| { | |
| "best_metric": 10.718358038768528, | |
| "best_model_checkpoint": "../whisper-NST-cons2e5/checkpoint-6000", | |
| "epoch": 1.0, | |
| "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": 9.88e-06, | |
| "loss": 1.2287, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.9880000000000003e-05, | |
| "loss": 0.2979, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2728, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2517, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.4131389260292053, | |
| "eval_runtime": 51.3583, | |
| "eval_samples_per_second": 1.947, | |
| "eval_steps_per_second": 0.136, | |
| "eval_wer": 18.47206385404789, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2254, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2115, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2018, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1931, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.3531236946582794, | |
| "eval_runtime": 74.7257, | |
| "eval_samples_per_second": 1.338, | |
| "eval_steps_per_second": 0.094, | |
| "eval_wer": 19.04218928164196, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1786, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1746, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1726, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1598, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.3605496883392334, | |
| "eval_runtime": 59.9526, | |
| "eval_samples_per_second": 1.668, | |
| "eval_steps_per_second": 0.117, | |
| "eval_wer": 16.875712656784494, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1636, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 2e-05, | |
| "loss": 0.158, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1494, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1541, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.33670979738235474, | |
| "eval_runtime": 50.4438, | |
| "eval_samples_per_second": 1.982, | |
| "eval_steps_per_second": 0.139, | |
| "eval_wer": 14.481185860889395, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1498, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1486, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1471, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1443, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.3274496793746948, | |
| "eval_runtime": 55.4869, | |
| "eval_samples_per_second": 1.802, | |
| "eval_steps_per_second": 0.126, | |
| "eval_wer": 13.340935005701255, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1337, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1358, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1232, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1301, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.34810879826545715, | |
| "eval_runtime": 52.0858, | |
| "eval_samples_per_second": 1.92, | |
| "eval_steps_per_second": 0.134, | |
| "eval_wer": 10.718358038768528, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1308, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 2e-05, | |
| "loss": 0.121, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1343, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1266, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.3451653718948364, | |
| "eval_runtime": 52.2228, | |
| "eval_samples_per_second": 1.915, | |
| "eval_steps_per_second": 0.134, | |
| "eval_wer": 12.998859749144811, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1223, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1264, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 2e-05, | |
| "loss": 0.119, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1216, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.32154810428619385, | |
| "eval_runtime": 82.8333, | |
| "eval_samples_per_second": 1.207, | |
| "eval_steps_per_second": 0.085, | |
| "eval_wer": 10.832383124287343, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1242, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1217, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1245, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1121, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.31604325771331787, | |
| "eval_runtime": 50.7948, | |
| "eval_samples_per_second": 1.969, | |
| "eval_steps_per_second": 0.138, | |
| "eval_wer": 11.516533637400228, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1161, | |
| "step": 9250 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2e-05, | |
| "loss": 0.112, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1167, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1171, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.3520647883415222, | |
| "eval_runtime": 84.9296, | |
| "eval_samples_per_second": 1.177, | |
| "eval_steps_per_second": 0.082, | |
| "eval_wer": 11.85860889395667, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 10000, | |
| "total_flos": 4.61736640512e+19, | |
| "train_loss": 0.18131806907653808, | |
| "train_runtime": 89475.4174, | |
| "train_samples_per_second": 3.576, | |
| "train_steps_per_second": 0.112 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 4.61736640512e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |