{
  "best_metric": 1.020882487297058,
  "best_model_checkpoint": "/kaggle/output/checkpoint-36000",
  "epoch": 1.4667535853976532,
  "eval_steps": 1000,
  "global_step": 36000,
  "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": 1.0722,
      "step": 1
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.7750000000000004e-08,
      "loss": 1.1407,
      "step": 1000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.32914171656686625,
      "eval_loss": 1.1048731803894043,
      "eval_runtime": 57.3781,
      "eval_samples_per_second": 87.316,
      "eval_steps_per_second": 10.928,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.5527777777777784e-08,
      "loss": 1.1253,
      "step": 2000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.3377245508982036,
      "eval_loss": 1.1033366918563843,
      "eval_runtime": 57.3237,
      "eval_samples_per_second": 87.398,
      "eval_steps_per_second": 10.938,
      "step": 2000
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.327777777777778e-08,
      "loss": 1.1179,
      "step": 3000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.3502994011976048,
      "eval_loss": 1.104222059249878,
      "eval_runtime": 57.4527,
      "eval_samples_per_second": 87.202,
      "eval_steps_per_second": 10.913,
      "step": 3000
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.1105555555555557e-07,
      "loss": 1.1197,
      "step": 4000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.3457085828343313,
      "eval_loss": 1.1000728607177734,
      "eval_runtime": 57.5257,
      "eval_samples_per_second": 87.091,
      "eval_steps_per_second": 10.899,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.3883333333333335e-07,
      "loss": 1.1202,
      "step": 5000
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.3383233532934132,
      "eval_loss": 1.099381923675537,
      "eval_runtime": 57.825,
      "eval_samples_per_second": 86.641,
      "eval_steps_per_second": 10.843,
      "step": 5000
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6661111111111112e-07,
      "loss": 1.1166,
      "step": 6000
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.35409181636726544,
      "eval_loss": 1.0975825786590576,
      "eval_runtime": 57.6509,
      "eval_samples_per_second": 86.902,
      "eval_steps_per_second": 10.876,
      "step": 6000
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.9436111111111112e-07,
      "loss": 1.1126,
      "step": 7000
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.35508982035928144,
      "eval_loss": 1.0957295894622803,
      "eval_runtime": 57.5423,
      "eval_samples_per_second": 87.066,
      "eval_steps_per_second": 10.896,
      "step": 7000
    },
    {
      "epoch": 0.33,
      "learning_rate": 2.2213888888888891e-07,
      "loss": 1.1091,
      "step": 8000
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.36666666666666664,
      "eval_loss": 1.093665599822998,
      "eval_runtime": 57.2594,
      "eval_samples_per_second": 87.497,
      "eval_steps_per_second": 10.95,
      "step": 8000
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.4988888888888893e-07,
      "loss": 1.1069,
      "step": 9000
    },
    {
      "epoch": 0.37,
      "eval_accuracy": 0.35129740518962077,
      "eval_loss": 1.1007317304611206,
      "eval_runtime": 57.5122,
      "eval_samples_per_second": 87.112,
      "eval_steps_per_second": 10.902,
      "step": 9000
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.776666666666667e-07,
      "loss": 1.1054,
      "step": 10000
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.38143712574850297,
      "eval_loss": 1.092604398727417,
      "eval_runtime": 57.5161,
      "eval_samples_per_second": 87.106,
      "eval_steps_per_second": 10.901,
      "step": 10000
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.0541666666666667e-07,
      "loss": 1.1041,
      "step": 11000
    },
    {
      "epoch": 0.45,
      "eval_accuracy": 0.37704590818363276,
      "eval_loss": 1.0923383235931396,
      "eval_runtime": 57.5839,
      "eval_samples_per_second": 87.003,
      "eval_steps_per_second": 10.888,
      "step": 11000
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.3319444444444444e-07,
      "loss": 1.1047,
      "step": 12000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.3914171656686627,
      "eval_loss": 1.0889469385147095,
      "eval_runtime": 57.877,
      "eval_samples_per_second": 86.563,
      "eval_steps_per_second": 10.833,
      "step": 12000
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.6094444444444446e-07,
      "loss": 1.1015,
      "step": 13000
    },
    {
      "epoch": 0.53,
      "eval_accuracy": 0.3437125748502994,
      "eval_loss": 1.095380187034607,
      "eval_runtime": 57.5214,
      "eval_samples_per_second": 87.098,
      "eval_steps_per_second": 10.9,
      "step": 13000
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.8872222222222223e-07,
      "loss": 1.0997,
      "step": 14000
    },
    {
      "epoch": 0.57,
      "eval_accuracy": 0.3596806387225549,
      "eval_loss": 1.096056580543518,
      "eval_runtime": 57.4747,
      "eval_samples_per_second": 87.169,
      "eval_steps_per_second": 10.909,
      "step": 14000
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.1650000000000006e-07,
      "loss": 1.1022,
      "step": 15000
    },
    {
      "epoch": 0.61,
      "eval_accuracy": 0.38882235528942116,
      "eval_loss": 1.089142084121704,
      "eval_runtime": 57.778,
      "eval_samples_per_second": 86.711,
      "eval_steps_per_second": 10.852,
      "step": 15000
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.4427777777777783e-07,
      "loss": 1.0985,
      "step": 16000
    },
    {
      "epoch": 0.65,
      "eval_accuracy": 0.3930139720558882,
      "eval_loss": 1.0844948291778564,
      "eval_runtime": 57.5695,
      "eval_samples_per_second": 87.025,
      "eval_steps_per_second": 10.891,
      "step": 16000
    },
    {
      "epoch": 0.69,
      "learning_rate": 4.7202777777777785e-07,
      "loss": 1.0966,
      "step": 17000
    },
    {
      "epoch": 0.69,
      "eval_accuracy": 0.40459081836327343,
      "eval_loss": 1.083156704902649,
      "eval_runtime": 57.5224,
      "eval_samples_per_second": 87.096,
      "eval_steps_per_second": 10.9,
      "step": 17000
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.998055555555556e-07,
      "loss": 1.0985,
      "step": 18000
    },
    {
      "epoch": 0.73,
      "eval_accuracy": 0.4055888223552894,
      "eval_loss": 1.0858113765716553,
      "eval_runtime": 57.413,
      "eval_samples_per_second": 87.263,
      "eval_steps_per_second": 10.921,
      "step": 18000
    },
    {
      "epoch": 0.77,
      "learning_rate": 5.275277777777778e-07,
      "loss": 1.0933,
      "step": 19000
    },
    {
      "epoch": 0.77,
      "eval_accuracy": 0.3996007984031936,
      "eval_loss": 1.0827423334121704,
      "eval_runtime": 57.2851,
      "eval_samples_per_second": 87.457,
      "eval_steps_per_second": 10.945,
      "step": 19000
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.553055555555556e-07,
      "loss": 1.094,
      "step": 20000
    },
    {
      "epoch": 0.81,
      "eval_accuracy": 0.4059880239520958,
      "eval_loss": 1.0834325551986694,
      "eval_runtime": 57.4875,
      "eval_samples_per_second": 87.149,
      "eval_steps_per_second": 10.907,
      "step": 20000
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.830555555555556e-07,
      "loss": 1.0924,
      "step": 21000
    },
    {
      "epoch": 0.86,
      "eval_accuracy": 0.3850299401197605,
      "eval_loss": 1.0823787450790405,
      "eval_runtime": 57.0913,
      "eval_samples_per_second": 87.754,
      "eval_steps_per_second": 10.982,
      "step": 21000
    },
    {
      "epoch": 0.9,
      "learning_rate": 6.108333333333333e-07,
      "loss": 1.0894,
      "step": 22000
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.4101796407185629,
      "eval_loss": 1.0732710361480713,
      "eval_runtime": 57.1315,
      "eval_samples_per_second": 87.692,
      "eval_steps_per_second": 10.975,
      "step": 22000
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.386111111111112e-07,
      "loss": 1.0812,
      "step": 23000
    },
    {
      "epoch": 0.94,
      "eval_accuracy": 0.43333333333333335,
      "eval_loss": 1.063122034072876,
      "eval_runtime": 57.1173,
      "eval_samples_per_second": 87.714,
      "eval_steps_per_second": 10.977,
      "step": 23000
    },
    {
      "epoch": 0.98,
      "learning_rate": 6.663611111111112e-07,
      "loss": 1.0751,
      "step": 24000
    },
    {
      "epoch": 0.98,
      "eval_accuracy": 0.43972055888223555,
      "eval_loss": 1.055448055267334,
      "eval_runtime": 57.3035,
      "eval_samples_per_second": 87.429,
      "eval_steps_per_second": 10.942,
      "step": 24000
    },
    {
      "epoch": 1.02,
      "learning_rate": 6.94138888888889e-07,
      "loss": 1.0724,
      "step": 25000
    },
    {
      "epoch": 1.02,
      "eval_accuracy": 0.436127744510978,
      "eval_loss": 1.0547080039978027,
      "eval_runtime": 57.4811,
      "eval_samples_per_second": 87.159,
      "eval_steps_per_second": 10.908,
      "step": 25000
    },
    {
      "epoch": 1.06,
      "learning_rate": 7.219166666666666e-07,
      "loss": 1.0682,
      "step": 26000
    },
    {
      "epoch": 1.06,
      "eval_accuracy": 0.4297405189620758,
      "eval_loss": 1.0550919771194458,
      "eval_runtime": 57.4375,
      "eval_samples_per_second": 87.225,
      "eval_steps_per_second": 10.916,
      "step": 26000
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.496666666666667e-07,
      "loss": 1.0675,
      "step": 27000
    },
    {
      "epoch": 1.1,
      "eval_accuracy": 0.43812375249500995,
      "eval_loss": 1.0507142543792725,
      "eval_runtime": 57.1993,
      "eval_samples_per_second": 87.589,
      "eval_steps_per_second": 10.962,
      "step": 27000
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.774444444444445e-07,
      "loss": 1.0642,
      "step": 28000
    },
    {
      "epoch": 1.14,
      "eval_accuracy": 0.4291417165668663,
      "eval_loss": 1.0485388040542603,
      "eval_runtime": 57.1444,
      "eval_samples_per_second": 87.673,
      "eval_steps_per_second": 10.972,
      "step": 28000
    },
    {
      "epoch": 1.18,
      "learning_rate": 8.051944444444445e-07,
      "loss": 1.0593,
      "step": 29000
    },
    {
      "epoch": 1.18,
      "eval_accuracy": 0.4413173652694611,
      "eval_loss": 1.0438047647476196,
      "eval_runtime": 57.4046,
      "eval_samples_per_second": 87.275,
      "eval_steps_per_second": 10.922,
      "step": 29000
    },
    {
      "epoch": 1.22,
      "learning_rate": 8.329722222222223e-07,
      "loss": 1.0601,
      "step": 30000
    },
    {
      "epoch": 1.22,
      "eval_accuracy": 0.43852295409181635,
      "eval_loss": 1.0402766466140747,
      "eval_runtime": 57.3546,
      "eval_samples_per_second": 87.351,
      "eval_steps_per_second": 10.932,
      "step": 30000
    },
    {
      "epoch": 1.26,
      "learning_rate": 8.607222222222223e-07,
      "loss": 1.0634,
      "step": 31000
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.4481037924151697,
      "eval_loss": 1.04641854763031,
      "eval_runtime": 57.3392,
      "eval_samples_per_second": 87.375,
      "eval_steps_per_second": 10.935,
      "step": 31000
    },
    {
      "epoch": 1.3,
      "learning_rate": 8.885e-07,
      "loss": 1.0589,
      "step": 32000
    },
    {
      "epoch": 1.3,
      "eval_accuracy": 0.4303393213572854,
      "eval_loss": 1.0385456085205078,
      "eval_runtime": 57.4244,
      "eval_samples_per_second": 87.245,
      "eval_steps_per_second": 10.919,
      "step": 32000
    },
    {
      "epoch": 1.34,
      "learning_rate": 9.162500000000001e-07,
      "loss": 1.0543,
      "step": 33000
    },
    {
      "epoch": 1.34,
      "eval_accuracy": 0.4536926147704591,
      "eval_loss": 1.033170223236084,
      "eval_runtime": 57.3319,
      "eval_samples_per_second": 87.386,
      "eval_steps_per_second": 10.936,
      "step": 33000
    },
    {
      "epoch": 1.39,
      "learning_rate": 9.440277777777779e-07,
      "loss": 1.0481,
      "step": 34000
    },
    {
      "epoch": 1.39,
      "eval_accuracy": 0.4489021956087824,
      "eval_loss": 1.0322452783584595,
      "eval_runtime": 57.6214,
      "eval_samples_per_second": 86.947,
      "eval_steps_per_second": 10.881,
      "step": 34000
    },
    {
      "epoch": 1.43,
      "learning_rate": 9.71777777777778e-07,
      "loss": 1.0525,
      "step": 35000
    },
    {
      "epoch": 1.43,
      "eval_accuracy": 0.45069860279441115,
      "eval_loss": 1.0243780612945557,
      "eval_runtime": 57.4229,
      "eval_samples_per_second": 87.247,
      "eval_steps_per_second": 10.919,
      "step": 35000
    },
    {
      "epoch": 1.47,
      "learning_rate": 9.995555555555557e-07,
      "loss": 1.05,
      "step": 36000
    },
    {
      "epoch": 1.47,
      "eval_accuracy": 0.4598802395209581,
      "eval_loss": 1.020882487297058,
      "eval_runtime": 57.5219,
      "eval_samples_per_second": 87.097,
      "eval_steps_per_second": 10.9,
      "step": 36000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10000000,
  "num_train_epochs": 408,
  "save_steps": 1000,
  "total_flos": 7.52525233988567e+16,
  "trial_name": null,
  "trial_params": null
}