{ "best_global_step": 1000, "best_metric": 26.638817480719794, "best_model_checkpoint": "./tarbiyah-whisper-medium-peft-tarteel/checkpoint-1000", "epoch": 1.09425, "eval_steps": 500, "global_step": 1000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0125, "grad_norm": 1.1746573448181152, "learning_rate": 9.600000000000001e-07, "loss": 1.6706, "step": 50 }, { "epoch": 0.025, "grad_norm": 1.521277904510498, "learning_rate": 1.9600000000000003e-06, "loss": 1.7122, "step": 100 }, { "epoch": 0.0375, "grad_norm": 1.4604511260986328, "learning_rate": 2.96e-06, "loss": 1.7295, "step": 150 }, { "epoch": 0.05, "grad_norm": 19.154804229736328, "learning_rate": 3.94e-06, "loss": 1.7764, "step": 200 }, { "epoch": 0.0625, "grad_norm": 1.852611780166626, "learning_rate": 4.94e-06, "loss": 1.5858, "step": 250 }, { "epoch": 0.075, "grad_norm": 0.8722860813140869, "learning_rate": 5.94e-06, "loss": 1.3009, "step": 300 }, { "epoch": 0.0875, "grad_norm": 0.8003638982772827, "learning_rate": 6.9400000000000005e-06, "loss": 1.1038, "step": 350 }, { "epoch": 0.1, "grad_norm": 0.6477518081665039, "learning_rate": 7.94e-06, "loss": 1.0106, "step": 400 }, { "epoch": 0.1125, "grad_norm": 0.9047778844833374, "learning_rate": 8.94e-06, "loss": 0.9663, "step": 450 }, { "epoch": 0.125, "grad_norm": 0.9506059288978577, "learning_rate": 9.940000000000001e-06, "loss": 0.9356, "step": 500 }, { "epoch": 0.125, "eval_loss": 0.5289004445075989, "eval_runtime": 4348.6348, "eval_samples_per_second": 0.23, "eval_steps_per_second": 0.029, "eval_wer": 28.301735218509, "step": 500 }, { "epoch": 0.1375, "grad_norm": 0.6594715118408203, "learning_rate": 9.865714285714285e-06, "loss": 0.6981, "step": 550 }, { "epoch": 0.15, "grad_norm": 0.6399093866348267, "learning_rate": 9.722857142857143e-06, "loss": 0.5712, "step": 600 }, { "epoch": 1.00675, "grad_norm": 0.7400449514389038, "learning_rate": 9.58e-06, "loss": 0.6562, "step": 650 }, { "epoch": 1.01925, "grad_norm": 0.7432519793510437, "learning_rate": 9.437142857142858e-06, "loss": 0.7226, "step": 700 }, { "epoch": 1.03175, "grad_norm": 0.8710829019546509, "learning_rate": 9.294285714285714e-06, "loss": 0.6667, "step": 750 }, { "epoch": 1.04425, "grad_norm": 0.940646767616272, "learning_rate": 9.151428571428572e-06, "loss": 0.701, "step": 800 }, { "epoch": 1.05675, "grad_norm": 1.0409663915634155, "learning_rate": 9.00857142857143e-06, "loss": 0.707, "step": 850 }, { "epoch": 1.06925, "grad_norm": 0.9208248853683472, "learning_rate": 8.865714285714287e-06, "loss": 0.7074, "step": 900 }, { "epoch": 1.08175, "grad_norm": 0.943744957447052, "learning_rate": 8.722857142857145e-06, "loss": 0.647, "step": 950 }, { "epoch": 1.09425, "grad_norm": 1.1141985654830933, "learning_rate": 8.580000000000001e-06, "loss": 0.5822, "step": 1000 }, { "epoch": 1.09425, "eval_loss": 0.3140052855014801, "eval_runtime": 4293.3635, "eval_samples_per_second": 0.233, "eval_steps_per_second": 0.029, "eval_wer": 26.638817480719794, "step": 1000 } ], "logging_steps": 50, "max_steps": 4000, "num_input_tokens_seen": 0, "num_train_epochs": 9223372036854775807, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 1.638608470474752e+19, "train_batch_size": 4, "trial_name": null, "trial_params": null }