| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.869440459110473, |
| "global_step": 8000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.18, |
| "learning_rate": 4.7034911525585844e-05, |
| "loss": 2.3749, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.36, |
| "learning_rate": 4.4051889048302244e-05, |
| "loss": 1.2761, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.54, |
| "learning_rate": 4.1062888570062176e-05, |
| "loss": 0.9438, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.72, |
| "learning_rate": 3.8073888091822094e-05, |
| "loss": 0.7756, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.9, |
| "learning_rate": 3.508488761358202e-05, |
| "loss": 0.7032, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_bleu": 0.0, |
| "eval_gen_len": 1.0, |
| "eval_loss": 0.34736406803131104, |
| "eval_runtime": 639.8738, |
| "eval_samples_per_second": 4.357, |
| "eval_steps_per_second": 1.089, |
| "step": 2788 |
| }, |
| { |
| "epoch": 1.08, |
| "learning_rate": 3.209588713534194e-05, |
| "loss": 0.5899, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.26, |
| "learning_rate": 2.9106886657101868e-05, |
| "loss": 0.4577, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.43, |
| "learning_rate": 2.6123864179818268e-05, |
| "loss": 0.4712, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.61, |
| "learning_rate": 2.3134863701578193e-05, |
| "loss": 0.4424, |
| "step": 4500 |
| }, |
| { |
| "epoch": 1.79, |
| "learning_rate": 2.0145863223338114e-05, |
| "loss": 0.3883, |
| "step": 5000 |
| }, |
| { |
| "epoch": 1.97, |
| "learning_rate": 1.715686274509804e-05, |
| "loss": 0.4077, |
| "step": 5500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_bleu": 0.0, |
| "eval_gen_len": 1.0, |
| "eval_loss": 0.1796296387910843, |
| "eval_runtime": 41.4373, |
| "eval_samples_per_second": 67.282, |
| "eval_steps_per_second": 16.821, |
| "step": 5576 |
| }, |
| { |
| "epoch": 2.15, |
| "learning_rate": 1.4167862266857965e-05, |
| "loss": 0.2923, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.33, |
| "learning_rate": 1.1178861788617887e-05, |
| "loss": 0.253, |
| "step": 6500 |
| }, |
| { |
| "epoch": 2.51, |
| "learning_rate": 8.18986131037781e-06, |
| "loss": 0.277, |
| "step": 7000 |
| }, |
| { |
| "epoch": 2.69, |
| "learning_rate": 5.200860832137733e-06, |
| "loss": 0.2555, |
| "step": 7500 |
| }, |
| { |
| "epoch": 2.87, |
| "learning_rate": 2.2118603538976567e-06, |
| "loss": 0.2414, |
| "step": 8000 |
| } |
| ], |
| "max_steps": 8364, |
| "num_train_epochs": 3, |
| "total_flos": 2825850447544320.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
