{
  "best_metric": 0.6810389161109924,
  "best_model_checkpoint": "YELP_full/checkpoint-10158",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 13544,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "grad_norm": 8.398588180541992,
      "learning_rate": 5.8227997637330186e-05,
      "loss": 0.9245,
      "step": 500
    },
    {
      "epoch": 0.3,
      "grad_norm": 3.5803122520446777,
      "learning_rate": 5.645599527466036e-05,
      "loss": 0.7913,
      "step": 1000
    },
    {
      "epoch": 0.44,
      "grad_norm": 3.3800671100616455,
      "learning_rate": 5.468399291199055e-05,
      "loss": 0.7613,
      "step": 1500
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.9862334728240967,
      "learning_rate": 5.291199054932074e-05,
      "loss": 0.7362,
      "step": 2000
    },
    {
      "epoch": 0.74,
      "grad_norm": 2.043750524520874,
      "learning_rate": 5.113998818665092e-05,
      "loss": 0.7262,
      "step": 2500
    },
    {
      "epoch": 0.89,
      "grad_norm": 2.8348655700683594,
      "learning_rate": 4.93679858239811e-05,
      "loss": 0.7165,
      "step": 3000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.69042,
      "eval_loss": 0.6984841227531433,
      "eval_runtime": 141.1206,
      "eval_samples_per_second": 354.307,
      "eval_steps_per_second": 1.849,
      "step": 3386
    },
    {
      "epoch": 1.03,
      "grad_norm": 2.2309489250183105,
      "learning_rate": 4.7595983461311283e-05,
      "loss": 0.7047,
      "step": 3500
    },
    {
      "epoch": 1.18,
      "grad_norm": 2.665053129196167,
      "learning_rate": 4.582398109864147e-05,
      "loss": 0.6764,
      "step": 4000
    },
    {
      "epoch": 1.33,
      "grad_norm": 2.590282678604126,
      "learning_rate": 4.4051978735971645e-05,
      "loss": 0.6733,
      "step": 4500
    },
    {
      "epoch": 1.48,
      "grad_norm": 2.890204429626465,
      "learning_rate": 4.227997637330183e-05,
      "loss": 0.6712,
      "step": 5000
    },
    {
      "epoch": 1.62,
      "grad_norm": 2.0544567108154297,
      "learning_rate": 4.050797401063202e-05,
      "loss": 0.6692,
      "step": 5500
    },
    {
      "epoch": 1.77,
      "grad_norm": 1.7416001558303833,
      "learning_rate": 3.87359716479622e-05,
      "loss": 0.6693,
      "step": 6000
    },
    {
      "epoch": 1.92,
      "grad_norm": 2.1172029972076416,
      "learning_rate": 3.696396928529238e-05,
      "loss": 0.6641,
      "step": 6500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.69884,
      "eval_loss": 0.6872273683547974,
      "eval_runtime": 141.4186,
      "eval_samples_per_second": 353.56,
      "eval_steps_per_second": 1.846,
      "step": 6772
    },
    {
      "epoch": 2.07,
      "grad_norm": 2.3501951694488525,
      "learning_rate": 3.5191966922622565e-05,
      "loss": 0.6458,
      "step": 7000
    },
    {
      "epoch": 2.22,
      "grad_norm": 1.6436865329742432,
      "learning_rate": 3.341996455995275e-05,
      "loss": 0.624,
      "step": 7500
    },
    {
      "epoch": 2.36,
      "grad_norm": 1.550398588180542,
      "learning_rate": 3.164796219728293e-05,
      "loss": 0.6243,
      "step": 8000
    },
    {
      "epoch": 2.51,
      "grad_norm": 3.080960512161255,
      "learning_rate": 2.9875959834613114e-05,
      "loss": 0.6246,
      "step": 8500
    },
    {
      "epoch": 2.66,
      "grad_norm": 2.6615045070648193,
      "learning_rate": 2.81039574719433e-05,
      "loss": 0.6246,
      "step": 9000
    },
    {
      "epoch": 2.81,
      "grad_norm": 1.667477011680603,
      "learning_rate": 2.633195510927348e-05,
      "loss": 0.622,
      "step": 9500
    },
    {
      "epoch": 2.95,
      "grad_norm": 2.1684648990631104,
      "learning_rate": 2.4559952746603663e-05,
      "loss": 0.62,
      "step": 10000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.70262,
      "eval_loss": 0.6810389161109924,
      "eval_runtime": 141.0454,
      "eval_samples_per_second": 354.496,
      "eval_steps_per_second": 1.85,
      "step": 10158
    },
    {
      "epoch": 3.1,
      "grad_norm": 2.221029043197632,
      "learning_rate": 2.2787950383933847e-05,
      "loss": 0.5931,
      "step": 10500
    },
    {
      "epoch": 3.25,
      "grad_norm": 2.1362805366516113,
      "learning_rate": 2.1015948021264028e-05,
      "loss": 0.5843,
      "step": 11000
    },
    {
      "epoch": 3.4,
      "grad_norm": 2.233245849609375,
      "learning_rate": 1.9243945658594212e-05,
      "loss": 0.5826,
      "step": 11500
    },
    {
      "epoch": 3.54,
      "grad_norm": 2.0913615226745605,
      "learning_rate": 1.7471943295924396e-05,
      "loss": 0.5824,
      "step": 12000
    },
    {
      "epoch": 3.69,
      "grad_norm": 2.6865012645721436,
      "learning_rate": 1.5699940933254577e-05,
      "loss": 0.5758,
      "step": 12500
    },
    {
      "epoch": 3.84,
      "grad_norm": 2.226073741912842,
      "learning_rate": 1.3927938570584761e-05,
      "loss": 0.5822,
      "step": 13000
    },
    {
      "epoch": 3.99,
      "grad_norm": 3.137810468673706,
      "learning_rate": 1.2155936207914945e-05,
      "loss": 0.5852,
      "step": 13500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.70036,
      "eval_loss": 0.6996423602104187,
      "eval_runtime": 141.3356,
      "eval_samples_per_second": 353.768,
      "eval_steps_per_second": 1.847,
      "step": 13544
    }
  ],
  "logging_steps": 500,
  "max_steps": 16930,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 1.247513087841362e+18,
  "train_batch_size": 192,
  "trial_name": null,
  "trial_params": null
}