| { | |
| "best_metric": 42.648490749756576, | |
| "best_model_checkpoint": "../models/whisper-base-te/checkpoint-5000", | |
| "epoch": 2.128, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.980000000000001e-06, | |
| "loss": 1.4811, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.980000000000001e-06, | |
| "loss": 0.6341, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.3893534541130066, | |
| "eval_runtime": 68.5923, | |
| "eval_samples_per_second": 1.458, | |
| "eval_steps_per_second": 0.029, | |
| "eval_wer": 60.71080817916261, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.446666666666667e-06, | |
| "loss": 0.424, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.891111111111111e-06, | |
| "loss": 0.349, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.30811938643455505, | |
| "eval_runtime": 69.2322, | |
| "eval_samples_per_second": 1.444, | |
| "eval_steps_per_second": 0.029, | |
| "eval_wer": 52.093476144109054, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 8.335555555555556e-06, | |
| "loss": 0.3067, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.78e-06, | |
| "loss": 0.2792, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.2873762547969818, | |
| "eval_runtime": 68.1266, | |
| "eval_samples_per_second": 1.468, | |
| "eval_steps_per_second": 0.029, | |
| "eval_wer": 49.7078870496592, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 7.224444444444445e-06, | |
| "loss": 0.2589, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 6.668888888888889e-06, | |
| "loss": 0.2433, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.271973192691803, | |
| "eval_runtime": 78.0972, | |
| "eval_samples_per_second": 1.28, | |
| "eval_steps_per_second": 0.026, | |
| "eval_wer": 47.565725413826684, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.1155555555555555e-06, | |
| "loss": 0.2314, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 5.560000000000001e-06, | |
| "loss": 0.2224, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_loss": 0.2631737291812897, | |
| "eval_runtime": 68.5355, | |
| "eval_samples_per_second": 1.459, | |
| "eval_steps_per_second": 0.029, | |
| "eval_wer": 45.228821811100296, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 5.004444444444445e-06, | |
| "loss": 0.2123, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.448888888888889e-06, | |
| "loss": 0.2058, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_loss": 0.25287795066833496, | |
| "eval_runtime": 68.6284, | |
| "eval_samples_per_second": 1.457, | |
| "eval_steps_per_second": 0.029, | |
| "eval_wer": 44.303797468354425, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 3.893333333333333e-06, | |
| "loss": 0.198, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.337777777777778e-06, | |
| "loss": 0.1944, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_loss": 0.25191909074783325, | |
| "eval_runtime": 68.631, | |
| "eval_samples_per_second": 1.457, | |
| "eval_steps_per_second": 0.029, | |
| "eval_wer": 44.595910418695226, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 2.7822222222222222e-06, | |
| "loss": 0.1901, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 2.226666666666667e-06, | |
| "loss": 0.1869, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 0.24745255708694458, | |
| "eval_runtime": 67.9563, | |
| "eval_samples_per_second": 1.472, | |
| "eval_steps_per_second": 0.029, | |
| "eval_wer": 43.71957156767283, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.6711111111111112e-06, | |
| "loss": 0.1841, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.1155555555555558e-06, | |
| "loss": 0.1811, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_loss": 0.24505974352359772, | |
| "eval_runtime": 69.3375, | |
| "eval_samples_per_second": 1.442, | |
| "eval_steps_per_second": 0.029, | |
| "eval_wer": 43.3300876338851, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 5.6e-07, | |
| "loss": 0.1812, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 4.444444444444445e-09, | |
| "loss": 0.1775, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_loss": 0.2454656958580017, | |
| "eval_runtime": 74.0826, | |
| "eval_samples_per_second": 1.35, | |
| "eval_steps_per_second": 0.027, | |
| "eval_wer": 42.648490749756576, | |
| "step": 5000 | |
| } | |
| ], | |
| "max_steps": 5000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 3.11315009568768e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |