| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.172839506172839, | |
| "eval_steps": 200, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.30864197530864196, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 176.8693, | |
| "eval_samples_per_second": 39.702, | |
| "eval_steps_per_second": 4.964, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6172839506172839, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.9503, | |
| "eval_samples_per_second": 39.909, | |
| "eval_steps_per_second": 4.99, | |
| "eval_wer": 0.8138755794476257, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7716049382716049, | |
| "grad_norm": 2.980449914932251, | |
| "learning_rate": 0.0002958, | |
| "loss": 3.6854, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9259259259259259, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 174.7168, | |
| "eval_samples_per_second": 40.191, | |
| "eval_steps_per_second": 5.025, | |
| "eval_wer": 0.6042746029891057, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.2345679012345678, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 176.3896, | |
| "eval_samples_per_second": 39.81, | |
| "eval_steps_per_second": 4.978, | |
| "eval_wer": 0.503447469907678, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.5432098765432098, | |
| "grad_norm": 0.7878520488739014, | |
| "learning_rate": 0.0002578285714285714, | |
| "loss": 0.7236, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5432098765432098, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.5785, | |
| "eval_samples_per_second": 39.994, | |
| "eval_steps_per_second": 5.001, | |
| "eval_wer": 0.44518457922688376, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.8518518518518519, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.3263, | |
| "eval_samples_per_second": 40.051, | |
| "eval_steps_per_second": 5.008, | |
| "eval_wer": 0.4584940204900471, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.1604938271604937, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 174.0761, | |
| "eval_samples_per_second": 40.339, | |
| "eval_steps_per_second": 5.044, | |
| "eval_wer": 0.4735564125537247, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.314814814814815, | |
| "grad_norm": 1.6092010736465454, | |
| "learning_rate": 0.0002153142857142857, | |
| "loss": 0.6244, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.4691358024691357, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 174.8676, | |
| "eval_samples_per_second": 40.156, | |
| "eval_steps_per_second": 5.021, | |
| "eval_wer": 0.4445742926518899, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.7777777777777777, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 174.4183, | |
| "eval_samples_per_second": 40.26, | |
| "eval_steps_per_second": 5.034, | |
| "eval_wer": 0.4992274031656993, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.0864197530864197, | |
| "grad_norm": 13.045821189880371, | |
| "learning_rate": 0.0001727142857142857, | |
| "loss": 0.8045, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0864197530864197, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 176.0054, | |
| "eval_samples_per_second": 39.897, | |
| "eval_steps_per_second": 4.988, | |
| "eval_wer": 0.699323490839209, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.3950617283950617, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.9958, | |
| "eval_samples_per_second": 39.899, | |
| "eval_steps_per_second": 4.989, | |
| "eval_wer": 0.9154558321322375, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.7037037037037037, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.4175, | |
| "eval_samples_per_second": 40.03, | |
| "eval_steps_per_second": 5.005, | |
| "eval_wer": 1.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.8580246913580245, | |
| "grad_norm": 1.601144552230835, | |
| "learning_rate": 0.0001302, | |
| "loss": 2.3067, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.012345679012346, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.9931, | |
| "eval_samples_per_second": 39.899, | |
| "eval_steps_per_second": 4.989, | |
| "eval_wer": 0.998558684897355, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.320987654320987, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.0274, | |
| "eval_samples_per_second": 40.119, | |
| "eval_steps_per_second": 5.016, | |
| "eval_wer": 1.0, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.62962962962963, | |
| "grad_norm": 0.7471032738685608, | |
| "learning_rate": 8.759999999999999e-05, | |
| "loss": 3.008, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.62962962962963, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.5031, | |
| "eval_samples_per_second": 40.011, | |
| "eval_steps_per_second": 5.003, | |
| "eval_wer": 1.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.2017, | |
| "eval_samples_per_second": 40.08, | |
| "eval_steps_per_second": 5.011, | |
| "eval_wer": 1.0, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.246913580246914, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 176.0971, | |
| "eval_samples_per_second": 39.876, | |
| "eval_steps_per_second": 4.986, | |
| "eval_wer": 1.0, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 5.401234567901234, | |
| "grad_norm": 1.3811966180801392, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 3.032, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.555555555555555, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 176.4078, | |
| "eval_samples_per_second": 39.805, | |
| "eval_steps_per_second": 4.977, | |
| "eval_wer": 1.0, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 5.864197530864198, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.4996, | |
| "eval_samples_per_second": 40.011, | |
| "eval_steps_per_second": 5.003, | |
| "eval_wer": 1.0, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 6.172839506172839, | |
| "grad_norm": 0.011152578517794609, | |
| "learning_rate": 2.314285714285714e-06, | |
| "loss": 3.0173, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.172839506172839, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 175.8629, | |
| "eval_samples_per_second": 39.929, | |
| "eval_steps_per_second": 4.993, | |
| "eval_wer": 1.0, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.172839506172839, | |
| "step": 4000, | |
| "total_flos": 3.3889468875697558e+19, | |
| "train_loss": 2.1502342529296876, | |
| "train_runtime": 11441.7295, | |
| "train_samples_per_second": 22.374, | |
| "train_steps_per_second": 0.35 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.3889468875697558e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |