{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.390625,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00390625,
      "grad_norm": 0.9358924627304077,
      "learning_rate": 1e-05,
      "loss": 0.4079,
      "step": 1
    },
    {
      "epoch": 0.00390625,
      "eval_loss": 0.4379751980304718,
      "eval_runtime": 37.4931,
      "eval_samples_per_second": 5.761,
      "eval_steps_per_second": 0.72,
      "step": 1
    },
    {
      "epoch": 0.0078125,
      "grad_norm": 0.9645170569419861,
      "learning_rate": 2e-05,
      "loss": 0.3906,
      "step": 2
    },
    {
      "epoch": 0.01171875,
      "grad_norm": 1.000429630279541,
      "learning_rate": 3e-05,
      "loss": 0.4281,
      "step": 3
    },
    {
      "epoch": 0.015625,
      "grad_norm": 0.6437779664993286,
      "learning_rate": 4e-05,
      "loss": 0.3714,
      "step": 4
    },
    {
      "epoch": 0.01953125,
      "grad_norm": 0.8947291374206543,
      "learning_rate": 5e-05,
      "loss": 0.3801,
      "step": 5
    },
    {
      "epoch": 0.0234375,
      "grad_norm": 0.7578244805335999,
      "learning_rate": 6e-05,
      "loss": 0.3648,
      "step": 6
    },
    {
      "epoch": 0.02734375,
      "grad_norm": 0.6235212087631226,
      "learning_rate": 7e-05,
      "loss": 0.3909,
      "step": 7
    },
    {
      "epoch": 0.03125,
      "grad_norm": 0.4007953405380249,
      "learning_rate": 8e-05,
      "loss": 0.3141,
      "step": 8
    },
    {
      "epoch": 0.03515625,
      "grad_norm": 0.3965928554534912,
      "learning_rate": 9e-05,
      "loss": 0.2924,
      "step": 9
    },
    {
      "epoch": 0.03515625,
      "eval_loss": 0.2510845959186554,
      "eval_runtime": 37.7104,
      "eval_samples_per_second": 5.728,
      "eval_steps_per_second": 0.716,
      "step": 9
    },
    {
      "epoch": 0.0390625,
      "grad_norm": 0.4188036322593689,
      "learning_rate": 0.0001,
      "loss": 0.2828,
      "step": 10
    },
    {
      "epoch": 0.04296875,
      "grad_norm": 0.3323066830635071,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.2415,
      "step": 11
    },
    {
      "epoch": 0.046875,
      "grad_norm": 0.2657626271247864,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.1928,
      "step": 12
    },
    {
      "epoch": 0.05078125,
      "grad_norm": 0.2819468080997467,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.1859,
      "step": 13
    },
    {
      "epoch": 0.0546875,
      "grad_norm": 0.21283726394176483,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.1895,
      "step": 14
    },
    {
      "epoch": 0.05859375,
      "grad_norm": 0.22380894422531128,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.1606,
      "step": 15
    },
    {
      "epoch": 0.0625,
      "grad_norm": 0.26086336374282837,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.1618,
      "step": 16
    },
    {
      "epoch": 0.06640625,
      "grad_norm": 0.17846892774105072,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.1524,
      "step": 17
    },
    {
      "epoch": 0.0703125,
      "grad_norm": 0.16768787801265717,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.1242,
      "step": 18
    },
    {
      "epoch": 0.0703125,
      "eval_loss": 0.14291226863861084,
      "eval_runtime": 37.7214,
      "eval_samples_per_second": 5.726,
      "eval_steps_per_second": 0.716,
      "step": 18
    },
    {
      "epoch": 0.07421875,
      "grad_norm": 0.16563105583190918,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.1247,
      "step": 19
    },
    {
      "epoch": 0.078125,
      "grad_norm": 0.17445707321166992,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.1217,
      "step": 20
    },
    {
      "epoch": 0.08203125,
      "grad_norm": 0.16826312243938446,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.118,
      "step": 21
    },
    {
      "epoch": 0.0859375,
      "grad_norm": 0.14247825741767883,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.1009,
      "step": 22
    },
    {
      "epoch": 0.08984375,
      "grad_norm": 0.20849590003490448,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.1035,
      "step": 23
    },
    {
      "epoch": 0.09375,
      "grad_norm": 0.17949005961418152,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.1156,
      "step": 24
    },
    {
      "epoch": 0.09765625,
      "grad_norm": 0.19471654295921326,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.1187,
      "step": 25
    },
    {
      "epoch": 0.1015625,
      "grad_norm": 0.18970464169979095,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.1302,
      "step": 26
    },
    {
      "epoch": 0.10546875,
      "grad_norm": 0.13743029534816742,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.103,
      "step": 27
    },
    {
      "epoch": 0.10546875,
      "eval_loss": 0.11523790657520294,
      "eval_runtime": 37.6775,
      "eval_samples_per_second": 5.733,
      "eval_steps_per_second": 0.717,
      "step": 27
    },
    {
      "epoch": 0.109375,
      "grad_norm": 0.16887696087360382,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0911,
      "step": 28
    },
    {
      "epoch": 0.11328125,
      "grad_norm": 0.1817857027053833,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.1184,
      "step": 29
    },
    {
      "epoch": 0.1171875,
      "grad_norm": 0.39407843351364136,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.1714,
      "step": 30
    },
    {
      "epoch": 0.12109375,
      "grad_norm": 0.14840836822986603,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.0822,
      "step": 31
    },
    {
      "epoch": 0.125,
      "grad_norm": 0.12886931002140045,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.0924,
      "step": 32
    },
    {
      "epoch": 0.12890625,
      "grad_norm": 0.131106898188591,
      "learning_rate": 8.473291852294987e-05,
      "loss": 0.1126,
      "step": 33
    },
    {
      "epoch": 0.1328125,
      "grad_norm": 0.20917432010173798,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.1214,
      "step": 34
    },
    {
      "epoch": 0.13671875,
      "grad_norm": 0.13371144235134125,
      "learning_rate": 8.213938048432697e-05,
      "loss": 0.1085,
      "step": 35
    },
    {
      "epoch": 0.140625,
      "grad_norm": 0.16854634881019592,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.1154,
      "step": 36
    },
    {
      "epoch": 0.140625,
      "eval_loss": 0.10245885699987411,
      "eval_runtime": 37.6755,
      "eval_samples_per_second": 5.733,
      "eval_steps_per_second": 0.717,
      "step": 36
    },
    {
      "epoch": 0.14453125,
      "grad_norm": 0.12487249076366425,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.109,
      "step": 37
    },
    {
      "epoch": 0.1484375,
      "grad_norm": 0.1517653465270996,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.0798,
      "step": 38
    },
    {
      "epoch": 0.15234375,
      "grad_norm": 0.16995568573474884,
      "learning_rate": 7.649596321166024e-05,
      "loss": 0.1015,
      "step": 39
    },
    {
      "epoch": 0.15625,
      "grad_norm": 0.09856864809989929,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.1026,
      "step": 40
    },
    {
      "epoch": 0.16015625,
      "grad_norm": 0.11969555914402008,
      "learning_rate": 7.347357813929454e-05,
      "loss": 0.099,
      "step": 41
    },
    {
      "epoch": 0.1640625,
      "grad_norm": 0.1942385733127594,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.069,
      "step": 42
    },
    {
      "epoch": 0.16796875,
      "grad_norm": 0.12041034549474716,
      "learning_rate": 7.033683215379002e-05,
      "loss": 0.1015,
      "step": 43
    },
    {
      "epoch": 0.171875,
      "grad_norm": 0.1598462611436844,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.0893,
      "step": 44
    },
    {
      "epoch": 0.17578125,
      "grad_norm": 0.17382952570915222,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.1179,
      "step": 45
    },
    {
      "epoch": 0.17578125,
      "eval_loss": 0.09511082619428635,
      "eval_runtime": 37.6196,
      "eval_samples_per_second": 5.742,
      "eval_steps_per_second": 0.718,
      "step": 45
    },
    {
      "epoch": 0.1796875,
      "grad_norm": 0.1521601378917694,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0992,
      "step": 46
    },
    {
      "epoch": 0.18359375,
      "grad_norm": 0.18973639607429504,
      "learning_rate": 6.378186779084995e-05,
      "loss": 0.0867,
      "step": 47
    },
    {
      "epoch": 0.1875,
      "grad_norm": 0.1595524698495865,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.1163,
      "step": 48
    },
    {
      "epoch": 0.19140625,
      "grad_norm": 0.2963557243347168,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 0.0805,
      "step": 49
    },
    {
      "epoch": 0.1953125,
      "grad_norm": 0.09695882350206375,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.0806,
      "step": 50
    },
    {
      "epoch": 0.19921875,
      "grad_norm": 0.14281536638736725,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.0742,
      "step": 51
    },
    {
      "epoch": 0.203125,
      "grad_norm": 0.1024247333407402,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.0903,
      "step": 52
    },
    {
      "epoch": 0.20703125,
      "grad_norm": 0.0918855369091034,
      "learning_rate": 5.348782368720626e-05,
      "loss": 0.0718,
      "step": 53
    },
    {
      "epoch": 0.2109375,
      "grad_norm": 0.1059468612074852,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.0716,
      "step": 54
    },
    {
      "epoch": 0.2109375,
      "eval_loss": 0.08964384347200394,
      "eval_runtime": 37.691,
      "eval_samples_per_second": 5.731,
      "eval_steps_per_second": 0.716,
      "step": 54
    },
    {
      "epoch": 0.21484375,
      "grad_norm": 0.10198230296373367,
      "learning_rate": 5e-05,
      "loss": 0.0917,
      "step": 55
    },
    {
      "epoch": 0.21875,
      "grad_norm": 0.11491645872592926,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.0781,
      "step": 56
    },
    {
      "epoch": 0.22265625,
      "grad_norm": 0.1265731006860733,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 0.0994,
      "step": 57
    },
    {
      "epoch": 0.2265625,
      "grad_norm": 0.11908170580863953,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.0937,
      "step": 58
    },
    {
      "epoch": 0.23046875,
      "grad_norm": 0.1450977921485901,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 0.1074,
      "step": 59
    },
    {
      "epoch": 0.234375,
      "grad_norm": 0.09565068036317825,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.0856,
      "step": 60
    },
    {
      "epoch": 0.23828125,
      "grad_norm": 0.11376962810754776,
      "learning_rate": 3.960441545911204e-05,
      "loss": 0.0783,
      "step": 61
    },
    {
      "epoch": 0.2421875,
      "grad_norm": 0.1292828917503357,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.075,
      "step": 62
    },
    {
      "epoch": 0.24609375,
      "grad_norm": 0.13481825590133667,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 0.0983,
      "step": 63
    },
    {
      "epoch": 0.24609375,
      "eval_loss": 0.08666232973337173,
      "eval_runtime": 37.6793,
      "eval_samples_per_second": 5.733,
      "eval_steps_per_second": 0.717,
      "step": 63
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.11709572374820709,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0946,
      "step": 64
    },
    {
      "epoch": 0.25390625,
      "grad_norm": 0.12389571219682693,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.0922,
      "step": 65
    },
    {
      "epoch": 0.2578125,
      "grad_norm": 0.09427645802497864,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.0862,
      "step": 66
    },
    {
      "epoch": 0.26171875,
      "grad_norm": 0.06743673980236053,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 0.0659,
      "step": 67
    },
    {
      "epoch": 0.265625,
      "grad_norm": 0.09126238524913788,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.0738,
      "step": 68
    },
    {
      "epoch": 0.26953125,
      "grad_norm": 0.07526295632123947,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 0.0663,
      "step": 69
    },
    {
      "epoch": 0.2734375,
      "grad_norm": 0.10929597169160843,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0779,
      "step": 70
    },
    {
      "epoch": 0.27734375,
      "grad_norm": 0.15462149679660797,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.1058,
      "step": 71
    },
    {
      "epoch": 0.28125,
      "grad_norm": 0.08204536885023117,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.0645,
      "step": 72
    },
    {
      "epoch": 0.28125,
      "eval_loss": 0.08572184294462204,
      "eval_runtime": 37.6882,
      "eval_samples_per_second": 5.731,
      "eval_steps_per_second": 0.716,
      "step": 72
    },
    {
      "epoch": 0.28515625,
      "grad_norm": 0.13382871448993683,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.0863,
      "step": 73
    },
    {
      "epoch": 0.2890625,
      "grad_norm": 0.11069894582033157,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.0793,
      "step": 74
    },
    {
      "epoch": 0.29296875,
      "grad_norm": 0.14329516887664795,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.1133,
      "step": 75
    },
    {
      "epoch": 0.296875,
      "grad_norm": 0.1122271865606308,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.0951,
      "step": 76
    },
    {
      "epoch": 0.30078125,
      "grad_norm": 0.0981534868478775,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.0696,
      "step": 77
    },
    {
      "epoch": 0.3046875,
      "grad_norm": 0.11013442277908325,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.0881,
      "step": 78
    },
    {
      "epoch": 0.30859375,
      "grad_norm": 0.1381031721830368,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.1089,
      "step": 79
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.1080598533153534,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.0956,
      "step": 80
    },
    {
      "epoch": 0.31640625,
      "grad_norm": 0.13694408535957336,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.0884,
      "step": 81
    },
    {
      "epoch": 0.31640625,
      "eval_loss": 0.0842214897274971,
      "eval_runtime": 37.6747,
      "eval_samples_per_second": 5.733,
      "eval_steps_per_second": 0.717,
      "step": 81
    },
    {
      "epoch": 0.3203125,
      "grad_norm": 0.11448404937982559,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0717,
      "step": 82
    },
    {
      "epoch": 0.32421875,
      "grad_norm": 0.09915163367986679,
      "learning_rate": 8.548121372247918e-06,
      "loss": 0.0859,
      "step": 83
    },
    {
      "epoch": 0.328125,
      "grad_norm": 0.13431188464164734,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.1183,
      "step": 84
    },
    {
      "epoch": 0.33203125,
      "grad_norm": 0.1266343742609024,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.0907,
      "step": 85
    },
    {
      "epoch": 0.3359375,
      "grad_norm": 0.08939848095178604,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.0936,
      "step": 86
    },
    {
      "epoch": 0.33984375,
      "grad_norm": 0.08177338540554047,
      "learning_rate": 5.060297685041659e-06,
      "loss": 0.081,
      "step": 87
    },
    {
      "epoch": 0.34375,
      "grad_norm": 0.11731616407632828,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.0617,
      "step": 88
    },
    {
      "epoch": 0.34765625,
      "grad_norm": 0.14580127596855164,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.1243,
      "step": 89
    },
    {
      "epoch": 0.3515625,
      "grad_norm": 0.125287264585495,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.0565,
      "step": 90
    },
    {
      "epoch": 0.3515625,
      "eval_loss": 0.08387485146522522,
      "eval_runtime": 37.6798,
      "eval_samples_per_second": 5.733,
      "eval_steps_per_second": 0.717,
      "step": 90
    },
    {
      "epoch": 0.35546875,
      "grad_norm": 0.100253626704216,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.0871,
      "step": 91
    },
    {
      "epoch": 0.359375,
      "grad_norm": 0.09870755672454834,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.0835,
      "step": 92
    },
    {
      "epoch": 0.36328125,
      "grad_norm": 0.1428849697113037,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 0.0803,
      "step": 93
    },
    {
      "epoch": 0.3671875,
      "grad_norm": 0.09685681015253067,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.0927,
      "step": 94
    },
    {
      "epoch": 0.37109375,
      "grad_norm": 0.10489041358232498,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.0825,
      "step": 95
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.09282656759023666,
      "learning_rate": 4.865965629214819e-07,
      "loss": 0.0892,
      "step": 96
    },
    {
      "epoch": 0.37890625,
      "grad_norm": 0.1284118890762329,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.0695,
      "step": 97
    },
    {
      "epoch": 0.3828125,
      "grad_norm": 0.08799555897712708,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.0832,
      "step": 98
    },
    {
      "epoch": 0.38671875,
      "grad_norm": 0.10245624929666519,
      "learning_rate": 3.04586490452119e-08,
      "loss": 0.0673,
      "step": 99
    },
    {
      "epoch": 0.38671875,
      "eval_loss": 0.08374549448490143,
      "eval_runtime": 37.6733,
      "eval_samples_per_second": 5.733,
      "eval_steps_per_second": 0.717,
      "step": 99
    },
    {
      "epoch": 0.390625,
      "grad_norm": 0.1315760463476181,
      "learning_rate": 0.0,
      "loss": 0.0925,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1271654692880384e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}