{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06872852233676977,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006872852233676976,
      "grad_norm": 0.4530450403690338,
      "learning_rate": 1e-05,
      "loss": 3.5055,
      "step": 1
    },
    {
      "epoch": 0.0006872852233676976,
      "eval_loss": 1.7271678447723389,
      "eval_runtime": 252.7415,
      "eval_samples_per_second": 4.851,
      "eval_steps_per_second": 0.609,
      "step": 1
    },
    {
      "epoch": 0.0013745704467353953,
      "grad_norm": 0.4202934205532074,
      "learning_rate": 2e-05,
      "loss": 3.4948,
      "step": 2
    },
    {
      "epoch": 0.002061855670103093,
      "grad_norm": 0.4171874225139618,
      "learning_rate": 3e-05,
      "loss": 3.4219,
      "step": 3
    },
    {
      "epoch": 0.0027491408934707906,
      "grad_norm": 0.5224863290786743,
      "learning_rate": 4e-05,
      "loss": 3.3726,
      "step": 4
    },
    {
      "epoch": 0.003436426116838488,
      "grad_norm": 0.4832131862640381,
      "learning_rate": 5e-05,
      "loss": 3.3309,
      "step": 5
    },
    {
      "epoch": 0.004123711340206186,
      "grad_norm": 0.5266788601875305,
      "learning_rate": 6e-05,
      "loss": 3.4297,
      "step": 6
    },
    {
      "epoch": 0.004810996563573883,
      "grad_norm": 0.5814648270606995,
      "learning_rate": 7e-05,
      "loss": 3.3866,
      "step": 7
    },
    {
      "epoch": 0.005498281786941581,
      "grad_norm": 0.7436932921409607,
      "learning_rate": 8e-05,
      "loss": 3.3173,
      "step": 8
    },
    {
      "epoch": 0.006185567010309278,
      "grad_norm": 0.7650415897369385,
      "learning_rate": 9e-05,
      "loss": 3.0903,
      "step": 9
    },
    {
      "epoch": 0.006185567010309278,
      "eval_loss": 1.549910068511963,
      "eval_runtime": 253.3835,
      "eval_samples_per_second": 4.839,
      "eval_steps_per_second": 0.608,
      "step": 9
    },
    {
      "epoch": 0.006872852233676976,
      "grad_norm": 0.8187753558158875,
      "learning_rate": 0.0001,
      "loss": 3.2222,
      "step": 10
    },
    {
      "epoch": 0.007560137457044674,
      "grad_norm": 0.8003529906272888,
      "learning_rate": 9.99695413509548e-05,
      "loss": 2.7961,
      "step": 11
    },
    {
      "epoch": 0.008247422680412371,
      "grad_norm": 0.813484251499176,
      "learning_rate": 9.987820251299122e-05,
      "loss": 2.6905,
      "step": 12
    },
    {
      "epoch": 0.008934707903780068,
      "grad_norm": 0.7743918299674988,
      "learning_rate": 9.972609476841367e-05,
      "loss": 2.5702,
      "step": 13
    },
    {
      "epoch": 0.009621993127147767,
      "grad_norm": 0.7361239194869995,
      "learning_rate": 9.951340343707852e-05,
      "loss": 2.3314,
      "step": 14
    },
    {
      "epoch": 0.010309278350515464,
      "grad_norm": 1.0487244129180908,
      "learning_rate": 9.924038765061042e-05,
      "loss": 2.308,
      "step": 15
    },
    {
      "epoch": 0.010996563573883162,
      "grad_norm": 0.8277301788330078,
      "learning_rate": 9.890738003669029e-05,
      "loss": 2.3839,
      "step": 16
    },
    {
      "epoch": 0.01168384879725086,
      "grad_norm": 0.8363406658172607,
      "learning_rate": 9.851478631379982e-05,
      "loss": 2.1383,
      "step": 17
    },
    {
      "epoch": 0.012371134020618556,
      "grad_norm": 0.7794749736785889,
      "learning_rate": 9.806308479691595e-05,
      "loss": 2.0868,
      "step": 18
    },
    {
      "epoch": 0.012371134020618556,
      "eval_loss": 0.9877583980560303,
      "eval_runtime": 254.3292,
      "eval_samples_per_second": 4.821,
      "eval_steps_per_second": 0.606,
      "step": 18
    },
    {
      "epoch": 0.013058419243986255,
      "grad_norm": 0.8351041078567505,
      "learning_rate": 9.755282581475769e-05,
      "loss": 2.1081,
      "step": 19
    },
    {
      "epoch": 0.013745704467353952,
      "grad_norm": 0.8560498356819153,
      "learning_rate": 9.698463103929542e-05,
      "loss": 1.8485,
      "step": 20
    },
    {
      "epoch": 0.01443298969072165,
      "grad_norm": 0.7435014843940735,
      "learning_rate": 9.635919272833938e-05,
      "loss": 1.7601,
      "step": 21
    },
    {
      "epoch": 0.015120274914089347,
      "grad_norm": 0.7566239833831787,
      "learning_rate": 9.567727288213005e-05,
      "loss": 1.6923,
      "step": 22
    },
    {
      "epoch": 0.015807560137457044,
      "grad_norm": 0.8233651518821716,
      "learning_rate": 9.493970231495835e-05,
      "loss": 1.724,
      "step": 23
    },
    {
      "epoch": 0.016494845360824743,
      "grad_norm": 0.8340699672698975,
      "learning_rate": 9.414737964294636e-05,
      "loss": 1.6131,
      "step": 24
    },
    {
      "epoch": 0.01718213058419244,
      "grad_norm": 0.7858492136001587,
      "learning_rate": 9.330127018922194e-05,
      "loss": 1.528,
      "step": 25
    },
    {
      "epoch": 0.017869415807560136,
      "grad_norm": 0.8513308763504028,
      "learning_rate": 9.24024048078213e-05,
      "loss": 1.5835,
      "step": 26
    },
    {
      "epoch": 0.018556701030927835,
      "grad_norm": 0.6795158982276917,
      "learning_rate": 9.145187862775209e-05,
      "loss": 1.5386,
      "step": 27
    },
    {
      "epoch": 0.018556701030927835,
      "eval_loss": 0.748094916343689,
      "eval_runtime": 253.4746,
      "eval_samples_per_second": 4.837,
      "eval_steps_per_second": 0.608,
      "step": 27
    },
    {
      "epoch": 0.019243986254295534,
      "grad_norm": 0.7328270077705383,
      "learning_rate": 9.045084971874738e-05,
      "loss": 1.551,
      "step": 28
    },
    {
      "epoch": 0.01993127147766323,
      "grad_norm": 0.7274560928344727,
      "learning_rate": 8.940053768033609e-05,
      "loss": 1.4977,
      "step": 29
    },
    {
      "epoch": 0.020618556701030927,
      "grad_norm": 0.7910637259483337,
      "learning_rate": 8.83022221559489e-05,
      "loss": 1.4872,
      "step": 30
    },
    {
      "epoch": 0.021305841924398626,
      "grad_norm": 0.8576248288154602,
      "learning_rate": 8.715724127386972e-05,
      "loss": 1.4049,
      "step": 31
    },
    {
      "epoch": 0.021993127147766325,
      "grad_norm": 0.9982084631919861,
      "learning_rate": 8.596699001693255e-05,
      "loss": 1.3457,
      "step": 32
    },
    {
      "epoch": 0.02268041237113402,
      "grad_norm": 0.7275662422180176,
      "learning_rate": 8.473291852294987e-05,
      "loss": 1.345,
      "step": 33
    },
    {
      "epoch": 0.02336769759450172,
      "grad_norm": 0.7337255477905273,
      "learning_rate": 8.345653031794292e-05,
      "loss": 1.3482,
      "step": 34
    },
    {
      "epoch": 0.024054982817869417,
      "grad_norm": 0.8252787590026855,
      "learning_rate": 8.213938048432697e-05,
      "loss": 1.3413,
      "step": 35
    },
    {
      "epoch": 0.024742268041237112,
      "grad_norm": 0.9308682084083557,
      "learning_rate": 8.07830737662829e-05,
      "loss": 1.3861,
      "step": 36
    },
    {
      "epoch": 0.024742268041237112,
      "eval_loss": 0.6290791034698486,
      "eval_runtime": 252.791,
      "eval_samples_per_second": 4.85,
      "eval_steps_per_second": 0.609,
      "step": 36
    },
    {
      "epoch": 0.02542955326460481,
      "grad_norm": 0.8834632635116577,
      "learning_rate": 7.938926261462366e-05,
      "loss": 1.2901,
      "step": 37
    },
    {
      "epoch": 0.02611683848797251,
      "grad_norm": 0.7757920026779175,
      "learning_rate": 7.795964517353735e-05,
      "loss": 1.238,
      "step": 38
    },
    {
      "epoch": 0.026804123711340205,
      "grad_norm": 0.9437708854675293,
      "learning_rate": 7.649596321166024e-05,
      "loss": 1.1715,
      "step": 39
    },
    {
      "epoch": 0.027491408934707903,
      "grad_norm": 0.9284512400627136,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.091,
      "step": 40
    },
    {
      "epoch": 0.028178694158075602,
      "grad_norm": 0.8415759205818176,
      "learning_rate": 7.347357813929454e-05,
      "loss": 1.1651,
      "step": 41
    },
    {
      "epoch": 0.0288659793814433,
      "grad_norm": 0.8733612895011902,
      "learning_rate": 7.191855733945387e-05,
      "loss": 1.1425,
      "step": 42
    },
    {
      "epoch": 0.029553264604810996,
      "grad_norm": 0.9334592819213867,
      "learning_rate": 7.033683215379002e-05,
      "loss": 1.1424,
      "step": 43
    },
    {
      "epoch": 0.030240549828178694,
      "grad_norm": 0.8202064037322998,
      "learning_rate": 6.873032967079561e-05,
      "loss": 1.1921,
      "step": 44
    },
    {
      "epoch": 0.030927835051546393,
      "grad_norm": 0.7855863571166992,
      "learning_rate": 6.710100716628344e-05,
      "loss": 1.1513,
      "step": 45
    },
    {
      "epoch": 0.030927835051546393,
      "eval_loss": 0.5508877038955688,
      "eval_runtime": 252.7564,
      "eval_samples_per_second": 4.851,
      "eval_steps_per_second": 0.609,
      "step": 45
    },
    {
      "epoch": 0.03161512027491409,
      "grad_norm": 0.8433973789215088,
      "learning_rate": 6.545084971874738e-05,
      "loss": 1.0264,
      "step": 46
    },
    {
      "epoch": 0.03230240549828179,
      "grad_norm": 0.8501434922218323,
      "learning_rate": 6.378186779084995e-05,
      "loss": 1.0172,
      "step": 47
    },
    {
      "epoch": 0.032989690721649485,
      "grad_norm": 0.8408077359199524,
      "learning_rate": 6.209609477998338e-05,
      "loss": 1.1105,
      "step": 48
    },
    {
      "epoch": 0.03367697594501718,
      "grad_norm": 0.8546326160430908,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 1.1052,
      "step": 49
    },
    {
      "epoch": 0.03436426116838488,
      "grad_norm": 0.8061237931251526,
      "learning_rate": 5.868240888334653e-05,
      "loss": 1.0345,
      "step": 50
    },
    {
      "epoch": 0.03505154639175258,
      "grad_norm": 0.8400550484657288,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.9842,
      "step": 51
    },
    {
      "epoch": 0.03573883161512027,
      "grad_norm": 0.8896331191062927,
      "learning_rate": 5.522642316338268e-05,
      "loss": 1.0689,
      "step": 52
    },
    {
      "epoch": 0.036426116838487975,
      "grad_norm": 0.984061598777771,
      "learning_rate": 5.348782368720626e-05,
      "loss": 1.1267,
      "step": 53
    },
    {
      "epoch": 0.03711340206185567,
      "grad_norm": 0.98240065574646,
      "learning_rate": 5.174497483512506e-05,
      "loss": 1.0024,
      "step": 54
    },
    {
      "epoch": 0.03711340206185567,
      "eval_loss": 0.496385782957077,
      "eval_runtime": 252.835,
      "eval_samples_per_second": 4.849,
      "eval_steps_per_second": 0.609,
      "step": 54
    },
    {
      "epoch": 0.037800687285223365,
      "grad_norm": 0.9597200155258179,
      "learning_rate": 5e-05,
      "loss": 1.0872,
      "step": 55
    },
    {
      "epoch": 0.03848797250859107,
      "grad_norm": 0.9907920360565186,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.9184,
      "step": 56
    },
    {
      "epoch": 0.03917525773195876,
      "grad_norm": 1.0494588613510132,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 1.0417,
      "step": 57
    },
    {
      "epoch": 0.03986254295532646,
      "grad_norm": 0.9095539450645447,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.8995,
      "step": 58
    },
    {
      "epoch": 0.04054982817869416,
      "grad_norm": 1.0099577903747559,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 0.9593,
      "step": 59
    },
    {
      "epoch": 0.041237113402061855,
      "grad_norm": 1.0103282928466797,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.9622,
      "step": 60
    },
    {
      "epoch": 0.04192439862542955,
      "grad_norm": 0.9153605103492737,
      "learning_rate": 3.960441545911204e-05,
      "loss": 0.8628,
      "step": 61
    },
    {
      "epoch": 0.04261168384879725,
      "grad_norm": 0.8908958435058594,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.922,
      "step": 62
    },
    {
      "epoch": 0.04329896907216495,
      "grad_norm": 0.9071544408798218,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 0.9928,
      "step": 63
    },
    {
      "epoch": 0.04329896907216495,
      "eval_loss": 0.464398592710495,
      "eval_runtime": 252.7256,
      "eval_samples_per_second": 4.851,
      "eval_steps_per_second": 0.609,
      "step": 63
    },
    {
      "epoch": 0.04398625429553265,
      "grad_norm": 0.9316085577011108,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.9378,
      "step": 64
    },
    {
      "epoch": 0.044673539518900345,
      "grad_norm": 0.8639270663261414,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.9379,
      "step": 65
    },
    {
      "epoch": 0.04536082474226804,
      "grad_norm": 0.9839565753936768,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.8882,
      "step": 66
    },
    {
      "epoch": 0.04604810996563574,
      "grad_norm": 0.9108591079711914,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 0.9424,
      "step": 67
    },
    {
      "epoch": 0.04673539518900344,
      "grad_norm": 0.9234270453453064,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.9626,
      "step": 68
    },
    {
      "epoch": 0.04742268041237113,
      "grad_norm": 0.9421894550323486,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 0.8883,
      "step": 69
    },
    {
      "epoch": 0.048109965635738834,
      "grad_norm": 0.9845061302185059,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.8485,
      "step": 70
    },
    {
      "epoch": 0.04879725085910653,
      "grad_norm": 0.9633055329322815,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.9693,
      "step": 71
    },
    {
      "epoch": 0.049484536082474224,
      "grad_norm": 0.9256303310394287,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.9221,
      "step": 72
    },
    {
      "epoch": 0.049484536082474224,
      "eval_loss": 0.44281116127967834,
      "eval_runtime": 252.7515,
      "eval_samples_per_second": 4.851,
      "eval_steps_per_second": 0.609,
      "step": 72
    },
    {
      "epoch": 0.05017182130584193,
      "grad_norm": 0.9681574106216431,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.9509,
      "step": 73
    },
    {
      "epoch": 0.05085910652920962,
      "grad_norm": 0.922154426574707,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.8309,
      "step": 74
    },
    {
      "epoch": 0.05154639175257732,
      "grad_norm": 0.9173290729522705,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.9151,
      "step": 75
    },
    {
      "epoch": 0.05223367697594502,
      "grad_norm": 0.9165816903114319,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.8993,
      "step": 76
    },
    {
      "epoch": 0.052920962199312714,
      "grad_norm": 0.7988548874855042,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.7629,
      "step": 77
    },
    {
      "epoch": 0.05360824742268041,
      "grad_norm": 0.8661845922470093,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.8056,
      "step": 78
    },
    {
      "epoch": 0.05429553264604811,
      "grad_norm": 0.8931153416633606,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.8286,
      "step": 79
    },
    {
      "epoch": 0.054982817869415807,
      "grad_norm": 0.9941338300704956,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.969,
      "step": 80
    },
    {
      "epoch": 0.05567010309278351,
      "grad_norm": 0.9468693137168884,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.9096,
      "step": 81
    },
    {
      "epoch": 0.05567010309278351,
      "eval_loss": 0.432786762714386,
      "eval_runtime": 252.8196,
      "eval_samples_per_second": 4.849,
      "eval_steps_per_second": 0.609,
      "step": 81
    },
    {
      "epoch": 0.056357388316151204,
      "grad_norm": 0.8518151640892029,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.738,
      "step": 82
    },
    {
      "epoch": 0.0570446735395189,
      "grad_norm": 0.8456606268882751,
      "learning_rate": 8.548121372247918e-06,
      "loss": 0.8421,
      "step": 83
    },
    {
      "epoch": 0.0577319587628866,
      "grad_norm": 0.916377067565918,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.7926,
      "step": 84
    },
    {
      "epoch": 0.058419243986254296,
      "grad_norm": 0.8955652713775635,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.8337,
      "step": 85
    },
    {
      "epoch": 0.05910652920962199,
      "grad_norm": 0.9039252996444702,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.8484,
      "step": 86
    },
    {
      "epoch": 0.05979381443298969,
      "grad_norm": 0.9059133529663086,
      "learning_rate": 5.060297685041659e-06,
      "loss": 0.8716,
      "step": 87
    },
    {
      "epoch": 0.06048109965635739,
      "grad_norm": 0.9183720350265503,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.9808,
      "step": 88
    },
    {
      "epoch": 0.061168384879725084,
      "grad_norm": 0.9803336262702942,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.8317,
      "step": 89
    },
    {
      "epoch": 0.061855670103092786,
      "grad_norm": 0.911630392074585,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.8261,
      "step": 90
    },
    {
      "epoch": 0.061855670103092786,
      "eval_loss": 0.4276270270347595,
      "eval_runtime": 252.8496,
      "eval_samples_per_second": 4.849,
      "eval_steps_per_second": 0.609,
      "step": 90
    },
    {
      "epoch": 0.06254295532646048,
      "grad_norm": 1.0115582942962646,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.8234,
      "step": 91
    },
    {
      "epoch": 0.06323024054982818,
      "grad_norm": 0.8366324305534363,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.8316,
      "step": 92
    },
    {
      "epoch": 0.06391752577319587,
      "grad_norm": 0.9966782331466675,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 0.7705,
      "step": 93
    },
    {
      "epoch": 0.06460481099656358,
      "grad_norm": 0.9167208075523376,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.8403,
      "step": 94
    },
    {
      "epoch": 0.06529209621993128,
      "grad_norm": 0.9588520526885986,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.8928,
      "step": 95
    },
    {
      "epoch": 0.06597938144329897,
      "grad_norm": 0.9358211159706116,
      "learning_rate": 4.865965629214819e-07,
      "loss": 0.8543,
      "step": 96
    },
    {
      "epoch": 0.06666666666666667,
      "grad_norm": 0.8442792892456055,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.8146,
      "step": 97
    },
    {
      "epoch": 0.06735395189003436,
      "grad_norm": 0.8882707357406616,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.8587,
      "step": 98
    },
    {
      "epoch": 0.06804123711340206,
      "grad_norm": 1.0984888076782227,
      "learning_rate": 3.04586490452119e-08,
      "loss": 0.9505,
      "step": 99
    },
    {
      "epoch": 0.06804123711340206,
      "eval_loss": 0.42630666494369507,
      "eval_runtime": 253.3629,
      "eval_samples_per_second": 4.839,
      "eval_steps_per_second": 0.608,
      "step": 99
    },
    {
      "epoch": 0.06872852233676977,
      "grad_norm": 0.9829119443893433,
      "learning_rate": 0.0,
      "loss": 0.9251,
      "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": 1.26956599246848e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}