{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.3937278985977173,
      "learning_rate": 0.00019,
      "loss": 1.0622,
      "step": 15
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5334441661834717,
      "learning_rate": 0.00018,
      "loss": 0.8896,
      "step": 30
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8579671382904053,
      "learning_rate": 0.00017,
      "loss": 0.7804,
      "step": 45
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.8313461542129517,
      "learning_rate": 0.00016,
      "loss": 0.6891,
      "step": 60
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.5978763103485107,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.5876,
      "step": 75
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.7626159191131592,
      "learning_rate": 0.00014,
      "loss": 0.5168,
      "step": 90
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.6013137102127075,
      "learning_rate": 0.00013000000000000002,
      "loss": 0.444,
      "step": 105
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.9760432243347168,
      "learning_rate": 0.00012,
      "loss": 0.3858,
      "step": 120
    },
    {
      "epoch": 9.0,
      "grad_norm": 1.9180957078933716,
      "learning_rate": 0.00011000000000000002,
      "loss": 0.3373,
      "step": 135
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.409602403640747,
      "learning_rate": 0.0001,
      "loss": 0.2891,
      "step": 150
    },
    {
      "epoch": 11.0,
      "grad_norm": 1.8191059827804565,
      "learning_rate": 9e-05,
      "loss": 0.2519,
      "step": 165
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.5950895547866821,
      "learning_rate": 8e-05,
      "loss": 0.2237,
      "step": 180
    },
    {
      "epoch": 13.0,
      "grad_norm": 1.4105021953582764,
      "learning_rate": 7e-05,
      "loss": 0.1976,
      "step": 195
    },
    {
      "epoch": 14.0,
      "grad_norm": 2.034846305847168,
      "learning_rate": 6e-05,
      "loss": 0.1792,
      "step": 210
    },
    {
      "epoch": 15.0,
      "grad_norm": 1.1015961170196533,
      "learning_rate": 5e-05,
      "loss": 0.1618,
      "step": 225
    },
    {
      "epoch": 16.0,
      "grad_norm": 1.245694637298584,
      "learning_rate": 4e-05,
      "loss": 0.1421,
      "step": 240
    },
    {
      "epoch": 17.0,
      "grad_norm": 1.5578792095184326,
      "learning_rate": 3e-05,
      "loss": 0.1361,
      "step": 255
    },
    {
      "epoch": 18.0,
      "grad_norm": 1.452083706855774,
      "learning_rate": 2e-05,
      "loss": 0.1245,
      "step": 270
    },
    {
      "epoch": 19.0,
      "grad_norm": 1.1172585487365723,
      "learning_rate": 1e-05,
      "loss": 0.1173,
      "step": 285
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.3190027475357056,
      "learning_rate": 0.0,
      "loss": 0.1142,
      "step": 300
    }
  ],
  "logging_steps": 500,
  "max_steps": 300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "total_flos": 4538269544755200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}