{ "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "global_step": 16425, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.09, "learning_rate": 0.00029086757990867576, "loss": 0.4741, "step": 500 }, { "epoch": 0.18, "learning_rate": 0.0002817351598173516, "loss": 0.2543, "step": 1000 }, { "epoch": 0.27, "learning_rate": 0.0002726027397260274, "loss": 0.2336, "step": 1500 }, { "epoch": 0.37, "learning_rate": 0.00026347031963470317, "loss": 0.21, "step": 2000 }, { "epoch": 0.46, "learning_rate": 0.00025433789954337895, "loss": 0.2007, "step": 2500 }, { "epoch": 0.55, "learning_rate": 0.0002452054794520548, "loss": 0.183, "step": 3000 }, { "epoch": 0.64, "learning_rate": 0.00023607305936073058, "loss": 0.1879, "step": 3500 }, { "epoch": 0.73, "learning_rate": 0.00022694063926940636, "loss": 0.1723, "step": 4000 }, { "epoch": 0.82, "learning_rate": 0.00021780821917808218, "loss": 0.1769, "step": 4500 }, { "epoch": 0.91, "learning_rate": 0.000208675799086758, "loss": 0.1656, "step": 5000 }, { "epoch": 1.0, "learning_rate": 0.00019954337899543377, "loss": 0.1722, "step": 5500 }, { "epoch": 1.1, "learning_rate": 0.0001904109589041096, "loss": 0.1434, "step": 6000 }, { "epoch": 1.19, "learning_rate": 0.00018127853881278537, "loss": 0.1427, "step": 6500 }, { "epoch": 1.28, "learning_rate": 0.00017214611872146118, "loss": 0.1394, "step": 7000 }, { "epoch": 1.37, "learning_rate": 0.00016301369863013697, "loss": 0.1316, "step": 7500 }, { "epoch": 1.46, "learning_rate": 0.00015388127853881278, "loss": 0.1404, "step": 8000 }, { "epoch": 1.55, "learning_rate": 0.00014474885844748857, "loss": 0.1345, "step": 8500 }, { "epoch": 1.64, "learning_rate": 0.00013561643835616435, "loss": 0.136, "step": 9000 }, { "epoch": 1.74, "learning_rate": 0.00012648401826484017, "loss": 0.1333, "step": 9500 }, { "epoch": 1.83, "learning_rate": 0.00011735159817351596, "loss": 0.1314, "step": 10000 }, { "epoch": 1.92, "learning_rate": 0.00010821917808219176, "loss": 0.1297, "step": 10500 }, { "epoch": 2.01, "learning_rate": 9.908675799086756e-05, "loss": 0.1235, "step": 11000 }, { "epoch": 2.1, "learning_rate": 8.995433789954336e-05, "loss": 0.1053, "step": 11500 }, { "epoch": 2.19, "learning_rate": 8.082191780821917e-05, "loss": 0.108, "step": 12000 }, { "epoch": 2.28, "learning_rate": 7.168949771689497e-05, "loss": 0.104, "step": 12500 }, { "epoch": 2.37, "learning_rate": 6.255707762557077e-05, "loss": 0.105, "step": 13000 }, { "epoch": 2.47, "learning_rate": 5.342465753424657e-05, "loss": 0.1053, "step": 13500 }, { "epoch": 2.56, "learning_rate": 4.429223744292237e-05, "loss": 0.1029, "step": 14000 }, { "epoch": 2.65, "learning_rate": 3.515981735159817e-05, "loss": 0.103, "step": 14500 }, { "epoch": 2.74, "learning_rate": 2.602739726027397e-05, "loss": 0.1012, "step": 15000 }, { "epoch": 2.83, "learning_rate": 1.6894977168949768e-05, "loss": 0.0993, "step": 15500 }, { "epoch": 2.92, "learning_rate": 7.76255707762557e-06, "loss": 0.0973, "step": 16000 } ], "max_steps": 16425, "num_train_epochs": 3, "total_flos": 4.828892652872663e+17, "trial_name": null, "trial_params": null }