| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 80.0, | |
| "global_step": 4560, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 4.8822115384615383e-05, | |
| "loss": 6.3683, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "eval_loss": 3.128021001815796, | |
| "eval_runtime": 61.7966, | |
| "eval_samples_per_second": 27.202, | |
| "eval_steps_per_second": 3.414, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "learning_rate": 4.28125e-05, | |
| "loss": 1.9915, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "eval_loss": 0.6600078344345093, | |
| "eval_runtime": 61.9497, | |
| "eval_samples_per_second": 27.135, | |
| "eval_steps_per_second": 3.406, | |
| "eval_wer": 0.6444381384790011, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 3.680288461538462e-05, | |
| "loss": 0.6565, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "eval_loss": 0.4207763671875, | |
| "eval_runtime": 61.6392, | |
| "eval_samples_per_second": 27.272, | |
| "eval_steps_per_second": 3.423, | |
| "eval_wer": 0.4486379114642452, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 35.09, | |
| "learning_rate": 3.079326923076923e-05, | |
| "loss": 0.4484, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 35.09, | |
| "eval_loss": 0.38846877217292786, | |
| "eval_runtime": 60.5282, | |
| "eval_samples_per_second": 27.772, | |
| "eval_steps_per_second": 3.486, | |
| "eval_wer": 0.4005864547862278, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 43.86, | |
| "learning_rate": 2.478365384615385e-05, | |
| "loss": 0.3573, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 43.86, | |
| "eval_loss": 0.3548009693622589, | |
| "eval_runtime": 60.4916, | |
| "eval_samples_per_second": 27.789, | |
| "eval_steps_per_second": 3.488, | |
| "eval_wer": 0.36256148316307224, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 52.63, | |
| "learning_rate": 1.8774038461538462e-05, | |
| "loss": 0.3063, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 52.63, | |
| "eval_loss": 0.3374779522418976, | |
| "eval_runtime": 60.6448, | |
| "eval_samples_per_second": 27.719, | |
| "eval_steps_per_second": 3.479, | |
| "eval_wer": 0.34298146046159667, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 61.4, | |
| "learning_rate": 1.2764423076923077e-05, | |
| "loss": 0.2751, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 61.4, | |
| "eval_loss": 0.3359163701534271, | |
| "eval_runtime": 61.2942, | |
| "eval_samples_per_second": 27.425, | |
| "eval_steps_per_second": 3.442, | |
| "eval_wer": 0.32406356413166854, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 70.18, | |
| "learning_rate": 6.754807692307692e-06, | |
| "loss": 0.2511, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 70.18, | |
| "eval_loss": 0.32216018438339233, | |
| "eval_runtime": 60.7546, | |
| "eval_samples_per_second": 27.669, | |
| "eval_steps_per_second": 3.473, | |
| "eval_wer": 0.31082103670071887, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 78.95, | |
| "learning_rate": 7.451923076923078e-07, | |
| "loss": 0.2361, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 78.95, | |
| "eval_loss": 0.3205258548259735, | |
| "eval_runtime": 61.518, | |
| "eval_samples_per_second": 27.325, | |
| "eval_steps_per_second": 3.43, | |
| "eval_wer": 0.3083617101778282, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "step": 4560, | |
| "total_flos": 3.6657223992885486e+19, | |
| "train_loss": 1.1972084400946634, | |
| "train_runtime": 14228.904, | |
| "train_samples_per_second": 20.387, | |
| "train_steps_per_second": 0.32 | |
| } | |
| ], | |
| "max_steps": 4560, | |
| "num_train_epochs": 80, | |
| "total_flos": 3.6657223992885486e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |