{
  "best_metric": 1.9801934740826255e-06,
  "best_model_checkpoint": "Models/t5-base-class-gen/checkpoint-10400",
  "epoch": 4.9440298507462686,
  "eval_steps": 100,
  "global_step": 10600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04664179104477612,
      "grad_norm": 1.2865939140319824,
      "learning_rate": 3.9630597014925376e-05,
      "loss": 0.437,
      "step": 100
    },
    {
      "epoch": 0.04664179104477612,
      "eval_loss": 0.020251411944627762,
      "eval_runtime": 0.1587,
      "eval_samples_per_second": 189.027,
      "eval_steps_per_second": 25.204,
      "step": 100
    },
    {
      "epoch": 0.09328358208955224,
      "grad_norm": 0.69548100233078,
      "learning_rate": 3.925746268656717e-05,
      "loss": 0.0333,
      "step": 200
    },
    {
      "epoch": 0.09328358208955224,
      "eval_loss": 0.0021759807132184505,
      "eval_runtime": 0.1592,
      "eval_samples_per_second": 188.446,
      "eval_steps_per_second": 25.126,
      "step": 200
    },
    {
      "epoch": 0.13992537313432835,
      "grad_norm": 0.07560934871435165,
      "learning_rate": 3.888432835820896e-05,
      "loss": 0.0203,
      "step": 300
    },
    {
      "epoch": 0.13992537313432835,
      "eval_loss": 0.004563063848763704,
      "eval_runtime": 0.1618,
      "eval_samples_per_second": 185.371,
      "eval_steps_per_second": 24.716,
      "step": 300
    },
    {
      "epoch": 0.1865671641791045,
      "grad_norm": 0.646364688873291,
      "learning_rate": 3.851119402985075e-05,
      "loss": 0.0144,
      "step": 400
    },
    {
      "epoch": 0.1865671641791045,
      "eval_loss": 0.00161658669821918,
      "eval_runtime": 0.1567,
      "eval_samples_per_second": 191.49,
      "eval_steps_per_second": 25.532,
      "step": 400
    },
    {
      "epoch": 0.2332089552238806,
      "grad_norm": 0.30101820826530457,
      "learning_rate": 3.813805970149254e-05,
      "loss": 0.0079,
      "step": 500
    },
    {
      "epoch": 0.2332089552238806,
      "eval_loss": 0.0010551114100962877,
      "eval_runtime": 0.1559,
      "eval_samples_per_second": 192.377,
      "eval_steps_per_second": 25.65,
      "step": 500
    },
    {
      "epoch": 0.2798507462686567,
      "grad_norm": 0.5621537566184998,
      "learning_rate": 3.7764925373134334e-05,
      "loss": 0.0076,
      "step": 600
    },
    {
      "epoch": 0.2798507462686567,
      "eval_loss": 0.0002234852290712297,
      "eval_runtime": 0.1571,
      "eval_samples_per_second": 191.018,
      "eval_steps_per_second": 25.469,
      "step": 600
    },
    {
      "epoch": 0.32649253731343286,
      "grad_norm": 0.10140960663557053,
      "learning_rate": 3.7391791044776125e-05,
      "loss": 0.0092,
      "step": 700
    },
    {
      "epoch": 0.32649253731343286,
      "eval_loss": 0.00010653692879714072,
      "eval_runtime": 0.1572,
      "eval_samples_per_second": 190.855,
      "eval_steps_per_second": 25.447,
      "step": 700
    },
    {
      "epoch": 0.373134328358209,
      "grad_norm": 0.02672586776316166,
      "learning_rate": 3.701865671641792e-05,
      "loss": 0.0053,
      "step": 800
    },
    {
      "epoch": 0.373134328358209,
      "eval_loss": 0.00011367371189408004,
      "eval_runtime": 0.1569,
      "eval_samples_per_second": 191.155,
      "eval_steps_per_second": 25.487,
      "step": 800
    },
    {
      "epoch": 0.4197761194029851,
      "grad_norm": 0.09280095249414444,
      "learning_rate": 3.66455223880597e-05,
      "loss": 0.0054,
      "step": 900
    },
    {
      "epoch": 0.4197761194029851,
      "eval_loss": 0.00013640630641020834,
      "eval_runtime": 0.1588,
      "eval_samples_per_second": 188.886,
      "eval_steps_per_second": 25.185,
      "step": 900
    },
    {
      "epoch": 0.4664179104477612,
      "grad_norm": 0.034318458288908005,
      "learning_rate": 3.62723880597015e-05,
      "loss": 0.0067,
      "step": 1000
    },
    {
      "epoch": 0.4664179104477612,
      "eval_loss": 7.585491403006017e-05,
      "eval_runtime": 0.1557,
      "eval_samples_per_second": 192.664,
      "eval_steps_per_second": 25.689,
      "step": 1000
    },
    {
      "epoch": 0.5130597014925373,
      "grad_norm": 0.018935877829790115,
      "learning_rate": 3.5899253731343285e-05,
      "loss": 0.0035,
      "step": 1100
    },
    {
      "epoch": 0.5130597014925373,
      "eval_loss": 4.489553612074815e-05,
      "eval_runtime": 0.1574,
      "eval_samples_per_second": 190.631,
      "eval_steps_per_second": 25.418,
      "step": 1100
    },
    {
      "epoch": 0.5597014925373134,
      "grad_norm": 0.8913596868515015,
      "learning_rate": 3.5526119402985077e-05,
      "loss": 0.0056,
      "step": 1200
    },
    {
      "epoch": 0.5597014925373134,
      "eval_loss": 4.118124343222007e-05,
      "eval_runtime": 0.1589,
      "eval_samples_per_second": 188.831,
      "eval_steps_per_second": 25.177,
      "step": 1200
    },
    {
      "epoch": 0.6063432835820896,
      "grad_norm": 0.017305633053183556,
      "learning_rate": 3.515298507462687e-05,
      "loss": 0.0041,
      "step": 1300
    },
    {
      "epoch": 0.6063432835820896,
      "eval_loss": 4.481070573092438e-05,
      "eval_runtime": 0.1552,
      "eval_samples_per_second": 193.329,
      "eval_steps_per_second": 25.777,
      "step": 1300
    },
    {
      "epoch": 0.6529850746268657,
      "grad_norm": 0.009413465857505798,
      "learning_rate": 3.477985074626866e-05,
      "loss": 0.0029,
      "step": 1400
    },
    {
      "epoch": 0.6529850746268657,
      "eval_loss": 2.9594228180940263e-05,
      "eval_runtime": 0.1654,
      "eval_samples_per_second": 181.422,
      "eval_steps_per_second": 24.19,
      "step": 1400
    },
    {
      "epoch": 0.6996268656716418,
      "grad_norm": 0.004255557898432016,
      "learning_rate": 3.440671641791045e-05,
      "loss": 0.0028,
      "step": 1500
    },
    {
      "epoch": 0.6996268656716418,
      "eval_loss": 3.3944801543839276e-05,
      "eval_runtime": 0.1543,
      "eval_samples_per_second": 194.435,
      "eval_steps_per_second": 25.925,
      "step": 1500
    },
    {
      "epoch": 0.746268656716418,
      "grad_norm": 0.6249216198921204,
      "learning_rate": 3.403358208955224e-05,
      "loss": 0.003,
      "step": 1600
    },
    {
      "epoch": 0.746268656716418,
      "eval_loss": 3.595968155423179e-05,
      "eval_runtime": 0.1621,
      "eval_samples_per_second": 185.072,
      "eval_steps_per_second": 24.676,
      "step": 1600
    },
    {
      "epoch": 0.792910447761194,
      "grad_norm": 0.014389215037226677,
      "learning_rate": 3.3660447761194034e-05,
      "loss": 0.002,
      "step": 1700
    },
    {
      "epoch": 0.792910447761194,
      "eval_loss": 3.822403959929943e-05,
      "eval_runtime": 0.1593,
      "eval_samples_per_second": 188.284,
      "eval_steps_per_second": 25.105,
      "step": 1700
    },
    {
      "epoch": 0.8395522388059702,
      "grad_norm": 0.1081342026591301,
      "learning_rate": 3.328731343283582e-05,
      "loss": 0.0039,
      "step": 1800
    },
    {
      "epoch": 0.8395522388059702,
      "eval_loss": 3.7738984246971086e-05,
      "eval_runtime": 0.1558,
      "eval_samples_per_second": 192.583,
      "eval_steps_per_second": 25.678,
      "step": 1800
    },
    {
      "epoch": 0.8861940298507462,
      "grad_norm": 1.123632550239563,
      "learning_rate": 3.291417910447762e-05,
      "loss": 0.0035,
      "step": 1900
    },
    {
      "epoch": 0.8861940298507462,
      "eval_loss": 2.4694763851584867e-05,
      "eval_runtime": 0.157,
      "eval_samples_per_second": 191.065,
      "eval_steps_per_second": 25.475,
      "step": 1900
    },
    {
      "epoch": 0.9328358208955224,
      "grad_norm": 0.00756313418969512,
      "learning_rate": 3.25410447761194e-05,
      "loss": 0.0027,
      "step": 2000
    },
    {
      "epoch": 0.9328358208955224,
      "eval_loss": 5.045727448305115e-05,
      "eval_runtime": 0.1664,
      "eval_samples_per_second": 180.272,
      "eval_steps_per_second": 24.036,
      "step": 2000
    },
    {
      "epoch": 0.9794776119402985,
      "grad_norm": 0.0035852426663041115,
      "learning_rate": 3.2167910447761194e-05,
      "loss": 0.0026,
      "step": 2100
    },
    {
      "epoch": 0.9794776119402985,
      "eval_loss": 2.137782212230377e-05,
      "eval_runtime": 0.1552,
      "eval_samples_per_second": 193.332,
      "eval_steps_per_second": 25.778,
      "step": 2100
    },
    {
      "epoch": 1.0261194029850746,
      "grad_norm": 0.01669035293161869,
      "learning_rate": 3.1794776119402986e-05,
      "loss": 0.0008,
      "step": 2200
    },
    {
      "epoch": 1.0261194029850746,
      "eval_loss": 2.3261545720743015e-05,
      "eval_runtime": 0.1554,
      "eval_samples_per_second": 193.073,
      "eval_steps_per_second": 25.743,
      "step": 2200
    },
    {
      "epoch": 1.0727611940298507,
      "grad_norm": 0.002784071722999215,
      "learning_rate": 3.142164179104478e-05,
      "loss": 0.0016,
      "step": 2300
    },
    {
      "epoch": 1.0727611940298507,
      "eval_loss": 1.877884460554924e-05,
      "eval_runtime": 0.1597,
      "eval_samples_per_second": 187.795,
      "eval_steps_per_second": 25.039,
      "step": 2300
    },
    {
      "epoch": 1.1194029850746268,
      "grad_norm": 0.010812056250870228,
      "learning_rate": 3.104850746268657e-05,
      "loss": 0.0013,
      "step": 2400
    },
    {
      "epoch": 1.1194029850746268,
      "eval_loss": 2.032700831477996e-05,
      "eval_runtime": 0.157,
      "eval_samples_per_second": 191.1,
      "eval_steps_per_second": 25.48,
      "step": 2400
    },
    {
      "epoch": 1.166044776119403,
      "grad_norm": 0.08882497251033783,
      "learning_rate": 3.067537313432836e-05,
      "loss": 0.0023,
      "step": 2500
    },
    {
      "epoch": 1.166044776119403,
      "eval_loss": 2.7997491997666657e-05,
      "eval_runtime": 0.1583,
      "eval_samples_per_second": 189.547,
      "eval_steps_per_second": 25.273,
      "step": 2500
    },
    {
      "epoch": 1.212686567164179,
      "grad_norm": 0.5638116002082825,
      "learning_rate": 3.030223880597015e-05,
      "loss": 0.0014,
      "step": 2600
    },
    {
      "epoch": 1.212686567164179,
      "eval_loss": 1.3365392987907398e-05,
      "eval_runtime": 0.1568,
      "eval_samples_per_second": 191.368,
      "eval_steps_per_second": 25.516,
      "step": 2600
    },
    {
      "epoch": 1.2593283582089552,
      "grad_norm": 0.6982096433639526,
      "learning_rate": 2.9929104477611944e-05,
      "loss": 0.0009,
      "step": 2700
    },
    {
      "epoch": 1.2593283582089552,
      "eval_loss": 1.1131261089758482e-05,
      "eval_runtime": 0.1571,
      "eval_samples_per_second": 191.001,
      "eval_steps_per_second": 25.467,
      "step": 2700
    },
    {
      "epoch": 1.3059701492537314,
      "grad_norm": 0.2126922756433487,
      "learning_rate": 2.9555970149253732e-05,
      "loss": 0.0007,
      "step": 2800
    },
    {
      "epoch": 1.3059701492537314,
      "eval_loss": 1.0051960998680443e-05,
      "eval_runtime": 0.1557,
      "eval_samples_per_second": 192.64,
      "eval_steps_per_second": 25.685,
      "step": 2800
    },
    {
      "epoch": 1.3526119402985075,
      "grad_norm": 0.07693888247013092,
      "learning_rate": 2.9182835820895527e-05,
      "loss": 0.0011,
      "step": 2900
    },
    {
      "epoch": 1.3526119402985075,
      "eval_loss": 8.506859558110591e-06,
      "eval_runtime": 0.1606,
      "eval_samples_per_second": 186.795,
      "eval_steps_per_second": 24.906,
      "step": 2900
    },
    {
      "epoch": 1.3992537313432836,
      "grad_norm": 0.005555067211389542,
      "learning_rate": 2.8809701492537315e-05,
      "loss": 0.0012,
      "step": 3000
    },
    {
      "epoch": 1.3992537313432836,
      "eval_loss": 1.0461569218023214e-05,
      "eval_runtime": 0.1562,
      "eval_samples_per_second": 192.09,
      "eval_steps_per_second": 25.612,
      "step": 3000
    },
    {
      "epoch": 1.4458955223880596,
      "grad_norm": 0.15883490443229675,
      "learning_rate": 2.8436567164179106e-05,
      "loss": 0.0015,
      "step": 3100
    },
    {
      "epoch": 1.4458955223880596,
      "eval_loss": 9.19290141609963e-06,
      "eval_runtime": 0.1574,
      "eval_samples_per_second": 190.61,
      "eval_steps_per_second": 25.415,
      "step": 3100
    },
    {
      "epoch": 1.4925373134328357,
      "grad_norm": 0.0036452861968427896,
      "learning_rate": 2.806716417910448e-05,
      "loss": 0.0017,
      "step": 3200
    },
    {
      "epoch": 1.4925373134328357,
      "eval_loss": 7.644109246029984e-06,
      "eval_runtime": 0.1573,
      "eval_samples_per_second": 190.724,
      "eval_steps_per_second": 25.43,
      "step": 3200
    },
    {
      "epoch": 1.539179104477612,
      "grad_norm": 0.1588873416185379,
      "learning_rate": 2.769402985074627e-05,
      "loss": 0.0009,
      "step": 3300
    },
    {
      "epoch": 1.539179104477612,
      "eval_loss": 9.155748557532206e-06,
      "eval_runtime": 0.1602,
      "eval_samples_per_second": 187.322,
      "eval_steps_per_second": 24.976,
      "step": 3300
    },
    {
      "epoch": 1.585820895522388,
      "grad_norm": 0.004792116116732359,
      "learning_rate": 2.7320895522388062e-05,
      "loss": 0.0018,
      "step": 3400
    },
    {
      "epoch": 1.585820895522388,
      "eval_loss": 8.091868039628025e-06,
      "eval_runtime": 0.1578,
      "eval_samples_per_second": 190.11,
      "eval_steps_per_second": 25.348,
      "step": 3400
    },
    {
      "epoch": 1.6324626865671643,
      "grad_norm": 2.761561870574951,
      "learning_rate": 2.694776119402985e-05,
      "loss": 0.0011,
      "step": 3500
    },
    {
      "epoch": 1.6324626865671643,
      "eval_loss": 8.603728929301724e-06,
      "eval_runtime": 0.1552,
      "eval_samples_per_second": 193.265,
      "eval_steps_per_second": 25.769,
      "step": 3500
    },
    {
      "epoch": 1.6791044776119404,
      "grad_norm": 0.001969451317563653,
      "learning_rate": 2.6574626865671645e-05,
      "loss": 0.0012,
      "step": 3600
    },
    {
      "epoch": 1.6791044776119404,
      "eval_loss": 8.435232302872464e-06,
      "eval_runtime": 0.1553,
      "eval_samples_per_second": 193.184,
      "eval_steps_per_second": 25.758,
      "step": 3600
    },
    {
      "epoch": 1.7257462686567164,
      "grad_norm": 0.005350825376808643,
      "learning_rate": 2.6201492537313434e-05,
      "loss": 0.0008,
      "step": 3700
    },
    {
      "epoch": 1.7257462686567164,
      "eval_loss": 7.432493021042319e-06,
      "eval_runtime": 0.1563,
      "eval_samples_per_second": 191.964,
      "eval_steps_per_second": 25.595,
      "step": 3700
    },
    {
      "epoch": 1.7723880597014925,
      "grad_norm": 0.004110433626919985,
      "learning_rate": 2.5828358208955225e-05,
      "loss": 0.0014,
      "step": 3800
    },
    {
      "epoch": 1.7723880597014925,
      "eval_loss": 7.303169240913121e-06,
      "eval_runtime": 0.1569,
      "eval_samples_per_second": 191.177,
      "eval_steps_per_second": 25.49,
      "step": 3800
    },
    {
      "epoch": 1.8190298507462686,
      "grad_norm": 0.0039463951252400875,
      "learning_rate": 2.5455223880597017e-05,
      "loss": 0.0015,
      "step": 3900
    },
    {
      "epoch": 1.8190298507462686,
      "eval_loss": 9.76303817878943e-06,
      "eval_runtime": 0.1571,
      "eval_samples_per_second": 190.973,
      "eval_steps_per_second": 25.463,
      "step": 3900
    },
    {
      "epoch": 1.8656716417910446,
      "grad_norm": 0.004805909004062414,
      "learning_rate": 2.508208955223881e-05,
      "loss": 0.0012,
      "step": 4000
    },
    {
      "epoch": 1.8656716417910446,
      "eval_loss": 8.14036320662126e-06,
      "eval_runtime": 0.157,
      "eval_samples_per_second": 191.027,
      "eval_steps_per_second": 25.47,
      "step": 4000
    },
    {
      "epoch": 1.912313432835821,
      "grad_norm": 0.0031389566138386726,
      "learning_rate": 2.4708955223880597e-05,
      "loss": 0.0004,
      "step": 4100
    },
    {
      "epoch": 1.912313432835821,
      "eval_loss": 6.821030183346011e-06,
      "eval_runtime": 0.1614,
      "eval_samples_per_second": 185.9,
      "eval_steps_per_second": 24.787,
      "step": 4100
    },
    {
      "epoch": 1.9589552238805972,
      "grad_norm": 0.0017104181461036205,
      "learning_rate": 2.433582089552239e-05,
      "loss": 0.0008,
      "step": 4200
    },
    {
      "epoch": 1.9589552238805972,
      "eval_loss": 7.834916686988436e-06,
      "eval_runtime": 0.1615,
      "eval_samples_per_second": 185.757,
      "eval_steps_per_second": 24.768,
      "step": 4200
    },
    {
      "epoch": 2.0055970149253732,
      "grad_norm": 0.013449718244373798,
      "learning_rate": 2.396268656716418e-05,
      "loss": 0.001,
      "step": 4300
    },
    {
      "epoch": 2.0055970149253732,
      "eval_loss": 7.582441867270973e-06,
      "eval_runtime": 0.1686,
      "eval_samples_per_second": 177.913,
      "eval_steps_per_second": 23.722,
      "step": 4300
    },
    {
      "epoch": 2.0522388059701493,
      "grad_norm": 0.0017907052533701062,
      "learning_rate": 2.358955223880597e-05,
      "loss": 0.0007,
      "step": 4400
    },
    {
      "epoch": 2.0522388059701493,
      "eval_loss": 6.949214366613887e-06,
      "eval_runtime": 0.1537,
      "eval_samples_per_second": 195.234,
      "eval_steps_per_second": 26.031,
      "step": 4400
    },
    {
      "epoch": 2.0988805970149254,
      "grad_norm": 0.001830141874961555,
      "learning_rate": 2.3216417910447763e-05,
      "loss": 0.0004,
      "step": 4500
    },
    {
      "epoch": 2.0988805970149254,
      "eval_loss": 7.074564564391039e-06,
      "eval_runtime": 0.1577,
      "eval_samples_per_second": 190.227,
      "eval_steps_per_second": 25.364,
      "step": 4500
    },
    {
      "epoch": 2.1455223880597014,
      "grad_norm": 0.0030846677254885435,
      "learning_rate": 2.2843283582089555e-05,
      "loss": 0.0006,
      "step": 4600
    },
    {
      "epoch": 2.1455223880597014,
      "eval_loss": 6.291250883805333e-06,
      "eval_runtime": 0.1566,
      "eval_samples_per_second": 191.563,
      "eval_steps_per_second": 25.542,
      "step": 4600
    },
    {
      "epoch": 2.1921641791044775,
      "grad_norm": 0.24780245125293732,
      "learning_rate": 2.2470149253731343e-05,
      "loss": 0.0003,
      "step": 4700
    },
    {
      "epoch": 2.1921641791044775,
      "eval_loss": 6.1347077462414745e-06,
      "eval_runtime": 0.1575,
      "eval_samples_per_second": 190.525,
      "eval_steps_per_second": 25.403,
      "step": 4700
    },
    {
      "epoch": 2.2388059701492535,
      "grad_norm": 0.006296923384070396,
      "learning_rate": 2.2097014925373138e-05,
      "loss": 0.0003,
      "step": 4800
    },
    {
      "epoch": 2.2388059701492535,
      "eval_loss": 5.517637873708736e-06,
      "eval_runtime": 0.1579,
      "eval_samples_per_second": 189.97,
      "eval_steps_per_second": 25.329,
      "step": 4800
    },
    {
      "epoch": 2.28544776119403,
      "grad_norm": 0.0017776070162653923,
      "learning_rate": 2.1723880597014926e-05,
      "loss": 0.0007,
      "step": 4900
    },
    {
      "epoch": 2.28544776119403,
      "eval_loss": 5.534998763323529e-06,
      "eval_runtime": 0.158,
      "eval_samples_per_second": 189.867,
      "eval_steps_per_second": 25.316,
      "step": 4900
    },
    {
      "epoch": 2.332089552238806,
      "grad_norm": 0.0026376782916486263,
      "learning_rate": 2.1350746268656717e-05,
      "loss": 0.0004,
      "step": 5000
    },
    {
      "epoch": 2.332089552238806,
      "eval_loss": 5.3646053856937215e-06,
      "eval_runtime": 0.158,
      "eval_samples_per_second": 189.825,
      "eval_steps_per_second": 25.31,
      "step": 5000
    },
    {
      "epoch": 2.378731343283582,
      "grad_norm": 0.008729691617190838,
      "learning_rate": 2.097761194029851e-05,
      "loss": 0.0009,
      "step": 5100
    },
    {
      "epoch": 2.378731343283582,
      "eval_loss": 5.699926987290382e-06,
      "eval_runtime": 0.1586,
      "eval_samples_per_second": 189.183,
      "eval_steps_per_second": 25.224,
      "step": 5100
    },
    {
      "epoch": 2.425373134328358,
      "grad_norm": 0.0009663441451266408,
      "learning_rate": 2.06044776119403e-05,
      "loss": 0.0002,
      "step": 5200
    },
    {
      "epoch": 2.425373134328358,
      "eval_loss": 5.018339379603276e-06,
      "eval_runtime": 0.1609,
      "eval_samples_per_second": 186.434,
      "eval_steps_per_second": 24.858,
      "step": 5200
    },
    {
      "epoch": 2.4720149253731343,
      "grad_norm": 0.0009370720363222063,
      "learning_rate": 2.023134328358209e-05,
      "loss": 0.0003,
      "step": 5300
    },
    {
      "epoch": 2.4720149253731343,
      "eval_loss": 4.319118488638196e-06,
      "eval_runtime": 0.1594,
      "eval_samples_per_second": 188.258,
      "eval_steps_per_second": 25.101,
      "step": 5300
    },
    {
      "epoch": 2.5186567164179103,
      "grad_norm": 0.00313656241632998,
      "learning_rate": 1.9858208955223884e-05,
      "loss": 0.0008,
      "step": 5400
    },
    {
      "epoch": 2.5186567164179103,
      "eval_loss": 4.456574970390648e-06,
      "eval_runtime": 0.1551,
      "eval_samples_per_second": 193.476,
      "eval_steps_per_second": 25.797,
      "step": 5400
    },
    {
      "epoch": 2.5652985074626864,
      "grad_norm": 0.0011074623325839639,
      "learning_rate": 1.9485074626865675e-05,
      "loss": 0.001,
      "step": 5500
    },
    {
      "epoch": 2.5652985074626864,
      "eval_loss": 3.924767952412367e-06,
      "eval_runtime": 0.1544,
      "eval_samples_per_second": 194.285,
      "eval_steps_per_second": 25.905,
      "step": 5500
    },
    {
      "epoch": 2.611940298507463,
      "grad_norm": 0.02510717324912548,
      "learning_rate": 1.9111940298507467e-05,
      "loss": 0.0003,
      "step": 5600
    },
    {
      "epoch": 2.611940298507463,
      "eval_loss": 4.046037247462664e-06,
      "eval_runtime": 0.1553,
      "eval_samples_per_second": 193.136,
      "eval_steps_per_second": 25.752,
      "step": 5600
    },
    {
      "epoch": 2.658582089552239,
      "grad_norm": 0.023706955835223198,
      "learning_rate": 1.8738805970149255e-05,
      "loss": 0.0003,
      "step": 5700
    },
    {
      "epoch": 2.658582089552239,
      "eval_loss": 4.111709131393582e-06,
      "eval_runtime": 0.1558,
      "eval_samples_per_second": 192.543,
      "eval_steps_per_second": 25.672,
      "step": 5700
    },
    {
      "epoch": 2.705223880597015,
      "grad_norm": 0.004163182340562344,
      "learning_rate": 1.8365671641791047e-05,
      "loss": 0.0006,
      "step": 5800
    },
    {
      "epoch": 2.705223880597015,
      "eval_loss": 4.339735369285336e-06,
      "eval_runtime": 0.1607,
      "eval_samples_per_second": 186.695,
      "eval_steps_per_second": 24.893,
      "step": 5800
    },
    {
      "epoch": 2.751865671641791,
      "grad_norm": 0.002963052364066243,
      "learning_rate": 1.7992537313432835e-05,
      "loss": 0.0005,
      "step": 5900
    },
    {
      "epoch": 2.751865671641791,
      "eval_loss": 3.9786968955013435e-06,
      "eval_runtime": 0.1575,
      "eval_samples_per_second": 190.477,
      "eval_steps_per_second": 25.397,
      "step": 5900
    },
    {
      "epoch": 2.798507462686567,
      "grad_norm": 0.045626550912857056,
      "learning_rate": 1.7619402985074627e-05,
      "loss": 0.0008,
      "step": 6000
    },
    {
      "epoch": 2.798507462686567,
      "eval_loss": 4.052158146805596e-06,
      "eval_runtime": 0.1579,
      "eval_samples_per_second": 189.965,
      "eval_steps_per_second": 25.329,
      "step": 6000
    },
    {
      "epoch": 2.845149253731343,
      "grad_norm": 0.008129669353365898,
      "learning_rate": 1.7246268656716418e-05,
      "loss": 0.0004,
      "step": 6100
    },
    {
      "epoch": 2.845149253731343,
      "eval_loss": 4.024254849355202e-06,
      "eval_runtime": 0.158,
      "eval_samples_per_second": 189.912,
      "eval_steps_per_second": 25.322,
      "step": 6100
    },
    {
      "epoch": 2.8917910447761193,
      "grad_norm": 0.002264247043058276,
      "learning_rate": 1.687313432835821e-05,
      "loss": 0.0007,
      "step": 6200
    },
    {
      "epoch": 2.8917910447761193,
      "eval_loss": 3.7680488276237156e-06,
      "eval_runtime": 0.1559,
      "eval_samples_per_second": 192.455,
      "eval_steps_per_second": 25.661,
      "step": 6200
    },
    {
      "epoch": 2.9384328358208958,
      "grad_norm": 0.013498159125447273,
      "learning_rate": 1.65e-05,
      "loss": 0.0002,
      "step": 6300
    },
    {
      "epoch": 2.9384328358208958,
      "eval_loss": 3.534241386660142e-06,
      "eval_runtime": 0.1562,
      "eval_samples_per_second": 192.066,
      "eval_steps_per_second": 25.609,
      "step": 6300
    },
    {
      "epoch": 2.9850746268656714,
      "grad_norm": 0.000992327812127769,
      "learning_rate": 1.6126865671641793e-05,
      "loss": 0.0009,
      "step": 6400
    },
    {
      "epoch": 2.9850746268656714,
      "eval_loss": 3.539219051162945e-06,
      "eval_runtime": 0.1581,
      "eval_samples_per_second": 189.704,
      "eval_steps_per_second": 25.294,
      "step": 6400
    },
    {
      "epoch": 3.031716417910448,
      "grad_norm": 0.002681414596736431,
      "learning_rate": 1.575373134328358e-05,
      "loss": 0.0003,
      "step": 6500
    },
    {
      "epoch": 3.031716417910448,
      "eval_loss": 3.239377292629797e-06,
      "eval_runtime": 0.1572,
      "eval_samples_per_second": 190.818,
      "eval_steps_per_second": 25.442,
      "step": 6500
    },
    {
      "epoch": 3.078358208955224,
      "grad_norm": 0.06693683564662933,
      "learning_rate": 1.5380597014925373e-05,
      "loss": 0.0003,
      "step": 6600
    },
    {
      "epoch": 3.078358208955224,
      "eval_loss": 3.1195118026516866e-06,
      "eval_runtime": 0.1554,
      "eval_samples_per_second": 192.996,
      "eval_steps_per_second": 25.733,
      "step": 6600
    },
    {
      "epoch": 3.125,
      "grad_norm": 0.007976911030709743,
      "learning_rate": 1.5007462686567164e-05,
      "loss": 0.0002,
      "step": 6700
    },
    {
      "epoch": 3.125,
      "eval_loss": 3.158217623422388e-06,
      "eval_runtime": 0.1613,
      "eval_samples_per_second": 186.004,
      "eval_steps_per_second": 24.801,
      "step": 6700
    },
    {
      "epoch": 3.171641791044776,
      "grad_norm": 0.0018902523443102837,
      "learning_rate": 1.4634328358208956e-05,
      "loss": 0.0002,
      "step": 6800
    },
    {
      "epoch": 3.171641791044776,
      "eval_loss": 3.020254553121049e-06,
      "eval_runtime": 0.1619,
      "eval_samples_per_second": 185.322,
      "eval_steps_per_second": 24.71,
      "step": 6800
    },
    {
      "epoch": 3.218283582089552,
      "grad_norm": 0.0031356920953840017,
      "learning_rate": 1.4261194029850747e-05,
      "loss": 0.0002,
      "step": 6900
    },
    {
      "epoch": 3.218283582089552,
      "eval_loss": 2.9624054604937555e-06,
      "eval_runtime": 0.1556,
      "eval_samples_per_second": 192.812,
      "eval_steps_per_second": 25.708,
      "step": 6900
    },
    {
      "epoch": 3.264925373134328,
      "grad_norm": 0.005576052237302065,
      "learning_rate": 1.3888059701492537e-05,
      "loss": 0.0005,
      "step": 7000
    },
    {
      "epoch": 3.264925373134328,
      "eval_loss": 2.9549014470831025e-06,
      "eval_runtime": 0.1578,
      "eval_samples_per_second": 190.15,
      "eval_steps_per_second": 25.353,
      "step": 7000
    },
    {
      "epoch": 3.3115671641791042,
      "grad_norm": 0.0018602815689519048,
      "learning_rate": 1.3514925373134329e-05,
      "loss": 0.0002,
      "step": 7100
    },
    {
      "epoch": 3.3115671641791042,
      "eval_loss": 2.8336680770735256e-06,
      "eval_runtime": 0.1552,
      "eval_samples_per_second": 193.246,
      "eval_steps_per_second": 25.766,
      "step": 7100
    },
    {
      "epoch": 3.3582089552238807,
      "grad_norm": 0.002030579838901758,
      "learning_rate": 1.314179104477612e-05,
      "loss": 0.0003,
      "step": 7200
    },
    {
      "epoch": 3.3582089552238807,
      "eval_loss": 2.7239029805059545e-06,
      "eval_runtime": 0.1624,
      "eval_samples_per_second": 184.688,
      "eval_steps_per_second": 24.625,
      "step": 7200
    },
    {
      "epoch": 3.404850746268657,
      "grad_norm": 0.027261830866336823,
      "learning_rate": 1.276865671641791e-05,
      "loss": 0.0003,
      "step": 7300
    },
    {
      "epoch": 3.404850746268657,
      "eval_loss": 2.706149643927347e-06,
      "eval_runtime": 0.1586,
      "eval_samples_per_second": 189.213,
      "eval_steps_per_second": 25.228,
      "step": 7300
    },
    {
      "epoch": 3.451492537313433,
      "grad_norm": 0.005287382751703262,
      "learning_rate": 1.2395522388059702e-05,
      "loss": 0.0002,
      "step": 7400
    },
    {
      "epoch": 3.451492537313433,
      "eval_loss": 2.5531167011649814e-06,
      "eval_runtime": 0.1543,
      "eval_samples_per_second": 194.469,
      "eval_steps_per_second": 25.929,
      "step": 7400
    },
    {
      "epoch": 3.498134328358209,
      "grad_norm": 0.0021771446336060762,
      "learning_rate": 1.2022388059701493e-05,
      "loss": 0.0003,
      "step": 7500
    },
    {
      "epoch": 3.498134328358209,
      "eval_loss": 2.8221681986906333e-06,
      "eval_runtime": 0.1601,
      "eval_samples_per_second": 187.36,
      "eval_steps_per_second": 24.981,
      "step": 7500
    },
    {
      "epoch": 3.544776119402985,
      "grad_norm": 0.0027294817846268415,
      "learning_rate": 1.1649253731343283e-05,
      "loss": 0.0003,
      "step": 7600
    },
    {
      "epoch": 3.544776119402985,
      "eval_loss": 2.784717707982054e-06,
      "eval_runtime": 0.1555,
      "eval_samples_per_second": 192.876,
      "eval_steps_per_second": 25.717,
      "step": 7600
    },
    {
      "epoch": 3.591417910447761,
      "grad_norm": 0.0013641875702887774,
      "learning_rate": 1.1276119402985075e-05,
      "loss": 0.0003,
      "step": 7700
    },
    {
      "epoch": 3.591417910447761,
      "eval_loss": 2.9222812827356393e-06,
      "eval_runtime": 0.1639,
      "eval_samples_per_second": 183.034,
      "eval_steps_per_second": 24.405,
      "step": 7700
    },
    {
      "epoch": 3.638059701492537,
      "grad_norm": 0.08378789573907852,
      "learning_rate": 1.0902985074626867e-05,
      "loss": 0.0002,
      "step": 7800
    },
    {
      "epoch": 3.638059701492537,
      "eval_loss": 2.9144475774955936e-06,
      "eval_runtime": 0.1585,
      "eval_samples_per_second": 189.266,
      "eval_steps_per_second": 25.235,
      "step": 7800
    },
    {
      "epoch": 3.6847014925373136,
      "grad_norm": 0.004190696403384209,
      "learning_rate": 1.0529850746268656e-05,
      "loss": 0.0002,
      "step": 7900
    },
    {
      "epoch": 3.6847014925373136,
      "eval_loss": 2.752276259343489e-06,
      "eval_runtime": 0.1641,
      "eval_samples_per_second": 182.838,
      "eval_steps_per_second": 24.378,
      "step": 7900
    },
    {
      "epoch": 3.7313432835820897,
      "grad_norm": 0.0028494582511484623,
      "learning_rate": 1.0156716417910448e-05,
      "loss": 0.0005,
      "step": 8000
    },
    {
      "epoch": 3.7313432835820897,
      "eval_loss": 2.7154821964359144e-06,
      "eval_runtime": 0.1578,
      "eval_samples_per_second": 190.11,
      "eval_steps_per_second": 25.348,
      "step": 8000
    },
    {
      "epoch": 3.7779850746268657,
      "grad_norm": 0.03807828575372696,
      "learning_rate": 9.787313432835822e-06,
      "loss": 0.0006,
      "step": 8100
    },
    {
      "epoch": 3.7779850746268657,
      "eval_loss": 2.827802290994441e-06,
      "eval_runtime": 0.1552,
      "eval_samples_per_second": 193.25,
      "eval_steps_per_second": 25.767,
      "step": 8100
    },
    {
      "epoch": 3.824626865671642,
      "grad_norm": 0.5222908854484558,
      "learning_rate": 9.414179104477614e-06,
      "loss": 0.0002,
      "step": 8200
    },
    {
      "epoch": 3.824626865671642,
      "eval_loss": 2.7299265639157966e-06,
      "eval_runtime": 0.156,
      "eval_samples_per_second": 192.29,
      "eval_steps_per_second": 25.639,
      "step": 8200
    },
    {
      "epoch": 3.871268656716418,
      "grad_norm": 0.0023298938758671284,
      "learning_rate": 9.041044776119404e-06,
      "loss": 0.0002,
      "step": 8300
    },
    {
      "epoch": 3.871268656716418,
      "eval_loss": 2.6636719212547177e-06,
      "eval_runtime": 0.1552,
      "eval_samples_per_second": 193.302,
      "eval_steps_per_second": 25.774,
      "step": 8300
    },
    {
      "epoch": 3.917910447761194,
      "grad_norm": 0.05465957522392273,
      "learning_rate": 8.667910447761195e-06,
      "loss": 0.0005,
      "step": 8400
    },
    {
      "epoch": 3.917910447761194,
      "eval_loss": 2.582373781478964e-06,
      "eval_runtime": 0.1571,
      "eval_samples_per_second": 190.991,
      "eval_steps_per_second": 25.465,
      "step": 8400
    },
    {
      "epoch": 3.96455223880597,
      "grad_norm": 0.048116568475961685,
      "learning_rate": 8.294776119402985e-06,
      "loss": 0.0002,
      "step": 8500
    },
    {
      "epoch": 3.96455223880597,
      "eval_loss": 2.487162873876514e-06,
      "eval_runtime": 0.1645,
      "eval_samples_per_second": 182.375,
      "eval_steps_per_second": 24.317,
      "step": 8500
    },
    {
      "epoch": 4.0111940298507465,
      "grad_norm": 0.0024153583217412233,
      "learning_rate": 7.921641791044777e-06,
      "loss": 0.0001,
      "step": 8600
    },
    {
      "epoch": 4.0111940298507465,
      "eval_loss": 2.429806954751257e-06,
      "eval_runtime": 0.182,
      "eval_samples_per_second": 164.853,
      "eval_steps_per_second": 21.98,
      "step": 8600
    },
    {
      "epoch": 4.057835820895522,
      "grad_norm": 0.002540579764172435,
      "learning_rate": 7.548507462686568e-06,
      "loss": 0.0001,
      "step": 8700
    },
    {
      "epoch": 4.057835820895522,
      "eval_loss": 2.3769621293467935e-06,
      "eval_runtime": 0.1595,
      "eval_samples_per_second": 188.114,
      "eval_steps_per_second": 25.082,
      "step": 8700
    },
    {
      "epoch": 4.104477611940299,
      "grad_norm": 0.0042966934852302074,
      "learning_rate": 7.175373134328358e-06,
      "loss": 0.0002,
      "step": 8800
    },
    {
      "epoch": 4.104477611940299,
      "eval_loss": 2.33703940466512e-06,
      "eval_runtime": 0.1575,
      "eval_samples_per_second": 190.469,
      "eval_steps_per_second": 25.396,
      "step": 8800
    },
    {
      "epoch": 4.151119402985074,
      "grad_norm": 0.0006062475731596351,
      "learning_rate": 6.802238805970149e-06,
      "loss": 0.0004,
      "step": 8900
    },
    {
      "epoch": 4.151119402985074,
      "eval_loss": 2.3632851480215322e-06,
      "eval_runtime": 0.1561,
      "eval_samples_per_second": 192.228,
      "eval_steps_per_second": 25.63,
      "step": 8900
    },
    {
      "epoch": 4.197761194029851,
      "grad_norm": 0.009168209508061409,
      "learning_rate": 6.429104477611941e-06,
      "loss": 0.0002,
      "step": 9000
    },
    {
      "epoch": 4.197761194029851,
      "eval_loss": 2.2578756215807516e-06,
      "eval_runtime": 0.1552,
      "eval_samples_per_second": 193.262,
      "eval_steps_per_second": 25.768,
      "step": 9000
    },
    {
      "epoch": 4.244402985074627,
      "grad_norm": 0.002683964092284441,
      "learning_rate": 6.055970149253731e-06,
      "loss": 0.0001,
      "step": 9100
    },
    {
      "epoch": 4.244402985074627,
      "eval_loss": 2.2409305984183447e-06,
      "eval_runtime": 0.1564,
      "eval_samples_per_second": 191.8,
      "eval_steps_per_second": 25.573,
      "step": 9100
    },
    {
      "epoch": 4.291044776119403,
      "grad_norm": 0.0012024191673845053,
      "learning_rate": 5.682835820895522e-06,
      "loss": 0.0002,
      "step": 9200
    },
    {
      "epoch": 4.291044776119403,
      "eval_loss": 2.202784344262909e-06,
      "eval_runtime": 0.1563,
      "eval_samples_per_second": 191.927,
      "eval_steps_per_second": 25.59,
      "step": 9200
    },
    {
      "epoch": 4.337686567164179,
      "grad_norm": 0.003109121695160866,
      "learning_rate": 5.309701492537314e-06,
      "loss": 0.0002,
      "step": 9300
    },
    {
      "epoch": 4.337686567164179,
      "eval_loss": 2.2312042347039096e-06,
      "eval_runtime": 0.1563,
      "eval_samples_per_second": 191.941,
      "eval_steps_per_second": 25.592,
      "step": 9300
    },
    {
      "epoch": 4.384328358208955,
      "grad_norm": 0.0009185061790049076,
      "learning_rate": 4.9365671641791045e-06,
      "loss": 0.0002,
      "step": 9400
    },
    {
      "epoch": 4.384328358208955,
      "eval_loss": 2.1759601622761693e-06,
      "eval_runtime": 0.159,
      "eval_samples_per_second": 188.695,
      "eval_steps_per_second": 25.159,
      "step": 9400
    },
    {
      "epoch": 4.4309701492537314,
      "grad_norm": 0.0008011642494238913,
      "learning_rate": 4.563432835820896e-06,
      "loss": 0.0004,
      "step": 9500
    },
    {
      "epoch": 4.4309701492537314,
      "eval_loss": 2.2056660782254767e-06,
      "eval_runtime": 0.1567,
      "eval_samples_per_second": 191.395,
      "eval_steps_per_second": 25.519,
      "step": 9500
    },
    {
      "epoch": 4.477611940298507,
      "grad_norm": 0.001676097046583891,
      "learning_rate": 4.190298507462687e-06,
      "loss": 0.0001,
      "step": 9600
    },
    {
      "epoch": 4.477611940298507,
      "eval_loss": 2.196997229475528e-06,
      "eval_runtime": 0.1554,
      "eval_samples_per_second": 193.034,
      "eval_steps_per_second": 25.738,
      "step": 9600
    },
    {
      "epoch": 4.524253731343284,
      "grad_norm": 0.001348722493276,
      "learning_rate": 3.8171641791044775e-06,
      "loss": 0.0002,
      "step": 9700
    },
    {
      "epoch": 4.524253731343284,
      "eval_loss": 2.159943960577948e-06,
      "eval_runtime": 0.1588,
      "eval_samples_per_second": 188.88,
      "eval_steps_per_second": 25.184,
      "step": 9700
    },
    {
      "epoch": 4.57089552238806,
      "grad_norm": 0.002949994755908847,
      "learning_rate": 3.4440298507462687e-06,
      "loss": 0.0001,
      "step": 9800
    },
    {
      "epoch": 4.57089552238806,
      "eval_loss": 2.117126769007882e-06,
      "eval_runtime": 0.1558,
      "eval_samples_per_second": 192.602,
      "eval_steps_per_second": 25.68,
      "step": 9800
    },
    {
      "epoch": 4.617537313432836,
      "grad_norm": 0.005024689715355635,
      "learning_rate": 3.07089552238806e-06,
      "loss": 0.0001,
      "step": 9900
    },
    {
      "epoch": 4.617537313432836,
      "eval_loss": 2.069621814371203e-06,
      "eval_runtime": 0.1558,
      "eval_samples_per_second": 192.576,
      "eval_steps_per_second": 25.677,
      "step": 9900
    },
    {
      "epoch": 4.664179104477612,
      "grad_norm": 0.0006656211335211992,
      "learning_rate": 2.697761194029851e-06,
      "loss": 0.0004,
      "step": 10000
    },
    {
      "epoch": 4.664179104477612,
      "eval_loss": 2.0410807337611914e-06,
      "eval_runtime": 0.1585,
      "eval_samples_per_second": 189.231,
      "eval_steps_per_second": 25.231,
      "step": 10000
    },
    {
      "epoch": 4.710820895522388,
      "grad_norm": 0.0003097167646046728,
      "learning_rate": 2.324626865671642e-06,
      "loss": 0.0001,
      "step": 10100
    },
    {
      "epoch": 4.710820895522388,
      "eval_loss": 2.0217082692397526e-06,
      "eval_runtime": 0.1557,
      "eval_samples_per_second": 192.62,
      "eval_steps_per_second": 25.683,
      "step": 10100
    },
    {
      "epoch": 4.757462686567164,
      "grad_norm": 0.0007842735503800213,
      "learning_rate": 1.951492537313433e-06,
      "loss": 0.0001,
      "step": 10200
    },
    {
      "epoch": 4.757462686567164,
      "eval_loss": 2.0089323697902728e-06,
      "eval_runtime": 0.174,
      "eval_samples_per_second": 172.426,
      "eval_steps_per_second": 22.99,
      "step": 10200
    },
    {
      "epoch": 4.80410447761194,
      "grad_norm": 0.004456710536032915,
      "learning_rate": 1.578358208955224e-06,
      "loss": 0.0001,
      "step": 10300
    },
    {
      "epoch": 4.80410447761194,
      "eval_loss": 2.0024342575197807e-06,
      "eval_runtime": 0.1647,
      "eval_samples_per_second": 182.154,
      "eval_steps_per_second": 24.287,
      "step": 10300
    },
    {
      "epoch": 4.850746268656716,
      "grad_norm": 0.0003174096636939794,
      "learning_rate": 1.205223880597015e-06,
      "loss": 0.0001,
      "step": 10400
    },
    {
      "epoch": 4.850746268656716,
      "eval_loss": 1.9801934740826255e-06,
      "eval_runtime": 0.1571,
      "eval_samples_per_second": 190.953,
      "eval_steps_per_second": 25.46,
      "step": 10400
    },
    {
      "epoch": 4.897388059701493,
      "grad_norm": 0.013769041746854782,
      "learning_rate": 8.320895522388061e-07,
      "loss": 0.0001,
      "step": 10500
    },
    {
      "epoch": 4.897388059701493,
      "eval_loss": 1.9813883227470797e-06,
      "eval_runtime": 0.1598,
      "eval_samples_per_second": 187.736,
      "eval_steps_per_second": 25.031,
      "step": 10500
    },
    {
      "epoch": 4.9440298507462686,
      "grad_norm": 0.0012828693725168705,
      "learning_rate": 4.5895522388059706e-07,
      "loss": 0.0002,
      "step": 10600
    },
    {
      "epoch": 4.9440298507462686,
      "eval_loss": 1.981977902687504e-06,
      "eval_runtime": 0.1586,
      "eval_samples_per_second": 189.182,
      "eval_steps_per_second": 25.224,
      "step": 10600
    }
  ],
  "logging_steps": 100,
  "max_steps": 10720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4161551619317760.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}