| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.168413444163354, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.9819298879653056e-05, | |
| "loss": 2.6489, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.9638597759306108e-05, | |
| "loss": 1.7223, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1.9457896638959163e-05, | |
| "loss": 1.4874, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.9277195518612218e-05, | |
| "loss": 1.3632, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.909649439826527e-05, | |
| "loss": 1.2963, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.1749399900436401, | |
| "eval_runtime": 25.9529, | |
| "eval_samples_per_second": 415.522, | |
| "eval_steps_per_second": 12.985, | |
| "step": 2767 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.8915793277918324e-05, | |
| "loss": 1.1963, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.873509215757138e-05, | |
| "loss": 1.0687, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.8554391037224433e-05, | |
| "loss": 1.0932, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.8373689916877485e-05, | |
| "loss": 1.0549, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.819298879653054e-05, | |
| "loss": 1.0376, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.8012287676183594e-05, | |
| "loss": 1.0317, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.1026474237442017, | |
| "eval_runtime": 25.9642, | |
| "eval_samples_per_second": 415.342, | |
| "eval_steps_per_second": 12.979, | |
| "step": 5534 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.783158655583665e-05, | |
| "loss": 0.861, | |
| "step": 6000 | |
| } | |
| ], | |
| "max_steps": 55340, | |
| "num_train_epochs": 20, | |
| "total_flos": 1.115469291829248e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |