{ "best_metric": null, "best_model_checkpoint": null, "epoch": 1.0, "global_step": 9441, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.05, "loss": 5.6498, "step": 500 }, { "epoch": 0.05, "eval_loss": 3.2174227237701416, "eval_runtime": 50.998, "eval_samples_per_second": 236.362, "eval_steps_per_second": 0.941, "eval_wer": 0.053516471892462315, "step": 500 }, { "epoch": 0.11, "loss": 2.7794, "step": 1000 }, { "epoch": 0.11, "eval_loss": 2.011007070541382, "eval_runtime": 51.5372, "eval_samples_per_second": 233.889, "eval_steps_per_second": 0.931, "eval_wer": 0.049977469992612156, "step": 1000 }, { "epoch": 0.16, "loss": 1.8179, "step": 1500 }, { "epoch": 0.16, "eval_loss": 0.8721506595611572, "eval_runtime": 53.762, "eval_samples_per_second": 224.21, "eval_steps_per_second": 0.893, "eval_wer": 0.04245802797052525, "step": 1500 }, { "epoch": 0.21, "loss": 0.7713, "step": 2000 }, { "epoch": 0.21, "eval_loss": 0.3388041853904724, "eval_runtime": 57.0462, "eval_samples_per_second": 211.303, "eval_steps_per_second": 0.841, "eval_wer": 0.036441974611471896, "step": 2000 }, { "epoch": 0.26, "loss": 0.4234, "step": 2500 }, { "epoch": 0.26, "eval_loss": 0.2034144103527069, "eval_runtime": 65.9674, "eval_samples_per_second": 182.727, "eval_steps_per_second": 0.728, "eval_wer": 0.03474535261335592, "step": 2500 }, { "epoch": 0.32, "loss": 0.2822, "step": 3000 }, { "epoch": 0.32, "eval_loss": 0.13999593257904053, "eval_runtime": 51.8439, "eval_samples_per_second": 232.506, "eval_steps_per_second": 0.926, "eval_wer": 0.03314867889200672, "step": 3000 }, { "epoch": 0.37, "loss": 0.211, "step": 3500 }, { "epoch": 0.37, "eval_loss": 0.10781644284725189, "eval_runtime": 50.8714, "eval_samples_per_second": 236.95, "eval_steps_per_second": 0.944, "eval_wer": 0.032414059057770936, "step": 3500 }, { "epoch": 0.42, "loss": 0.1715, "step": 4000 }, { "epoch": 0.42, "eval_loss": 0.09290529042482376, "eval_runtime": 51.4425, "eval_samples_per_second": 234.32, "eval_steps_per_second": 0.933, "eval_wer": 0.03216918577969234, "step": 4000 }, { "epoch": 0.48, "loss": 0.1441, "step": 4500 }, { "epoch": 0.48, "eval_loss": 0.07368380576372147, "eval_runtime": 51.0907, "eval_samples_per_second": 235.933, "eval_steps_per_second": 0.94, "eval_wer": 0.031723583045773815, "step": 4500 }, { "epoch": 0.53, "loss": 0.1263, "step": 5000 }, { "epoch": 0.53, "eval_loss": 0.06525683403015137, "eval_runtime": 51.1104, "eval_samples_per_second": 235.843, "eval_steps_per_second": 0.939, "eval_wer": 0.03161363994133036, "step": 5000 }, { "epoch": 0.58, "loss": 0.1114, "step": 5500 }, { "epoch": 0.58, "eval_loss": 0.06087009608745575, "eval_runtime": 54.1523, "eval_samples_per_second": 222.595, "eval_steps_per_second": 0.886, "eval_wer": 0.03163362959668371, "step": 5500 }, { "epoch": 0.64, "loss": 0.1007, "step": 6000 }, { "epoch": 0.64, "eval_loss": 0.05339992418885231, "eval_runtime": 51.3822, "eval_samples_per_second": 234.595, "eval_steps_per_second": 0.934, "eval_wer": 0.031447892382358796, "step": 6000 }, { "epoch": 0.69, "loss": 0.0912, "step": 6500 }, { "epoch": 0.69, "eval_loss": 0.050159111618995667, "eval_runtime": 51.2486, "eval_samples_per_second": 235.206, "eval_steps_per_second": 0.937, "eval_wer": 0.031327121547932275, "step": 6500 }, { "epoch": 0.74, "loss": 0.085, "step": 7000 }, { "epoch": 0.74, "eval_loss": 0.04742148146033287, "eval_runtime": 51.4277, "eval_samples_per_second": 234.387, "eval_steps_per_second": 0.933, "eval_wer": 0.03132129123178755, "step": 7000 }, { "epoch": 0.79, "loss": 0.0797, "step": 7500 }, { "epoch": 0.79, "eval_loss": 0.044490132480859756, "eval_runtime": 51.2937, "eval_samples_per_second": 235.0, "eval_steps_per_second": 0.936, "eval_wer": 0.031255491949582756, "step": 7500 }, { "epoch": 0.85, "loss": 0.0741, "step": 8000 }, { "epoch": 0.85, "eval_loss": 0.04160303622484207, "eval_runtime": 51.2978, "eval_samples_per_second": 234.981, "eval_steps_per_second": 0.936, "eval_wer": 0.031156376575122373, "step": 8000 }, { "epoch": 0.9, "loss": 0.072, "step": 8500 }, { "epoch": 0.9, "eval_loss": 0.04107438027858734, "eval_runtime": 51.3225, "eval_samples_per_second": 234.868, "eval_steps_per_second": 0.935, "eval_wer": 0.031193857178909913, "step": 8500 }, { "epoch": 0.95, "loss": 0.0682, "step": 9000 }, { "epoch": 0.95, "eval_loss": 0.03936346620321274, "eval_runtime": 51.3106, "eval_samples_per_second": 234.922, "eval_steps_per_second": 0.935, "eval_wer": 0.031177199132782118, "step": 9000 } ], "max_steps": 9441, "num_train_epochs": 1, "total_flos": 2.3280237219423744e+17, "trial_name": null, "trial_params": null }