{
  "best_metric": 0.9889349937438965,
  "best_model_checkpoint": "/kaggle/output/checkpoint-49000",
  "epoch": 2.077900912646675,
  "eval_steps": 1000,
  "global_step": 51000,
  "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
    },
    {
      "epoch": 1.22,
      "learning_rate": 8.329722222222223e-07,
      "loss": 1.0593,
      "step": 30000
    },
    {
      "epoch": 1.22,
      "eval_accuracy": 0.4375249500998004,
      "eval_loss": 1.0382143259048462,
      "eval_runtime": 54.3397,
      "eval_samples_per_second": 92.198,
      "eval_steps_per_second": 11.539,
      "step": 30000
    },
    {
      "epoch": 1.26,
      "learning_rate": 8.607500000000001e-07,
      "loss": 1.0642,
      "step": 31000
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.4367265469061876,
      "eval_loss": 1.043815016746521,
      "eval_runtime": 54.3991,
      "eval_samples_per_second": 92.097,
      "eval_steps_per_second": 11.526,
      "step": 31000
    },
    {
      "epoch": 1.3,
      "learning_rate": 8.885e-07,
      "loss": 1.0568,
      "step": 32000
    },
    {
      "epoch": 1.3,
      "eval_accuracy": 0.43253493013972055,
      "eval_loss": 1.040160059928894,
      "eval_runtime": 54.3558,
      "eval_samples_per_second": 92.171,
      "eval_steps_per_second": 11.535,
      "step": 32000
    },
    {
      "epoch": 1.34,
      "learning_rate": 9.162777777777779e-07,
      "loss": 1.0555,
      "step": 33000
    },
    {
      "epoch": 1.34,
      "eval_accuracy": 0.4471057884231537,
      "eval_loss": 1.0343328714370728,
      "eval_runtime": 54.3506,
      "eval_samples_per_second": 92.179,
      "eval_steps_per_second": 11.536,
      "step": 33000
    },
    {
      "epoch": 1.39,
      "learning_rate": 9.440555555555557e-07,
      "loss": 1.0502,
      "step": 34000
    },
    {
      "epoch": 1.39,
      "eval_accuracy": 0.4540918163672655,
      "eval_loss": 1.0305495262145996,
      "eval_runtime": 54.504,
      "eval_samples_per_second": 91.92,
      "eval_steps_per_second": 11.504,
      "step": 34000
    },
    {
      "epoch": 1.43,
      "learning_rate": 9.718055555555557e-07,
      "loss": 1.054,
      "step": 35000
    },
    {
      "epoch": 1.43,
      "eval_accuracy": 0.45169660678642715,
      "eval_loss": 1.0271915197372437,
      "eval_runtime": 54.3119,
      "eval_samples_per_second": 92.245,
      "eval_steps_per_second": 11.544,
      "step": 35000
    },
    {
      "epoch": 1.47,
      "learning_rate": 9.995833333333334e-07,
      "loss": 1.049,
      "step": 36000
    },
    {
      "epoch": 1.47,
      "eval_accuracy": 0.4540918163672655,
      "eval_loss": 1.024204969406128,
      "eval_runtime": 54.5422,
      "eval_samples_per_second": 91.855,
      "eval_steps_per_second": 11.496,
      "step": 36000
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0273333333333335e-06,
      "loss": 1.0466,
      "step": 37000
    },
    {
      "epoch": 1.51,
      "eval_accuracy": 0.45848303393213574,
      "eval_loss": 1.0162004232406616,
      "eval_runtime": 54.455,
      "eval_samples_per_second": 92.003,
      "eval_steps_per_second": 11.514,
      "step": 37000
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.0551111111111113e-06,
      "loss": 1.0443,
      "step": 38000
    },
    {
      "epoch": 1.55,
      "eval_accuracy": 0.46447105788423154,
      "eval_loss": 1.018208622932434,
      "eval_runtime": 54.3633,
      "eval_samples_per_second": 92.158,
      "eval_steps_per_second": 11.534,
      "step": 38000
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.082888888888889e-06,
      "loss": 1.0468,
      "step": 39000
    },
    {
      "epoch": 1.59,
      "eval_accuracy": 0.4620758483033932,
      "eval_loss": 1.0207008123397827,
      "eval_runtime": 54.3964,
      "eval_samples_per_second": 92.102,
      "eval_steps_per_second": 11.527,
      "step": 39000
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1106666666666668e-06,
      "loss": 1.0374,
      "step": 40000
    },
    {
      "epoch": 1.63,
      "eval_accuracy": 0.469061876247505,
      "eval_loss": 1.0093586444854736,
      "eval_runtime": 54.4765,
      "eval_samples_per_second": 91.966,
      "eval_steps_per_second": 11.51,
      "step": 40000
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.1384166666666669e-06,
      "loss": 1.0398,
      "step": 41000
    },
    {
      "epoch": 1.67,
      "eval_accuracy": 0.4590818363273453,
      "eval_loss": 1.0263862609863281,
      "eval_runtime": 54.4342,
      "eval_samples_per_second": 92.038,
      "eval_steps_per_second": 11.518,
      "step": 41000
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.1661944444444447e-06,
      "loss": 1.0353,
      "step": 42000
    },
    {
      "epoch": 1.71,
      "eval_accuracy": 0.4748502994011976,
      "eval_loss": 1.005388617515564,
      "eval_runtime": 54.5239,
      "eval_samples_per_second": 91.886,
      "eval_steps_per_second": 11.5,
      "step": 42000
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.1939444444444445e-06,
      "loss": 1.0366,
      "step": 43000
    },
    {
      "epoch": 1.75,
      "eval_accuracy": 0.4479041916167665,
      "eval_loss": 1.0292167663574219,
      "eval_runtime": 54.3843,
      "eval_samples_per_second": 92.122,
      "eval_steps_per_second": 11.529,
      "step": 43000
    },
    {
      "epoch": 1.79,
      "learning_rate": 1.2217222222222223e-06,
      "loss": 1.0371,
      "step": 44000
    },
    {
      "epoch": 1.79,
      "eval_accuracy": 0.48243512974051894,
      "eval_loss": 1.0023199319839478,
      "eval_runtime": 54.4813,
      "eval_samples_per_second": 91.958,
      "eval_steps_per_second": 11.509,
      "step": 44000
    },
    {
      "epoch": 1.83,
      "learning_rate": 1.2494722222222224e-06,
      "loss": 1.0347,
      "step": 45000
    },
    {
      "epoch": 1.83,
      "eval_accuracy": 0.47924151696606787,
      "eval_loss": 1.0037392377853394,
      "eval_runtime": 54.6308,
      "eval_samples_per_second": 91.706,
      "eval_steps_per_second": 11.477,
      "step": 45000
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.2772500000000001e-06,
      "loss": 1.0346,
      "step": 46000
    },
    {
      "epoch": 1.87,
      "eval_accuracy": 0.47764471057884234,
      "eval_loss": 1.008847951889038,
      "eval_runtime": 54.5558,
      "eval_samples_per_second": 91.833,
      "eval_steps_per_second": 11.493,
      "step": 46000
    },
    {
      "epoch": 1.91,
      "learning_rate": 1.3050000000000002e-06,
      "loss": 1.0267,
      "step": 47000
    },
    {
      "epoch": 1.91,
      "eval_accuracy": 0.4934131736526946,
      "eval_loss": 0.9937484264373779,
      "eval_runtime": 54.422,
      "eval_samples_per_second": 92.058,
      "eval_steps_per_second": 11.521,
      "step": 47000
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.3327777777777778e-06,
      "loss": 1.0319,
      "step": 48000
    },
    {
      "epoch": 1.96,
      "eval_accuracy": 0.475249500998004,
      "eval_loss": 1.0211399793624878,
      "eval_runtime": 54.503,
      "eval_samples_per_second": 91.921,
      "eval_steps_per_second": 11.504,
      "step": 48000
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.3605555555555555e-06,
      "loss": 1.024,
      "step": 49000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.49860279441117766,
      "eval_loss": 0.9889349937438965,
      "eval_runtime": 54.5076,
      "eval_samples_per_second": 91.914,
      "eval_steps_per_second": 11.503,
      "step": 49000
    },
    {
      "epoch": 2.04,
      "learning_rate": 1.3883055555555556e-06,
      "loss": 1.025,
      "step": 50000
    },
    {
      "epoch": 2.04,
      "eval_accuracy": 0.4872255489021956,
      "eval_loss": 1.0002667903900146,
      "eval_runtime": 54.5376,
      "eval_samples_per_second": 91.863,
      "eval_steps_per_second": 11.497,
      "step": 50000
    },
    {
      "epoch": 2.08,
      "learning_rate": 1.4160833333333334e-06,
      "loss": 1.0258,
      "step": 51000
    },
    {
      "epoch": 2.08,
      "eval_accuracy": 0.49101796407185627,
      "eval_loss": 0.9935734272003174,
      "eval_runtime": 54.4157,
      "eval_samples_per_second": 92.069,
      "eval_steps_per_second": 11.522,
      "step": 51000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10000000,
  "num_train_epochs": 408,
  "save_steps": 1000,
  "total_flos": 1.066075890599854e+17,
  "trial_name": null,
  "trial_params": null
}