{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.028161081385525203,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000281610813855252,
      "grad_norm": 2.2820143699645996,
      "learning_rate": 1e-05,
      "loss": 7.6012,
      "step": 1
    },
    {
      "epoch": 0.000281610813855252,
      "eval_loss": 3.763329029083252,
      "eval_runtime": 518.8953,
      "eval_samples_per_second": 5.762,
      "eval_steps_per_second": 0.721,
      "step": 1
    },
    {
      "epoch": 0.000563221627710504,
      "grad_norm": 2.415229082107544,
      "learning_rate": 2e-05,
      "loss": 7.2807,
      "step": 2
    },
    {
      "epoch": 0.0008448324415657561,
      "grad_norm": 2.2119874954223633,
      "learning_rate": 3e-05,
      "loss": 7.6457,
      "step": 3
    },
    {
      "epoch": 0.001126443255421008,
      "grad_norm": 2.285097599029541,
      "learning_rate": 4e-05,
      "loss": 7.5695,
      "step": 4
    },
    {
      "epoch": 0.0014080540692762602,
      "grad_norm": 2.235811948776245,
      "learning_rate": 5e-05,
      "loss": 7.744,
      "step": 5
    },
    {
      "epoch": 0.0016896648831315122,
      "grad_norm": 2.445420503616333,
      "learning_rate": 6e-05,
      "loss": 7.4837,
      "step": 6
    },
    {
      "epoch": 0.0019712756969867645,
      "grad_norm": 2.834716320037842,
      "learning_rate": 7e-05,
      "loss": 7.4933,
      "step": 7
    },
    {
      "epoch": 0.002252886510842016,
      "grad_norm": 3.3109493255615234,
      "learning_rate": 8e-05,
      "loss": 7.3508,
      "step": 8
    },
    {
      "epoch": 0.0025344973246972683,
      "grad_norm": 3.259788990020752,
      "learning_rate": 9e-05,
      "loss": 6.869,
      "step": 9
    },
    {
      "epoch": 0.0025344973246972683,
      "eval_loss": 3.2674660682678223,
      "eval_runtime": 519.0713,
      "eval_samples_per_second": 5.76,
      "eval_steps_per_second": 0.721,
      "step": 9
    },
    {
      "epoch": 0.0028161081385525205,
      "grad_norm": 3.805819511413574,
      "learning_rate": 0.0001,
      "loss": 6.5288,
      "step": 10
    },
    {
      "epoch": 0.0030977189524077726,
      "grad_norm": 3.735010862350464,
      "learning_rate": 9.99695413509548e-05,
      "loss": 5.7594,
      "step": 11
    },
    {
      "epoch": 0.0033793297662630243,
      "grad_norm": 3.8842203617095947,
      "learning_rate": 9.987820251299122e-05,
      "loss": 5.8022,
      "step": 12
    },
    {
      "epoch": 0.0036609405801182764,
      "grad_norm": 4.158329963684082,
      "learning_rate": 9.972609476841367e-05,
      "loss": 5.0675,
      "step": 13
    },
    {
      "epoch": 0.003942551393973529,
      "grad_norm": 3.7810776233673096,
      "learning_rate": 9.951340343707852e-05,
      "loss": 4.3389,
      "step": 14
    },
    {
      "epoch": 0.00422416220782878,
      "grad_norm": 3.8304190635681152,
      "learning_rate": 9.924038765061042e-05,
      "loss": 4.1328,
      "step": 15
    },
    {
      "epoch": 0.004505773021684032,
      "grad_norm": 3.2623438835144043,
      "learning_rate": 9.890738003669029e-05,
      "loss": 4.335,
      "step": 16
    },
    {
      "epoch": 0.0047873838355392846,
      "grad_norm": 2.291598081588745,
      "learning_rate": 9.851478631379982e-05,
      "loss": 3.5827,
      "step": 17
    },
    {
      "epoch": 0.005068994649394537,
      "grad_norm": 2.769401788711548,
      "learning_rate": 9.806308479691595e-05,
      "loss": 3.4517,
      "step": 18
    },
    {
      "epoch": 0.005068994649394537,
      "eval_loss": 1.5952740907669067,
      "eval_runtime": 519.0308,
      "eval_samples_per_second": 5.761,
      "eval_steps_per_second": 0.721,
      "step": 18
    },
    {
      "epoch": 0.005350605463249789,
      "grad_norm": 3.678323268890381,
      "learning_rate": 9.755282581475769e-05,
      "loss": 3.5372,
      "step": 19
    },
    {
      "epoch": 0.005632216277105041,
      "grad_norm": 5.350372791290283,
      "learning_rate": 9.698463103929542e-05,
      "loss": 2.6173,
      "step": 20
    },
    {
      "epoch": 0.005913827090960293,
      "grad_norm": 5.400213241577148,
      "learning_rate": 9.635919272833938e-05,
      "loss": 3.0168,
      "step": 21
    },
    {
      "epoch": 0.006195437904815545,
      "grad_norm": 5.264115810394287,
      "learning_rate": 9.567727288213005e-05,
      "loss": 2.5549,
      "step": 22
    },
    {
      "epoch": 0.006477048718670797,
      "grad_norm": 3.276190757751465,
      "learning_rate": 9.493970231495835e-05,
      "loss": 2.5311,
      "step": 23
    },
    {
      "epoch": 0.006758659532526049,
      "grad_norm": 2.4248852729797363,
      "learning_rate": 9.414737964294636e-05,
      "loss": 2.3755,
      "step": 24
    },
    {
      "epoch": 0.007040270346381301,
      "grad_norm": 3.3110451698303223,
      "learning_rate": 9.330127018922194e-05,
      "loss": 2.5545,
      "step": 25
    },
    {
      "epoch": 0.007321881160236553,
      "grad_norm": 3.004551887512207,
      "learning_rate": 9.24024048078213e-05,
      "loss": 2.1855,
      "step": 26
    },
    {
      "epoch": 0.007603491974091805,
      "grad_norm": 2.196512460708618,
      "learning_rate": 9.145187862775209e-05,
      "loss": 1.9427,
      "step": 27
    },
    {
      "epoch": 0.007603491974091805,
      "eval_loss": 0.9296455383300781,
      "eval_runtime": 518.8812,
      "eval_samples_per_second": 5.762,
      "eval_steps_per_second": 0.721,
      "step": 27
    },
    {
      "epoch": 0.007885102787947058,
      "grad_norm": 1.2902026176452637,
      "learning_rate": 9.045084971874738e-05,
      "loss": 1.6342,
      "step": 28
    },
    {
      "epoch": 0.008166713601802308,
      "grad_norm": 2.955904483795166,
      "learning_rate": 8.940053768033609e-05,
      "loss": 1.8788,
      "step": 29
    },
    {
      "epoch": 0.00844832441565756,
      "grad_norm": 3.204195976257324,
      "learning_rate": 8.83022221559489e-05,
      "loss": 1.7209,
      "step": 30
    },
    {
      "epoch": 0.008729935229512813,
      "grad_norm": 1.4495835304260254,
      "learning_rate": 8.715724127386972e-05,
      "loss": 1.5066,
      "step": 31
    },
    {
      "epoch": 0.009011546043368065,
      "grad_norm": 1.1010254621505737,
      "learning_rate": 8.596699001693255e-05,
      "loss": 1.4999,
      "step": 32
    },
    {
      "epoch": 0.009293156857223317,
      "grad_norm": 0.9905458688735962,
      "learning_rate": 8.473291852294987e-05,
      "loss": 1.4371,
      "step": 33
    },
    {
      "epoch": 0.009574767671078569,
      "grad_norm": 0.7902398705482483,
      "learning_rate": 8.345653031794292e-05,
      "loss": 1.433,
      "step": 34
    },
    {
      "epoch": 0.009856378484933821,
      "grad_norm": 0.7530960440635681,
      "learning_rate": 8.213938048432697e-05,
      "loss": 1.4247,
      "step": 35
    },
    {
      "epoch": 0.010137989298789073,
      "grad_norm": 0.981115996837616,
      "learning_rate": 8.07830737662829e-05,
      "loss": 1.4491,
      "step": 36
    },
    {
      "epoch": 0.010137989298789073,
      "eval_loss": 0.7092130184173584,
      "eval_runtime": 518.7764,
      "eval_samples_per_second": 5.764,
      "eval_steps_per_second": 0.721,
      "step": 36
    },
    {
      "epoch": 0.010419600112644326,
      "grad_norm": 0.5802685618400574,
      "learning_rate": 7.938926261462366e-05,
      "loss": 1.4158,
      "step": 37
    },
    {
      "epoch": 0.010701210926499578,
      "grad_norm": 0.7508324384689331,
      "learning_rate": 7.795964517353735e-05,
      "loss": 1.4274,
      "step": 38
    },
    {
      "epoch": 0.01098282174035483,
      "grad_norm": 0.8648625612258911,
      "learning_rate": 7.649596321166024e-05,
      "loss": 1.4218,
      "step": 39
    },
    {
      "epoch": 0.011264432554210082,
      "grad_norm": 0.8676905632019043,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.4303,
      "step": 40
    },
    {
      "epoch": 0.011546043368065334,
      "grad_norm": 0.3803287148475647,
      "learning_rate": 7.347357813929454e-05,
      "loss": 1.4144,
      "step": 41
    },
    {
      "epoch": 0.011827654181920586,
      "grad_norm": 0.3803757429122925,
      "learning_rate": 7.191855733945387e-05,
      "loss": 1.4255,
      "step": 42
    },
    {
      "epoch": 0.012109264995775838,
      "grad_norm": 0.34761127829551697,
      "learning_rate": 7.033683215379002e-05,
      "loss": 1.4156,
      "step": 43
    },
    {
      "epoch": 0.01239087580963109,
      "grad_norm": 0.3840619921684265,
      "learning_rate": 6.873032967079561e-05,
      "loss": 1.4275,
      "step": 44
    },
    {
      "epoch": 0.012672486623486343,
      "grad_norm": 0.32926231622695923,
      "learning_rate": 6.710100716628344e-05,
      "loss": 1.4243,
      "step": 45
    },
    {
      "epoch": 0.012672486623486343,
      "eval_loss": 0.7055554389953613,
      "eval_runtime": 518.8734,
      "eval_samples_per_second": 5.762,
      "eval_steps_per_second": 0.721,
      "step": 45
    },
    {
      "epoch": 0.012954097437341595,
      "grad_norm": 0.2857879400253296,
      "learning_rate": 6.545084971874738e-05,
      "loss": 1.41,
      "step": 46
    },
    {
      "epoch": 0.013235708251196845,
      "grad_norm": 0.2769199013710022,
      "learning_rate": 6.378186779084995e-05,
      "loss": 1.409,
      "step": 47
    },
    {
      "epoch": 0.013517319065052097,
      "grad_norm": 0.2755632996559143,
      "learning_rate": 6.209609477998338e-05,
      "loss": 1.4072,
      "step": 48
    },
    {
      "epoch": 0.01379892987890735,
      "grad_norm": 0.3106071650981903,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 1.4099,
      "step": 49
    },
    {
      "epoch": 0.014080540692762602,
      "grad_norm": 0.32463961839675903,
      "learning_rate": 5.868240888334653e-05,
      "loss": 1.3935,
      "step": 50
    },
    {
      "epoch": 0.014362151506617854,
      "grad_norm": 0.2624410092830658,
      "learning_rate": 5.695865504800327e-05,
      "loss": 1.4003,
      "step": 51
    },
    {
      "epoch": 0.014643762320473106,
      "grad_norm": 0.4355788826942444,
      "learning_rate": 5.522642316338268e-05,
      "loss": 1.4076,
      "step": 52
    },
    {
      "epoch": 0.014925373134328358,
      "grad_norm": 0.3189755380153656,
      "learning_rate": 5.348782368720626e-05,
      "loss": 1.3951,
      "step": 53
    },
    {
      "epoch": 0.01520698394818361,
      "grad_norm": 0.2545646131038666,
      "learning_rate": 5.174497483512506e-05,
      "loss": 1.4063,
      "step": 54
    },
    {
      "epoch": 0.01520698394818361,
      "eval_loss": 0.705012321472168,
      "eval_runtime": 518.9348,
      "eval_samples_per_second": 5.762,
      "eval_steps_per_second": 0.721,
      "step": 54
    },
    {
      "epoch": 0.015488594762038862,
      "grad_norm": 0.4052259922027588,
      "learning_rate": 5e-05,
      "loss": 1.4113,
      "step": 55
    },
    {
      "epoch": 0.015770205575894116,
      "grad_norm": 0.2603490948677063,
      "learning_rate": 4.825502516487497e-05,
      "loss": 1.4027,
      "step": 56
    },
    {
      "epoch": 0.016051816389749365,
      "grad_norm": 0.3026479482650757,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 1.4092,
      "step": 57
    },
    {
      "epoch": 0.016333427203604617,
      "grad_norm": 0.25429239869117737,
      "learning_rate": 4.477357683661734e-05,
      "loss": 1.4131,
      "step": 58
    },
    {
      "epoch": 0.01661503801745987,
      "grad_norm": 0.2743135094642639,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 1.408,
      "step": 59
    },
    {
      "epoch": 0.01689664883131512,
      "grad_norm": 0.8984668254852295,
      "learning_rate": 4.131759111665349e-05,
      "loss": 1.3968,
      "step": 60
    },
    {
      "epoch": 0.017178259645170373,
      "grad_norm": 0.2367839813232422,
      "learning_rate": 3.960441545911204e-05,
      "loss": 1.4044,
      "step": 61
    },
    {
      "epoch": 0.017459870459025625,
      "grad_norm": 0.25028061866760254,
      "learning_rate": 3.790390522001662e-05,
      "loss": 1.3982,
      "step": 62
    },
    {
      "epoch": 0.017741481272880878,
      "grad_norm": 0.2683260142803192,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 1.408,
      "step": 63
    },
    {
      "epoch": 0.017741481272880878,
      "eval_loss": 0.7046836018562317,
      "eval_runtime": 518.929,
      "eval_samples_per_second": 5.762,
      "eval_steps_per_second": 0.721,
      "step": 63
    },
    {
      "epoch": 0.01802309208673613,
      "grad_norm": 0.3018679916858673,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 1.3995,
      "step": 64
    },
    {
      "epoch": 0.018304702900591382,
      "grad_norm": 0.29377129673957825,
      "learning_rate": 3.289899283371657e-05,
      "loss": 1.3886,
      "step": 65
    },
    {
      "epoch": 0.018586313714446634,
      "grad_norm": 0.19938716292381287,
      "learning_rate": 3.12696703292044e-05,
      "loss": 1.4021,
      "step": 66
    },
    {
      "epoch": 0.018867924528301886,
      "grad_norm": 0.2533423602581024,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 1.4022,
      "step": 67
    },
    {
      "epoch": 0.019149535342157138,
      "grad_norm": 0.20846250653266907,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 1.4101,
      "step": 68
    },
    {
      "epoch": 0.01943114615601239,
      "grad_norm": 0.20643027126789093,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 1.4013,
      "step": 69
    },
    {
      "epoch": 0.019712756969867642,
      "grad_norm": 0.2405877709388733,
      "learning_rate": 2.500000000000001e-05,
      "loss": 1.3946,
      "step": 70
    },
    {
      "epoch": 0.019994367783722895,
      "grad_norm": 0.2167934626340866,
      "learning_rate": 2.350403678833976e-05,
      "loss": 1.4087,
      "step": 71
    },
    {
      "epoch": 0.020275978597578147,
      "grad_norm": 0.17771920561790466,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 1.3985,
      "step": 72
    },
    {
      "epoch": 0.020275978597578147,
      "eval_loss": 0.7046549320220947,
      "eval_runtime": 518.9209,
      "eval_samples_per_second": 5.762,
      "eval_steps_per_second": 0.721,
      "step": 72
    },
    {
      "epoch": 0.0205575894114334,
      "grad_norm": 0.1883406639099121,
      "learning_rate": 2.061073738537635e-05,
      "loss": 1.4099,
      "step": 73
    },
    {
      "epoch": 0.02083920022528865,
      "grad_norm": 0.2247111201286316,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 1.4117,
      "step": 74
    },
    {
      "epoch": 0.021120811039143903,
      "grad_norm": 0.19664792716503143,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 1.4106,
      "step": 75
    },
    {
      "epoch": 0.021402421852999155,
      "grad_norm": 0.20162613689899445,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 1.3914,
      "step": 76
    },
    {
      "epoch": 0.021684032666854407,
      "grad_norm": 0.17795982956886292,
      "learning_rate": 1.526708147705013e-05,
      "loss": 1.4005,
      "step": 77
    },
    {
      "epoch": 0.02196564348070966,
      "grad_norm": 0.21461506187915802,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 1.4016,
      "step": 78
    },
    {
      "epoch": 0.02224725429456491,
      "grad_norm": 0.15580390393733978,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 1.4033,
      "step": 79
    },
    {
      "epoch": 0.022528865108420164,
      "grad_norm": 0.26418066024780273,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 1.4117,
      "step": 80
    },
    {
      "epoch": 0.022810475922275416,
      "grad_norm": 0.22687722742557526,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 1.4085,
      "step": 81
    },
    {
      "epoch": 0.022810475922275416,
      "eval_loss": 0.704688310623169,
      "eval_runtime": 518.8374,
      "eval_samples_per_second": 5.763,
      "eval_steps_per_second": 0.721,
      "step": 81
    },
    {
      "epoch": 0.023092086736130668,
      "grad_norm": 0.2125590443611145,
      "learning_rate": 9.549150281252633e-06,
      "loss": 1.406,
      "step": 82
    },
    {
      "epoch": 0.02337369754998592,
      "grad_norm": 0.2060941457748413,
      "learning_rate": 8.548121372247918e-06,
      "loss": 1.4117,
      "step": 83
    },
    {
      "epoch": 0.023655308363841172,
      "grad_norm": 0.22980840504169464,
      "learning_rate": 7.597595192178702e-06,
      "loss": 1.398,
      "step": 84
    },
    {
      "epoch": 0.023936919177696424,
      "grad_norm": 0.1792149692773819,
      "learning_rate": 6.698729810778065e-06,
      "loss": 1.4137,
      "step": 85
    },
    {
      "epoch": 0.024218529991551677,
      "grad_norm": 1.1202366352081299,
      "learning_rate": 5.852620357053651e-06,
      "loss": 1.4726,
      "step": 86
    },
    {
      "epoch": 0.02450014080540693,
      "grad_norm": 0.18409371376037598,
      "learning_rate": 5.060297685041659e-06,
      "loss": 1.3985,
      "step": 87
    },
    {
      "epoch": 0.02478175161926218,
      "grad_norm": 0.1613732874393463,
      "learning_rate": 4.322727117869951e-06,
      "loss": 1.4006,
      "step": 88
    },
    {
      "epoch": 0.025063362433117433,
      "grad_norm": 0.2666328549385071,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 1.4024,
      "step": 89
    },
    {
      "epoch": 0.025344973246972685,
      "grad_norm": 0.19227731227874756,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 1.4133,
      "step": 90
    },
    {
      "epoch": 0.025344973246972685,
      "eval_loss": 0.7046293616294861,
      "eval_runtime": 518.987,
      "eval_samples_per_second": 5.761,
      "eval_steps_per_second": 0.721,
      "step": 90
    },
    {
      "epoch": 0.025626584060827937,
      "grad_norm": 0.18015547096729279,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 1.3942,
      "step": 91
    },
    {
      "epoch": 0.02590819487468319,
      "grad_norm": 0.20062336325645447,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 1.4004,
      "step": 92
    },
    {
      "epoch": 0.02618980568853844,
      "grad_norm": 0.20917601883411407,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 1.4097,
      "step": 93
    },
    {
      "epoch": 0.02647141650239369,
      "grad_norm": 0.1842080056667328,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 1.4086,
      "step": 94
    },
    {
      "epoch": 0.026753027316248942,
      "grad_norm": 0.2337813377380371,
      "learning_rate": 7.596123493895991e-07,
      "loss": 1.414,
      "step": 95
    },
    {
      "epoch": 0.027034638130104195,
      "grad_norm": 0.1556297391653061,
      "learning_rate": 4.865965629214819e-07,
      "loss": 1.4119,
      "step": 96
    },
    {
      "epoch": 0.027316248943959447,
      "grad_norm": 0.18028058111667633,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 1.3931,
      "step": 97
    },
    {
      "epoch": 0.0275978597578147,
      "grad_norm": 0.24062392115592957,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 1.4069,
      "step": 98
    },
    {
      "epoch": 0.02787947057166995,
      "grad_norm": 0.16729937493801117,
      "learning_rate": 3.04586490452119e-08,
      "loss": 1.3994,
      "step": 99
    },
    {
      "epoch": 0.02787947057166995,
      "eval_loss": 0.7044768929481506,
      "eval_runtime": 518.85,
      "eval_samples_per_second": 5.763,
      "eval_steps_per_second": 0.721,
      "step": 99
    },
    {
      "epoch": 0.028161081385525203,
      "grad_norm": 0.17885710299015045,
      "learning_rate": 0.0,
      "loss": 1.3979,
      "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": 1.2677214100468531e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}