| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 3.5624999999999998e-06, | |
| "loss": 137.0348, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 7.275e-06, | |
| "loss": 106.075, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.1024999999999999e-05, | |
| "loss": 95.1437, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.4775e-05, | |
| "loss": 88.0386, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 1.8525e-05, | |
| "loss": 82.628, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 79.55873107910156, | |
| "eval_runtime": 25.5475, | |
| "eval_samples_per_second": 19.297, | |
| "eval_steps_per_second": 9.668, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2.2274999999999996e-05, | |
| "loss": 73.1146, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 2.6024999999999996e-05, | |
| "loss": 62.7571, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 2.9775e-05, | |
| "loss": 48.9782, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 3.3524999999999995e-05, | |
| "loss": 33.2033, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.7275e-05, | |
| "loss": 17.5635, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 11.592885971069336, | |
| "eval_runtime": 25.2356, | |
| "eval_samples_per_second": 19.536, | |
| "eval_steps_per_second": 9.788, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 4.1025e-05, | |
| "loss": 8.6246, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4775e-05, | |
| "loss": 6.5245, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 1200, | |
| "total_flos": 1.2304867447957263e+18, | |
| "train_loss": 63.30716496785482, | |
| "train_runtime": 1396.5226, | |
| "train_samples_per_second": 27.411, | |
| "train_steps_per_second": 0.859 | |
| } | |
| ], | |
| "max_steps": 1200, | |
| "num_train_epochs": 6, | |
| "total_flos": 1.2304867447957263e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |