{
  "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": 1.8812966346740723,
      "learning_rate": 0.0,
      "loss": 3.0332,
      "step": 1
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.918473720550537,
      "learning_rate": 2e-05,
      "loss": 2.9492,
      "step": 2
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.9667052030563354,
      "learning_rate": 4e-05,
      "loss": 2.9997,
      "step": 3
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.808651328086853,
      "learning_rate": 6e-05,
      "loss": 2.9066,
      "step": 4
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.882343053817749,
      "learning_rate": 8e-05,
      "loss": 2.8918,
      "step": 5
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.577702283859253,
      "learning_rate": 0.0001,
      "loss": 2.8003,
      "step": 6
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.7149527072906494,
      "learning_rate": 9.375e-05,
      "loss": 2.8392,
      "step": 7
    },
    {
      "epoch": 1.16,
      "grad_norm": 1.394409418106079,
      "learning_rate": 8.75e-05,
      "loss": 2.5173,
      "step": 8
    },
    {
      "epoch": 1.32,
      "grad_norm": 1.2313398122787476,
      "learning_rate": 8.125000000000001e-05,
      "loss": 2.5424,
      "step": 9
    },
    {
      "epoch": 1.48,
      "grad_norm": 1.3582338094711304,
      "learning_rate": 7.500000000000001e-05,
      "loss": 2.3947,
      "step": 10
    },
    {
      "epoch": 1.6400000000000001,
      "grad_norm": 1.2914677858352661,
      "learning_rate": 6.875e-05,
      "loss": 2.3496,
      "step": 11
    },
    {
      "epoch": 1.8,
      "grad_norm": 1.6446493864059448,
      "learning_rate": 6.25e-05,
      "loss": 2.3543,
      "step": 12
    },
    {
      "epoch": 1.96,
      "grad_norm": 1.2854161262512207,
      "learning_rate": 5.6250000000000005e-05,
      "loss": 2.3406,
      "step": 13
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.074948310852051,
      "learning_rate": 5e-05,
      "loss": 2.2983,
      "step": 14
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.2562259435653687,
      "learning_rate": 4.375e-05,
      "loss": 2.2847,
      "step": 15
    },
    {
      "epoch": 2.32,
      "grad_norm": 1.2016979455947876,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 2.2045,
      "step": 16
    },
    {
      "epoch": 2.48,
      "grad_norm": 1.2354278564453125,
      "learning_rate": 3.125e-05,
      "loss": 2.1844,
      "step": 17
    },
    {
      "epoch": 2.64,
      "grad_norm": 1.1094051599502563,
      "learning_rate": 2.5e-05,
      "loss": 2.1625,
      "step": 18
    },
    {
      "epoch": 2.8,
      "grad_norm": 1.1314666271209717,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 2.1791,
      "step": 19
    },
    {
      "epoch": 2.96,
      "grad_norm": 1.5989243984222412,
      "learning_rate": 1.25e-05,
      "loss": 2.2109,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.1544666290283203,
      "learning_rate": 6.25e-06,
      "loss": 2.2118,
      "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": 861504379699200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}