| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 58.064516129032256, | |
| "global_step": 3600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 0.0002906832298136646, | |
| "loss": 3.3346, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "eval_loss": 0.4317544102668762, | |
| "eval_runtime": 130.0321, | |
| "eval_samples_per_second": 14.181, | |
| "eval_wer": 0.5394368823419727, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 0.00023478260869565215, | |
| "loss": 0.1371, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "eval_loss": 0.39734482765197754, | |
| "eval_runtime": 130.3612, | |
| "eval_samples_per_second": 14.145, | |
| "eval_wer": 0.4464851762073466, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "learning_rate": 0.00017888198757763974, | |
| "loss": 0.0616, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 29.03, | |
| "eval_loss": 0.46754732728004456, | |
| "eval_runtime": 132.7444, | |
| "eval_samples_per_second": 13.891, | |
| "eval_wer": 0.43147492075330973, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "learning_rate": 0.00012298136645962733, | |
| "loss": 0.0415, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "eval_loss": 0.45845311880111694, | |
| "eval_runtime": 133.8611, | |
| "eval_samples_per_second": 13.775, | |
| "eval_wer": 0.4201006899123625, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 48.39, | |
| "learning_rate": 6.70807453416149e-05, | |
| "loss": 0.0304, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 48.39, | |
| "eval_loss": 0.47607213258743286, | |
| "eval_runtime": 131.552, | |
| "eval_samples_per_second": 14.017, | |
| "eval_wer": 0.4080738392690658, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 58.06, | |
| "learning_rate": 1.1180124223602484e-05, | |
| "loss": 0.0244, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 58.06, | |
| "eval_loss": 0.4667948782444, | |
| "eval_runtime": 135.2111, | |
| "eval_samples_per_second": 13.638, | |
| "eval_wer": 0.40313257505127725, | |
| "step": 3600 | |
| } | |
| ], | |
| "max_steps": 3720, | |
| "num_train_epochs": 60, | |
| "total_flos": 2.7090797391585608e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |