{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 60.0,
  "global_step": 18720,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.555956678700361,
      "eval_loss": 0.3513844907283783,
      "eval_runtime": 3.2893,
      "eval_samples_per_second": 84.213,
      "eval_steps_per_second": 10.641,
      "step": 312
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 1.0,
      "step": 312
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.004866452991452991,
      "loss": 0.6229,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5487364620938628,
      "eval_loss": 0.6273205876350403,
      "eval_runtime": 3.2918,
      "eval_samples_per_second": 84.149,
      "eval_steps_per_second": 10.633,
      "step": 624
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 2.0,
      "step": 624
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.8084543943405151,
      "eval_runtime": 3.2989,
      "eval_samples_per_second": 83.968,
      "eval_steps_per_second": 10.61,
      "step": 936
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 3.0,
      "step": 936
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.004732905982905983,
      "loss": 0.5188,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.6059607267379761,
      "eval_runtime": 3.2975,
      "eval_samples_per_second": 84.002,
      "eval_steps_per_second": 10.614,
      "step": 1248
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 4.0,
      "step": 1248
    },
    {
      "epoch": 4.81,
      "learning_rate": 0.004599358974358974,
      "loss": 0.4693,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.49458483754512633,
      "eval_loss": 0.360698938369751,
      "eval_runtime": 3.3006,
      "eval_samples_per_second": 83.925,
      "eval_steps_per_second": 10.604,
      "step": 1560
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 5.0,
      "step": 1560
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.48014440433212996,
      "eval_loss": 0.3896714150905609,
      "eval_runtime": 3.302,
      "eval_samples_per_second": 83.888,
      "eval_steps_per_second": 10.6,
      "step": 1872
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 6.0,
      "step": 1872
    },
    {
      "epoch": 6.41,
      "learning_rate": 0.004465811965811966,
      "loss": 0.4249,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.5828221440315247,
      "eval_runtime": 3.2974,
      "eval_samples_per_second": 84.006,
      "eval_steps_per_second": 10.615,
      "step": 2184
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 7.0,
      "step": 2184
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5306859205776173,
      "eval_loss": 0.4718327224254608,
      "eval_runtime": 3.2969,
      "eval_samples_per_second": 84.018,
      "eval_steps_per_second": 10.616,
      "step": 2496
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 8.0,
      "step": 2496
    },
    {
      "epoch": 8.01,
      "learning_rate": 0.004332264957264957,
      "loss": 0.4226,
      "step": 2500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.534294843673706,
      "eval_runtime": 3.2965,
      "eval_samples_per_second": 84.028,
      "eval_steps_per_second": 10.617,
      "step": 2808
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 9.0,
      "step": 2808
    },
    {
      "epoch": 9.62,
      "learning_rate": 0.004198717948717949,
      "loss": 0.42,
      "step": 3000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5451263537906137,
      "eval_loss": 0.3477543592453003,
      "eval_runtime": 3.2981,
      "eval_samples_per_second": 83.988,
      "eval_steps_per_second": 10.612,
      "step": 3120
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 10.0,
      "step": 3120
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.4041728079319,
      "eval_runtime": 3.3027,
      "eval_samples_per_second": 83.87,
      "eval_steps_per_second": 10.597,
      "step": 3432
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 11.0,
      "step": 3432
    },
    {
      "epoch": 11.22,
      "learning_rate": 0.00406517094017094,
      "loss": 0.407,
      "step": 3500
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.4693140794223827,
      "eval_loss": 0.5783184170722961,
      "eval_runtime": 3.2982,
      "eval_samples_per_second": 83.986,
      "eval_steps_per_second": 10.612,
      "step": 3744
    },
    {
      "best_epoch": 0,
      "best_eval_accuracy": 0.555956678700361,
      "epoch": 12.0,
      "step": 3744
    },
    {
      "epoch": 12.82,
      "learning_rate": 0.003931623931623931,
      "loss": 0.4156,
      "step": 4000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5740072202166066,
      "eval_loss": 0.3465688228607178,
      "eval_runtime": 3.3071,
      "eval_samples_per_second": 83.76,
      "eval_steps_per_second": 10.583,
      "step": 4056
    },
    {
      "best_epoch": 12,
      "best_eval_accuracy": 0.5740072202166066,
      "epoch": 13.0,
      "step": 4056
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.5379061371841155,
      "eval_loss": 0.37199193239212036,
      "eval_runtime": 3.304,
      "eval_samples_per_second": 83.838,
      "eval_steps_per_second": 10.593,
      "step": 4368
    },
    {
      "best_epoch": 12,
      "best_eval_accuracy": 0.5740072202166066,
      "epoch": 14.0,
      "step": 4368
    },
    {
      "epoch": 14.42,
      "learning_rate": 0.003798076923076923,
      "loss": 0.3784,
      "step": 4500
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.631768953068592,
      "eval_loss": 0.3414437770843506,
      "eval_runtime": 3.2995,
      "eval_samples_per_second": 83.953,
      "eval_steps_per_second": 10.608,
      "step": 4680
    },
    {
      "best_epoch": 14,
      "best_eval_accuracy": 0.631768953068592,
      "epoch": 15.0,
      "step": 4680
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.631768953068592,
      "eval_loss": 0.33300819993019104,
      "eval_runtime": 3.2992,
      "eval_samples_per_second": 83.96,
      "eval_steps_per_second": 10.609,
      "step": 4992
    },
    {
      "best_epoch": 14,
      "best_eval_accuracy": 0.631768953068592,
      "epoch": 16.0,
      "step": 4992
    },
    {
      "epoch": 16.03,
      "learning_rate": 0.003664529914529914,
      "loss": 0.3734,
      "step": 5000
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5956678700361011,
      "eval_loss": 0.46307575702667236,
      "eval_runtime": 3.321,
      "eval_samples_per_second": 83.409,
      "eval_steps_per_second": 10.539,
      "step": 5304
    },
    {
      "best_epoch": 14,
      "best_eval_accuracy": 0.631768953068592,
      "epoch": 17.0,
      "step": 5304
    },
    {
      "epoch": 17.63,
      "learning_rate": 0.003530982905982906,
      "loss": 0.3573,
      "step": 5500
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5848375451263538,
      "eval_loss": 0.3375145494937897,
      "eval_runtime": 3.3052,
      "eval_samples_per_second": 83.807,
      "eval_steps_per_second": 10.589,
      "step": 5616
    },
    {
      "best_epoch": 14,
      "best_eval_accuracy": 0.631768953068592,
      "epoch": 18.0,
      "step": 5616
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.6606498194945848,
      "eval_loss": 0.34285253286361694,
      "eval_runtime": 3.3028,
      "eval_samples_per_second": 83.868,
      "eval_steps_per_second": 10.597,
      "step": 5928
    },
    {
      "best_epoch": 18,
      "best_eval_accuracy": 0.6606498194945848,
      "epoch": 19.0,
      "step": 5928
    },
    {
      "epoch": 19.23,
      "learning_rate": 0.0033974358974358976,
      "loss": 0.3516,
      "step": 6000
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6606498194945848,
      "eval_loss": 0.33439937233924866,
      "eval_runtime": 3.3006,
      "eval_samples_per_second": 83.925,
      "eval_steps_per_second": 10.604,
      "step": 6240
    },
    {
      "best_epoch": 18,
      "best_eval_accuracy": 0.6606498194945848,
      "epoch": 20.0,
      "step": 6240
    },
    {
      "epoch": 20.83,
      "learning_rate": 0.003263888888888889,
      "loss": 0.3399,
      "step": 6500
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.6678700361010831,
      "eval_loss": 0.3670630156993866,
      "eval_runtime": 3.2987,
      "eval_samples_per_second": 83.972,
      "eval_steps_per_second": 10.61,
      "step": 6552
    },
    {
      "best_epoch": 20,
      "best_eval_accuracy": 0.6678700361010831,
      "epoch": 21.0,
      "step": 6552
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.6642599277978339,
      "eval_loss": 0.3484797775745392,
      "eval_runtime": 3.2984,
      "eval_samples_per_second": 83.981,
      "eval_steps_per_second": 10.611,
      "step": 6864
    },
    {
      "best_epoch": 20,
      "best_eval_accuracy": 0.6678700361010831,
      "epoch": 22.0,
      "step": 6864
    },
    {
      "epoch": 22.44,
      "learning_rate": 0.0031303418803418806,
      "loss": 0.3345,
      "step": 7000
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.6678700361010831,
      "eval_loss": 0.3415996730327606,
      "eval_runtime": 3.3001,
      "eval_samples_per_second": 83.938,
      "eval_steps_per_second": 10.606,
      "step": 7176
    },
    {
      "best_epoch": 20,
      "best_eval_accuracy": 0.6678700361010831,
      "epoch": 23.0,
      "step": 7176
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6967509025270758,
      "eval_loss": 0.3263380825519562,
      "eval_runtime": 3.2986,
      "eval_samples_per_second": 83.976,
      "eval_steps_per_second": 10.611,
      "step": 7488
    },
    {
      "best_epoch": 23,
      "best_eval_accuracy": 0.6967509025270758,
      "epoch": 24.0,
      "step": 7488
    },
    {
      "epoch": 24.04,
      "learning_rate": 0.0029967948717948716,
      "loss": 0.325,
      "step": 7500
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.6895306859205776,
      "eval_loss": 0.3330814838409424,
      "eval_runtime": 3.2958,
      "eval_samples_per_second": 84.047,
      "eval_steps_per_second": 10.62,
      "step": 7800
    },
    {
      "best_epoch": 23,
      "best_eval_accuracy": 0.6967509025270758,
      "epoch": 25.0,
      "step": 7800
    },
    {
      "epoch": 25.64,
      "learning_rate": 0.002863247863247863,
      "loss": 0.3197,
      "step": 8000
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.6787003610108303,
      "eval_loss": 0.3590630888938904,
      "eval_runtime": 3.295,
      "eval_samples_per_second": 84.067,
      "eval_steps_per_second": 10.622,
      "step": 8112
    },
    {
      "best_epoch": 23,
      "best_eval_accuracy": 0.6967509025270758,
      "epoch": 26.0,
      "step": 8112
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.7292418772563177,
      "eval_loss": 0.3174915611743927,
      "eval_runtime": 3.2978,
      "eval_samples_per_second": 83.996,
      "eval_steps_per_second": 10.613,
      "step": 8424
    },
    {
      "best_epoch": 26,
      "best_eval_accuracy": 0.7292418772563177,
      "epoch": 27.0,
      "step": 8424
    },
    {
      "epoch": 27.24,
      "learning_rate": 0.0027297008547008546,
      "loss": 0.3165,
      "step": 8500
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.7148014440433214,
      "eval_loss": 0.3208056390285492,
      "eval_runtime": 3.2957,
      "eval_samples_per_second": 84.049,
      "eval_steps_per_second": 10.62,
      "step": 8736
    },
    {
      "best_epoch": 26,
      "best_eval_accuracy": 0.7292418772563177,
      "epoch": 28.0,
      "step": 8736
    },
    {
      "epoch": 28.85,
      "learning_rate": 0.0025961538461538466,
      "loss": 0.3122,
      "step": 9000
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.7292418772563177,
      "eval_loss": 0.31998708844184875,
      "eval_runtime": 3.2973,
      "eval_samples_per_second": 84.009,
      "eval_steps_per_second": 10.615,
      "step": 9048
    },
    {
      "best_epoch": 26,
      "best_eval_accuracy": 0.7292418772563177,
      "epoch": 29.0,
      "step": 9048
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.6570397111913358,
      "eval_loss": 0.37904560565948486,
      "eval_runtime": 3.2972,
      "eval_samples_per_second": 84.01,
      "eval_steps_per_second": 10.615,
      "step": 9360
    },
    {
      "best_epoch": 26,
      "best_eval_accuracy": 0.7292418772563177,
      "epoch": 30.0,
      "step": 9360
    },
    {
      "epoch": 30.45,
      "learning_rate": 0.0024626068376068376,
      "loss": 0.3072,
      "step": 9500
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.7111913357400722,
      "eval_loss": 0.32207196950912476,
      "eval_runtime": 3.2959,
      "eval_samples_per_second": 84.044,
      "eval_steps_per_second": 10.619,
      "step": 9672
    },
    {
      "best_epoch": 26,
      "best_eval_accuracy": 0.7292418772563177,
      "epoch": 31.0,
      "step": 9672
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.7364620938628159,
      "eval_loss": 0.3263281583786011,
      "eval_runtime": 3.2973,
      "eval_samples_per_second": 84.008,
      "eval_steps_per_second": 10.615,
      "step": 9984
    },
    {
      "best_epoch": 31,
      "best_eval_accuracy": 0.7364620938628159,
      "epoch": 32.0,
      "step": 9984
    },
    {
      "epoch": 32.05,
      "learning_rate": 0.002329059829059829,
      "loss": 0.3041,
      "step": 10000
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.7292418772563177,
      "eval_loss": 0.33217206597328186,
      "eval_runtime": 3.2981,
      "eval_samples_per_second": 83.989,
      "eval_steps_per_second": 10.612,
      "step": 10296
    },
    {
      "best_epoch": 31,
      "best_eval_accuracy": 0.7364620938628159,
      "epoch": 33.0,
      "step": 10296
    },
    {
      "epoch": 33.65,
      "learning_rate": 0.0021955128205128206,
      "loss": 0.2885,
      "step": 10500
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.7364620938628159,
      "eval_loss": 0.32957446575164795,
      "eval_runtime": 3.2979,
      "eval_samples_per_second": 83.992,
      "eval_steps_per_second": 10.613,
      "step": 10608
    },
    {
      "best_epoch": 31,
      "best_eval_accuracy": 0.7364620938628159,
      "epoch": 34.0,
      "step": 10608
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.7220216606498195,
      "eval_loss": 0.32645183801651,
      "eval_runtime": 3.2945,
      "eval_samples_per_second": 84.081,
      "eval_steps_per_second": 10.624,
      "step": 10920
    },
    {
      "best_epoch": 31,
      "best_eval_accuracy": 0.7364620938628159,
      "epoch": 35.0,
      "step": 10920
    },
    {
      "epoch": 35.26,
      "learning_rate": 0.002061965811965812,
      "loss": 0.2875,
      "step": 11000
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.7509025270758123,
      "eval_loss": 0.3236461281776428,
      "eval_runtime": 3.2925,
      "eval_samples_per_second": 84.13,
      "eval_steps_per_second": 10.63,
      "step": 11232
    },
    {
      "best_epoch": 35,
      "best_eval_accuracy": 0.7509025270758123,
      "epoch": 36.0,
      "step": 11232
    },
    {
      "epoch": 36.86,
      "learning_rate": 0.0019284188034188036,
      "loss": 0.2848,
      "step": 11500
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.7111913357400722,
      "eval_loss": 0.3483838438987732,
      "eval_runtime": 3.2953,
      "eval_samples_per_second": 84.06,
      "eval_steps_per_second": 10.621,
      "step": 11544
    },
    {
      "best_epoch": 35,
      "best_eval_accuracy": 0.7509025270758123,
      "epoch": 37.0,
      "step": 11544
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.7364620938628159,
      "eval_loss": 0.3266400396823883,
      "eval_runtime": 3.2944,
      "eval_samples_per_second": 84.083,
      "eval_steps_per_second": 10.624,
      "step": 11856
    },
    {
      "best_epoch": 35,
      "best_eval_accuracy": 0.7509025270758123,
      "epoch": 38.0,
      "step": 11856
    },
    {
      "epoch": 38.46,
      "learning_rate": 0.0017948717948717949,
      "loss": 0.2766,
      "step": 12000
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.7472924187725631,
      "eval_loss": 0.3303963840007782,
      "eval_runtime": 3.2974,
      "eval_samples_per_second": 84.007,
      "eval_steps_per_second": 10.615,
      "step": 12168
    },
    {
      "best_epoch": 35,
      "best_eval_accuracy": 0.7509025270758123,
      "epoch": 39.0,
      "step": 12168
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.740072202166065,
      "eval_loss": 0.330539733171463,
      "eval_runtime": 3.2965,
      "eval_samples_per_second": 84.029,
      "eval_steps_per_second": 10.617,
      "step": 12480
    },
    {
      "best_epoch": 35,
      "best_eval_accuracy": 0.7509025270758123,
      "epoch": 40.0,
      "step": 12480
    },
    {
      "epoch": 40.06,
      "learning_rate": 0.0016613247863247864,
      "loss": 0.2743,
      "step": 12500
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.7545126353790613,
      "eval_loss": 0.3287397027015686,
      "eval_runtime": 3.2968,
      "eval_samples_per_second": 84.021,
      "eval_steps_per_second": 10.616,
      "step": 12792
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 41.0,
      "step": 12792
    },
    {
      "epoch": 41.67,
      "learning_rate": 0.0015277777777777779,
      "loss": 0.2708,
      "step": 13000
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.7364620938628159,
      "eval_loss": 0.32920366525650024,
      "eval_runtime": 3.3008,
      "eval_samples_per_second": 83.92,
      "eval_steps_per_second": 10.604,
      "step": 13104
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 42.0,
      "step": 13104
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.7256317689530686,
      "eval_loss": 0.33632931113243103,
      "eval_runtime": 3.2974,
      "eval_samples_per_second": 84.006,
      "eval_steps_per_second": 10.614,
      "step": 13416
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 43.0,
      "step": 13416
    },
    {
      "epoch": 43.27,
      "learning_rate": 0.0013942307692307694,
      "loss": 0.2662,
      "step": 13500
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.7328519855595668,
      "eval_loss": 0.32028645277023315,
      "eval_runtime": 3.3021,
      "eval_samples_per_second": 83.887,
      "eval_steps_per_second": 10.599,
      "step": 13728
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 44.0,
      "step": 13728
    },
    {
      "epoch": 44.87,
      "learning_rate": 0.0012606837606837606,
      "loss": 0.2636,
      "step": 14000
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.740072202166065,
      "eval_loss": 0.3337833881378174,
      "eval_runtime": 3.3037,
      "eval_samples_per_second": 83.846,
      "eval_steps_per_second": 10.594,
      "step": 14040
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 45.0,
      "step": 14040
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.7364620938628159,
      "eval_loss": 0.3480018973350525,
      "eval_runtime": 3.303,
      "eval_samples_per_second": 83.862,
      "eval_steps_per_second": 10.596,
      "step": 14352
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 46.0,
      "step": 14352
    },
    {
      "epoch": 46.47,
      "learning_rate": 0.0011271367521367521,
      "loss": 0.261,
      "step": 14500
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.740072202166065,
      "eval_loss": 0.3282296061515808,
      "eval_runtime": 3.3097,
      "eval_samples_per_second": 83.694,
      "eval_steps_per_second": 10.575,
      "step": 14664
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 47.0,
      "step": 14664
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.7328519855595668,
      "eval_loss": 0.33301836252212524,
      "eval_runtime": 3.2975,
      "eval_samples_per_second": 84.003,
      "eval_steps_per_second": 10.614,
      "step": 14976
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 48.0,
      "step": 14976
    },
    {
      "epoch": 48.08,
      "learning_rate": 0.0009935897435897436,
      "loss": 0.2585,
      "step": 15000
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.7292418772563177,
      "eval_loss": 0.35189110040664673,
      "eval_runtime": 3.3,
      "eval_samples_per_second": 83.939,
      "eval_steps_per_second": 10.606,
      "step": 15288
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 49.0,
      "step": 15288
    },
    {
      "epoch": 49.68,
      "learning_rate": 0.0008600427350427351,
      "loss": 0.2561,
      "step": 15500
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.7472924187725631,
      "eval_loss": 0.3215104937553406,
      "eval_runtime": 3.2994,
      "eval_samples_per_second": 83.954,
      "eval_steps_per_second": 10.608,
      "step": 15600
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 50.0,
      "step": 15600
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.740072202166065,
      "eval_loss": 0.33880314230918884,
      "eval_runtime": 3.2982,
      "eval_samples_per_second": 83.986,
      "eval_steps_per_second": 10.612,
      "step": 15912
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 51.0,
      "step": 15912
    },
    {
      "epoch": 51.28,
      "learning_rate": 0.0007264957264957266,
      "loss": 0.2569,
      "step": 16000
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.7364620938628159,
      "eval_loss": 0.3326878845691681,
      "eval_runtime": 3.3013,
      "eval_samples_per_second": 83.905,
      "eval_steps_per_second": 10.602,
      "step": 16224
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 52.0,
      "step": 16224
    },
    {
      "epoch": 52.88,
      "learning_rate": 0.000592948717948718,
      "loss": 0.2544,
      "step": 16500
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.740072202166065,
      "eval_loss": 0.34021586179733276,
      "eval_runtime": 3.2965,
      "eval_samples_per_second": 84.029,
      "eval_steps_per_second": 10.617,
      "step": 16536
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 53.0,
      "step": 16536
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.7436823104693141,
      "eval_loss": 0.3312588930130005,
      "eval_runtime": 3.2993,
      "eval_samples_per_second": 83.958,
      "eval_steps_per_second": 10.608,
      "step": 16848
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 54.0,
      "step": 16848
    },
    {
      "epoch": 54.49,
      "learning_rate": 0.00045940170940170943,
      "loss": 0.2499,
      "step": 17000
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.740072202166065,
      "eval_loss": 0.33167052268981934,
      "eval_runtime": 3.2972,
      "eval_samples_per_second": 84.01,
      "eval_steps_per_second": 10.615,
      "step": 17160
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 55.0,
      "step": 17160
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.7328519855595668,
      "eval_loss": 0.3464829623699188,
      "eval_runtime": 3.2973,
      "eval_samples_per_second": 84.008,
      "eval_steps_per_second": 10.615,
      "step": 17472
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 56.0,
      "step": 17472
    },
    {
      "epoch": 56.09,
      "learning_rate": 0.00032585470085470087,
      "loss": 0.2505,
      "step": 17500
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.7436823104693141,
      "eval_loss": 0.3397899866104126,
      "eval_runtime": 3.2963,
      "eval_samples_per_second": 84.034,
      "eval_steps_per_second": 10.618,
      "step": 17784
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 57.0,
      "step": 17784
    },
    {
      "epoch": 57.69,
      "learning_rate": 0.00019230769230769233,
      "loss": 0.2468,
      "step": 18000
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.7436823104693141,
      "eval_loss": 0.33796247839927673,
      "eval_runtime": 3.2977,
      "eval_samples_per_second": 83.997,
      "eval_steps_per_second": 10.613,
      "step": 18096
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 58.0,
      "step": 18096
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.7436823104693141,
      "eval_loss": 0.3370416462421417,
      "eval_runtime": 3.3081,
      "eval_samples_per_second": 83.734,
      "eval_steps_per_second": 10.58,
      "step": 18408
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 59.0,
      "step": 18408
    },
    {
      "epoch": 59.29,
      "learning_rate": 5.876068376068376e-05,
      "loss": 0.2487,
      "step": 18500
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.740072202166065,
      "eval_loss": 0.34018510580062866,
      "eval_runtime": 3.3027,
      "eval_samples_per_second": 83.871,
      "eval_steps_per_second": 10.597,
      "step": 18720
    },
    {
      "best_epoch": 40,
      "best_eval_accuracy": 0.7545126353790613,
      "epoch": 60.0,
      "step": 18720
    },
    {
      "epoch": 60.0,
      "step": 18720,
      "total_flos": 6.96152728406016e+16,
      "train_loss": 0.3291880776739528,
      "train_runtime": 3070.3912,
      "train_samples_per_second": 48.658,
      "train_steps_per_second": 6.097
    }
  ],
  "max_steps": 18720,
  "num_train_epochs": 60,
  "total_flos": 6.96152728406016e+16,
  "trial_name": null,
  "trial_params": null
}