{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 25000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 1.2684,
      "step": 500
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 1.2858,
      "step": 1000
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.88e-05,
      "loss": 1.2162,
      "step": 1500
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 1.2343,
      "step": 2000
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.8e-05,
      "loss": 1.2604,
      "step": 2500
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.76e-05,
      "loss": 1.2677,
      "step": 3000
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.72e-05,
      "loss": 1.2578,
      "step": 3500
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 1.2211,
      "step": 4000
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.64e-05,
      "loss": 1.259,
      "step": 4500
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.224,
      "step": 5000
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.5600000000000003e-05,
      "loss": 1.2511,
      "step": 5500
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 1.2556,
      "step": 6000
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.48e-05,
      "loss": 1.2077,
      "step": 6500
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 1.2261,
      "step": 7000
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.4e-05,
      "loss": 1.2021,
      "step": 7500
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 1.2381,
      "step": 8000
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.3200000000000002e-05,
      "loss": 1.2355,
      "step": 8500
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 1.2332,
      "step": 9000
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.2400000000000002e-05,
      "loss": 1.199,
      "step": 9500
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.2e-05,
      "loss": 1.2086,
      "step": 10000
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.16e-05,
      "loss": 1.2177,
      "step": 10500
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 1.2126,
      "step": 11000
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.0800000000000002e-05,
      "loss": 1.1851,
      "step": 11500
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.04e-05,
      "loss": 1.219,
      "step": 12000
    },
    {
      "epoch": 1.0,
      "learning_rate": 1e-05,
      "loss": 1.2116,
      "step": 12500
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.323660969734192,
      "eval_runtime": 19.2676,
      "eval_samples_per_second": 103.801,
      "eval_steps_per_second": 12.975,
      "step": 12500
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.600000000000001e-06,
      "loss": 1.1815,
      "step": 13000
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.200000000000002e-06,
      "loss": 1.167,
      "step": 13500
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.8e-06,
      "loss": 1.1743,
      "step": 14000
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.400000000000001e-06,
      "loss": 1.1567,
      "step": 14500
    },
    {
      "epoch": 1.2,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.1831,
      "step": 15000
    },
    {
      "epoch": 1.24,
      "learning_rate": 7.600000000000001e-06,
      "loss": 1.1804,
      "step": 15500
    },
    {
      "epoch": 1.28,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 1.1467,
      "step": 16000
    },
    {
      "epoch": 1.32,
      "learning_rate": 6.800000000000001e-06,
      "loss": 1.1758,
      "step": 16500
    },
    {
      "epoch": 1.36,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 1.182,
      "step": 17000
    },
    {
      "epoch": 1.4,
      "learning_rate": 6e-06,
      "loss": 1.1705,
      "step": 17500
    },
    {
      "epoch": 1.44,
      "learning_rate": 5.600000000000001e-06,
      "loss": 1.1814,
      "step": 18000
    },
    {
      "epoch": 1.48,
      "learning_rate": 5.2e-06,
      "loss": 1.1931,
      "step": 18500
    },
    {
      "epoch": 1.52,
      "learning_rate": 4.800000000000001e-06,
      "loss": 1.1727,
      "step": 19000
    },
    {
      "epoch": 1.56,
      "learning_rate": 4.4e-06,
      "loss": 1.176,
      "step": 19500
    },
    {
      "epoch": 1.6,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.1597,
      "step": 20000
    },
    {
      "epoch": 1.64,
      "learning_rate": 3.6000000000000003e-06,
      "loss": 1.1616,
      "step": 20500
    },
    {
      "epoch": 1.68,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 1.1087,
      "step": 21000
    },
    {
      "epoch": 1.72,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 1.1478,
      "step": 21500
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 1.1471,
      "step": 22000
    },
    {
      "epoch": 1.8,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.1366,
      "step": 22500
    },
    {
      "epoch": 1.84,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 1.1719,
      "step": 23000
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 1.183,
      "step": 23500
    },
    {
      "epoch": 1.92,
      "learning_rate": 8.000000000000001e-07,
      "loss": 1.1675,
      "step": 24000
    },
    {
      "epoch": 1.96,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 1.1829,
      "step": 24500
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0,
      "loss": 1.1523,
      "step": 25000
    }
  ],
  "logging_steps": 500,
  "max_steps": 25000,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 1.686792444807168e+16,
  "trial_name": null,
  "trial_params": null
}