{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.05, "global_step": 50000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0, "learning_rate": 2.0000000000000003e-06, "loss": 5.8658, "step": 1000 }, { "epoch": 0.0, "learning_rate": 4.000000000000001e-06, "loss": 3.9693, "step": 2000 }, { "epoch": 0.0, "learning_rate": 6e-06, "loss": 3.5073, "step": 3000 }, { "epoch": 0.0, "learning_rate": 8.000000000000001e-06, "loss": 3.2872, "step": 4000 }, { "epoch": 0.01, "learning_rate": 1e-05, "loss": 2.8352, "step": 5000 }, { "epoch": 0.01, "learning_rate": 1.2e-05, "loss": 2.6156, "step": 6000 }, { "epoch": 0.01, "learning_rate": 1.4000000000000001e-05, "loss": 2.7728, "step": 7000 }, { "epoch": 0.01, "learning_rate": 1.6000000000000003e-05, "loss": 2.626, "step": 8000 }, { "epoch": 0.01, "learning_rate": 1.8e-05, "loss": 2.3897, "step": 9000 }, { "epoch": 0.01, "learning_rate": 2e-05, "loss": 2.4122, "step": 10000 }, { "epoch": 0.01, "learning_rate": 2.2000000000000003e-05, "loss": 2.2424, "step": 11000 }, { "epoch": 0.01, "learning_rate": 2.4e-05, "loss": 2.1699, "step": 12000 }, { "epoch": 0.01, "learning_rate": 2.6000000000000002e-05, "loss": 2.0725, "step": 13000 }, { "epoch": 0.01, "learning_rate": 2.8000000000000003e-05, "loss": 2.1961, "step": 14000 }, { "epoch": 0.01, "learning_rate": 3e-05, "loss": 2.1299, "step": 15000 }, { "epoch": 0.02, "learning_rate": 3.2000000000000005e-05, "loss": 1.8074, "step": 16000 }, { "epoch": 0.02, "learning_rate": 3.4000000000000007e-05, "loss": 1.8742, "step": 17000 }, { "epoch": 0.02, "learning_rate": 3.6e-05, "loss": 1.9853, "step": 18000 }, { "epoch": 0.02, "learning_rate": 3.8e-05, "loss": 1.9888, "step": 19000 }, { "epoch": 0.02, "learning_rate": 4e-05, "loss": 1.8076, "step": 20000 }, { "epoch": 0.02, "learning_rate": 4.2e-05, "loss": 1.6815, "step": 21000 }, { "epoch": 0.02, "learning_rate": 4.4000000000000006e-05, "loss": 1.8142, "step": 22000 }, { "epoch": 0.02, "learning_rate": 4.600000000000001e-05, "loss": 1.8181, "step": 23000 }, { "epoch": 0.02, "learning_rate": 4.8e-05, "loss": 1.7406, "step": 24000 }, { "epoch": 0.03, "learning_rate": 5e-05, "loss": 1.7678, "step": 25000 }, { "epoch": 0.03, "learning_rate": 5.2000000000000004e-05, "loss": 1.43, "step": 26000 }, { "epoch": 0.03, "learning_rate": 5.4000000000000005e-05, "loss": 1.2935, "step": 27000 }, { "epoch": 0.03, "learning_rate": 5.6000000000000006e-05, "loss": 1.3209, "step": 28000 }, { "epoch": 0.03, "learning_rate": 5.8e-05, "loss": 1.3775, "step": 29000 }, { "epoch": 0.03, "learning_rate": 6e-05, "loss": 1.3187, "step": 30000 }, { "epoch": 0.03, "learning_rate": 6.2e-05, "loss": 1.0829, "step": 31000 }, { "epoch": 0.03, "learning_rate": 6.400000000000001e-05, "loss": 1.0854, "step": 32000 }, { "epoch": 0.03, "learning_rate": 6.6e-05, "loss": 1.1996, "step": 33000 }, { "epoch": 0.03, "learning_rate": 6.800000000000001e-05, "loss": 1.2452, "step": 34000 }, { "epoch": 0.04, "learning_rate": 7e-05, "loss": 1.1567, "step": 35000 }, { "epoch": 0.04, "learning_rate": 7.2e-05, "loss": 1.0192, "step": 36000 }, { "epoch": 0.04, "learning_rate": 7.4e-05, "loss": 1.0225, "step": 37000 }, { "epoch": 0.04, "learning_rate": 7.6e-05, "loss": 1.1064, "step": 38000 }, { "epoch": 0.04, "learning_rate": 7.800000000000001e-05, "loss": 1.1388, "step": 39000 }, { "epoch": 0.04, "learning_rate": 8e-05, "loss": 1.1252, "step": 40000 }, { "epoch": 0.04, "learning_rate": 8.2e-05, "loss": 0.9804, "step": 41000 }, { "epoch": 0.04, "learning_rate": 8.4e-05, "loss": 0.9372, "step": 42000 }, { "epoch": 0.04, "learning_rate": 8.6e-05, "loss": 1.0405, "step": 43000 }, { "epoch": 0.04, "learning_rate": 8.800000000000001e-05, "loss": 1.1122, "step": 44000 }, { "epoch": 0.04, "learning_rate": 9e-05, "loss": 1.0926, "step": 45000 }, { "epoch": 0.05, "learning_rate": 9.200000000000001e-05, "loss": 0.9018, "step": 46000 }, { "epoch": 0.05, "learning_rate": 9.4e-05, "loss": 0.9155, "step": 47000 }, { "epoch": 0.05, "learning_rate": 9.6e-05, "loss": 1.0172, "step": 48000 }, { "epoch": 0.05, "learning_rate": 9.8e-05, "loss": 1.0905, "step": 49000 }, { "epoch": 0.05, "learning_rate": 0.0001, "loss": 1.0248, "step": 50000 }, { "epoch": 0.05, "eval_loss": 0.8032926321029663, "eval_runtime": 16.619, "eval_samples_per_second": 300.86, "eval_steps_per_second": 2.407, "step": 50000 } ], "max_steps": 1000000, "num_train_epochs": 9223372036854775807, "total_flos": 8.422699302912e+17, "trial_name": null, "trial_params": null }