| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 6.028131279303416, |
| "global_step": 9000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.33, |
| "learning_rate": 2.5e-05, |
| "loss": 0.2041, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.67, |
| "learning_rate": 5e-05, |
| "loss": 0.0505, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "learning_rate": 4.722222222222222e-05, |
| "loss": 0.036, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.34, |
| "learning_rate": 4.4444444444444447e-05, |
| "loss": 0.0217, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.67, |
| "learning_rate": 4.166666666666667e-05, |
| "loss": 0.0214, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.01, |
| "learning_rate": 3.888888888888889e-05, |
| "loss": 0.0181, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.01, |
| "eval_loss": 1.230069875717163, |
| "eval_runtime": 43.3592, |
| "eval_samples_per_second": 75.601, |
| "eval_steps_per_second": 18.912, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.34, |
| "learning_rate": 3.611111111111111e-05, |
| "loss": 0.0115, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.68, |
| "learning_rate": 3.3333333333333335e-05, |
| "loss": 0.0106, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.01, |
| "learning_rate": 3.055555555555556e-05, |
| "loss": 0.0111, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.35, |
| "learning_rate": 2.777777777777778e-05, |
| "loss": 0.0058, |
| "step": 5000 |
| }, |
| { |
| "epoch": 3.68, |
| "learning_rate": 2.5e-05, |
| "loss": 0.0067, |
| "step": 5500 |
| }, |
| { |
| "epoch": 4.02, |
| "learning_rate": 2.2222222222222223e-05, |
| "loss": 0.0059, |
| "step": 6000 |
| }, |
| { |
| "epoch": 4.02, |
| "eval_loss": 1.4619778394699097, |
| "eval_runtime": 43.1064, |
| "eval_samples_per_second": 76.044, |
| "eval_steps_per_second": 19.023, |
| "step": 6000 |
| }, |
| { |
| "epoch": 4.35, |
| "learning_rate": 1.9444444444444445e-05, |
| "loss": 0.0036, |
| "step": 6500 |
| }, |
| { |
| "epoch": 4.69, |
| "learning_rate": 1.6666666666666667e-05, |
| "loss": 0.0034, |
| "step": 7000 |
| }, |
| { |
| "epoch": 5.02, |
| "learning_rate": 1.388888888888889e-05, |
| "loss": 0.0031, |
| "step": 7500 |
| }, |
| { |
| "epoch": 5.36, |
| "learning_rate": 1.1111111111111112e-05, |
| "loss": 0.0019, |
| "step": 8000 |
| }, |
| { |
| "epoch": 5.69, |
| "learning_rate": 8.333333333333334e-06, |
| "loss": 0.0019, |
| "step": 8500 |
| }, |
| { |
| "epoch": 6.03, |
| "learning_rate": 5.555555555555556e-06, |
| "loss": 0.0019, |
| "step": 9000 |
| }, |
| { |
| "epoch": 6.03, |
| "eval_loss": 1.6374868154525757, |
| "eval_runtime": 43.607, |
| "eval_samples_per_second": 75.171, |
| "eval_steps_per_second": 18.804, |
| "step": 9000 |
| } |
| ], |
| "max_steps": 10000, |
| "num_train_epochs": 7, |
| "total_flos": 5.642598943011226e+16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
