{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.774057048004438,
  "global_step": 180000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 4.892535079835384e-05,
      "loss": 3.2017,
      "step": 5000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.4179681193308354,
      "eval_loss": 2.969822883605957,
      "eval_runtime": 1929.4803,
      "eval_samples_per_second": 19.496,
      "eval_steps_per_second": 2.437,
      "step": 5000
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.785070159670768e-05,
      "loss": 2.9116,
      "step": 10000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.4297956163670618,
      "eval_loss": 2.8731179237365723,
      "eval_runtime": 1919.4124,
      "eval_samples_per_second": 19.598,
      "eval_steps_per_second": 2.45,
      "step": 10000
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.677605239506152e-05,
      "loss": 2.8412,
      "step": 15000
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.4368075611910975,
      "eval_loss": 2.81874942779541,
      "eval_runtime": 2015.9454,
      "eval_samples_per_second": 18.66,
      "eval_steps_per_second": 2.333,
      "step": 15000
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.5701403193415356e-05,
      "loss": 2.7937,
      "step": 20000
    },
    {
      "epoch": 0.09,
      "eval_accuracy": 0.44204520475458375,
      "eval_loss": 2.778646230697632,
      "eval_runtime": 1919.08,
      "eval_samples_per_second": 19.602,
      "eval_steps_per_second": 2.451,
      "step": 20000
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.4626753991769196e-05,
      "loss": 2.7614,
      "step": 25000
    },
    {
      "epoch": 0.11,
      "eval_accuracy": 0.44610035327770187,
      "eval_loss": 2.7471842765808105,
      "eval_runtime": 1919.775,
      "eval_samples_per_second": 19.594,
      "eval_steps_per_second": 2.45,
      "step": 25000
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.355188977427636e-05,
      "loss": 2.7295,
      "step": 30000
    },
    {
      "epoch": 0.13,
      "eval_accuracy": 0.4501440419543679,
      "eval_loss": 2.7196261882781982,
      "eval_runtime": 1928.9026,
      "eval_samples_per_second": 19.502,
      "eval_steps_per_second": 2.438,
      "step": 30000
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.24772405726302e-05,
      "loss": 2.7008,
      "step": 35000
    },
    {
      "epoch": 0.15,
      "eval_accuracy": 0.45296033170252703,
      "eval_loss": 2.6961302757263184,
      "eval_runtime": 1926.8386,
      "eval_samples_per_second": 19.523,
      "eval_steps_per_second": 2.441,
      "step": 35000
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.140237635513737e-05,
      "loss": 2.6827,
      "step": 40000
    },
    {
      "epoch": 0.17,
      "eval_accuracy": 0.4561378015092748,
      "eval_loss": 2.6738195419311523,
      "eval_runtime": 1920.5318,
      "eval_samples_per_second": 19.587,
      "eval_steps_per_second": 2.449,
      "step": 40000
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.032772715349121e-05,
      "loss": 2.6545,
      "step": 45000
    },
    {
      "epoch": 0.19,
      "eval_accuracy": 0.4590067649734393,
      "eval_loss": 2.6545751094818115,
      "eval_runtime": 1933.7083,
      "eval_samples_per_second": 19.453,
      "eval_steps_per_second": 2.432,
      "step": 45000
    },
    {
      "epoch": 0.22,
      "learning_rate": 3.9252862935998384e-05,
      "loss": 2.6419,
      "step": 50000
    },
    {
      "epoch": 0.22,
      "eval_accuracy": 0.4616434651550895,
      "eval_loss": 2.635937452316284,
      "eval_runtime": 1931.385,
      "eval_samples_per_second": 19.477,
      "eval_steps_per_second": 2.435,
      "step": 50000
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.8178213734352224e-05,
      "loss": 2.6208,
      "step": 55000
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.4644954077588773,
      "eval_loss": 2.617478847503662,
      "eval_runtime": 1935.6188,
      "eval_samples_per_second": 19.434,
      "eval_steps_per_second": 2.43,
      "step": 55000
    },
    {
      "epoch": 0.26,
      "learning_rate": 3.7103564532706064e-05,
      "loss": 2.604,
      "step": 60000
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.4666273030036154,
      "eval_loss": 2.6017534732818604,
      "eval_runtime": 1934.3532,
      "eval_samples_per_second": 19.447,
      "eval_steps_per_second": 2.431,
      "step": 60000
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.60289153310599e-05,
      "loss": 2.5883,
      "step": 65000
    },
    {
      "epoch": 0.28,
      "eval_accuracy": 0.4690363654190064,
      "eval_loss": 2.5858962535858154,
      "eval_runtime": 1937.9051,
      "eval_samples_per_second": 19.411,
      "eval_steps_per_second": 2.427,
      "step": 65000
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.495426612941374e-05,
      "loss": 2.5733,
      "step": 70000
    },
    {
      "epoch": 0.3,
      "eval_accuracy": 0.47110511457104925,
      "eval_loss": 2.5711781978607178,
      "eval_runtime": 1929.1817,
      "eval_samples_per_second": 19.499,
      "eval_steps_per_second": 2.438,
      "step": 70000
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.387940191192091e-05,
      "loss": 2.5603,
      "step": 75000
    },
    {
      "epoch": 0.32,
      "eval_accuracy": 0.4735584312234197,
      "eval_loss": 2.5554990768432617,
      "eval_runtime": 1924.3705,
      "eval_samples_per_second": 19.548,
      "eval_steps_per_second": 2.444,
      "step": 75000
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.280475271027475e-05,
      "loss": 2.5462,
      "step": 80000
    },
    {
      "epoch": 0.34,
      "eval_accuracy": 0.4755388538038284,
      "eval_loss": 2.541663646697998,
      "eval_runtime": 1932.1982,
      "eval_samples_per_second": 19.469,
      "eval_steps_per_second": 2.434,
      "step": 80000
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.173010350862859e-05,
      "loss": 2.5267,
      "step": 85000
    },
    {
      "epoch": 0.37,
      "eval_accuracy": 0.47766700705788817,
      "eval_loss": 2.528374433517456,
      "eval_runtime": 1933.9592,
      "eval_samples_per_second": 19.451,
      "eval_steps_per_second": 2.432,
      "step": 85000
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.0655454306982425e-05,
      "loss": 2.5153,
      "step": 90000
    },
    {
      "epoch": 0.39,
      "eval_accuracy": 0.47989050311610376,
      "eval_loss": 2.514660596847534,
      "eval_runtime": 1929.164,
      "eval_samples_per_second": 19.499,
      "eval_steps_per_second": 2.438,
      "step": 90000
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.9580805105336262e-05,
      "loss": 2.5028,
      "step": 95000
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.481903668114947,
      "eval_loss": 2.501615285873413,
      "eval_runtime": 1929.1634,
      "eval_samples_per_second": 19.499,
      "eval_steps_per_second": 2.438,
      "step": 95000
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.8505940887843435e-05,
      "loss": 2.4881,
      "step": 100000
    },
    {
      "epoch": 0.43,
      "eval_accuracy": 0.4839338799602133,
      "eval_loss": 2.488723039627075,
      "eval_runtime": 1930.4001,
      "eval_samples_per_second": 19.487,
      "eval_steps_per_second": 2.436,
      "step": 100000
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.7431291686197276e-05,
      "loss": 2.4758,
      "step": 105000
    },
    {
      "epoch": 0.45,
      "eval_accuracy": 0.48596689829848827,
      "eval_loss": 2.475806951522827,
      "eval_runtime": 1923.9804,
      "eval_samples_per_second": 19.552,
      "eval_steps_per_second": 2.444,
      "step": 105000
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.6356642484551112e-05,
      "loss": 2.4632,
      "step": 110000
    },
    {
      "epoch": 0.47,
      "eval_accuracy": 0.4880078683669545,
      "eval_loss": 2.463679552078247,
      "eval_runtime": 1943.6457,
      "eval_samples_per_second": 19.354,
      "eval_steps_per_second": 2.42,
      "step": 110000
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.5281993282904953e-05,
      "loss": 2.4543,
      "step": 115000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.4898118976645586,
      "eval_loss": 2.4515490531921387,
      "eval_runtime": 1940.3072,
      "eval_samples_per_second": 19.387,
      "eval_steps_per_second": 2.424,
      "step": 115000
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.4206914049565453e-05,
      "loss": 2.4416,
      "step": 120000
    },
    {
      "epoch": 0.52,
      "eval_accuracy": 0.49199028194626443,
      "eval_loss": 2.4388370513916016,
      "eval_runtime": 1928.5903,
      "eval_samples_per_second": 19.505,
      "eval_steps_per_second": 2.439,
      "step": 120000
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.313247986376596e-05,
      "loss": 2.4256,
      "step": 125000
    },
    {
      "epoch": 0.54,
      "eval_accuracy": 0.49386564294117247,
      "eval_loss": 2.4272964000701904,
      "eval_runtime": 1927.1096,
      "eval_samples_per_second": 19.52,
      "eval_steps_per_second": 2.44,
      "step": 125000
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.20578306621198e-05,
      "loss": 2.4166,
      "step": 130000
    },
    {
      "epoch": 0.56,
      "eval_accuracy": 0.4959010260096677,
      "eval_loss": 2.4158496856689453,
      "eval_runtime": 1932.6805,
      "eval_samples_per_second": 19.464,
      "eval_steps_per_second": 2.433,
      "step": 130000
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.0982751428780303e-05,
      "loss": 2.4052,
      "step": 135000
    },
    {
      "epoch": 0.58,
      "eval_accuracy": 0.4977953568184306,
      "eval_loss": 2.4038643836975098,
      "eval_runtime": 1925.4197,
      "eval_samples_per_second": 19.537,
      "eval_steps_per_second": 2.443,
      "step": 135000
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.990831724298081e-05,
      "loss": 2.3936,
      "step": 140000
    },
    {
      "epoch": 0.6,
      "eval_accuracy": 0.49973316228278164,
      "eval_loss": 2.392191171646118,
      "eval_runtime": 1928.0691,
      "eval_samples_per_second": 19.51,
      "eval_steps_per_second": 2.439,
      "step": 140000
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.8833668041334647e-05,
      "loss": 2.379,
      "step": 145000
    },
    {
      "epoch": 0.62,
      "eval_accuracy": 0.5019234221876816,
      "eval_loss": 2.38037371635437,
      "eval_runtime": 1923.4067,
      "eval_samples_per_second": 19.557,
      "eval_steps_per_second": 2.445,
      "step": 145000
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.7759018839688484e-05,
      "loss": 2.3659,
      "step": 150000
    },
    {
      "epoch": 0.65,
      "eval_accuracy": 0.5037539832386363,
      "eval_loss": 2.3692569732666016,
      "eval_runtime": 1925.9734,
      "eval_samples_per_second": 19.531,
      "eval_steps_per_second": 2.442,
      "step": 150000
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.6684154622195657e-05,
      "loss": 2.3588,
      "step": 155000
    },
    {
      "epoch": 0.67,
      "eval_accuracy": 0.5057738006653519,
      "eval_loss": 2.357806444168091,
      "eval_runtime": 1927.4777,
      "eval_samples_per_second": 19.516,
      "eval_steps_per_second": 2.44,
      "step": 155000
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.5609505420549494e-05,
      "loss": 2.3529,
      "step": 160000
    },
    {
      "epoch": 0.69,
      "eval_accuracy": 0.5079495603563737,
      "eval_loss": 2.3464465141296387,
      "eval_runtime": 1926.6683,
      "eval_samples_per_second": 19.524,
      "eval_steps_per_second": 2.441,
      "step": 160000
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.4535071234750002e-05,
      "loss": 2.3338,
      "step": 165000
    },
    {
      "epoch": 0.71,
      "eval_accuracy": 0.5098386939558613,
      "eval_loss": 2.335519313812256,
      "eval_runtime": 1925.6719,
      "eval_samples_per_second": 19.534,
      "eval_steps_per_second": 2.442,
      "step": 165000
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.3460422033103839e-05,
      "loss": 2.3238,
      "step": 170000
    },
    {
      "epoch": 0.73,
      "eval_accuracy": 0.5118405290384843,
      "eval_loss": 2.3250160217285156,
      "eval_runtime": 1925.0126,
      "eval_samples_per_second": 19.541,
      "eval_steps_per_second": 2.443,
      "step": 170000
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.2385557815611011e-05,
      "loss": 2.3139,
      "step": 175000
    },
    {
      "epoch": 0.75,
      "eval_accuracy": 0.5137200997729053,
      "eval_loss": 2.314518451690674,
      "eval_runtime": 1927.2965,
      "eval_samples_per_second": 19.518,
      "eval_steps_per_second": 2.44,
      "step": 175000
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.1311123629811518e-05,
      "loss": 2.3035,
      "step": 180000
    },
    {
      "epoch": 0.77,
      "eval_accuracy": 0.5153991621734844,
      "eval_loss": 2.3050577640533447,
      "eval_runtime": 1927.482,
      "eval_samples_per_second": 19.516,
      "eval_steps_per_second": 2.44,
      "step": 180000
    }
  ],
  "max_steps": 232541,
  "num_train_epochs": 1,
  "total_flos": 2.138588502294528e+19,
  "trial_name": null,
  "trial_params": null
}