{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.9823182711198428, "eval_steps": 500, "global_step": 2000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.02, "learning_rate": 8.999464304375242e-06, "loss": 0.7019, "step": 50 }, { "epoch": 0.05, "learning_rate": 8.997857345043108e-06, "loss": 0.6739, "step": 100 }, { "epoch": 0.07, "learning_rate": 8.99517950459963e-06, "loss": 0.6677, "step": 150 }, { "epoch": 0.1, "learning_rate": 8.99143142060366e-06, "loss": 0.6625, "step": 200 }, { "epoch": 0.12, "learning_rate": 8.98661398542506e-06, "loss": 0.6649, "step": 250 }, { "epoch": 0.15, "learning_rate": 8.980728346032255e-06, "loss": 0.6528, "step": 300 }, { "epoch": 0.17, "learning_rate": 8.973775903719142e-06, "loss": 0.6442, "step": 350 }, { "epoch": 0.2, "learning_rate": 8.965758313771466e-06, "loss": 0.6375, "step": 400 }, { "epoch": 0.22, "learning_rate": 8.956677485072722e-06, "loss": 0.6303, "step": 450 }, { "epoch": 0.25, "learning_rate": 8.946535579649664e-06, "loss": 0.634, "step": 500 }, { "epoch": 0.25, "eval_accuracy": 0.0, "eval_loss": 0.6457224488258362, "eval_runtime": 154.2695, "eval_samples_per_second": 87.989, "eval_steps_per_second": 14.669, "step": 500 }, { "epoch": 0.27, "learning_rate": 8.935335012157564e-06, "loss": 0.6236, "step": 550 }, { "epoch": 0.29, "learning_rate": 8.92307844930531e-06, "loss": 0.6235, "step": 600 }, { "epoch": 0.32, "learning_rate": 8.909768809220503e-06, "loss": 0.6301, "step": 650 }, { "epoch": 0.34, "learning_rate": 8.895409260754679e-06, "loss": 0.6227, "step": 700 }, { "epoch": 0.37, "learning_rate": 8.880003222728855e-06, "loss": 0.6187, "step": 750 }, { "epoch": 0.39, "learning_rate": 8.863554363119547e-06, "loss": 0.6201, "step": 800 }, { "epoch": 0.42, "learning_rate": 8.846066598185479e-06, "loss": 0.614, "step": 850 }, { "epoch": 0.44, "learning_rate": 8.827544091535166e-06, "loss": 0.607, "step": 900 }, { "epoch": 0.47, "learning_rate": 8.80799125313562e-06, "loss": 0.6176, "step": 950 }, { "epoch": 0.49, "learning_rate": 8.787412738262386e-06, "loss": 0.5999, "step": 1000 }, { "epoch": 0.49, "eval_accuracy": 0.0, "eval_loss": 0.6475000977516174, "eval_runtime": 154.4877, "eval_samples_per_second": 87.865, "eval_steps_per_second": 14.648, "step": 1000 }, { "epoch": 0.52, "learning_rate": 8.765813446391192e-06, "loss": 0.6055, "step": 1050 }, { "epoch": 0.54, "learning_rate": 8.743198520031437e-06, "loss": 0.6142, "step": 1100 }, { "epoch": 0.56, "learning_rate": 8.719573343501835e-06, "loss": 0.6006, "step": 1150 }, { "epoch": 0.59, "learning_rate": 8.694943541648477e-06, "loss": 0.5965, "step": 1200 }, { "epoch": 0.61, "learning_rate": 8.669314978505624e-06, "loss": 0.5951, "step": 1250 }, { "epoch": 0.64, "learning_rate": 8.642693755899563e-06, "loss": 0.5695, "step": 1300 }, { "epoch": 0.66, "learning_rate": 8.615086211995838e-06, "loss": 0.5758, "step": 1350 }, { "epoch": 0.69, "learning_rate": 8.586498919790219e-06, "loss": 0.5733, "step": 1400 }, { "epoch": 0.71, "learning_rate": 8.556938685543754e-06, "loss": 0.5728, "step": 1450 }, { "epoch": 0.74, "learning_rate": 8.526412547162289e-06, "loss": 0.5616, "step": 1500 }, { "epoch": 0.74, "eval_accuracy": 0.0, "eval_loss": 0.6478520035743713, "eval_runtime": 154.2746, "eval_samples_per_second": 87.986, "eval_steps_per_second": 14.669, "step": 1500 }, { "epoch": 0.76, "learning_rate": 8.494927772520829e-06, "loss": 0.5707, "step": 1550 }, { "epoch": 0.79, "learning_rate": 8.46249185773317e-06, "loss": 0.5655, "step": 1600 }, { "epoch": 0.81, "learning_rate": 8.429112525367142e-06, "loss": 0.5504, "step": 1650 }, { "epoch": 0.83, "learning_rate": 8.394797722605998e-06, "loss": 0.5526, "step": 1700 }, { "epoch": 0.86, "learning_rate": 8.35955561935627e-06, "loss": 0.5489, "step": 1750 }, { "epoch": 0.88, "learning_rate": 8.323394606302632e-06, "loss": 0.5429, "step": 1800 }, { "epoch": 0.91, "learning_rate": 8.28632329291019e-06, "loss": 0.5427, "step": 1850 }, { "epoch": 0.93, "learning_rate": 8.248350505374668e-06, "loss": 0.536, "step": 1900 }, { "epoch": 0.96, "learning_rate": 8.20948528452102e-06, "loss": 0.5352, "step": 1950 }, { "epoch": 0.98, "learning_rate": 8.169736883650921e-06, "loss": 0.5149, "step": 2000 }, { "epoch": 0.98, "eval_accuracy": 0.0, "eval_loss": 0.6605007648468018, "eval_runtime": 154.4529, "eval_samples_per_second": 87.884, "eval_steps_per_second": 14.652, "step": 2000 } ], "logging_steps": 50, "max_steps": 10180, "num_train_epochs": 5, "save_steps": 500, "total_flos": 0.0, "trial_name": null, "trial_params": null }