{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 31250,
  "global_step": 125000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.3911694288253784,
      "learning_rate": 0.0002988,
      "loss": 0.1122,
      "step": 500
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.48731762170791626,
      "learning_rate": 0.00029759999999999997,
      "loss": 0.1165,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.6664834022521973,
      "learning_rate": 0.0002964,
      "loss": 0.1364,
      "step": 1500
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.37042468786239624,
      "learning_rate": 0.00029519999999999997,
      "loss": 0.1335,
      "step": 2000
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.42679572105407715,
      "learning_rate": 0.000294,
      "loss": 0.1434,
      "step": 2500
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.6656017899513245,
      "learning_rate": 0.00029279999999999996,
      "loss": 0.1353,
      "step": 3000
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.6870421171188354,
      "learning_rate": 0.0002916,
      "loss": 0.1316,
      "step": 3500
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.993772566318512,
      "learning_rate": 0.00029039999999999996,
      "loss": 0.1354,
      "step": 4000
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8672360181808472,
      "learning_rate": 0.0002892,
      "loss": 0.1536,
      "step": 4500
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8261292576789856,
      "learning_rate": 0.00028799999999999995,
      "loss": 0.1469,
      "step": 5000
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.0536205768585205,
      "learning_rate": 0.0002868,
      "loss": 0.1446,
      "step": 5500
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.5625632405281067,
      "learning_rate": 0.00028559999999999995,
      "loss": 0.1454,
      "step": 6000
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.6709353923797607,
      "learning_rate": 0.0002844,
      "loss": 0.1441,
      "step": 6500
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.0651500225067139,
      "learning_rate": 0.00028319999999999994,
      "loss": 0.1427,
      "step": 7000
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.2373555898666382,
      "learning_rate": 0.00028199999999999997,
      "loss": 0.1439,
      "step": 7500
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.6624026298522949,
      "learning_rate": 0.0002808,
      "loss": 0.152,
      "step": 8000
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.4510903358459473,
      "learning_rate": 0.00027959999999999997,
      "loss": 0.1508,
      "step": 8500
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.47858643531799316,
      "learning_rate": 0.0002784,
      "loss": 0.1599,
      "step": 9000
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5771904587745667,
      "learning_rate": 0.0002772,
      "loss": 0.1519,
      "step": 9500
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6509794592857361,
      "learning_rate": 0.000276,
      "loss": 0.1568,
      "step": 10000
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.7169878482818604,
      "learning_rate": 0.0002748,
      "loss": 0.1531,
      "step": 10500
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.7649238109588623,
      "learning_rate": 0.0002736,
      "loss": 0.1545,
      "step": 11000
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.1057517528533936,
      "learning_rate": 0.0002724,
      "loss": 0.1523,
      "step": 11500
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.4265155792236328,
      "learning_rate": 0.0002712,
      "loss": 0.1496,
      "step": 12000
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.9195236563682556,
      "learning_rate": 0.00027,
      "loss": 0.156,
      "step": 12500
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.6371893882751465,
      "learning_rate": 0.0002688,
      "loss": 0.1538,
      "step": 13000
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.3511596620082855,
      "learning_rate": 0.0002676,
      "loss": 0.1535,
      "step": 13500
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.599895715713501,
      "learning_rate": 0.00026639999999999997,
      "loss": 0.1512,
      "step": 14000
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.9206880927085876,
      "learning_rate": 0.0002652,
      "loss": 0.1611,
      "step": 14500
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.49088287353515625,
      "learning_rate": 0.00026399999999999997,
      "loss": 0.1459,
      "step": 15000
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.42937585711479187,
      "learning_rate": 0.0002628,
      "loss": 0.1605,
      "step": 15500
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.413435161113739,
      "learning_rate": 0.00026159999999999996,
      "loss": 0.1482,
      "step": 16000
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.9658032655715942,
      "learning_rate": 0.0002604,
      "loss": 0.1462,
      "step": 16500
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.7052133083343506,
      "learning_rate": 0.00025919999999999996,
      "loss": 0.1494,
      "step": 17000
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.18123334646224976,
      "learning_rate": 0.000258,
      "loss": 0.1517,
      "step": 17500
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.4006165862083435,
      "learning_rate": 0.00025679999999999995,
      "loss": 0.1533,
      "step": 18000
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.49044716358184814,
      "learning_rate": 0.0002556,
      "loss": 0.1616,
      "step": 18500
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.7895859479904175,
      "learning_rate": 0.00025439999999999995,
      "loss": 0.1575,
      "step": 19000
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7281160950660706,
      "learning_rate": 0.0002532,
      "loss": 0.1519,
      "step": 19500
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.84168541431427,
      "learning_rate": 0.00025199999999999995,
      "loss": 0.1513,
      "step": 20000
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.320218563079834,
      "learning_rate": 0.00025079999999999997,
      "loss": 0.1512,
      "step": 20500
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.6346606612205505,
      "learning_rate": 0.00024959999999999994,
      "loss": 0.1512,
      "step": 21000
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.6091498136520386,
      "learning_rate": 0.00024839999999999997,
      "loss": 0.1528,
      "step": 21500
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.37541925907135,
      "learning_rate": 0.0002472,
      "loss": 0.1472,
      "step": 22000
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.41054484248161316,
      "learning_rate": 0.00024599999999999996,
      "loss": 0.1548,
      "step": 22500
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.3798141479492188,
      "learning_rate": 0.0002448,
      "loss": 0.1471,
      "step": 23000
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.5459690690040588,
      "learning_rate": 0.00024359999999999999,
      "loss": 0.1426,
      "step": 23500
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6976202726364136,
      "learning_rate": 0.00024239999999999998,
      "loss": 0.1552,
      "step": 24000
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4615379869937897,
      "learning_rate": 0.00024119999999999998,
      "loss": 0.1625,
      "step": 24500
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.9051592350006104,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.1626,
      "step": 25000
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.3928375840187073,
      "learning_rate": 0.0002388,
      "loss": 0.1532,
      "step": 25500
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.570114254951477,
      "learning_rate": 0.0002376,
      "loss": 0.1519,
      "step": 26000
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.360300898551941,
      "learning_rate": 0.0002364,
      "loss": 0.1571,
      "step": 26500
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.0905797481536865,
      "learning_rate": 0.0002352,
      "loss": 0.1544,
      "step": 27000
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.8806708455085754,
      "learning_rate": 0.000234,
      "loss": 0.1483,
      "step": 27500
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.876801073551178,
      "learning_rate": 0.0002328,
      "loss": 0.1533,
      "step": 28000
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.6989384293556213,
      "learning_rate": 0.0002316,
      "loss": 0.1478,
      "step": 28500
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.1341756582260132,
      "learning_rate": 0.0002304,
      "loss": 0.1525,
      "step": 29000
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5615713000297546,
      "learning_rate": 0.0002292,
      "loss": 0.1442,
      "step": 29500
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.6419674754142761,
      "learning_rate": 0.00022799999999999999,
      "loss": 0.1546,
      "step": 30000
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.48240768909454346,
      "learning_rate": 0.00022679999999999998,
      "loss": 0.1526,
      "step": 30500
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5017668008804321,
      "learning_rate": 0.00022559999999999998,
      "loss": 0.1522,
      "step": 31000
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.1550917774438858,
      "eval_runtime": 203.798,
      "eval_samples_per_second": 98.136,
      "eval_steps_per_second": 24.534,
      "step": 31250
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.9850152134895325,
      "learning_rate": 0.00022439999999999998,
      "loss": 0.1478,
      "step": 31500
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5590870976448059,
      "learning_rate": 0.00022319999999999998,
      "loss": 0.1501,
      "step": 32000
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.128318428993225,
      "learning_rate": 0.00022199999999999998,
      "loss": 0.156,
      "step": 32500
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5681318044662476,
      "learning_rate": 0.00022079999999999997,
      "loss": 0.1492,
      "step": 33000
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.7050376534461975,
      "learning_rate": 0.00021959999999999997,
      "loss": 0.1632,
      "step": 33500
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.3747576177120209,
      "learning_rate": 0.00021839999999999997,
      "loss": 0.1492,
      "step": 34000
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.44152018427848816,
      "learning_rate": 0.00021719999999999997,
      "loss": 0.1505,
      "step": 34500
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.40513116121292114,
      "learning_rate": 0.00021599999999999996,
      "loss": 0.1484,
      "step": 35000
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.7306041121482849,
      "learning_rate": 0.00021479999999999996,
      "loss": 0.1448,
      "step": 35500
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.8463369011878967,
      "learning_rate": 0.00021359999999999996,
      "loss": 0.1534,
      "step": 36000
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.9181163311004639,
      "learning_rate": 0.00021239999999999996,
      "loss": 0.1493,
      "step": 36500
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.0870494842529297,
      "learning_rate": 0.00021119999999999996,
      "loss": 0.1572,
      "step": 37000
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.41071075201034546,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.1542,
      "step": 37500
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.1432327032089233,
      "learning_rate": 0.00020879999999999998,
      "loss": 0.1448,
      "step": 38000
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.465071439743042,
      "learning_rate": 0.00020759999999999998,
      "loss": 0.1452,
      "step": 38500
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.7455276250839233,
      "learning_rate": 0.00020639999999999998,
      "loss": 0.1491,
      "step": 39000
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6694576144218445,
      "learning_rate": 0.0002052,
      "loss": 0.1439,
      "step": 39500
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6440109014511108,
      "learning_rate": 0.000204,
      "loss": 0.1497,
      "step": 40000
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6777958273887634,
      "learning_rate": 0.0002028,
      "loss": 0.1489,
      "step": 40500
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.4175439476966858,
      "learning_rate": 0.0002016,
      "loss": 0.1549,
      "step": 41000
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.4022385776042938,
      "learning_rate": 0.0002004,
      "loss": 0.1519,
      "step": 41500
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.1414192914962769,
      "learning_rate": 0.0001992,
      "loss": 0.1478,
      "step": 42000
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.484553575515747,
      "learning_rate": 0.000198,
      "loss": 0.1571,
      "step": 42500
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.7761896848678589,
      "learning_rate": 0.00019679999999999999,
      "loss": 0.1539,
      "step": 43000
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.6768946051597595,
      "learning_rate": 0.00019559999999999998,
      "loss": 0.1501,
      "step": 43500
    },
    {
      "epoch": 0.35,
      "grad_norm": 2.1360421180725098,
      "learning_rate": 0.00019439999999999998,
      "loss": 0.1531,
      "step": 44000
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5634905099868774,
      "learning_rate": 0.00019319999999999998,
      "loss": 0.1452,
      "step": 44500
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3370414972305298,
      "learning_rate": 0.00019199999999999998,
      "loss": 0.1486,
      "step": 45000
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.3636302947998047,
      "learning_rate": 0.00019079999999999998,
      "loss": 0.1506,
      "step": 45500
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.5136697292327881,
      "learning_rate": 0.00018959999999999997,
      "loss": 0.1453,
      "step": 46000
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.4970598518848419,
      "learning_rate": 0.00018839999999999997,
      "loss": 0.1447,
      "step": 46500
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.8702558279037476,
      "learning_rate": 0.0001872,
      "loss": 0.1505,
      "step": 47000
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.3983975648880005,
      "learning_rate": 0.000186,
      "loss": 0.157,
      "step": 47500
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.6854807734489441,
      "learning_rate": 0.0001848,
      "loss": 0.1482,
      "step": 48000
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.45792344212532043,
      "learning_rate": 0.0001836,
      "loss": 0.1522,
      "step": 48500
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.8602573871612549,
      "learning_rate": 0.0001824,
      "loss": 0.1466,
      "step": 49000
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.4061354100704193,
      "learning_rate": 0.00018119999999999999,
      "loss": 0.143,
      "step": 49500
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.1822609901428223,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.1377,
      "step": 50000
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.3173510730266571,
      "learning_rate": 0.00017879999999999998,
      "loss": 0.1445,
      "step": 50500
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.9117286205291748,
      "learning_rate": 0.00017759999999999998,
      "loss": 0.148,
      "step": 51000
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.5054415464401245,
      "learning_rate": 0.00017639999999999998,
      "loss": 0.1397,
      "step": 51500
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.496282935142517,
      "learning_rate": 0.00017519999999999998,
      "loss": 0.1487,
      "step": 52000
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.48316746950149536,
      "learning_rate": 0.00017399999999999997,
      "loss": 0.1588,
      "step": 52500
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.00754976272583,
      "learning_rate": 0.00017279999999999997,
      "loss": 0.1559,
      "step": 53000
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.4439372420310974,
      "learning_rate": 0.00017159999999999997,
      "loss": 0.1422,
      "step": 53500
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.2646929025650024,
      "learning_rate": 0.00017039999999999997,
      "loss": 0.144,
      "step": 54000
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.8087254762649536,
      "learning_rate": 0.00016919999999999997,
      "loss": 0.1538,
      "step": 54500
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.2770575284957886,
      "learning_rate": 0.000168,
      "loss": 0.1454,
      "step": 55000
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.0025547742843628,
      "learning_rate": 0.0001668,
      "loss": 0.149,
      "step": 55500
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5146603584289551,
      "learning_rate": 0.0001656,
      "loss": 0.1512,
      "step": 56000
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5651877522468567,
      "learning_rate": 0.0001644,
      "loss": 0.1494,
      "step": 56500
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.3547286987304688,
      "learning_rate": 0.0001632,
      "loss": 0.1435,
      "step": 57000
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.29811692237854004,
      "learning_rate": 0.000162,
      "loss": 0.1536,
      "step": 57500
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.111806869506836,
      "learning_rate": 0.0001608,
      "loss": 0.1468,
      "step": 58000
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.6532767415046692,
      "learning_rate": 0.0001596,
      "loss": 0.1525,
      "step": 58500
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.31857454776763916,
      "learning_rate": 0.0001584,
      "loss": 0.139,
      "step": 59000
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.8610038757324219,
      "learning_rate": 0.0001572,
      "loss": 0.1503,
      "step": 59500
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5853925347328186,
      "learning_rate": 0.000156,
      "loss": 0.1448,
      "step": 60000
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.6353251934051514,
      "learning_rate": 0.0001548,
      "loss": 0.1538,
      "step": 60500
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.0176310539245605,
      "learning_rate": 0.0001536,
      "loss": 0.1443,
      "step": 61000
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.3258388638496399,
      "learning_rate": 0.0001524,
      "loss": 0.1386,
      "step": 61500
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.5362368822097778,
      "learning_rate": 0.0001512,
      "loss": 0.1465,
      "step": 62000
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.46234962344169617,
      "learning_rate": 0.00015,
      "loss": 0.1449,
      "step": 62500
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.1452270895242691,
      "eval_runtime": 201.7763,
      "eval_samples_per_second": 99.12,
      "eval_steps_per_second": 24.78,
      "step": 62500
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3215588331222534,
      "learning_rate": 0.00014879999999999998,
      "loss": 0.1436,
      "step": 63000
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.9984625577926636,
      "learning_rate": 0.00014759999999999998,
      "loss": 0.1453,
      "step": 63500
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.3026807308197021,
      "learning_rate": 0.00014639999999999998,
      "loss": 0.145,
      "step": 64000
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.2961193323135376,
      "learning_rate": 0.00014519999999999998,
      "loss": 0.1458,
      "step": 64500
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.6397677659988403,
      "learning_rate": 0.00014399999999999998,
      "loss": 0.1442,
      "step": 65000
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.2444109916687012,
      "learning_rate": 0.00014279999999999997,
      "loss": 0.1533,
      "step": 65500
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.3235061764717102,
      "learning_rate": 0.00014159999999999997,
      "loss": 0.1473,
      "step": 66000
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.508976936340332,
      "learning_rate": 0.0001404,
      "loss": 0.1455,
      "step": 66500
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.44948485493659973,
      "learning_rate": 0.0001392,
      "loss": 0.1473,
      "step": 67000
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.8090860247612,
      "learning_rate": 0.000138,
      "loss": 0.144,
      "step": 67500
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.7624330520629883,
      "learning_rate": 0.0001368,
      "loss": 0.1513,
      "step": 68000
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.3432750105857849,
      "learning_rate": 0.0001356,
      "loss": 0.1518,
      "step": 68500
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.5248464345932007,
      "learning_rate": 0.0001344,
      "loss": 0.1481,
      "step": 69000
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5486373901367188,
      "learning_rate": 0.00013319999999999999,
      "loss": 0.1493,
      "step": 69500
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.1218228340148926,
      "learning_rate": 0.00013199999999999998,
      "loss": 0.1513,
      "step": 70000
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.6330295205116272,
      "learning_rate": 0.00013079999999999998,
      "loss": 0.1426,
      "step": 70500
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.347390055656433,
      "learning_rate": 0.00012959999999999998,
      "loss": 0.1481,
      "step": 71000
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.37359416484832764,
      "learning_rate": 0.00012839999999999998,
      "loss": 0.147,
      "step": 71500
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.5048085451126099,
      "learning_rate": 0.00012719999999999997,
      "loss": 0.1454,
      "step": 72000
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4849306344985962,
      "learning_rate": 0.00012599999999999997,
      "loss": 0.1519,
      "step": 72500
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.022958755493164,
      "learning_rate": 0.00012479999999999997,
      "loss": 0.1514,
      "step": 73000
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.671461284160614,
      "learning_rate": 0.0001236,
      "loss": 0.143,
      "step": 73500
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.6191059947013855,
      "learning_rate": 0.0001224,
      "loss": 0.1509,
      "step": 74000
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.8567871451377869,
      "learning_rate": 0.00012119999999999999,
      "loss": 0.1505,
      "step": 74500
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5671202540397644,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.1472,
      "step": 75000
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.9271816611289978,
      "learning_rate": 0.0001188,
      "loss": 0.154,
      "step": 75500
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.8567375540733337,
      "learning_rate": 0.0001176,
      "loss": 0.1458,
      "step": 76000
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.40565225481987,
      "learning_rate": 0.0001164,
      "loss": 0.1529,
      "step": 76500
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.1839731931686401,
      "learning_rate": 0.0001152,
      "loss": 0.1416,
      "step": 77000
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.0027508735656738,
      "learning_rate": 0.00011399999999999999,
      "loss": 0.1494,
      "step": 77500
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.7692805528640747,
      "learning_rate": 0.00011279999999999999,
      "loss": 0.1426,
      "step": 78000
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.5682281255722046,
      "learning_rate": 0.00011159999999999999,
      "loss": 0.1496,
      "step": 78500
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.4132564961910248,
      "learning_rate": 0.00011039999999999999,
      "loss": 0.1537,
      "step": 79000
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.06421959400177,
      "learning_rate": 0.00010919999999999998,
      "loss": 0.148,
      "step": 79500
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.649330198764801,
      "learning_rate": 0.00010799999999999998,
      "loss": 0.1485,
      "step": 80000
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5109010338783264,
      "learning_rate": 0.00010679999999999998,
      "loss": 0.1512,
      "step": 80500
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.40312623977661133,
      "learning_rate": 0.00010559999999999998,
      "loss": 0.1533,
      "step": 81000
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.678674578666687,
      "learning_rate": 0.00010439999999999999,
      "loss": 0.1522,
      "step": 81500
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4731667935848236,
      "learning_rate": 0.00010319999999999999,
      "loss": 0.154,
      "step": 82000
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.742722749710083,
      "learning_rate": 0.000102,
      "loss": 0.1511,
      "step": 82500
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5631982684135437,
      "learning_rate": 0.0001008,
      "loss": 0.1442,
      "step": 83000
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.3675552010536194,
      "learning_rate": 9.96e-05,
      "loss": 0.1448,
      "step": 83500
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.697551429271698,
      "learning_rate": 9.839999999999999e-05,
      "loss": 0.1434,
      "step": 84000
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.7331182956695557,
      "learning_rate": 9.719999999999999e-05,
      "loss": 0.1459,
      "step": 84500
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.344137042760849,
      "learning_rate": 9.599999999999999e-05,
      "loss": 0.1537,
      "step": 85000
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.1674726009368896,
      "learning_rate": 9.479999999999999e-05,
      "loss": 0.1515,
      "step": 85500
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5482670664787292,
      "learning_rate": 9.36e-05,
      "loss": 0.1487,
      "step": 86000
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.8054832220077515,
      "learning_rate": 9.24e-05,
      "loss": 0.1545,
      "step": 86500
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.1015689373016357,
      "learning_rate": 9.12e-05,
      "loss": 0.1475,
      "step": 87000
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.3283549547195435,
      "learning_rate": 8.999999999999999e-05,
      "loss": 0.1546,
      "step": 87500
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.41955164074897766,
      "learning_rate": 8.879999999999999e-05,
      "loss": 0.1457,
      "step": 88000
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.647566556930542,
      "learning_rate": 8.759999999999999e-05,
      "loss": 0.149,
      "step": 88500
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.6643834710121155,
      "learning_rate": 8.639999999999999e-05,
      "loss": 0.1531,
      "step": 89000
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.21100400388240814,
      "learning_rate": 8.519999999999998e-05,
      "loss": 0.1452,
      "step": 89500
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.537507951259613,
      "learning_rate": 8.4e-05,
      "loss": 0.1441,
      "step": 90000
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.9034580588340759,
      "learning_rate": 8.28e-05,
      "loss": 0.1513,
      "step": 90500
    },
    {
      "epoch": 0.73,
      "grad_norm": 2.2073426246643066,
      "learning_rate": 8.16e-05,
      "loss": 0.1534,
      "step": 91000
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.3272952735424042,
      "learning_rate": 8.04e-05,
      "loss": 0.1343,
      "step": 91500
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.18572063744068146,
      "learning_rate": 7.92e-05,
      "loss": 0.1525,
      "step": 92000
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.6784604787826538,
      "learning_rate": 7.8e-05,
      "loss": 0.1498,
      "step": 92500
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.6597641110420227,
      "learning_rate": 7.68e-05,
      "loss": 0.1482,
      "step": 93000
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.2626825273036957,
      "learning_rate": 7.56e-05,
      "loss": 0.1483,
      "step": 93500
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.13335928320884705,
      "eval_runtime": 202.0526,
      "eval_samples_per_second": 98.984,
      "eval_steps_per_second": 24.746,
      "step": 93750
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.8486179113388062,
      "learning_rate": 7.439999999999999e-05,
      "loss": 0.1397,
      "step": 94000
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.38229885697364807,
      "learning_rate": 7.319999999999999e-05,
      "loss": 0.1599,
      "step": 94500
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.6869509220123291,
      "learning_rate": 7.199999999999999e-05,
      "loss": 0.1467,
      "step": 95000
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5972515344619751,
      "learning_rate": 7.079999999999999e-05,
      "loss": 0.1465,
      "step": 95500
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.3241545557975769,
      "learning_rate": 6.96e-05,
      "loss": 0.1539,
      "step": 96000
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.406515508890152,
      "learning_rate": 6.84e-05,
      "loss": 0.1464,
      "step": 96500
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.8518499135971069,
      "learning_rate": 6.72e-05,
      "loss": 0.1481,
      "step": 97000
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.6099169254302979,
      "learning_rate": 6.599999999999999e-05,
      "loss": 0.1492,
      "step": 97500
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.002388834953308,
      "learning_rate": 6.479999999999999e-05,
      "loss": 0.152,
      "step": 98000
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.734794557094574,
      "learning_rate": 6.359999999999999e-05,
      "loss": 0.1443,
      "step": 98500
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.0499773025512695,
      "learning_rate": 6.239999999999999e-05,
      "loss": 0.1572,
      "step": 99000
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.2162612676620483,
      "learning_rate": 6.12e-05,
      "loss": 0.1523,
      "step": 99500
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.9487478733062744,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 0.1563,
      "step": 100000
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.7570409178733826,
      "learning_rate": 5.88e-05,
      "loss": 0.1555,
      "step": 100500
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.8198479413986206,
      "learning_rate": 5.76e-05,
      "loss": 0.1529,
      "step": 101000
    },
    {
      "epoch": 0.81,
      "grad_norm": 1.0460014343261719,
      "learning_rate": 5.6399999999999995e-05,
      "loss": 0.1447,
      "step": 101500
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.6016435027122498,
      "learning_rate": 5.519999999999999e-05,
      "loss": 0.1527,
      "step": 102000
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.0050832033157349,
      "learning_rate": 5.399999999999999e-05,
      "loss": 0.1482,
      "step": 102500
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5460426807403564,
      "learning_rate": 5.279999999999999e-05,
      "loss": 0.1616,
      "step": 103000
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.7548345327377319,
      "learning_rate": 5.1599999999999994e-05,
      "loss": 0.1477,
      "step": 103500
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.590357780456543,
      "learning_rate": 5.04e-05,
      "loss": 0.156,
      "step": 104000
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5072462558746338,
      "learning_rate": 4.9199999999999997e-05,
      "loss": 0.1597,
      "step": 104500
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.5731006264686584,
      "learning_rate": 4.7999999999999994e-05,
      "loss": 0.1463,
      "step": 105000
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.1715773344039917,
      "learning_rate": 4.68e-05,
      "loss": 0.1499,
      "step": 105500
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.20436818897724152,
      "learning_rate": 4.56e-05,
      "loss": 0.1547,
      "step": 106000
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.46894416213035583,
      "learning_rate": 4.4399999999999995e-05,
      "loss": 0.1579,
      "step": 106500
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5912017822265625,
      "learning_rate": 4.319999999999999e-05,
      "loss": 0.1428,
      "step": 107000
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.9041949510574341,
      "learning_rate": 4.2e-05,
      "loss": 0.15,
      "step": 107500
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.5391335487365723,
      "learning_rate": 4.08e-05,
      "loss": 0.1634,
      "step": 108000
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.38163840770721436,
      "learning_rate": 3.96e-05,
      "loss": 0.1578,
      "step": 108500
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4478495419025421,
      "learning_rate": 3.84e-05,
      "loss": 0.1588,
      "step": 109000
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.6843103766441345,
      "learning_rate": 3.7199999999999996e-05,
      "loss": 0.1507,
      "step": 109500
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.832815408706665,
      "learning_rate": 3.5999999999999994e-05,
      "loss": 0.1588,
      "step": 110000
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.35042092204093933,
      "learning_rate": 3.48e-05,
      "loss": 0.1581,
      "step": 110500
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.49709609150886536,
      "learning_rate": 3.36e-05,
      "loss": 0.1546,
      "step": 111000
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.85088711977005,
      "learning_rate": 3.2399999999999995e-05,
      "loss": 0.1639,
      "step": 111500
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.7669118642807007,
      "learning_rate": 3.119999999999999e-05,
      "loss": 0.1585,
      "step": 112000
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.6558898091316223,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 0.1552,
      "step": 112500
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.9164992570877075,
      "learning_rate": 2.88e-05,
      "loss": 0.1552,
      "step": 113000
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.1778265237808228,
      "learning_rate": 2.7599999999999997e-05,
      "loss": 0.1603,
      "step": 113500
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.4928775727748871,
      "learning_rate": 2.6399999999999995e-05,
      "loss": 0.1558,
      "step": 114000
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.10056832432746887,
      "learning_rate": 2.52e-05,
      "loss": 0.1581,
      "step": 114500
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.104854054749012,
      "learning_rate": 2.3999999999999997e-05,
      "loss": 0.1584,
      "step": 115000
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.1790800094604492,
      "learning_rate": 2.28e-05,
      "loss": 0.1634,
      "step": 115500
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.7926896214485168,
      "learning_rate": 2.1599999999999996e-05,
      "loss": 0.1546,
      "step": 116000
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.6564472317695618,
      "learning_rate": 2.04e-05,
      "loss": 0.1628,
      "step": 116500
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5036044716835022,
      "learning_rate": 1.92e-05,
      "loss": 0.1684,
      "step": 117000
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.9592491984367371,
      "learning_rate": 1.7999999999999997e-05,
      "loss": 0.1628,
      "step": 117500
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5925645232200623,
      "learning_rate": 1.68e-05,
      "loss": 0.1658,
      "step": 118000
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.5361761450767517,
      "learning_rate": 1.5599999999999996e-05,
      "loss": 0.1639,
      "step": 118500
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.6238356828689575,
      "learning_rate": 1.44e-05,
      "loss": 0.1506,
      "step": 119000
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.3824297189712524,
      "learning_rate": 1.3199999999999997e-05,
      "loss": 0.1642,
      "step": 119500
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.9775289297103882,
      "learning_rate": 1.1999999999999999e-05,
      "loss": 0.1631,
      "step": 120000
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.29735296964645386,
      "learning_rate": 1.0799999999999998e-05,
      "loss": 0.1693,
      "step": 120500
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5140572786331177,
      "learning_rate": 9.6e-06,
      "loss": 0.1651,
      "step": 121000
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5154848098754883,
      "learning_rate": 8.4e-06,
      "loss": 0.1639,
      "step": 121500
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.21747368574142456,
      "learning_rate": 7.2e-06,
      "loss": 0.1679,
      "step": 122000
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.51055908203125,
      "learning_rate": 5.999999999999999e-06,
      "loss": 0.1723,
      "step": 122500
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.451370120048523,
      "learning_rate": 4.8e-06,
      "loss": 0.1694,
      "step": 123000
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.4937916398048401,
      "learning_rate": 3.6e-06,
      "loss": 0.1545,
      "step": 123500
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.46715646982192993,
      "learning_rate": 2.4e-06,
      "loss": 0.1748,
      "step": 124000
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.35662218928337097,
      "learning_rate": 1.2e-06,
      "loss": 0.1627,
      "step": 124500
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.8029049634933472,
      "learning_rate": 0.0,
      "loss": 0.1636,
      "step": 125000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.12450261414051056,
      "eval_runtime": 202.3721,
      "eval_samples_per_second": 98.828,
      "eval_steps_per_second": 24.707,
      "step": 125000
    }
  ],
  "logging_steps": 500,
  "max_steps": 125000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 125000,
  "total_flos": 6.527143521828864e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}