{ "best_metric": null, "best_model_checkpoint": null, "epoch": 4.999816075041383, "eval_steps": 500, "global_step": 3398, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.74, "learning_rate": 1.7565046637211588e-05, "loss": 1.4025, "step": 500 }, { "epoch": 1.0, "eval_loss": 1.291014552116394, "eval_runtime": 49.8857, "eval_samples_per_second": 238.325, "eval_steps_per_second": 14.914, "step": 679 }, { "epoch": 1.47, "learning_rate": 1.5110456553755524e-05, "loss": 0.8512, "step": 1000 }, { "epoch": 2.0, "eval_loss": 1.2626343965530396, "eval_runtime": 49.871, "eval_samples_per_second": 238.395, "eval_steps_per_second": 14.918, "step": 1359 }, { "epoch": 2.21, "learning_rate": 1.2665684830633284e-05, "loss": 0.7231, "step": 1500 }, { "epoch": 2.94, "learning_rate": 1.0211094747177222e-05, "loss": 0.6013, "step": 2000 }, { "epoch": 3.0, "eval_loss": 1.350563883781433, "eval_runtime": 49.8485, "eval_samples_per_second": 238.503, "eval_steps_per_second": 14.925, "step": 2038 }, { "epoch": 3.68, "learning_rate": 7.756504663721159e-06, "loss": 0.4873, "step": 2500 }, { "epoch": 4.0, "eval_loss": 1.428869605064392, "eval_runtime": 49.815, "eval_samples_per_second": 238.663, "eval_steps_per_second": 14.935, "step": 2718 }, { "epoch": 4.41, "learning_rate": 5.301914580265096e-06, "loss": 0.4222, "step": 3000 }, { "epoch": 5.0, "eval_loss": 1.6059815883636475, "eval_runtime": 49.8621, "eval_samples_per_second": 238.437, "eval_steps_per_second": 14.921, "step": 3398 } ], "logging_steps": 500, "max_steps": 4074, "num_train_epochs": 6, "save_steps": 500, "total_flos": 3.0293145111101184e+17, "trial_name": null, "trial_params": null }