| { | |
| "best_metric": 19.93811881188119, | |
| "best_model_checkpoint": "o0dimplz0o/Whisper-Large-v3-turbo-STT-Zeroth-KO-v2/checkpoint-4500", | |
| "epoch": 1.0513036164844407, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00021026072329688813, | |
| "grad_norm": 0.9765625, | |
| "learning_rate": 2e-08, | |
| "loss": 0.0149, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.10513036164844407, | |
| "grad_norm": 4.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0107, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.10513036164844407, | |
| "eval_cer": 0.10140471258415329, | |
| "eval_loss": 0.01961674913764, | |
| "eval_runtime": 150.7373, | |
| "eval_samples_per_second": 4.75, | |
| "eval_steps_per_second": 0.597, | |
| "eval_wer": 28.849009900990097, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.21026072329688814, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.0202, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.21026072329688814, | |
| "eval_cer": 0.0718539616778871, | |
| "eval_loss": 0.018409011885523796, | |
| "eval_runtime": 146.6316, | |
| "eval_samples_per_second": 4.883, | |
| "eval_steps_per_second": 0.614, | |
| "eval_wer": 21.806930693069308, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3153910849453322, | |
| "grad_norm": 0.5078125, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.0197, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3153910849453322, | |
| "eval_cer": 0.06975012946659762, | |
| "eval_loss": 0.017869649454951286, | |
| "eval_runtime": 146.8815, | |
| "eval_samples_per_second": 4.875, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 21.633663366336634, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.42052144659377627, | |
| "grad_norm": 1.5546875, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.0177, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.42052144659377627, | |
| "eval_cer": 0.11176204039357845, | |
| "eval_loss": 0.017242593690752983, | |
| "eval_runtime": 154.9659, | |
| "eval_samples_per_second": 4.62, | |
| "eval_steps_per_second": 0.581, | |
| "eval_wer": 26.683168316831683, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5256518082422204, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.0182, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5256518082422204, | |
| "eval_cer": 0.10477084412221647, | |
| "eval_loss": 0.016888286918401718, | |
| "eval_runtime": 155.4479, | |
| "eval_samples_per_second": 4.606, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 24.047029702970296, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.6307821698906644, | |
| "grad_norm": 1.859375, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.017, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6307821698906644, | |
| "eval_cer": 0.08761651993785603, | |
| "eval_loss": 0.0166668388992548, | |
| "eval_runtime": 158.8576, | |
| "eval_samples_per_second": 4.507, | |
| "eval_steps_per_second": 0.567, | |
| "eval_wer": 22.834158415841586, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7359125315391085, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.0174, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.7359125315391085, | |
| "eval_cer": 0.08716338684619368, | |
| "eval_loss": 0.01654214970767498, | |
| "eval_runtime": 151.2266, | |
| "eval_samples_per_second": 4.735, | |
| "eval_steps_per_second": 0.595, | |
| "eval_wer": 22.636138613861387, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.8410428931875525, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.0183, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.8410428931875525, | |
| "eval_cer": 0.08654842050750906, | |
| "eval_loss": 0.01647871918976307, | |
| "eval_runtime": 151.8997, | |
| "eval_samples_per_second": 4.714, | |
| "eval_steps_per_second": 0.592, | |
| "eval_wer": 22.5990099009901, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.9461732548359967, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.0168, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.9461732548359967, | |
| "eval_cer": 0.06573666494044536, | |
| "eval_loss": 0.016473352909088135, | |
| "eval_runtime": 147.2823, | |
| "eval_samples_per_second": 4.861, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 19.93811881188119, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.0513036164844407, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 0.0, | |
| "loss": 0.015, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.0513036164844407, | |
| "eval_cer": 0.08683972035214915, | |
| "eval_loss": 0.016479654237627983, | |
| "eval_runtime": 151.4679, | |
| "eval_samples_per_second": 4.727, | |
| "eval_steps_per_second": 0.594, | |
| "eval_wer": 22.54950495049505, | |
| "step": 5000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 1.704780657524736e+19, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |