{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 2000,
  "global_step": 100000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 13.3125,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.5082,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "grad_norm": 11.75,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.2924,
      "step": 2000
    },
    {
      "epoch": 0.02,
      "eval_loss": 1.2315222024917603,
      "eval_runtime": 78.9092,
      "eval_samples_per_second": 49.563,
      "eval_steps_per_second": 12.394,
      "step": 2000
    },
    {
      "epoch": 0.03,
      "grad_norm": 11.0625,
      "learning_rate": 6e-06,
      "loss": 1.18,
      "step": 3000
    },
    {
      "epoch": 0.04,
      "grad_norm": 11.4375,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.101,
      "step": 4000
    },
    {
      "epoch": 0.04,
      "eval_loss": 1.073156714439392,
      "eval_runtime": 79.3697,
      "eval_samples_per_second": 49.276,
      "eval_steps_per_second": 12.322,
      "step": 4000
    },
    {
      "epoch": 0.05,
      "grad_norm": 11.125,
      "learning_rate": 1e-05,
      "loss": 1.0474,
      "step": 5000
    },
    {
      "epoch": 0.06,
      "grad_norm": 10.125,
      "learning_rate": 9.99726628670463e-06,
      "loss": 0.9949,
      "step": 6000
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.978751003742218,
      "eval_runtime": 79.5844,
      "eval_samples_per_second": 49.143,
      "eval_steps_per_second": 12.289,
      "step": 6000
    },
    {
      "epoch": 0.07,
      "grad_norm": 8.8125,
      "learning_rate": 9.989068136093873e-06,
      "loss": 0.9562,
      "step": 7000
    },
    {
      "epoch": 0.08,
      "grad_norm": 8.375,
      "learning_rate": 9.975414512725058e-06,
      "loss": 0.9365,
      "step": 8000
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.9295967221260071,
      "eval_runtime": 78.9753,
      "eval_samples_per_second": 49.522,
      "eval_steps_per_second": 12.384,
      "step": 8000
    },
    {
      "epoch": 0.09,
      "grad_norm": 9.8125,
      "learning_rate": 9.956320346634877e-06,
      "loss": 0.9169,
      "step": 9000
    },
    {
      "epoch": 0.1,
      "grad_norm": 8.9375,
      "learning_rate": 9.931806517013612e-06,
      "loss": 0.9036,
      "step": 10000
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.9004048705101013,
      "eval_runtime": 79.5265,
      "eval_samples_per_second": 49.179,
      "eval_steps_per_second": 12.298,
      "step": 10000
    },
    {
      "epoch": 0.11,
      "grad_norm": 9.0625,
      "learning_rate": 9.901899829374048e-06,
      "loss": 0.89,
      "step": 11000
    },
    {
      "epoch": 0.12,
      "grad_norm": 9.125,
      "learning_rate": 9.86663298624003e-06,
      "loss": 0.8811,
      "step": 12000
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.8809531331062317,
      "eval_runtime": 79.2169,
      "eval_samples_per_second": 49.371,
      "eval_steps_per_second": 12.346,
      "step": 12000
    },
    {
      "epoch": 0.13,
      "grad_norm": 8.8125,
      "learning_rate": 9.826044551386743e-06,
      "loss": 0.8724,
      "step": 13000
    },
    {
      "epoch": 0.14,
      "grad_norm": 8.6875,
      "learning_rate": 9.780178907671788e-06,
      "loss": 0.8656,
      "step": 14000
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.8673146963119507,
      "eval_runtime": 78.9692,
      "eval_samples_per_second": 49.526,
      "eval_steps_per_second": 12.385,
      "step": 14000
    },
    {
      "epoch": 0.15,
      "grad_norm": 9.25,
      "learning_rate": 9.729086208503174e-06,
      "loss": 0.8562,
      "step": 15000
    },
    {
      "epoch": 0.16,
      "grad_norm": 8.8125,
      "learning_rate": 9.672822322997305e-06,
      "loss": 0.8493,
      "step": 16000
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.8561869859695435,
      "eval_runtime": 79.0551,
      "eval_samples_per_second": 49.472,
      "eval_steps_per_second": 12.371,
      "step": 16000
    },
    {
      "epoch": 0.17,
      "grad_norm": 9.375,
      "learning_rate": 9.611448774886925e-06,
      "loss": 0.8459,
      "step": 17000
    },
    {
      "epoch": 0.18,
      "grad_norm": 9.1875,
      "learning_rate": 9.545032675245814e-06,
      "loss": 0.8433,
      "step": 18000
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.8474856615066528,
      "eval_runtime": 79.8666,
      "eval_samples_per_second": 48.969,
      "eval_steps_per_second": 12.245,
      "step": 18000
    },
    {
      "epoch": 0.19,
      "grad_norm": 9.0,
      "learning_rate": 9.473646649103819e-06,
      "loss": 0.8409,
      "step": 19000
    },
    {
      "epoch": 0.2,
      "grad_norm": 8.6875,
      "learning_rate": 9.397368756032445e-06,
      "loss": 0.8352,
      "step": 20000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.8410937786102295,
      "eval_runtime": 79.0729,
      "eval_samples_per_second": 49.461,
      "eval_steps_per_second": 12.368,
      "step": 20000
    },
    {
      "epoch": 0.21,
      "grad_norm": 10.0,
      "learning_rate": 9.31628240478787e-06,
      "loss": 0.8347,
      "step": 21000
    },
    {
      "epoch": 0.22,
      "grad_norm": 9.3125,
      "learning_rate": 9.230476262104678e-06,
      "loss": 0.8295,
      "step": 22000
    },
    {
      "epoch": 0.22,
      "eval_loss": 0.8347375392913818,
      "eval_runtime": 79.1246,
      "eval_samples_per_second": 49.428,
      "eval_steps_per_second": 12.36,
      "step": 22000
    },
    {
      "epoch": 0.23,
      "grad_norm": 9.875,
      "learning_rate": 9.140044155740102e-06,
      "loss": 0.8275,
      "step": 23000
    },
    {
      "epoch": 0.24,
      "grad_norm": 10.1875,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.822,
      "step": 24000
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.8303326368331909,
      "eval_runtime": 78.9475,
      "eval_samples_per_second": 49.539,
      "eval_steps_per_second": 12.388,
      "step": 24000
    },
    {
      "epoch": 0.25,
      "grad_norm": 8.875,
      "learning_rate": 8.94570254698197e-06,
      "loss": 0.8245,
      "step": 25000
    },
    {
      "epoch": 0.26,
      "grad_norm": 9.0625,
      "learning_rate": 8.842005554284296e-06,
      "loss": 0.8216,
      "step": 26000
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.8262131810188293,
      "eval_runtime": 78.9434,
      "eval_samples_per_second": 49.542,
      "eval_steps_per_second": 12.389,
      "step": 26000
    },
    {
      "epoch": 0.27,
      "grad_norm": 8.875,
      "learning_rate": 8.734107384920771e-06,
      "loss": 0.8174,
      "step": 27000
    },
    {
      "epoch": 0.28,
      "grad_norm": 10.125,
      "learning_rate": 8.622126023955446e-06,
      "loss": 0.8116,
      "step": 28000
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.8211948275566101,
      "eval_runtime": 78.9986,
      "eval_samples_per_second": 49.507,
      "eval_steps_per_second": 12.38,
      "step": 28000
    },
    {
      "epoch": 0.29,
      "grad_norm": 8.375,
      "learning_rate": 8.506183921362443e-06,
      "loss": 0.8073,
      "step": 29000
    },
    {
      "epoch": 0.3,
      "grad_norm": 8.3125,
      "learning_rate": 8.386407858128707e-06,
      "loss": 0.7966,
      "step": 30000
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.8195165991783142,
      "eval_runtime": 78.951,
      "eval_samples_per_second": 49.537,
      "eval_steps_per_second": 12.387,
      "step": 30000
    },
    {
      "epoch": 0.31,
      "grad_norm": 9.25,
      "learning_rate": 8.262928807620843e-06,
      "loss": 0.7853,
      "step": 31000
    },
    {
      "epoch": 0.32,
      "grad_norm": 8.8125,
      "learning_rate": 8.135881792367686e-06,
      "loss": 0.7807,
      "step": 32000
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.8171260952949524,
      "eval_runtime": 79.0444,
      "eval_samples_per_second": 49.479,
      "eval_steps_per_second": 12.373,
      "step": 32000
    },
    {
      "epoch": 0.33,
      "grad_norm": 8.6875,
      "learning_rate": 8.005405736415127e-06,
      "loss": 0.7828,
      "step": 33000
    },
    {
      "epoch": 0.34,
      "grad_norm": 9.375,
      "learning_rate": 7.871643313414718e-06,
      "loss": 0.7794,
      "step": 34000
    },
    {
      "epoch": 0.34,
      "eval_loss": 0.8152210712432861,
      "eval_runtime": 78.8407,
      "eval_samples_per_second": 49.606,
      "eval_steps_per_second": 12.405,
      "step": 34000
    },
    {
      "epoch": 0.35,
      "grad_norm": 9.375,
      "learning_rate": 7.734740790612137e-06,
      "loss": 0.7846,
      "step": 35000
    },
    {
      "epoch": 0.36,
      "grad_norm": 9.1875,
      "learning_rate": 7.594847868906076e-06,
      "loss": 0.7812,
      "step": 36000
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.8131356239318848,
      "eval_runtime": 79.0096,
      "eval_samples_per_second": 49.5,
      "eval_steps_per_second": 12.378,
      "step": 36000
    },
    {
      "epoch": 0.37,
      "grad_norm": 9.75,
      "learning_rate": 7.452117519152542e-06,
      "loss": 0.7798,
      "step": 37000
    },
    {
      "epoch": 0.38,
      "grad_norm": 9.4375,
      "learning_rate": 7.30670581489344e-06,
      "loss": 0.7819,
      "step": 38000
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.8113033175468445,
      "eval_runtime": 79.0043,
      "eval_samples_per_second": 49.504,
      "eval_steps_per_second": 12.379,
      "step": 38000
    },
    {
      "epoch": 0.39,
      "grad_norm": 9.375,
      "learning_rate": 7.158771761692464e-06,
      "loss": 0.7825,
      "step": 39000
    },
    {
      "epoch": 0.4,
      "grad_norm": 10.0625,
      "learning_rate": 7.008477123264849e-06,
      "loss": 0.7798,
      "step": 40000
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.8099530935287476,
      "eval_runtime": 79.3148,
      "eval_samples_per_second": 49.31,
      "eval_steps_per_second": 12.331,
      "step": 40000
    },
    {
      "epoch": 0.41,
      "grad_norm": 9.0,
      "learning_rate": 6.855986244591104e-06,
      "loss": 0.7785,
      "step": 41000
    },
    {
      "epoch": 0.42,
      "grad_norm": 10.625,
      "learning_rate": 6.701465872208216e-06,
      "loss": 0.7777,
      "step": 42000
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.8080056309700012,
      "eval_runtime": 79.2187,
      "eval_samples_per_second": 49.37,
      "eval_steps_per_second": 12.346,
      "step": 42000
    },
    {
      "epoch": 0.43,
      "grad_norm": 9.75,
      "learning_rate": 6.545084971874738e-06,
      "loss": 0.7799,
      "step": 43000
    },
    {
      "epoch": 0.44,
      "grad_norm": 9.1875,
      "learning_rate": 6.387014543809224e-06,
      "loss": 0.7775,
      "step": 44000
    },
    {
      "epoch": 0.44,
      "eval_loss": 0.8067404627799988,
      "eval_runtime": 79.9447,
      "eval_samples_per_second": 48.921,
      "eval_steps_per_second": 12.233,
      "step": 44000
    },
    {
      "epoch": 0.45,
      "grad_norm": 9.3125,
      "learning_rate": 6.227427435703997e-06,
      "loss": 0.7792,
      "step": 45000
    },
    {
      "epoch": 0.46,
      "grad_norm": 9.0,
      "learning_rate": 6.066498153718735e-06,
      "loss": 0.7789,
      "step": 46000
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.8063809871673584,
      "eval_runtime": 79.5507,
      "eval_samples_per_second": 49.164,
      "eval_steps_per_second": 12.294,
      "step": 46000
    },
    {
      "epoch": 0.47,
      "grad_norm": 9.625,
      "learning_rate": 5.904402671660551e-06,
      "loss": 0.7813,
      "step": 47000
    },
    {
      "epoch": 0.48,
      "grad_norm": 10.6875,
      "learning_rate": 5.74131823855921e-06,
      "loss": 0.7795,
      "step": 48000
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.8048525452613831,
      "eval_runtime": 79.0053,
      "eval_samples_per_second": 49.503,
      "eval_steps_per_second": 12.379,
      "step": 48000
    },
    {
      "epoch": 0.49,
      "grad_norm": 9.875,
      "learning_rate": 5.577423184847932e-06,
      "loss": 0.7787,
      "step": 49000
    },
    {
      "epoch": 0.5,
      "grad_norm": 10.0625,
      "learning_rate": 5.412896727361663e-06,
      "loss": 0.781,
      "step": 50000
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.8037334084510803,
      "eval_runtime": 80.9609,
      "eval_samples_per_second": 48.307,
      "eval_steps_per_second": 12.08,
      "step": 50000
    },
    {
      "epoch": 0.51,
      "grad_norm": 8.875,
      "learning_rate": 5.247918773366112e-06,
      "loss": 0.7814,
      "step": 51000
    },
    {
      "epoch": 0.52,
      "grad_norm": 9.6875,
      "learning_rate": 5.082669723831793e-06,
      "loss": 0.7782,
      "step": 52000
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.8030668497085571,
      "eval_runtime": 79.8706,
      "eval_samples_per_second": 48.967,
      "eval_steps_per_second": 12.245,
      "step": 52000
    },
    {
      "epoch": 0.53,
      "grad_norm": 8.6875,
      "learning_rate": 4.917330276168208e-06,
      "loss": 0.7771,
      "step": 53000
    },
    {
      "epoch": 0.54,
      "grad_norm": 9.0625,
      "learning_rate": 4.752081226633888e-06,
      "loss": 0.773,
      "step": 54000
    },
    {
      "epoch": 0.54,
      "eval_loss": 0.8017913103103638,
      "eval_runtime": 80.4358,
      "eval_samples_per_second": 48.623,
      "eval_steps_per_second": 12.159,
      "step": 54000
    },
    {
      "epoch": 0.55,
      "grad_norm": 8.25,
      "learning_rate": 4.587103272638339e-06,
      "loss": 0.7708,
      "step": 55000
    },
    {
      "epoch": 0.56,
      "grad_norm": 9.625,
      "learning_rate": 4.42257681515207e-06,
      "loss": 0.7645,
      "step": 56000
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.8031152486801147,
      "eval_runtime": 80.4118,
      "eval_samples_per_second": 48.637,
      "eval_steps_per_second": 12.162,
      "step": 56000
    },
    {
      "epoch": 0.57,
      "grad_norm": 9.1875,
      "learning_rate": 4.25868176144079e-06,
      "loss": 0.7504,
      "step": 57000
    },
    {
      "epoch": 0.58,
      "grad_norm": 10.0,
      "learning_rate": 4.0955973283394525e-06,
      "loss": 0.7534,
      "step": 58000
    },
    {
      "epoch": 0.58,
      "eval_loss": 0.8028683662414551,
      "eval_runtime": 79.2432,
      "eval_samples_per_second": 49.354,
      "eval_steps_per_second": 12.342,
      "step": 58000
    },
    {
      "epoch": 0.59,
      "grad_norm": 9.4375,
      "learning_rate": 3.9335018462812664e-06,
      "loss": 0.7563,
      "step": 59000
    },
    {
      "epoch": 0.6,
      "grad_norm": 9.625,
      "learning_rate": 3.7725725642960047e-06,
      "loss": 0.7572,
      "step": 60000
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.8028994202613831,
      "eval_runtime": 80.8868,
      "eval_samples_per_second": 48.352,
      "eval_steps_per_second": 12.091,
      "step": 60000
    },
    {
      "epoch": 0.61,
      "grad_norm": 10.0,
      "learning_rate": 3.6129854561907786e-06,
      "loss": 0.7569,
      "step": 61000
    },
    {
      "epoch": 0.62,
      "grad_norm": 9.125,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 0.7549,
      "step": 62000
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.8032099604606628,
      "eval_runtime": 80.2623,
      "eval_samples_per_second": 48.728,
      "eval_steps_per_second": 12.185,
      "step": 62000
    },
    {
      "epoch": 0.63,
      "grad_norm": 9.375,
      "learning_rate": 3.298534127791785e-06,
      "loss": 0.7605,
      "step": 63000
    },
    {
      "epoch": 0.64,
      "grad_norm": 9.0625,
      "learning_rate": 3.1440137554088957e-06,
      "loss": 0.759,
      "step": 64000
    },
    {
      "epoch": 0.64,
      "eval_loss": 0.8016828894615173,
      "eval_runtime": 79.8549,
      "eval_samples_per_second": 48.976,
      "eval_steps_per_second": 12.247,
      "step": 64000
    },
    {
      "epoch": 0.65,
      "grad_norm": 13.1875,
      "learning_rate": 2.991522876735154e-06,
      "loss": 0.7626,
      "step": 65000
    },
    {
      "epoch": 0.66,
      "grad_norm": 9.1875,
      "learning_rate": 2.8412282383075362e-06,
      "loss": 0.7597,
      "step": 66000
    },
    {
      "epoch": 0.66,
      "eval_loss": 0.8016167879104614,
      "eval_runtime": 80.2412,
      "eval_samples_per_second": 48.741,
      "eval_steps_per_second": 12.188,
      "step": 66000
    },
    {
      "epoch": 0.67,
      "grad_norm": 9.375,
      "learning_rate": 2.693294185106562e-06,
      "loss": 0.7626,
      "step": 67000
    },
    {
      "epoch": 0.68,
      "grad_norm": 9.75,
      "learning_rate": 2.5478824808474613e-06,
      "loss": 0.7612,
      "step": 68000
    },
    {
      "epoch": 0.68,
      "eval_loss": 0.8017856478691101,
      "eval_runtime": 80.53,
      "eval_samples_per_second": 48.566,
      "eval_steps_per_second": 12.145,
      "step": 68000
    },
    {
      "epoch": 0.69,
      "grad_norm": 9.0,
      "learning_rate": 2.4051521310939258e-06,
      "loss": 0.7659,
      "step": 69000
    },
    {
      "epoch": 0.7,
      "grad_norm": 8.9375,
      "learning_rate": 2.265259209387867e-06,
      "loss": 0.7643,
      "step": 70000
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.8011899590492249,
      "eval_runtime": 80.4265,
      "eval_samples_per_second": 48.628,
      "eval_steps_per_second": 12.16,
      "step": 70000
    },
    {
      "epoch": 0.71,
      "grad_norm": 9.5625,
      "learning_rate": 2.1283566865852824e-06,
      "loss": 0.7658,
      "step": 71000
    },
    {
      "epoch": 0.72,
      "grad_norm": 9.5625,
      "learning_rate": 1.9945942635848745e-06,
      "loss": 0.768,
      "step": 72000
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.8010539412498474,
      "eval_runtime": 79.2118,
      "eval_samples_per_second": 49.374,
      "eval_steps_per_second": 12.347,
      "step": 72000
    },
    {
      "epoch": 0.73,
      "grad_norm": 10.3125,
      "learning_rate": 1.864118207632315e-06,
      "loss": 0.7683,
      "step": 73000
    },
    {
      "epoch": 0.74,
      "grad_norm": 9.375,
      "learning_rate": 1.7370711923791567e-06,
      "loss": 0.7669,
      "step": 74000
    },
    {
      "epoch": 0.74,
      "eval_loss": 0.8013535737991333,
      "eval_runtime": 79.7314,
      "eval_samples_per_second": 49.052,
      "eval_steps_per_second": 12.266,
      "step": 74000
    },
    {
      "epoch": 0.75,
      "grad_norm": 8.6875,
      "learning_rate": 1.6135921418712959e-06,
      "loss": 0.7683,
      "step": 75000
    },
    {
      "epoch": 0.76,
      "grad_norm": 9.3125,
      "learning_rate": 1.4938160786375571e-06,
      "loss": 0.7728,
      "step": 76000
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.8007571697235107,
      "eval_runtime": 79.1576,
      "eval_samples_per_second": 49.408,
      "eval_steps_per_second": 12.355,
      "step": 76000
    },
    {
      "epoch": 0.77,
      "grad_norm": 9.375,
      "learning_rate": 1.3778739760445552e-06,
      "loss": 0.7729,
      "step": 77000
    },
    {
      "epoch": 0.78,
      "grad_norm": 10.0625,
      "learning_rate": 1.2658926150792321e-06,
      "loss": 0.7728,
      "step": 78000
    },
    {
      "epoch": 0.78,
      "eval_loss": 0.8001759052276611,
      "eval_runtime": 79.1337,
      "eval_samples_per_second": 49.423,
      "eval_steps_per_second": 12.359,
      "step": 78000
    },
    {
      "epoch": 0.79,
      "grad_norm": 10.6875,
      "learning_rate": 1.157994445715706e-06,
      "loss": 0.771,
      "step": 79000
    },
    {
      "epoch": 0.8,
      "grad_norm": 9.4375,
      "learning_rate": 1.0542974530180327e-06,
      "loss": 0.7661,
      "step": 80000
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.8007293343544006,
      "eval_runtime": 79.1754,
      "eval_samples_per_second": 49.397,
      "eval_steps_per_second": 12.352,
      "step": 80000
    },
    {
      "epoch": 0.81,
      "grad_norm": 9.375,
      "learning_rate": 9.549150281252633e-07,
      "loss": 0.7593,
      "step": 81000
    },
    {
      "epoch": 0.82,
      "grad_norm": 9.4375,
      "learning_rate": 8.599558442598998e-07,
      "loss": 0.7505,
      "step": 82000
    },
    {
      "epoch": 0.82,
      "eval_loss": 0.8015545606613159,
      "eval_runtime": 79.021,
      "eval_samples_per_second": 49.493,
      "eval_steps_per_second": 12.376,
      "step": 82000
    },
    {
      "epoch": 0.83,
      "grad_norm": 9.6875,
      "learning_rate": 7.695237378953224e-07,
      "loss": 0.7452,
      "step": 83000
    },
    {
      "epoch": 0.84,
      "grad_norm": 9.125,
      "learning_rate": 6.837175952121305e-07,
      "loss": 0.7529,
      "step": 84000
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.8017655611038208,
      "eval_runtime": 79.164,
      "eval_samples_per_second": 49.404,
      "eval_steps_per_second": 12.354,
      "step": 84000
    },
    {
      "epoch": 0.85,
      "grad_norm": 8.9375,
      "learning_rate": 6.026312439675553e-07,
      "loss": 0.7481,
      "step": 85000
    },
    {
      "epoch": 0.86,
      "grad_norm": 9.625,
      "learning_rate": 5.263533508961827e-07,
      "loss": 0.7542,
      "step": 86000
    },
    {
      "epoch": 0.86,
      "eval_loss": 0.8010063767433167,
      "eval_runtime": 79.0232,
      "eval_samples_per_second": 49.492,
      "eval_steps_per_second": 12.376,
      "step": 86000
    },
    {
      "epoch": 0.87,
      "grad_norm": 10.25,
      "learning_rate": 4.549673247541875e-07,
      "loss": 0.7531,
      "step": 87000
    },
    {
      "epoch": 0.88,
      "grad_norm": 10.5,
      "learning_rate": 3.885512251130763e-07,
      "loss": 0.7531,
      "step": 88000
    },
    {
      "epoch": 0.88,
      "eval_loss": 0.8015376329421997,
      "eval_runtime": 79.1757,
      "eval_samples_per_second": 49.396,
      "eval_steps_per_second": 12.352,
      "step": 88000
    },
    {
      "epoch": 0.89,
      "grad_norm": 10.0625,
      "learning_rate": 3.271776770026963e-07,
      "loss": 0.7566,
      "step": 89000
    },
    {
      "epoch": 0.9,
      "grad_norm": 9.5625,
      "learning_rate": 2.7091379149682683e-07,
      "loss": 0.7601,
      "step": 90000
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.8013355135917664,
      "eval_runtime": 79.4846,
      "eval_samples_per_second": 49.205,
      "eval_steps_per_second": 12.304,
      "step": 90000
    },
    {
      "epoch": 0.91,
      "grad_norm": 9.5,
      "learning_rate": 2.198210923282118e-07,
      "loss": 0.7581,
      "step": 91000
    },
    {
      "epoch": 0.92,
      "grad_norm": 9.125,
      "learning_rate": 1.7395544861325718e-07,
      "loss": 0.7569,
      "step": 92000
    },
    {
      "epoch": 0.92,
      "eval_loss": 0.8014345765113831,
      "eval_runtime": 79.6557,
      "eval_samples_per_second": 49.099,
      "eval_steps_per_second": 12.278,
      "step": 92000
    },
    {
      "epoch": 0.93,
      "grad_norm": 9.8125,
      "learning_rate": 1.333670137599713e-07,
      "loss": 0.7599,
      "step": 93000
    },
    {
      "epoch": 0.94,
      "grad_norm": 9.125,
      "learning_rate": 9.810017062595322e-08,
      "loss": 0.7617,
      "step": 94000
    },
    {
      "epoch": 0.94,
      "eval_loss": 0.8014338612556458,
      "eval_runtime": 79.0466,
      "eval_samples_per_second": 49.477,
      "eval_steps_per_second": 12.372,
      "step": 94000
    },
    {
      "epoch": 0.95,
      "grad_norm": 10.1875,
      "learning_rate": 6.819348298638839e-08,
      "loss": 0.7641,
      "step": 95000
    },
    {
      "epoch": 0.96,
      "grad_norm": 9.8125,
      "learning_rate": 4.367965336512403e-08,
      "loss": 0.766,
      "step": 96000
    },
    {
      "epoch": 0.96,
      "eval_loss": 0.8010128140449524,
      "eval_runtime": 79.5861,
      "eval_samples_per_second": 49.142,
      "eval_steps_per_second": 12.289,
      "step": 96000
    },
    {
      "epoch": 0.97,
      "grad_norm": 9.75,
      "learning_rate": 2.4585487274942922e-08,
      "loss": 0.7646,
      "step": 97000
    },
    {
      "epoch": 0.98,
      "grad_norm": 10.0625,
      "learning_rate": 1.0931863906127327e-08,
      "loss": 0.7673,
      "step": 98000
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.8010921478271484,
      "eval_runtime": 79.6671,
      "eval_samples_per_second": 49.092,
      "eval_steps_per_second": 12.276,
      "step": 98000
    },
    {
      "epoch": 0.99,
      "grad_norm": 10.125,
      "learning_rate": 2.7337132953697555e-09,
      "loss": 0.7674,
      "step": 99000
    },
    {
      "epoch": 1.0,
      "grad_norm": 9.5625,
      "learning_rate": 0.0,
      "loss": 0.7658,
      "step": 100000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8011825680732727,
      "eval_runtime": 79.4501,
      "eval_samples_per_second": 49.226,
      "eval_steps_per_second": 12.31,
      "step": 100000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 100000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.68882057478144e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}