{ "best_metric": null, "best_model_checkpoint": null, "epoch": 11.80327868852459, "eval_steps": 500, "global_step": 18000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.32786885245901637, "grad_norm": 1.5182467699050903, "learning_rate": 9.838688524590164e-06, "loss": 0.3532, "step": 500 }, { "epoch": 0.6557377049180327, "grad_norm": 1.5529841184616089, "learning_rate": 9.674754098360658e-06, "loss": 0.2186, "step": 1000 }, { "epoch": 0.9836065573770492, "grad_norm": 1.2810306549072266, "learning_rate": 9.51081967213115e-06, "loss": 0.2085, "step": 1500 }, { "epoch": 1.3114754098360657, "grad_norm": 1.7146570682525635, "learning_rate": 9.34688524590164e-06, "loss": 0.1153, "step": 2000 }, { "epoch": 1.639344262295082, "grad_norm": 1.82291579246521, "learning_rate": 9.182950819672131e-06, "loss": 0.1108, "step": 2500 }, { "epoch": 1.9672131147540983, "grad_norm": 1.3579846620559692, "learning_rate": 9.019016393442623e-06, "loss": 0.1137, "step": 3000 }, { "epoch": 2.2950819672131146, "grad_norm": 1.2722268104553223, "learning_rate": 8.855081967213115e-06, "loss": 0.0613, "step": 3500 }, { "epoch": 2.6229508196721314, "grad_norm": 1.0057650804519653, "learning_rate": 8.691147540983608e-06, "loss": 0.0573, "step": 4000 }, { "epoch": 2.9508196721311473, "grad_norm": 1.5272691249847412, "learning_rate": 8.5272131147541e-06, "loss": 0.0585, "step": 4500 }, { "epoch": 3.278688524590164, "grad_norm": 0.9407837986946106, "learning_rate": 8.36327868852459e-06, "loss": 0.0391, "step": 5000 }, { "epoch": 3.6065573770491803, "grad_norm": 1.2755032777786255, "learning_rate": 8.199344262295082e-06, "loss": 0.0377, "step": 5500 }, { "epoch": 3.9344262295081966, "grad_norm": 0.9030640125274658, "learning_rate": 8.035409836065574e-06, "loss": 0.0405, "step": 6000 }, { "epoch": 4.262295081967213, "grad_norm": 0.8674460649490356, "learning_rate": 7.871475409836066e-06, "loss": 0.0298, "step": 6500 }, { "epoch": 4.590163934426229, "grad_norm": 0.7920001745223999, "learning_rate": 7.707540983606559e-06, "loss": 0.0277, "step": 7000 }, { "epoch": 4.918032786885246, "grad_norm": 0.5429362654685974, "learning_rate": 7.543934426229509e-06, "loss": 0.0295, "step": 7500 }, { "epoch": 5.245901639344262, "grad_norm": 0.613649845123291, "learning_rate": 7.3800000000000005e-06, "loss": 0.0257, "step": 8000 }, { "epoch": 5.573770491803279, "grad_norm": 0.6390576362609863, "learning_rate": 7.216065573770492e-06, "loss": 0.0248, "step": 8500 }, { "epoch": 5.901639344262295, "grad_norm": 0.6235594153404236, "learning_rate": 7.052131147540984e-06, "loss": 0.025, "step": 9000 }, { "epoch": 6.229508196721311, "grad_norm": 0.5020778179168701, "learning_rate": 6.888196721311477e-06, "loss": 0.0241, "step": 9500 }, { "epoch": 6.557377049180328, "grad_norm": 0.7309982776641846, "learning_rate": 6.724262295081968e-06, "loss": 0.0227, "step": 10000 }, { "epoch": 6.885245901639344, "grad_norm": 0.38623395562171936, "learning_rate": 6.5603278688524594e-06, "loss": 0.0229, "step": 10500 }, { "epoch": 7.213114754098361, "grad_norm": 0.5087317228317261, "learning_rate": 6.396393442622951e-06, "loss": 0.0227, "step": 11000 }, { "epoch": 7.540983606557377, "grad_norm": 0.5436781048774719, "learning_rate": 6.232459016393443e-06, "loss": 0.021, "step": 11500 }, { "epoch": 7.868852459016393, "grad_norm": 0.756500780582428, "learning_rate": 6.068524590163935e-06, "loss": 0.0218, "step": 12000 }, { "epoch": 8.19672131147541, "grad_norm": 0.20513100922107697, "learning_rate": 5.9045901639344274e-06, "loss": 0.0206, "step": 12500 }, { "epoch": 8.524590163934427, "grad_norm": 0.6370705366134644, "learning_rate": 5.740655737704918e-06, "loss": 0.0194, "step": 13000 }, { "epoch": 8.852459016393443, "grad_norm": 0.509864091873169, "learning_rate": 5.577049180327869e-06, "loss": 0.0207, "step": 13500 }, { "epoch": 9.180327868852459, "grad_norm": 0.4402160942554474, "learning_rate": 5.413114754098361e-06, "loss": 0.0194, "step": 14000 }, { "epoch": 9.508196721311476, "grad_norm": 0.29329636693000793, "learning_rate": 5.2491803278688525e-06, "loss": 0.0188, "step": 14500 }, { "epoch": 9.836065573770492, "grad_norm": 0.25181844830513, "learning_rate": 5.085245901639345e-06, "loss": 0.0194, "step": 15000 }, { "epoch": 10.163934426229508, "grad_norm": 0.36046847701072693, "learning_rate": 4.921311475409836e-06, "loss": 0.0188, "step": 15500 }, { "epoch": 10.491803278688524, "grad_norm": 0.3692318797111511, "learning_rate": 4.757377049180329e-06, "loss": 0.0183, "step": 16000 }, { "epoch": 10.819672131147541, "grad_norm": 0.6048111319541931, "learning_rate": 4.59344262295082e-06, "loss": 0.0185, "step": 16500 }, { "epoch": 11.147540983606557, "grad_norm": 0.26122042536735535, "learning_rate": 4.4295081967213114e-06, "loss": 0.018, "step": 17000 }, { "epoch": 11.475409836065573, "grad_norm": 0.30455446243286133, "learning_rate": 4.265573770491804e-06, "loss": 0.0174, "step": 17500 }, { "epoch": 11.80327868852459, "grad_norm": 0.2333904653787613, "learning_rate": 4.101639344262295e-06, "loss": 0.0179, "step": 18000 } ], "logging_steps": 500, "max_steps": 30500, "num_input_tokens_seen": 0, "num_train_epochs": 20, "save_steps": 1000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 1.17961666854912e+18, "train_batch_size": 2, "trial_name": null, "trial_params": null }