{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 99.87956643918105,
  "eval_steps": 3110,
  "global_step": 31100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 9.987956643918105,
      "grad_norm": 0.008763309568166733,
      "learning_rate": 9.002572347266882e-06,
      "loss": 2.156,
      "step": 3110
    },
    {
      "epoch": 9.987956643918105,
      "eval_accuracy": 0.04915439834529748,
      "eval_loss": 2.9957287311553955,
      "eval_runtime": 36.4037,
      "eval_samples_per_second": 225.774,
      "eval_steps_per_second": 11.29,
      "step": 3110
    },
    {
      "epoch": 19.97591328783621,
      "grad_norm": 1.9511775970458984,
      "learning_rate": 8.002572347266881e-06,
      "loss": 2.1411,
      "step": 6220
    },
    {
      "epoch": 19.97591328783621,
      "eval_accuracy": 0.05371699720160603,
      "eval_loss": 3.0536019802093506,
      "eval_runtime": 36.6549,
      "eval_samples_per_second": 224.227,
      "eval_steps_per_second": 11.213,
      "step": 6220
    },
    {
      "epoch": 29.963869931754317,
      "grad_norm": 2.7666614055633545,
      "learning_rate": 7.002572347266882e-06,
      "loss": 1.9888,
      "step": 9330
    },
    {
      "epoch": 29.963869931754317,
      "eval_accuracy": 0.05665733868678266,
      "eval_loss": 3.3323957920074463,
      "eval_runtime": 36.5927,
      "eval_samples_per_second": 224.607,
      "eval_steps_per_second": 11.232,
      "step": 9330
    },
    {
      "epoch": 39.95182657567242,
      "grad_norm": 3.709224224090576,
      "learning_rate": 6.002572347266882e-06,
      "loss": 1.8759,
      "step": 12440
    },
    {
      "epoch": 39.95182657567242,
      "eval_accuracy": 0.05706290302956564,
      "eval_loss": 3.6092300415039062,
      "eval_runtime": 36.6501,
      "eval_samples_per_second": 224.256,
      "eval_steps_per_second": 11.214,
      "step": 12440
    },
    {
      "epoch": 49.939783219590524,
      "grad_norm": 3.455683708190918,
      "learning_rate": 5.002893890675241e-06,
      "loss": 1.8129,
      "step": 15550
    },
    {
      "epoch": 49.939783219590524,
      "eval_accuracy": 0.05752524638033824,
      "eval_loss": 3.809110641479492,
      "eval_runtime": 36.6247,
      "eval_samples_per_second": 224.411,
      "eval_steps_per_second": 11.222,
      "step": 15550
    },
    {
      "epoch": 59.927739863508634,
      "grad_norm": 3.957350254058838,
      "learning_rate": 4.003536977491962e-06,
      "loss": 1.7708,
      "step": 18660
    },
    {
      "epoch": 59.927739863508634,
      "eval_accuracy": 0.05777264062943586,
      "eval_loss": 3.9897844791412354,
      "eval_runtime": 36.4616,
      "eval_samples_per_second": 225.415,
      "eval_steps_per_second": 11.272,
      "step": 18660
    },
    {
      "epoch": 69.91569650742673,
      "grad_norm": 4.713634014129639,
      "learning_rate": 3.003858520900322e-06,
      "loss": 1.7413,
      "step": 21770
    },
    {
      "epoch": 69.91569650742673,
      "eval_accuracy": 0.057914588149409904,
      "eval_loss": 4.2735114097595215,
      "eval_runtime": 36.4969,
      "eval_samples_per_second": 225.197,
      "eval_steps_per_second": 11.261,
      "step": 21770
    },
    {
      "epoch": 79.90365315134484,
      "grad_norm": 3.9962944984436035,
      "learning_rate": 2.0045016077170422e-06,
      "loss": 1.7172,
      "step": 24880
    },
    {
      "epoch": 79.90365315134484,
      "eval_accuracy": 0.0580362574522448,
      "eval_loss": 4.343382835388184,
      "eval_runtime": 36.6256,
      "eval_samples_per_second": 224.406,
      "eval_steps_per_second": 11.222,
      "step": 24880
    },
    {
      "epoch": 89.89160979526295,
      "grad_norm": 2.9578592777252197,
      "learning_rate": 1.0048231511254019e-06,
      "loss": 1.7056,
      "step": 27990
    },
    {
      "epoch": 89.89160979526295,
      "eval_accuracy": 0.058063295075097,
      "eval_loss": 4.511988162994385,
      "eval_runtime": 36.4648,
      "eval_samples_per_second": 225.395,
      "eval_steps_per_second": 11.271,
      "step": 27990
    },
    {
      "epoch": 99.87956643918105,
      "grad_norm": 3.108112335205078,
      "learning_rate": 5.466237942122187e-09,
      "loss": 1.7018,
      "step": 31100
    },
    {
      "epoch": 99.87956643918105,
      "eval_accuracy": 0.05819442754593016,
      "eval_loss": 4.51352071762085,
      "eval_runtime": 36.5052,
      "eval_samples_per_second": 225.146,
      "eval_steps_per_second": 11.259,
      "step": 31100
    },
    {
      "epoch": 99.87956643918105,
      "step": 31100,
      "total_flos": 1.4711495868588088e+18,
      "train_loss": 1.861133707420619,
      "train_runtime": 58944.542,
      "train_samples_per_second": 105.616,
      "train_steps_per_second": 0.528
    }
  ],
  "logging_steps": 3110,
  "max_steps": 31100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.4711495868588088e+18,
  "train_batch_size": 25,
  "trial_name": null,
  "trial_params": null
}