{ "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": 5.813, "step": 1000 }, { "epoch": 0.0, "learning_rate": 4.000000000000001e-06, "loss": 3.8522, "step": 2000 }, { "epoch": 0.0, "learning_rate": 6e-06, "loss": 3.1123, "step": 3000 }, { "epoch": 0.0, "learning_rate": 8.000000000000001e-06, "loss": 2.8214, "step": 4000 }, { "epoch": 0.01, "learning_rate": 1e-05, "loss": 2.6386, "step": 5000 }, { "epoch": 0.01, "learning_rate": 1.2e-05, "loss": 2.4912, "step": 6000 }, { "epoch": 0.01, "learning_rate": 1.4000000000000001e-05, "loss": 2.3574, "step": 7000 }, { "epoch": 0.01, "learning_rate": 1.6000000000000003e-05, "loss": 2.2448, "step": 8000 }, { "epoch": 0.01, "learning_rate": 1.8e-05, "loss": 2.14, "step": 9000 }, { "epoch": 0.01, "learning_rate": 2e-05, "loss": 2.0264, "step": 10000 }, { "epoch": 0.01, "learning_rate": 2.2000000000000003e-05, "loss": 1.9757, "step": 11000 }, { "epoch": 0.01, "learning_rate": 2.4e-05, "loss": 1.9128, "step": 12000 }, { "epoch": 0.01, "learning_rate": 2.6000000000000002e-05, "loss": 1.8556, "step": 13000 }, { "epoch": 0.01, "learning_rate": 2.8000000000000003e-05, "loss": 1.8063, "step": 14000 }, { "epoch": 0.01, "learning_rate": 3e-05, "loss": 1.7684, "step": 15000 }, { "epoch": 0.02, "learning_rate": 3.2000000000000005e-05, "loss": 1.7355, "step": 16000 }, { "epoch": 0.02, "learning_rate": 3.4000000000000007e-05, "loss": 1.6983, "step": 17000 }, { "epoch": 0.02, "learning_rate": 3.6e-05, "loss": 1.669, "step": 18000 }, { "epoch": 0.02, "learning_rate": 3.8e-05, "loss": 1.6363, "step": 19000 }, { "epoch": 0.02, "learning_rate": 4e-05, "loss": 1.5894, "step": 20000 }, { "epoch": 0.02, "learning_rate": 4.2e-05, "loss": 1.5883, "step": 21000 }, { "epoch": 0.02, "learning_rate": 4.4000000000000006e-05, "loss": 1.5656, "step": 22000 }, { "epoch": 0.02, "learning_rate": 4.600000000000001e-05, "loss": 1.5479, "step": 23000 }, { "epoch": 0.02, "learning_rate": 4.8e-05, "loss": 1.5247, "step": 24000 }, { "epoch": 0.03, "learning_rate": 5e-05, "loss": 1.5167, "step": 25000 }, { "epoch": 0.03, "learning_rate": 5.2000000000000004e-05, "loss": 1.3467, "step": 26000 }, { "epoch": 0.03, "learning_rate": 5.4000000000000005e-05, "loss": 1.2348, "step": 27000 }, { "epoch": 0.03, "learning_rate": 5.6000000000000006e-05, "loss": 1.1802, "step": 28000 }, { "epoch": 0.03, "learning_rate": 5.8e-05, "loss": 1.1369, "step": 29000 }, { "epoch": 0.03, "learning_rate": 6e-05, "loss": 1.1024, "step": 30000 }, { "epoch": 0.03, "learning_rate": 6.2e-05, "loss": 1.0741, "step": 31000 }, { "epoch": 0.03, "learning_rate": 6.400000000000001e-05, "loss": 1.048, "step": 32000 }, { "epoch": 0.03, "learning_rate": 6.6e-05, "loss": 1.0313, "step": 33000 }, { "epoch": 0.03, "learning_rate": 6.800000000000001e-05, "loss": 1.0111, "step": 34000 }, { "epoch": 0.04, "learning_rate": 7e-05, "loss": 1.0046, "step": 35000 }, { "epoch": 0.04, "learning_rate": 7.2e-05, "loss": 0.9983, "step": 36000 }, { "epoch": 0.04, "learning_rate": 7.4e-05, "loss": 0.9828, "step": 37000 }, { "epoch": 0.04, "learning_rate": 7.6e-05, "loss": 0.9726, "step": 38000 }, { "epoch": 0.04, "learning_rate": 7.800000000000001e-05, "loss": 0.9636, "step": 39000 }, { "epoch": 0.04, "learning_rate": 8e-05, "loss": 0.9526, "step": 40000 }, { "epoch": 0.04, "learning_rate": 8.2e-05, "loss": 0.9359, "step": 41000 }, { "epoch": 0.04, "learning_rate": 8.4e-05, "loss": 0.9267, "step": 42000 }, { "epoch": 0.04, "learning_rate": 8.6e-05, "loss": 0.9192, "step": 43000 }, { "epoch": 0.04, "learning_rate": 8.800000000000001e-05, "loss": 0.9105, "step": 44000 }, { "epoch": 0.04, "learning_rate": 9e-05, "loss": 0.908, "step": 45000 }, { "epoch": 0.05, "learning_rate": 9.200000000000001e-05, "loss": 0.9087, "step": 46000 }, { "epoch": 0.05, "learning_rate": 9.4e-05, "loss": 0.9023, "step": 47000 }, { "epoch": 0.05, "learning_rate": 9.6e-05, "loss": 0.8975, "step": 48000 }, { "epoch": 0.05, "learning_rate": 9.8e-05, "loss": 0.8956, "step": 49000 }, { "epoch": 0.05, "learning_rate": 0.0001, "loss": 0.8857, "step": 50000 }, { "epoch": 0.05, "eval_loss": 0.6918957233428955, "eval_runtime": 16.7, "eval_samples_per_second": 299.401, "eval_steps_per_second": 2.395, "step": 50000 } ], "max_steps": 1000000, "num_train_epochs": 9223372036854775807, "total_flos": 8.422699302912e+17, "trial_name": null, "trial_params": null }