| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.997402597402598, | |
| "global_step": 576, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 18.22025489807129, | |
| "eval_runtime": 188.3513, | |
| "eval_samples_per_second": 26.785, | |
| "eval_steps_per_second": 0.839, | |
| "eval_wer": 1.0021133629565406, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 1.5156249999999998e-06, | |
| "loss": 15.219, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 17.49703025817871, | |
| "eval_runtime": 186.0672, | |
| "eval_samples_per_second": 27.114, | |
| "eval_steps_per_second": 0.849, | |
| "eval_wer": 1.0, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3.0781249999999994e-06, | |
| "loss": 14.7046, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 11.056963920593262, | |
| "eval_runtime": 184.9586, | |
| "eval_samples_per_second": 27.276, | |
| "eval_steps_per_second": 0.854, | |
| "eval_wer": 1.0, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 4.640624999999999e-06, | |
| "loss": 11.6089, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 5.172672271728516, | |
| "eval_runtime": 185.1159, | |
| "eval_samples_per_second": 27.253, | |
| "eval_steps_per_second": 0.854, | |
| "eval_wer": 1.0, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 6.203124999999999e-06, | |
| "loss": 5.9408, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 3.6944665908813477, | |
| "eval_runtime": 186.0698, | |
| "eval_samples_per_second": 27.113, | |
| "eval_steps_per_second": 0.849, | |
| "eval_wer": 1.0, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 7.765624999999999e-06, | |
| "loss": 3.8196, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 3.3380837440490723, | |
| "eval_runtime": 183.9322, | |
| "eval_samples_per_second": 27.429, | |
| "eval_steps_per_second": 0.859, | |
| "eval_wer": 1.0, | |
| "step": 576 | |
| } | |
| ], | |
| "max_steps": 19200, | |
| "num_train_epochs": 200, | |
| "total_flos": 8.945907324105985e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |