| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.46129450771226754, | |
| "eval_steps": 500, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "ep_loss": 4.7343, | |
| "epoch": 0.04, | |
| "learning_rate": 2.5e-05, | |
| "loss": 11.6218, | |
| "mlm_loss": 6.8875, | |
| "step": 500 | |
| }, | |
| { | |
| "ep_loss": 0.4315, | |
| "epoch": 0.08, | |
| "learning_rate": 5e-05, | |
| "loss": 3.0285, | |
| "mlm_loss": 2.5969, | |
| "step": 1000 | |
| }, | |
| { | |
| "ep_loss": 0.3811, | |
| "epoch": 0.12, | |
| "learning_rate": 7.5e-05, | |
| "loss": 2.2571, | |
| "mlm_loss": 1.8761, | |
| "step": 1500 | |
| }, | |
| { | |
| "ep_loss": 0.37, | |
| "epoch": 0.15, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9695, | |
| "mlm_loss": 1.5994, | |
| "step": 2000 | |
| }, | |
| { | |
| "ep_loss": 0.3813, | |
| "epoch": 0.19, | |
| "learning_rate": 0.00012495, | |
| "loss": 1.5597, | |
| "mlm_loss": 1.1785, | |
| "step": 2500 | |
| }, | |
| { | |
| "ep_loss": 0.3624, | |
| "epoch": 0.23, | |
| "learning_rate": 0.00014995, | |
| "loss": 1.4228, | |
| "mlm_loss": 1.0604, | |
| "step": 3000 | |
| }, | |
| { | |
| "ep_loss": 0.338, | |
| "epoch": 0.27, | |
| "learning_rate": 0.00017495, | |
| "loss": 1.3108, | |
| "mlm_loss": 0.9729, | |
| "step": 3500 | |
| }, | |
| { | |
| "ep_loss": 0.3524, | |
| "epoch": 0.31, | |
| "learning_rate": 0.00019994999999999998, | |
| "loss": 1.2734, | |
| "mlm_loss": 0.921, | |
| "step": 4000 | |
| }, | |
| { | |
| "ep_loss": 0.3423, | |
| "epoch": 0.35, | |
| "learning_rate": 0.00022495000000000002, | |
| "loss": 1.217, | |
| "mlm_loss": 0.8747, | |
| "step": 4500 | |
| }, | |
| { | |
| "ep_loss": 0.3229, | |
| "epoch": 0.38, | |
| "learning_rate": 0.0002499, | |
| "loss": 1.1373, | |
| "mlm_loss": 0.8144, | |
| "step": 5000 | |
| }, | |
| { | |
| "ep_loss": 0.3326, | |
| "epoch": 0.42, | |
| "learning_rate": 0.00027489999999999996, | |
| "loss": 1.117, | |
| "mlm_loss": 0.7844, | |
| "step": 5500 | |
| }, | |
| { | |
| "ep_loss": 0.332, | |
| "epoch": 0.46, | |
| "learning_rate": 0.00029985, | |
| "loss": 1.0958, | |
| "mlm_loss": 0.7638, | |
| "step": 6000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 520240, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 5.728903631701279e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |