{
  "best_metric": 0.3561370372772217,
  "best_model_checkpoint": "results/checkpoint-8500",
  "epoch": 0.8119209093514185,
  "eval_steps": 500,
  "global_step": 8500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04776005349125991,
      "grad_norm": 6.31744909286499,
      "learning_rate": 4.8805998662718505e-05,
      "loss": 0.4525,
      "step": 500
    },
    {
      "epoch": 0.04776005349125991,
      "eval_loss": 0.38868293166160583,
      "eval_runtime": 105.2805,
      "eval_samples_per_second": 248.393,
      "eval_steps_per_second": 3.885,
      "step": 500
    },
    {
      "epoch": 0.09552010698251982,
      "grad_norm": 8.537174224853516,
      "learning_rate": 4.761199732543701e-05,
      "loss": 0.4516,
      "step": 1000
    },
    {
      "epoch": 0.09552010698251982,
      "eval_loss": 0.3857233226299286,
      "eval_runtime": 110.5179,
      "eval_samples_per_second": 236.622,
      "eval_steps_per_second": 3.701,
      "step": 1000
    },
    {
      "epoch": 0.14328016047377973,
      "grad_norm": 3.55123233795166,
      "learning_rate": 4.641799598815551e-05,
      "loss": 0.4412,
      "step": 1500
    },
    {
      "epoch": 0.14328016047377973,
      "eval_loss": 0.3843745291233063,
      "eval_runtime": 108.5384,
      "eval_samples_per_second": 240.938,
      "eval_steps_per_second": 3.768,
      "step": 1500
    },
    {
      "epoch": 0.19104021396503965,
      "grad_norm": 4.456860065460205,
      "learning_rate": 4.522399465087401e-05,
      "loss": 0.4411,
      "step": 2000
    },
    {
      "epoch": 0.19104021396503965,
      "eval_loss": 0.3724469542503357,
      "eval_runtime": 104.9479,
      "eval_samples_per_second": 249.181,
      "eval_steps_per_second": 3.897,
      "step": 2000
    },
    {
      "epoch": 0.23880026745629956,
      "grad_norm": 8.174205780029297,
      "learning_rate": 4.402999331359251e-05,
      "loss": 0.4269,
      "step": 2500
    },
    {
      "epoch": 0.23880026745629956,
      "eval_loss": 0.3822018504142761,
      "eval_runtime": 109.3926,
      "eval_samples_per_second": 239.056,
      "eval_steps_per_second": 3.739,
      "step": 2500
    },
    {
      "epoch": 0.28656032094755945,
      "grad_norm": 8.617950439453125,
      "learning_rate": 4.2835991976311015e-05,
      "loss": 0.4417,
      "step": 3000
    },
    {
      "epoch": 0.28656032094755945,
      "eval_loss": 0.37791815400123596,
      "eval_runtime": 110.9874,
      "eval_samples_per_second": 235.621,
      "eval_steps_per_second": 3.685,
      "step": 3000
    },
    {
      "epoch": 0.33432037443881935,
      "grad_norm": 4.857789993286133,
      "learning_rate": 4.164199063902952e-05,
      "loss": 0.4324,
      "step": 3500
    },
    {
      "epoch": 0.33432037443881935,
      "eval_loss": 0.37730905413627625,
      "eval_runtime": 110.4008,
      "eval_samples_per_second": 236.873,
      "eval_steps_per_second": 3.705,
      "step": 3500
    },
    {
      "epoch": 0.3820804279300793,
      "grad_norm": 4.581517219543457,
      "learning_rate": 4.044798930174802e-05,
      "loss": 0.4184,
      "step": 4000
    },
    {
      "epoch": 0.3820804279300793,
      "eval_loss": 0.3715578615665436,
      "eval_runtime": 111.0466,
      "eval_samples_per_second": 235.496,
      "eval_steps_per_second": 3.683,
      "step": 4000
    },
    {
      "epoch": 0.4298404814213392,
      "grad_norm": 7.005139350891113,
      "learning_rate": 3.925398796446652e-05,
      "loss": 0.4135,
      "step": 4500
    },
    {
      "epoch": 0.4298404814213392,
      "eval_loss": 0.3852500319480896,
      "eval_runtime": 110.8005,
      "eval_samples_per_second": 236.019,
      "eval_steps_per_second": 3.691,
      "step": 4500
    },
    {
      "epoch": 0.47760053491259913,
      "grad_norm": 7.208944797515869,
      "learning_rate": 3.8059986627185024e-05,
      "loss": 0.4172,
      "step": 5000
    },
    {
      "epoch": 0.47760053491259913,
      "eval_loss": 0.3664040267467499,
      "eval_runtime": 110.6088,
      "eval_samples_per_second": 236.428,
      "eval_steps_per_second": 3.698,
      "step": 5000
    },
    {
      "epoch": 0.525360588403859,
      "grad_norm": 4.011592864990234,
      "learning_rate": 3.6865985289903526e-05,
      "loss": 0.4164,
      "step": 5500
    },
    {
      "epoch": 0.525360588403859,
      "eval_loss": 0.36664679646492004,
      "eval_runtime": 108.0591,
      "eval_samples_per_second": 242.006,
      "eval_steps_per_second": 3.785,
      "step": 5500
    },
    {
      "epoch": 0.5731206418951189,
      "grad_norm": 3.7559821605682373,
      "learning_rate": 3.567198395262203e-05,
      "loss": 0.4124,
      "step": 6000
    },
    {
      "epoch": 0.5731206418951189,
      "eval_loss": 0.36337631940841675,
      "eval_runtime": 105.397,
      "eval_samples_per_second": 248.119,
      "eval_steps_per_second": 3.881,
      "step": 6000
    },
    {
      "epoch": 0.6208806953863788,
      "grad_norm": 4.904799938201904,
      "learning_rate": 3.447798261534053e-05,
      "loss": 0.4133,
      "step": 6500
    },
    {
      "epoch": 0.6208806953863788,
      "eval_loss": 0.3614007234573364,
      "eval_runtime": 110.9753,
      "eval_samples_per_second": 235.647,
      "eval_steps_per_second": 3.686,
      "step": 6500
    },
    {
      "epoch": 0.6686407488776387,
      "grad_norm": 8.097270011901855,
      "learning_rate": 3.328398127805903e-05,
      "loss": 0.4132,
      "step": 7000
    },
    {
      "epoch": 0.6686407488776387,
      "eval_loss": 0.360387921333313,
      "eval_runtime": 110.5513,
      "eval_samples_per_second": 236.551,
      "eval_steps_per_second": 3.7,
      "step": 7000
    },
    {
      "epoch": 0.7164008023688987,
      "grad_norm": 5.6349968910217285,
      "learning_rate": 3.2089979940777535e-05,
      "loss": 0.3994,
      "step": 7500
    },
    {
      "epoch": 0.7164008023688987,
      "eval_loss": 0.3582770824432373,
      "eval_runtime": 109.3558,
      "eval_samples_per_second": 239.137,
      "eval_steps_per_second": 3.74,
      "step": 7500
    },
    {
      "epoch": 0.7641608558601586,
      "grad_norm": 5.345180988311768,
      "learning_rate": 3.089597860349604e-05,
      "loss": 0.4032,
      "step": 8000
    },
    {
      "epoch": 0.7641608558601586,
      "eval_loss": 0.3689400851726532,
      "eval_runtime": 110.4919,
      "eval_samples_per_second": 236.678,
      "eval_steps_per_second": 3.702,
      "step": 8000
    },
    {
      "epoch": 0.8119209093514185,
      "grad_norm": 3.314952850341797,
      "learning_rate": 2.9701977266214542e-05,
      "loss": 0.4101,
      "step": 8500
    },
    {
      "epoch": 0.8119209093514185,
      "eval_loss": 0.3561370372772217,
      "eval_runtime": 112.2203,
      "eval_samples_per_second": 233.033,
      "eval_steps_per_second": 3.645,
      "step": 8500
    }
  ],
  "logging_steps": 500,
  "max_steps": 20938,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.43163257192448e+17,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}