{
  "best_metric": 1.0414146184921265,
  "best_model_checkpoint": "/kaggle/output/checkpoint-29000",
  "epoch": 1.1815514993481095,
  "eval_steps": 1000,
  "global_step": 29000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2.7777777777777777e-11,
      "loss": 1.1065,
      "step": 1
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.7750000000000004e-08,
      "loss": 1.1586,
      "step": 1000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.32954091816367265,
      "eval_loss": 1.1065887212753296,
      "eval_runtime": 54.4372,
      "eval_samples_per_second": 92.033,
      "eval_steps_per_second": 11.518,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.5527777777777784e-08,
      "loss": 1.1259,
      "step": 2000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.3435129740518962,
      "eval_loss": 1.1045255661010742,
      "eval_runtime": 54.458,
      "eval_samples_per_second": 91.997,
      "eval_steps_per_second": 11.513,
      "step": 2000
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.327777777777778e-08,
      "loss": 1.1229,
      "step": 3000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.3441117764471058,
      "eval_loss": 1.1050777435302734,
      "eval_runtime": 54.349,
      "eval_samples_per_second": 92.182,
      "eval_steps_per_second": 11.537,
      "step": 3000
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.1105555555555557e-07,
      "loss": 1.1237,
      "step": 4000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.3411177644710579,
      "eval_loss": 1.1015312671661377,
      "eval_runtime": 54.4388,
      "eval_samples_per_second": 92.03,
      "eval_steps_per_second": 11.518,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.3883333333333335e-07,
      "loss": 1.1202,
      "step": 5000
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.3405189620758483,
      "eval_loss": 1.1006112098693848,
      "eval_runtime": 54.4109,
      "eval_samples_per_second": 92.077,
      "eval_steps_per_second": 11.523,
      "step": 5000
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6658333333333335e-07,
      "loss": 1.1182,
      "step": 6000
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.35489021956087824,
      "eval_loss": 1.0982259511947632,
      "eval_runtime": 54.3841,
      "eval_samples_per_second": 92.123,
      "eval_steps_per_second": 11.529,
      "step": 6000
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.9436111111111112e-07,
      "loss": 1.1125,
      "step": 7000
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.35748502994011977,
      "eval_loss": 1.0954252481460571,
      "eval_runtime": 54.3975,
      "eval_samples_per_second": 92.1,
      "eval_steps_per_second": 11.526,
      "step": 7000
    },
    {
      "epoch": 0.33,
      "learning_rate": 2.2213888888888891e-07,
      "loss": 1.1096,
      "step": 8000
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.3664670658682635,
      "eval_loss": 1.0949230194091797,
      "eval_runtime": 54.3263,
      "eval_samples_per_second": 92.221,
      "eval_steps_per_second": 11.541,
      "step": 8000
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.4988888888888893e-07,
      "loss": 1.1057,
      "step": 9000
    },
    {
      "epoch": 0.37,
      "eval_accuracy": 0.34890219560878244,
      "eval_loss": 1.1037237644195557,
      "eval_runtime": 54.3657,
      "eval_samples_per_second": 92.154,
      "eval_steps_per_second": 11.533,
      "step": 9000
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.776666666666667e-07,
      "loss": 1.1056,
      "step": 10000
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.3838323353293413,
      "eval_loss": 1.0928014516830444,
      "eval_runtime": 54.6518,
      "eval_samples_per_second": 91.671,
      "eval_steps_per_second": 11.473,
      "step": 10000
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.054444444444444e-07,
      "loss": 1.1036,
      "step": 11000
    },
    {
      "epoch": 0.45,
      "eval_accuracy": 0.3838323353293413,
      "eval_loss": 1.0908663272857666,
      "eval_runtime": 54.579,
      "eval_samples_per_second": 91.794,
      "eval_steps_per_second": 11.488,
      "step": 11000
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.3322222222222225e-07,
      "loss": 1.1012,
      "step": 12000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.3872255489021956,
      "eval_loss": 1.0890672206878662,
      "eval_runtime": 54.5139,
      "eval_samples_per_second": 91.903,
      "eval_steps_per_second": 11.502,
      "step": 12000
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.609722222222222e-07,
      "loss": 1.103,
      "step": 13000
    },
    {
      "epoch": 0.53,
      "eval_accuracy": 0.34311377245508984,
      "eval_loss": 1.0942790508270264,
      "eval_runtime": 54.3951,
      "eval_samples_per_second": 92.104,
      "eval_steps_per_second": 11.527,
      "step": 13000
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.8875e-07,
      "loss": 1.0997,
      "step": 14000
    },
    {
      "epoch": 0.57,
      "eval_accuracy": 0.3712574850299401,
      "eval_loss": 1.0931994915008545,
      "eval_runtime": 54.3749,
      "eval_samples_per_second": 92.138,
      "eval_steps_per_second": 11.531,
      "step": 14000
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.1650000000000006e-07,
      "loss": 1.0993,
      "step": 15000
    },
    {
      "epoch": 0.61,
      "eval_accuracy": 0.3872255489021956,
      "eval_loss": 1.0887969732284546,
      "eval_runtime": 54.4149,
      "eval_samples_per_second": 92.07,
      "eval_steps_per_second": 11.523,
      "step": 15000
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.4427777777777783e-07,
      "loss": 1.0983,
      "step": 16000
    },
    {
      "epoch": 0.65,
      "eval_accuracy": 0.3980039920159681,
      "eval_loss": 1.0836474895477295,
      "eval_runtime": 54.3692,
      "eval_samples_per_second": 92.148,
      "eval_steps_per_second": 11.532,
      "step": 16000
    },
    {
      "epoch": 0.69,
      "learning_rate": 4.7202777777777785e-07,
      "loss": 1.0954,
      "step": 17000
    },
    {
      "epoch": 0.69,
      "eval_accuracy": 0.3996007984031936,
      "eval_loss": 1.0822749137878418,
      "eval_runtime": 54.3313,
      "eval_samples_per_second": 92.212,
      "eval_steps_per_second": 11.54,
      "step": 17000
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.998055555555556e-07,
      "loss": 1.097,
      "step": 18000
    },
    {
      "epoch": 0.73,
      "eval_accuracy": 0.4033932135728543,
      "eval_loss": 1.0845097303390503,
      "eval_runtime": 54.3565,
      "eval_samples_per_second": 92.169,
      "eval_steps_per_second": 11.535,
      "step": 18000
    },
    {
      "epoch": 0.77,
      "learning_rate": 5.275555555555556e-07,
      "loss": 1.0942,
      "step": 19000
    },
    {
      "epoch": 0.77,
      "eval_accuracy": 0.4039920159680639,
      "eval_loss": 1.0809950828552246,
      "eval_runtime": 54.3973,
      "eval_samples_per_second": 92.1,
      "eval_steps_per_second": 11.526,
      "step": 19000
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.553333333333334e-07,
      "loss": 1.0921,
      "step": 20000
    },
    {
      "epoch": 0.81,
      "eval_accuracy": 0.4055888223552894,
      "eval_loss": 1.0818568468093872,
      "eval_runtime": 54.8395,
      "eval_samples_per_second": 91.358,
      "eval_steps_per_second": 11.433,
      "step": 20000
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.830833333333334e-07,
      "loss": 1.0914,
      "step": 21000
    },
    {
      "epoch": 0.86,
      "eval_accuracy": 0.38163672654690617,
      "eval_loss": 1.08183753490448,
      "eval_runtime": 54.4076,
      "eval_samples_per_second": 92.083,
      "eval_steps_per_second": 11.524,
      "step": 21000
    },
    {
      "epoch": 0.9,
      "learning_rate": 6.108611111111111e-07,
      "loss": 1.091,
      "step": 22000
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.40658682634730536,
      "eval_loss": 1.0757951736450195,
      "eval_runtime": 54.4113,
      "eval_samples_per_second": 92.077,
      "eval_steps_per_second": 11.523,
      "step": 22000
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.386388888888889e-07,
      "loss": 1.0856,
      "step": 23000
    },
    {
      "epoch": 0.94,
      "eval_accuracy": 0.43313373253493015,
      "eval_loss": 1.0659974813461304,
      "eval_runtime": 54.3958,
      "eval_samples_per_second": 92.103,
      "eval_steps_per_second": 11.527,
      "step": 23000
    },
    {
      "epoch": 0.98,
      "learning_rate": 6.663888888888889e-07,
      "loss": 1.0794,
      "step": 24000
    },
    {
      "epoch": 0.98,
      "eval_accuracy": 0.43952095808383235,
      "eval_loss": 1.056550145149231,
      "eval_runtime": 54.6408,
      "eval_samples_per_second": 91.69,
      "eval_steps_per_second": 11.475,
      "step": 24000
    },
    {
      "epoch": 1.02,
      "learning_rate": 6.941666666666667e-07,
      "loss": 1.0714,
      "step": 25000
    },
    {
      "epoch": 1.02,
      "eval_accuracy": 0.4411177644710579,
      "eval_loss": 1.0531866550445557,
      "eval_runtime": 54.5122,
      "eval_samples_per_second": 91.906,
      "eval_steps_per_second": 11.502,
      "step": 25000
    },
    {
      "epoch": 1.06,
      "learning_rate": 7.219166666666666e-07,
      "loss": 1.0703,
      "step": 26000
    },
    {
      "epoch": 1.06,
      "eval_accuracy": 0.43273453093812375,
      "eval_loss": 1.0532232522964478,
      "eval_runtime": 54.5098,
      "eval_samples_per_second": 91.91,
      "eval_steps_per_second": 11.503,
      "step": 26000
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.496944444444444e-07,
      "loss": 1.0682,
      "step": 27000
    },
    {
      "epoch": 1.1,
      "eval_accuracy": 0.4353293413173653,
      "eval_loss": 1.050220012664795,
      "eval_runtime": 54.3702,
      "eval_samples_per_second": 92.146,
      "eval_steps_per_second": 11.532,
      "step": 27000
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.774444444444445e-07,
      "loss": 1.0671,
      "step": 28000
    },
    {
      "epoch": 1.14,
      "eval_accuracy": 0.43333333333333335,
      "eval_loss": 1.0470824241638184,
      "eval_runtime": 54.4316,
      "eval_samples_per_second": 92.042,
      "eval_steps_per_second": 11.519,
      "step": 28000
    },
    {
      "epoch": 1.18,
      "learning_rate": 8.052222222222223e-07,
      "loss": 1.0584,
      "step": 29000
    },
    {
      "epoch": 1.18,
      "eval_accuracy": 0.43972055888223555,
      "eval_loss": 1.0414146184921265,
      "eval_runtime": 54.4332,
      "eval_samples_per_second": 92.039,
      "eval_steps_per_second": 11.519,
      "step": 29000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10000000,
  "num_train_epochs": 408,
  "save_steps": 1000,
  "total_flos": 6.06200374862807e+16,
  "trial_name": null,
  "trial_params": null
}