| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 40.0, | |
| "global_step": 2520, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.35, | |
| "eval_loss": 0.9379647374153137, | |
| "eval_runtime": 95.9228, | |
| "eval_samples_per_second": 19.62, | |
| "eval_steps_per_second": 2.46, | |
| "eval_wer": 0.7901558654634947, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 0.00029939999999999996, | |
| "loss": 3.2674, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.7, | |
| "eval_loss": 0.5793638825416565, | |
| "eval_runtime": 96.8489, | |
| "eval_samples_per_second": 19.432, | |
| "eval_steps_per_second": 2.437, | |
| "eval_wer": 0.530872299699207, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 15.87, | |
| "learning_rate": 0.00022589108910891088, | |
| "loss": 0.7531, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "eval_loss": 0.5748894810676575, | |
| "eval_runtime": 94.3353, | |
| "eval_samples_per_second": 19.95, | |
| "eval_steps_per_second": 2.502, | |
| "eval_wer": 0.4814875581077386, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 23.81, | |
| "learning_rate": 0.0001516336633663366, | |
| "loss": 0.5382, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 25.4, | |
| "eval_loss": 0.5529599785804749, | |
| "eval_runtime": 100.0797, | |
| "eval_samples_per_second": 18.805, | |
| "eval_steps_per_second": 2.358, | |
| "eval_wer": 0.44473612250478534, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 31.75, | |
| "learning_rate": 7.737623762376237e-05, | |
| "loss": 0.4293, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 31.75, | |
| "eval_loss": 0.5709025263786316, | |
| "eval_runtime": 93.9383, | |
| "eval_samples_per_second": 20.034, | |
| "eval_steps_per_second": 2.512, | |
| "eval_wer": 0.42373530216024063, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 38.1, | |
| "eval_loss": 0.5476023554801941, | |
| "eval_runtime": 95.0441, | |
| "eval_samples_per_second": 19.801, | |
| "eval_steps_per_second": 2.483, | |
| "eval_wer": 0.4059064807219032, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 39.68, | |
| "learning_rate": 3.1188118811881187e-06, | |
| "loss": 0.3635, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "step": 2520, | |
| "total_flos": 2.64515330270157e+19, | |
| "train_loss": 1.0643680714425587, | |
| "train_runtime": 10312.3076, | |
| "train_samples_per_second": 15.543, | |
| "train_steps_per_second": 0.244 | |
| } | |
| ], | |
| "max_steps": 2520, | |
| "num_train_epochs": 40, | |
| "total_flos": 2.64515330270157e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |