{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004735521144101909,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.735521144101908e-05,
      "grad_norm": 19.782573699951172,
      "learning_rate": 1e-05,
      "loss": 11.7383,
      "step": 1
    },
    {
      "epoch": 4.735521144101908e-05,
      "eval_loss": 11.375913619995117,
      "eval_runtime": 1954.4568,
      "eval_samples_per_second": 9.099,
      "eval_steps_per_second": 1.137,
      "step": 1
    },
    {
      "epoch": 9.471042288203816e-05,
      "grad_norm": 12.204630851745605,
      "learning_rate": 2e-05,
      "loss": 12.498,
      "step": 2
    },
    {
      "epoch": 0.00014206563432305724,
      "grad_norm": 13.953908920288086,
      "learning_rate": 3e-05,
      "loss": 11.9622,
      "step": 3
    },
    {
      "epoch": 0.00018942084576407633,
      "grad_norm": 15.582470893859863,
      "learning_rate": 4e-05,
      "loss": 11.1347,
      "step": 4
    },
    {
      "epoch": 0.0002367760572050954,
      "grad_norm": 11.932799339294434,
      "learning_rate": 5e-05,
      "loss": 10.6174,
      "step": 5
    },
    {
      "epoch": 0.0002841312686461145,
      "grad_norm": 18.4592227935791,
      "learning_rate": 6e-05,
      "loss": 9.4297,
      "step": 6
    },
    {
      "epoch": 0.0003314864800871336,
      "grad_norm": 20.50937843322754,
      "learning_rate": 7e-05,
      "loss": 8.5525,
      "step": 7
    },
    {
      "epoch": 0.00037884169152815265,
      "grad_norm": 19.198875427246094,
      "learning_rate": 8e-05,
      "loss": 8.3855,
      "step": 8
    },
    {
      "epoch": 0.00042619690296917176,
      "grad_norm": 20.136850357055664,
      "learning_rate": 9e-05,
      "loss": 8.5612,
      "step": 9
    },
    {
      "epoch": 0.00042619690296917176,
      "eval_loss": 7.14232063293457,
      "eval_runtime": 1958.7927,
      "eval_samples_per_second": 9.079,
      "eval_steps_per_second": 1.135,
      "step": 9
    },
    {
      "epoch": 0.0004735521144101908,
      "grad_norm": 16.510393142700195,
      "learning_rate": 0.0001,
      "loss": 6.899,
      "step": 10
    },
    {
      "epoch": 0.00052090732585121,
      "grad_norm": 22.659875869750977,
      "learning_rate": 9.99695413509548e-05,
      "loss": 6.3529,
      "step": 11
    },
    {
      "epoch": 0.000568262537292229,
      "grad_norm": 19.63251495361328,
      "learning_rate": 9.987820251299122e-05,
      "loss": 5.9256,
      "step": 12
    },
    {
      "epoch": 0.0006156177487332481,
      "grad_norm": 16.59630584716797,
      "learning_rate": 9.972609476841367e-05,
      "loss": 5.244,
      "step": 13
    },
    {
      "epoch": 0.0006629729601742672,
      "grad_norm": 17.385766983032227,
      "learning_rate": 9.951340343707852e-05,
      "loss": 4.6911,
      "step": 14
    },
    {
      "epoch": 0.0007103281716152863,
      "grad_norm": 13.79629135131836,
      "learning_rate": 9.924038765061042e-05,
      "loss": 3.633,
      "step": 15
    },
    {
      "epoch": 0.0007576833830563053,
      "grad_norm": 14.321866035461426,
      "learning_rate": 9.890738003669029e-05,
      "loss": 3.4303,
      "step": 16
    },
    {
      "epoch": 0.0008050385944973244,
      "grad_norm": 16.999752044677734,
      "learning_rate": 9.851478631379982e-05,
      "loss": 2.8823,
      "step": 17
    },
    {
      "epoch": 0.0008523938059383435,
      "grad_norm": 15.764219284057617,
      "learning_rate": 9.806308479691595e-05,
      "loss": 2.8359,
      "step": 18
    },
    {
      "epoch": 0.0008523938059383435,
      "eval_loss": 2.567370891571045,
      "eval_runtime": 1958.998,
      "eval_samples_per_second": 9.078,
      "eval_steps_per_second": 1.135,
      "step": 18
    },
    {
      "epoch": 0.0008997490173793626,
      "grad_norm": 13.595306396484375,
      "learning_rate": 9.755282581475769e-05,
      "loss": 2.534,
      "step": 19
    },
    {
      "epoch": 0.0009471042288203816,
      "grad_norm": 12.287227630615234,
      "learning_rate": 9.698463103929542e-05,
      "loss": 3.0072,
      "step": 20
    },
    {
      "epoch": 0.0009944594402614007,
      "grad_norm": 13.478318214416504,
      "learning_rate": 9.635919272833938e-05,
      "loss": 2.1003,
      "step": 21
    },
    {
      "epoch": 0.00104181465170242,
      "grad_norm": 8.168448448181152,
      "learning_rate": 9.567727288213005e-05,
      "loss": 2.0879,
      "step": 22
    },
    {
      "epoch": 0.001089169863143439,
      "grad_norm": 15.11435317993164,
      "learning_rate": 9.493970231495835e-05,
      "loss": 2.2822,
      "step": 23
    },
    {
      "epoch": 0.001136525074584458,
      "grad_norm": 9.29191780090332,
      "learning_rate": 9.414737964294636e-05,
      "loss": 2.0676,
      "step": 24
    },
    {
      "epoch": 0.0011838802860254772,
      "grad_norm": 8.91670036315918,
      "learning_rate": 9.330127018922194e-05,
      "loss": 2.274,
      "step": 25
    },
    {
      "epoch": 0.0012312354974664962,
      "grad_norm": 14.876287460327148,
      "learning_rate": 9.24024048078213e-05,
      "loss": 2.9341,
      "step": 26
    },
    {
      "epoch": 0.0012785907089075152,
      "grad_norm": 10.619921684265137,
      "learning_rate": 9.145187862775209e-05,
      "loss": 2.7626,
      "step": 27
    },
    {
      "epoch": 0.0012785907089075152,
      "eval_loss": 2.165811061859131,
      "eval_runtime": 1958.6671,
      "eval_samples_per_second": 9.079,
      "eval_steps_per_second": 1.135,
      "step": 27
    },
    {
      "epoch": 0.0013259459203485344,
      "grad_norm": 10.87781810760498,
      "learning_rate": 9.045084971874738e-05,
      "loss": 2.246,
      "step": 28
    },
    {
      "epoch": 0.0013733011317895534,
      "grad_norm": 8.4467134475708,
      "learning_rate": 8.940053768033609e-05,
      "loss": 1.8875,
      "step": 29
    },
    {
      "epoch": 0.0014206563432305726,
      "grad_norm": 11.084881782531738,
      "learning_rate": 8.83022221559489e-05,
      "loss": 1.9841,
      "step": 30
    },
    {
      "epoch": 0.0014680115546715916,
      "grad_norm": 7.075559616088867,
      "learning_rate": 8.715724127386972e-05,
      "loss": 2.0835,
      "step": 31
    },
    {
      "epoch": 0.0015153667661126106,
      "grad_norm": 8.6925048828125,
      "learning_rate": 8.596699001693255e-05,
      "loss": 2.0031,
      "step": 32
    },
    {
      "epoch": 0.0015627219775536298,
      "grad_norm": 7.5939507484436035,
      "learning_rate": 8.473291852294987e-05,
      "loss": 2.1421,
      "step": 33
    },
    {
      "epoch": 0.0016100771889946488,
      "grad_norm": 6.9573516845703125,
      "learning_rate": 8.345653031794292e-05,
      "loss": 2.0526,
      "step": 34
    },
    {
      "epoch": 0.001657432400435668,
      "grad_norm": 9.302016258239746,
      "learning_rate": 8.213938048432697e-05,
      "loss": 1.7914,
      "step": 35
    },
    {
      "epoch": 0.001704787611876687,
      "grad_norm": 10.62584400177002,
      "learning_rate": 8.07830737662829e-05,
      "loss": 2.3807,
      "step": 36
    },
    {
      "epoch": 0.001704787611876687,
      "eval_loss": 2.0060789585113525,
      "eval_runtime": 1958.2153,
      "eval_samples_per_second": 9.081,
      "eval_steps_per_second": 1.135,
      "step": 36
    },
    {
      "epoch": 0.001752142823317706,
      "grad_norm": 6.709832191467285,
      "learning_rate": 7.938926261462366e-05,
      "loss": 1.7162,
      "step": 37
    },
    {
      "epoch": 0.0017994980347587253,
      "grad_norm": 6.176967620849609,
      "learning_rate": 7.795964517353735e-05,
      "loss": 1.9218,
      "step": 38
    },
    {
      "epoch": 0.0018468532461997443,
      "grad_norm": 8.372739791870117,
      "learning_rate": 7.649596321166024e-05,
      "loss": 1.879,
      "step": 39
    },
    {
      "epoch": 0.0018942084576407633,
      "grad_norm": 7.651803493499756,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.7878,
      "step": 40
    },
    {
      "epoch": 0.0019415636690817825,
      "grad_norm": 10.77332592010498,
      "learning_rate": 7.347357813929454e-05,
      "loss": 1.783,
      "step": 41
    },
    {
      "epoch": 0.0019889188805228015,
      "grad_norm": 6.215115070343018,
      "learning_rate": 7.191855733945387e-05,
      "loss": 1.5281,
      "step": 42
    },
    {
      "epoch": 0.0020362740919638205,
      "grad_norm": 7.770496845245361,
      "learning_rate": 7.033683215379002e-05,
      "loss": 1.4795,
      "step": 43
    },
    {
      "epoch": 0.00208362930340484,
      "grad_norm": 5.736296653747559,
      "learning_rate": 6.873032967079561e-05,
      "loss": 1.9338,
      "step": 44
    },
    {
      "epoch": 0.002130984514845859,
      "grad_norm": 7.136412143707275,
      "learning_rate": 6.710100716628344e-05,
      "loss": 2.0324,
      "step": 45
    },
    {
      "epoch": 0.002130984514845859,
      "eval_loss": 1.9209612607955933,
      "eval_runtime": 1957.0123,
      "eval_samples_per_second": 9.087,
      "eval_steps_per_second": 1.136,
      "step": 45
    },
    {
      "epoch": 0.002178339726286878,
      "grad_norm": 6.550395965576172,
      "learning_rate": 6.545084971874738e-05,
      "loss": 2.0345,
      "step": 46
    },
    {
      "epoch": 0.002225694937727897,
      "grad_norm": 7.430251598358154,
      "learning_rate": 6.378186779084995e-05,
      "loss": 2.5047,
      "step": 47
    },
    {
      "epoch": 0.002273050149168916,
      "grad_norm": 6.692016124725342,
      "learning_rate": 6.209609477998338e-05,
      "loss": 1.957,
      "step": 48
    },
    {
      "epoch": 0.002320405360609935,
      "grad_norm": 6.386138439178467,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 1.9522,
      "step": 49
    },
    {
      "epoch": 0.0023677605720509544,
      "grad_norm": 5.718717098236084,
      "learning_rate": 5.868240888334653e-05,
      "loss": 1.6807,
      "step": 50
    },
    {
      "epoch": 0.0024151157834919734,
      "grad_norm": 6.81873893737793,
      "learning_rate": 5.695865504800327e-05,
      "loss": 2.0401,
      "step": 51
    },
    {
      "epoch": 0.0024624709949329924,
      "grad_norm": 13.12895679473877,
      "learning_rate": 5.522642316338268e-05,
      "loss": 2.5685,
      "step": 52
    },
    {
      "epoch": 0.0025098262063740114,
      "grad_norm": 6.2279372215271,
      "learning_rate": 5.348782368720626e-05,
      "loss": 1.9409,
      "step": 53
    },
    {
      "epoch": 0.0025571814178150304,
      "grad_norm": 7.927605152130127,
      "learning_rate": 5.174497483512506e-05,
      "loss": 1.8451,
      "step": 54
    },
    {
      "epoch": 0.0025571814178150304,
      "eval_loss": 1.9416064023971558,
      "eval_runtime": 1958.1372,
      "eval_samples_per_second": 9.082,
      "eval_steps_per_second": 1.135,
      "step": 54
    },
    {
      "epoch": 0.00260453662925605,
      "grad_norm": 6.186990261077881,
      "learning_rate": 5e-05,
      "loss": 2.1235,
      "step": 55
    },
    {
      "epoch": 0.002651891840697069,
      "grad_norm": 9.007567405700684,
      "learning_rate": 4.825502516487497e-05,
      "loss": 1.8976,
      "step": 56
    },
    {
      "epoch": 0.002699247052138088,
      "grad_norm": 12.486298561096191,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 1.9031,
      "step": 57
    },
    {
      "epoch": 0.002746602263579107,
      "grad_norm": 5.143292427062988,
      "learning_rate": 4.477357683661734e-05,
      "loss": 1.6532,
      "step": 58
    },
    {
      "epoch": 0.002793957475020126,
      "grad_norm": 6.0199689865112305,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 1.8914,
      "step": 59
    },
    {
      "epoch": 0.0028413126864611452,
      "grad_norm": 5.626400947570801,
      "learning_rate": 4.131759111665349e-05,
      "loss": 1.8248,
      "step": 60
    },
    {
      "epoch": 0.0028886678979021642,
      "grad_norm": 6.385285377502441,
      "learning_rate": 3.960441545911204e-05,
      "loss": 2.0729,
      "step": 61
    },
    {
      "epoch": 0.0029360231093431832,
      "grad_norm": 6.212454319000244,
      "learning_rate": 3.790390522001662e-05,
      "loss": 1.7437,
      "step": 62
    },
    {
      "epoch": 0.0029833783207842022,
      "grad_norm": 7.248876094818115,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 1.7327,
      "step": 63
    },
    {
      "epoch": 0.0029833783207842022,
      "eval_loss": 1.874548316001892,
      "eval_runtime": 1956.879,
      "eval_samples_per_second": 9.087,
      "eval_steps_per_second": 1.136,
      "step": 63
    },
    {
      "epoch": 0.0030307335322252212,
      "grad_norm": 9.840112686157227,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 2.2445,
      "step": 64
    },
    {
      "epoch": 0.0030780887436662407,
      "grad_norm": 5.413729190826416,
      "learning_rate": 3.289899283371657e-05,
      "loss": 1.6007,
      "step": 65
    },
    {
      "epoch": 0.0031254439551072597,
      "grad_norm": 12.843977928161621,
      "learning_rate": 3.12696703292044e-05,
      "loss": 2.075,
      "step": 66
    },
    {
      "epoch": 0.0031727991665482787,
      "grad_norm": 16.38304328918457,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 1.5473,
      "step": 67
    },
    {
      "epoch": 0.0032201543779892977,
      "grad_norm": 9.26452922821045,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 1.893,
      "step": 68
    },
    {
      "epoch": 0.0032675095894303167,
      "grad_norm": 19.458349227905273,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 1.9362,
      "step": 69
    },
    {
      "epoch": 0.003314864800871336,
      "grad_norm": 5.507326126098633,
      "learning_rate": 2.500000000000001e-05,
      "loss": 1.8801,
      "step": 70
    },
    {
      "epoch": 0.003362220012312355,
      "grad_norm": 20.090539932250977,
      "learning_rate": 2.350403678833976e-05,
      "loss": 2.1517,
      "step": 71
    },
    {
      "epoch": 0.003409575223753374,
      "grad_norm": 13.709304809570312,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 2.2432,
      "step": 72
    },
    {
      "epoch": 0.003409575223753374,
      "eval_loss": 1.9207587242126465,
      "eval_runtime": 1957.5396,
      "eval_samples_per_second": 9.084,
      "eval_steps_per_second": 1.136,
      "step": 72
    },
    {
      "epoch": 0.003456930435194393,
      "grad_norm": 15.840682029724121,
      "learning_rate": 2.061073738537635e-05,
      "loss": 1.8682,
      "step": 73
    },
    {
      "epoch": 0.003504285646635412,
      "grad_norm": 8.384814262390137,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 1.7125,
      "step": 74
    },
    {
      "epoch": 0.003551640858076431,
      "grad_norm": 8.691859245300293,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 1.3733,
      "step": 75
    },
    {
      "epoch": 0.0035989960695174505,
      "grad_norm": 5.541296482086182,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 1.9249,
      "step": 76
    },
    {
      "epoch": 0.0036463512809584695,
      "grad_norm": 5.917468070983887,
      "learning_rate": 1.526708147705013e-05,
      "loss": 1.8697,
      "step": 77
    },
    {
      "epoch": 0.0036937064923994885,
      "grad_norm": 10.085787773132324,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 1.8634,
      "step": 78
    },
    {
      "epoch": 0.0037410617038405075,
      "grad_norm": 5.36349630355835,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 1.8555,
      "step": 79
    },
    {
      "epoch": 0.0037884169152815265,
      "grad_norm": 7.90194845199585,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 1.8127,
      "step": 80
    },
    {
      "epoch": 0.003835772126722546,
      "grad_norm": 7.089127540588379,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 2.1554,
      "step": 81
    },
    {
      "epoch": 0.003835772126722546,
      "eval_loss": 1.8363431692123413,
      "eval_runtime": 1963.8665,
      "eval_samples_per_second": 9.055,
      "eval_steps_per_second": 1.132,
      "step": 81
    },
    {
      "epoch": 0.003883127338163565,
      "grad_norm": 6.435274600982666,
      "learning_rate": 9.549150281252633e-06,
      "loss": 1.8305,
      "step": 82
    },
    {
      "epoch": 0.003930482549604584,
      "grad_norm": 5.313261032104492,
      "learning_rate": 8.548121372247918e-06,
      "loss": 1.7793,
      "step": 83
    },
    {
      "epoch": 0.003977837761045603,
      "grad_norm": 5.327849388122559,
      "learning_rate": 7.597595192178702e-06,
      "loss": 1.8134,
      "step": 84
    },
    {
      "epoch": 0.004025192972486622,
      "grad_norm": 8.038222312927246,
      "learning_rate": 6.698729810778065e-06,
      "loss": 2.091,
      "step": 85
    },
    {
      "epoch": 0.004072548183927641,
      "grad_norm": 6.482438564300537,
      "learning_rate": 5.852620357053651e-06,
      "loss": 2.0731,
      "step": 86
    },
    {
      "epoch": 0.00411990339536866,
      "grad_norm": 8.912972450256348,
      "learning_rate": 5.060297685041659e-06,
      "loss": 2.1404,
      "step": 87
    },
    {
      "epoch": 0.00416725860680968,
      "grad_norm": 6.126862049102783,
      "learning_rate": 4.322727117869951e-06,
      "loss": 1.9181,
      "step": 88
    },
    {
      "epoch": 0.004214613818250699,
      "grad_norm": 7.689204216003418,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 1.7556,
      "step": 89
    },
    {
      "epoch": 0.004261969029691718,
      "grad_norm": 7.858786582946777,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 1.848,
      "step": 90
    },
    {
      "epoch": 0.004261969029691718,
      "eval_loss": 1.8280308246612549,
      "eval_runtime": 1965.1163,
      "eval_samples_per_second": 9.049,
      "eval_steps_per_second": 1.131,
      "step": 90
    },
    {
      "epoch": 0.004309324241132737,
      "grad_norm": 5.3724493980407715,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 1.8187,
      "step": 91
    },
    {
      "epoch": 0.004356679452573756,
      "grad_norm": 10.32592487335205,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 1.8049,
      "step": 92
    },
    {
      "epoch": 0.004404034664014775,
      "grad_norm": 6.439472198486328,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 1.6192,
      "step": 93
    },
    {
      "epoch": 0.004451389875455794,
      "grad_norm": 9.626279830932617,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 2.053,
      "step": 94
    },
    {
      "epoch": 0.004498745086896813,
      "grad_norm": 5.963990688323975,
      "learning_rate": 7.596123493895991e-07,
      "loss": 1.8517,
      "step": 95
    },
    {
      "epoch": 0.004546100298337832,
      "grad_norm": 5.205244541168213,
      "learning_rate": 4.865965629214819e-07,
      "loss": 2.0463,
      "step": 96
    },
    {
      "epoch": 0.004593455509778851,
      "grad_norm": 5.28290319442749,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 1.7786,
      "step": 97
    },
    {
      "epoch": 0.00464081072121987,
      "grad_norm": 5.157119274139404,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 1.855,
      "step": 98
    },
    {
      "epoch": 0.00468816593266089,
      "grad_norm": 5.792222499847412,
      "learning_rate": 3.04586490452119e-08,
      "loss": 1.8657,
      "step": 99
    },
    {
      "epoch": 0.00468816593266089,
      "eval_loss": 1.8215854167938232,
      "eval_runtime": 1960.5219,
      "eval_samples_per_second": 9.071,
      "eval_steps_per_second": 1.134,
      "step": 99
    },
    {
      "epoch": 0.004735521144101909,
      "grad_norm": 6.510435581207275,
      "learning_rate": 0.0,
      "loss": 1.9326,
      "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": 7.41887283560448e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}