{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004518114816592777,
  "eval_steps": 5,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00022590574082963883,
      "grad_norm": 6.500443458557129,
      "learning_rate": 1e-05,
      "loss": 18.6481,
      "step": 1
    },
    {
      "epoch": 0.00022590574082963883,
      "eval_loss": 18.568599700927734,
      "eval_runtime": 290.449,
      "eval_samples_per_second": 25.671,
      "eval_steps_per_second": 12.835,
      "step": 1
    },
    {
      "epoch": 0.00045181148165927766,
      "grad_norm": 6.984323024749756,
      "learning_rate": 2e-05,
      "loss": 18.093,
      "step": 2
    },
    {
      "epoch": 0.0006777172224889165,
      "grad_norm": 7.316300392150879,
      "learning_rate": 3e-05,
      "loss": 18.673,
      "step": 3
    },
    {
      "epoch": 0.0009036229633185553,
      "grad_norm": 7.175333499908447,
      "learning_rate": 4e-05,
      "loss": 18.6659,
      "step": 4
    },
    {
      "epoch": 0.0011295287041481943,
      "grad_norm": 8.042078971862793,
      "learning_rate": 5e-05,
      "loss": 18.8358,
      "step": 5
    },
    {
      "epoch": 0.0011295287041481943,
      "eval_loss": 18.0319766998291,
      "eval_runtime": 266.1876,
      "eval_samples_per_second": 28.01,
      "eval_steps_per_second": 14.005,
      "step": 5
    },
    {
      "epoch": 0.001355434444977833,
      "grad_norm": 9.501138687133789,
      "learning_rate": 6e-05,
      "loss": 17.5967,
      "step": 6
    },
    {
      "epoch": 0.001581340185807472,
      "grad_norm": 11.359487533569336,
      "learning_rate": 7e-05,
      "loss": 18.0327,
      "step": 7
    },
    {
      "epoch": 0.0018072459266371107,
      "grad_norm": 12.280532836914062,
      "learning_rate": 8e-05,
      "loss": 17.2021,
      "step": 8
    },
    {
      "epoch": 0.0020331516674667494,
      "grad_norm": 16.318531036376953,
      "learning_rate": 9e-05,
      "loss": 16.39,
      "step": 9
    },
    {
      "epoch": 0.0022590574082963885,
      "grad_norm": 20.797056198120117,
      "learning_rate": 0.0001,
      "loss": 14.8167,
      "step": 10
    },
    {
      "epoch": 0.0022590574082963885,
      "eval_loss": 13.161385536193848,
      "eval_runtime": 183.1191,
      "eval_samples_per_second": 40.717,
      "eval_steps_per_second": 20.358,
      "step": 10
    },
    {
      "epoch": 0.0024849631491260273,
      "grad_norm": 25.225400924682617,
      "learning_rate": 9.755282581475769e-05,
      "loss": 13.105,
      "step": 11
    },
    {
      "epoch": 0.002710868889955666,
      "grad_norm": 26.978565216064453,
      "learning_rate": 9.045084971874738e-05,
      "loss": 10.561,
      "step": 12
    },
    {
      "epoch": 0.0029367746307853047,
      "grad_norm": 22.66143035888672,
      "learning_rate": 7.938926261462366e-05,
      "loss": 8.4321,
      "step": 13
    },
    {
      "epoch": 0.003162680371614944,
      "grad_norm": 20.655855178833008,
      "learning_rate": 6.545084971874738e-05,
      "loss": 6.691,
      "step": 14
    },
    {
      "epoch": 0.0033885861124445826,
      "grad_norm": 21.936189651489258,
      "learning_rate": 5e-05,
      "loss": 5.965,
      "step": 15
    },
    {
      "epoch": 0.0033885861124445826,
      "eval_loss": 4.28361701965332,
      "eval_runtime": 196.0362,
      "eval_samples_per_second": 38.034,
      "eval_steps_per_second": 19.017,
      "step": 15
    },
    {
      "epoch": 0.0036144918532742213,
      "grad_norm": 20.060565948486328,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 4.59,
      "step": 16
    },
    {
      "epoch": 0.00384039759410386,
      "grad_norm": 17.778438568115234,
      "learning_rate": 2.061073738537635e-05,
      "loss": 3.3184,
      "step": 17
    },
    {
      "epoch": 0.004066303334933499,
      "grad_norm": 16.09156036376953,
      "learning_rate": 9.549150281252633e-06,
      "loss": 3.2847,
      "step": 18
    },
    {
      "epoch": 0.004292209075763138,
      "grad_norm": 15.134675025939941,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 2.8811,
      "step": 19
    },
    {
      "epoch": 0.004518114816592777,
      "grad_norm": 14.81490707397461,
      "learning_rate": 0.0,
      "loss": 2.5459,
      "step": 20
    },
    {
      "epoch": 0.004518114816592777,
      "eval_loss": 2.7253942489624023,
      "eval_runtime": 193.5802,
      "eval_samples_per_second": 38.516,
      "eval_steps_per_second": 19.258,
      "step": 20
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.479651398516736e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}