| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "global_step": 1635, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 3.6067986488342285, | |
| "eval_runtime": 67.1306, | |
| "eval_samples_per_second": 24.534, | |
| "eval_steps_per_second": 3.069, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_loss": 3.041459321975708, | |
| "eval_runtime": 67.0963, | |
| "eval_samples_per_second": 24.547, | |
| "eval_steps_per_second": 3.07, | |
| "eval_wer": 0.9998978653865795, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_loss": 0.8838828206062317, | |
| "eval_runtime": 66.9473, | |
| "eval_samples_per_second": 24.601, | |
| "eval_steps_per_second": 3.077, | |
| "eval_wer": 0.7398631396180165, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "eval_loss": 0.5810018181800842, | |
| "eval_runtime": 65.6828, | |
| "eval_samples_per_second": 25.075, | |
| "eval_steps_per_second": 3.136, | |
| "eval_wer": 0.5892145848227964, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.0002982, | |
| "loss": 3.2096, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "eval_loss": 0.5231282711029053, | |
| "eval_runtime": 67.353, | |
| "eval_samples_per_second": 24.453, | |
| "eval_steps_per_second": 3.059, | |
| "eval_wer": 0.5485650086814421, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "eval_loss": 0.449034720659256, | |
| "eval_runtime": 66.057, | |
| "eval_samples_per_second": 24.933, | |
| "eval_steps_per_second": 3.119, | |
| "eval_wer": 0.5066898171790419, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "eval_loss": 0.4039786159992218, | |
| "eval_runtime": 65.8832, | |
| "eval_samples_per_second": 24.999, | |
| "eval_steps_per_second": 3.127, | |
| "eval_wer": 0.4405065876825656, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "eval_loss": 0.4107377827167511, | |
| "eval_runtime": 65.8157, | |
| "eval_samples_per_second": 25.024, | |
| "eval_steps_per_second": 3.13, | |
| "eval_wer": 0.43080379940761926, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "eval_loss": 0.4097311794757843, | |
| "eval_runtime": 65.3892, | |
| "eval_samples_per_second": 25.188, | |
| "eval_steps_per_second": 3.15, | |
| "eval_wer": 0.4143601266469206, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 0.00016863436123348017, | |
| "loss": 0.224, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "eval_loss": 0.3750542402267456, | |
| "eval_runtime": 65.3056, | |
| "eval_samples_per_second": 25.22, | |
| "eval_steps_per_second": 3.154, | |
| "eval_wer": 0.3914819732407313, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.09, | |
| "eval_loss": 0.4037274420261383, | |
| "eval_runtime": 66.4673, | |
| "eval_samples_per_second": 24.779, | |
| "eval_steps_per_second": 3.099, | |
| "eval_wer": 0.3872944540904913, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "eval_loss": 0.39948078989982605, | |
| "eval_runtime": 65.7895, | |
| "eval_samples_per_second": 25.034, | |
| "eval_steps_per_second": 3.131, | |
| "eval_wer": 0.37943008885711366, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "eval_loss": 0.380903959274292, | |
| "eval_runtime": 66.0515, | |
| "eval_samples_per_second": 24.935, | |
| "eval_steps_per_second": 3.119, | |
| "eval_wer": 0.36737820447349606, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 12.84, | |
| "eval_loss": 0.3864530920982361, | |
| "eval_runtime": 65.226, | |
| "eval_samples_per_second": 25.251, | |
| "eval_steps_per_second": 3.158, | |
| "eval_wer": 0.3573690123582882, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "learning_rate": 3.647577092511013e-05, | |
| "loss": 0.1027, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "eval_loss": 0.38098737597465515, | |
| "eval_runtime": 66.3137, | |
| "eval_samples_per_second": 24.837, | |
| "eval_steps_per_second": 3.106, | |
| "eval_wer": 0.3534878970483097, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "eval_loss": 0.37903106212615967, | |
| "eval_runtime": 65.9553, | |
| "eval_samples_per_second": 24.971, | |
| "eval_steps_per_second": 3.123, | |
| "eval_wer": 0.35021958941885406, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 1635, | |
| "total_flos": 6.429866751096261e+18, | |
| "train_loss": 1.0876544129957846, | |
| "train_runtime": 3803.1822, | |
| "train_samples_per_second": 13.717, | |
| "train_steps_per_second": 0.43 | |
| } | |
| ], | |
| "max_steps": 1635, | |
| "num_train_epochs": 15, | |
| "total_flos": 6.429866751096261e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |