| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.3311258278145695, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.9999999999999996e-05, | |
| "loss": 13.2048, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 7.46938775510204e-05, | |
| "loss": 26.5657, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 17.199691772460938, | |
| "eval_runtime": 5.248, | |
| "eval_samples_per_second": 18.674, | |
| "eval_steps_per_second": 0.762, | |
| "eval_wer": 1.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 7.36734693877551e-05, | |
| "loss": 8.0091, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 7.265306122448979e-05, | |
| "loss": 7.6332, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 4.783994197845459, | |
| "eval_runtime": 4.9818, | |
| "eval_samples_per_second": 19.672, | |
| "eval_steps_per_second": 0.803, | |
| "eval_wer": 1.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 7.163265306122449e-05, | |
| "loss": 3.8261, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 7.061224489795918e-05, | |
| "loss": 3.8717, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_loss": 3.6783852577209473, | |
| "eval_runtime": 4.9141, | |
| "eval_samples_per_second": 19.943, | |
| "eval_steps_per_second": 0.814, | |
| "eval_wer": 1.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 6.959183673469387e-05, | |
| "loss": 3.7607, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 6.857142857142857e-05, | |
| "loss": 3.638, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 3.4979612827301025, | |
| "eval_runtime": 5.0758, | |
| "eval_samples_per_second": 19.307, | |
| "eval_steps_per_second": 0.788, | |
| "eval_wer": 1.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 6.755102040816325e-05, | |
| "loss": 3.4601, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 6.653061224489796e-05, | |
| "loss": 3.4138, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 3.489065170288086, | |
| "eval_runtime": 4.9074, | |
| "eval_samples_per_second": 19.97, | |
| "eval_steps_per_second": 0.815, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| } | |
| ], | |
| "max_steps": 750, | |
| "num_train_epochs": 10, | |
| "total_flos": 2.00880445675488e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |