| { | |
| "best_metric": 0.3861003861003861, | |
| "best_model_checkpoint": "D:\\development\\whisper-finetune\\output_tiny\\checkpoint-1000", | |
| "epoch": 1000.0, | |
| "eval_steps": 1000, | |
| "global_step": 15000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 9.88e-06, | |
| "loss": 0.8261, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 66.67, | |
| "learning_rate": 9.659310344827587e-06, | |
| "loss": 0.0028, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 66.67, | |
| "eval_loss": 0.002442890079692006, | |
| "eval_runtime": 21.9145, | |
| "eval_samples_per_second": 5.339, | |
| "eval_steps_per_second": 0.684, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 9.314482758620691e-06, | |
| "loss": 0.0016, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 133.33, | |
| "learning_rate": 8.969655172413794e-06, | |
| "loss": 0.0014, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 133.33, | |
| "eval_loss": 0.0016565920086577535, | |
| "eval_runtime": 21.8481, | |
| "eval_samples_per_second": 5.355, | |
| "eval_steps_per_second": 0.687, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 166.67, | |
| "learning_rate": 8.624827586206898e-06, | |
| "loss": 0.0014, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "learning_rate": 8.28e-06, | |
| "loss": 0.0012, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "eval_loss": 0.0012075488921254873, | |
| "eval_runtime": 21.8068, | |
| "eval_samples_per_second": 5.365, | |
| "eval_steps_per_second": 0.688, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 233.33, | |
| "learning_rate": 7.935172413793105e-06, | |
| "loss": 0.0012, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 266.67, | |
| "learning_rate": 7.590344827586208e-06, | |
| "loss": 0.0011, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 266.67, | |
| "eval_loss": 0.0011881846003234386, | |
| "eval_runtime": 22.3448, | |
| "eval_samples_per_second": 5.236, | |
| "eval_steps_per_second": 0.671, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 300.0, | |
| "learning_rate": 7.2468965517241385e-06, | |
| "loss": 0.0013, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 333.33, | |
| "learning_rate": 6.902068965517243e-06, | |
| "loss": 0.0011, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 333.33, | |
| "eval_loss": 0.0012107096845284104, | |
| "eval_runtime": 21.9233, | |
| "eval_samples_per_second": 5.337, | |
| "eval_steps_per_second": 0.684, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 366.67, | |
| "learning_rate": 6.557241379310345e-06, | |
| "loss": 0.0011, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "learning_rate": 6.2124137931034485e-06, | |
| "loss": 0.0011, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "eval_loss": 0.0011809396091848612, | |
| "eval_runtime": 21.8718, | |
| "eval_samples_per_second": 5.349, | |
| "eval_steps_per_second": 0.686, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 433.33, | |
| "learning_rate": 5.868275862068966e-06, | |
| "loss": 0.0011, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 466.67, | |
| "learning_rate": 5.523448275862069e-06, | |
| "loss": 0.001, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 466.67, | |
| "eval_loss": 0.0011794030433520675, | |
| "eval_runtime": 21.8283, | |
| "eval_samples_per_second": 5.36, | |
| "eval_steps_per_second": 0.687, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "learning_rate": 5.178620689655173e-06, | |
| "loss": 0.001, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 533.33, | |
| "learning_rate": 4.833793103448276e-06, | |
| "loss": 0.001, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 533.33, | |
| "eval_loss": 0.0011808406561613083, | |
| "eval_runtime": 21.7413, | |
| "eval_samples_per_second": 5.381, | |
| "eval_steps_per_second": 0.69, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 566.67, | |
| "learning_rate": 4.489655172413793e-06, | |
| "loss": 0.001, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 600.0, | |
| "learning_rate": 4.144827586206897e-06, | |
| "loss": 0.001, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 600.0, | |
| "eval_loss": 0.00120567309204489, | |
| "eval_runtime": 21.8405, | |
| "eval_samples_per_second": 5.357, | |
| "eval_steps_per_second": 0.687, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 633.33, | |
| "learning_rate": 3.8000000000000005e-06, | |
| "loss": 0.001, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 666.67, | |
| "learning_rate": 3.455172413793104e-06, | |
| "loss": 0.001, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 666.67, | |
| "eval_loss": 0.0011991177452728152, | |
| "eval_runtime": 21.7109, | |
| "eval_samples_per_second": 5.389, | |
| "eval_steps_per_second": 0.691, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 700.0, | |
| "learning_rate": 3.111034482758621e-06, | |
| "loss": 0.001, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 733.33, | |
| "learning_rate": 2.766206896551724e-06, | |
| "loss": 0.001, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 733.33, | |
| "eval_loss": 0.0011818531202152371, | |
| "eval_runtime": 21.7427, | |
| "eval_samples_per_second": 5.381, | |
| "eval_steps_per_second": 0.69, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 766.67, | |
| "learning_rate": 2.4220689655172416e-06, | |
| "loss": 0.0009, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 800.0, | |
| "learning_rate": 2.077931034482759e-06, | |
| "loss": 0.001, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 800.0, | |
| "eval_loss": 0.0011808592826128006, | |
| "eval_runtime": 21.9404, | |
| "eval_samples_per_second": 5.333, | |
| "eval_steps_per_second": 0.684, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 833.33, | |
| "learning_rate": 1.733103448275862e-06, | |
| "loss": 0.0009, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 866.67, | |
| "learning_rate": 1.3882758620689656e-06, | |
| "loss": 0.0009, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 866.67, | |
| "eval_loss": 0.0012090284144505858, | |
| "eval_runtime": 21.6103, | |
| "eval_samples_per_second": 5.414, | |
| "eval_steps_per_second": 0.694, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 900.0, | |
| "learning_rate": 1.043448275862069e-06, | |
| "loss": 0.0009, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 933.33, | |
| "learning_rate": 6.993103448275862e-07, | |
| "loss": 0.0009, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 933.33, | |
| "eval_loss": 0.0012141974875703454, | |
| "eval_runtime": 21.6816, | |
| "eval_samples_per_second": 5.396, | |
| "eval_steps_per_second": 0.692, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 966.67, | |
| "learning_rate": 3.558620689655173e-07, | |
| "loss": 0.0009, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1000.0, | |
| "learning_rate": 1.1034482758620692e-08, | |
| "loss": 0.0009, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1000.0, | |
| "eval_loss": 0.0012086295755580068, | |
| "eval_runtime": 22.0752, | |
| "eval_samples_per_second": 5.3, | |
| "eval_steps_per_second": 0.679, | |
| "eval_wer": 0.3861003861003861, | |
| "step": 15000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 15000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1000, | |
| "save_steps": 1000, | |
| "total_flos": 2.88040955904e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |