| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 72700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 12.345376968383789, | |
| "learning_rate": 9.000962861072903e-06, | |
| "loss": 0.4404, | |
| "step": 7270 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.3582181930541992, | |
| "eval_runtime": 460.7541, | |
| "eval_samples_per_second": 3.507, | |
| "eval_steps_per_second": 1.754, | |
| "eval_wer": 27.027607691481364, | |
| "step": 7270 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 11.297811508178711, | |
| "learning_rate": 8.00123796423659e-06, | |
| "loss": 0.2295, | |
| "step": 14540 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.337088942527771, | |
| "eval_runtime": 446.7683, | |
| "eval_samples_per_second": 3.617, | |
| "eval_steps_per_second": 1.809, | |
| "eval_wer": 24.08421957245676, | |
| "step": 14540 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 6.743635654449463, | |
| "learning_rate": 7.0015130674002755e-06, | |
| "loss": 0.1288, | |
| "step": 21810 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.36680105328559875, | |
| "eval_runtime": 448.0545, | |
| "eval_samples_per_second": 3.607, | |
| "eval_steps_per_second": 1.803, | |
| "eval_wer": 22.762917606617254, | |
| "step": 21810 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.48354658484458923, | |
| "learning_rate": 6.001788170563962e-06, | |
| "loss": 0.0683, | |
| "step": 29080 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.4089459776878357, | |
| "eval_runtime": 445.7699, | |
| "eval_samples_per_second": 3.625, | |
| "eval_steps_per_second": 1.813, | |
| "eval_wer": 21.253625523686758, | |
| "step": 29080 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 6.894495010375977, | |
| "learning_rate": 5.0022008253094915e-06, | |
| "loss": 0.0345, | |
| "step": 36350 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.4668106138706207, | |
| "eval_runtime": 442.4372, | |
| "eval_samples_per_second": 3.652, | |
| "eval_steps_per_second": 1.826, | |
| "eval_wer": 20.888387581909978, | |
| "step": 36350 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.9548712372779846, | |
| "learning_rate": 4.002475928473178e-06, | |
| "loss": 0.0181, | |
| "step": 43620 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.4947289228439331, | |
| "eval_runtime": 441.6646, | |
| "eval_samples_per_second": 3.659, | |
| "eval_steps_per_second": 1.829, | |
| "eval_wer": 20.02900418949404, | |
| "step": 43620 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.00920580979436636, | |
| "learning_rate": 3.0030261348005508e-06, | |
| "loss": 0.0099, | |
| "step": 50890 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.5387442708015442, | |
| "eval_runtime": 444.3641, | |
| "eval_samples_per_second": 3.637, | |
| "eval_steps_per_second": 1.818, | |
| "eval_wer": 19.959179288860245, | |
| "step": 50890 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.07406998425722122, | |
| "learning_rate": 2.0033012379642366e-06, | |
| "loss": 0.0058, | |
| "step": 58160 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.5515362024307251, | |
| "eval_runtime": 445.8425, | |
| "eval_samples_per_second": 3.625, | |
| "eval_steps_per_second": 1.812, | |
| "eval_wer": 20.15791169835643, | |
| "step": 58160 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.14787828922271729, | |
| "learning_rate": 1.003576341127923e-06, | |
| "loss": 0.0029, | |
| "step": 65430 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.572388768196106, | |
| "eval_runtime": 443.4288, | |
| "eval_samples_per_second": 3.644, | |
| "eval_steps_per_second": 1.822, | |
| "eval_wer": 19.1212804812547, | |
| "step": 65430 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.0005218397127464414, | |
| "learning_rate": 3.851444291609354e-09, | |
| "loss": 0.0014, | |
| "step": 72700 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.5804322361946106, | |
| "eval_runtime": 443.5368, | |
| "eval_samples_per_second": 3.643, | |
| "eval_steps_per_second": 1.822, | |
| "eval_wer": 19.003115264797508, | |
| "step": 72700 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 72700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 1.48395953553408e+20, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |