| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "global_step": 2200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 13.64, | |
| "learning_rate": 2.99e-05, | |
| "loss": 7.5798, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "eval_loss": 3.4349324703216553, | |
| "eval_runtime": 13.2159, | |
| "eval_samples_per_second": 22.776, | |
| "eval_steps_per_second": 2.875, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 5.989999999999999e-05, | |
| "loss": 3.1252, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "eval_loss": 3.0706284046173096, | |
| "eval_runtime": 13.3085, | |
| "eval_samples_per_second": 22.617, | |
| "eval_steps_per_second": 2.855, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 40.91, | |
| "learning_rate": 6.729310344827586e-05, | |
| "loss": 2.2546, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 40.91, | |
| "eval_loss": 0.8426555395126343, | |
| "eval_runtime": 12.9127, | |
| "eval_samples_per_second": 23.31, | |
| "eval_steps_per_second": 2.943, | |
| "eval_wer": 0.7762882279011312, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 54.55, | |
| "learning_rate": 5.177586206896551e-05, | |
| "loss": 0.7564, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 54.55, | |
| "eval_loss": 0.6129118800163269, | |
| "eval_runtime": 13.2137, | |
| "eval_samples_per_second": 22.779, | |
| "eval_steps_per_second": 2.876, | |
| "eval_wer": 0.637620444072057, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 68.18, | |
| "learning_rate": 3.625862068965517e-05, | |
| "loss": 0.5239, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 68.18, | |
| "eval_loss": 0.5769144892692566, | |
| "eval_runtime": 12.822, | |
| "eval_samples_per_second": 23.475, | |
| "eval_steps_per_second": 2.964, | |
| "eval_wer": 0.6036866359447005, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 81.82, | |
| "learning_rate": 2.074137931034483e-05, | |
| "loss": 0.438, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 81.82, | |
| "eval_loss": 0.5937696099281311, | |
| "eval_runtime": 13.4551, | |
| "eval_samples_per_second": 22.371, | |
| "eval_steps_per_second": 2.824, | |
| "eval_wer": 0.5915374947633012, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 95.45, | |
| "learning_rate": 5.224137931034482e-06, | |
| "loss": 0.3945, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 95.45, | |
| "eval_loss": 0.5868746042251587, | |
| "eval_runtime": 13.1379, | |
| "eval_samples_per_second": 22.911, | |
| "eval_steps_per_second": 2.892, | |
| "eval_wer": 0.5860913280268119, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 2200, | |
| "total_flos": 1.3980125790314312e+19, | |
| "train_loss": 2.0729129652543503, | |
| "train_runtime": 5588.1144, | |
| "train_samples_per_second": 12.598, | |
| "train_steps_per_second": 0.394 | |
| } | |
| ], | |
| "max_steps": 2200, | |
| "num_train_epochs": 100, | |
| "total_flos": 1.3980125790314312e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |