| { | |
| "best_metric": 0.9914958477020264, | |
| "best_model_checkpoint": "test-squad/checkpoint-6000", | |
| "epoch": 2.1849963583394025, | |
| "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": 1.9635833940276766e-05, | |
| "loss": 1.2131, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.9271667880553534e-05, | |
| "loss": 1.2772, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.8907501820830302e-05, | |
| "loss": 1.1877, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.8543335761107067e-05, | |
| "loss": 1.1523, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.0681313276290894, | |
| "eval_runtime": 119.1833, | |
| "eval_samples_per_second": 89.417, | |
| "eval_steps_per_second": 2.802, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.8179169701383832e-05, | |
| "loss": 1.0754, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1.7815003641660596e-05, | |
| "loss": 0.987, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.7450837581937364e-05, | |
| "loss": 0.9121, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 1.7086671522214133e-05, | |
| "loss": 0.8929, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 0.9955193996429443, | |
| "eval_runtime": 56.1478, | |
| "eval_samples_per_second": 189.803, | |
| "eval_steps_per_second": 5.949, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 1.6722505462490897e-05, | |
| "loss": 0.8909, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.6358339402767662e-05, | |
| "loss": 0.8772, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.599417334304443e-05, | |
| "loss": 0.8696, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1.5630007283321195e-05, | |
| "loss": 0.7111, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_loss": 0.9914958477020264, | |
| "eval_runtime": 56.0391, | |
| "eval_samples_per_second": 190.171, | |
| "eval_steps_per_second": 5.96, | |
| "step": 6000 | |
| } | |
| ], | |
| "max_steps": 27460, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.63649378451456e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |