{
  "best_global_step": 70,
  "best_metric": 0.17395253479480743,
  "best_model_checkpoint": "./lora_adapter/checkpoint-70",
  "epoch": 4.140969162995595,
  "eval_steps": 10,
  "global_step": 120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0,
      "eval_loss": 1.7887901067733765,
      "eval_runtime": 5.8251,
      "eval_samples_per_second": 8.584,
      "eval_steps_per_second": 8.584,
      "step": 0
    },
    {
      "epoch": 0,
      "eval_perplexity": 5.982210250450309,
      "step": 0
    },
    {
      "epoch": 0.03524229074889868,
      "grad_norm": 0.4597412645816803,
      "learning_rate": 0.0,
      "loss": 1.5720514059066772,
      "step": 1
    },
    {
      "epoch": 0.07048458149779736,
      "grad_norm": 0.4013294577598572,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.5010937452316284,
      "step": 2
    },
    {
      "epoch": 0.10572687224669604,
      "grad_norm": 0.5167919397354126,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.782500147819519,
      "step": 3
    },
    {
      "epoch": 0.14096916299559473,
      "grad_norm": 0.42857807874679565,
      "learning_rate": 2.5e-05,
      "loss": 1.5116475820541382,
      "step": 4
    },
    {
      "epoch": 0.1762114537444934,
      "grad_norm": 0.457490473985672,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.6051511764526367,
      "step": 5
    },
    {
      "epoch": 0.21145374449339208,
      "grad_norm": 0.34170931577682495,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.330802321434021,
      "step": 6
    },
    {
      "epoch": 0.24669603524229075,
      "grad_norm": 0.2989931106567383,
      "learning_rate": 5e-05,
      "loss": 1.0146887302398682,
      "step": 7
    },
    {
      "epoch": 0.28193832599118945,
      "grad_norm": 0.2887096703052521,
      "learning_rate": 5.833333333333334e-05,
      "loss": 0.935509979724884,
      "step": 8
    },
    {
      "epoch": 0.31718061674008813,
      "grad_norm": 0.1863260716199875,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.5329318642616272,
      "step": 9
    },
    {
      "epoch": 0.3524229074889868,
      "grad_norm": 0.20654812455177307,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.502984344959259,
      "step": 10
    },
    {
      "epoch": 0.3524229074889868,
      "eval_loss": 0.37128761410713196,
      "eval_runtime": 5.5912,
      "eval_samples_per_second": 8.943,
      "eval_steps_per_second": 8.943,
      "step": 10
    },
    {
      "epoch": 0.3524229074889868,
      "eval_perplexity": 1.4495999388276846,
      "step": 10
    },
    {
      "epoch": 0.3876651982378855,
      "grad_norm": 0.15473979711532593,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.38561761379241943,
      "step": 11
    },
    {
      "epoch": 0.42290748898678415,
      "grad_norm": 0.10418996214866638,
      "learning_rate": 9.166666666666667e-05,
      "loss": 0.33394187688827515,
      "step": 12
    },
    {
      "epoch": 0.4581497797356828,
      "grad_norm": 0.1276063770055771,
      "learning_rate": 0.0001,
      "loss": 0.26392874121665955,
      "step": 13
    },
    {
      "epoch": 0.4933920704845815,
      "grad_norm": 0.10101140290498734,
      "learning_rate": 9.999490215047167e-05,
      "loss": 0.31664976477622986,
      "step": 14
    },
    {
      "epoch": 0.5286343612334802,
      "grad_norm": 0.09670906513929367,
      "learning_rate": 9.997960964140947e-05,
      "loss": 0.4327008128166199,
      "step": 15
    },
    {
      "epoch": 0.5638766519823789,
      "grad_norm": 0.10477341711521149,
      "learning_rate": 9.995412559116979e-05,
      "loss": 0.36697056889533997,
      "step": 16
    },
    {
      "epoch": 0.5991189427312775,
      "grad_norm": 0.07639091461896896,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.2757386267185211,
      "step": 17
    },
    {
      "epoch": 0.6343612334801763,
      "grad_norm": 0.05634988844394684,
      "learning_rate": 9.987260573051269e-05,
      "loss": 0.2590947151184082,
      "step": 18
    },
    {
      "epoch": 0.6696035242290749,
      "grad_norm": 0.07752630114555359,
      "learning_rate": 9.981658654313457e-05,
      "loss": 0.3277636170387268,
      "step": 19
    },
    {
      "epoch": 0.7048458149779736,
      "grad_norm": 0.07242827862501144,
      "learning_rate": 9.975040905726798e-05,
      "loss": 0.24663390219211578,
      "step": 20
    },
    {
      "epoch": 0.7048458149779736,
      "eval_loss": 0.20978213846683502,
      "eval_runtime": 5.5285,
      "eval_samples_per_second": 9.044,
      "eval_steps_per_second": 9.044,
      "step": 20
    },
    {
      "epoch": 0.7048458149779736,
      "eval_perplexity": 1.2334093182384733,
      "step": 20
    },
    {
      "epoch": 0.7400881057268722,
      "grad_norm": 0.056974418461322784,
      "learning_rate": 9.967408676742751e-05,
      "loss": 0.2576606571674347,
      "step": 21
    },
    {
      "epoch": 0.775330396475771,
      "grad_norm": 0.06447745859622955,
      "learning_rate": 9.958763523679514e-05,
      "loss": 0.27974751591682434,
      "step": 22
    },
    {
      "epoch": 0.8105726872246696,
      "grad_norm": 0.06282766163349152,
      "learning_rate": 9.949107209404665e-05,
      "loss": 0.2800998091697693,
      "step": 23
    },
    {
      "epoch": 0.8458149779735683,
      "grad_norm": 0.05366954952478409,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.24023258686065674,
      "step": 24
    },
    {
      "epoch": 0.8810572687224669,
      "grad_norm": 0.06681771576404572,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.21368750929832458,
      "step": 25
    },
    {
      "epoch": 0.9162995594713657,
      "grad_norm": 0.04931754246354103,
      "learning_rate": 9.914092018383778e-05,
      "loss": 0.14743126928806305,
      "step": 26
    },
    {
      "epoch": 0.9515418502202643,
      "grad_norm": 0.0491667315363884,
      "learning_rate": 9.900412805461967e-05,
      "loss": 0.18975743651390076,
      "step": 27
    },
    {
      "epoch": 0.986784140969163,
      "grad_norm": 0.07496297359466553,
      "learning_rate": 9.885734329855798e-05,
      "loss": 0.2644786536693573,
      "step": 28
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.0941501259803772,
      "learning_rate": 9.870059584711668e-05,
      "loss": 0.32210999727249146,
      "step": 29
    },
    {
      "epoch": 1.0352422907488987,
      "grad_norm": 0.056953802704811096,
      "learning_rate": 9.853391766329263e-05,
      "loss": 0.24877111613750458,
      "step": 30
    },
    {
      "epoch": 1.0352422907488987,
      "eval_loss": 0.19354841113090515,
      "eval_runtime": 5.5598,
      "eval_samples_per_second": 8.993,
      "eval_steps_per_second": 8.993,
      "step": 30
    },
    {
      "epoch": 1.0352422907488987,
      "eval_perplexity": 1.2135481343670569,
      "step": 30
    },
    {
      "epoch": 1.0704845814977975,
      "grad_norm": 0.04779611527919769,
      "learning_rate": 9.835734273509786e-05,
      "loss": 0.18907222151756287,
      "step": 31
    },
    {
      "epoch": 1.105726872246696,
      "grad_norm": 0.05204286426305771,
      "learning_rate": 9.817090706862895e-05,
      "loss": 0.2008168250322342,
      "step": 32
    },
    {
      "epoch": 1.1409691629955947,
      "grad_norm": 0.06194741651415825,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.3145204782485962,
      "step": 33
    },
    {
      "epoch": 1.1762114537444934,
      "grad_norm": 0.051133181899785995,
      "learning_rate": 9.776860759121484e-05,
      "loss": 0.17028148472309113,
      "step": 34
    },
    {
      "epoch": 1.2114537444933922,
      "grad_norm": 0.034042421728372574,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.11714069545269012,
      "step": 35
    },
    {
      "epoch": 1.2466960352422907,
      "grad_norm": 0.05539604276418686,
      "learning_rate": 9.73273473522745e-05,
      "loss": 0.2758980691432953,
      "step": 36
    },
    {
      "epoch": 1.2819383259911894,
      "grad_norm": 0.057937804609537125,
      "learning_rate": 9.709221818197624e-05,
      "loss": 0.20445898175239563,
      "step": 37
    },
    {
      "epoch": 1.3171806167400881,
      "grad_norm": 0.05714031681418419,
      "learning_rate": 9.68474862499881e-05,
      "loss": 0.23848702013492584,
      "step": 38
    },
    {
      "epoch": 1.3524229074889869,
      "grad_norm": 0.04392428323626518,
      "learning_rate": 9.659320146057262e-05,
      "loss": 0.19320310652256012,
      "step": 39
    },
    {
      "epoch": 1.3876651982378854,
      "grad_norm": 0.0482228621840477,
      "learning_rate": 9.632941566595357e-05,
      "loss": 0.22485677897930145,
      "step": 40
    },
    {
      "epoch": 1.3876651982378854,
      "eval_loss": 0.18485793471336365,
      "eval_runtime": 5.5445,
      "eval_samples_per_second": 9.018,
      "eval_steps_per_second": 9.018,
      "step": 40
    },
    {
      "epoch": 1.3876651982378854,
      "eval_perplexity": 1.203047516696543,
      "step": 40
    },
    {
      "epoch": 1.422907488986784,
      "grad_norm": 0.03985185548663139,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.13736587762832642,
      "step": 41
    },
    {
      "epoch": 1.4581497797356828,
      "grad_norm": 0.04527609050273895,
      "learning_rate": 9.577355814597031e-05,
      "loss": 0.23677022755146027,
      "step": 42
    },
    {
      "epoch": 1.4933920704845816,
      "grad_norm": 0.0444558784365654,
      "learning_rate": 9.548159976772592e-05,
      "loss": 0.15576449036598206,
      "step": 43
    },
    {
      "epoch": 1.5286343612334803,
      "grad_norm": 0.03990132361650467,
      "learning_rate": 9.518036705540458e-05,
      "loss": 0.16122624278068542,
      "step": 44
    },
    {
      "epoch": 1.5638766519823788,
      "grad_norm": 0.06517821550369263,
      "learning_rate": 9.486992143456792e-05,
      "loss": 0.20816665887832642,
      "step": 45
    },
    {
      "epoch": 1.5991189427312775,
      "grad_norm": 0.054203156381845474,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.22152908146381378,
      "step": 46
    },
    {
      "epoch": 1.6343612334801763,
      "grad_norm": 0.05166909471154213,
      "learning_rate": 9.422164654989072e-05,
      "loss": 0.23038849234580994,
      "step": 47
    },
    {
      "epoch": 1.6696035242290748,
      "grad_norm": 0.051149919629096985,
      "learning_rate": 9.388394947836279e-05,
      "loss": 0.20961874723434448,
      "step": 48
    },
    {
      "epoch": 1.7048458149779737,
      "grad_norm": 0.039978332817554474,
      "learning_rate": 9.353730385598887e-05,
      "loss": 0.14660704135894775,
      "step": 49
    },
    {
      "epoch": 1.7400881057268722,
      "grad_norm": 0.05832689255475998,
      "learning_rate": 9.318178036865785e-05,
      "loss": 0.1376156210899353,
      "step": 50
    },
    {
      "epoch": 1.7400881057268722,
      "eval_loss": 0.18398527801036835,
      "eval_runtime": 5.5458,
      "eval_samples_per_second": 9.016,
      "eval_steps_per_second": 9.016,
      "step": 50
    },
    {
      "epoch": 1.7400881057268722,
      "eval_perplexity": 1.2019981271620768,
      "step": 50
    },
    {
      "epoch": 1.775330396475771,
      "grad_norm": 0.06165598705410957,
      "learning_rate": 9.281745151257946e-05,
      "loss": 0.18745188415050507,
      "step": 51
    },
    {
      "epoch": 1.8105726872246697,
      "grad_norm": 0.04807993024587631,
      "learning_rate": 9.244439157950114e-05,
      "loss": 0.14113658666610718,
      "step": 52
    },
    {
      "epoch": 1.8458149779735682,
      "grad_norm": 0.07304494827985764,
      "learning_rate": 9.206267664155907e-05,
      "loss": 0.2687993347644806,
      "step": 53
    },
    {
      "epoch": 1.881057268722467,
      "grad_norm": 0.055231839418411255,
      "learning_rate": 9.167238453576589e-05,
      "loss": 0.17293335497379303,
      "step": 54
    },
    {
      "epoch": 1.9162995594713657,
      "grad_norm": 0.04506454989314079,
      "learning_rate": 9.12735948481387e-05,
      "loss": 0.21196959912776947,
      "step": 55
    },
    {
      "epoch": 1.9515418502202642,
      "grad_norm": 0.043365973979234695,
      "learning_rate": 9.086638889747035e-05,
      "loss": 0.17916671931743622,
      "step": 56
    },
    {
      "epoch": 1.9867841409691631,
      "grad_norm": 0.04912644252181053,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.24066588282585144,
      "step": 57
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.08831573277711868,
      "learning_rate": 9.002706204621803e-05,
      "loss": 0.19998322427272797,
      "step": 58
    },
    {
      "epoch": 2.0352422907488985,
      "grad_norm": 0.06712567806243896,
      "learning_rate": 8.959511229611376e-05,
      "loss": 0.1661558896303177,
      "step": 59
    },
    {
      "epoch": 2.0704845814977975,
      "grad_norm": 0.03976866602897644,
      "learning_rate": 8.915508854902778e-05,
      "loss": 0.173383891582489,
      "step": 60
    },
    {
      "epoch": 2.0704845814977975,
      "eval_loss": 0.17691832780838013,
      "eval_runtime": 5.559,
      "eval_samples_per_second": 8.994,
      "eval_steps_per_second": 8.994,
      "step": 60
    },
    {
      "epoch": 2.0704845814977975,
      "eval_perplexity": 1.1935336106406358,
      "step": 60
    },
    {
      "epoch": 2.105726872246696,
      "grad_norm": 0.03575924411416054,
      "learning_rate": 8.870708053195413e-05,
      "loss": 0.11559620499610901,
      "step": 61
    },
    {
      "epoch": 2.140969162995595,
      "grad_norm": 0.03323924168944359,
      "learning_rate": 8.825117959999116e-05,
      "loss": 0.1270863115787506,
      "step": 62
    },
    {
      "epoch": 2.1762114537444934,
      "grad_norm": 0.038361676037311554,
      "learning_rate": 8.778747871771292e-05,
      "loss": 0.15825499594211578,
      "step": 63
    },
    {
      "epoch": 2.211453744493392,
      "grad_norm": 0.0353211909532547,
      "learning_rate": 8.731607244021236e-05,
      "loss": 0.14611881971359253,
      "step": 64
    },
    {
      "epoch": 2.246696035242291,
      "grad_norm": 0.04414643347263336,
      "learning_rate": 8.683705689382024e-05,
      "loss": 0.1989956796169281,
      "step": 65
    },
    {
      "epoch": 2.2819383259911894,
      "grad_norm": 0.035386525094509125,
      "learning_rate": 8.635052975650369e-05,
      "loss": 0.1573476791381836,
      "step": 66
    },
    {
      "epoch": 2.317180616740088,
      "grad_norm": 0.04224468767642975,
      "learning_rate": 8.585659023794818e-05,
      "loss": 0.17944753170013428,
      "step": 67
    },
    {
      "epoch": 2.352422907488987,
      "grad_norm": 0.040473032742738724,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.19669276475906372,
      "step": 68
    },
    {
      "epoch": 2.3876651982378854,
      "grad_norm": 0.04288116469979286,
      "learning_rate": 8.484687843276469e-05,
      "loss": 0.168636754155159,
      "step": 69
    },
    {
      "epoch": 2.4229074889867843,
      "grad_norm": 0.04106265306472778,
      "learning_rate": 8.433131204049067e-05,
      "loss": 0.14968255162239075,
      "step": 70
    },
    {
      "epoch": 2.4229074889867843,
      "eval_loss": 0.17395253479480743,
      "eval_runtime": 5.5663,
      "eval_samples_per_second": 8.983,
      "eval_steps_per_second": 8.983,
      "step": 70
    },
    {
      "epoch": 2.4229074889867843,
      "eval_perplexity": 1.1899990809292844,
      "step": 70
    },
    {
      "epoch": 2.458149779735683,
      "grad_norm": 0.039962347596883774,
      "learning_rate": 8.380874501370097e-05,
      "loss": 0.15449365973472595,
      "step": 71
    },
    {
      "epoch": 2.4933920704845813,
      "grad_norm": 0.04701816663146019,
      "learning_rate": 8.327928391111841e-05,
      "loss": 0.1417522430419922,
      "step": 72
    },
    {
      "epoch": 2.5286343612334803,
      "grad_norm": 0.0489797480404377,
      "learning_rate": 8.274303669726426e-05,
      "loss": 0.14920863509178162,
      "step": 73
    },
    {
      "epoch": 2.563876651982379,
      "grad_norm": 0.035862814635038376,
      "learning_rate": 8.220011272044277e-05,
      "loss": 0.13807804882526398,
      "step": 74
    },
    {
      "epoch": 2.5991189427312777,
      "grad_norm": 0.03754030168056488,
      "learning_rate": 8.165062269044353e-05,
      "loss": 0.1002129390835762,
      "step": 75
    },
    {
      "epoch": 2.6343612334801763,
      "grad_norm": 0.03721809759736061,
      "learning_rate": 8.109467865596612e-05,
      "loss": 0.14590686559677124,
      "step": 76
    },
    {
      "epoch": 2.6696035242290748,
      "grad_norm": 0.04426711052656174,
      "learning_rate": 8.053239398177191e-05,
      "loss": 0.1654958873987198,
      "step": 77
    },
    {
      "epoch": 2.7048458149779737,
      "grad_norm": 0.04694978520274162,
      "learning_rate": 7.996388332556735e-05,
      "loss": 0.17505832016468048,
      "step": 78
    },
    {
      "epoch": 2.7400881057268722,
      "grad_norm": 0.04963630810379982,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.17087210714817047,
      "step": 79
    },
    {
      "epoch": 2.7753303964757707,
      "grad_norm": 0.04072735831141472,
      "learning_rate": 7.880864902213765e-05,
      "loss": 0.11690441519021988,
      "step": 80
    },
    {
      "epoch": 2.7753303964757707,
      "eval_loss": 0.1763938069343567,
      "eval_runtime": 5.5438,
      "eval_samples_per_second": 9.019,
      "eval_steps_per_second": 9.019,
      "step": 80
    },
    {
      "epoch": 2.7753303964757707,
      "eval_perplexity": 1.192907741503069,
      "step": 80
    },
    {
      "epoch": 2.8105726872246697,
      "grad_norm": 0.04362448304891586,
      "learning_rate": 7.822216094333847e-05,
      "loss": 0.18380989134311676,
      "step": 81
    },
    {
      "epoch": 2.845814977973568,
      "grad_norm": 0.08166193962097168,
      "learning_rate": 7.762991797134514e-05,
      "loss": 0.18010962009429932,
      "step": 82
    },
    {
      "epoch": 2.8810572687224667,
      "grad_norm": 0.041152581572532654,
      "learning_rate": 7.703204087277988e-05,
      "loss": 0.1840750277042389,
      "step": 83
    },
    {
      "epoch": 2.9162995594713657,
      "grad_norm": 0.0513804629445076,
      "learning_rate": 7.64286515631421e-05,
      "loss": 0.14389707148075104,
      "step": 84
    },
    {
      "epoch": 2.951541850220264,
      "grad_norm": 0.048202160745859146,
      "learning_rate": 7.58198730819481e-05,
      "loss": 0.15499596297740936,
      "step": 85
    },
    {
      "epoch": 2.986784140969163,
      "grad_norm": 0.048344533890485764,
      "learning_rate": 7.52058295676416e-05,
      "loss": 0.1284043788909912,
      "step": 86
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.06424666941165924,
      "learning_rate": 7.45866462322802e-05,
      "loss": 0.12437701970338821,
      "step": 87
    },
    {
      "epoch": 3.0352422907488985,
      "grad_norm": 0.05029122158885002,
      "learning_rate": 7.396244933600285e-05,
      "loss": 0.16445861756801605,
      "step": 88
    },
    {
      "epoch": 3.0704845814977975,
      "grad_norm": 0.03273235633969307,
      "learning_rate": 7.333336616128369e-05,
      "loss": 0.08856770396232605,
      "step": 89
    },
    {
      "epoch": 3.105726872246696,
      "grad_norm": 0.040203846991062164,
      "learning_rate": 7.269952498697734e-05,
      "loss": 0.11728361248970032,
      "step": 90
    },
    {
      "epoch": 3.105726872246696,
      "eval_loss": 0.18123260140419006,
      "eval_runtime": 5.5343,
      "eval_samples_per_second": 9.035,
      "eval_steps_per_second": 9.035,
      "step": 90
    },
    {
      "epoch": 3.105726872246696,
      "eval_perplexity": 1.1986939647683954,
      "step": 90
    },
    {
      "epoch": 3.140969162995595,
      "grad_norm": 0.03822289779782295,
      "learning_rate": 7.206105506216106e-05,
      "loss": 0.12806665897369385,
      "step": 91
    },
    {
      "epoch": 3.1762114537444934,
      "grad_norm": 0.03790947049856186,
      "learning_rate": 7.141808657977907e-05,
      "loss": 0.12213566899299622,
      "step": 92
    },
    {
      "epoch": 3.211453744493392,
      "grad_norm": 0.03983893617987633,
      "learning_rate": 7.077075065009433e-05,
      "loss": 0.14553172886371613,
      "step": 93
    },
    {
      "epoch": 3.246696035242291,
      "grad_norm": 0.0287310890853405,
      "learning_rate": 7.01191792739534e-05,
      "loss": 0.068100206553936,
      "step": 94
    },
    {
      "epoch": 3.2819383259911894,
      "grad_norm": 0.04404347762465477,
      "learning_rate": 6.946350531586959e-05,
      "loss": 0.07530707120895386,
      "step": 95
    },
    {
      "epoch": 3.317180616740088,
      "grad_norm": 0.03737332299351692,
      "learning_rate": 6.880386247692999e-05,
      "loss": 0.13074061274528503,
      "step": 96
    },
    {
      "epoch": 3.352422907488987,
      "grad_norm": 0.04556211084127426,
      "learning_rate": 6.814038526753205e-05,
      "loss": 0.1442827582359314,
      "step": 97
    },
    {
      "epoch": 3.3876651982378854,
      "grad_norm": 0.03330394998192787,
      "learning_rate": 6.747320897995493e-05,
      "loss": 0.06684030592441559,
      "step": 98
    },
    {
      "epoch": 3.4229074889867843,
      "grad_norm": 0.044004376977682114,
      "learning_rate": 6.680246966077151e-05,
      "loss": 0.06947988271713257,
      "step": 99
    },
    {
      "epoch": 3.458149779735683,
      "grad_norm": 0.04061594605445862,
      "learning_rate": 6.61283040831067e-05,
      "loss": 0.12199011445045471,
      "step": 100
    },
    {
      "epoch": 3.458149779735683,
      "eval_loss": 0.17973366379737854,
      "eval_runtime": 5.5457,
      "eval_samples_per_second": 9.016,
      "eval_steps_per_second": 9.016,
      "step": 100
    },
    {
      "epoch": 3.458149779735683,
      "eval_perplexity": 1.1968985432541248,
      "step": 100
    },
    {
      "epoch": 3.4933920704845813,
      "grad_norm": 0.035715501755476,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.09066961705684662,
      "step": 101
    },
    {
      "epoch": 3.5286343612334803,
      "grad_norm": 0.03552844747900963,
      "learning_rate": 6.477024471011001e-05,
      "loss": 0.11516473442316055,
      "step": 102
    },
    {
      "epoch": 3.563876651982379,
      "grad_norm": 0.037987079471349716,
      "learning_rate": 6.408662784207149e-05,
      "loss": 0.12247294187545776,
      "step": 103
    },
    {
      "epoch": 3.5991189427312777,
      "grad_norm": 0.05642999708652496,
      "learning_rate": 6.340013851366896e-05,
      "loss": 0.17614305019378662,
      "step": 104
    },
    {
      "epoch": 3.6343612334801763,
      "grad_norm": 0.046289026737213135,
      "learning_rate": 6.271091670967436e-05,
      "loss": 0.08439283818006516,
      "step": 105
    },
    {
      "epoch": 3.6696035242290748,
      "grad_norm": 0.05821594223380089,
      "learning_rate": 6.201910297204962e-05,
      "loss": 0.1878018081188202,
      "step": 106
    },
    {
      "epoch": 3.7048458149779737,
      "grad_norm": 0.05191818252205849,
      "learning_rate": 6.132483837128823e-05,
      "loss": 0.1220342367887497,
      "step": 107
    },
    {
      "epoch": 3.7400881057268722,
      "grad_norm": 0.041689906269311905,
      "learning_rate": 6.062826447764883e-05,
      "loss": 0.07030203938484192,
      "step": 108
    },
    {
      "epoch": 3.7753303964757707,
      "grad_norm": 0.06289155781269073,
      "learning_rate": 5.992952333228728e-05,
      "loss": 0.1382126659154892,
      "step": 109
    },
    {
      "epoch": 3.8105726872246697,
      "grad_norm": 0.05354708805680275,
      "learning_rate": 5.9228757418292266e-05,
      "loss": 0.10265767574310303,
      "step": 110
    },
    {
      "epoch": 3.8105726872246697,
      "eval_loss": 0.18481433391571045,
      "eval_runtime": 5.5391,
      "eval_samples_per_second": 9.027,
      "eval_steps_per_second": 9.027,
      "step": 110
    },
    {
      "epoch": 3.8105726872246697,
      "eval_perplexity": 1.2029950640086982,
      "step": 110
    },
    {
      "epoch": 3.845814977973568,
      "grad_norm": 0.048672065138816833,
      "learning_rate": 5.85261096316312e-05,
      "loss": 0.10617683827877045,
      "step": 111
    },
    {
      "epoch": 3.8810572687224667,
      "grad_norm": 0.050876542925834656,
      "learning_rate": 5.782172325201155e-05,
      "loss": 0.12120889127254486,
      "step": 112
    },
    {
      "epoch": 3.9162995594713657,
      "grad_norm": 0.041362058371305466,
      "learning_rate": 5.7115741913664264e-05,
      "loss": 0.08605430275201797,
      "step": 113
    },
    {
      "epoch": 3.951541850220264,
      "grad_norm": 0.04938527196645737,
      "learning_rate": 5.640830957605465e-05,
      "loss": 0.13313618302345276,
      "step": 114
    },
    {
      "epoch": 3.986784140969163,
      "grad_norm": 0.06675253063440323,
      "learning_rate": 5.569957049452703e-05,
      "loss": 0.12465330958366394,
      "step": 115
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.09207858890295029,
      "learning_rate": 5.4989669190889136e-05,
      "loss": 0.14120639860630035,
      "step": 116
    },
    {
      "epoch": 4.035242290748899,
      "grad_norm": 0.03314574807882309,
      "learning_rate": 5.427875042394199e-05,
      "loss": 0.06799043715000153,
      "step": 117
    },
    {
      "epoch": 4.070484581497797,
      "grad_norm": 0.03621016442775726,
      "learning_rate": 5.3566959159961615e-05,
      "loss": 0.06281277537345886,
      "step": 118
    },
    {
      "epoch": 4.105726872246696,
      "grad_norm": 0.05796584114432335,
      "learning_rate": 5.2854440543138406e-05,
      "loss": 0.12782785296440125,
      "step": 119
    },
    {
      "epoch": 4.140969162995595,
      "grad_norm": 0.045323941856622696,
      "learning_rate": 5.2141339865980134e-05,
      "loss": 0.11147820204496384,
      "step": 120
    },
    {
      "epoch": 4.140969162995595,
      "eval_loss": 0.20198915898799896,
      "eval_runtime": 5.5511,
      "eval_samples_per_second": 9.007,
      "eval_steps_per_second": 9.007,
      "step": 120
    },
    {
      "epoch": 4.140969162995595,
      "eval_perplexity": 1.2238347404323324,
      "step": 120
    },
    {
      "epoch": 4.140969162995595,
      "step": 120,
      "total_flos": 1.423789286185943e+17,
      "train_loss": 0.26457088105380533,
      "train_runtime": 827.7026,
      "train_samples_per_second": 4.388,
      "train_steps_per_second": 0.28
    },
    {
      "epoch": 4.140969162995595,
      "eval_loss": 0.17395253479480743,
      "eval_runtime": 5.5408,
      "eval_samples_per_second": 9.024,
      "eval_steps_per_second": 9.024,
      "step": 120
    },
    {
      "epoch": 4.140969162995595,
      "eval_perplexity": 1.1899990809292844,
      "step": 120
    }
  ],
  "logging_steps": 1,
  "max_steps": 232,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 10,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.001
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.423789286185943e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}