{
  "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": 15.92177677154541,
      "learning_rate": 0.0,
      "loss": 3.9818,
      "step": 1
    },
    {
      "epoch": 0.32,
      "grad_norm": 14.490289688110352,
      "learning_rate": 2e-05,
      "loss": 3.7845,
      "step": 2
    },
    {
      "epoch": 0.48,
      "grad_norm": 13.69474983215332,
      "learning_rate": 4e-05,
      "loss": 3.7913,
      "step": 3
    },
    {
      "epoch": 0.64,
      "grad_norm": 8.53797721862793,
      "learning_rate": 6e-05,
      "loss": 3.461,
      "step": 4
    },
    {
      "epoch": 0.8,
      "grad_norm": 7.636988162994385,
      "learning_rate": 8e-05,
      "loss": 3.2987,
      "step": 5
    },
    {
      "epoch": 0.96,
      "grad_norm": 3.4956459999084473,
      "learning_rate": 0.0001,
      "loss": 3.0411,
      "step": 6
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.3727364540100098,
      "learning_rate": 9.375e-05,
      "loss": 3.0688,
      "step": 7
    },
    {
      "epoch": 1.16,
      "grad_norm": 1.7858895063400269,
      "learning_rate": 8.75e-05,
      "loss": 2.7543,
      "step": 8
    },
    {
      "epoch": 1.32,
      "grad_norm": 1.6203562021255493,
      "learning_rate": 8.125000000000001e-05,
      "loss": 2.7939,
      "step": 9
    },
    {
      "epoch": 1.48,
      "grad_norm": 1.4051226377487183,
      "learning_rate": 7.500000000000001e-05,
      "loss": 2.6177,
      "step": 10
    },
    {
      "epoch": 1.6400000000000001,
      "grad_norm": 1.298180341720581,
      "learning_rate": 6.875e-05,
      "loss": 2.5475,
      "step": 11
    },
    {
      "epoch": 1.8,
      "grad_norm": 1.2686516046524048,
      "learning_rate": 6.25e-05,
      "loss": 2.543,
      "step": 12
    },
    {
      "epoch": 1.96,
      "grad_norm": 1.2775018215179443,
      "learning_rate": 5.6250000000000005e-05,
      "loss": 2.5224,
      "step": 13
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.78619384765625,
      "learning_rate": 5e-05,
      "loss": 2.4166,
      "step": 14
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.0871943235397339,
      "learning_rate": 4.375e-05,
      "loss": 2.4308,
      "step": 15
    },
    {
      "epoch": 2.32,
      "grad_norm": 1.1050121784210205,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 2.3177,
      "step": 16
    },
    {
      "epoch": 2.48,
      "grad_norm": 1.0715268850326538,
      "learning_rate": 3.125e-05,
      "loss": 2.3069,
      "step": 17
    },
    {
      "epoch": 2.64,
      "grad_norm": 1.0779463052749634,
      "learning_rate": 2.5e-05,
      "loss": 2.2828,
      "step": 18
    },
    {
      "epoch": 2.8,
      "grad_norm": 1.0815116167068481,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 2.3069,
      "step": 19
    },
    {
      "epoch": 2.96,
      "grad_norm": 1.1218745708465576,
      "learning_rate": 1.25e-05,
      "loss": 2.2895,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.921610951423645,
      "learning_rate": 6.25e-06,
      "loss": 2.3531,
      "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": 814709863219200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}