000000000004e-06, | |
| "loss": 0.089, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.9400000000000002e-06, | |
| "loss": 0.0359, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 3.44e-06, | |
| "loss": 0.0119, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 3.94e-06, | |
| "loss": 0.0063, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "eval_loss": 0.5237645506858826, | |
| "eval_runtime": 642.7046, | |
| "eval_samples_per_second": 1.925, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 19.291511296336918, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 4.440000000000001e-06, | |
| "loss": 0.0044, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 4.94e-06, | |
| "loss": 0.003, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 5.4400000000000004e-06, | |
| "loss": 0.0051, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "learning_rate": 5.94e-06, | |
| "loss": 0.0044, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 6.440000000000001e-06, | |
| "loss": 0.0045, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 6.9400000000000005e-06, | |
| "loss": 0.0041, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 7.440000000000001e-06, | |
| "loss": 0.0042, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 24.01, | |
| "learning_rate": 7.94e-06, | |
| "loss": 0.0046, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 24.01, | |
| "eval_loss": 0.5686367750167847, | |
| "eval_runtime": 644.5254, | |
| "eval_samples_per_second": 1.919, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 18.863785917964464, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 26.01, | |
| "learning_rate": 8.44e-06, | |
| "loss": 0.0042, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 8.94e-06, | |
| "loss": 0.0078, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "learning_rate": 9.440000000000001e-06, | |
| "loss": 0.01, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "learning_rate": 9.940000000000001e-06, | |
| "loss": 0.0112, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 32.01, | |
| "learning_rate": 9.755555555555556e-06, | |
| "loss": 0.0147, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 9.47777777777778e-06, | |
| "loss": 0.01, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 35.01, | |
| "learning_rate": 9.200000000000002e-06, | |
| "loss": 0.0092, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "learning_rate": 8.922222222222224e-06, | |
| "loss": 0.0067, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "eval_loss": 0.5923575758934021, | |
| "eval_runtime": 650.0985, | |
| "eval_samples_per_second": 1.903, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 20.607589383636764, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "learning_rate": 8.644444444444445e-06, | |
| "loss": 0.0062, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 40.01, | |
| "learning_rate": 8.366666666666667e-06, | |
| "loss": 0.0033, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "learning_rate": 8.08888888888889e-06, | |
| "loss": 0.0017, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "learning_rate": 7.811111111111111e-06, | |
| "loss": 0.0078, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "learning_rate": 7.533333333333334e-06, | |
| "loss": 0.0014, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 46.01, | |
| "learning_rate": 7.255555555555556e-06, | |
| "loss": 0.0017, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 48.01, | |
| "learning_rate": 6.977777777777779e-06, | |
| "loss": 0.0004, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 49.01, | |
| "learning_rate": 6.700000000000001e-06, | |
| "loss": 0.0004, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 49.01, | |
| "eval_loss": 0.6238680481910706, | |
| "eval_runtime": 645.2778, | |
| "eval_samples_per_second": 1.917, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 19.80697521386269, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 51.01, | |
| "learning_rate": 6.422222222222223e-06, | |
| "loss": 0.0005, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "learning_rate": 6.144444444444445e-06, | |
| "loss": 0.0003, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 54.01, | |
| "learning_rate": 5.8666666666666675e-06, | |
| "loss": 0.0003, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "learning_rate": 5.588888888888889e-06, | |
| "loss": 0.0002, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 57.01, | |
| "learning_rate": 5.311111111111111e-06, | |
| "loss": 0.0003, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "learning_rate": 5.033333333333333e-06, | |
| "loss": 0.0005, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 60.01, | |
| "learning_rate": 4.755555555555556e-06, | |
| "loss": 0.0008, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 62.01, | |
| "learning_rate": 4.477777777777778e-06, | |
| "loss": 0.0005, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 62.01, | |
| "eval_loss": 0.6354185938835144, | |
| "eval_runtime": 650.6168, | |
| "eval_samples_per_second": 1.901, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 19.76310594428603, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "loss": 0.0002, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 65.01, | |
| "learning_rate": 3.922222222222223e-06, | |
| "loss": 0.0003, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "learning_rate": 3.644444444444445e-06, | |
| "loss": 0.0002, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 68.01, | |
| "learning_rate": 3.366666666666667e-06, | |
| "loss": 0.0002, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "learning_rate": 3.088888888888889e-06, | |
| "loss": 0.0002, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 71.01, | |
| "learning_rate": 2.811111111111111e-06, | |
| "loss": 0.0002, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "learning_rate": 2.5333333333333338e-06, | |
| "loss": 0.0002, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 74.01, | |
| "learning_rate": 2.2555555555555557e-06, | |
| "loss": 0.0001, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 74.01, | |
| "eval_loss": 0.6446589231491089, | |
| "eval_runtime": 652.0364, | |
| "eval_samples_per_second": 1.897, | |
| "eval_steps_per_second": 0.12, | |
| "eval_wer": 19.554726913796884, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 76.01, | |
| "learning_rate": 1.977777777777778e-06, | |
| "loss": 0.0001, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "learning_rate": 1.7000000000000002e-06, | |
| "loss": 0.0001, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 79.01, | |
| "learning_rate": 1.4222222222222223e-06, | |
| "loss": 0.0001, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "learning_rate": 1.1444444444444446e-06, | |
| "loss": 0.0001, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 82.01, | |
| "learning_rate": 8.666666666666668e-07, | |
| "loss": 0.0001, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "learning_rate": 5.888888888888889e-07, | |
| "loss": 0.0001, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 85.01, | |
| "learning_rate": 3.111111111111111e-07, | |
| "loss": 0.0001, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 87.01, | |
| "learning_rate": 3.333333333333334e-08, | |
| "loss": 0.0001, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 87.01, | |
| "eval_loss": 0.6472771763801575, | |
| "eval_runtime": 647.0101, | |
| "eval_samples_per_second": 1.912, | |
| "eval_steps_per_second": 0.121, | |
| "eval_wer": 19.554726913796884, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 87.01, | |
| "step": 1400, | |
| "total_flos": 4.430241782562816e+19, | |
| "train_loss": 0.07293392582147915, | |
| "train_runtime": 12725.5944, | |
| "train_samples_per_second": 3.52, | |
| "train_steps_per_second": 0.11 | |
| } | |
| ], | |
| "max_steps": 1400, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 4.430241782562816e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |