{ "best_metric": null, "best_model_checkpoint": null, "epoch": 1.1614295994243224, "eval_steps": 500, "global_step": 160, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 1.9999928710990414e-05, "loss": 1.3025, "step": 1 }, { "epoch": 0.05, "learning_rate": 1.9971497912068014e-05, "loss": 1.0331, "step": 20 }, { "epoch": 0.1, "learning_rate": 1.9886154122075344e-05, "loss": 0.9573, "step": 40 }, { "epoch": 0.14, "learning_rate": 1.974445512526336e-05, "loss": 0.9535, "step": 60 }, { "epoch": 0.19, "learning_rate": 1.954720866508546e-05, "loss": 0.9354, "step": 80 }, { "epoch": 0.22, "eval_loss": 0.9210964441299438, "eval_runtime": 48.914, "eval_samples_per_second": 19.544, "eval_steps_per_second": 9.772, "step": 92 }, { "epoch": 1.02, "learning_rate": 1.9295539128732096e-05, "loss": 0.8984, "step": 100 }, { "epoch": 1.07, "learning_rate": 1.899088113765426e-05, "loss": 0.8842, "step": 120 }, { "epoch": 1.11, "learning_rate": 1.863497136962213e-05, "loss": 0.8736, "step": 140 }, { "epoch": 1.16, "learning_rate": 1.8229838658936566e-05, "loss": 0.8752, "step": 160 }, { "epoch": 1.16, "eval_loss": 0.9012895226478577, "eval_runtime": 48.6193, "eval_samples_per_second": 19.663, "eval_steps_per_second": 9.831, "step": 160 }, { "epoch": 1.16, "step": 160, "total_flos": 8.13446546367447e+16, "train_loss": 0.0, "train_runtime": 80.9781, "train_samples_per_second": 823.729, "train_steps_per_second": 20.549 } ], "logging_steps": 20, "max_steps": 1664, "num_train_epochs": 4, "save_steps": 40, "total_flos": 8.13446546367447e+16, "trial_name": null, "trial_params": null }