| { |
| "best_metric": 0.2075, |
| "best_model_checkpoint": "/content/model_asr/checkpoint-1100", |
| "epoch": 80.0, |
| "global_step": 1200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 3.33, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 8.6183, |
| "step": 50 |
| }, |
| { |
| "epoch": 3.33, |
| "eval_loss": 4.025334358215332, |
| "eval_runtime": 15.3335, |
| "eval_samples_per_second": 31.304, |
| "eval_steps_per_second": 3.913, |
| "eval_wer": 1.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 6.67, |
| "learning_rate": 9.6e-05, |
| "loss": 3.2794, |
| "step": 100 |
| }, |
| { |
| "epoch": 6.67, |
| "eval_loss": 3.0245492458343506, |
| "eval_runtime": 15.5923, |
| "eval_samples_per_second": 30.784, |
| "eval_steps_per_second": 3.848, |
| "eval_wer": 1.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 10.0, |
| "learning_rate": 9.956913093678348e-05, |
| "loss": 2.9692, |
| "step": 150 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 2.8865509033203125, |
| "eval_runtime": 15.6178, |
| "eval_samples_per_second": 30.734, |
| "eval_steps_per_second": 3.842, |
| "eval_wer": 1.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 13.33, |
| "learning_rate": 9.813243995692098e-05, |
| "loss": 2.8042, |
| "step": 200 |
| }, |
| { |
| "epoch": 13.33, |
| "eval_loss": 2.635584831237793, |
| "eval_runtime": 15.4284, |
| "eval_samples_per_second": 31.112, |
| "eval_steps_per_second": 3.889, |
| "eval_wer": 1.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 16.67, |
| "learning_rate": 9.57159113020304e-05, |
| "loss": 2.5114, |
| "step": 250 |
| }, |
| { |
| "epoch": 16.67, |
| "eval_loss": 2.116630792617798, |
| "eval_runtime": 15.5962, |
| "eval_samples_per_second": 30.777, |
| "eval_steps_per_second": 3.847, |
| "eval_wer": 1.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 20.0, |
| "learning_rate": 9.236873852683213e-05, |
| "loss": 1.7677, |
| "step": 300 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_loss": 1.0035134553909302, |
| "eval_runtime": 15.4247, |
| "eval_samples_per_second": 31.119, |
| "eval_steps_per_second": 3.89, |
| "eval_wer": 0.89375, |
| "step": 300 |
| }, |
| { |
| "epoch": 23.33, |
| "learning_rate": 8.815906041658002e-05, |
| "loss": 0.8802, |
| "step": 350 |
| }, |
| { |
| "epoch": 23.33, |
| "eval_loss": 0.3643040060997009, |
| "eval_runtime": 15.6657, |
| "eval_samples_per_second": 30.64, |
| "eval_steps_per_second": 3.83, |
| "eval_wer": 0.5025, |
| "step": 350 |
| }, |
| { |
| "epoch": 26.67, |
| "learning_rate": 8.31725738778896e-05, |
| "loss": 0.4318, |
| "step": 400 |
| }, |
| { |
| "epoch": 26.67, |
| "eval_loss": 0.14456388354301453, |
| "eval_runtime": 15.4959, |
| "eval_samples_per_second": 30.976, |
| "eval_steps_per_second": 3.872, |
| "eval_wer": 0.3075, |
| "step": 400 |
| }, |
| { |
| "epoch": 30.0, |
| "learning_rate": 7.751078939684885e-05, |
| "loss": 0.2275, |
| "step": 450 |
| }, |
| { |
| "epoch": 30.0, |
| "eval_loss": 0.07944890111684799, |
| "eval_runtime": 16.2382, |
| "eval_samples_per_second": 29.56, |
| "eval_steps_per_second": 3.695, |
| "eval_wer": 0.25, |
| "step": 450 |
| }, |
| { |
| "epoch": 33.33, |
| "learning_rate": 7.128896457825364e-05, |
| "loss": 0.1407, |
| "step": 500 |
| }, |
| { |
| "epoch": 33.33, |
| "eval_loss": 0.05435527116060257, |
| "eval_runtime": 15.4181, |
| "eval_samples_per_second": 31.132, |
| "eval_steps_per_second": 3.892, |
| "eval_wer": 0.22125, |
| "step": 500 |
| }, |
| { |
| "epoch": 36.67, |
| "learning_rate": 6.463375783314937e-05, |
| "loss": 0.1191, |
| "step": 550 |
| }, |
| { |
| "epoch": 36.67, |
| "eval_loss": 0.046384669840335846, |
| "eval_runtime": 15.8291, |
| "eval_samples_per_second": 30.324, |
| "eval_steps_per_second": 3.79, |
| "eval_wer": 0.23375, |
| "step": 550 |
| }, |
| { |
| "epoch": 40.0, |
| "learning_rate": 5.768064997883498e-05, |
| "loss": 0.0825, |
| "step": 600 |
| }, |
| { |
| "epoch": 40.0, |
| "eval_loss": 0.03952096030116081, |
| "eval_runtime": 15.4288, |
| "eval_samples_per_second": 31.111, |
| "eval_steps_per_second": 3.889, |
| "eval_wer": 0.25625, |
| "step": 600 |
| }, |
| { |
| "epoch": 43.33, |
| "learning_rate": 5.057118624011702e-05, |
| "loss": 0.0699, |
| "step": 650 |
| }, |
| { |
| "epoch": 43.33, |
| "eval_loss": 0.03604818880558014, |
| "eval_runtime": 15.4828, |
| "eval_samples_per_second": 31.002, |
| "eval_steps_per_second": 3.875, |
| "eval_wer": 0.2125, |
| "step": 650 |
| }, |
| { |
| "epoch": 46.67, |
| "learning_rate": 4.345009479671536e-05, |
| "loss": 0.0592, |
| "step": 700 |
| }, |
| { |
| "epoch": 46.67, |
| "eval_loss": 0.02805432118475437, |
| "eval_runtime": 15.5578, |
| "eval_samples_per_second": 30.853, |
| "eval_steps_per_second": 3.857, |
| "eval_wer": 0.23875, |
| "step": 700 |
| }, |
| { |
| "epoch": 50.0, |
| "learning_rate": 3.646234053488574e-05, |
| "loss": 0.0546, |
| "step": 750 |
| }, |
| { |
| "epoch": 50.0, |
| "eval_loss": 0.023771686479449272, |
| "eval_runtime": 15.8929, |
| "eval_samples_per_second": 30.202, |
| "eval_steps_per_second": 3.775, |
| "eval_wer": 0.26375, |
| "step": 750 |
| }, |
| { |
| "epoch": 53.33, |
| "learning_rate": 2.9750173980379736e-05, |
| "loss": 0.0455, |
| "step": 800 |
| }, |
| { |
| "epoch": 53.33, |
| "eval_loss": 0.021949028596282005, |
| "eval_runtime": 15.824, |
| "eval_samples_per_second": 30.334, |
| "eval_steps_per_second": 3.792, |
| "eval_wer": 0.24125, |
| "step": 800 |
| }, |
| { |
| "epoch": 56.67, |
| "learning_rate": 2.3450235487957135e-05, |
| "loss": 0.0391, |
| "step": 850 |
| }, |
| { |
| "epoch": 56.67, |
| "eval_loss": 0.020865805447101593, |
| "eval_runtime": 16.0515, |
| "eval_samples_per_second": 29.904, |
| "eval_steps_per_second": 3.738, |
| "eval_wer": 0.24375, |
| "step": 850 |
| }, |
| { |
| "epoch": 60.0, |
| "learning_rate": 1.7690773637801294e-05, |
| "loss": 0.0367, |
| "step": 900 |
| }, |
| { |
| "epoch": 60.0, |
| "eval_loss": 0.022958721965551376, |
| "eval_runtime": 15.4228, |
| "eval_samples_per_second": 31.123, |
| "eval_steps_per_second": 3.89, |
| "eval_wer": 0.21625, |
| "step": 900 |
| }, |
| { |
| "epoch": 63.33, |
| "learning_rate": 1.2589034464264932e-05, |
| "loss": 0.0333, |
| "step": 950 |
| }, |
| { |
| "epoch": 63.33, |
| "eval_loss": 0.021822014823555946, |
| "eval_runtime": 15.5217, |
| "eval_samples_per_second": 30.924, |
| "eval_steps_per_second": 3.866, |
| "eval_wer": 0.23, |
| "step": 950 |
| }, |
| { |
| "epoch": 66.67, |
| "learning_rate": 8.248874664720375e-06, |
| "loss": 0.0327, |
| "step": 1000 |
| }, |
| { |
| "epoch": 66.67, |
| "eval_loss": 0.02000792883336544, |
| "eval_runtime": 15.8435, |
| "eval_samples_per_second": 30.296, |
| "eval_steps_per_second": 3.787, |
| "eval_wer": 0.21625, |
| "step": 1000 |
| }, |
| { |
| "epoch": 70.0, |
| "learning_rate": 4.758647376699032e-06, |
| "loss": 0.0306, |
| "step": 1050 |
| }, |
| { |
| "epoch": 70.0, |
| "eval_loss": 0.02014043927192688, |
| "eval_runtime": 15.7796, |
| "eval_samples_per_second": 30.419, |
| "eval_steps_per_second": 3.802, |
| "eval_wer": 0.21625, |
| "step": 1050 |
| }, |
| { |
| "epoch": 73.33, |
| "learning_rate": 2.1894035628006517e-06, |
| "loss": 0.0269, |
| "step": 1100 |
| }, |
| { |
| "epoch": 73.33, |
| "eval_loss": 0.02098565548658371, |
| "eval_runtime": 16.1184, |
| "eval_samples_per_second": 29.78, |
| "eval_steps_per_second": 3.722, |
| "eval_wer": 0.2075, |
| "step": 1100 |
| }, |
| { |
| "epoch": 76.67, |
| "learning_rate": 5.934456179888804e-07, |
| "loss": 0.0288, |
| "step": 1150 |
| }, |
| { |
| "epoch": 76.67, |
| "eval_loss": 0.021614952012896538, |
| "eval_runtime": 15.8598, |
| "eval_samples_per_second": 30.265, |
| "eval_steps_per_second": 3.783, |
| "eval_wer": 0.21625, |
| "step": 1150 |
| }, |
| { |
| "epoch": 80.0, |
| "learning_rate": 3.262643657425679e-09, |
| "loss": 0.0289, |
| "step": 1200 |
| }, |
| { |
| "epoch": 80.0, |
| "eval_loss": 0.02152070961892605, |
| "eval_runtime": 16.1024, |
| "eval_samples_per_second": 29.809, |
| "eval_steps_per_second": 3.726, |
| "eval_wer": 0.21625, |
| "step": 1200 |
| } |
| ], |
| "max_steps": 1200, |
| "num_train_epochs": 80, |
| "total_flos": 3.3688058529081743e+18, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
