{
  "best_metric": 1.0100239515304565,
  "best_model_checkpoint": "/kaggle/output/checkpoint-47000",
  "epoch": 1.9556714471968708,
  "eval_steps": 1000,
  "global_step": 48000,
  "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": 1.05,
      "step": 1
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.7750000000000004e-08,
      "loss": 1.1403,
      "step": 1000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.33712574850299404,
      "eval_loss": 1.1030837297439575,
      "eval_runtime": 29.7882,
      "eval_samples_per_second": 168.187,
      "eval_steps_per_second": 21.049,
      "step": 1000
    },
    {
      "epoch": 0.08,
      "learning_rate": 5.5527777777777784e-08,
      "loss": 1.1194,
      "step": 2000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.34331337325349304,
      "eval_loss": 1.1002165079116821,
      "eval_runtime": 29.8685,
      "eval_samples_per_second": 167.735,
      "eval_steps_per_second": 20.992,
      "step": 2000
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.327777777777778e-08,
      "loss": 1.1194,
      "step": 3000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.3471057884231537,
      "eval_loss": 1.101266622543335,
      "eval_runtime": 29.709,
      "eval_samples_per_second": 168.636,
      "eval_steps_per_second": 21.105,
      "step": 3000
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.1105555555555557e-07,
      "loss": 1.1166,
      "step": 4000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.34311377245508984,
      "eval_loss": 1.0984796285629272,
      "eval_runtime": 29.9401,
      "eval_samples_per_second": 167.334,
      "eval_steps_per_second": 20.942,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.3883333333333335e-07,
      "loss": 1.1137,
      "step": 5000
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.3403193612774451,
      "eval_loss": 1.0974279642105103,
      "eval_runtime": 29.8185,
      "eval_samples_per_second": 168.016,
      "eval_steps_per_second": 21.027,
      "step": 5000
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.6658333333333335e-07,
      "loss": 1.1109,
      "step": 6000
    },
    {
      "epoch": 0.24,
      "eval_accuracy": 0.3592814371257485,
      "eval_loss": 1.0984892845153809,
      "eval_runtime": 29.893,
      "eval_samples_per_second": 167.598,
      "eval_steps_per_second": 20.975,
      "step": 6000
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.9436111111111112e-07,
      "loss": 1.1122,
      "step": 7000
    },
    {
      "epoch": 0.29,
      "eval_accuracy": 0.36327345309381237,
      "eval_loss": 1.0960686206817627,
      "eval_runtime": 29.8315,
      "eval_samples_per_second": 167.943,
      "eval_steps_per_second": 21.018,
      "step": 7000
    },
    {
      "epoch": 0.33,
      "learning_rate": 2.2213888888888891e-07,
      "loss": 1.1091,
      "step": 8000
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.35728542914171657,
      "eval_loss": 1.0958523750305176,
      "eval_runtime": 29.9494,
      "eval_samples_per_second": 167.282,
      "eval_steps_per_second": 20.935,
      "step": 8000
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.4988888888888893e-07,
      "loss": 1.1077,
      "step": 9000
    },
    {
      "epoch": 0.37,
      "eval_accuracy": 0.35528942115768464,
      "eval_loss": 1.1003851890563965,
      "eval_runtime": 29.9762,
      "eval_samples_per_second": 167.133,
      "eval_steps_per_second": 20.917,
      "step": 9000
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.776666666666667e-07,
      "loss": 1.1091,
      "step": 10000
    },
    {
      "epoch": 0.41,
      "eval_accuracy": 0.3790419161676647,
      "eval_loss": 1.0951762199401855,
      "eval_runtime": 29.8593,
      "eval_samples_per_second": 167.787,
      "eval_steps_per_second": 20.998,
      "step": 10000
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.0541666666666667e-07,
      "loss": 1.1047,
      "step": 11000
    },
    {
      "epoch": 0.45,
      "eval_accuracy": 0.3780439121756487,
      "eval_loss": 1.0928910970687866,
      "eval_runtime": 29.8768,
      "eval_samples_per_second": 167.689,
      "eval_steps_per_second": 20.986,
      "step": 11000
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.3319444444444444e-07,
      "loss": 1.1053,
      "step": 12000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.38323353293413176,
      "eval_loss": 1.0910215377807617,
      "eval_runtime": 29.8703,
      "eval_samples_per_second": 167.725,
      "eval_steps_per_second": 20.991,
      "step": 12000
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.6094444444444446e-07,
      "loss": 1.1035,
      "step": 13000
    },
    {
      "epoch": 0.53,
      "eval_accuracy": 0.34850299401197604,
      "eval_loss": 1.0930161476135254,
      "eval_runtime": 29.8865,
      "eval_samples_per_second": 167.634,
      "eval_steps_per_second": 20.979,
      "step": 13000
    },
    {
      "epoch": 0.57,
      "learning_rate": 3.8872222222222223e-07,
      "loss": 1.1002,
      "step": 14000
    },
    {
      "epoch": 0.57,
      "eval_accuracy": 0.3646706586826347,
      "eval_loss": 1.093497633934021,
      "eval_runtime": 30.0222,
      "eval_samples_per_second": 166.877,
      "eval_steps_per_second": 20.885,
      "step": 14000
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.1647222222222225e-07,
      "loss": 1.1025,
      "step": 15000
    },
    {
      "epoch": 0.61,
      "eval_accuracy": 0.37924151696606784,
      "eval_loss": 1.0900779962539673,
      "eval_runtime": 30.0049,
      "eval_samples_per_second": 166.973,
      "eval_steps_per_second": 20.897,
      "step": 15000
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.4425e-07,
      "loss": 1.0992,
      "step": 16000
    },
    {
      "epoch": 0.65,
      "eval_accuracy": 0.40119760479041916,
      "eval_loss": 1.0855979919433594,
      "eval_runtime": 30.1095,
      "eval_samples_per_second": 166.393,
      "eval_steps_per_second": 20.824,
      "step": 16000
    },
    {
      "epoch": 0.69,
      "learning_rate": 4.7200000000000004e-07,
      "loss": 1.0965,
      "step": 17000
    },
    {
      "epoch": 0.69,
      "eval_accuracy": 0.4171656686626746,
      "eval_loss": 1.0845381021499634,
      "eval_runtime": 30.0586,
      "eval_samples_per_second": 166.674,
      "eval_steps_per_second": 20.859,
      "step": 17000
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.997777777777779e-07,
      "loss": 1.0962,
      "step": 18000
    },
    {
      "epoch": 0.73,
      "eval_accuracy": 0.4087824351297405,
      "eval_loss": 1.0860552787780762,
      "eval_runtime": 30.2343,
      "eval_samples_per_second": 165.706,
      "eval_steps_per_second": 20.738,
      "step": 18000
    },
    {
      "epoch": 0.77,
      "learning_rate": 5.275277777777778e-07,
      "loss": 1.0936,
      "step": 19000
    },
    {
      "epoch": 0.77,
      "eval_accuracy": 0.4039920159680639,
      "eval_loss": 1.0822259187698364,
      "eval_runtime": 30.216,
      "eval_samples_per_second": 165.806,
      "eval_steps_per_second": 20.751,
      "step": 19000
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.553055555555556e-07,
      "loss": 1.0923,
      "step": 20000
    },
    {
      "epoch": 0.81,
      "eval_accuracy": 0.4095808383233533,
      "eval_loss": 1.0820951461791992,
      "eval_runtime": 29.968,
      "eval_samples_per_second": 167.178,
      "eval_steps_per_second": 20.922,
      "step": 20000
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.830555555555556e-07,
      "loss": 1.0897,
      "step": 21000
    },
    {
      "epoch": 0.86,
      "eval_accuracy": 0.4149700598802395,
      "eval_loss": 1.0763800144195557,
      "eval_runtime": 30.021,
      "eval_samples_per_second": 166.883,
      "eval_steps_per_second": 20.885,
      "step": 21000
    },
    {
      "epoch": 0.9,
      "learning_rate": 6.108333333333333e-07,
      "loss": 1.0888,
      "step": 22000
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.43193612774451096,
      "eval_loss": 1.0687414407730103,
      "eval_runtime": 30.1577,
      "eval_samples_per_second": 166.127,
      "eval_steps_per_second": 20.791,
      "step": 22000
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.385833333333334e-07,
      "loss": 1.0823,
      "step": 23000
    },
    {
      "epoch": 0.94,
      "eval_accuracy": 0.4285429141716567,
      "eval_loss": 1.0642070770263672,
      "eval_runtime": 30.0701,
      "eval_samples_per_second": 166.611,
      "eval_steps_per_second": 20.851,
      "step": 23000
    },
    {
      "epoch": 0.98,
      "learning_rate": 6.663611111111112e-07,
      "loss": 1.0786,
      "step": 24000
    },
    {
      "epoch": 0.98,
      "eval_accuracy": 0.4363273453093812,
      "eval_loss": 1.0598841905593872,
      "eval_runtime": 30.0689,
      "eval_samples_per_second": 166.617,
      "eval_steps_per_second": 20.852,
      "step": 24000
    },
    {
      "epoch": 1.02,
      "learning_rate": 6.941111111111112e-07,
      "loss": 1.0708,
      "step": 25000
    },
    {
      "epoch": 1.02,
      "eval_accuracy": 0.4429141716566866,
      "eval_loss": 1.0561795234680176,
      "eval_runtime": 30.1301,
      "eval_samples_per_second": 166.279,
      "eval_steps_per_second": 20.81,
      "step": 25000
    },
    {
      "epoch": 1.06,
      "learning_rate": 7.218888888888889e-07,
      "loss": 1.072,
      "step": 26000
    },
    {
      "epoch": 1.06,
      "eval_accuracy": 0.4303393213572854,
      "eval_loss": 1.0574887990951538,
      "eval_runtime": 30.0473,
      "eval_samples_per_second": 166.737,
      "eval_steps_per_second": 20.867,
      "step": 26000
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.496666666666667e-07,
      "loss": 1.0701,
      "step": 27000
    },
    {
      "epoch": 1.1,
      "eval_accuracy": 0.4421157684630739,
      "eval_loss": 1.055459976196289,
      "eval_runtime": 30.1049,
      "eval_samples_per_second": 166.418,
      "eval_steps_per_second": 20.827,
      "step": 27000
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.774166666666668e-07,
      "loss": 1.0677,
      "step": 28000
    },
    {
      "epoch": 1.14,
      "eval_accuracy": 0.4285429141716567,
      "eval_loss": 1.0527067184448242,
      "eval_runtime": 30.1476,
      "eval_samples_per_second": 166.182,
      "eval_steps_per_second": 20.798,
      "step": 28000
    },
    {
      "epoch": 1.18,
      "learning_rate": 8.051944444444445e-07,
      "loss": 1.065,
      "step": 29000
    },
    {
      "epoch": 1.18,
      "eval_accuracy": 0.43852295409181635,
      "eval_loss": 1.0494154691696167,
      "eval_runtime": 30.036,
      "eval_samples_per_second": 166.8,
      "eval_steps_per_second": 20.875,
      "step": 29000
    },
    {
      "epoch": 1.22,
      "learning_rate": 8.329444444444445e-07,
      "loss": 1.0628,
      "step": 30000
    },
    {
      "epoch": 1.22,
      "eval_accuracy": 0.4377245508982036,
      "eval_loss": 1.0461777448654175,
      "eval_runtime": 30.1005,
      "eval_samples_per_second": 166.442,
      "eval_steps_per_second": 20.83,
      "step": 30000
    },
    {
      "epoch": 1.26,
      "learning_rate": 8.607222222222223e-07,
      "loss": 1.0672,
      "step": 31000
    },
    {
      "epoch": 1.26,
      "eval_accuracy": 0.4435129740518962,
      "eval_loss": 1.0531694889068604,
      "eval_runtime": 30.1733,
      "eval_samples_per_second": 166.041,
      "eval_steps_per_second": 20.78,
      "step": 31000
    },
    {
      "epoch": 1.3,
      "learning_rate": 8.884722222222224e-07,
      "loss": 1.0638,
      "step": 32000
    },
    {
      "epoch": 1.3,
      "eval_accuracy": 0.4297405189620758,
      "eval_loss": 1.0501775741577148,
      "eval_runtime": 30.058,
      "eval_samples_per_second": 166.678,
      "eval_steps_per_second": 20.86,
      "step": 32000
    },
    {
      "epoch": 1.34,
      "learning_rate": 9.162500000000001e-07,
      "loss": 1.0637,
      "step": 33000
    },
    {
      "epoch": 1.34,
      "eval_accuracy": 0.44491017964071855,
      "eval_loss": 1.0452704429626465,
      "eval_runtime": 30.3623,
      "eval_samples_per_second": 165.007,
      "eval_steps_per_second": 20.651,
      "step": 33000
    },
    {
      "epoch": 1.39,
      "learning_rate": 9.440000000000001e-07,
      "loss": 1.0558,
      "step": 34000
    },
    {
      "epoch": 1.39,
      "eval_accuracy": 0.44510978043912175,
      "eval_loss": 1.0434744358062744,
      "eval_runtime": 30.2239,
      "eval_samples_per_second": 165.763,
      "eval_steps_per_second": 20.745,
      "step": 34000
    },
    {
      "epoch": 1.43,
      "learning_rate": 9.71777777777778e-07,
      "loss": 1.0598,
      "step": 35000
    },
    {
      "epoch": 1.43,
      "eval_accuracy": 0.4473053892215569,
      "eval_loss": 1.039801001548767,
      "eval_runtime": 30.107,
      "eval_samples_per_second": 166.407,
      "eval_steps_per_second": 20.826,
      "step": 35000
    },
    {
      "epoch": 1.47,
      "learning_rate": 9.995277777777778e-07,
      "loss": 1.0601,
      "step": 36000
    },
    {
      "epoch": 1.47,
      "eval_accuracy": 0.4469061876247505,
      "eval_loss": 1.037074327468872,
      "eval_runtime": 30.1954,
      "eval_samples_per_second": 165.919,
      "eval_steps_per_second": 20.765,
      "step": 36000
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.0273055555555556e-06,
      "loss": 1.0529,
      "step": 37000
    },
    {
      "epoch": 1.51,
      "eval_accuracy": 0.45229540918163674,
      "eval_loss": 1.0346434116363525,
      "eval_runtime": 30.2856,
      "eval_samples_per_second": 165.425,
      "eval_steps_per_second": 20.703,
      "step": 37000
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.0550555555555557e-06,
      "loss": 1.0544,
      "step": 38000
    },
    {
      "epoch": 1.55,
      "eval_accuracy": 0.45129740518962075,
      "eval_loss": 1.0338674783706665,
      "eval_runtime": 30.1511,
      "eval_samples_per_second": 166.163,
      "eval_steps_per_second": 20.795,
      "step": 38000
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.0828333333333334e-06,
      "loss": 1.0528,
      "step": 39000
    },
    {
      "epoch": 1.59,
      "eval_accuracy": 0.44550898203592815,
      "eval_loss": 1.0376836061477661,
      "eval_runtime": 30.304,
      "eval_samples_per_second": 165.324,
      "eval_steps_per_second": 20.69,
      "step": 39000
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.1106111111111112e-06,
      "loss": 1.0482,
      "step": 40000
    },
    {
      "epoch": 1.63,
      "eval_accuracy": 0.4540918163672655,
      "eval_loss": 1.031366229057312,
      "eval_runtime": 30.2062,
      "eval_samples_per_second": 165.86,
      "eval_steps_per_second": 20.757,
      "step": 40000
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.1383611111111113e-06,
      "loss": 1.0494,
      "step": 41000
    },
    {
      "epoch": 1.67,
      "eval_accuracy": 0.4552894211576846,
      "eval_loss": 1.032992959022522,
      "eval_runtime": 30.1667,
      "eval_samples_per_second": 166.077,
      "eval_steps_per_second": 20.784,
      "step": 41000
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.166138888888889e-06,
      "loss": 1.0474,
      "step": 42000
    },
    {
      "epoch": 1.71,
      "eval_accuracy": 0.45129740518962075,
      "eval_loss": 1.0255250930786133,
      "eval_runtime": 30.2122,
      "eval_samples_per_second": 165.827,
      "eval_steps_per_second": 20.753,
      "step": 42000
    },
    {
      "epoch": 1.75,
      "learning_rate": 1.193888888888889e-06,
      "loss": 1.0448,
      "step": 43000
    },
    {
      "epoch": 1.75,
      "eval_accuracy": 0.4483033932135729,
      "eval_loss": 1.0412585735321045,
      "eval_runtime": 30.2083,
      "eval_samples_per_second": 165.849,
      "eval_steps_per_second": 20.756,
      "step": 43000
    },
    {
      "epoch": 1.79,
      "learning_rate": 1.2216666666666667e-06,
      "loss": 1.0458,
      "step": 44000
    },
    {
      "epoch": 1.79,
      "eval_accuracy": 0.46427145708582834,
      "eval_loss": 1.0211690664291382,
      "eval_runtime": 30.4051,
      "eval_samples_per_second": 164.775,
      "eval_steps_per_second": 20.622,
      "step": 44000
    },
    {
      "epoch": 1.83,
      "learning_rate": 1.2494166666666668e-06,
      "loss": 1.0468,
      "step": 45000
    },
    {
      "epoch": 1.83,
      "eval_accuracy": 0.47105788423153694,
      "eval_loss": 1.0185887813568115,
      "eval_runtime": 30.1465,
      "eval_samples_per_second": 166.188,
      "eval_steps_per_second": 20.798,
      "step": 45000
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.2771944444444445e-06,
      "loss": 1.0414,
      "step": 46000
    },
    {
      "epoch": 1.87,
      "eval_accuracy": 0.46966067864271455,
      "eval_loss": 1.0215678215026855,
      "eval_runtime": 30.1615,
      "eval_samples_per_second": 166.106,
      "eval_steps_per_second": 20.788,
      "step": 46000
    },
    {
      "epoch": 1.91,
      "learning_rate": 1.3049444444444446e-06,
      "loss": 1.0374,
      "step": 47000
    },
    {
      "epoch": 1.91,
      "eval_accuracy": 0.47684630738522954,
      "eval_loss": 1.0100239515304565,
      "eval_runtime": 30.0919,
      "eval_samples_per_second": 166.49,
      "eval_steps_per_second": 20.836,
      "step": 47000
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.3327222222222224e-06,
      "loss": 1.0423,
      "step": 48000
    },
    {
      "epoch": 1.96,
      "eval_accuracy": 0.462874251497006,
      "eval_loss": 1.0277544260025024,
      "eval_runtime": 30.0926,
      "eval_samples_per_second": 166.486,
      "eval_steps_per_second": 20.836,
      "step": 48000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10000000,
  "num_train_epochs": 408,
  "save_steps": 1000,
  "total_flos": 5.016975143259341e+16,
  "trial_name": null,
  "trial_params": null
}