{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "grad_norm": 8.644428253173828,
      "learning_rate": 0.0,
      "loss": 3.5197,
      "step": 1
    },
    {
      "epoch": 0.32,
      "grad_norm": 8.070281982421875,
      "learning_rate": 2e-05,
      "loss": 3.3666,
      "step": 2
    },
    {
      "epoch": 0.48,
      "grad_norm": 7.960400581359863,
      "learning_rate": 4e-05,
      "loss": 3.4233,
      "step": 3
    },
    {
      "epoch": 0.64,
      "grad_norm": 5.458095550537109,
      "learning_rate": 6e-05,
      "loss": 3.0845,
      "step": 4
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.2934956550598145,
      "learning_rate": 8e-05,
      "loss": 2.8616,
      "step": 5
    },
    {
      "epoch": 0.96,
      "grad_norm": 2.040363073348999,
      "learning_rate": 0.0001,
      "loss": 2.7408,
      "step": 6
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.2898333072662354,
      "learning_rate": 9.375e-05,
      "loss": 2.6703,
      "step": 7
    },
    {
      "epoch": 1.16,
      "grad_norm": 1.2641125917434692,
      "learning_rate": 8.75e-05,
      "loss": 2.4129,
      "step": 8
    },
    {
      "epoch": 1.32,
      "grad_norm": 1.3355458974838257,
      "learning_rate": 8.125000000000001e-05,
      "loss": 2.4037,
      "step": 9
    },
    {
      "epoch": 1.48,
      "grad_norm": 1.2536065578460693,
      "learning_rate": 7.500000000000001e-05,
      "loss": 2.2565,
      "step": 10
    },
    {
      "epoch": 1.6400000000000001,
      "grad_norm": 1.197403907775879,
      "learning_rate": 6.875e-05,
      "loss": 2.196,
      "step": 11
    },
    {
      "epoch": 1.8,
      "grad_norm": 1.1043906211853027,
      "learning_rate": 6.25e-05,
      "loss": 2.1441,
      "step": 12
    },
    {
      "epoch": 1.96,
      "grad_norm": 1.0590766668319702,
      "learning_rate": 5.6250000000000005e-05,
      "loss": 2.1485,
      "step": 13
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6279308795928955,
      "learning_rate": 5e-05,
      "loss": 2.0498,
      "step": 14
    },
    {
      "epoch": 2.16,
      "grad_norm": 0.9995439648628235,
      "learning_rate": 4.375e-05,
      "loss": 2.0478,
      "step": 15
    },
    {
      "epoch": 2.32,
      "grad_norm": 1.0121654272079468,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 1.9138,
      "step": 16
    },
    {
      "epoch": 2.48,
      "grad_norm": 1.0135365724563599,
      "learning_rate": 3.125e-05,
      "loss": 1.9102,
      "step": 17
    },
    {
      "epoch": 2.64,
      "grad_norm": 0.9728643298149109,
      "learning_rate": 2.5e-05,
      "loss": 1.9128,
      "step": 18
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.9662680625915527,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 1.902,
      "step": 19
    },
    {
      "epoch": 2.96,
      "grad_norm": 1.0662381649017334,
      "learning_rate": 1.25e-05,
      "loss": 1.89,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8702627420425415,
      "learning_rate": 6.25e-06,
      "loss": 1.8779,
      "step": 21
    }
  ],
  "logging_steps": 1,
  "max_steps": 21,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2086913893269504.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}