| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.8125, |
| "global_step": 180, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.16, |
| "learning_rate": 4.739583333333333e-05, |
| "loss": 0.2329, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 4.4791666666666673e-05, |
| "loss": 0.0657, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.47, |
| "learning_rate": 4.21875e-05, |
| "loss": 0.0442, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.47, |
| "eval_cer": 0.09341614906832298, |
| "eval_loss": 0.03504600375890732, |
| "eval_runtime": 433.0095, |
| "eval_samples_per_second": 1.18, |
| "eval_steps_per_second": 0.037, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 3.958333333333333e-05, |
| "loss": 0.0327, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.78, |
| "learning_rate": 3.697916666666667e-05, |
| "loss": 0.0316, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.94, |
| "learning_rate": 3.4375e-05, |
| "loss": 0.0242, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.94, |
| "eval_cer": 0.02981366459627329, |
| "eval_loss": 0.02102348767220974, |
| "eval_runtime": 445.6571, |
| "eval_samples_per_second": 1.147, |
| "eval_steps_per_second": 0.036, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.09, |
| "learning_rate": 3.177083333333333e-05, |
| "loss": 0.0252, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.25, |
| "learning_rate": 2.916666666666667e-05, |
| "loss": 0.0208, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.41, |
| "learning_rate": 2.6562500000000002e-05, |
| "loss": 0.0203, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.41, |
| "eval_cer": 0.01937888198757764, |
| "eval_loss": 0.018760638311505318, |
| "eval_runtime": 489.631, |
| "eval_samples_per_second": 1.044, |
| "eval_steps_per_second": 0.033, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.56, |
| "learning_rate": 2.3958333333333334e-05, |
| "loss": 0.02, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.72, |
| "learning_rate": 2.1354166666666666e-05, |
| "loss": 0.0178, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.88, |
| "learning_rate": 1.8750000000000002e-05, |
| "loss": 0.0169, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.88, |
| "eval_cer": 0.010186335403726708, |
| "eval_loss": 0.01577191799879074, |
| "eval_runtime": 572.6491, |
| "eval_samples_per_second": 0.892, |
| "eval_steps_per_second": 0.028, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.03, |
| "learning_rate": 1.6145833333333334e-05, |
| "loss": 0.0161, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.19, |
| "learning_rate": 1.3541666666666666e-05, |
| "loss": 0.0151, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.34, |
| "learning_rate": 1.09375e-05, |
| "loss": 0.0151, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.34, |
| "eval_cer": 0.0084472049689441, |
| "eval_loss": 0.014565763995051384, |
| "eval_runtime": 618.7531, |
| "eval_samples_per_second": 0.826, |
| "eval_steps_per_second": 0.026, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.5, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 0.015, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.66, |
| "learning_rate": 5.729166666666667e-06, |
| "loss": 0.0139, |
| "step": 170 |
| }, |
| { |
| "epoch": 2.81, |
| "learning_rate": 3.125e-06, |
| "loss": 0.0133, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.81, |
| "eval_cer": 0.007950310559006211, |
| "eval_loss": 0.014070576056838036, |
| "eval_runtime": 564.3264, |
| "eval_samples_per_second": 0.906, |
| "eval_steps_per_second": 0.028, |
| "step": 180 |
| } |
| ], |
| "max_steps": 192, |
| "num_train_epochs": 3, |
| "total_flos": 5.084609259497325e+18, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
