{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 4.947040498442368, "eval_steps": 500, "global_step": 400, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.12461059190031153, "grad_norm": 2.996091481245303, "learning_rate": 2.5e-06, "loss": 0.441, "step": 10 }, { "epoch": 0.24922118380062305, "grad_norm": 1.509457164389346, "learning_rate": 5e-06, "loss": 0.3384, "step": 20 }, { "epoch": 0.37383177570093457, "grad_norm": 1.5540882760833508, "learning_rate": 7.500000000000001e-06, "loss": 0.2443, "step": 30 }, { "epoch": 0.4984423676012461, "grad_norm": 0.5837019331950773, "learning_rate": 1e-05, "loss": 0.2139, "step": 40 }, { "epoch": 0.6230529595015576, "grad_norm": 0.25066734190038364, "learning_rate": 9.980973490458728e-06, "loss": 0.1431, "step": 50 }, { "epoch": 0.7476635514018691, "grad_norm": 0.3847855364487568, "learning_rate": 9.924038765061042e-06, "loss": 0.1575, "step": 60 }, { "epoch": 0.8722741433021807, "grad_norm": 0.5863140592759178, "learning_rate": 9.829629131445342e-06, "loss": 0.1293, "step": 70 }, { "epoch": 0.9968847352024922, "grad_norm": 0.24163113888653256, "learning_rate": 9.698463103929542e-06, "loss": 0.1275, "step": 80 }, { "epoch": 1.1121495327102804, "grad_norm": 0.17545609714785104, "learning_rate": 9.531538935183252e-06, "loss": 0.1211, "step": 90 }, { "epoch": 1.236760124610592, "grad_norm": 0.4129092473453207, "learning_rate": 9.330127018922195e-06, "loss": 0.1112, "step": 100 }, { "epoch": 1.3613707165109035, "grad_norm": 0.17027543196447287, "learning_rate": 9.09576022144496e-06, "loss": 0.0986, "step": 110 }, { "epoch": 1.485981308411215, "grad_norm": 0.38351119544670076, "learning_rate": 8.83022221559489e-06, "loss": 0.099, "step": 120 }, { "epoch": 1.6105919003115265, "grad_norm": 0.35022981475765697, "learning_rate": 8.535533905932739e-06, "loss": 0.1018, "step": 130 }, { "epoch": 1.735202492211838, "grad_norm": 0.21063308126485916, "learning_rate": 8.213938048432697e-06, "loss": 0.108, "step": 140 }, { "epoch": 1.8598130841121496, "grad_norm": 0.320753628121036, "learning_rate": 7.86788218175523e-06, "loss": 0.0896, "step": 150 }, { "epoch": 1.9844236760124612, "grad_norm": 0.17394262959246917, "learning_rate": 7.500000000000001e-06, "loss": 0.0899, "step": 160 }, { "epoch": 2.0996884735202492, "grad_norm": 0.1604164195468027, "learning_rate": 7.113091308703498e-06, "loss": 0.0743, "step": 170 }, { "epoch": 2.2242990654205608, "grad_norm": 0.1653097854719759, "learning_rate": 6.710100716628345e-06, "loss": 0.0698, "step": 180 }, { "epoch": 2.3489096573208723, "grad_norm": 0.17110336957386751, "learning_rate": 6.294095225512604e-06, "loss": 0.0642, "step": 190 }, { "epoch": 2.473520249221184, "grad_norm": 0.19194873459564457, "learning_rate": 5.8682408883346535e-06, "loss": 0.0768, "step": 200 }, { "epoch": 2.5981308411214954, "grad_norm": 0.18685473707282946, "learning_rate": 5.435778713738292e-06, "loss": 0.0662, "step": 210 }, { "epoch": 2.722741433021807, "grad_norm": 0.1552135550801311, "learning_rate": 5e-06, "loss": 0.0773, "step": 220 }, { "epoch": 2.8473520249221185, "grad_norm": 0.15189156158692, "learning_rate": 4.564221286261709e-06, "loss": 0.0641, "step": 230 }, { "epoch": 2.97196261682243, "grad_norm": 0.1416868449041196, "learning_rate": 4.131759111665349e-06, "loss": 0.0586, "step": 240 }, { "epoch": 3.087227414330218, "grad_norm": 0.15860045704097103, "learning_rate": 3.705904774487396e-06, "loss": 0.0621, "step": 250 }, { "epoch": 3.2118380062305296, "grad_norm": 0.1909754560694067, "learning_rate": 3.289899283371657e-06, "loss": 0.048, "step": 260 }, { "epoch": 3.336448598130841, "grad_norm": 0.14738345978872078, "learning_rate": 2.886908691296504e-06, "loss": 0.0506, "step": 270 }, { "epoch": 3.4610591900311527, "grad_norm": 0.1636519624319427, "learning_rate": 2.5000000000000015e-06, "loss": 0.0499, "step": 280 }, { "epoch": 3.5856697819314642, "grad_norm": 0.18481461929920587, "learning_rate": 2.132117818244771e-06, "loss": 0.05, "step": 290 }, { "epoch": 3.710280373831776, "grad_norm": 0.1564034342013883, "learning_rate": 1.7860619515673034e-06, "loss": 0.0504, "step": 300 }, { "epoch": 3.8348909657320873, "grad_norm": 0.2007406802949044, "learning_rate": 1.4644660940672628e-06, "loss": 0.0474, "step": 310 }, { "epoch": 3.959501557632399, "grad_norm": 0.19659343625181785, "learning_rate": 1.1697777844051105e-06, "loss": 0.0456, "step": 320 }, { "epoch": 4.074766355140187, "grad_norm": 0.1289550110072798, "learning_rate": 9.042397785550405e-07, "loss": 0.0533, "step": 330 }, { "epoch": 4.1993769470404985, "grad_norm": 0.14424785730829176, "learning_rate": 6.698729810778065e-07, "loss": 0.0383, "step": 340 }, { "epoch": 4.32398753894081, "grad_norm": 0.16849835996827525, "learning_rate": 4.6846106481675035e-07, "loss": 0.0472, "step": 350 }, { "epoch": 4.4485981308411215, "grad_norm": 0.0886043880713109, "learning_rate": 3.015368960704584e-07, "loss": 0.0412, "step": 360 }, { "epoch": 4.573208722741433, "grad_norm": 0.16168692516023012, "learning_rate": 1.7037086855465902e-07, "loss": 0.0432, "step": 370 }, { "epoch": 4.697819314641745, "grad_norm": 0.13391205742884732, "learning_rate": 7.59612349389599e-08, "loss": 0.0415, "step": 380 }, { "epoch": 4.822429906542056, "grad_norm": 0.1185627294691602, "learning_rate": 1.9026509541272276e-08, "loss": 0.0373, "step": 390 }, { "epoch": 4.947040498442368, "grad_norm": 0.14371848122220537, "learning_rate": 0.0, "loss": 0.0453, "step": 400 }, { "epoch": 4.947040498442368, "step": 400, "total_flos": 3.1025515205545165e+17, "train_loss": 0.09792151391506194, "train_runtime": 85393.5514, "train_samples_per_second": 0.112, "train_steps_per_second": 0.005 } ], "logging_steps": 10, "max_steps": 400, "num_input_tokens_seen": 0, "num_train_epochs": 5, "save_steps": 200, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 3.1025515205545165e+17, "train_batch_size": 1, "trial_name": null, "trial_params": null }