{ "best_metric": 0.36799120903015137, "best_model_checkpoint": "/tmp/model/checkpoint-4000", "epoch": 2.0, "eval_steps": 500, "global_step": 4000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.05, "learning_rate": 5e-06, "loss": 0.581, "step": 100 }, { "epoch": 0.1, "learning_rate": 1e-05, "loss": 0.5379, "step": 200 }, { "epoch": 0.15, "learning_rate": 1.5e-05, "loss": 0.4945, "step": 300 }, { "epoch": 0.2, "learning_rate": 2e-05, "loss": 0.4066, "step": 400 }, { "epoch": 0.25, "learning_rate": 2.5e-05, "loss": 0.4182, "step": 500 }, { "epoch": 0.3, "learning_rate": 3e-05, "loss": 0.4407, "step": 600 }, { "epoch": 0.35, "learning_rate": 3.5e-05, "loss": 0.418, "step": 700 }, { "epoch": 0.4, "learning_rate": 4e-05, "loss": 0.4258, "step": 800 }, { "epoch": 0.45, "learning_rate": 4.5e-05, "loss": 0.3885, "step": 900 }, { "epoch": 0.5, "learning_rate": 5e-05, "loss": 0.4283, "step": 1000 }, { "epoch": 0.55, "learning_rate": 4.9444444444444446e-05, "loss": 0.4148, "step": 1100 }, { "epoch": 0.6, "learning_rate": 4.888888888888889e-05, "loss": 0.3951, "step": 1200 }, { "epoch": 0.65, "learning_rate": 4.8333333333333334e-05, "loss": 0.4255, "step": 1300 }, { "epoch": 0.7, "learning_rate": 4.7777777777777784e-05, "loss": 0.3842, "step": 1400 }, { "epoch": 0.75, "learning_rate": 4.722222222222222e-05, "loss": 0.4461, "step": 1500 }, { "epoch": 0.8, "learning_rate": 4.666666666666667e-05, "loss": 0.3789, "step": 1600 }, { "epoch": 0.85, "learning_rate": 4.6111111111111115e-05, "loss": 0.3893, "step": 1700 }, { "epoch": 0.9, "learning_rate": 4.555555555555556e-05, "loss": 0.3833, "step": 1800 }, { "epoch": 0.95, "learning_rate": 4.5e-05, "loss": 0.3817, "step": 1900 }, { "epoch": 1.0, "learning_rate": 4.4444444444444447e-05, "loss": 0.3654, "step": 2000 }, { "epoch": 1.0, "eval_accuracy": 0.84475, "eval_auc": 0.8835736098526796, "eval_f1": 0.689344672336168, "eval_loss": 0.40313366055488586, "eval_precision": 0.6828543111992071, "eval_recall": 0.695959595959596, "eval_runtime": 738.7835, "eval_samples_per_second": 5.414, "eval_steps_per_second": 0.338, "step": 2000 }, { "epoch": 1.05, "learning_rate": 4.388888888888889e-05, "loss": 0.3236, "step": 2100 }, { "epoch": 1.1, "learning_rate": 4.3333333333333334e-05, "loss": 0.3882, "step": 2200 }, { "epoch": 1.15, "learning_rate": 4.277777777777778e-05, "loss": 0.3413, "step": 2300 }, { "epoch": 1.2, "learning_rate": 4.222222222222222e-05, "loss": 0.303, "step": 2400 }, { "epoch": 1.25, "learning_rate": 4.166666666666667e-05, "loss": 0.352, "step": 2500 }, { "epoch": 1.3, "learning_rate": 4.111111111111111e-05, "loss": 0.323, "step": 2600 }, { "epoch": 1.35, "learning_rate": 4.055555555555556e-05, "loss": 0.3494, "step": 2700 }, { "epoch": 1.4, "learning_rate": 4e-05, "loss": 0.3125, "step": 2800 }, { "epoch": 1.45, "learning_rate": 3.944444444444445e-05, "loss": 0.314, "step": 2900 }, { "epoch": 1.5, "learning_rate": 3.888888888888889e-05, "loss": 0.394, "step": 3000 }, { "epoch": 1.55, "learning_rate": 3.8333333333333334e-05, "loss": 0.3142, "step": 3100 }, { "epoch": 1.6, "learning_rate": 3.777777777777778e-05, "loss": 0.3179, "step": 3200 }, { "epoch": 1.65, "learning_rate": 3.722222222222222e-05, "loss": 0.2756, "step": 3300 }, { "epoch": 1.7, "learning_rate": 3.6666666666666666e-05, "loss": 0.3178, "step": 3400 }, { "epoch": 1.75, "learning_rate": 3.611111111111111e-05, "loss": 0.3101, "step": 3500 }, { "epoch": 1.8, "learning_rate": 3.555555555555556e-05, "loss": 0.3615, "step": 3600 }, { "epoch": 1.85, "learning_rate": 3.5e-05, "loss": 0.3084, "step": 3700 }, { "epoch": 1.9, "learning_rate": 3.444444444444445e-05, "loss": 0.308, "step": 3800 }, { "epoch": 1.95, "learning_rate": 3.388888888888889e-05, "loss": 0.3686, "step": 3900 }, { "epoch": 2.0, "learning_rate": 3.3333333333333335e-05, "loss": 0.3642, "step": 4000 }, { "epoch": 2.0, "eval_accuracy": 0.85775, "eval_auc": 0.8958072418537535, "eval_f1": 0.7044155844155844, "eval_loss": 0.36799120903015137, "eval_precision": 0.7251336898395722, "eval_recall": 0.6848484848484848, "eval_runtime": 782.6408, "eval_samples_per_second": 5.111, "eval_steps_per_second": 0.319, "step": 4000 } ], "logging_steps": 100, "max_steps": 10000, "num_train_epochs": 5, "save_steps": 500, "total_flos": 2104888442880000.0, "trial_name": null, "trial_params": null }