| { | |
| "best_metric": 69.98899144720129, | |
| "best_model_checkpoint": "./whisper-tiny-hi2_test/checkpoint-120", | |
| "epoch": 6.46875, | |
| "eval_steps": 40, | |
| "global_step": 207, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 10.95101547241211, | |
| "learning_rate": 9.200000000000002e-06, | |
| "loss": 2.3747, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 7.747081279754639, | |
| "learning_rate": 8.2e-06, | |
| "loss": 1.2169, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 1.0360029935836792, | |
| "eval_runtime": 574.5339, | |
| "eval_samples_per_second": 1.741, | |
| "eval_steps_per_second": 1.741, | |
| "eval_wer": 91.60809552036582, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 7.32460355758667, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 0.9058, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 6.835532188415527, | |
| "learning_rate": 6.200000000000001e-06, | |
| "loss": 0.7302, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 0.7368654608726501, | |
| "eval_runtime": 549.562, | |
| "eval_samples_per_second": 1.82, | |
| "eval_steps_per_second": 1.82, | |
| "eval_wer": 79.80353967313066, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 6.408384323120117, | |
| "learning_rate": 5.2e-06, | |
| "loss": 0.6453, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 5.809150695800781, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "loss": 0.5917, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "eval_loss": 0.6504533886909485, | |
| "eval_runtime": 548.8688, | |
| "eval_samples_per_second": 1.822, | |
| "eval_steps_per_second": 1.822, | |
| "eval_wer": 69.98899144720129, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 5.724637508392334, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 0.5356, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 10.124310493469238, | |
| "learning_rate": 2.2e-06, | |
| "loss": 0.5156, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.6157090663909912, | |
| "eval_runtime": 549.7895, | |
| "eval_samples_per_second": 1.819, | |
| "eval_steps_per_second": 1.819, | |
| "eval_wer": 70.4293335591498, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.625, | |
| "grad_norm": 4.947811126708984, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 0.4986, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 4.541861057281494, | |
| "learning_rate": 2.0000000000000002e-07, | |
| "loss": 0.481, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "eval_loss": 0.6050636768341064, | |
| "eval_runtime": 552.964, | |
| "eval_samples_per_second": 1.808, | |
| "eval_steps_per_second": 1.808, | |
| "eval_wer": 70.03980015242611, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "step": 200, | |
| "total_flos": 1.5401574531072e+17, | |
| "train_loss": 0.8495243072509766, | |
| "train_runtime": 3592.94, | |
| "train_samples_per_second": 1.781, | |
| "train_steps_per_second": 0.056 | |
| }, | |
| { | |
| "epoch": 6.28125, | |
| "step": 201, | |
| "total_flos": 1.5480354963456e+17, | |
| "train_loss": 0.003230639654605543, | |
| "train_runtime": 7.5043, | |
| "train_samples_per_second": 852.844, | |
| "train_steps_per_second": 26.651 | |
| }, | |
| { | |
| "epoch": 6.28125, | |
| "eval_loss": 0.6504533886909485, | |
| "eval_runtime": 548.5863, | |
| "eval_samples_per_second": 1.823, | |
| "eval_steps_per_second": 1.823, | |
| "eval_wer": 69.98899144720129, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 6.3125, | |
| "step": 202, | |
| "total_flos": 1.555913539584e+17, | |
| "train_loss": 0.0035205459830784562, | |
| "train_runtime": 7.6486, | |
| "train_samples_per_second": 836.753, | |
| "train_steps_per_second": 26.149 | |
| }, | |
| { | |
| "epoch": 6.3125, | |
| "eval_loss": 0.6504533886909485, | |
| "eval_runtime": 548.2984, | |
| "eval_samples_per_second": 1.824, | |
| "eval_steps_per_second": 1.824, | |
| "eval_wer": 69.98899144720129, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 6.34375, | |
| "step": 203, | |
| "total_flos": 1.5637915828224e+17, | |
| "train_loss": 0.002875891812329222, | |
| "train_runtime": 7.5227, | |
| "train_samples_per_second": 850.756, | |
| "train_steps_per_second": 26.586 | |
| }, | |
| { | |
| "epoch": 6.34375, | |
| "eval_loss": 0.6504533886909485, | |
| "eval_runtime": 547.5722, | |
| "eval_samples_per_second": 1.826, | |
| "eval_steps_per_second": 1.826, | |
| "eval_wer": 69.98899144720129, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 6.375, | |
| "step": 204, | |
| "total_flos": 1.5716696260608e+17, | |
| "train_loss": 0.003042735305486941, | |
| "train_runtime": 7.3464, | |
| "train_samples_per_second": 871.18, | |
| "train_steps_per_second": 27.224 | |
| }, | |
| { | |
| "epoch": 6.375, | |
| "eval_loss": 0.6504533886909485, | |
| "eval_runtime": 547.015, | |
| "eval_samples_per_second": 1.828, | |
| "eval_steps_per_second": 1.828, | |
| "eval_wer": 69.98899144720129, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.40625, | |
| "step": 205, | |
| "total_flos": 1.5795476692992e+17, | |
| "train_loss": 0.0033008325390699433, | |
| "train_runtime": 7.3265, | |
| "train_samples_per_second": 873.547, | |
| "train_steps_per_second": 27.298 | |
| }, | |
| { | |
| "epoch": 6.40625, | |
| "eval_loss": 0.6504533886909485, | |
| "eval_runtime": 545.7731, | |
| "eval_samples_per_second": 1.832, | |
| "eval_steps_per_second": 1.832, | |
| "eval_wer": 69.98899144720129, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 6.4375, | |
| "step": 206, | |
| "total_flos": 1.5874257125376e+17, | |
| "train_loss": 0.0030977731769524734, | |
| "train_runtime": 7.5608, | |
| "train_samples_per_second": 846.467, | |
| "train_steps_per_second": 26.452 | |
| }, | |
| { | |
| "epoch": 6.4375, | |
| "eval_loss": 0.6504533886909485, | |
| "eval_runtime": 546.3323, | |
| "eval_samples_per_second": 1.83, | |
| "eval_steps_per_second": 1.83, | |
| "eval_wer": 69.98899144720129, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 6.46875, | |
| "step": 207, | |
| "total_flos": 1.595303755776e+17, | |
| "train_loss": 0.0028965755361289795, | |
| "train_runtime": 7.5526, | |
| "train_samples_per_second": 847.394, | |
| "train_steps_per_second": 26.481 | |
| }, | |
| { | |
| "epoch": 6.46875, | |
| "eval_loss": 0.6504533886909485, | |
| "eval_runtime": 546.1485, | |
| "eval_samples_per_second": 1.831, | |
| "eval_steps_per_second": 1.831, | |
| "eval_wer": 69.98899144720129, | |
| "step": 207 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 40, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.595303755776e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |