{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 5460,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 0.0019010665128007531,
      "learning_rate": 0.0005,
      "loss": 0.0009,
      "step": 25
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.33505979180336,
      "learning_rate": 0.001,
      "loss": 0.0059,
      "step": 50
    },
    {
      "epoch": 0.14,
      "grad_norm": 2.521899700164795,
      "learning_rate": 0.0009953789279112755,
      "loss": 0.2569,
      "step": 75
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.3200310468673706,
      "learning_rate": 0.000990757855822551,
      "loss": 2.1666,
      "step": 100
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.1342906951904297,
      "learning_rate": 0.0009861367837338264,
      "loss": 0.3677,
      "step": 125
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.6839067935943604,
      "learning_rate": 0.0009815157116451016,
      "loss": 0.2573,
      "step": 150
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.120778203010559,
      "learning_rate": 0.0009768946395563771,
      "loss": 0.1784,
      "step": 175
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.39630964398384094,
      "learning_rate": 0.0009722735674676525,
      "loss": 0.1007,
      "step": 200
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.8115565776824951,
      "learning_rate": 0.0009676524953789279,
      "loss": 0.1009,
      "step": 225
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.4256753921508789,
      "learning_rate": 0.0009630314232902033,
      "loss": 0.0613,
      "step": 250
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.3296879529953003,
      "learning_rate": 0.0009584103512014787,
      "loss": 0.0656,
      "step": 275
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.2332513928413391,
      "learning_rate": 0.0009537892791127542,
      "loss": 0.1056,
      "step": 300
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.361102819442749,
      "learning_rate": 0.0009491682070240297,
      "loss": 0.0714,
      "step": 325
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.3176443576812744,
      "learning_rate": 0.000944547134935305,
      "loss": 0.051,
      "step": 350
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.5561681389808655,
      "learning_rate": 0.0009399260628465805,
      "loss": 0.0987,
      "step": 375
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.17937970161437988,
      "learning_rate": 0.0009353049907578558,
      "loss": 0.0705,
      "step": 400
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.11134567111730576,
      "learning_rate": 0.0009306839186691313,
      "loss": 0.0449,
      "step": 425
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.1100844293832779,
      "learning_rate": 0.0009260628465804066,
      "loss": 0.0605,
      "step": 450
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.5044831037521362,
      "learning_rate": 0.0009214417744916821,
      "loss": 0.0682,
      "step": 475
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.1506507396697998,
      "learning_rate": 0.0009168207024029575,
      "loss": 0.0382,
      "step": 500
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.358964741230011,
      "learning_rate": 0.0009121996303142329,
      "loss": 0.0416,
      "step": 525
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.22685140371322632,
      "eval_runtime": 467.2357,
      "eval_samples_per_second": 1.74,
      "eval_steps_per_second": 0.291,
      "step": 546
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.21843843162059784,
      "learning_rate": 0.0009075785582255084,
      "loss": 0.0283,
      "step": 550
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.3907661437988281,
      "learning_rate": 0.0009029574861367837,
      "loss": 0.0322,
      "step": 575
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.026330502703785896,
      "learning_rate": 0.0008983364140480592,
      "loss": 0.021,
      "step": 600
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.015783503651618958,
      "learning_rate": 0.0008937153419593346,
      "loss": 0.0237,
      "step": 625
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.07014349848031998,
      "learning_rate": 0.00088909426987061,
      "loss": 0.0307,
      "step": 650
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.05321989953517914,
      "learning_rate": 0.0008844731977818854,
      "loss": 0.0217,
      "step": 675
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.07509706914424896,
      "learning_rate": 0.0008798521256931608,
      "loss": 0.0319,
      "step": 700
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.239598348736763,
      "learning_rate": 0.0008752310536044362,
      "loss": 0.0373,
      "step": 725
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.1276847869157791,
      "learning_rate": 0.0008706099815157116,
      "loss": 0.0308,
      "step": 750
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.17856040596961975,
      "learning_rate": 0.000865988909426987,
      "loss": 0.0326,
      "step": 775
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.5299984216690063,
      "learning_rate": 0.0008613678373382626,
      "loss": 0.034,
      "step": 800
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.18570055067539215,
      "learning_rate": 0.0008567467652495379,
      "loss": 0.0255,
      "step": 825
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.012400169856846333,
      "learning_rate": 0.0008521256931608134,
      "loss": 0.0133,
      "step": 850
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.18435439467430115,
      "learning_rate": 0.0008475046210720887,
      "loss": 0.0392,
      "step": 875
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.20227985084056854,
      "learning_rate": 0.0008428835489833642,
      "loss": 0.0339,
      "step": 900
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.09818145632743835,
      "learning_rate": 0.0008382624768946395,
      "loss": 0.039,
      "step": 925
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.17143449187278748,
      "learning_rate": 0.000833641404805915,
      "loss": 0.0256,
      "step": 950
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.3052353858947754,
      "learning_rate": 0.0008290203327171904,
      "loss": 0.0279,
      "step": 975
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.09069275110960007,
      "learning_rate": 0.0008243992606284658,
      "loss": 0.0253,
      "step": 1000
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.7721070647239685,
      "learning_rate": 0.0008197781885397413,
      "loss": 0.0222,
      "step": 1025
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.2378959357738495,
      "learning_rate": 0.0008151571164510166,
      "loss": 0.0386,
      "step": 1050
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.004904525820165873,
      "learning_rate": 0.0008105360443622921,
      "loss": 0.0243,
      "step": 1075
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.20544852316379547,
      "eval_runtime": 463.5714,
      "eval_samples_per_second": 1.754,
      "eval_steps_per_second": 0.293,
      "step": 1092
    },
    {
      "epoch": 2.01,
      "grad_norm": 0.13851934671401978,
      "learning_rate": 0.0008059149722735675,
      "loss": 0.0267,
      "step": 1100
    },
    {
      "epoch": 2.06,
      "grad_norm": 0.01875193975865841,
      "learning_rate": 0.0008012939001848429,
      "loss": 0.0117,
      "step": 1125
    },
    {
      "epoch": 2.11,
      "grad_norm": 0.0016854548593983054,
      "learning_rate": 0.0007966728280961183,
      "loss": 0.0317,
      "step": 1150
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.10259977728128433,
      "learning_rate": 0.0007920517560073937,
      "loss": 0.0253,
      "step": 1175
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.054936520755290985,
      "learning_rate": 0.0007874306839186691,
      "loss": 0.0228,
      "step": 1200
    },
    {
      "epoch": 2.24,
      "grad_norm": 0.08183781057596207,
      "learning_rate": 0.0007828096118299445,
      "loss": 0.0163,
      "step": 1225
    },
    {
      "epoch": 2.29,
      "grad_norm": 0.03448671102523804,
      "learning_rate": 0.0007781885397412199,
      "loss": 0.0252,
      "step": 1250
    },
    {
      "epoch": 2.34,
      "grad_norm": 0.026254719123244286,
      "learning_rate": 0.0007735674676524955,
      "loss": 0.0263,
      "step": 1275
    },
    {
      "epoch": 2.38,
      "grad_norm": 0.07833431661128998,
      "learning_rate": 0.0007689463955637708,
      "loss": 0.0268,
      "step": 1300
    },
    {
      "epoch": 2.43,
      "grad_norm": 0.3215916156768799,
      "learning_rate": 0.0007643253234750463,
      "loss": 0.0184,
      "step": 1325
    },
    {
      "epoch": 2.47,
      "grad_norm": 0.1949949562549591,
      "learning_rate": 0.0007597042513863216,
      "loss": 0.0234,
      "step": 1350
    },
    {
      "epoch": 2.52,
      "grad_norm": 0.10761301964521408,
      "learning_rate": 0.0007550831792975971,
      "loss": 0.0214,
      "step": 1375
    },
    {
      "epoch": 2.56,
      "grad_norm": 0.23488566279411316,
      "learning_rate": 0.0007504621072088724,
      "loss": 0.0321,
      "step": 1400
    },
    {
      "epoch": 2.61,
      "grad_norm": 0.16079489886760712,
      "learning_rate": 0.0007458410351201479,
      "loss": 0.041,
      "step": 1425
    },
    {
      "epoch": 2.66,
      "grad_norm": 0.3500367999076843,
      "learning_rate": 0.0007412199630314234,
      "loss": 0.0218,
      "step": 1450
    },
    {
      "epoch": 2.7,
      "grad_norm": 0.01966880075633526,
      "learning_rate": 0.0007365988909426987,
      "loss": 0.0185,
      "step": 1475
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.10084854066371918,
      "learning_rate": 0.0007319778188539742,
      "loss": 0.0158,
      "step": 1500
    },
    {
      "epoch": 2.79,
      "grad_norm": 0.045843809843063354,
      "learning_rate": 0.0007273567467652495,
      "loss": 0.0193,
      "step": 1525
    },
    {
      "epoch": 2.84,
      "grad_norm": 0.19230197370052338,
      "learning_rate": 0.000722735674676525,
      "loss": 0.0115,
      "step": 1550
    },
    {
      "epoch": 2.88,
      "grad_norm": 0.10168833285570145,
      "learning_rate": 0.0007181146025878004,
      "loss": 0.0173,
      "step": 1575
    },
    {
      "epoch": 2.93,
      "grad_norm": 0.24770613014698029,
      "learning_rate": 0.0007134935304990758,
      "loss": 0.019,
      "step": 1600
    },
    {
      "epoch": 2.98,
      "grad_norm": 0.04277370125055313,
      "learning_rate": 0.0007088724584103512,
      "loss": 0.0262,
      "step": 1625
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.18655328452587128,
      "eval_runtime": 463.8687,
      "eval_samples_per_second": 1.753,
      "eval_steps_per_second": 0.293,
      "step": 1638
    },
    {
      "epoch": 3.02,
      "grad_norm": 0.03143952414393425,
      "learning_rate": 0.0007042513863216266,
      "loss": 0.0185,
      "step": 1650
    },
    {
      "epoch": 3.07,
      "grad_norm": 0.061480745673179626,
      "learning_rate": 0.000699630314232902,
      "loss": 0.0158,
      "step": 1675
    },
    {
      "epoch": 3.11,
      "grad_norm": 0.05645143985748291,
      "learning_rate": 0.0006950092421441774,
      "loss": 0.0163,
      "step": 1700
    },
    {
      "epoch": 3.16,
      "grad_norm": 0.3927539885044098,
      "learning_rate": 0.0006903881700554528,
      "loss": 0.0257,
      "step": 1725
    },
    {
      "epoch": 3.21,
      "grad_norm": 0.1579461544752121,
      "learning_rate": 0.0006857670979667284,
      "loss": 0.0203,
      "step": 1750
    },
    {
      "epoch": 3.25,
      "grad_norm": 0.003284105099737644,
      "learning_rate": 0.0006811460258780037,
      "loss": 0.012,
      "step": 1775
    },
    {
      "epoch": 3.3,
      "grad_norm": 0.0939943715929985,
      "learning_rate": 0.0006765249537892792,
      "loss": 0.0139,
      "step": 1800
    },
    {
      "epoch": 3.34,
      "grad_norm": 0.08114974200725555,
      "learning_rate": 0.0006719038817005545,
      "loss": 0.0134,
      "step": 1825
    },
    {
      "epoch": 3.39,
      "grad_norm": 0.008277042768895626,
      "learning_rate": 0.00066728280961183,
      "loss": 0.0203,
      "step": 1850
    },
    {
      "epoch": 3.43,
      "grad_norm": 0.014137201942503452,
      "learning_rate": 0.0006626617375231053,
      "loss": 0.018,
      "step": 1875
    },
    {
      "epoch": 3.48,
      "grad_norm": 0.04209378361701965,
      "learning_rate": 0.0006580406654343808,
      "loss": 0.0107,
      "step": 1900
    },
    {
      "epoch": 3.53,
      "grad_norm": 0.015557551756501198,
      "learning_rate": 0.0006534195933456563,
      "loss": 0.0104,
      "step": 1925
    },
    {
      "epoch": 3.57,
      "grad_norm": 0.021405475214123726,
      "learning_rate": 0.0006487985212569316,
      "loss": 0.0117,
      "step": 1950
    },
    {
      "epoch": 3.62,
      "grad_norm": 0.0015239958884194493,
      "learning_rate": 0.0006441774491682071,
      "loss": 0.0176,
      "step": 1975
    },
    {
      "epoch": 3.66,
      "grad_norm": 0.0997876301407814,
      "learning_rate": 0.0006395563770794824,
      "loss": 0.0183,
      "step": 2000
    },
    {
      "epoch": 3.71,
      "grad_norm": 0.004715020768344402,
      "learning_rate": 0.0006349353049907579,
      "loss": 0.0199,
      "step": 2025
    },
    {
      "epoch": 3.75,
      "grad_norm": 0.1075858548283577,
      "learning_rate": 0.0006303142329020333,
      "loss": 0.0201,
      "step": 2050
    },
    {
      "epoch": 3.8,
      "grad_norm": 0.020496558398008347,
      "learning_rate": 0.0006256931608133087,
      "loss": 0.0145,
      "step": 2075
    },
    {
      "epoch": 3.85,
      "grad_norm": 0.11063025891780853,
      "learning_rate": 0.0006210720887245841,
      "loss": 0.0201,
      "step": 2100
    },
    {
      "epoch": 3.89,
      "grad_norm": 0.1012192815542221,
      "learning_rate": 0.0006164510166358595,
      "loss": 0.0223,
      "step": 2125
    },
    {
      "epoch": 3.94,
      "grad_norm": 0.04694315418601036,
      "learning_rate": 0.0006118299445471349,
      "loss": 0.0163,
      "step": 2150
    },
    {
      "epoch": 3.98,
      "grad_norm": 0.05395512282848358,
      "learning_rate": 0.0006072088724584103,
      "loss": 0.009,
      "step": 2175
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.20004291832447052,
      "eval_runtime": 464.3668,
      "eval_samples_per_second": 1.751,
      "eval_steps_per_second": 0.293,
      "step": 2184
    },
    {
      "epoch": 4.03,
      "grad_norm": 0.08517912030220032,
      "learning_rate": 0.0006025878003696857,
      "loss": 0.0152,
      "step": 2200
    },
    {
      "epoch": 4.08,
      "grad_norm": 0.23693686723709106,
      "learning_rate": 0.0005979667282809613,
      "loss": 0.0123,
      "step": 2225
    },
    {
      "epoch": 4.12,
      "grad_norm": 0.04390133172273636,
      "learning_rate": 0.0005933456561922366,
      "loss": 0.0104,
      "step": 2250
    },
    {
      "epoch": 4.17,
      "grad_norm": 0.048480305820703506,
      "learning_rate": 0.0005887245841035121,
      "loss": 0.0191,
      "step": 2275
    },
    {
      "epoch": 4.21,
      "grad_norm": 0.07334431260824203,
      "learning_rate": 0.0005841035120147874,
      "loss": 0.0079,
      "step": 2300
    },
    {
      "epoch": 4.26,
      "grad_norm": 0.26686009764671326,
      "learning_rate": 0.0005794824399260629,
      "loss": 0.0134,
      "step": 2325
    },
    {
      "epoch": 4.3,
      "grad_norm": 0.18834412097930908,
      "learning_rate": 0.0005748613678373382,
      "loss": 0.0108,
      "step": 2350
    },
    {
      "epoch": 4.35,
      "grad_norm": 0.11365604400634766,
      "learning_rate": 0.0005702402957486137,
      "loss": 0.0116,
      "step": 2375
    },
    {
      "epoch": 4.4,
      "grad_norm": 0.21077445149421692,
      "learning_rate": 0.0005656192236598892,
      "loss": 0.017,
      "step": 2400
    },
    {
      "epoch": 4.44,
      "grad_norm": 0.14450936019420624,
      "learning_rate": 0.0005609981515711645,
      "loss": 0.0056,
      "step": 2425
    },
    {
      "epoch": 4.49,
      "grad_norm": 0.07659462839365005,
      "learning_rate": 0.00055637707948244,
      "loss": 0.0128,
      "step": 2450
    },
    {
      "epoch": 4.53,
      "grad_norm": 0.07819797843694687,
      "learning_rate": 0.0005517560073937153,
      "loss": 0.0085,
      "step": 2475
    },
    {
      "epoch": 4.58,
      "grad_norm": 0.10529200732707977,
      "learning_rate": 0.0005471349353049908,
      "loss": 0.0156,
      "step": 2500
    },
    {
      "epoch": 4.62,
      "grad_norm": 0.034541305154561996,
      "learning_rate": 0.0005425138632162662,
      "loss": 0.0114,
      "step": 2525
    },
    {
      "epoch": 4.67,
      "grad_norm": 0.0043388293124735355,
      "learning_rate": 0.0005378927911275416,
      "loss": 0.0114,
      "step": 2550
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.09843795001506805,
      "learning_rate": 0.000533271719038817,
      "loss": 0.0097,
      "step": 2575
    },
    {
      "epoch": 4.76,
      "grad_norm": 0.1924191564321518,
      "learning_rate": 0.0005286506469500924,
      "loss": 0.0138,
      "step": 2600
    },
    {
      "epoch": 4.81,
      "grad_norm": 0.0032940045930445194,
      "learning_rate": 0.0005240295748613678,
      "loss": 0.009,
      "step": 2625
    },
    {
      "epoch": 4.85,
      "grad_norm": 0.17411276698112488,
      "learning_rate": 0.0005194085027726432,
      "loss": 0.005,
      "step": 2650
    },
    {
      "epoch": 4.9,
      "grad_norm": 0.0008068850729614496,
      "learning_rate": 0.0005147874306839186,
      "loss": 0.0091,
      "step": 2675
    },
    {
      "epoch": 4.95,
      "grad_norm": 0.013785873539745808,
      "learning_rate": 0.0005101663585951941,
      "loss": 0.0174,
      "step": 2700
    },
    {
      "epoch": 4.99,
      "grad_norm": 0.06957102566957474,
      "learning_rate": 0.0005055452865064695,
      "loss": 0.0196,
      "step": 2725
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.1927657425403595,
      "eval_runtime": 464.5332,
      "eval_samples_per_second": 1.75,
      "eval_steps_per_second": 0.293,
      "step": 2730
    },
    {
      "epoch": 5.04,
      "grad_norm": 0.1873362511396408,
      "learning_rate": 0.000500924214417745,
      "loss": 0.0114,
      "step": 2750
    },
    {
      "epoch": 5.08,
      "grad_norm": 0.013944294303655624,
      "learning_rate": 0.0004963031423290203,
      "loss": 0.0047,
      "step": 2775
    },
    {
      "epoch": 5.13,
      "grad_norm": 0.14739681780338287,
      "learning_rate": 0.0004916820702402958,
      "loss": 0.0064,
      "step": 2800
    },
    {
      "epoch": 5.17,
      "grad_norm": 0.039295587688684464,
      "learning_rate": 0.00048706099815157115,
      "loss": 0.0061,
      "step": 2825
    },
    {
      "epoch": 5.22,
      "grad_norm": 0.009731476195156574,
      "learning_rate": 0.0004824399260628466,
      "loss": 0.0064,
      "step": 2850
    },
    {
      "epoch": 5.27,
      "grad_norm": 0.009130421094596386,
      "learning_rate": 0.000477818853974122,
      "loss": 0.0056,
      "step": 2875
    },
    {
      "epoch": 5.31,
      "grad_norm": 0.10517439246177673,
      "learning_rate": 0.0004731977818853974,
      "loss": 0.0095,
      "step": 2900
    },
    {
      "epoch": 5.36,
      "grad_norm": 0.03147244080901146,
      "learning_rate": 0.00046857670979667283,
      "loss": 0.0069,
      "step": 2925
    },
    {
      "epoch": 5.4,
      "grad_norm": 0.07550155371427536,
      "learning_rate": 0.00046395563770794824,
      "loss": 0.0084,
      "step": 2950
    },
    {
      "epoch": 5.45,
      "grad_norm": 0.09899873286485672,
      "learning_rate": 0.00045933456561922365,
      "loss": 0.0087,
      "step": 2975
    },
    {
      "epoch": 5.49,
      "grad_norm": 0.062454238533973694,
      "learning_rate": 0.00045471349353049906,
      "loss": 0.0114,
      "step": 3000
    },
    {
      "epoch": 5.54,
      "grad_norm": 0.14996998012065887,
      "learning_rate": 0.00045009242144177446,
      "loss": 0.0091,
      "step": 3025
    },
    {
      "epoch": 5.59,
      "grad_norm": 0.19108814001083374,
      "learning_rate": 0.00044547134935304987,
      "loss": 0.0147,
      "step": 3050
    },
    {
      "epoch": 5.63,
      "grad_norm": 0.14450325071811676,
      "learning_rate": 0.00044085027726432533,
      "loss": 0.0152,
      "step": 3075
    },
    {
      "epoch": 5.68,
      "grad_norm": 0.04423892870545387,
      "learning_rate": 0.0004362292051756008,
      "loss": 0.006,
      "step": 3100
    },
    {
      "epoch": 5.72,
      "grad_norm": 0.13844439387321472,
      "learning_rate": 0.0004316081330868762,
      "loss": 0.009,
      "step": 3125
    },
    {
      "epoch": 5.77,
      "grad_norm": 0.0006735218339599669,
      "learning_rate": 0.0004269870609981516,
      "loss": 0.0058,
      "step": 3150
    },
    {
      "epoch": 5.82,
      "grad_norm": 0.011760660447180271,
      "learning_rate": 0.000422365988909427,
      "loss": 0.0049,
      "step": 3175
    },
    {
      "epoch": 5.86,
      "grad_norm": 0.08969856053590775,
      "learning_rate": 0.0004177449168207024,
      "loss": 0.0065,
      "step": 3200
    },
    {
      "epoch": 5.91,
      "grad_norm": 0.12556907534599304,
      "learning_rate": 0.00041312384473197783,
      "loss": 0.0089,
      "step": 3225
    },
    {
      "epoch": 5.95,
      "grad_norm": 0.017725255340337753,
      "learning_rate": 0.00040850277264325324,
      "loss": 0.0088,
      "step": 3250
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.009897828102111816,
      "learning_rate": 0.00040388170055452864,
      "loss": 0.0071,
      "step": 3275
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.20994354784488678,
      "eval_runtime": 463.0078,
      "eval_samples_per_second": 1.756,
      "eval_steps_per_second": 0.294,
      "step": 3276
    },
    {
      "epoch": 6.04,
      "grad_norm": 0.0028004159685224295,
      "learning_rate": 0.00039926062846580405,
      "loss": 0.0093,
      "step": 3300
    },
    {
      "epoch": 6.09,
      "grad_norm": 0.10490375012159348,
      "learning_rate": 0.0003946395563770795,
      "loss": 0.0053,
      "step": 3325
    },
    {
      "epoch": 6.14,
      "grad_norm": 0.019779745489358902,
      "learning_rate": 0.0003900184842883549,
      "loss": 0.0084,
      "step": 3350
    },
    {
      "epoch": 6.18,
      "grad_norm": 0.00020589173072949052,
      "learning_rate": 0.00038539741219963033,
      "loss": 0.0029,
      "step": 3375
    },
    {
      "epoch": 6.23,
      "grad_norm": 0.003221085062250495,
      "learning_rate": 0.00038077634011090574,
      "loss": 0.0051,
      "step": 3400
    },
    {
      "epoch": 6.27,
      "grad_norm": 0.00455264188349247,
      "learning_rate": 0.00037615526802218114,
      "loss": 0.0063,
      "step": 3425
    },
    {
      "epoch": 6.32,
      "grad_norm": 0.00967650581151247,
      "learning_rate": 0.00037153419593345655,
      "loss": 0.0035,
      "step": 3450
    },
    {
      "epoch": 6.36,
      "grad_norm": 0.009352604858577251,
      "learning_rate": 0.00036691312384473196,
      "loss": 0.0065,
      "step": 3475
    },
    {
      "epoch": 6.41,
      "grad_norm": 0.002876508515328169,
      "learning_rate": 0.00036229205175600736,
      "loss": 0.0049,
      "step": 3500
    },
    {
      "epoch": 6.46,
      "grad_norm": 0.002914861775934696,
      "learning_rate": 0.00035767097966728277,
      "loss": 0.0043,
      "step": 3525
    },
    {
      "epoch": 6.5,
      "grad_norm": 0.021481545642018318,
      "learning_rate": 0.0003530499075785583,
      "loss": 0.0072,
      "step": 3550
    },
    {
      "epoch": 6.55,
      "grad_norm": 0.08110266923904419,
      "learning_rate": 0.0003484288354898337,
      "loss": 0.0044,
      "step": 3575
    },
    {
      "epoch": 6.59,
      "grad_norm": 0.020943278446793556,
      "learning_rate": 0.0003438077634011091,
      "loss": 0.0072,
      "step": 3600
    },
    {
      "epoch": 6.64,
      "grad_norm": 0.005692564882338047,
      "learning_rate": 0.0003391866913123845,
      "loss": 0.0078,
      "step": 3625
    },
    {
      "epoch": 6.68,
      "grad_norm": 0.11609622091054916,
      "learning_rate": 0.0003345656192236599,
      "loss": 0.008,
      "step": 3650
    },
    {
      "epoch": 6.73,
      "grad_norm": 0.05904560536146164,
      "learning_rate": 0.0003299445471349353,
      "loss": 0.0061,
      "step": 3675
    },
    {
      "epoch": 6.78,
      "grad_norm": 0.03346557542681694,
      "learning_rate": 0.00032532347504621073,
      "loss": 0.0069,
      "step": 3700
    },
    {
      "epoch": 6.82,
      "grad_norm": 0.04848520830273628,
      "learning_rate": 0.00032070240295748614,
      "loss": 0.0078,
      "step": 3725
    },
    {
      "epoch": 6.87,
      "grad_norm": 0.11064545810222626,
      "learning_rate": 0.00031608133086876155,
      "loss": 0.0083,
      "step": 3750
    },
    {
      "epoch": 6.91,
      "grad_norm": 0.001821186626330018,
      "learning_rate": 0.00031146025878003695,
      "loss": 0.0084,
      "step": 3775
    },
    {
      "epoch": 6.96,
      "grad_norm": 0.03919747844338417,
      "learning_rate": 0.0003068391866913124,
      "loss": 0.0054,
      "step": 3800
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.20703129470348358,
      "eval_runtime": 463.5868,
      "eval_samples_per_second": 1.754,
      "eval_steps_per_second": 0.293,
      "step": 3822
    },
    {
      "epoch": 7.01,
      "grad_norm": 0.008144889958202839,
      "learning_rate": 0.0003022181146025878,
      "loss": 0.0057,
      "step": 3825
    },
    {
      "epoch": 7.05,
      "grad_norm": 0.005378293804824352,
      "learning_rate": 0.00029759704251386323,
      "loss": 0.004,
      "step": 3850
    },
    {
      "epoch": 7.1,
      "grad_norm": 0.03501349315047264,
      "learning_rate": 0.00029297597042513864,
      "loss": 0.003,
      "step": 3875
    },
    {
      "epoch": 7.14,
      "grad_norm": 0.07073014974594116,
      "learning_rate": 0.00028835489833641404,
      "loss": 0.0029,
      "step": 3900
    },
    {
      "epoch": 7.19,
      "grad_norm": 0.09017951786518097,
      "learning_rate": 0.00028373382624768945,
      "loss": 0.0027,
      "step": 3925
    },
    {
      "epoch": 7.23,
      "grad_norm": 0.009881277568638325,
      "learning_rate": 0.00027911275415896486,
      "loss": 0.0044,
      "step": 3950
    },
    {
      "epoch": 7.28,
      "grad_norm": 0.0018990118987858295,
      "learning_rate": 0.00027449168207024027,
      "loss": 0.0031,
      "step": 3975
    },
    {
      "epoch": 7.33,
      "grad_norm": 0.004116680007427931,
      "learning_rate": 0.00026987060998151567,
      "loss": 0.0026,
      "step": 4000
    },
    {
      "epoch": 7.37,
      "grad_norm": 0.03917807340621948,
      "learning_rate": 0.00026524953789279113,
      "loss": 0.0038,
      "step": 4025
    },
    {
      "epoch": 7.42,
      "grad_norm": 0.0030583201441913843,
      "learning_rate": 0.0002606284658040666,
      "loss": 0.0032,
      "step": 4050
    },
    {
      "epoch": 7.46,
      "grad_norm": 0.0014874553307890892,
      "learning_rate": 0.000256007393715342,
      "loss": 0.0054,
      "step": 4075
    },
    {
      "epoch": 7.51,
      "grad_norm": 0.0008628646028228104,
      "learning_rate": 0.0002513863216266174,
      "loss": 0.0019,
      "step": 4100
    },
    {
      "epoch": 7.55,
      "grad_norm": 0.02715575322508812,
      "learning_rate": 0.00024676524953789276,
      "loss": 0.0037,
      "step": 4125
    },
    {
      "epoch": 7.6,
      "grad_norm": 0.0031906655058264732,
      "learning_rate": 0.00024214417744916822,
      "loss": 0.0058,
      "step": 4150
    },
    {
      "epoch": 7.65,
      "grad_norm": 0.011863148771226406,
      "learning_rate": 0.00023752310536044363,
      "loss": 0.0022,
      "step": 4175
    },
    {
      "epoch": 7.69,
      "grad_norm": 0.0015202141366899014,
      "learning_rate": 0.00023290203327171904,
      "loss": 0.0045,
      "step": 4200
    },
    {
      "epoch": 7.74,
      "grad_norm": 0.02240474335849285,
      "learning_rate": 0.00022828096118299447,
      "loss": 0.0039,
      "step": 4225
    },
    {
      "epoch": 7.78,
      "grad_norm": 0.00918908603489399,
      "learning_rate": 0.00022365988909426988,
      "loss": 0.0014,
      "step": 4250
    },
    {
      "epoch": 7.83,
      "grad_norm": 0.005950120277702808,
      "learning_rate": 0.0002190388170055453,
      "loss": 0.0031,
      "step": 4275
    },
    {
      "epoch": 7.88,
      "grad_norm": 0.07433830946683884,
      "learning_rate": 0.0002144177449168207,
      "loss": 0.002,
      "step": 4300
    },
    {
      "epoch": 7.92,
      "grad_norm": 0.09878811240196228,
      "learning_rate": 0.0002097966728280961,
      "loss": 0.0039,
      "step": 4325
    },
    {
      "epoch": 7.97,
      "grad_norm": 0.004627088084816933,
      "learning_rate": 0.00020517560073937154,
      "loss": 0.0066,
      "step": 4350
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.21887589991092682,
      "eval_runtime": 464.2458,
      "eval_samples_per_second": 1.751,
      "eval_steps_per_second": 0.293,
      "step": 4368
    },
    {
      "epoch": 8.01,
      "grad_norm": 0.016953645274043083,
      "learning_rate": 0.00020055452865064697,
      "loss": 0.002,
      "step": 4375
    },
    {
      "epoch": 8.06,
      "grad_norm": 0.00016232863708864897,
      "learning_rate": 0.00019593345656192238,
      "loss": 0.0022,
      "step": 4400
    },
    {
      "epoch": 8.1,
      "grad_norm": 0.00045125139877200127,
      "learning_rate": 0.00019131238447319779,
      "loss": 0.0016,
      "step": 4425
    },
    {
      "epoch": 8.15,
      "grad_norm": 0.02065761759877205,
      "learning_rate": 0.0001866913123844732,
      "loss": 0.0017,
      "step": 4450
    },
    {
      "epoch": 8.2,
      "grad_norm": 0.042185261845588684,
      "learning_rate": 0.00018207024029574863,
      "loss": 0.0027,
      "step": 4475
    },
    {
      "epoch": 8.24,
      "grad_norm": 0.003087196499109268,
      "learning_rate": 0.00017744916820702404,
      "loss": 0.0018,
      "step": 4500
    },
    {
      "epoch": 8.29,
      "grad_norm": 0.02859407104551792,
      "learning_rate": 0.00017282809611829944,
      "loss": 0.0015,
      "step": 4525
    },
    {
      "epoch": 8.33,
      "grad_norm": 0.00041793755372054875,
      "learning_rate": 0.00016820702402957485,
      "loss": 0.0035,
      "step": 4550
    },
    {
      "epoch": 8.38,
      "grad_norm": 0.0037734461948275566,
      "learning_rate": 0.00016358595194085026,
      "loss": 0.002,
      "step": 4575
    },
    {
      "epoch": 8.42,
      "grad_norm": 0.0030207443051040173,
      "learning_rate": 0.00015896487985212572,
      "loss": 0.0022,
      "step": 4600
    },
    {
      "epoch": 8.47,
      "grad_norm": 0.0026946039870381355,
      "learning_rate": 0.00015434380776340113,
      "loss": 0.0028,
      "step": 4625
    },
    {
      "epoch": 8.52,
      "grad_norm": 0.041892848908901215,
      "learning_rate": 0.00014972273567467653,
      "loss": 0.001,
      "step": 4650
    },
    {
      "epoch": 8.56,
      "grad_norm": 0.06906843930482864,
      "learning_rate": 0.00014510166358595194,
      "loss": 0.0013,
      "step": 4675
    },
    {
      "epoch": 8.61,
      "grad_norm": 0.00067297019995749,
      "learning_rate": 0.00014048059149722737,
      "loss": 0.0029,
      "step": 4700
    },
    {
      "epoch": 8.65,
      "grad_norm": 0.011746911332011223,
      "learning_rate": 0.00013585951940850278,
      "loss": 0.0012,
      "step": 4725
    },
    {
      "epoch": 8.7,
      "grad_norm": 0.0013995037879794836,
      "learning_rate": 0.0001312384473197782,
      "loss": 0.0017,
      "step": 4750
    },
    {
      "epoch": 8.75,
      "grad_norm": 0.009580204263329506,
      "learning_rate": 0.0001266173752310536,
      "loss": 0.001,
      "step": 4775
    },
    {
      "epoch": 8.79,
      "grad_norm": 0.0008843488758429885,
      "learning_rate": 0.00012199630314232903,
      "loss": 0.0015,
      "step": 4800
    },
    {
      "epoch": 8.84,
      "grad_norm": 0.0013571062590926886,
      "learning_rate": 0.00011737523105360444,
      "loss": 0.0024,
      "step": 4825
    },
    {
      "epoch": 8.88,
      "grad_norm": 0.01475840900093317,
      "learning_rate": 0.00011275415896487985,
      "loss": 0.0009,
      "step": 4850
    },
    {
      "epoch": 8.93,
      "grad_norm": 0.008486463688313961,
      "learning_rate": 0.00010813308687615527,
      "loss": 0.0022,
      "step": 4875
    },
    {
      "epoch": 8.97,
      "grad_norm": 0.0005981879075989127,
      "learning_rate": 0.00010351201478743069,
      "loss": 0.0006,
      "step": 4900
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.23254649341106415,
      "eval_runtime": 463.7576,
      "eval_samples_per_second": 1.753,
      "eval_steps_per_second": 0.293,
      "step": 4914
    },
    {
      "epoch": 9.02,
      "grad_norm": 0.011784604750573635,
      "learning_rate": 9.889094269870611e-05,
      "loss": 0.0028,
      "step": 4925
    },
    {
      "epoch": 9.07,
      "grad_norm": 0.0005488657625392079,
      "learning_rate": 9.426987060998152e-05,
      "loss": 0.0016,
      "step": 4950
    },
    {
      "epoch": 9.11,
      "grad_norm": 0.0024228901602327824,
      "learning_rate": 8.964879852125694e-05,
      "loss": 0.001,
      "step": 4975
    },
    {
      "epoch": 9.16,
      "grad_norm": 0.0021140037570148706,
      "learning_rate": 8.502772643253234e-05,
      "loss": 0.0014,
      "step": 5000
    },
    {
      "epoch": 9.2,
      "grad_norm": 0.0011844311375170946,
      "learning_rate": 8.040665434380776e-05,
      "loss": 0.001,
      "step": 5025
    },
    {
      "epoch": 9.25,
      "grad_norm": 0.011841055937111378,
      "learning_rate": 7.578558225508319e-05,
      "loss": 0.0009,
      "step": 5050
    },
    {
      "epoch": 9.29,
      "grad_norm": 0.013395372778177261,
      "learning_rate": 7.116451016635859e-05,
      "loss": 0.0009,
      "step": 5075
    },
    {
      "epoch": 9.34,
      "grad_norm": 0.05545121058821678,
      "learning_rate": 6.654343807763401e-05,
      "loss": 0.0012,
      "step": 5100
    },
    {
      "epoch": 9.39,
      "grad_norm": 0.01891588233411312,
      "learning_rate": 6.192236598890943e-05,
      "loss": 0.0006,
      "step": 5125
    },
    {
      "epoch": 9.43,
      "grad_norm": 0.0025335114914923906,
      "learning_rate": 5.730129390018484e-05,
      "loss": 0.0006,
      "step": 5150
    },
    {
      "epoch": 9.48,
      "grad_norm": 0.0021167423110455275,
      "learning_rate": 5.268022181146026e-05,
      "loss": 0.0007,
      "step": 5175
    },
    {
      "epoch": 9.52,
      "grad_norm": 0.0011415353510528803,
      "learning_rate": 4.8059149722735676e-05,
      "loss": 0.0014,
      "step": 5200
    },
    {
      "epoch": 9.57,
      "grad_norm": 0.00026013093884103,
      "learning_rate": 4.343807763401109e-05,
      "loss": 0.0007,
      "step": 5225
    },
    {
      "epoch": 9.62,
      "grad_norm": 0.03879648819565773,
      "learning_rate": 3.8817005545286504e-05,
      "loss": 0.0007,
      "step": 5250
    },
    {
      "epoch": 9.66,
      "grad_norm": 0.006720875855535269,
      "learning_rate": 3.4195933456561925e-05,
      "loss": 0.0009,
      "step": 5275
    },
    {
      "epoch": 9.71,
      "grad_norm": 0.006371485069394112,
      "learning_rate": 2.957486136783734e-05,
      "loss": 0.0009,
      "step": 5300
    },
    {
      "epoch": 9.75,
      "grad_norm": 0.012291524559259415,
      "learning_rate": 2.4953789279112753e-05,
      "loss": 0.0012,
      "step": 5325
    },
    {
      "epoch": 9.8,
      "grad_norm": 0.012388636358082294,
      "learning_rate": 2.033271719038817e-05,
      "loss": 0.0006,
      "step": 5350
    },
    {
      "epoch": 9.84,
      "grad_norm": 0.0905984491109848,
      "learning_rate": 1.5711645101663588e-05,
      "loss": 0.0011,
      "step": 5375
    },
    {
      "epoch": 9.89,
      "grad_norm": 0.0024207117967307568,
      "learning_rate": 1.1090573012939002e-05,
      "loss": 0.001,
      "step": 5400
    },
    {
      "epoch": 9.94,
      "grad_norm": 0.003070174716413021,
      "learning_rate": 6.469500924214418e-06,
      "loss": 0.0008,
      "step": 5425
    },
    {
      "epoch": 9.98,
      "grad_norm": 0.012533812783658504,
      "learning_rate": 1.8484288354898337e-06,
      "loss": 0.001,
      "step": 5450
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.23037216067314148,
      "eval_runtime": 463.1996,
      "eval_samples_per_second": 1.755,
      "eval_steps_per_second": 0.294,
      "step": 5460
    },
    {
      "epoch": 10.0,
      "step": 5460,
      "total_flos": 1.135723105419264e+20,
      "train_loss": 0.029495533068592733,
      "train_runtime": 29108.855,
      "train_samples_per_second": 1.124,
      "train_steps_per_second": 0.188
    }
  ],
  "logging_steps": 25,
  "max_steps": 5460,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.135723105419264e+20,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}