{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.02248959856066569,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002248959856066569,
      "grad_norm": 0.6395772099494934,
      "learning_rate": 1e-05,
      "loss": 0.2596,
      "step": 1
    },
    {
      "epoch": 0.0002248959856066569,
      "eval_loss": 0.24757881462574005,
      "eval_runtime": 817.293,
      "eval_samples_per_second": 4.582,
      "eval_steps_per_second": 0.574,
      "step": 1
    },
    {
      "epoch": 0.0004497919712133138,
      "grad_norm": 0.3925122022628784,
      "learning_rate": 2e-05,
      "loss": 0.2105,
      "step": 2
    },
    {
      "epoch": 0.0006746879568199708,
      "grad_norm": 0.5591574907302856,
      "learning_rate": 3e-05,
      "loss": 0.2967,
      "step": 3
    },
    {
      "epoch": 0.0008995839424266276,
      "grad_norm": 0.40480154752731323,
      "learning_rate": 4e-05,
      "loss": 0.2416,
      "step": 4
    },
    {
      "epoch": 0.0011244799280332846,
      "grad_norm": 0.40806055068969727,
      "learning_rate": 5e-05,
      "loss": 0.2484,
      "step": 5
    },
    {
      "epoch": 0.0013493759136399416,
      "grad_norm": 0.47302746772766113,
      "learning_rate": 6e-05,
      "loss": 0.2439,
      "step": 6
    },
    {
      "epoch": 0.0015742718992465985,
      "grad_norm": 0.595953643321991,
      "learning_rate": 7e-05,
      "loss": 0.2163,
      "step": 7
    },
    {
      "epoch": 0.0017991678848532553,
      "grad_norm": 0.4065772593021393,
      "learning_rate": 8e-05,
      "loss": 0.1615,
      "step": 8
    },
    {
      "epoch": 0.0020240638704599125,
      "grad_norm": 0.34312620759010315,
      "learning_rate": 9e-05,
      "loss": 0.1355,
      "step": 9
    },
    {
      "epoch": 0.0020240638704599125,
      "eval_loss": 0.12818512320518494,
      "eval_runtime": 818.844,
      "eval_samples_per_second": 4.574,
      "eval_steps_per_second": 0.573,
      "step": 9
    },
    {
      "epoch": 0.0022489598560665692,
      "grad_norm": 0.25303488969802856,
      "learning_rate": 0.0001,
      "loss": 0.0991,
      "step": 10
    },
    {
      "epoch": 0.002473855841673226,
      "grad_norm": 0.2839983403682709,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.0974,
      "step": 11
    },
    {
      "epoch": 0.002698751827279883,
      "grad_norm": 0.32634347677230835,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.1081,
      "step": 12
    },
    {
      "epoch": 0.00292364781288654,
      "grad_norm": 0.4044373333454132,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.0978,
      "step": 13
    },
    {
      "epoch": 0.003148543798493197,
      "grad_norm": 0.2824390232563019,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.0631,
      "step": 14
    },
    {
      "epoch": 0.003373439784099854,
      "grad_norm": 0.45528894662857056,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.1004,
      "step": 15
    },
    {
      "epoch": 0.0035983357697065106,
      "grad_norm": 0.3091089129447937,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.052,
      "step": 16
    },
    {
      "epoch": 0.0038232317553131678,
      "grad_norm": 0.5007195472717285,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.0665,
      "step": 17
    },
    {
      "epoch": 0.004048127740919825,
      "grad_norm": 0.5512179136276245,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.0478,
      "step": 18
    },
    {
      "epoch": 0.004048127740919825,
      "eval_loss": 0.04387344792485237,
      "eval_runtime": 818.3909,
      "eval_samples_per_second": 4.576,
      "eval_steps_per_second": 0.573,
      "step": 18
    },
    {
      "epoch": 0.004273023726526481,
      "grad_norm": 0.3249228894710541,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0364,
      "step": 19
    },
    {
      "epoch": 0.0044979197121331384,
      "grad_norm": 0.30455729365348816,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0466,
      "step": 20
    },
    {
      "epoch": 0.004722815697739796,
      "grad_norm": 0.23767390847206116,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.0372,
      "step": 21
    },
    {
      "epoch": 0.004947711683346452,
      "grad_norm": 0.2876013219356537,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0343,
      "step": 22
    },
    {
      "epoch": 0.005172607668953109,
      "grad_norm": 0.4053821563720703,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.0246,
      "step": 23
    },
    {
      "epoch": 0.005397503654559766,
      "grad_norm": 0.2427462935447693,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0258,
      "step": 24
    },
    {
      "epoch": 0.005622399640166423,
      "grad_norm": 0.19849058985710144,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0192,
      "step": 25
    },
    {
      "epoch": 0.00584729562577308,
      "grad_norm": 0.19882732629776,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0161,
      "step": 26
    },
    {
      "epoch": 0.006072191611379737,
      "grad_norm": 0.37108710408210754,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.017,
      "step": 27
    },
    {
      "epoch": 0.006072191611379737,
      "eval_loss": 0.015909897163510323,
      "eval_runtime": 818.5239,
      "eval_samples_per_second": 4.575,
      "eval_steps_per_second": 0.573,
      "step": 27
    },
    {
      "epoch": 0.006297087596986394,
      "grad_norm": 0.24871313571929932,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0088,
      "step": 28
    },
    {
      "epoch": 0.0065219835825930505,
      "grad_norm": 0.27095913887023926,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0154,
      "step": 29
    },
    {
      "epoch": 0.006746879568199708,
      "grad_norm": 0.3403993844985962,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.0122,
      "step": 30
    },
    {
      "epoch": 0.006971775553806365,
      "grad_norm": 0.22196494042873383,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.0209,
      "step": 31
    },
    {
      "epoch": 0.007196671539413021,
      "grad_norm": 0.2796581983566284,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.011,
      "step": 32
    },
    {
      "epoch": 0.007421567525019678,
      "grad_norm": 0.2604925334453583,
      "learning_rate": 8.473291852294987e-05,
      "loss": 0.0177,
      "step": 33
    },
    {
      "epoch": 0.0076464635106263355,
      "grad_norm": 0.49495238065719604,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.0213,
      "step": 34
    },
    {
      "epoch": 0.007871359496232992,
      "grad_norm": 0.261060506105423,
      "learning_rate": 8.213938048432697e-05,
      "loss": 0.0097,
      "step": 35
    },
    {
      "epoch": 0.00809625548183965,
      "grad_norm": 0.26140645146369934,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.012,
      "step": 36
    },
    {
      "epoch": 0.00809625548183965,
      "eval_loss": 0.012119167484343052,
      "eval_runtime": 818.628,
      "eval_samples_per_second": 4.575,
      "eval_steps_per_second": 0.573,
      "step": 36
    },
    {
      "epoch": 0.008321151467446306,
      "grad_norm": 0.1571914255619049,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.0036,
      "step": 37
    },
    {
      "epoch": 0.008546047453052963,
      "grad_norm": 0.22023926675319672,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.0067,
      "step": 38
    },
    {
      "epoch": 0.00877094343865962,
      "grad_norm": 0.2149086743593216,
      "learning_rate": 7.649596321166024e-05,
      "loss": 0.0074,
      "step": 39
    },
    {
      "epoch": 0.008995839424266277,
      "grad_norm": 0.27121996879577637,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0062,
      "step": 40
    },
    {
      "epoch": 0.009220735409872933,
      "grad_norm": 0.28654319047927856,
      "learning_rate": 7.347357813929454e-05,
      "loss": 0.0138,
      "step": 41
    },
    {
      "epoch": 0.009445631395479591,
      "grad_norm": 0.11791090667247772,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.0053,
      "step": 42
    },
    {
      "epoch": 0.009670527381086248,
      "grad_norm": 0.20881877839565277,
      "learning_rate": 7.033683215379002e-05,
      "loss": 0.016,
      "step": 43
    },
    {
      "epoch": 0.009895423366692904,
      "grad_norm": 0.27633312344551086,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.0082,
      "step": 44
    },
    {
      "epoch": 0.010120319352299562,
      "grad_norm": 0.3581463396549225,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.0226,
      "step": 45
    },
    {
      "epoch": 0.010120319352299562,
      "eval_loss": 0.011704214848577976,
      "eval_runtime": 818.4677,
      "eval_samples_per_second": 4.576,
      "eval_steps_per_second": 0.573,
      "step": 45
    },
    {
      "epoch": 0.010345215337906218,
      "grad_norm": 0.17337502539157867,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0052,
      "step": 46
    },
    {
      "epoch": 0.010570111323512875,
      "grad_norm": 0.3789822459220886,
      "learning_rate": 6.378186779084995e-05,
      "loss": 0.0107,
      "step": 47
    },
    {
      "epoch": 0.010795007309119533,
      "grad_norm": 0.22627811133861542,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.0054,
      "step": 48
    },
    {
      "epoch": 0.011019903294726189,
      "grad_norm": 0.10004477947950363,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 0.0037,
      "step": 49
    },
    {
      "epoch": 0.011244799280332845,
      "grad_norm": 0.5107955932617188,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.0153,
      "step": 50
    },
    {
      "epoch": 0.011469695265939503,
      "grad_norm": 0.1338202804327011,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.003,
      "step": 51
    },
    {
      "epoch": 0.01169459125154616,
      "grad_norm": 0.35056552290916443,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.0064,
      "step": 52
    },
    {
      "epoch": 0.011919487237152818,
      "grad_norm": 0.40434548258781433,
      "learning_rate": 5.348782368720626e-05,
      "loss": 0.0119,
      "step": 53
    },
    {
      "epoch": 0.012144383222759474,
      "grad_norm": 0.2577972710132599,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.0052,
      "step": 54
    },
    {
      "epoch": 0.012144383222759474,
      "eval_loss": 0.008480534888803959,
      "eval_runtime": 818.3944,
      "eval_samples_per_second": 4.576,
      "eval_steps_per_second": 0.573,
      "step": 54
    },
    {
      "epoch": 0.01236927920836613,
      "grad_norm": 0.5249858498573303,
      "learning_rate": 5e-05,
      "loss": 0.0104,
      "step": 55
    },
    {
      "epoch": 0.012594175193972788,
      "grad_norm": 0.280838280916214,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.0103,
      "step": 56
    },
    {
      "epoch": 0.012819071179579445,
      "grad_norm": 0.5980654358863831,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 0.012,
      "step": 57
    },
    {
      "epoch": 0.013043967165186101,
      "grad_norm": 0.2870578169822693,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.0111,
      "step": 58
    },
    {
      "epoch": 0.013268863150792759,
      "grad_norm": 0.09381887316703796,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 0.0015,
      "step": 59
    },
    {
      "epoch": 0.013493759136399415,
      "grad_norm": 0.1809445321559906,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.0046,
      "step": 60
    },
    {
      "epoch": 0.013718655122006072,
      "grad_norm": 0.1818263679742813,
      "learning_rate": 3.960441545911204e-05,
      "loss": 0.0077,
      "step": 61
    },
    {
      "epoch": 0.01394355110761273,
      "grad_norm": 0.34547847509384155,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.0156,
      "step": 62
    },
    {
      "epoch": 0.014168447093219386,
      "grad_norm": 0.5689454674720764,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 0.0058,
      "step": 63
    },
    {
      "epoch": 0.014168447093219386,
      "eval_loss": 0.0059502702206373215,
      "eval_runtime": 818.6217,
      "eval_samples_per_second": 4.575,
      "eval_steps_per_second": 0.573,
      "step": 63
    },
    {
      "epoch": 0.014393343078826042,
      "grad_norm": 0.3550148606300354,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0166,
      "step": 64
    },
    {
      "epoch": 0.0146182390644327,
      "grad_norm": 0.1510383039712906,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.0095,
      "step": 65
    },
    {
      "epoch": 0.014843135050039357,
      "grad_norm": 0.10363886505365372,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.0053,
      "step": 66
    },
    {
      "epoch": 0.015068031035646013,
      "grad_norm": 0.12097151577472687,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 0.0022,
      "step": 67
    },
    {
      "epoch": 0.015292927021252671,
      "grad_norm": 0.2542628347873688,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.0098,
      "step": 68
    },
    {
      "epoch": 0.015517823006859327,
      "grad_norm": 0.25181394815444946,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 0.005,
      "step": 69
    },
    {
      "epoch": 0.015742718992465984,
      "grad_norm": 0.14270396530628204,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0016,
      "step": 70
    },
    {
      "epoch": 0.015967614978072642,
      "grad_norm": 0.6511343121528625,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.0052,
      "step": 71
    },
    {
      "epoch": 0.0161925109636793,
      "grad_norm": 0.11533847451210022,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.0064,
      "step": 72
    },
    {
      "epoch": 0.0161925109636793,
      "eval_loss": 0.0056665195152163506,
      "eval_runtime": 818.3432,
      "eval_samples_per_second": 4.576,
      "eval_steps_per_second": 0.573,
      "step": 72
    },
    {
      "epoch": 0.016417406949285954,
      "grad_norm": 0.17094752192497253,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.0037,
      "step": 73
    },
    {
      "epoch": 0.016642302934892612,
      "grad_norm": 0.2323194444179535,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.0051,
      "step": 74
    },
    {
      "epoch": 0.01686719892049927,
      "grad_norm": 0.2130383402109146,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.0021,
      "step": 75
    },
    {
      "epoch": 0.017092094906105925,
      "grad_norm": 0.21683244407176971,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.0055,
      "step": 76
    },
    {
      "epoch": 0.017316990891712583,
      "grad_norm": 0.4317927956581116,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.0152,
      "step": 77
    },
    {
      "epoch": 0.01754188687731924,
      "grad_norm": 0.09037106484174728,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.0017,
      "step": 78
    },
    {
      "epoch": 0.017766782862925896,
      "grad_norm": 0.21414412558078766,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.0079,
      "step": 79
    },
    {
      "epoch": 0.017991678848532554,
      "grad_norm": 0.11598372459411621,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.0019,
      "step": 80
    },
    {
      "epoch": 0.018216574834139212,
      "grad_norm": 0.18292973935604095,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.0033,
      "step": 81
    },
    {
      "epoch": 0.018216574834139212,
      "eval_loss": 0.005181905347853899,
      "eval_runtime": 818.2261,
      "eval_samples_per_second": 4.577,
      "eval_steps_per_second": 0.573,
      "step": 81
    },
    {
      "epoch": 0.018441470819745866,
      "grad_norm": 0.37185707688331604,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0087,
      "step": 82
    },
    {
      "epoch": 0.018666366805352524,
      "grad_norm": 0.08201336115598679,
      "learning_rate": 8.548121372247918e-06,
      "loss": 0.0024,
      "step": 83
    },
    {
      "epoch": 0.018891262790959183,
      "grad_norm": 0.11379610002040863,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.0033,
      "step": 84
    },
    {
      "epoch": 0.019116158776565837,
      "grad_norm": 0.1044936403632164,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.0021,
      "step": 85
    },
    {
      "epoch": 0.019341054762172495,
      "grad_norm": 0.299877405166626,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.0048,
      "step": 86
    },
    {
      "epoch": 0.019565950747779153,
      "grad_norm": 0.039606399834156036,
      "learning_rate": 5.060297685041659e-06,
      "loss": 0.0011,
      "step": 87
    },
    {
      "epoch": 0.019790846733385808,
      "grad_norm": 0.1582029163837433,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.0047,
      "step": 88
    },
    {
      "epoch": 0.020015742718992466,
      "grad_norm": 0.2292255014181137,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.0033,
      "step": 89
    },
    {
      "epoch": 0.020240638704599124,
      "grad_norm": 0.07060254365205765,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.0019,
      "step": 90
    },
    {
      "epoch": 0.020240638704599124,
      "eval_loss": 0.004966777749359608,
      "eval_runtime": 818.3255,
      "eval_samples_per_second": 4.576,
      "eval_steps_per_second": 0.573,
      "step": 90
    },
    {
      "epoch": 0.02046553469020578,
      "grad_norm": 0.0997166708111763,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.0035,
      "step": 91
    },
    {
      "epoch": 0.020690430675812437,
      "grad_norm": 0.0572785846889019,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.0027,
      "step": 92
    },
    {
      "epoch": 0.020915326661419095,
      "grad_norm": 0.18832778930664062,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 0.0106,
      "step": 93
    },
    {
      "epoch": 0.02114022264702575,
      "grad_norm": 0.14147458970546722,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.0078,
      "step": 94
    },
    {
      "epoch": 0.021365118632632407,
      "grad_norm": 0.08944231271743774,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.0047,
      "step": 95
    },
    {
      "epoch": 0.021590014618239065,
      "grad_norm": 0.07234779745340347,
      "learning_rate": 4.865965629214819e-07,
      "loss": 0.0009,
      "step": 96
    },
    {
      "epoch": 0.02181491060384572,
      "grad_norm": 0.3288421034812927,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.0055,
      "step": 97
    },
    {
      "epoch": 0.022039806589452378,
      "grad_norm": 0.09930330514907837,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.0017,
      "step": 98
    },
    {
      "epoch": 0.022264702575059036,
      "grad_norm": 0.45012977719306946,
      "learning_rate": 3.04586490452119e-08,
      "loss": 0.0016,
      "step": 99
    },
    {
      "epoch": 0.022264702575059036,
      "eval_loss": 0.004946997854858637,
      "eval_runtime": 818.2178,
      "eval_samples_per_second": 4.577,
      "eval_steps_per_second": 0.573,
      "step": 99
    },
    {
      "epoch": 0.02248959856066569,
      "grad_norm": 0.10031512379646301,
      "learning_rate": 0.0,
      "loss": 0.0014,
      "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": 1.4652273850318848e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}