| { | |
| "best_metric": 59.720575783234544, | |
| "best_model_checkpoint": "./whisper-tiny-hi2_test/checkpoint-120", | |
| "epoch": 6.349206349206349, | |
| "eval_steps": 40, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6349206349206349, | |
| "grad_norm": 8.253315925598145, | |
| "learning_rate": 3.75e-05, | |
| "loss": 1.7555, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2698412698412698, | |
| "grad_norm": 6.022846698760986, | |
| "learning_rate": 3.75e-05, | |
| "loss": 0.6766, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.2698412698412698, | |
| "eval_loss": 0.6153799295425415, | |
| "eval_runtime": 368.3001, | |
| "eval_samples_per_second": 2.715, | |
| "eval_steps_per_second": 0.679, | |
| "eval_wer": 81.47332768839966, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 6.249065399169922, | |
| "learning_rate": 3.75e-05, | |
| "loss": 0.4907, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.5396825396825395, | |
| "grad_norm": 5.881297588348389, | |
| "learning_rate": 3.75e-05, | |
| "loss": 0.3599, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.5396825396825395, | |
| "eval_loss": 0.5078147649765015, | |
| "eval_runtime": 314.278, | |
| "eval_samples_per_second": 3.182, | |
| "eval_steps_per_second": 0.795, | |
| "eval_wer": 67.01100762066045, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1746031746031744, | |
| "grad_norm": 4.2090606689453125, | |
| "learning_rate": 3.75e-05, | |
| "loss": 0.301, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 4.868574142456055, | |
| "learning_rate": 3.75e-05, | |
| "loss": 0.2297, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "eval_loss": 0.49397116899490356, | |
| "eval_runtime": 304.836, | |
| "eval_samples_per_second": 3.28, | |
| "eval_steps_per_second": 0.82, | |
| "eval_wer": 59.720575783234544, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 4.324865341186523, | |
| "learning_rate": 3.75e-05, | |
| "loss": 0.1754, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 5.079365079365079, | |
| "grad_norm": 3.0074007511138916, | |
| "learning_rate": 3.75e-05, | |
| "loss": 0.153, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.079365079365079, | |
| "eval_loss": 0.5192885994911194, | |
| "eval_runtime": 309.7886, | |
| "eval_samples_per_second": 3.228, | |
| "eval_steps_per_second": 0.807, | |
| "eval_wer": 62.07451312447078, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 2.9954638481140137, | |
| "learning_rate": 3.75e-05, | |
| "loss": 0.1049, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.349206349206349, | |
| "grad_norm": 3.4090383052825928, | |
| "learning_rate": 3.75e-05, | |
| "loss": 0.0885, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.349206349206349, | |
| "eval_loss": 0.5557394027709961, | |
| "eval_runtime": 308.2851, | |
| "eval_samples_per_second": 3.244, | |
| "eval_steps_per_second": 0.811, | |
| "eval_wer": 60.584250635055035, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.349206349206349, | |
| "step": 200, | |
| "total_flos": 1.5637915828224e+17, | |
| "train_loss": 0.4335097998380661, | |
| "train_runtime": 2402.3095, | |
| "train_samples_per_second": 2.664, | |
| "train_steps_per_second": 0.083 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 40, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.5637915828224e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |