{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5242392141022614,
  "global_step": 25000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 1.9951222486433756e-05,
      "loss": 0.0754,
      "step": 200
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.990244497286751e-05,
      "loss": 0.0747,
      "step": 400
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.9853667459301263e-05,
      "loss": 0.0737,
      "step": 600
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9804889945735017e-05,
      "loss": 0.073,
      "step": 800
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.975611243216877e-05,
      "loss": 0.0723,
      "step": 1000
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9707334918602525e-05,
      "loss": 0.072,
      "step": 1200
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.965855740503628e-05,
      "loss": 0.0709,
      "step": 1400
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9609779891470033e-05,
      "loss": 0.0703,
      "step": 1600
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9561002377903787e-05,
      "loss": 0.0708,
      "step": 1800
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.951222486433754e-05,
      "loss": 0.0711,
      "step": 2000
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.9463447350771295e-05,
      "loss": 0.0699,
      "step": 2200
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.941466983720505e-05,
      "loss": 0.0701,
      "step": 2400
    },
    {
      "epoch": 0.15,
      "eval_loss": 0.055753033608198166,
      "eval_runtime": 1008.1775,
      "eval_samples_per_second": 82.199,
      "eval_steps_per_second": 13.7,
      "step": 2500
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.9365892323638806e-05,
      "loss": 0.071,
      "step": 2600
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.931711481007256e-05,
      "loss": 0.0692,
      "step": 2800
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.9268337296506314e-05,
      "loss": 0.0687,
      "step": 3000
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.9219559782940067e-05,
      "loss": 0.0689,
      "step": 3200
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.917078226937382e-05,
      "loss": 0.0681,
      "step": 3400
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.9122004755807575e-05,
      "loss": 0.0678,
      "step": 3600
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.907322724224133e-05,
      "loss": 0.0666,
      "step": 3800
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.9024449728675083e-05,
      "loss": 0.0676,
      "step": 4000
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.8975672215108837e-05,
      "loss": 0.0662,
      "step": 4200
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.892689470154259e-05,
      "loss": 0.0676,
      "step": 4400
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8878117187976345e-05,
      "loss": 0.0663,
      "step": 4600
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.88293396744101e-05,
      "loss": 0.0661,
      "step": 4800
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.8780562160843853e-05,
      "loss": 0.0661,
      "step": 5000
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.05344025790691376,
      "eval_runtime": 1009.8843,
      "eval_samples_per_second": 82.06,
      "eval_steps_per_second": 13.677,
      "step": 5000
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8731784647277606e-05,
      "loss": 0.0674,
      "step": 5200
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.868300713371136e-05,
      "loss": 0.066,
      "step": 5400
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.8634229620145114e-05,
      "loss": 0.0642,
      "step": 5600
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.8585452106578868e-05,
      "loss": 0.0655,
      "step": 5800
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.8536674593012622e-05,
      "loss": 0.0654,
      "step": 6000
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.8487897079446376e-05,
      "loss": 0.0641,
      "step": 6200
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.843911956588013e-05,
      "loss": 0.0636,
      "step": 6400
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.8390342052313884e-05,
      "loss": 0.0658,
      "step": 6600
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.8341564538747638e-05,
      "loss": 0.0639,
      "step": 6800
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.8292787025181395e-05,
      "loss": 0.0642,
      "step": 7000
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.8244009511615145e-05,
      "loss": 0.0648,
      "step": 7200
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.8195231998048903e-05,
      "loss": 0.0645,
      "step": 7400
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.05153830349445343,
      "eval_runtime": 1007.6883,
      "eval_samples_per_second": 82.239,
      "eval_steps_per_second": 13.707,
      "step": 7500
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.8146454484482653e-05,
      "loss": 0.0639,
      "step": 7600
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.809767697091641e-05,
      "loss": 0.0628,
      "step": 7800
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.8048899457350164e-05,
      "loss": 0.0647,
      "step": 8000
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.800012194378392e-05,
      "loss": 0.0629,
      "step": 8200
    },
    {
      "epoch": 0.51,
      "learning_rate": 1.7951344430217672e-05,
      "loss": 0.0638,
      "step": 8400
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.7902566916651426e-05,
      "loss": 0.0635,
      "step": 8600
    },
    {
      "epoch": 0.54,
      "learning_rate": 1.785378940308518e-05,
      "loss": 0.0628,
      "step": 8800
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.7805011889518934e-05,
      "loss": 0.063,
      "step": 9000
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.7756234375952688e-05,
      "loss": 0.0636,
      "step": 9200
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.7707456862386442e-05,
      "loss": 0.0633,
      "step": 9400
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.7658679348820196e-05,
      "loss": 0.0624,
      "step": 9600
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.760990183525395e-05,
      "loss": 0.0615,
      "step": 9800
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.7561124321687703e-05,
      "loss": 0.062,
      "step": 10000
    },
    {
      "epoch": 0.61,
      "eval_loss": 0.05109347030520439,
      "eval_runtime": 1005.611,
      "eval_samples_per_second": 82.409,
      "eval_steps_per_second": 13.735,
      "step": 10000
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.7512346808121457e-05,
      "loss": 0.0616,
      "step": 10200
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.746356929455521e-05,
      "loss": 0.0613,
      "step": 10400
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.7414791780988965e-05,
      "loss": 0.0616,
      "step": 10600
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.736601426742272e-05,
      "loss": 0.0623,
      "step": 10800
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.7317236753856473e-05,
      "loss": 0.062,
      "step": 11000
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.7268459240290227e-05,
      "loss": 0.0632,
      "step": 11200
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.7219681726723984e-05,
      "loss": 0.0623,
      "step": 11400
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.7170904213157735e-05,
      "loss": 0.0608,
      "step": 11600
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.7122126699591492e-05,
      "loss": 0.061,
      "step": 11800
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.7073349186025242e-05,
      "loss": 0.061,
      "step": 12000
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.7024571672459e-05,
      "loss": 0.0605,
      "step": 12200
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.697579415889275e-05,
      "loss": 0.0607,
      "step": 12400
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.04968786612153053,
      "eval_runtime": 1006.7134,
      "eval_samples_per_second": 82.318,
      "eval_steps_per_second": 13.72,
      "step": 12500
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.6927016645326508e-05,
      "loss": 0.0616,
      "step": 12600
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.6878239131760258e-05,
      "loss": 0.0596,
      "step": 12800
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.6829461618194015e-05,
      "loss": 0.0615,
      "step": 13000
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.6780684104627766e-05,
      "loss": 0.0599,
      "step": 13200
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.6731906591061523e-05,
      "loss": 0.0598,
      "step": 13400
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.6683129077495277e-05,
      "loss": 0.0606,
      "step": 13600
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.663435156392903e-05,
      "loss": 0.0596,
      "step": 13800
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.6585574050362785e-05,
      "loss": 0.0585,
      "step": 14000
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.653679653679654e-05,
      "loss": 0.0597,
      "step": 14200
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.6488019023230293e-05,
      "loss": 0.0606,
      "step": 14400
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.6439241509664047e-05,
      "loss": 0.0591,
      "step": 14600
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.63904639960978e-05,
      "loss": 0.06,
      "step": 14800
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.6341686482531554e-05,
      "loss": 0.0586,
      "step": 15000
    },
    {
      "epoch": 0.91,
      "eval_loss": 0.04891153797507286,
      "eval_runtime": 1009.9343,
      "eval_samples_per_second": 82.056,
      "eval_steps_per_second": 13.676,
      "step": 15000
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.6292908968965308e-05,
      "loss": 0.06,
      "step": 15200
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.6244131455399062e-05,
      "loss": 0.0587,
      "step": 15400
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.6195353941832816e-05,
      "loss": 0.0584,
      "step": 15600
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.614657642826657e-05,
      "loss": 0.0602,
      "step": 15800
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.6097798914700324e-05,
      "loss": 0.0587,
      "step": 16000
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.604902140113408e-05,
      "loss": 0.0578,
      "step": 16200
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.600024388756783e-05,
      "loss": 0.0578,
      "step": 16400
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.595146637400159e-05,
      "loss": 0.0598,
      "step": 16600
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.590268886043534e-05,
      "loss": 0.0583,
      "step": 16800
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.5853911346869097e-05,
      "loss": 0.0577,
      "step": 17000
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.5805133833302847e-05,
      "loss": 0.0571,
      "step": 17200
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.5756356319736604e-05,
      "loss": 0.0583,
      "step": 17400
    },
    {
      "epoch": 1.07,
      "eval_loss": 0.048093125224113464,
      "eval_runtime": 1004.9925,
      "eval_samples_per_second": 82.459,
      "eval_steps_per_second": 13.743,
      "step": 17500
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.5707578806170355e-05,
      "loss": 0.0588,
      "step": 17600
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.5658801292604112e-05,
      "loss": 0.0578,
      "step": 17800
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.5610023779037863e-05,
      "loss": 0.0584,
      "step": 18000
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.556124626547162e-05,
      "loss": 0.0569,
      "step": 18200
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.551246875190537e-05,
      "loss": 0.0575,
      "step": 18400
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.5463691238339128e-05,
      "loss": 0.058,
      "step": 18600
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.5414913724772882e-05,
      "loss": 0.0562,
      "step": 18800
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.5366136211206636e-05,
      "loss": 0.0586,
      "step": 19000
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.531735869764039e-05,
      "loss": 0.0576,
      "step": 19200
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.5268581184074143e-05,
      "loss": 0.0578,
      "step": 19400
    },
    {
      "epoch": 1.2,
      "learning_rate": 1.5219803670507897e-05,
      "loss": 0.0571,
      "step": 19600
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.5171026156941651e-05,
      "loss": 0.0564,
      "step": 19800
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.5122248643375405e-05,
      "loss": 0.0578,
      "step": 20000
    },
    {
      "epoch": 1.22,
      "eval_loss": 0.047511521726846695,
      "eval_runtime": 1012.2966,
      "eval_samples_per_second": 81.864,
      "eval_steps_per_second": 13.644,
      "step": 20000
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.5073471129809159e-05,
      "loss": 0.0573,
      "step": 20200
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.5024693616242913e-05,
      "loss": 0.0577,
      "step": 20400
    },
    {
      "epoch": 1.26,
      "learning_rate": 1.4975916102676667e-05,
      "loss": 0.0571,
      "step": 20600
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.492713858911042e-05,
      "loss": 0.0571,
      "step": 20800
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.4878361075544175e-05,
      "loss": 0.0573,
      "step": 21000
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.4829583561977929e-05,
      "loss": 0.0571,
      "step": 21200
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.4780806048411684e-05,
      "loss": 0.0574,
      "step": 21400
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.4732028534845438e-05,
      "loss": 0.0565,
      "step": 21600
    },
    {
      "epoch": 1.33,
      "learning_rate": 1.4683251021279192e-05,
      "loss": 0.0564,
      "step": 21800
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.4634473507712946e-05,
      "loss": 0.0563,
      "step": 22000
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.45856959941467e-05,
      "loss": 0.0564,
      "step": 22200
    },
    {
      "epoch": 1.37,
      "learning_rate": 1.4536918480580454e-05,
      "loss": 0.0557,
      "step": 22400
    },
    {
      "epoch": 1.37,
      "eval_loss": 0.04695958271622658,
      "eval_runtime": 1007.3223,
      "eval_samples_per_second": 82.269,
      "eval_steps_per_second": 13.712,
      "step": 22500
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.4488140967014208e-05,
      "loss": 0.0563,
      "step": 22600
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.4439363453447961e-05,
      "loss": 0.0555,
      "step": 22800
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.4390585939881715e-05,
      "loss": 0.0567,
      "step": 23000
    },
    {
      "epoch": 1.41,
      "learning_rate": 1.434180842631547e-05,
      "loss": 0.0563,
      "step": 23200
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.4293030912749223e-05,
      "loss": 0.0568,
      "step": 23400
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.4244253399182977e-05,
      "loss": 0.0552,
      "step": 23600
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.4195475885616733e-05,
      "loss": 0.0564,
      "step": 23800
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.4146698372050487e-05,
      "loss": 0.0558,
      "step": 24000
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.409792085848424e-05,
      "loss": 0.0553,
      "step": 24200
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.4049143344917994e-05,
      "loss": 0.0545,
      "step": 24400
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.4000365831351748e-05,
      "loss": 0.0551,
      "step": 24600
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.3951588317785502e-05,
      "loss": 0.0556,
      "step": 24800
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.3902810804219256e-05,
      "loss": 0.0543,
      "step": 25000
    },
    {
      "epoch": 1.52,
      "eval_loss": 0.04659077152609825,
      "eval_runtime": 1008.4315,
      "eval_samples_per_second": 82.178,
      "eval_steps_per_second": 13.697,
      "step": 25000
    }
  ],
  "max_steps": 82005,
  "num_train_epochs": 5,
  "total_flos": 1.2689982006126182e+18,
  "trial_name": null,
  "trial_params": null
}