{
  "best_metric": 2.785961627960205,
  "best_model_checkpoint": "experiments/qg/google/mt5-large_all/checkpoint-126000",
  "epoch": 1.9977723341340248,
  "global_step": 126000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3e-05,
      "loss": 8.8865,
      "step": 500
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.9976277372262774e-05,
      "loss": 4.0325,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_loss": 3.3621604442596436,
      "eval_runtime": 444.6073,
      "eval_samples_per_second": 89.967,
      "eval_steps_per_second": 11.246,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.9952554744525548e-05,
      "loss": 3.6162,
      "step": 1500
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.992883211678832e-05,
      "loss": 3.4961,
      "step": 2000
    },
    {
      "epoch": 0.03,
      "eval_loss": 3.1849684715270996,
      "eval_runtime": 443.8466,
      "eval_samples_per_second": 90.121,
      "eval_steps_per_second": 11.265,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.9905109489051095e-05,
      "loss": 3.4531,
      "step": 2500
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.988138686131387e-05,
      "loss": 3.3883,
      "step": 3000
    },
    {
      "epoch": 0.05,
      "eval_loss": 3.0837316513061523,
      "eval_runtime": 451.6674,
      "eval_samples_per_second": 88.561,
      "eval_steps_per_second": 11.07,
      "step": 3000
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.9857664233576645e-05,
      "loss": 3.3483,
      "step": 3500
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.9833941605839416e-05,
      "loss": 3.281,
      "step": 4000
    },
    {
      "epoch": 0.06,
      "eval_loss": 3.053595781326294,
      "eval_runtime": 451.6413,
      "eval_samples_per_second": 88.566,
      "eval_steps_per_second": 11.071,
      "step": 4000
    },
    {
      "epoch": 0.07,
      "learning_rate": 2.9810218978102192e-05,
      "loss": 3.2901,
      "step": 4500
    },
    {
      "epoch": 0.08,
      "learning_rate": 2.9786496350364963e-05,
      "loss": 3.2764,
      "step": 5000
    },
    {
      "epoch": 0.08,
      "eval_loss": 3.045032262802124,
      "eval_runtime": 450.7805,
      "eval_samples_per_second": 88.735,
      "eval_steps_per_second": 11.092,
      "step": 5000
    },
    {
      "epoch": 0.09,
      "learning_rate": 2.9762773722627736e-05,
      "loss": 3.255,
      "step": 5500
    },
    {
      "epoch": 0.1,
      "learning_rate": 2.9739051094890513e-05,
      "loss": 3.2352,
      "step": 6000
    },
    {
      "epoch": 0.1,
      "eval_loss": 3.018794059753418,
      "eval_runtime": 451.4803,
      "eval_samples_per_second": 88.597,
      "eval_steps_per_second": 11.075,
      "step": 6000
    },
    {
      "epoch": 0.1,
      "learning_rate": 2.9715328467153283e-05,
      "loss": 3.1954,
      "step": 6500
    },
    {
      "epoch": 0.11,
      "learning_rate": 2.969160583941606e-05,
      "loss": 3.1939,
      "step": 7000
    },
    {
      "epoch": 0.11,
      "eval_loss": 3.000720500946045,
      "eval_runtime": 450.9264,
      "eval_samples_per_second": 88.706,
      "eval_steps_per_second": 11.088,
      "step": 7000
    },
    {
      "epoch": 0.12,
      "learning_rate": 2.966788321167883e-05,
      "loss": 3.1893,
      "step": 7500
    },
    {
      "epoch": 0.13,
      "learning_rate": 2.9644160583941607e-05,
      "loss": 3.1936,
      "step": 8000
    },
    {
      "epoch": 0.13,
      "eval_loss": 2.988145589828491,
      "eval_runtime": 451.5214,
      "eval_samples_per_second": 88.589,
      "eval_steps_per_second": 11.074,
      "step": 8000
    },
    {
      "epoch": 0.13,
      "learning_rate": 2.962043795620438e-05,
      "loss": 3.1604,
      "step": 8500
    },
    {
      "epoch": 0.14,
      "learning_rate": 2.9596715328467154e-05,
      "loss": 3.1765,
      "step": 9000
    },
    {
      "epoch": 0.14,
      "eval_loss": 2.9744112491607666,
      "eval_runtime": 449.7283,
      "eval_samples_per_second": 88.943,
      "eval_steps_per_second": 11.118,
      "step": 9000
    },
    {
      "epoch": 0.15,
      "learning_rate": 2.9572992700729928e-05,
      "loss": 3.1478,
      "step": 9500
    },
    {
      "epoch": 0.16,
      "learning_rate": 2.95492700729927e-05,
      "loss": 3.1532,
      "step": 10000
    },
    {
      "epoch": 0.16,
      "eval_loss": 2.960232734680176,
      "eval_runtime": 450.2869,
      "eval_samples_per_second": 88.832,
      "eval_steps_per_second": 11.104,
      "step": 10000
    },
    {
      "epoch": 0.17,
      "learning_rate": 2.9525547445255475e-05,
      "loss": 3.1187,
      "step": 10500
    },
    {
      "epoch": 0.17,
      "learning_rate": 2.950182481751825e-05,
      "loss": 3.1318,
      "step": 11000
    },
    {
      "epoch": 0.17,
      "eval_loss": 2.956113338470459,
      "eval_runtime": 450.2104,
      "eval_samples_per_second": 88.847,
      "eval_steps_per_second": 11.106,
      "step": 11000
    },
    {
      "epoch": 0.18,
      "learning_rate": 2.9478102189781022e-05,
      "loss": 3.1086,
      "step": 11500
    },
    {
      "epoch": 0.19,
      "learning_rate": 2.9454379562043796e-05,
      "loss": 3.0946,
      "step": 12000
    },
    {
      "epoch": 0.19,
      "eval_loss": 2.946828842163086,
      "eval_runtime": 450.0294,
      "eval_samples_per_second": 88.883,
      "eval_steps_per_second": 11.11,
      "step": 12000
    },
    {
      "epoch": 0.2,
      "learning_rate": 2.9430656934306573e-05,
      "loss": 3.114,
      "step": 12500
    },
    {
      "epoch": 0.21,
      "learning_rate": 2.9406934306569343e-05,
      "loss": 3.0903,
      "step": 13000
    },
    {
      "epoch": 0.21,
      "eval_loss": 2.936713457107544,
      "eval_runtime": 449.1132,
      "eval_samples_per_second": 89.064,
      "eval_steps_per_second": 11.133,
      "step": 13000
    },
    {
      "epoch": 0.21,
      "learning_rate": 2.938321167883212e-05,
      "loss": 3.102,
      "step": 13500
    },
    {
      "epoch": 0.22,
      "learning_rate": 2.935948905109489e-05,
      "loss": 3.0916,
      "step": 14000
    },
    {
      "epoch": 0.22,
      "eval_loss": 2.9305248260498047,
      "eval_runtime": 449.5477,
      "eval_samples_per_second": 88.978,
      "eval_steps_per_second": 11.122,
      "step": 14000
    },
    {
      "epoch": 0.23,
      "learning_rate": 2.9335766423357667e-05,
      "loss": 3.0578,
      "step": 14500
    },
    {
      "epoch": 0.24,
      "learning_rate": 2.931204379562044e-05,
      "loss": 3.0754,
      "step": 15000
    },
    {
      "epoch": 0.24,
      "eval_loss": 2.92580246925354,
      "eval_runtime": 449.9439,
      "eval_samples_per_second": 88.9,
      "eval_steps_per_second": 11.112,
      "step": 15000
    },
    {
      "epoch": 0.25,
      "learning_rate": 2.928832116788321e-05,
      "loss": 3.0673,
      "step": 15500
    },
    {
      "epoch": 0.25,
      "learning_rate": 2.9264598540145987e-05,
      "loss": 3.0606,
      "step": 16000
    },
    {
      "epoch": 0.25,
      "eval_loss": 2.9198672771453857,
      "eval_runtime": 450.1912,
      "eval_samples_per_second": 88.851,
      "eval_steps_per_second": 11.106,
      "step": 16000
    },
    {
      "epoch": 0.26,
      "learning_rate": 2.9240875912408757e-05,
      "loss": 3.0472,
      "step": 16500
    },
    {
      "epoch": 0.27,
      "learning_rate": 2.9217153284671534e-05,
      "loss": 3.0598,
      "step": 17000
    },
    {
      "epoch": 0.27,
      "eval_loss": 2.91711163520813,
      "eval_runtime": 449.4944,
      "eval_samples_per_second": 88.989,
      "eval_steps_per_second": 11.124,
      "step": 17000
    },
    {
      "epoch": 0.28,
      "learning_rate": 2.9193430656934308e-05,
      "loss": 3.0376,
      "step": 17500
    },
    {
      "epoch": 0.29,
      "learning_rate": 2.916970802919708e-05,
      "loss": 3.0369,
      "step": 18000
    },
    {
      "epoch": 0.29,
      "eval_loss": 2.9090449810028076,
      "eval_runtime": 450.1639,
      "eval_samples_per_second": 88.857,
      "eval_steps_per_second": 11.107,
      "step": 18000
    },
    {
      "epoch": 0.29,
      "learning_rate": 2.9145985401459855e-05,
      "loss": 3.0517,
      "step": 18500
    },
    {
      "epoch": 0.3,
      "learning_rate": 2.912226277372263e-05,
      "loss": 3.0312,
      "step": 19000
    },
    {
      "epoch": 0.3,
      "eval_loss": 2.903899669647217,
      "eval_runtime": 450.5358,
      "eval_samples_per_second": 88.783,
      "eval_steps_per_second": 11.098,
      "step": 19000
    },
    {
      "epoch": 0.31,
      "learning_rate": 2.9098540145985402e-05,
      "loss": 3.0135,
      "step": 19500
    },
    {
      "epoch": 0.32,
      "learning_rate": 2.9074817518248176e-05,
      "loss": 3.0304,
      "step": 20000
    },
    {
      "epoch": 0.32,
      "eval_loss": 2.8915276527404785,
      "eval_runtime": 449.8834,
      "eval_samples_per_second": 88.912,
      "eval_steps_per_second": 11.114,
      "step": 20000
    },
    {
      "epoch": 0.33,
      "learning_rate": 2.905109489051095e-05,
      "loss": 3.0157,
      "step": 20500
    },
    {
      "epoch": 0.33,
      "learning_rate": 2.9027372262773723e-05,
      "loss": 3.008,
      "step": 21000
    },
    {
      "epoch": 0.33,
      "eval_loss": 2.89363431930542,
      "eval_runtime": 450.0699,
      "eval_samples_per_second": 88.875,
      "eval_steps_per_second": 11.109,
      "step": 21000
    },
    {
      "epoch": 0.34,
      "learning_rate": 2.9003649635036496e-05,
      "loss": 3.0125,
      "step": 21500
    },
    {
      "epoch": 0.35,
      "learning_rate": 2.897992700729927e-05,
      "loss": 3.0373,
      "step": 22000
    },
    {
      "epoch": 0.35,
      "eval_loss": 2.8869340419769287,
      "eval_runtime": 450.8527,
      "eval_samples_per_second": 88.721,
      "eval_steps_per_second": 11.09,
      "step": 22000
    },
    {
      "epoch": 0.36,
      "learning_rate": 2.8956204379562047e-05,
      "loss": 3.0246,
      "step": 22500
    },
    {
      "epoch": 0.36,
      "learning_rate": 2.8932481751824817e-05,
      "loss": 3.0107,
      "step": 23000
    },
    {
      "epoch": 0.36,
      "eval_loss": 2.882735252380371,
      "eval_runtime": 451.0618,
      "eval_samples_per_second": 88.68,
      "eval_steps_per_second": 11.085,
      "step": 23000
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.8908759124087594e-05,
      "loss": 3.0006,
      "step": 23500
    },
    {
      "epoch": 0.38,
      "learning_rate": 2.8885036496350364e-05,
      "loss": 2.9781,
      "step": 24000
    },
    {
      "epoch": 0.38,
      "eval_loss": 2.8770592212677,
      "eval_runtime": 450.672,
      "eval_samples_per_second": 88.756,
      "eval_steps_per_second": 11.095,
      "step": 24000
    },
    {
      "epoch": 0.39,
      "learning_rate": 2.886131386861314e-05,
      "loss": 2.9744,
      "step": 24500
    },
    {
      "epoch": 0.4,
      "learning_rate": 2.8837591240875914e-05,
      "loss": 3.0146,
      "step": 25000
    },
    {
      "epoch": 0.4,
      "eval_loss": 2.8825135231018066,
      "eval_runtime": 450.4517,
      "eval_samples_per_second": 88.8,
      "eval_steps_per_second": 11.1,
      "step": 25000
    },
    {
      "epoch": 0.4,
      "learning_rate": 2.8813868613138685e-05,
      "loss": 2.997,
      "step": 25500
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.879014598540146e-05,
      "loss": 2.9856,
      "step": 26000
    },
    {
      "epoch": 0.41,
      "eval_loss": 2.8855531215667725,
      "eval_runtime": 450.7359,
      "eval_samples_per_second": 88.744,
      "eval_steps_per_second": 11.093,
      "step": 26000
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.8766423357664232e-05,
      "loss": 2.986,
      "step": 26500
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.874270072992701e-05,
      "loss": 3.0001,
      "step": 27000
    },
    {
      "epoch": 0.43,
      "eval_loss": 2.886821985244751,
      "eval_runtime": 450.0432,
      "eval_samples_per_second": 88.88,
      "eval_steps_per_second": 11.11,
      "step": 27000
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.8718978102189782e-05,
      "loss": 2.9924,
      "step": 27500
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.8695255474452556e-05,
      "loss": 2.9633,
      "step": 28000
    },
    {
      "epoch": 0.44,
      "eval_loss": 2.8713860511779785,
      "eval_runtime": 450.4908,
      "eval_samples_per_second": 88.792,
      "eval_steps_per_second": 11.099,
      "step": 28000
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.867153284671533e-05,
      "loss": 2.9684,
      "step": 28500
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.8647810218978103e-05,
      "loss": 2.9563,
      "step": 29000
    },
    {
      "epoch": 0.46,
      "eval_loss": 2.8724424839019775,
      "eval_runtime": 450.9726,
      "eval_samples_per_second": 88.697,
      "eval_steps_per_second": 11.087,
      "step": 29000
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.8624087591240876e-05,
      "loss": 2.9573,
      "step": 29500
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.860036496350365e-05,
      "loss": 2.9812,
      "step": 30000
    },
    {
      "epoch": 0.48,
      "eval_loss": 2.86970591545105,
      "eval_runtime": 450.3055,
      "eval_samples_per_second": 88.829,
      "eval_steps_per_second": 11.104,
      "step": 30000
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.8576642335766423e-05,
      "loss": 2.9677,
      "step": 30500
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.8552919708029197e-05,
      "loss": 2.9773,
      "step": 31000
    },
    {
      "epoch": 0.49,
      "eval_loss": 2.8727774620056152,
      "eval_runtime": 449.455,
      "eval_samples_per_second": 88.997,
      "eval_steps_per_second": 11.125,
      "step": 31000
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.8529197080291974e-05,
      "loss": 2.971,
      "step": 31500
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.8505474452554744e-05,
      "loss": 2.9743,
      "step": 32000
    },
    {
      "epoch": 0.51,
      "eval_loss": 2.8665173053741455,
      "eval_runtime": 449.7846,
      "eval_samples_per_second": 88.931,
      "eval_steps_per_second": 11.116,
      "step": 32000
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.848175182481752e-05,
      "loss": 2.9585,
      "step": 32500
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.845802919708029e-05,
      "loss": 2.9428,
      "step": 33000
    },
    {
      "epoch": 0.52,
      "eval_loss": 2.8613767623901367,
      "eval_runtime": 450.8331,
      "eval_samples_per_second": 88.725,
      "eval_steps_per_second": 11.091,
      "step": 33000
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.8434306569343068e-05,
      "loss": 2.9509,
      "step": 33500
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.841058394160584e-05,
      "loss": 2.9683,
      "step": 34000
    },
    {
      "epoch": 0.54,
      "eval_loss": 2.86824369430542,
      "eval_runtime": 450.4958,
      "eval_samples_per_second": 88.791,
      "eval_steps_per_second": 11.099,
      "step": 34000
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.8386861313868612e-05,
      "loss": 2.9332,
      "step": 34500
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.836313868613139e-05,
      "loss": 2.9567,
      "step": 35000
    },
    {
      "epoch": 0.55,
      "eval_loss": 2.854598045349121,
      "eval_runtime": 450.2268,
      "eval_samples_per_second": 88.844,
      "eval_steps_per_second": 11.106,
      "step": 35000
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.833941605839416e-05,
      "loss": 2.9529,
      "step": 35500
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.8315693430656936e-05,
      "loss": 2.9515,
      "step": 36000
    },
    {
      "epoch": 0.57,
      "eval_loss": 2.8551025390625,
      "eval_runtime": 450.4445,
      "eval_samples_per_second": 88.801,
      "eval_steps_per_second": 11.1,
      "step": 36000
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.829197080291971e-05,
      "loss": 2.9273,
      "step": 36500
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.8268248175182483e-05,
      "loss": 2.965,
      "step": 37000
    },
    {
      "epoch": 0.59,
      "eval_loss": 2.8546838760375977,
      "eval_runtime": 450.6163,
      "eval_samples_per_second": 88.767,
      "eval_steps_per_second": 11.096,
      "step": 37000
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.8244525547445256e-05,
      "loss": 2.9441,
      "step": 37500
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.822080291970803e-05,
      "loss": 2.9326,
      "step": 38000
    },
    {
      "epoch": 0.6,
      "eval_loss": 2.848287582397461,
      "eval_runtime": 450.6373,
      "eval_samples_per_second": 88.763,
      "eval_steps_per_second": 11.095,
      "step": 38000
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.8197080291970803e-05,
      "loss": 2.9547,
      "step": 38500
    },
    {
      "epoch": 0.62,
      "learning_rate": 2.817335766423358e-05,
      "loss": 2.9096,
      "step": 39000
    },
    {
      "epoch": 0.62,
      "eval_loss": 2.8502256870269775,
      "eval_runtime": 450.3731,
      "eval_samples_per_second": 88.815,
      "eval_steps_per_second": 11.102,
      "step": 39000
    },
    {
      "epoch": 0.63,
      "learning_rate": 2.814963503649635e-05,
      "loss": 2.9222,
      "step": 39500
    },
    {
      "epoch": 0.63,
      "learning_rate": 2.8125912408759124e-05,
      "loss": 2.9132,
      "step": 40000
    },
    {
      "epoch": 0.63,
      "eval_loss": 2.8453314304351807,
      "eval_runtime": 450.4787,
      "eval_samples_per_second": 88.794,
      "eval_steps_per_second": 11.099,
      "step": 40000
    },
    {
      "epoch": 0.64,
      "learning_rate": 2.8102189781021898e-05,
      "loss": 2.9172,
      "step": 40500
    },
    {
      "epoch": 0.65,
      "learning_rate": 2.807846715328467e-05,
      "loss": 2.9373,
      "step": 41000
    },
    {
      "epoch": 0.65,
      "eval_loss": 2.8487000465393066,
      "eval_runtime": 450.8104,
      "eval_samples_per_second": 88.729,
      "eval_steps_per_second": 11.091,
      "step": 41000
    },
    {
      "epoch": 0.66,
      "learning_rate": 2.8054744525547448e-05,
      "loss": 2.9451,
      "step": 41500
    },
    {
      "epoch": 0.67,
      "learning_rate": 2.8031021897810218e-05,
      "loss": 2.9333,
      "step": 42000
    },
    {
      "epoch": 0.67,
      "eval_loss": 2.8426833152770996,
      "eval_runtime": 449.7638,
      "eval_samples_per_second": 88.936,
      "eval_steps_per_second": 11.117,
      "step": 42000
    },
    {
      "epoch": 0.67,
      "learning_rate": 2.8007299270072995e-05,
      "loss": 2.9263,
      "step": 42500
    },
    {
      "epoch": 0.68,
      "learning_rate": 2.7983576642335765e-05,
      "loss": 2.9242,
      "step": 43000
    },
    {
      "epoch": 0.68,
      "eval_loss": 2.8440945148468018,
      "eval_runtime": 450.8592,
      "eval_samples_per_second": 88.719,
      "eval_steps_per_second": 11.09,
      "step": 43000
    },
    {
      "epoch": 0.69,
      "learning_rate": 2.795985401459854e-05,
      "loss": 2.9226,
      "step": 43500
    },
    {
      "epoch": 0.7,
      "learning_rate": 2.7936131386861316e-05,
      "loss": 2.9058,
      "step": 44000
    },
    {
      "epoch": 0.7,
      "eval_loss": 2.8383543491363525,
      "eval_runtime": 450.5891,
      "eval_samples_per_second": 88.773,
      "eval_steps_per_second": 11.097,
      "step": 44000
    },
    {
      "epoch": 0.71,
      "learning_rate": 2.7912408759124086e-05,
      "loss": 2.9113,
      "step": 44500
    },
    {
      "epoch": 0.71,
      "learning_rate": 2.7888686131386863e-05,
      "loss": 2.9237,
      "step": 45000
    },
    {
      "epoch": 0.71,
      "eval_loss": 2.8444628715515137,
      "eval_runtime": 450.7048,
      "eval_samples_per_second": 88.75,
      "eval_steps_per_second": 11.094,
      "step": 45000
    },
    {
      "epoch": 0.72,
      "learning_rate": 2.7864963503649633e-05,
      "loss": 2.904,
      "step": 45500
    },
    {
      "epoch": 0.73,
      "learning_rate": 2.784124087591241e-05,
      "loss": 2.9101,
      "step": 46000
    },
    {
      "epoch": 0.73,
      "eval_loss": 2.8378028869628906,
      "eval_runtime": 450.6185,
      "eval_samples_per_second": 88.767,
      "eval_steps_per_second": 11.096,
      "step": 46000
    },
    {
      "epoch": 0.74,
      "learning_rate": 2.7817518248175184e-05,
      "loss": 2.913,
      "step": 46500
    },
    {
      "epoch": 0.75,
      "learning_rate": 2.7793795620437957e-05,
      "loss": 2.9053,
      "step": 47000
    },
    {
      "epoch": 0.75,
      "eval_loss": 2.840611219406128,
      "eval_runtime": 448.4181,
      "eval_samples_per_second": 89.202,
      "eval_steps_per_second": 11.15,
      "step": 47000
    },
    {
      "epoch": 0.75,
      "learning_rate": 2.777007299270073e-05,
      "loss": 2.8649,
      "step": 47500
    },
    {
      "epoch": 0.76,
      "learning_rate": 2.7746350364963504e-05,
      "loss": 2.9016,
      "step": 48000
    },
    {
      "epoch": 0.76,
      "eval_loss": 2.832401752471924,
      "eval_runtime": 448.5723,
      "eval_samples_per_second": 89.172,
      "eval_steps_per_second": 11.146,
      "step": 48000
    },
    {
      "epoch": 0.77,
      "learning_rate": 2.7722627737226278e-05,
      "loss": 2.9178,
      "step": 48500
    },
    {
      "epoch": 0.78,
      "learning_rate": 2.769890510948905e-05,
      "loss": 2.8928,
      "step": 49000
    },
    {
      "epoch": 0.78,
      "eval_loss": 2.8320999145507812,
      "eval_runtime": 449.0119,
      "eval_samples_per_second": 89.084,
      "eval_steps_per_second": 11.136,
      "step": 49000
    },
    {
      "epoch": 0.78,
      "learning_rate": 2.7675182481751825e-05,
      "loss": 2.903,
      "step": 49500
    },
    {
      "epoch": 0.79,
      "learning_rate": 2.76514598540146e-05,
      "loss": 2.8792,
      "step": 50000
    },
    {
      "epoch": 0.79,
      "eval_loss": 2.836050033569336,
      "eval_runtime": 447.2366,
      "eval_samples_per_second": 89.438,
      "eval_steps_per_second": 11.18,
      "step": 50000
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.7627737226277375e-05,
      "loss": 2.8807,
      "step": 50500
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.7604014598540145e-05,
      "loss": 2.896,
      "step": 51000
    },
    {
      "epoch": 0.81,
      "eval_loss": 2.8326449394226074,
      "eval_runtime": 447.8025,
      "eval_samples_per_second": 89.325,
      "eval_steps_per_second": 11.166,
      "step": 51000
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.7580291970802922e-05,
      "loss": 2.8805,
      "step": 51500
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.7556569343065692e-05,
      "loss": 2.8708,
      "step": 52000
    },
    {
      "epoch": 0.82,
      "eval_loss": 2.830998420715332,
      "eval_runtime": 441.655,
      "eval_samples_per_second": 90.568,
      "eval_steps_per_second": 11.321,
      "step": 52000
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.753284671532847e-05,
      "loss": 2.8656,
      "step": 52500
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.7509124087591243e-05,
      "loss": 2.8882,
      "step": 53000
    },
    {
      "epoch": 0.84,
      "eval_loss": 2.8259706497192383,
      "eval_runtime": 442.0896,
      "eval_samples_per_second": 90.479,
      "eval_steps_per_second": 11.31,
      "step": 53000
    },
    {
      "epoch": 0.85,
      "learning_rate": 2.7485401459854017e-05,
      "loss": 2.8818,
      "step": 53500
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.746167883211679e-05,
      "loss": 2.8682,
      "step": 54000
    },
    {
      "epoch": 0.86,
      "eval_loss": 2.8257410526275635,
      "eval_runtime": 441.6514,
      "eval_samples_per_second": 90.569,
      "eval_steps_per_second": 11.321,
      "step": 54000
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.743795620437956e-05,
      "loss": 2.8765,
      "step": 54500
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.7414233576642337e-05,
      "loss": 2.8843,
      "step": 55000
    },
    {
      "epoch": 0.87,
      "eval_loss": 2.8210554122924805,
      "eval_runtime": 441.9171,
      "eval_samples_per_second": 90.515,
      "eval_steps_per_second": 11.314,
      "step": 55000
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.739051094890511e-05,
      "loss": 2.8785,
      "step": 55500
    },
    {
      "epoch": 0.89,
      "learning_rate": 2.7366788321167884e-05,
      "loss": 2.8631,
      "step": 56000
    },
    {
      "epoch": 0.89,
      "eval_loss": 2.825152635574341,
      "eval_runtime": 442.1075,
      "eval_samples_per_second": 90.476,
      "eval_steps_per_second": 11.309,
      "step": 56000
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.7343065693430658e-05,
      "loss": 2.8557,
      "step": 56500
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.731934306569343e-05,
      "loss": 2.8643,
      "step": 57000
    },
    {
      "epoch": 0.9,
      "eval_loss": 2.8247392177581787,
      "eval_runtime": 441.9526,
      "eval_samples_per_second": 90.507,
      "eval_steps_per_second": 11.313,
      "step": 57000
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.7295620437956205e-05,
      "loss": 2.8976,
      "step": 57500
    },
    {
      "epoch": 0.92,
      "learning_rate": 2.727189781021898e-05,
      "loss": 2.8658,
      "step": 58000
    },
    {
      "epoch": 0.92,
      "eval_loss": 2.8236825466156006,
      "eval_runtime": 441.9598,
      "eval_samples_per_second": 90.506,
      "eval_steps_per_second": 11.313,
      "step": 58000
    },
    {
      "epoch": 0.93,
      "learning_rate": 2.7248175182481752e-05,
      "loss": 2.8608,
      "step": 58500
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.7224452554744525e-05,
      "loss": 2.875,
      "step": 59000
    },
    {
      "epoch": 0.94,
      "eval_loss": 2.820819139480591,
      "eval_runtime": 442.0566,
      "eval_samples_per_second": 90.486,
      "eval_steps_per_second": 11.311,
      "step": 59000
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.72007299270073e-05,
      "loss": 2.8704,
      "step": 59500
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.7177007299270073e-05,
      "loss": 2.8531,
      "step": 60000
    },
    {
      "epoch": 0.95,
      "eval_loss": 2.821174144744873,
      "eval_runtime": 442.2056,
      "eval_samples_per_second": 90.456,
      "eval_steps_per_second": 11.307,
      "step": 60000
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.715328467153285e-05,
      "loss": 2.8748,
      "step": 60500
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.712956204379562e-05,
      "loss": 2.8654,
      "step": 61000
    },
    {
      "epoch": 0.97,
      "eval_loss": 2.825101613998413,
      "eval_runtime": 442.0952,
      "eval_samples_per_second": 90.478,
      "eval_steps_per_second": 11.31,
      "step": 61000
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.7105839416058397e-05,
      "loss": 2.8515,
      "step": 61500
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.7082116788321167e-05,
      "loss": 2.8432,
      "step": 62000
    },
    {
      "epoch": 0.98,
      "eval_loss": 2.819878101348877,
      "eval_runtime": 441.9364,
      "eval_samples_per_second": 90.511,
      "eval_steps_per_second": 11.314,
      "step": 62000
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.7058394160583944e-05,
      "loss": 2.8591,
      "step": 62500
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.7034671532846717e-05,
      "loss": 2.8871,
      "step": 63000
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.8195338249206543,
      "eval_runtime": 442.0595,
      "eval_samples_per_second": 90.486,
      "eval_steps_per_second": 11.311,
      "step": 63000
    },
    {
      "epoch": 1.01,
      "learning_rate": 2.7010948905109487e-05,
      "loss": 2.7957,
      "step": 63500
    },
    {
      "epoch": 1.01,
      "learning_rate": 2.6987226277372264e-05,
      "loss": 2.7854,
      "step": 64000
    },
    {
      "epoch": 1.01,
      "eval_loss": 2.817440986633301,
      "eval_runtime": 442.0252,
      "eval_samples_per_second": 90.493,
      "eval_steps_per_second": 11.312,
      "step": 64000
    },
    {
      "epoch": 1.02,
      "learning_rate": 2.6963503649635034e-05,
      "loss": 2.7874,
      "step": 64500
    },
    {
      "epoch": 1.03,
      "learning_rate": 2.693978102189781e-05,
      "loss": 2.7703,
      "step": 65000
    },
    {
      "epoch": 1.03,
      "eval_loss": 2.8208272457122803,
      "eval_runtime": 442.1489,
      "eval_samples_per_second": 90.467,
      "eval_steps_per_second": 11.308,
      "step": 65000
    },
    {
      "epoch": 1.04,
      "learning_rate": 2.6916058394160585e-05,
      "loss": 2.7773,
      "step": 65500
    },
    {
      "epoch": 1.05,
      "learning_rate": 2.689233576642336e-05,
      "loss": 2.7764,
      "step": 66000
    },
    {
      "epoch": 1.05,
      "eval_loss": 2.8173325061798096,
      "eval_runtime": 443.1184,
      "eval_samples_per_second": 90.269,
      "eval_steps_per_second": 11.284,
      "step": 66000
    },
    {
      "epoch": 1.05,
      "learning_rate": 2.6868613138686132e-05,
      "loss": 2.7838,
      "step": 66500
    },
    {
      "epoch": 1.06,
      "learning_rate": 2.6844890510948906e-05,
      "loss": 2.7734,
      "step": 67000
    },
    {
      "epoch": 1.06,
      "eval_loss": 2.812627077102661,
      "eval_runtime": 443.0624,
      "eval_samples_per_second": 90.281,
      "eval_steps_per_second": 11.285,
      "step": 67000
    },
    {
      "epoch": 1.07,
      "learning_rate": 2.682116788321168e-05,
      "loss": 2.7541,
      "step": 67500
    },
    {
      "epoch": 1.08,
      "learning_rate": 2.6797445255474456e-05,
      "loss": 2.7637,
      "step": 68000
    },
    {
      "epoch": 1.08,
      "eval_loss": 2.817030668258667,
      "eval_runtime": 443.2279,
      "eval_samples_per_second": 90.247,
      "eval_steps_per_second": 11.281,
      "step": 68000
    },
    {
      "epoch": 1.09,
      "learning_rate": 2.6773722627737226e-05,
      "loss": 2.7809,
      "step": 68500
    },
    {
      "epoch": 1.09,
      "learning_rate": 2.675e-05,
      "loss": 2.775,
      "step": 69000
    },
    {
      "epoch": 1.09,
      "eval_loss": 2.819810628890991,
      "eval_runtime": 442.8797,
      "eval_samples_per_second": 90.318,
      "eval_steps_per_second": 11.29,
      "step": 69000
    },
    {
      "epoch": 1.1,
      "learning_rate": 2.6726277372262777e-05,
      "loss": 2.786,
      "step": 69500
    },
    {
      "epoch": 1.11,
      "learning_rate": 2.6702554744525547e-05,
      "loss": 2.747,
      "step": 70000
    },
    {
      "epoch": 1.11,
      "eval_loss": 2.820467233657837,
      "eval_runtime": 443.0381,
      "eval_samples_per_second": 90.286,
      "eval_steps_per_second": 11.286,
      "step": 70000
    },
    {
      "epoch": 1.12,
      "learning_rate": 2.6678832116788324e-05,
      "loss": 2.801,
      "step": 70500
    },
    {
      "epoch": 1.13,
      "learning_rate": 2.6655109489051094e-05,
      "loss": 2.7589,
      "step": 71000
    },
    {
      "epoch": 1.13,
      "eval_loss": 2.8193023204803467,
      "eval_runtime": 443.2053,
      "eval_samples_per_second": 90.252,
      "eval_steps_per_second": 11.281,
      "step": 71000
    },
    {
      "epoch": 1.13,
      "learning_rate": 2.663138686131387e-05,
      "loss": 2.7801,
      "step": 71500
    },
    {
      "epoch": 1.14,
      "learning_rate": 2.6607664233576644e-05,
      "loss": 2.7521,
      "step": 72000
    },
    {
      "epoch": 1.14,
      "eval_loss": 2.813857078552246,
      "eval_runtime": 443.2947,
      "eval_samples_per_second": 90.233,
      "eval_steps_per_second": 11.279,
      "step": 72000
    },
    {
      "epoch": 1.15,
      "learning_rate": 2.6583941605839414e-05,
      "loss": 2.7876,
      "step": 72500
    },
    {
      "epoch": 1.16,
      "learning_rate": 2.656021897810219e-05,
      "loss": 2.772,
      "step": 73000
    },
    {
      "epoch": 1.16,
      "eval_loss": 2.814068555831909,
      "eval_runtime": 443.3415,
      "eval_samples_per_second": 90.224,
      "eval_steps_per_second": 11.278,
      "step": 73000
    },
    {
      "epoch": 1.17,
      "learning_rate": 2.653649635036496e-05,
      "loss": 2.7618,
      "step": 73500
    },
    {
      "epoch": 1.17,
      "learning_rate": 2.651277372262774e-05,
      "loss": 2.785,
      "step": 74000
    },
    {
      "epoch": 1.17,
      "eval_loss": 2.8143815994262695,
      "eval_runtime": 443.2942,
      "eval_samples_per_second": 90.234,
      "eval_steps_per_second": 11.279,
      "step": 74000
    },
    {
      "epoch": 1.18,
      "learning_rate": 2.6489051094890512e-05,
      "loss": 2.7878,
      "step": 74500
    },
    {
      "epoch": 1.19,
      "learning_rate": 2.6465328467153286e-05,
      "loss": 2.743,
      "step": 75000
    },
    {
      "epoch": 1.19,
      "eval_loss": 2.8167881965637207,
      "eval_runtime": 443.0056,
      "eval_samples_per_second": 90.292,
      "eval_steps_per_second": 11.287,
      "step": 75000
    },
    {
      "epoch": 1.2,
      "learning_rate": 2.644160583941606e-05,
      "loss": 2.7564,
      "step": 75500
    },
    {
      "epoch": 1.21,
      "learning_rate": 2.6417883211678833e-05,
      "loss": 2.7694,
      "step": 76000
    },
    {
      "epoch": 1.21,
      "eval_loss": 2.8128414154052734,
      "eval_runtime": 443.1543,
      "eval_samples_per_second": 90.262,
      "eval_steps_per_second": 11.283,
      "step": 76000
    },
    {
      "epoch": 1.21,
      "learning_rate": 2.6394160583941606e-05,
      "loss": 2.7583,
      "step": 76500
    },
    {
      "epoch": 1.22,
      "learning_rate": 2.6370437956204383e-05,
      "loss": 2.7769,
      "step": 77000
    },
    {
      "epoch": 1.22,
      "eval_loss": 2.8131983280181885,
      "eval_runtime": 443.0764,
      "eval_samples_per_second": 90.278,
      "eval_steps_per_second": 11.285,
      "step": 77000
    },
    {
      "epoch": 1.23,
      "learning_rate": 2.6346715328467153e-05,
      "loss": 2.7623,
      "step": 77500
    },
    {
      "epoch": 1.24,
      "learning_rate": 2.6322992700729927e-05,
      "loss": 2.7523,
      "step": 78000
    },
    {
      "epoch": 1.24,
      "eval_loss": 2.813045024871826,
      "eval_runtime": 443.0803,
      "eval_samples_per_second": 90.277,
      "eval_steps_per_second": 11.285,
      "step": 78000
    },
    {
      "epoch": 1.24,
      "learning_rate": 2.62992700729927e-05,
      "loss": 2.7569,
      "step": 78500
    },
    {
      "epoch": 1.25,
      "learning_rate": 2.6275547445255474e-05,
      "loss": 2.7572,
      "step": 79000
    },
    {
      "epoch": 1.25,
      "eval_loss": 2.8214099407196045,
      "eval_runtime": 442.9545,
      "eval_samples_per_second": 90.303,
      "eval_steps_per_second": 11.288,
      "step": 79000
    },
    {
      "epoch": 1.26,
      "learning_rate": 2.625182481751825e-05,
      "loss": 2.7564,
      "step": 79500
    },
    {
      "epoch": 1.27,
      "learning_rate": 2.622810218978102e-05,
      "loss": 2.7573,
      "step": 80000
    },
    {
      "epoch": 1.27,
      "eval_loss": 2.811310052871704,
      "eval_runtime": 442.9337,
      "eval_samples_per_second": 90.307,
      "eval_steps_per_second": 11.288,
      "step": 80000
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.6204379562043798e-05,
      "loss": 2.7635,
      "step": 80500
    },
    {
      "epoch": 1.28,
      "learning_rate": 2.6180656934306568e-05,
      "loss": 2.7514,
      "step": 81000
    },
    {
      "epoch": 1.28,
      "eval_loss": 2.8111774921417236,
      "eval_runtime": 442.9468,
      "eval_samples_per_second": 90.304,
      "eval_steps_per_second": 11.288,
      "step": 81000
    },
    {
      "epoch": 1.29,
      "learning_rate": 2.615693430656934e-05,
      "loss": 2.7534,
      "step": 81500
    },
    {
      "epoch": 1.3,
      "learning_rate": 2.613321167883212e-05,
      "loss": 2.7676,
      "step": 82000
    },
    {
      "epoch": 1.3,
      "eval_loss": 2.8070504665374756,
      "eval_runtime": 442.7918,
      "eval_samples_per_second": 90.336,
      "eval_steps_per_second": 11.292,
      "step": 82000
    },
    {
      "epoch": 1.31,
      "learning_rate": 2.6109489051094892e-05,
      "loss": 2.7212,
      "step": 82500
    },
    {
      "epoch": 1.32,
      "learning_rate": 2.6085766423357666e-05,
      "loss": 2.7044,
      "step": 83000
    },
    {
      "epoch": 1.32,
      "eval_loss": 2.8113818168640137,
      "eval_runtime": 440.1567,
      "eval_samples_per_second": 90.877,
      "eval_steps_per_second": 11.36,
      "step": 83000
    },
    {
      "epoch": 1.32,
      "learning_rate": 2.6062043795620436e-05,
      "loss": 2.742,
      "step": 83500
    },
    {
      "epoch": 1.33,
      "learning_rate": 2.6038321167883213e-05,
      "loss": 2.7205,
      "step": 84000
    },
    {
      "epoch": 1.33,
      "eval_loss": 2.810499906539917,
      "eval_runtime": 440.19,
      "eval_samples_per_second": 90.87,
      "eval_steps_per_second": 11.359,
      "step": 84000
    },
    {
      "epoch": 1.34,
      "learning_rate": 2.6014598540145986e-05,
      "loss": 2.7065,
      "step": 84500
    },
    {
      "epoch": 1.35,
      "learning_rate": 2.599087591240876e-05,
      "loss": 2.7168,
      "step": 85000
    },
    {
      "epoch": 1.35,
      "eval_loss": 2.813863754272461,
      "eval_runtime": 440.1828,
      "eval_samples_per_second": 90.871,
      "eval_steps_per_second": 11.359,
      "step": 85000
    },
    {
      "epoch": 1.36,
      "learning_rate": 2.5967153284671533e-05,
      "loss": 2.7367,
      "step": 85500
    },
    {
      "epoch": 1.36,
      "learning_rate": 2.5943430656934307e-05,
      "loss": 2.7337,
      "step": 86000
    },
    {
      "epoch": 1.36,
      "eval_loss": 2.80753231048584,
      "eval_runtime": 439.9369,
      "eval_samples_per_second": 90.922,
      "eval_steps_per_second": 11.365,
      "step": 86000
    },
    {
      "epoch": 1.37,
      "learning_rate": 2.591970802919708e-05,
      "loss": 2.718,
      "step": 86500
    },
    {
      "epoch": 1.38,
      "learning_rate": 2.5895985401459854e-05,
      "loss": 2.709,
      "step": 87000
    },
    {
      "epoch": 1.38,
      "eval_loss": 2.8110671043395996,
      "eval_runtime": 439.984,
      "eval_samples_per_second": 90.912,
      "eval_steps_per_second": 11.364,
      "step": 87000
    },
    {
      "epoch": 1.39,
      "learning_rate": 2.5872262773722628e-05,
      "loss": 2.7045,
      "step": 87500
    },
    {
      "epoch": 1.4,
      "learning_rate": 2.58485401459854e-05,
      "loss": 2.7449,
      "step": 88000
    },
    {
      "epoch": 1.4,
      "eval_loss": 2.806777238845825,
      "eval_runtime": 439.8672,
      "eval_samples_per_second": 90.937,
      "eval_steps_per_second": 11.367,
      "step": 88000
    },
    {
      "epoch": 1.4,
      "learning_rate": 2.5824817518248178e-05,
      "loss": 2.704,
      "step": 88500
    },
    {
      "epoch": 1.41,
      "learning_rate": 2.5801094890510948e-05,
      "loss": 2.7175,
      "step": 89000
    },
    {
      "epoch": 1.41,
      "eval_loss": 2.8058323860168457,
      "eval_runtime": 439.8993,
      "eval_samples_per_second": 90.93,
      "eval_steps_per_second": 11.366,
      "step": 89000
    },
    {
      "epoch": 1.42,
      "learning_rate": 2.5777372262773725e-05,
      "loss": 2.7292,
      "step": 89500
    },
    {
      "epoch": 1.43,
      "learning_rate": 2.5753649635036495e-05,
      "loss": 2.7096,
      "step": 90000
    },
    {
      "epoch": 1.43,
      "eval_loss": 2.808027744293213,
      "eval_runtime": 439.7855,
      "eval_samples_per_second": 90.953,
      "eval_steps_per_second": 11.369,
      "step": 90000
    },
    {
      "epoch": 1.43,
      "learning_rate": 2.572992700729927e-05,
      "loss": 2.7125,
      "step": 90500
    },
    {
      "epoch": 1.44,
      "learning_rate": 2.5706204379562046e-05,
      "loss": 2.7272,
      "step": 91000
    },
    {
      "epoch": 1.44,
      "eval_loss": 2.8089358806610107,
      "eval_runtime": 439.8804,
      "eval_samples_per_second": 90.934,
      "eval_steps_per_second": 11.367,
      "step": 91000
    },
    {
      "epoch": 1.45,
      "learning_rate": 2.568248175182482e-05,
      "loss": 2.719,
      "step": 91500
    },
    {
      "epoch": 1.46,
      "learning_rate": 2.5658759124087593e-05,
      "loss": 2.7092,
      "step": 92000
    },
    {
      "epoch": 1.46,
      "eval_loss": 2.806529998779297,
      "eval_runtime": 439.7422,
      "eval_samples_per_second": 90.962,
      "eval_steps_per_second": 11.37,
      "step": 92000
    },
    {
      "epoch": 1.47,
      "learning_rate": 2.5635036496350363e-05,
      "loss": 2.7121,
      "step": 92500
    },
    {
      "epoch": 1.47,
      "learning_rate": 2.561131386861314e-05,
      "loss": 2.6846,
      "step": 93000
    },
    {
      "epoch": 1.47,
      "eval_loss": 2.804901123046875,
      "eval_runtime": 440.0389,
      "eval_samples_per_second": 90.901,
      "eval_steps_per_second": 11.363,
      "step": 93000
    },
    {
      "epoch": 1.48,
      "learning_rate": 2.5587591240875913e-05,
      "loss": 2.7301,
      "step": 93500
    },
    {
      "epoch": 1.49,
      "learning_rate": 2.5563868613138687e-05,
      "loss": 2.6976,
      "step": 94000
    },
    {
      "epoch": 1.49,
      "eval_loss": 2.8061363697052,
      "eval_runtime": 439.8304,
      "eval_samples_per_second": 90.944,
      "eval_steps_per_second": 11.368,
      "step": 94000
    },
    {
      "epoch": 1.5,
      "learning_rate": 2.554014598540146e-05,
      "loss": 2.7008,
      "step": 94500
    },
    {
      "epoch": 1.51,
      "learning_rate": 2.5516423357664234e-05,
      "loss": 2.7057,
      "step": 95000
    },
    {
      "epoch": 1.51,
      "eval_loss": 2.815894603729248,
      "eval_runtime": 439.7552,
      "eval_samples_per_second": 90.96,
      "eval_steps_per_second": 11.37,
      "step": 95000
    },
    {
      "epoch": 1.51,
      "learning_rate": 2.5492700729927008e-05,
      "loss": 2.724,
      "step": 95500
    },
    {
      "epoch": 1.52,
      "learning_rate": 2.546897810218978e-05,
      "loss": 2.6922,
      "step": 96000
    },
    {
      "epoch": 1.52,
      "eval_loss": 2.8022561073303223,
      "eval_runtime": 439.8096,
      "eval_samples_per_second": 90.948,
      "eval_steps_per_second": 11.369,
      "step": 96000
    },
    {
      "epoch": 1.53,
      "learning_rate": 2.5445255474452555e-05,
      "loss": 2.7095,
      "step": 96500
    },
    {
      "epoch": 1.54,
      "learning_rate": 2.542153284671533e-05,
      "loss": 2.7011,
      "step": 97000
    },
    {
      "epoch": 1.54,
      "eval_loss": 2.801751136779785,
      "eval_runtime": 439.7019,
      "eval_samples_per_second": 90.971,
      "eval_steps_per_second": 11.371,
      "step": 97000
    },
    {
      "epoch": 1.55,
      "learning_rate": 2.5397810218978102e-05,
      "loss": 2.7212,
      "step": 97500
    },
    {
      "epoch": 1.55,
      "learning_rate": 2.5374087591240875e-05,
      "loss": 2.7029,
      "step": 98000
    },
    {
      "epoch": 1.55,
      "eval_loss": 2.8029119968414307,
      "eval_runtime": 439.744,
      "eval_samples_per_second": 90.962,
      "eval_steps_per_second": 11.37,
      "step": 98000
    },
    {
      "epoch": 1.56,
      "learning_rate": 2.5350364963503652e-05,
      "loss": 2.7055,
      "step": 98500
    },
    {
      "epoch": 1.57,
      "learning_rate": 2.5326642335766422e-05,
      "loss": 2.701,
      "step": 99000
    },
    {
      "epoch": 1.57,
      "eval_loss": 2.800555944442749,
      "eval_runtime": 439.7759,
      "eval_samples_per_second": 90.955,
      "eval_steps_per_second": 11.369,
      "step": 99000
    },
    {
      "epoch": 1.58,
      "learning_rate": 2.53029197080292e-05,
      "loss": 2.7136,
      "step": 99500
    },
    {
      "epoch": 1.59,
      "learning_rate": 2.527919708029197e-05,
      "loss": 2.7044,
      "step": 100000
    },
    {
      "epoch": 1.59,
      "eval_loss": 2.8019235134124756,
      "eval_runtime": 439.7518,
      "eval_samples_per_second": 90.96,
      "eval_steps_per_second": 11.37,
      "step": 100000
    },
    {
      "epoch": 1.59,
      "learning_rate": 2.5255474452554746e-05,
      "loss": 2.7141,
      "step": 100500
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.523175182481752e-05,
      "loss": 2.7311,
      "step": 101000
    },
    {
      "epoch": 1.6,
      "eval_loss": 2.8053483963012695,
      "eval_runtime": 439.6812,
      "eval_samples_per_second": 90.975,
      "eval_steps_per_second": 11.372,
      "step": 101000
    },
    {
      "epoch": 1.61,
      "learning_rate": 2.520802919708029e-05,
      "loss": 2.7107,
      "step": 101500
    },
    {
      "epoch": 1.62,
      "learning_rate": 2.5184306569343067e-05,
      "loss": 2.711,
      "step": 102000
    },
    {
      "epoch": 1.62,
      "eval_loss": 2.805739641189575,
      "eval_runtime": 439.6645,
      "eval_samples_per_second": 90.978,
      "eval_steps_per_second": 11.372,
      "step": 102000
    },
    {
      "epoch": 1.63,
      "learning_rate": 2.5160583941605837e-05,
      "loss": 2.7212,
      "step": 102500
    },
    {
      "epoch": 1.63,
      "learning_rate": 2.5136861313868614e-05,
      "loss": 2.7344,
      "step": 103000
    },
    {
      "epoch": 1.63,
      "eval_loss": 2.799654722213745,
      "eval_runtime": 439.7601,
      "eval_samples_per_second": 90.959,
      "eval_steps_per_second": 11.37,
      "step": 103000
    },
    {
      "epoch": 1.64,
      "learning_rate": 2.5113138686131388e-05,
      "loss": 2.7122,
      "step": 103500
    },
    {
      "epoch": 1.65,
      "learning_rate": 2.508941605839416e-05,
      "loss": 2.7236,
      "step": 104000
    },
    {
      "epoch": 1.65,
      "eval_loss": 2.803441047668457,
      "eval_runtime": 439.7253,
      "eval_samples_per_second": 90.966,
      "eval_steps_per_second": 11.371,
      "step": 104000
    },
    {
      "epoch": 1.66,
      "learning_rate": 2.5065693430656935e-05,
      "loss": 2.7226,
      "step": 104500
    },
    {
      "epoch": 1.66,
      "learning_rate": 2.5041970802919705e-05,
      "loss": 2.6879,
      "step": 105000
    },
    {
      "epoch": 1.66,
      "eval_loss": 2.801588773727417,
      "eval_runtime": 439.6315,
      "eval_samples_per_second": 90.985,
      "eval_steps_per_second": 11.373,
      "step": 105000
    },
    {
      "epoch": 1.67,
      "learning_rate": 2.5018248175182482e-05,
      "loss": 2.732,
      "step": 105500
    },
    {
      "epoch": 1.68,
      "learning_rate": 2.499452554744526e-05,
      "loss": 2.7188,
      "step": 106000
    },
    {
      "epoch": 1.68,
      "eval_loss": 2.7991223335266113,
      "eval_runtime": 439.7752,
      "eval_samples_per_second": 90.956,
      "eval_steps_per_second": 11.369,
      "step": 106000
    },
    {
      "epoch": 1.69,
      "learning_rate": 2.497080291970803e-05,
      "loss": 2.7164,
      "step": 106500
    },
    {
      "epoch": 1.7,
      "learning_rate": 2.4947080291970802e-05,
      "loss": 2.6994,
      "step": 107000
    },
    {
      "epoch": 1.7,
      "eval_loss": 2.8013699054718018,
      "eval_runtime": 439.7722,
      "eval_samples_per_second": 90.956,
      "eval_steps_per_second": 11.37,
      "step": 107000
    },
    {
      "epoch": 1.7,
      "learning_rate": 2.492335766423358e-05,
      "loss": 2.6989,
      "step": 107500
    },
    {
      "epoch": 1.71,
      "learning_rate": 2.489963503649635e-05,
      "loss": 2.7129,
      "step": 108000
    },
    {
      "epoch": 1.71,
      "eval_loss": 2.7973504066467285,
      "eval_runtime": 439.8695,
      "eval_samples_per_second": 90.936,
      "eval_steps_per_second": 11.367,
      "step": 108000
    },
    {
      "epoch": 1.72,
      "learning_rate": 2.4875912408759126e-05,
      "loss": 2.7056,
      "step": 108500
    },
    {
      "epoch": 1.73,
      "learning_rate": 2.4852189781021897e-05,
      "loss": 2.7095,
      "step": 109000
    },
    {
      "epoch": 1.73,
      "eval_loss": 2.8017683029174805,
      "eval_runtime": 439.7668,
      "eval_samples_per_second": 90.957,
      "eval_steps_per_second": 11.37,
      "step": 109000
    },
    {
      "epoch": 1.74,
      "learning_rate": 2.4828467153284674e-05,
      "loss": 2.7021,
      "step": 109500
    },
    {
      "epoch": 1.74,
      "learning_rate": 2.4804744525547447e-05,
      "loss": 2.7029,
      "step": 110000
    },
    {
      "epoch": 1.74,
      "eval_loss": 2.7975335121154785,
      "eval_runtime": 439.9718,
      "eval_samples_per_second": 90.915,
      "eval_steps_per_second": 11.364,
      "step": 110000
    },
    {
      "epoch": 1.75,
      "learning_rate": 2.4781021897810217e-05,
      "loss": 2.7046,
      "step": 110500
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.4757299270072994e-05,
      "loss": 2.7042,
      "step": 111000
    },
    {
      "epoch": 1.76,
      "eval_loss": 2.7961161136627197,
      "eval_runtime": 439.788,
      "eval_samples_per_second": 90.953,
      "eval_steps_per_second": 11.369,
      "step": 111000
    },
    {
      "epoch": 1.77,
      "learning_rate": 2.4733576642335768e-05,
      "loss": 2.7234,
      "step": 111500
    },
    {
      "epoch": 1.78,
      "learning_rate": 2.470985401459854e-05,
      "loss": 2.7134,
      "step": 112000
    },
    {
      "epoch": 1.78,
      "eval_loss": 2.795431613922119,
      "eval_runtime": 439.911,
      "eval_samples_per_second": 90.927,
      "eval_steps_per_second": 11.366,
      "step": 112000
    },
    {
      "epoch": 1.78,
      "learning_rate": 2.4686131386861315e-05,
      "loss": 2.6937,
      "step": 112500
    },
    {
      "epoch": 1.79,
      "learning_rate": 2.4662408759124088e-05,
      "loss": 2.6905,
      "step": 113000
    },
    {
      "epoch": 1.79,
      "eval_loss": 2.7966976165771484,
      "eval_runtime": 439.7831,
      "eval_samples_per_second": 90.954,
      "eval_steps_per_second": 11.369,
      "step": 113000
    },
    {
      "epoch": 1.8,
      "learning_rate": 2.4638686131386862e-05,
      "loss": 2.7027,
      "step": 113500
    },
    {
      "epoch": 1.81,
      "learning_rate": 2.4614963503649635e-05,
      "loss": 2.6871,
      "step": 114000
    },
    {
      "epoch": 1.81,
      "eval_loss": 2.7926506996154785,
      "eval_runtime": 440.0624,
      "eval_samples_per_second": 90.896,
      "eval_steps_per_second": 11.362,
      "step": 114000
    },
    {
      "epoch": 1.82,
      "learning_rate": 2.459124087591241e-05,
      "loss": 2.697,
      "step": 114500
    },
    {
      "epoch": 1.82,
      "learning_rate": 2.4567518248175186e-05,
      "loss": 2.6931,
      "step": 115000
    },
    {
      "epoch": 1.82,
      "eval_loss": 2.8011257648468018,
      "eval_runtime": 439.8116,
      "eval_samples_per_second": 90.948,
      "eval_steps_per_second": 11.369,
      "step": 115000
    },
    {
      "epoch": 1.83,
      "learning_rate": 2.4543795620437956e-05,
      "loss": 2.7136,
      "step": 115500
    },
    {
      "epoch": 1.84,
      "learning_rate": 2.452007299270073e-05,
      "loss": 2.702,
      "step": 116000
    },
    {
      "epoch": 1.84,
      "eval_loss": 2.7955451011657715,
      "eval_runtime": 439.914,
      "eval_samples_per_second": 90.927,
      "eval_steps_per_second": 11.366,
      "step": 116000
    },
    {
      "epoch": 1.85,
      "learning_rate": 2.4496350364963503e-05,
      "loss": 2.6859,
      "step": 116500
    },
    {
      "epoch": 1.86,
      "learning_rate": 2.4472627737226277e-05,
      "loss": 2.712,
      "step": 117000
    },
    {
      "epoch": 1.86,
      "eval_loss": 2.7944655418395996,
      "eval_runtime": 439.809,
      "eval_samples_per_second": 90.949,
      "eval_steps_per_second": 11.369,
      "step": 117000
    },
    {
      "epoch": 1.86,
      "learning_rate": 2.4448905109489054e-05,
      "loss": 2.6984,
      "step": 117500
    },
    {
      "epoch": 1.87,
      "learning_rate": 2.4425182481751824e-05,
      "loss": 2.7091,
      "step": 118000
    },
    {
      "epoch": 1.87,
      "eval_loss": 2.7959797382354736,
      "eval_runtime": 439.9158,
      "eval_samples_per_second": 90.926,
      "eval_steps_per_second": 11.366,
      "step": 118000
    },
    {
      "epoch": 1.88,
      "learning_rate": 2.44014598540146e-05,
      "loss": 2.6978,
      "step": 118500
    },
    {
      "epoch": 1.89,
      "learning_rate": 2.437773722627737e-05,
      "loss": 2.6913,
      "step": 119000
    },
    {
      "epoch": 1.89,
      "eval_loss": 2.79091739654541,
      "eval_runtime": 439.7923,
      "eval_samples_per_second": 90.952,
      "eval_steps_per_second": 11.369,
      "step": 119000
    },
    {
      "epoch": 1.89,
      "learning_rate": 2.4354014598540144e-05,
      "loss": 2.6964,
      "step": 119500
    },
    {
      "epoch": 1.9,
      "learning_rate": 2.433029197080292e-05,
      "loss": 2.7008,
      "step": 120000
    },
    {
      "epoch": 1.9,
      "eval_loss": 2.7938427925109863,
      "eval_runtime": 439.9595,
      "eval_samples_per_second": 90.917,
      "eval_steps_per_second": 11.365,
      "step": 120000
    },
    {
      "epoch": 1.91,
      "learning_rate": 2.4306569343065695e-05,
      "loss": 2.7015,
      "step": 120500
    },
    {
      "epoch": 1.92,
      "learning_rate": 2.428284671532847e-05,
      "loss": 2.6968,
      "step": 121000
    },
    {
      "epoch": 1.92,
      "eval_loss": 2.796417236328125,
      "eval_runtime": 439.8462,
      "eval_samples_per_second": 90.941,
      "eval_steps_per_second": 11.368,
      "step": 121000
    },
    {
      "epoch": 1.93,
      "learning_rate": 2.425912408759124e-05,
      "loss": 2.6936,
      "step": 121500
    },
    {
      "epoch": 1.93,
      "learning_rate": 2.4235401459854015e-05,
      "loss": 2.688,
      "step": 122000
    },
    {
      "epoch": 1.93,
      "eval_loss": 2.790297269821167,
      "eval_runtime": 439.89,
      "eval_samples_per_second": 90.932,
      "eval_steps_per_second": 11.366,
      "step": 122000
    },
    {
      "epoch": 1.94,
      "learning_rate": 2.421167883211679e-05,
      "loss": 2.6885,
      "step": 122500
    },
    {
      "epoch": 1.95,
      "learning_rate": 2.4187956204379563e-05,
      "loss": 2.7015,
      "step": 123000
    },
    {
      "epoch": 1.95,
      "eval_loss": 2.7968406677246094,
      "eval_runtime": 439.8161,
      "eval_samples_per_second": 90.947,
      "eval_steps_per_second": 11.368,
      "step": 123000
    },
    {
      "epoch": 1.96,
      "learning_rate": 2.4164233576642336e-05,
      "loss": 2.7095,
      "step": 123500
    },
    {
      "epoch": 1.97,
      "learning_rate": 2.414051094890511e-05,
      "loss": 2.7038,
      "step": 124000
    },
    {
      "epoch": 1.97,
      "eval_loss": 2.79244327545166,
      "eval_runtime": 439.8929,
      "eval_samples_per_second": 90.931,
      "eval_steps_per_second": 11.366,
      "step": 124000
    },
    {
      "epoch": 1.97,
      "learning_rate": 2.4116788321167883e-05,
      "loss": 2.6901,
      "step": 124500
    },
    {
      "epoch": 1.98,
      "learning_rate": 2.4093065693430657e-05,
      "loss": 2.6991,
      "step": 125000
    },
    {
      "epoch": 1.98,
      "eval_loss": 2.7943027019500732,
      "eval_runtime": 439.7901,
      "eval_samples_per_second": 90.952,
      "eval_steps_per_second": 11.369,
      "step": 125000
    },
    {
      "epoch": 1.99,
      "learning_rate": 2.406934306569343e-05,
      "loss": 2.6954,
      "step": 125500
    },
    {
      "epoch": 2.0,
      "learning_rate": 2.4045620437956207e-05,
      "loss": 2.6804,
      "step": 126000
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.785961627960205,
      "eval_runtime": 440.0425,
      "eval_samples_per_second": 90.9,
      "eval_steps_per_second": 11.363,
      "step": 126000
    }
  ],
  "max_steps": 630700,
  "num_train_epochs": 10,
  "total_flos": 8.47096165805568e+17,
  "trial_name": null,
  "trial_params": null
}