| { | |
| "best_metric": 108.59188544152745, | |
| "best_model_checkpoint": "ASFEMA_SEV/checkpoint-800", | |
| "epoch": 10.385852090032154, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.6000000000000004e-07, | |
| "loss": 4.0894, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 3.7, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.46e-06, | |
| "loss": 3.0759, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.9600000000000003e-06, | |
| "loss": 2.5729, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 2.46e-06, | |
| "loss": 2.1885, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 2.96e-06, | |
| "loss": 1.9431, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 3.46e-06, | |
| "loss": 1.5753, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 3.96e-06, | |
| "loss": 1.3319, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 4.4600000000000005e-06, | |
| "loss": 1.2076, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 4.960000000000001e-06, | |
| "loss": 0.9211, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 5.460000000000001e-06, | |
| "loss": 0.7332, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 0.6601, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 6.460000000000001e-06, | |
| "loss": 0.4824, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 6.96e-06, | |
| "loss": 0.3455, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 7.4600000000000006e-06, | |
| "loss": 0.3741, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 0.2448, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "eval_loss": 0.9975764751434326, | |
| "eval_runtime": 24.147, | |
| "eval_samples_per_second": 5.756, | |
| "eval_steps_per_second": 0.745, | |
| "eval_wer": 132.57756563245823, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 8.46e-06, | |
| "loss": 0.1713, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 8.96e-06, | |
| "loss": 0.1906, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 9.460000000000001e-06, | |
| "loss": 0.1479, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 0.103, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 9.975789473684211e-06, | |
| "loss": 0.1271, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 9.949473684210526e-06, | |
| "loss": 0.1033, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 9.923157894736844e-06, | |
| "loss": 0.0799, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 9.89684210526316e-06, | |
| "loss": 0.0814, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 9.870526315789474e-06, | |
| "loss": 0.069, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 9.84421052631579e-06, | |
| "loss": 0.0546, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 9.817894736842106e-06, | |
| "loss": 0.0627, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 9.791578947368422e-06, | |
| "loss": 0.0459, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 9.765263157894737e-06, | |
| "loss": 0.034, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 9.738947368421054e-06, | |
| "loss": 0.036, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 9.712631578947369e-06, | |
| "loss": 0.0367, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 10.39, | |
| "learning_rate": 9.686315789473684e-06, | |
| "loss": 0.0248, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 10.39, | |
| "eval_loss": 0.8046127557754517, | |
| "eval_runtime": 19.9036, | |
| "eval_samples_per_second": 6.984, | |
| "eval_steps_per_second": 0.904, | |
| "eval_wer": 108.59188544152745, | |
| "step": 800 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 130, | |
| "total_flos": 1.31862154051584e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |