| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 14.990291262135923, | |
| "global_step": 765, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 6.859605312347412, | |
| "eval_runtime": 25.516, | |
| "eval_samples_per_second": 8.034, | |
| "eval_steps_per_second": 1.019, | |
| "eval_wer": 1.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_loss": 3.7317371368408203, | |
| "eval_runtime": 25.4287, | |
| "eval_samples_per_second": 8.062, | |
| "eval_steps_per_second": 1.022, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_loss": 3.4334847927093506, | |
| "eval_runtime": 25.371, | |
| "eval_samples_per_second": 8.08, | |
| "eval_steps_per_second": 1.025, | |
| "eval_wer": 1.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "eval_loss": 1.2018941640853882, | |
| "eval_runtime": 25.2525, | |
| "eval_samples_per_second": 8.118, | |
| "eval_steps_per_second": 1.03, | |
| "eval_wer": 0.7789185489390829, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "eval_loss": 0.3254333734512329, | |
| "eval_runtime": 25.4191, | |
| "eval_samples_per_second": 8.065, | |
| "eval_steps_per_second": 1.023, | |
| "eval_wer": 0.3333333333333333, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "eval_loss": 0.2722693979740143, | |
| "eval_runtime": 25.4112, | |
| "eval_samples_per_second": 8.067, | |
| "eval_steps_per_second": 1.023, | |
| "eval_wer": 0.3073237508555784, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "eval_loss": 0.23940488696098328, | |
| "eval_runtime": 25.3486, | |
| "eval_samples_per_second": 8.087, | |
| "eval_steps_per_second": 1.026, | |
| "eval_wer": 0.2464065708418891, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "eval_loss": 0.22206060588359833, | |
| "eval_runtime": 25.3043, | |
| "eval_samples_per_second": 8.101, | |
| "eval_steps_per_second": 1.027, | |
| "eval_wer": 0.22587268993839835, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "eval_loss": 0.213778555393219, | |
| "eval_runtime": 25.4337, | |
| "eval_samples_per_second": 8.06, | |
| "eval_steps_per_second": 1.022, | |
| "eval_wer": 0.22039698836413416, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 0.00012090225563909773, | |
| "loss": 4.0907, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "eval_loss": 0.2070361226797104, | |
| "eval_runtime": 25.2632, | |
| "eval_samples_per_second": 8.115, | |
| "eval_steps_per_second": 1.029, | |
| "eval_wer": 0.21765913757700206, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 10.78, | |
| "eval_loss": 0.2013179063796997, | |
| "eval_runtime": 25.4168, | |
| "eval_samples_per_second": 8.066, | |
| "eval_steps_per_second": 1.023, | |
| "eval_wer": 0.20123203285420946, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "eval_loss": 0.2033400684595108, | |
| "eval_runtime": 25.6076, | |
| "eval_samples_per_second": 8.005, | |
| "eval_steps_per_second": 1.015, | |
| "eval_wer": 0.20465434633812457, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "eval_loss": 0.20385989546775818, | |
| "eval_runtime": 25.6153, | |
| "eval_samples_per_second": 8.003, | |
| "eval_steps_per_second": 1.015, | |
| "eval_wer": 0.2026009582477755, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "eval_loss": 0.2015053778886795, | |
| "eval_runtime": 25.3511, | |
| "eval_samples_per_second": 8.086, | |
| "eval_steps_per_second": 1.026, | |
| "eval_wer": 0.2053388090349076, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 14.7, | |
| "eval_loss": 0.19900637865066528, | |
| "eval_runtime": 25.0763, | |
| "eval_samples_per_second": 8.175, | |
| "eval_steps_per_second": 1.037, | |
| "eval_wer": 0.20191649555099248, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "step": 765, | |
| "total_flos": 5.601028718005862e+18, | |
| "train_loss": 2.7169430053312014, | |
| "train_runtime": 6457.8468, | |
| "train_samples_per_second": 3.828, | |
| "train_steps_per_second": 0.118 | |
| } | |
| ], | |
| "max_steps": 765, | |
| "num_train_epochs": 15, | |
| "total_flos": 5.601028718005862e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |