{ "best_metric": null, "best_model_checkpoint": null, "epoch": 29.625766871165645, "global_step": 2400, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 4.93, "learning_rate": 6.400000000000001e-05, "loss": 0.0363, "step": 400 }, { "epoch": 4.93, "eval_loss": 0.6197775602340698, "eval_runtime": 84.7989, "eval_samples_per_second": 10.389, "eval_wer": 0.5268051434223541, "step": 400 }, { "epoch": 9.87, "learning_rate": 6.756476683937824e-05, "loss": 0.0221, "step": 800 }, { "epoch": 9.87, "eval_loss": 0.7040311694145203, "eval_runtime": 86.0378, "eval_samples_per_second": 10.24, "eval_wer": 0.5272007912957468, "step": 800 }, { "epoch": 14.81, "learning_rate": 5.098445595854923e-05, "loss": 0.0174, "step": 1200 }, { "epoch": 14.81, "eval_loss": 0.7597708702087402, "eval_runtime": 85.9482, "eval_samples_per_second": 10.25, "eval_wer": 0.5339268051434224, "step": 1200 }, { "epoch": 19.75, "learning_rate": 3.440414507772021e-05, "loss": 0.0126, "step": 1600 }, { "epoch": 19.75, "eval_loss": 0.7824531197547913, "eval_runtime": 85.9896, "eval_samples_per_second": 10.245, "eval_wer": 0.5224530168150346, "step": 1600 }, { "epoch": 24.69, "learning_rate": 1.7823834196891192e-05, "loss": 0.0125, "step": 2000 }, { "epoch": 24.69, "eval_loss": 0.769702672958374, "eval_runtime": 86.2655, "eval_samples_per_second": 10.213, "eval_wer": 0.5224530168150346, "step": 2000 }, { "epoch": 29.63, "learning_rate": 1.2435233160621762e-06, "loss": 0.0167, "step": 2400 }, { "epoch": 29.63, "eval_loss": 0.7395063638687134, "eval_runtime": 83.5861, "eval_samples_per_second": 10.54, "eval_wer": 0.5198813056379822, "step": 2400 } ], "max_steps": 2430, "num_train_epochs": 30, "total_flos": 9.370015200295258e+18, "trial_name": null, "trial_params": null }