| { | |
| "best_metric": 7.639680729760548, | |
| "best_model_checkpoint": "../whisper-NST2/checkpoint-8000", | |
| "epoch": 1.3959, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.976e-05, | |
| "loss": 0.9006, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.9760000000000006e-05, | |
| "loss": 0.2407, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3.896e-05, | |
| "loss": 0.2145, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 3.790736842105263e-05, | |
| "loss": 0.1846, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.34600791335105896, | |
| "eval_runtime": 44.1083, | |
| "eval_samples_per_second": 2.267, | |
| "eval_steps_per_second": 0.159, | |
| "eval_wer": 14.937286202964653, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 3.6854736842105265e-05, | |
| "loss": 0.1663, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.58021052631579e-05, | |
| "loss": 0.1567, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.474947368421053e-05, | |
| "loss": 0.1449, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.369684210526316e-05, | |
| "loss": 0.1325, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.34131211042404175, | |
| "eval_runtime": 45.4411, | |
| "eval_samples_per_second": 2.201, | |
| "eval_steps_per_second": 0.154, | |
| "eval_wer": 11.402508551881414, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 3.2644210526315796e-05, | |
| "loss": 0.1292, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3.159157894736842e-05, | |
| "loss": 0.1225, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.0538947368421055e-05, | |
| "loss": 0.1168, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.9486315789473688e-05, | |
| "loss": 0.1135, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.3427869379520416, | |
| "eval_runtime": 43.6701, | |
| "eval_samples_per_second": 2.29, | |
| "eval_steps_per_second": 0.16, | |
| "eval_wer": 12.656784492588368, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2.843368421052632e-05, | |
| "loss": 0.1074, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 2.738105263157895e-05, | |
| "loss": 0.1039, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 2.6328421052631582e-05, | |
| "loss": 0.0973, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.5275789473684215e-05, | |
| "loss": 0.0955, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.31404200196266174, | |
| "eval_runtime": 43.7248, | |
| "eval_samples_per_second": 2.287, | |
| "eval_steps_per_second": 0.16, | |
| "eval_wer": 10.718358038768528, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 2.4223157894736848e-05, | |
| "loss": 0.0914, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 2.3170526315789474e-05, | |
| "loss": 0.0878, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.2117894736842107e-05, | |
| "loss": 0.0857, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.106526315789474e-05, | |
| "loss": 0.0871, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.2906649112701416, | |
| "eval_runtime": 53.0801, | |
| "eval_samples_per_second": 1.884, | |
| "eval_steps_per_second": 0.132, | |
| "eval_wer": 9.464082098061574, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.001263157894737e-05, | |
| "loss": 0.0806, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.896e-05, | |
| "loss": 0.078, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1.7907368421052634e-05, | |
| "loss": 0.0779, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.6854736842105264e-05, | |
| "loss": 0.0774, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.3019491136074066, | |
| "eval_runtime": 49.1546, | |
| "eval_samples_per_second": 2.034, | |
| "eval_steps_per_second": 0.142, | |
| "eval_wer": 11.402508551881414, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.5802105263157896e-05, | |
| "loss": 0.0674, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.4749473684210527e-05, | |
| "loss": 0.0526, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.369684210526316e-05, | |
| "loss": 0.0451, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.264421052631579e-05, | |
| "loss": 0.041, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.2897065281867981, | |
| "eval_runtime": 42.2915, | |
| "eval_samples_per_second": 2.365, | |
| "eval_steps_per_second": 0.166, | |
| "eval_wer": 9.007981755986318, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1.1591578947368422e-05, | |
| "loss": 0.0381, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.0538947368421053e-05, | |
| "loss": 0.0365, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 9.486315789473684e-06, | |
| "loss": 0.0343, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 8.433684210526317e-06, | |
| "loss": 0.0306, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.30130401253700256, | |
| "eval_runtime": 48.5193, | |
| "eval_samples_per_second": 2.061, | |
| "eval_steps_per_second": 0.144, | |
| "eval_wer": 7.639680729760548, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 7.385263157894737e-06, | |
| "loss": 0.0294, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 6.332631578947369e-06, | |
| "loss": 0.0288, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 5.28e-06, | |
| "loss": 0.0276, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 4.227368421052631e-06, | |
| "loss": 0.0279, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.29579514265060425, | |
| "eval_runtime": 42.0972, | |
| "eval_samples_per_second": 2.375, | |
| "eval_steps_per_second": 0.166, | |
| "eval_wer": 9.122006841505131, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.174736842105263e-06, | |
| "loss": 0.0256, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.1221052631578947e-06, | |
| "loss": 0.0256, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.0694736842105266e-06, | |
| "loss": 0.0226, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.6842105263157897e-08, | |
| "loss": 0.0239, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_loss": 0.29898393154144287, | |
| "eval_runtime": 56.0138, | |
| "eval_samples_per_second": 1.785, | |
| "eval_steps_per_second": 0.125, | |
| "eval_wer": 7.753705815279361, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "step": 10000, | |
| "total_flos": 1.3851118024998912e+20, | |
| "train_loss": 0.10624286437034607, | |
| "train_runtime": 115360.2706, | |
| "train_samples_per_second": 8.322, | |
| "train_steps_per_second": 0.087 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.3851118024998912e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |