{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.13531799729364005,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0013531799729364006,
      "grad_norm": 8.728355407714844,
      "learning_rate": 1e-05,
      "loss": 1.6889,
      "step": 1
    },
    {
      "epoch": 0.0013531799729364006,
      "eval_loss": 1.6367640495300293,
      "eval_runtime": 33.1413,
      "eval_samples_per_second": 18.798,
      "eval_steps_per_second": 2.354,
      "step": 1
    },
    {
      "epoch": 0.0027063599458728013,
      "grad_norm": 8.341392517089844,
      "learning_rate": 2e-05,
      "loss": 1.7952,
      "step": 2
    },
    {
      "epoch": 0.0040595399188092015,
      "grad_norm": 8.935415267944336,
      "learning_rate": 3e-05,
      "loss": 1.5444,
      "step": 3
    },
    {
      "epoch": 0.005412719891745603,
      "grad_norm": 7.361211776733398,
      "learning_rate": 4e-05,
      "loss": 1.452,
      "step": 4
    },
    {
      "epoch": 0.006765899864682003,
      "grad_norm": 5.186268329620361,
      "learning_rate": 5e-05,
      "loss": 1.3568,
      "step": 5
    },
    {
      "epoch": 0.008119079837618403,
      "grad_norm": 4.2360663414001465,
      "learning_rate": 6e-05,
      "loss": 1.1432,
      "step": 6
    },
    {
      "epoch": 0.009472259810554804,
      "grad_norm": 3.4211342334747314,
      "learning_rate": 7e-05,
      "loss": 0.8908,
      "step": 7
    },
    {
      "epoch": 0.010825439783491205,
      "grad_norm": 2.8270256519317627,
      "learning_rate": 8e-05,
      "loss": 0.7026,
      "step": 8
    },
    {
      "epoch": 0.012178619756427604,
      "grad_norm": 4.626265525817871,
      "learning_rate": 9e-05,
      "loss": 0.7591,
      "step": 9
    },
    {
      "epoch": 0.012178619756427604,
      "eval_loss": 0.4093446731567383,
      "eval_runtime": 32.5672,
      "eval_samples_per_second": 19.13,
      "eval_steps_per_second": 2.395,
      "step": 9
    },
    {
      "epoch": 0.013531799729364006,
      "grad_norm": 1.9083399772644043,
      "learning_rate": 0.0001,
      "loss": 0.3278,
      "step": 10
    },
    {
      "epoch": 0.014884979702300407,
      "grad_norm": 2.498229503631592,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.4199,
      "step": 11
    },
    {
      "epoch": 0.016238159675236806,
      "grad_norm": 1.37482750415802,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.1413,
      "step": 12
    },
    {
      "epoch": 0.017591339648173207,
      "grad_norm": 1.1291427612304688,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.137,
      "step": 13
    },
    {
      "epoch": 0.018944519621109608,
      "grad_norm": 1.5424270629882812,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.1136,
      "step": 14
    },
    {
      "epoch": 0.02029769959404601,
      "grad_norm": 1.4762475490570068,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.0587,
      "step": 15
    },
    {
      "epoch": 0.02165087956698241,
      "grad_norm": 0.367654025554657,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.0216,
      "step": 16
    },
    {
      "epoch": 0.023004059539918808,
      "grad_norm": 1.203076958656311,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.0476,
      "step": 17
    },
    {
      "epoch": 0.02435723951285521,
      "grad_norm": 0.837530791759491,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.0208,
      "step": 18
    },
    {
      "epoch": 0.02435723951285521,
      "eval_loss": 0.03585303574800491,
      "eval_runtime": 32.5664,
      "eval_samples_per_second": 19.13,
      "eval_steps_per_second": 2.395,
      "step": 18
    },
    {
      "epoch": 0.02571041948579161,
      "grad_norm": 0.6422433853149414,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0111,
      "step": 19
    },
    {
      "epoch": 0.02706359945872801,
      "grad_norm": 1.1846771240234375,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0873,
      "step": 20
    },
    {
      "epoch": 0.028416779431664412,
      "grad_norm": 0.19324840605258942,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.0102,
      "step": 21
    },
    {
      "epoch": 0.029769959404600813,
      "grad_norm": 0.296364426612854,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0174,
      "step": 22
    },
    {
      "epoch": 0.03112313937753721,
      "grad_norm": 0.7968310117721558,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.047,
      "step": 23
    },
    {
      "epoch": 0.03247631935047361,
      "grad_norm": 0.12593711912631989,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0061,
      "step": 24
    },
    {
      "epoch": 0.03382949932341001,
      "grad_norm": 0.28613799810409546,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0075,
      "step": 25
    },
    {
      "epoch": 0.035182679296346414,
      "grad_norm": 0.23756200075149536,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0165,
      "step": 26
    },
    {
      "epoch": 0.036535859269282815,
      "grad_norm": 0.12324630469083786,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.0029,
      "step": 27
    },
    {
      "epoch": 0.036535859269282815,
      "eval_loss": 0.020997991785407066,
      "eval_runtime": 32.6042,
      "eval_samples_per_second": 19.108,
      "eval_steps_per_second": 2.392,
      "step": 27
    },
    {
      "epoch": 0.037889039242219216,
      "grad_norm": 0.08755990862846375,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0074,
      "step": 28
    },
    {
      "epoch": 0.03924221921515562,
      "grad_norm": 0.18577858805656433,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0084,
      "step": 29
    },
    {
      "epoch": 0.04059539918809202,
      "grad_norm": 0.09318247437477112,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.0049,
      "step": 30
    },
    {
      "epoch": 0.04194857916102842,
      "grad_norm": 0.2695228159427643,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.0159,
      "step": 31
    },
    {
      "epoch": 0.04330175913396482,
      "grad_norm": 0.08464854955673218,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.0065,
      "step": 32
    },
    {
      "epoch": 0.044654939106901215,
      "grad_norm": 0.17591221630573273,
      "learning_rate": 8.473291852294987e-05,
      "loss": 0.0049,
      "step": 33
    },
    {
      "epoch": 0.046008119079837616,
      "grad_norm": 0.2792809009552002,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.0171,
      "step": 34
    },
    {
      "epoch": 0.04736129905277402,
      "grad_norm": 0.12685370445251465,
      "learning_rate": 8.213938048432697e-05,
      "loss": 0.0043,
      "step": 35
    },
    {
      "epoch": 0.04871447902571042,
      "grad_norm": 0.5813450813293457,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.0135,
      "step": 36
    },
    {
      "epoch": 0.04871447902571042,
      "eval_loss": 0.019910942763090134,
      "eval_runtime": 32.5995,
      "eval_samples_per_second": 19.111,
      "eval_steps_per_second": 2.393,
      "step": 36
    },
    {
      "epoch": 0.05006765899864682,
      "grad_norm": 0.08670137077569962,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.0062,
      "step": 37
    },
    {
      "epoch": 0.05142083897158322,
      "grad_norm": 0.09463904052972794,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.0021,
      "step": 38
    },
    {
      "epoch": 0.05277401894451962,
      "grad_norm": 0.05870044231414795,
      "learning_rate": 7.649596321166024e-05,
      "loss": 0.006,
      "step": 39
    },
    {
      "epoch": 0.05412719891745602,
      "grad_norm": 0.09653154015541077,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.007,
      "step": 40
    },
    {
      "epoch": 0.05548037889039242,
      "grad_norm": 0.06797784566879272,
      "learning_rate": 7.347357813929454e-05,
      "loss": 0.0094,
      "step": 41
    },
    {
      "epoch": 0.056833558863328824,
      "grad_norm": 0.9252069592475891,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.0755,
      "step": 42
    },
    {
      "epoch": 0.058186738836265225,
      "grad_norm": 0.08743719756603241,
      "learning_rate": 7.033683215379002e-05,
      "loss": 0.007,
      "step": 43
    },
    {
      "epoch": 0.05953991880920163,
      "grad_norm": 0.22028687596321106,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.0107,
      "step": 44
    },
    {
      "epoch": 0.06089309878213803,
      "grad_norm": 0.24319298565387726,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.0168,
      "step": 45
    },
    {
      "epoch": 0.06089309878213803,
      "eval_loss": 0.019319333136081696,
      "eval_runtime": 32.603,
      "eval_samples_per_second": 19.109,
      "eval_steps_per_second": 2.392,
      "step": 45
    },
    {
      "epoch": 0.06224627875507442,
      "grad_norm": 0.6694943904876709,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.081,
      "step": 46
    },
    {
      "epoch": 0.06359945872801083,
      "grad_norm": 0.06383314728736877,
      "learning_rate": 6.378186779084995e-05,
      "loss": 0.0016,
      "step": 47
    },
    {
      "epoch": 0.06495263870094722,
      "grad_norm": 0.05941140279173851,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.0045,
      "step": 48
    },
    {
      "epoch": 0.06630581867388363,
      "grad_norm": 0.19987758994102478,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 0.0132,
      "step": 49
    },
    {
      "epoch": 0.06765899864682003,
      "grad_norm": 0.11729590594768524,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.0058,
      "step": 50
    },
    {
      "epoch": 0.06901217861975643,
      "grad_norm": 0.0919797345995903,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.0045,
      "step": 51
    },
    {
      "epoch": 0.07036535859269283,
      "grad_norm": 0.05789254978299141,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.0048,
      "step": 52
    },
    {
      "epoch": 0.07171853856562922,
      "grad_norm": 0.18244007229804993,
      "learning_rate": 5.348782368720626e-05,
      "loss": 0.0155,
      "step": 53
    },
    {
      "epoch": 0.07307171853856563,
      "grad_norm": 0.6028981804847717,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.1107,
      "step": 54
    },
    {
      "epoch": 0.07307171853856563,
      "eval_loss": 0.018029918894171715,
      "eval_runtime": 32.6237,
      "eval_samples_per_second": 19.097,
      "eval_steps_per_second": 2.391,
      "step": 54
    },
    {
      "epoch": 0.07442489851150202,
      "grad_norm": 0.5009127855300903,
      "learning_rate": 5e-05,
      "loss": 0.0893,
      "step": 55
    },
    {
      "epoch": 0.07577807848443843,
      "grad_norm": 0.04764901474118233,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.0064,
      "step": 56
    },
    {
      "epoch": 0.07713125845737483,
      "grad_norm": 0.8551574945449829,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 0.1503,
      "step": 57
    },
    {
      "epoch": 0.07848443843031123,
      "grad_norm": 0.0777273029088974,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.005,
      "step": 58
    },
    {
      "epoch": 0.07983761840324763,
      "grad_norm": 0.6547033786773682,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 0.1104,
      "step": 59
    },
    {
      "epoch": 0.08119079837618404,
      "grad_norm": 0.15890103578567505,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.0155,
      "step": 60
    },
    {
      "epoch": 0.08254397834912043,
      "grad_norm": 1.0533157587051392,
      "learning_rate": 3.960441545911204e-05,
      "loss": 0.1344,
      "step": 61
    },
    {
      "epoch": 0.08389715832205684,
      "grad_norm": 0.13047480583190918,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.0116,
      "step": 62
    },
    {
      "epoch": 0.08525033829499323,
      "grad_norm": 0.14830543100833893,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 0.0111,
      "step": 63
    },
    {
      "epoch": 0.08525033829499323,
      "eval_loss": 0.01703736186027527,
      "eval_runtime": 32.5933,
      "eval_samples_per_second": 19.114,
      "eval_steps_per_second": 2.393,
      "step": 63
    },
    {
      "epoch": 0.08660351826792964,
      "grad_norm": 0.10914517939090729,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.008,
      "step": 64
    },
    {
      "epoch": 0.08795669824086604,
      "grad_norm": 0.13250844180583954,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.0078,
      "step": 65
    },
    {
      "epoch": 0.08930987821380243,
      "grad_norm": 0.14327938854694366,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.0085,
      "step": 66
    },
    {
      "epoch": 0.09066305818673884,
      "grad_norm": 0.14970730245113373,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 0.0042,
      "step": 67
    },
    {
      "epoch": 0.09201623815967523,
      "grad_norm": 0.08402867615222931,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.0104,
      "step": 68
    },
    {
      "epoch": 0.09336941813261164,
      "grad_norm": 0.08050940930843353,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 0.005,
      "step": 69
    },
    {
      "epoch": 0.09472259810554803,
      "grad_norm": 0.5739089250564575,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0547,
      "step": 70
    },
    {
      "epoch": 0.09607577807848444,
      "grad_norm": 0.11337218433618546,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.0091,
      "step": 71
    },
    {
      "epoch": 0.09742895805142084,
      "grad_norm": 0.06483836472034454,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.0072,
      "step": 72
    },
    {
      "epoch": 0.09742895805142084,
      "eval_loss": 0.015148372389376163,
      "eval_runtime": 32.5938,
      "eval_samples_per_second": 19.114,
      "eval_steps_per_second": 2.393,
      "step": 72
    },
    {
      "epoch": 0.09878213802435724,
      "grad_norm": 0.12975576519966125,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.0114,
      "step": 73
    },
    {
      "epoch": 0.10013531799729364,
      "grad_norm": 0.08989189565181732,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.0041,
      "step": 74
    },
    {
      "epoch": 0.10148849797023005,
      "grad_norm": 0.6734408140182495,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.0926,
      "step": 75
    },
    {
      "epoch": 0.10284167794316644,
      "grad_norm": 0.1009981706738472,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.009,
      "step": 76
    },
    {
      "epoch": 0.10419485791610285,
      "grad_norm": 0.044376980513334274,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.0047,
      "step": 77
    },
    {
      "epoch": 0.10554803788903924,
      "grad_norm": 0.07079390436410904,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.0048,
      "step": 78
    },
    {
      "epoch": 0.10690121786197564,
      "grad_norm": 0.1579139083623886,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.006,
      "step": 79
    },
    {
      "epoch": 0.10825439783491204,
      "grad_norm": 0.3575461208820343,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.0312,
      "step": 80
    },
    {
      "epoch": 0.10960757780784844,
      "grad_norm": 0.054556429386138916,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.0069,
      "step": 81
    },
    {
      "epoch": 0.10960757780784844,
      "eval_loss": 0.013904375955462456,
      "eval_runtime": 32.5975,
      "eval_samples_per_second": 19.112,
      "eval_steps_per_second": 2.393,
      "step": 81
    },
    {
      "epoch": 0.11096075778078485,
      "grad_norm": 0.059846166521310806,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0073,
      "step": 82
    },
    {
      "epoch": 0.11231393775372124,
      "grad_norm": 0.07215581089258194,
      "learning_rate": 8.548121372247918e-06,
      "loss": 0.0051,
      "step": 83
    },
    {
      "epoch": 0.11366711772665765,
      "grad_norm": 0.1450567990541458,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.0122,
      "step": 84
    },
    {
      "epoch": 0.11502029769959404,
      "grad_norm": 0.03491105139255524,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.0043,
      "step": 85
    },
    {
      "epoch": 0.11637347767253045,
      "grad_norm": 0.5953935384750366,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.0621,
      "step": 86
    },
    {
      "epoch": 0.11772665764546685,
      "grad_norm": 0.4230695068836212,
      "learning_rate": 5.060297685041659e-06,
      "loss": 0.0422,
      "step": 87
    },
    {
      "epoch": 0.11907983761840325,
      "grad_norm": 0.033300649374723434,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.0045,
      "step": 88
    },
    {
      "epoch": 0.12043301759133965,
      "grad_norm": 0.30535688996315,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.0485,
      "step": 89
    },
    {
      "epoch": 0.12178619756427606,
      "grad_norm": 0.07103042304515839,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.004,
      "step": 90
    },
    {
      "epoch": 0.12178619756427606,
      "eval_loss": 0.013438088819384575,
      "eval_runtime": 32.5892,
      "eval_samples_per_second": 19.117,
      "eval_steps_per_second": 2.393,
      "step": 90
    },
    {
      "epoch": 0.12313937753721245,
      "grad_norm": 0.11808881163597107,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.0081,
      "step": 91
    },
    {
      "epoch": 0.12449255751014884,
      "grad_norm": 0.033296674489974976,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.0045,
      "step": 92
    },
    {
      "epoch": 0.12584573748308525,
      "grad_norm": 0.19024379551410675,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 0.0157,
      "step": 93
    },
    {
      "epoch": 0.12719891745602166,
      "grad_norm": 0.7022133469581604,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.0563,
      "step": 94
    },
    {
      "epoch": 0.12855209742895804,
      "grad_norm": 0.110652394592762,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.0102,
      "step": 95
    },
    {
      "epoch": 0.12990527740189445,
      "grad_norm": 0.110796257853508,
      "learning_rate": 4.865965629214819e-07,
      "loss": 0.0116,
      "step": 96
    },
    {
      "epoch": 0.13125845737483086,
      "grad_norm": 0.03754672780632973,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.0038,
      "step": 97
    },
    {
      "epoch": 0.13261163734776726,
      "grad_norm": 0.07318272441625595,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.0019,
      "step": 98
    },
    {
      "epoch": 0.13396481732070364,
      "grad_norm": 0.4888281226158142,
      "learning_rate": 3.04586490452119e-08,
      "loss": 0.0511,
      "step": 99
    },
    {
      "epoch": 0.13396481732070364,
      "eval_loss": 0.013342314399778843,
      "eval_runtime": 32.607,
      "eval_samples_per_second": 19.106,
      "eval_steps_per_second": 2.392,
      "step": 99
    },
    {
      "epoch": 0.13531799729364005,
      "grad_norm": 0.3825893998146057,
      "learning_rate": 0.0,
      "loss": 0.0563,
      "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
}