| { | |
| "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": 0.1726, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0862, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.079, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.0733, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.0722, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0715, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.42010074853897095, | |
| "eval_runtime": 43.6469, | |
| "eval_samples_per_second": 75.103, | |
| "eval_steps_per_second": 18.787, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.0683, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0671, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.067, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0653, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0652, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0646, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_loss": 0.43611547350883484, | |
| "eval_runtime": 45.1004, | |
| "eval_samples_per_second": 72.682, | |
| "eval_steps_per_second": 18.182, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.0632, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0639, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0629, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0626, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0615, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0623, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "eval_loss": 0.43588343262672424, | |
| "eval_runtime": 43.624, | |
| "eval_samples_per_second": 75.142, | |
| "eval_steps_per_second": 18.797, | |
| "step": 9000 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 7, | |
| "total_flos": 5.642598943011226e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |