| { | |
| "best_metric": 61.34371957156767, | |
| "best_model_checkpoint": "../models/whisper-tiny-te/checkpoint-3000", | |
| "epoch": 1.031, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.94e-06, | |
| "loss": 2.1902, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3.94e-06, | |
| "loss": 1.4572, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 5.94e-06, | |
| "loss": 1.3178, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.4030134677886963, | |
| "eval_runtime": 61.0962, | |
| "eval_samples_per_second": 1.637, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 100.82765335929894, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 7.94e-06, | |
| "loss": 1.2405, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.940000000000001e-06, | |
| "loss": 1.1747, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.616e-06, | |
| "loss": 1.1082, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 1.2772541046142578, | |
| "eval_runtime": 60.8796, | |
| "eval_samples_per_second": 1.643, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 101.65530671859786, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 9.216000000000001e-06, | |
| "loss": 1.0402, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 8.816000000000002e-06, | |
| "loss": 0.9201, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 8.416e-06, | |
| "loss": 0.7341, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.6280691623687744, | |
| "eval_runtime": 61.6707, | |
| "eval_samples_per_second": 1.622, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 84.85881207400195, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 8.016e-06, | |
| "loss": 0.6138, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 7.616000000000001e-06, | |
| "loss": 0.5609, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 7.216000000000001e-06, | |
| "loss": 0.5161, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.4645693898200989, | |
| "eval_runtime": 61.4927, | |
| "eval_samples_per_second": 1.626, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 69.62025316455697, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 6.8160000000000005e-06, | |
| "loss": 0.489, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 6.416e-06, | |
| "loss": 0.4683, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 6.0160000000000005e-06, | |
| "loss": 0.4468, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.42932942509651184, | |
| "eval_runtime": 60.8369, | |
| "eval_samples_per_second": 1.644, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 67.28334956183058, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 5.616e-06, | |
| "loss": 0.4331, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 5.216e-06, | |
| "loss": 0.4202, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.816e-06, | |
| "loss": 0.41, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.41048839688301086, | |
| "eval_runtime": 60.1608, | |
| "eval_samples_per_second": 1.662, | |
| "eval_steps_per_second": 0.05, | |
| "eval_wer": 63.583252190847126, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.416000000000001e-06, | |
| "loss": 0.3992, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.016e-06, | |
| "loss": 0.3911, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 3.616e-06, | |
| "loss": 0.3865, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.3972886800765991, | |
| "eval_runtime": 62.6118, | |
| "eval_samples_per_second": 1.597, | |
| "eval_steps_per_second": 0.048, | |
| "eval_wer": 61.92794547224927, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 3.216e-06, | |
| "loss": 0.3813, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 2.8160000000000002e-06, | |
| "loss": 0.3729, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.4160000000000002e-06, | |
| "loss": 0.3672, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.3908187747001648, | |
| "eval_runtime": 61.0288, | |
| "eval_samples_per_second": 1.639, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 61.53846153846154, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 2.0160000000000003e-06, | |
| "loss": 0.3623, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.616e-06, | |
| "loss": 0.3613, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.216e-06, | |
| "loss": 0.3627, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.38463568687438965, | |
| "eval_runtime": 61.1228, | |
| "eval_samples_per_second": 1.636, | |
| "eval_steps_per_second": 0.049, | |
| "eval_wer": 61.39240506329114, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 8.160000000000001e-07, | |
| "loss": 0.3545, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 4.16e-07, | |
| "loss": 0.3525, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.6e-08, | |
| "loss": 0.3535, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_loss": 0.3824814558029175, | |
| "eval_runtime": 60.5274, | |
| "eval_samples_per_second": 1.652, | |
| "eval_steps_per_second": 0.05, | |
| "eval_wer": 61.34371957156767, | |
| "step": 3000 | |
| } | |
| ], | |
| "max_steps": 3000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 5.31684214382592e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |