{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.789678675754625,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.31,
      "grad_norm": 0.48931118845939636,
      "learning_rate": 1.98125e-05,
      "loss": 1.7417,
      "step": 20
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.56327223777771,
      "learning_rate": 1.9604166666666668e-05,
      "loss": 1.7626,
      "step": 40
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.6054126620292664,
      "learning_rate": 1.9395833333333335e-05,
      "loss": 1.5774,
      "step": 60
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.8035449385643005,
      "learning_rate": 1.9187500000000002e-05,
      "loss": 1.4772,
      "step": 80
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.9972027540206909,
      "learning_rate": 1.897916666666667e-05,
      "loss": 1.409,
      "step": 100
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.7526600360870361,
      "learning_rate": 1.8781250000000003e-05,
      "loss": 1.3088,
      "step": 120
    },
    {
      "epoch": 2.18,
      "grad_norm": 0.8370587825775146,
      "learning_rate": 1.8572916666666666e-05,
      "loss": 1.2086,
      "step": 140
    },
    {
      "epoch": 2.49,
      "grad_norm": 0.8337924480438232,
      "learning_rate": 1.8364583333333334e-05,
      "loss": 1.1321,
      "step": 160
    },
    {
      "epoch": 2.8,
      "grad_norm": 1.0028570890426636,
      "learning_rate": 1.815625e-05,
      "loss": 1.0439,
      "step": 180
    },
    {
      "epoch": 3.12,
      "grad_norm": 0.7663702368736267,
      "learning_rate": 1.7947916666666668e-05,
      "loss": 1.026,
      "step": 200
    },
    {
      "epoch": 3.43,
      "grad_norm": 0.8835870623588562,
      "learning_rate": 1.7739583333333335e-05,
      "loss": 0.9994,
      "step": 220
    },
    {
      "epoch": 3.74,
      "grad_norm": 1.010624885559082,
      "learning_rate": 1.7531250000000003e-05,
      "loss": 0.9381,
      "step": 240
    },
    {
      "epoch": 4.05,
      "grad_norm": 0.9645354151725769,
      "learning_rate": 1.7322916666666666e-05,
      "loss": 0.8667,
      "step": 260
    },
    {
      "epoch": 4.36,
      "grad_norm": 0.9926008582115173,
      "learning_rate": 1.7114583333333334e-05,
      "loss": 0.8729,
      "step": 280
    },
    {
      "epoch": 4.67,
      "grad_norm": 1.0364482402801514,
      "learning_rate": 1.690625e-05,
      "loss": 0.836,
      "step": 300
    },
    {
      "epoch": 4.99,
      "grad_norm": 1.3519001007080078,
      "learning_rate": 1.6697916666666668e-05,
      "loss": 0.8126,
      "step": 320
    },
    {
      "epoch": 5.3,
      "grad_norm": 1.2280508279800415,
      "learning_rate": 1.6489583333333335e-05,
      "loss": 0.7953,
      "step": 340
    },
    {
      "epoch": 5.61,
      "grad_norm": 0.8931779265403748,
      "learning_rate": 1.6281250000000003e-05,
      "loss": 0.7813,
      "step": 360
    },
    {
      "epoch": 5.92,
      "grad_norm": 2.7053027153015137,
      "learning_rate": 1.6072916666666667e-05,
      "loss": 0.7256,
      "step": 380
    },
    {
      "epoch": 6.23,
      "grad_norm": 1.411024808883667,
      "learning_rate": 1.5864583333333334e-05,
      "loss": 0.7336,
      "step": 400
    },
    {
      "epoch": 6.54,
      "grad_norm": 1.0111807584762573,
      "learning_rate": 1.565625e-05,
      "loss": 0.7207,
      "step": 420
    },
    {
      "epoch": 6.85,
      "grad_norm": 2.213623523712158,
      "learning_rate": 1.544791666666667e-05,
      "loss": 0.664,
      "step": 440
    },
    {
      "epoch": 7.17,
      "grad_norm": 1.3642323017120361,
      "learning_rate": 1.5239583333333334e-05,
      "loss": 0.6848,
      "step": 460
    },
    {
      "epoch": 7.48,
      "grad_norm": 1.3692028522491455,
      "learning_rate": 1.5031250000000001e-05,
      "loss": 0.6663,
      "step": 480
    },
    {
      "epoch": 7.79,
      "grad_norm": 1.9850131273269653,
      "learning_rate": 1.4822916666666667e-05,
      "loss": 0.6199,
      "step": 500
    }
  ],
  "logging_steps": 20,
  "max_steps": 1920,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 1.2995638935552e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}