| { | |
| "best_global_step": null, | |
| "best_metric": 0.47698996707825364, | |
| "best_model_checkpoint": "/raid/p-storage/slp01_565a7357/bagas-fine-tune-whisper/whisper-tiny-javanese-openslr-v2/checkpoint-500", | |
| "epoch": 4.32152117545376, | |
| "eval_steps": 500, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17286084701815038, | |
| "grad_norm": 5.079843521118164, | |
| "learning_rate": 7.840000000000001e-06, | |
| "loss": 2.5618, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.34572169403630076, | |
| "grad_norm": 4.103299617767334, | |
| "learning_rate": 1.584e-05, | |
| "loss": 1.021, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5185825410544511, | |
| "grad_norm": 3.8726439476013184, | |
| "learning_rate": 1.9573333333333335e-05, | |
| "loss": 0.7363, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6914433880726015, | |
| "grad_norm": 3.9386720657348633, | |
| "learning_rate": 1.8684444444444446e-05, | |
| "loss": 0.6066, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8643042350907519, | |
| "grad_norm": 3.273592472076416, | |
| "learning_rate": 1.7795555555555557e-05, | |
| "loss": 0.528, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8643042350907519, | |
| "eval_loss": 0.4466875493526459, | |
| "eval_runtime": 5073.2528, | |
| "eval_samples_per_second": 3.647, | |
| "eval_steps_per_second": 0.228, | |
| "eval_wer": 0.47698996707825364, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0363007778738116, | |
| "grad_norm": 3.0979788303375244, | |
| "learning_rate": 1.690666666666667e-05, | |
| "loss": 0.4728, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.2091616248919619, | |
| "grad_norm": 3.1844944953918457, | |
| "learning_rate": 1.601777777777778e-05, | |
| "loss": 0.4222, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.3820224719101124, | |
| "grad_norm": 2.981661319732666, | |
| "learning_rate": 1.5128888888888891e-05, | |
| "loss": 0.404, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.5548833189282627, | |
| "grad_norm": 3.146735906600952, | |
| "learning_rate": 1.4240000000000001e-05, | |
| "loss": 0.3867, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.727744165946413, | |
| "grad_norm": 2.873340368270874, | |
| "learning_rate": 1.3351111111111112e-05, | |
| "loss": 0.3702, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.727744165946413, | |
| "eval_loss": 0.34238535165786743, | |
| "eval_runtime": 6401.1103, | |
| "eval_samples_per_second": 2.891, | |
| "eval_steps_per_second": 0.181, | |
| "eval_wer": 0.5528429218245372, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.902333621434745, | |
| "grad_norm": 2.7809109687805176, | |
| "learning_rate": 1.2462222222222222e-05, | |
| "loss": 0.3611, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.076058772687986, | |
| "grad_norm": 2.62384295463562, | |
| "learning_rate": 1.1573333333333335e-05, | |
| "loss": 0.3371, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.2489196197061365, | |
| "grad_norm": 2.793154716491699, | |
| "learning_rate": 1.0684444444444446e-05, | |
| "loss": 0.3134, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.421780466724287, | |
| "grad_norm": 2.5597171783447266, | |
| "learning_rate": 9.795555555555556e-06, | |
| "loss": 0.3069, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.5946413137424376, | |
| "grad_norm": 2.6140079498291016, | |
| "learning_rate": 8.906666666666667e-06, | |
| "loss": 0.2988, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.5946413137424376, | |
| "eval_loss": 0.30312925577163696, | |
| "eval_runtime": 2962.721, | |
| "eval_samples_per_second": 6.246, | |
| "eval_steps_per_second": 0.391, | |
| "eval_wer": 0.5552201690660703, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.767502160760588, | |
| "grad_norm": 2.635817050933838, | |
| "learning_rate": 8.017777777777779e-06, | |
| "loss": 0.2948, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.940363007778738, | |
| "grad_norm": 2.918149709701538, | |
| "learning_rate": 7.12888888888889e-06, | |
| "loss": 0.2903, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.1123595505617976, | |
| "grad_norm": 2.2907419204711914, | |
| "learning_rate": 6.24e-06, | |
| "loss": 0.2711, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.2852203975799483, | |
| "grad_norm": 2.640625238418579, | |
| "learning_rate": 5.351111111111112e-06, | |
| "loss": 0.2667, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.4580812445980986, | |
| "grad_norm": 2.5180225372314453, | |
| "learning_rate": 4.462222222222223e-06, | |
| "loss": 0.2607, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.4580812445980986, | |
| "eval_loss": 0.28587546944618225, | |
| "eval_runtime": 2830.646, | |
| "eval_samples_per_second": 6.537, | |
| "eval_steps_per_second": 0.409, | |
| "eval_wer": 0.6484880238350316, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.630942091616249, | |
| "grad_norm": 2.6452524662017822, | |
| "learning_rate": 3.5733333333333336e-06, | |
| "loss": 0.2637, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.8038029386343992, | |
| "grad_norm": 2.3364474773406982, | |
| "learning_rate": 2.6844444444444445e-06, | |
| "loss": 0.2615, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.9766637856525495, | |
| "grad_norm": 2.414165496826172, | |
| "learning_rate": 1.7955555555555556e-06, | |
| "loss": 0.2603, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.148660328435609, | |
| "grad_norm": 2.3572444915771484, | |
| "learning_rate": 9.066666666666668e-07, | |
| "loss": 0.2483, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.32152117545376, | |
| "grad_norm": 2.6238534450531006, | |
| "learning_rate": 1.777777777777778e-08, | |
| "loss": 0.2481, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.32152117545376, | |
| "eval_loss": 0.2792496979236603, | |
| "eval_runtime": 2806.8606, | |
| "eval_samples_per_second": 6.592, | |
| "eval_steps_per_second": 0.412, | |
| "eval_wer": 0.6471586421539112, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.32152117545376, | |
| "step": 2500, | |
| "total_flos": 1.574545111842816e+19, | |
| "train_loss": 0.17131386260986328, | |
| "train_runtime": 60097.0332, | |
| "train_samples_per_second": 10.649, | |
| "train_steps_per_second": 0.042 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.574545111842816e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |