{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.994535519125683,
  "eval_steps": 500,
  "global_step": 91,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01092896174863388,
      "grad_norm": 67.9917221069336,
      "learning_rate": 2e-05,
      "loss": 2.2916,
      "step": 1
    },
    {
      "epoch": 0.0546448087431694,
      "grad_norm": 2.097611665725708,
      "learning_rate": 0.0001,
      "loss": 2.2127,
      "step": 5
    },
    {
      "epoch": 0.1092896174863388,
      "grad_norm": 1.9431463479995728,
      "learning_rate": 0.0002,
      "loss": 2.0887,
      "step": 10
    },
    {
      "epoch": 0.16393442622950818,
      "grad_norm": 1.8884824514389038,
      "learning_rate": 0.00019812553106273847,
      "loss": 1.9685,
      "step": 15
    },
    {
      "epoch": 0.2185792349726776,
      "grad_norm": 2.330018997192383,
      "learning_rate": 0.00019257239692688907,
      "loss": 1.8856,
      "step": 20
    },
    {
      "epoch": 0.273224043715847,
      "grad_norm": 0.6340972781181335,
      "learning_rate": 0.00018354878114129367,
      "loss": 1.8482,
      "step": 25
    },
    {
      "epoch": 0.32786885245901637,
      "grad_norm": 0.45715248584747314,
      "learning_rate": 0.00017139297345578994,
      "loss": 1.7705,
      "step": 30
    },
    {
      "epoch": 0.3825136612021858,
      "grad_norm": 1.0445672273635864,
      "learning_rate": 0.00015656068754865387,
      "loss": 1.7627,
      "step": 35
    },
    {
      "epoch": 0.4371584699453552,
      "grad_norm": 0.503304123878479,
      "learning_rate": 0.0001396079766039157,
      "loss": 1.7488,
      "step": 40
    },
    {
      "epoch": 0.4918032786885246,
      "grad_norm": 0.5089389681816101,
      "learning_rate": 0.0001211703872229411,
      "loss": 1.7165,
      "step": 45
    },
    {
      "epoch": 0.546448087431694,
      "grad_norm": 0.416858971118927,
      "learning_rate": 0.00010193913317718244,
      "loss": 1.7396,
      "step": 50
    },
    {
      "epoch": 0.6010928961748634,
      "grad_norm": 0.29764389991760254,
      "learning_rate": 8.263518223330697e-05,
      "loss": 1.7045,
      "step": 55
    },
    {
      "epoch": 0.6557377049180327,
      "grad_norm": 0.3781444728374481,
      "learning_rate": 6.398222751952899e-05,
      "loss": 1.6977,
      "step": 60
    },
    {
      "epoch": 0.7103825136612022,
      "grad_norm": 0.3005172610282898,
      "learning_rate": 4.66795567198309e-05,
      "loss": 1.715,
      "step": 65
    },
    {
      "epoch": 0.7650273224043715,
      "grad_norm": 0.39595624804496765,
      "learning_rate": 3.137583621312665e-05,
      "loss": 1.7143,
      "step": 70
    },
    {
      "epoch": 0.819672131147541,
      "grad_norm": 0.4393070638179779,
      "learning_rate": 1.864479297370325e-05,
      "loss": 1.6919,
      "step": 75
    },
    {
      "epoch": 0.8743169398907104,
      "grad_norm": 0.38544324040412903,
      "learning_rate": 8.963705903385345e-06,
      "loss": 1.7241,
      "step": 80
    },
    {
      "epoch": 0.9289617486338798,
      "grad_norm": 0.5594367980957031,
      "learning_rate": 2.6955129420176196e-06,
      "loss": 1.6885,
      "step": 85
    },
    {
      "epoch": 0.9836065573770492,
      "grad_norm": 0.2916119396686554,
      "learning_rate": 7.520474957699586e-08,
      "loss": 1.7241,
      "step": 90
    },
    {
      "epoch": 0.994535519125683,
      "eval_loss": 1.7210174798965454,
      "eval_runtime": 1.2379,
      "eval_samples_per_second": 140.557,
      "eval_steps_per_second": 1.616,
      "step": 91
    },
    {
      "epoch": 0.994535519125683,
      "step": 91,
      "total_flos": 4.03002565094015e+17,
      "train_loss": 1.799644404715234,
      "train_runtime": 332.7348,
      "train_samples_per_second": 52.685,
      "train_steps_per_second": 0.273
    }
  ],
  "logging_steps": 5,
  "max_steps": 91,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.03002565094015e+17,
  "train_batch_size": 12,
  "trial_name": null,
  "trial_params": null
}