| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "global_step": 14794, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 2.8986075435987562e-05, | |
| "loss": 2.1093, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.7972150871975126e-05, | |
| "loss": 1.015, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.695822630796269e-05, | |
| "loss": 0.991, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.594430174395025e-05, | |
| "loss": 0.9608, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 2.4930377179937813e-05, | |
| "loss": 0.9279, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 2.3916452615925377e-05, | |
| "loss": 0.8864, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.2902528051912938e-05, | |
| "loss": 0.8804, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.18886034879005e-05, | |
| "loss": 1.6836, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 2.0874678923888063e-05, | |
| "loss": 0.8609, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.9860754359875628e-05, | |
| "loss": 1.6934, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.884682979586319e-05, | |
| "loss": 0.8344, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.783290523185075e-05, | |
| "loss": 0.8253, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.681898066783831e-05, | |
| "loss": 0.8228, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.5805056103825875e-05, | |
| "loss": 0.8247, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 1.4791131539813438e-05, | |
| "loss": 0.7209, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.3777206975801e-05, | |
| "loss": 1.3878, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.2763282411788563e-05, | |
| "loss": 1.4032, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.1749357847776126e-05, | |
| "loss": 0.5432, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.0735433283763689e-05, | |
| "loss": 0.5517, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 9.721508719751251e-06, | |
| "loss": 0.5571, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 8.707584155738812e-06, | |
| "loss": 1.3683, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 7.693659591726377e-06, | |
| "loss": 0.5425, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 6.679735027713938e-06, | |
| "loss": 0.521, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 5.6658104637015e-06, | |
| "loss": 0.5489, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 4.651885899689063e-06, | |
| "loss": 0.5275, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 3.6379613356766258e-06, | |
| "loss": 0.5213, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 2.624036771664188e-06, | |
| "loss": 0.5213, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.6101122076517507e-06, | |
| "loss": 0.5143, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 5.961876436393133e-07, | |
| "loss": 0.5155, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 14794, | |
| "total_flos": 3.62396401532928e+16, | |
| "train_loss": 0.8917488705649381, | |
| "train_runtime": 3121.8744, | |
| "train_samples_per_second": 56.863, | |
| "train_steps_per_second": 4.739 | |
| } | |
| ], | |
| "max_steps": 14794, | |
| "num_train_epochs": 2, | |
| "total_flos": 3.62396401532928e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |