| { |
| "best_metric": 22.905706191825175, |
| "best_model_checkpoint": "./checkpoint-600", |
| "epoch": 40.0, |
| "global_step": 800, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.25, |
| "learning_rate": 4.2000000000000006e-07, |
| "loss": 2.1103, |
| "step": 25 |
| }, |
| { |
| "epoch": 2.5, |
| "learning_rate": 9.200000000000001e-07, |
| "loss": 1.6262, |
| "step": 50 |
| }, |
| { |
| "epoch": 3.75, |
| "learning_rate": 1.42e-06, |
| "loss": 0.9697, |
| "step": 75 |
| }, |
| { |
| "epoch": 5.0, |
| "learning_rate": 1.9200000000000003e-06, |
| "loss": 0.4257, |
| "step": 100 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.43767037987709045, |
| "eval_runtime": 132.7315, |
| "eval_samples_per_second": 5.101, |
| "eval_steps_per_second": 0.166, |
| "eval_wer": 32.41602590044516, |
| "step": 100 |
| }, |
| { |
| "epoch": 6.25, |
| "learning_rate": 2.42e-06, |
| "loss": 0.267, |
| "step": 125 |
| }, |
| { |
| "epoch": 7.5, |
| "learning_rate": 2.92e-06, |
| "loss": 0.1836, |
| "step": 150 |
| }, |
| { |
| "epoch": 8.75, |
| "learning_rate": 3.4200000000000007e-06, |
| "loss": 0.1212, |
| "step": 175 |
| }, |
| { |
| "epoch": 10.0, |
| "learning_rate": 3.920000000000001e-06, |
| "loss": 0.0779, |
| "step": 200 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.3927859365940094, |
| "eval_runtime": 170.4399, |
| "eval_samples_per_second": 3.972, |
| "eval_steps_per_second": 0.129, |
| "eval_wer": 23.75556454876568, |
| "step": 200 |
| }, |
| { |
| "epoch": 11.25, |
| "learning_rate": 4.42e-06, |
| "loss": 0.0479, |
| "step": 225 |
| }, |
| { |
| "epoch": 12.5, |
| "learning_rate": 4.92e-06, |
| "loss": 0.027, |
| "step": 250 |
| }, |
| { |
| "epoch": 13.75, |
| "learning_rate": 5.420000000000001e-06, |
| "loss": 0.0155, |
| "step": 275 |
| }, |
| { |
| "epoch": 15.0, |
| "learning_rate": 5.92e-06, |
| "loss": 0.0108, |
| "step": 300 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_loss": 0.48562100529670715, |
| "eval_runtime": 132.2653, |
| "eval_samples_per_second": 5.118, |
| "eval_steps_per_second": 0.166, |
| "eval_wer": 23.431808984216918, |
| "step": 300 |
| }, |
| { |
| "epoch": 16.25, |
| "learning_rate": 6.42e-06, |
| "loss": 0.0077, |
| "step": 325 |
| }, |
| { |
| "epoch": 17.5, |
| "learning_rate": 6.92e-06, |
| "loss": 0.0056, |
| "step": 350 |
| }, |
| { |
| "epoch": 18.75, |
| "learning_rate": 7.420000000000001e-06, |
| "loss": 0.0064, |
| "step": 375 |
| }, |
| { |
| "epoch": 20.0, |
| "learning_rate": 7.92e-06, |
| "loss": 0.0104, |
| "step": 400 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_loss": 0.5637333989143372, |
| "eval_runtime": 131.0951, |
| "eval_samples_per_second": 5.164, |
| "eval_steps_per_second": 0.168, |
| "eval_wer": 25.495750708215297, |
| "step": 400 |
| }, |
| { |
| "epoch": 21.25, |
| "learning_rate": 8.42e-06, |
| "loss": 0.0091, |
| "step": 425 |
| }, |
| { |
| "epoch": 22.5, |
| "learning_rate": 8.920000000000001e-06, |
| "loss": 0.0073, |
| "step": 450 |
| }, |
| { |
| "epoch": 23.75, |
| "learning_rate": 9.42e-06, |
| "loss": 0.0072, |
| "step": 475 |
| }, |
| { |
| "epoch": 25.0, |
| "learning_rate": 9.920000000000002e-06, |
| "loss": 0.0069, |
| "step": 500 |
| }, |
| { |
| "epoch": 25.0, |
| "eval_loss": 0.5289409756660461, |
| "eval_runtime": 131.511, |
| "eval_samples_per_second": 5.148, |
| "eval_steps_per_second": 0.167, |
| "eval_wer": 23.148522865236746, |
| "step": 500 |
| }, |
| { |
| "epoch": 26.25, |
| "learning_rate": 9.3e-06, |
| "loss": 0.0056, |
| "step": 525 |
| }, |
| { |
| "epoch": 27.5, |
| "learning_rate": 8.466666666666668e-06, |
| "loss": 0.0057, |
| "step": 550 |
| }, |
| { |
| "epoch": 28.75, |
| "learning_rate": 7.633333333333334e-06, |
| "loss": 0.0035, |
| "step": 575 |
| }, |
| { |
| "epoch": 30.0, |
| "learning_rate": 6.800000000000001e-06, |
| "loss": 0.0022, |
| "step": 600 |
| }, |
| { |
| "epoch": 30.0, |
| "eval_loss": 0.5670017004013062, |
| "eval_runtime": 130.771, |
| "eval_samples_per_second": 5.177, |
| "eval_steps_per_second": 0.168, |
| "eval_wer": 22.905706191825175, |
| "step": 600 |
| }, |
| { |
| "epoch": 31.25, |
| "learning_rate": 5.966666666666667e-06, |
| "loss": 0.0016, |
| "step": 625 |
| }, |
| { |
| "epoch": 32.5, |
| "learning_rate": 5.133333333333334e-06, |
| "loss": 0.001, |
| "step": 650 |
| }, |
| { |
| "epoch": 33.75, |
| "learning_rate": 4.3e-06, |
| "loss": 0.0012, |
| "step": 675 |
| }, |
| { |
| "epoch": 35.0, |
| "learning_rate": 3.4666666666666672e-06, |
| "loss": 0.0012, |
| "step": 700 |
| }, |
| { |
| "epoch": 35.0, |
| "eval_loss": 0.574644923210144, |
| "eval_runtime": 131.2536, |
| "eval_samples_per_second": 5.158, |
| "eval_steps_per_second": 0.168, |
| "eval_wer": 23.02711452853096, |
| "step": 700 |
| }, |
| { |
| "epoch": 36.25, |
| "learning_rate": 2.6333333333333332e-06, |
| "loss": 0.0006, |
| "step": 725 |
| }, |
| { |
| "epoch": 37.5, |
| "learning_rate": 1.8000000000000001e-06, |
| "loss": 0.0006, |
| "step": 750 |
| }, |
| { |
| "epoch": 38.75, |
| "learning_rate": 9.666666666666668e-07, |
| "loss": 0.0006, |
| "step": 775 |
| }, |
| { |
| "epoch": 40.0, |
| "learning_rate": 1.3333333333333336e-07, |
| "loss": 0.0006, |
| "step": 800 |
| }, |
| { |
| "epoch": 40.0, |
| "eval_loss": 0.5809842944145203, |
| "eval_runtime": 131.1241, |
| "eval_samples_per_second": 5.163, |
| "eval_steps_per_second": 0.168, |
| "eval_wer": 23.188992310805343, |
| "step": 800 |
| }, |
| { |
| "epoch": 40.0, |
| "step": 800, |
| "total_flos": 1.43830963519488e+19, |
| "train_loss": 0.18650398002238944, |
| "train_runtime": 4114.7394, |
| "train_samples_per_second": 12.443, |
| "train_steps_per_second": 0.194 |
| } |
| ], |
| "max_steps": 800, |
| "num_train_epochs": 40, |
| "total_flos": 1.43830963519488e+19, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
