{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 100,
  "global_step": 4233,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01771793054571226,
      "grad_norm": 0.24673239886760712,
      "learning_rate": 0.0001,
      "loss": 3.8313,
      "step": 25
    },
    {
      "epoch": 0.03543586109142452,
      "grad_norm": 0.23432838916778564,
      "learning_rate": 0.0001,
      "loss": 3.6164,
      "step": 50
    },
    {
      "epoch": 0.05315379163713678,
      "grad_norm": 0.4262351393699646,
      "learning_rate": 0.0001,
      "loss": 3.4113,
      "step": 75
    },
    {
      "epoch": 0.07087172218284904,
      "grad_norm": 0.329630047082901,
      "learning_rate": 0.0001,
      "loss": 3.1874,
      "step": 100
    },
    {
      "epoch": 0.07087172218284904,
      "eval_loss": 3.0117650032043457,
      "eval_runtime": 31.2497,
      "eval_samples_per_second": 9.536,
      "eval_steps_per_second": 2.4,
      "step": 100
    },
    {
      "epoch": 0.0885896527285613,
      "grad_norm": 0.39532458782196045,
      "learning_rate": 0.0001,
      "loss": 3.0928,
      "step": 125
    },
    {
      "epoch": 0.10630758327427356,
      "grad_norm": 0.38724836707115173,
      "learning_rate": 0.0001,
      "loss": 2.8781,
      "step": 150
    },
    {
      "epoch": 0.12402551381998582,
      "grad_norm": 0.2721017897129059,
      "learning_rate": 0.0001,
      "loss": 2.9166,
      "step": 175
    },
    {
      "epoch": 0.14174344436569808,
      "grad_norm": 0.40747421979904175,
      "learning_rate": 0.0001,
      "loss": 2.8756,
      "step": 200
    },
    {
      "epoch": 0.14174344436569808,
      "eval_loss": 2.822810411453247,
      "eval_runtime": 32.4779,
      "eval_samples_per_second": 9.175,
      "eval_steps_per_second": 2.309,
      "step": 200
    },
    {
      "epoch": 0.15946137491141035,
      "grad_norm": 0.4922279119491577,
      "learning_rate": 0.0001,
      "loss": 2.9015,
      "step": 225
    },
    {
      "epoch": 0.1771793054571226,
      "grad_norm": 0.3577287197113037,
      "learning_rate": 0.0001,
      "loss": 2.8639,
      "step": 250
    },
    {
      "epoch": 0.19489723600283487,
      "grad_norm": 0.26928216218948364,
      "learning_rate": 0.0001,
      "loss": 2.6776,
      "step": 275
    },
    {
      "epoch": 0.21261516654854712,
      "grad_norm": 0.3623720109462738,
      "learning_rate": 0.0001,
      "loss": 2.7134,
      "step": 300
    },
    {
      "epoch": 0.21261516654854712,
      "eval_loss": 2.735769271850586,
      "eval_runtime": 32.6152,
      "eval_samples_per_second": 9.137,
      "eval_steps_per_second": 2.3,
      "step": 300
    },
    {
      "epoch": 0.2303330970942594,
      "grad_norm": 0.48846569657325745,
      "learning_rate": 0.0001,
      "loss": 2.7118,
      "step": 325
    },
    {
      "epoch": 0.24805102763997164,
      "grad_norm": 0.555242657661438,
      "learning_rate": 0.0001,
      "loss": 2.7072,
      "step": 350
    },
    {
      "epoch": 0.2657689581856839,
      "grad_norm": 0.3421138823032379,
      "learning_rate": 0.0001,
      "loss": 2.7114,
      "step": 375
    },
    {
      "epoch": 0.28348688873139616,
      "grad_norm": 0.3518970012664795,
      "learning_rate": 0.0001,
      "loss": 2.6948,
      "step": 400
    },
    {
      "epoch": 0.28348688873139616,
      "eval_loss": 2.6832730770111084,
      "eval_runtime": 32.6428,
      "eval_samples_per_second": 9.129,
      "eval_steps_per_second": 2.298,
      "step": 400
    },
    {
      "epoch": 0.30120481927710846,
      "grad_norm": 0.7523891925811768,
      "learning_rate": 0.0001,
      "loss": 2.7691,
      "step": 425
    },
    {
      "epoch": 0.3189227498228207,
      "grad_norm": 0.36019715666770935,
      "learning_rate": 0.0001,
      "loss": 2.7116,
      "step": 450
    },
    {
      "epoch": 0.33664068036853295,
      "grad_norm": 0.47855111956596375,
      "learning_rate": 0.0001,
      "loss": 2.5935,
      "step": 475
    },
    {
      "epoch": 0.3543586109142452,
      "grad_norm": 0.3621235489845276,
      "learning_rate": 0.0001,
      "loss": 2.6386,
      "step": 500
    },
    {
      "epoch": 0.3543586109142452,
      "eval_loss": 2.6440742015838623,
      "eval_runtime": 32.6151,
      "eval_samples_per_second": 9.137,
      "eval_steps_per_second": 2.3,
      "step": 500
    },
    {
      "epoch": 0.3720765414599575,
      "grad_norm": 0.6827765703201294,
      "learning_rate": 0.0001,
      "loss": 2.6515,
      "step": 525
    },
    {
      "epoch": 0.38979447200566975,
      "grad_norm": 0.5584849119186401,
      "learning_rate": 0.0001,
      "loss": 2.7017,
      "step": 550
    },
    {
      "epoch": 0.407512402551382,
      "grad_norm": 0.42108777165412903,
      "learning_rate": 0.0001,
      "loss": 2.6605,
      "step": 575
    },
    {
      "epoch": 0.42523033309709424,
      "grad_norm": 0.4811398684978485,
      "learning_rate": 0.0001,
      "loss": 2.6525,
      "step": 600
    },
    {
      "epoch": 0.42523033309709424,
      "eval_loss": 2.6149747371673584,
      "eval_runtime": 32.6256,
      "eval_samples_per_second": 9.134,
      "eval_steps_per_second": 2.299,
      "step": 600
    },
    {
      "epoch": 0.44294826364280654,
      "grad_norm": 0.3859870731830597,
      "learning_rate": 0.0001,
      "loss": 2.6805,
      "step": 625
    },
    {
      "epoch": 0.4606661941885188,
      "grad_norm": 0.5074867010116577,
      "learning_rate": 0.0001,
      "loss": 2.66,
      "step": 650
    },
    {
      "epoch": 0.47838412473423103,
      "grad_norm": 0.4459812343120575,
      "learning_rate": 0.0001,
      "loss": 2.645,
      "step": 675
    },
    {
      "epoch": 0.4961020552799433,
      "grad_norm": 0.3960532248020172,
      "learning_rate": 0.0001,
      "loss": 2.6242,
      "step": 700
    },
    {
      "epoch": 0.4961020552799433,
      "eval_loss": 2.5855672359466553,
      "eval_runtime": 32.5667,
      "eval_samples_per_second": 9.15,
      "eval_steps_per_second": 2.303,
      "step": 700
    },
    {
      "epoch": 0.5138199858256556,
      "grad_norm": 0.28771647810935974,
      "learning_rate": 0.0001,
      "loss": 2.6985,
      "step": 725
    },
    {
      "epoch": 0.5315379163713678,
      "grad_norm": 0.3489636182785034,
      "learning_rate": 0.0001,
      "loss": 2.7126,
      "step": 750
    },
    {
      "epoch": 0.5492558469170801,
      "grad_norm": 0.5809263586997986,
      "learning_rate": 0.0001,
      "loss": 2.558,
      "step": 775
    },
    {
      "epoch": 0.5669737774627923,
      "grad_norm": 0.5202405452728271,
      "learning_rate": 0.0001,
      "loss": 2.6444,
      "step": 800
    },
    {
      "epoch": 0.5669737774627923,
      "eval_loss": 2.5701286792755127,
      "eval_runtime": 32.5138,
      "eval_samples_per_second": 9.165,
      "eval_steps_per_second": 2.307,
      "step": 800
    },
    {
      "epoch": 0.5846917080085046,
      "grad_norm": 0.4100205600261688,
      "learning_rate": 0.0001,
      "loss": 2.5452,
      "step": 825
    },
    {
      "epoch": 0.6024096385542169,
      "grad_norm": 0.494243323802948,
      "learning_rate": 0.0001,
      "loss": 2.566,
      "step": 850
    },
    {
      "epoch": 0.6201275690999292,
      "grad_norm": 0.41979244351387024,
      "learning_rate": 0.0001,
      "loss": 2.6271,
      "step": 875
    },
    {
      "epoch": 0.6378454996456414,
      "grad_norm": 0.4423984885215759,
      "learning_rate": 0.0001,
      "loss": 2.6007,
      "step": 900
    },
    {
      "epoch": 0.6378454996456414,
      "eval_loss": 2.554037094116211,
      "eval_runtime": 32.562,
      "eval_samples_per_second": 9.152,
      "eval_steps_per_second": 2.303,
      "step": 900
    },
    {
      "epoch": 0.6555634301913537,
      "grad_norm": 0.49124300479888916,
      "learning_rate": 0.0001,
      "loss": 2.6006,
      "step": 925
    },
    {
      "epoch": 0.6732813607370659,
      "grad_norm": 0.4179531931877136,
      "learning_rate": 0.0001,
      "loss": 2.634,
      "step": 950
    },
    {
      "epoch": 0.6909992912827781,
      "grad_norm": 0.35822227597236633,
      "learning_rate": 0.0001,
      "loss": 2.5601,
      "step": 975
    },
    {
      "epoch": 0.7087172218284904,
      "grad_norm": 0.44015607237815857,
      "learning_rate": 0.0001,
      "loss": 2.462,
      "step": 1000
    },
    {
      "epoch": 0.7087172218284904,
      "eval_loss": 2.5418050289154053,
      "eval_runtime": 32.5949,
      "eval_samples_per_second": 9.143,
      "eval_steps_per_second": 2.301,
      "step": 1000
    },
    {
      "epoch": 0.7264351523742026,
      "grad_norm": 0.3899957835674286,
      "learning_rate": 0.0001,
      "loss": 2.538,
      "step": 1025
    },
    {
      "epoch": 0.744153082919915,
      "grad_norm": 0.5775083303451538,
      "learning_rate": 0.0001,
      "loss": 2.5553,
      "step": 1050
    },
    {
      "epoch": 0.7618710134656272,
      "grad_norm": 0.6485064625740051,
      "learning_rate": 0.0001,
      "loss": 2.5878,
      "step": 1075
    },
    {
      "epoch": 0.7795889440113395,
      "grad_norm": 0.5756754875183105,
      "learning_rate": 0.0001,
      "loss": 2.5641,
      "step": 1100
    },
    {
      "epoch": 0.7795889440113395,
      "eval_loss": 2.5314512252807617,
      "eval_runtime": 32.673,
      "eval_samples_per_second": 9.121,
      "eval_steps_per_second": 2.295,
      "step": 1100
    },
    {
      "epoch": 0.7973068745570517,
      "grad_norm": 0.46359655261039734,
      "learning_rate": 0.0001,
      "loss": 2.5062,
      "step": 1125
    },
    {
      "epoch": 0.815024805102764,
      "grad_norm": 0.6569529175758362,
      "learning_rate": 0.0001,
      "loss": 2.5921,
      "step": 1150
    },
    {
      "epoch": 0.8327427356484762,
      "grad_norm": 0.5323280096054077,
      "learning_rate": 0.0001,
      "loss": 2.4839,
      "step": 1175
    },
    {
      "epoch": 0.8504606661941885,
      "grad_norm": 0.46269166469573975,
      "learning_rate": 0.0001,
      "loss": 2.4672,
      "step": 1200
    },
    {
      "epoch": 0.8504606661941885,
      "eval_loss": 2.5238418579101562,
      "eval_runtime": 32.6451,
      "eval_samples_per_second": 9.128,
      "eval_steps_per_second": 2.297,
      "step": 1200
    },
    {
      "epoch": 0.8681785967399008,
      "grad_norm": 0.4144786596298218,
      "learning_rate": 0.0001,
      "loss": 2.5174,
      "step": 1225
    },
    {
      "epoch": 0.8858965272856131,
      "grad_norm": 0.6366602182388306,
      "learning_rate": 0.0001,
      "loss": 2.512,
      "step": 1250
    },
    {
      "epoch": 0.9036144578313253,
      "grad_norm": 0.490589439868927,
      "learning_rate": 0.0001,
      "loss": 2.5243,
      "step": 1275
    },
    {
      "epoch": 0.9213323883770376,
      "grad_norm": 0.5880224108695984,
      "learning_rate": 0.0001,
      "loss": 2.5017,
      "step": 1300
    },
    {
      "epoch": 0.9213323883770376,
      "eval_loss": 2.51455020904541,
      "eval_runtime": 32.6422,
      "eval_samples_per_second": 9.129,
      "eval_steps_per_second": 2.298,
      "step": 1300
    },
    {
      "epoch": 0.9390503189227498,
      "grad_norm": 0.5042569637298584,
      "learning_rate": 0.0001,
      "loss": 2.5238,
      "step": 1325
    },
    {
      "epoch": 0.9567682494684621,
      "grad_norm": 0.3827306032180786,
      "learning_rate": 0.0001,
      "loss": 2.5785,
      "step": 1350
    },
    {
      "epoch": 0.9744861800141743,
      "grad_norm": 0.4069231450557709,
      "learning_rate": 0.0001,
      "loss": 2.5312,
      "step": 1375
    },
    {
      "epoch": 0.9922041105598866,
      "grad_norm": 0.5515998005867004,
      "learning_rate": 0.0001,
      "loss": 2.6389,
      "step": 1400
    },
    {
      "epoch": 0.9922041105598866,
      "eval_loss": 2.5082926750183105,
      "eval_runtime": 32.6146,
      "eval_samples_per_second": 9.137,
      "eval_steps_per_second": 2.3,
      "step": 1400
    },
    {
      "epoch": 1.009922041105599,
      "grad_norm": 0.34569019079208374,
      "learning_rate": 0.0001,
      "loss": 2.4996,
      "step": 1425
    },
    {
      "epoch": 1.0276399716513112,
      "grad_norm": 0.4051329493522644,
      "learning_rate": 0.0001,
      "loss": 2.4935,
      "step": 1450
    },
    {
      "epoch": 1.0453579021970234,
      "grad_norm": 0.5617458820343018,
      "learning_rate": 0.0001,
      "loss": 2.5528,
      "step": 1475
    },
    {
      "epoch": 1.0630758327427356,
      "grad_norm": 0.7300685048103333,
      "learning_rate": 0.0001,
      "loss": 2.4869,
      "step": 1500
    },
    {
      "epoch": 1.0630758327427356,
      "eval_loss": 2.502122640609741,
      "eval_runtime": 32.6521,
      "eval_samples_per_second": 9.127,
      "eval_steps_per_second": 2.297,
      "step": 1500
    },
    {
      "epoch": 1.080793763288448,
      "grad_norm": 0.3424386978149414,
      "learning_rate": 0.0001,
      "loss": 2.5036,
      "step": 1525
    },
    {
      "epoch": 1.0985116938341601,
      "grad_norm": 0.48358920216560364,
      "learning_rate": 0.0001,
      "loss": 2.4546,
      "step": 1550
    },
    {
      "epoch": 1.1162296243798724,
      "grad_norm": 0.620297372341156,
      "learning_rate": 0.0001,
      "loss": 2.4505,
      "step": 1575
    },
    {
      "epoch": 1.1339475549255846,
      "grad_norm": 0.41402992606163025,
      "learning_rate": 0.0001,
      "loss": 2.5302,
      "step": 1600
    },
    {
      "epoch": 1.1339475549255846,
      "eval_loss": 2.4941623210906982,
      "eval_runtime": 32.6717,
      "eval_samples_per_second": 9.121,
      "eval_steps_per_second": 2.296,
      "step": 1600
    },
    {
      "epoch": 1.1516654854712969,
      "grad_norm": 0.6208804249763489,
      "learning_rate": 0.0001,
      "loss": 2.5035,
      "step": 1625
    },
    {
      "epoch": 1.1693834160170091,
      "grad_norm": 0.3958500623703003,
      "learning_rate": 0.0001,
      "loss": 2.6296,
      "step": 1650
    },
    {
      "epoch": 1.1871013465627214,
      "grad_norm": 0.5418444275856018,
      "learning_rate": 0.0001,
      "loss": 2.5194,
      "step": 1675
    },
    {
      "epoch": 1.2048192771084336,
      "grad_norm": 0.7059912085533142,
      "learning_rate": 0.0001,
      "loss": 2.497,
      "step": 1700
    },
    {
      "epoch": 1.2048192771084336,
      "eval_loss": 2.4886298179626465,
      "eval_runtime": 32.6428,
      "eval_samples_per_second": 9.129,
      "eval_steps_per_second": 2.298,
      "step": 1700
    },
    {
      "epoch": 1.222537207654146,
      "grad_norm": 0.5447824001312256,
      "learning_rate": 0.0001,
      "loss": 2.5098,
      "step": 1725
    },
    {
      "epoch": 1.2402551381998583,
      "grad_norm": 0.508121132850647,
      "learning_rate": 0.0001,
      "loss": 2.4463,
      "step": 1750
    },
    {
      "epoch": 1.2579730687455706,
      "grad_norm": 0.5141641497612,
      "learning_rate": 0.0001,
      "loss": 2.4407,
      "step": 1775
    },
    {
      "epoch": 1.2756909992912828,
      "grad_norm": 0.7895596623420715,
      "learning_rate": 0.0001,
      "loss": 2.4965,
      "step": 1800
    },
    {
      "epoch": 1.2756909992912828,
      "eval_loss": 2.4845895767211914,
      "eval_runtime": 32.6353,
      "eval_samples_per_second": 9.131,
      "eval_steps_per_second": 2.298,
      "step": 1800
    },
    {
      "epoch": 1.293408929836995,
      "grad_norm": 0.33818376064300537,
      "learning_rate": 0.0001,
      "loss": 2.3678,
      "step": 1825
    },
    {
      "epoch": 1.3111268603827073,
      "grad_norm": 0.36670777201652527,
      "learning_rate": 0.0001,
      "loss": 2.4522,
      "step": 1850
    },
    {
      "epoch": 1.3288447909284196,
      "grad_norm": 0.46683812141418457,
      "learning_rate": 0.0001,
      "loss": 2.5355,
      "step": 1875
    },
    {
      "epoch": 1.3465627214741318,
      "grad_norm": 0.45993196964263916,
      "learning_rate": 0.0001,
      "loss": 2.5535,
      "step": 1900
    },
    {
      "epoch": 1.3465627214741318,
      "eval_loss": 2.4782519340515137,
      "eval_runtime": 32.6522,
      "eval_samples_per_second": 9.126,
      "eval_steps_per_second": 2.297,
      "step": 1900
    },
    {
      "epoch": 1.364280652019844,
      "grad_norm": 0.4003180265426636,
      "learning_rate": 0.0001,
      "loss": 2.4353,
      "step": 1925
    },
    {
      "epoch": 1.3819985825655563,
      "grad_norm": 0.4715386629104614,
      "learning_rate": 0.0001,
      "loss": 2.4086,
      "step": 1950
    },
    {
      "epoch": 1.3997165131112685,
      "grad_norm": 0.4463669955730438,
      "learning_rate": 0.0001,
      "loss": 2.5299,
      "step": 1975
    },
    {
      "epoch": 1.4174344436569808,
      "grad_norm": 0.6194645762443542,
      "learning_rate": 0.0001,
      "loss": 2.5747,
      "step": 2000
    },
    {
      "epoch": 1.4174344436569808,
      "eval_loss": 2.4732108116149902,
      "eval_runtime": 32.5875,
      "eval_samples_per_second": 9.145,
      "eval_steps_per_second": 2.301,
      "step": 2000
    },
    {
      "epoch": 1.4351523742026933,
      "grad_norm": 0.4712235927581787,
      "learning_rate": 0.0001,
      "loss": 2.433,
      "step": 2025
    },
    {
      "epoch": 1.4528703047484055,
      "grad_norm": 0.5619576573371887,
      "learning_rate": 0.0001,
      "loss": 2.4883,
      "step": 2050
    },
    {
      "epoch": 1.4705882352941178,
      "grad_norm": 0.44078579545021057,
      "learning_rate": 0.0001,
      "loss": 2.5575,
      "step": 2075
    },
    {
      "epoch": 1.48830616583983,
      "grad_norm": 0.44347578287124634,
      "learning_rate": 0.0001,
      "loss": 2.4534,
      "step": 2100
    },
    {
      "epoch": 1.48830616583983,
      "eval_loss": 2.467924118041992,
      "eval_runtime": 32.6469,
      "eval_samples_per_second": 9.128,
      "eval_steps_per_second": 2.297,
      "step": 2100
    },
    {
      "epoch": 1.5060240963855422,
      "grad_norm": 0.4806898832321167,
      "learning_rate": 0.0001,
      "loss": 2.4605,
      "step": 2125
    },
    {
      "epoch": 1.5237420269312545,
      "grad_norm": 0.4710817337036133,
      "learning_rate": 0.0001,
      "loss": 2.5124,
      "step": 2150
    },
    {
      "epoch": 1.5414599574769667,
      "grad_norm": 0.4697023630142212,
      "learning_rate": 0.0001,
      "loss": 2.5251,
      "step": 2175
    },
    {
      "epoch": 1.559177888022679,
      "grad_norm": 0.4275937080383301,
      "learning_rate": 0.0001,
      "loss": 2.4909,
      "step": 2200
    },
    {
      "epoch": 1.559177888022679,
      "eval_loss": 2.465703248977661,
      "eval_runtime": 32.66,
      "eval_samples_per_second": 9.124,
      "eval_steps_per_second": 2.296,
      "step": 2200
    },
    {
      "epoch": 1.5768958185683912,
      "grad_norm": 0.6271502375602722,
      "learning_rate": 0.0001,
      "loss": 2.4563,
      "step": 2225
    },
    {
      "epoch": 1.5946137491141035,
      "grad_norm": 0.5484246015548706,
      "learning_rate": 0.0001,
      "loss": 2.4781,
      "step": 2250
    },
    {
      "epoch": 1.6123316796598157,
      "grad_norm": 0.7340563535690308,
      "learning_rate": 0.0001,
      "loss": 2.5204,
      "step": 2275
    },
    {
      "epoch": 1.630049610205528,
      "grad_norm": 0.43257761001586914,
      "learning_rate": 0.0001,
      "loss": 2.5192,
      "step": 2300
    },
    {
      "epoch": 1.630049610205528,
      "eval_loss": 2.461730718612671,
      "eval_runtime": 32.5818,
      "eval_samples_per_second": 9.146,
      "eval_steps_per_second": 2.302,
      "step": 2300
    },
    {
      "epoch": 1.6477675407512402,
      "grad_norm": 0.7394423484802246,
      "learning_rate": 0.0001,
      "loss": 2.5862,
      "step": 2325
    },
    {
      "epoch": 1.6654854712969525,
      "grad_norm": 0.48102429509162903,
      "learning_rate": 0.0001,
      "loss": 2.519,
      "step": 2350
    },
    {
      "epoch": 1.6832034018426647,
      "grad_norm": 0.5994846820831299,
      "learning_rate": 0.0001,
      "loss": 2.4566,
      "step": 2375
    },
    {
      "epoch": 1.700921332388377,
      "grad_norm": 0.4805436134338379,
      "learning_rate": 0.0001,
      "loss": 2.4271,
      "step": 2400
    },
    {
      "epoch": 1.700921332388377,
      "eval_loss": 2.457273006439209,
      "eval_runtime": 32.6806,
      "eval_samples_per_second": 9.119,
      "eval_steps_per_second": 2.295,
      "step": 2400
    },
    {
      "epoch": 1.7186392629340892,
      "grad_norm": 0.6208567023277283,
      "learning_rate": 0.0001,
      "loss": 2.3581,
      "step": 2425
    },
    {
      "epoch": 1.7363571934798014,
      "grad_norm": 0.44081413745880127,
      "learning_rate": 0.0001,
      "loss": 2.4295,
      "step": 2450
    },
    {
      "epoch": 1.7540751240255137,
      "grad_norm": 0.4629543721675873,
      "learning_rate": 0.0001,
      "loss": 2.4569,
      "step": 2475
    },
    {
      "epoch": 1.771793054571226,
      "grad_norm": 0.518991231918335,
      "learning_rate": 0.0001,
      "loss": 2.4855,
      "step": 2500
    },
    {
      "epoch": 1.771793054571226,
      "eval_loss": 2.454190731048584,
      "eval_runtime": 32.6137,
      "eval_samples_per_second": 9.137,
      "eval_steps_per_second": 2.3,
      "step": 2500
    },
    {
      "epoch": 1.7895109851169382,
      "grad_norm": 0.6166653037071228,
      "learning_rate": 0.0001,
      "loss": 2.4761,
      "step": 2525
    },
    {
      "epoch": 1.8072289156626506,
      "grad_norm": 0.5490785241127014,
      "learning_rate": 0.0001,
      "loss": 2.5232,
      "step": 2550
    },
    {
      "epoch": 1.824946846208363,
      "grad_norm": 0.6279402375221252,
      "learning_rate": 0.0001,
      "loss": 2.4425,
      "step": 2575
    },
    {
      "epoch": 1.8426647767540751,
      "grad_norm": 0.606396496295929,
      "learning_rate": 0.0001,
      "loss": 2.4599,
      "step": 2600
    },
    {
      "epoch": 1.8426647767540751,
      "eval_loss": 2.4530327320098877,
      "eval_runtime": 32.6381,
      "eval_samples_per_second": 9.13,
      "eval_steps_per_second": 2.298,
      "step": 2600
    },
    {
      "epoch": 1.8603827072997874,
      "grad_norm": 0.5355327129364014,
      "learning_rate": 0.0001,
      "loss": 2.5615,
      "step": 2625
    },
    {
      "epoch": 1.8781006378454996,
      "grad_norm": 0.3971356451511383,
      "learning_rate": 0.0001,
      "loss": 2.445,
      "step": 2650
    },
    {
      "epoch": 1.8958185683912119,
      "grad_norm": 0.48701226711273193,
      "learning_rate": 0.0001,
      "loss": 2.5405,
      "step": 2675
    },
    {
      "epoch": 1.9135364989369241,
      "grad_norm": 0.5117021203041077,
      "learning_rate": 0.0001,
      "loss": 2.4482,
      "step": 2700
    },
    {
      "epoch": 1.9135364989369241,
      "eval_loss": 2.444391965866089,
      "eval_runtime": 32.6264,
      "eval_samples_per_second": 9.134,
      "eval_steps_per_second": 2.299,
      "step": 2700
    },
    {
      "epoch": 1.9312544294826366,
      "grad_norm": 0.40752479434013367,
      "learning_rate": 0.0001,
      "loss": 2.421,
      "step": 2725
    },
    {
      "epoch": 1.9489723600283488,
      "grad_norm": 0.4466327428817749,
      "learning_rate": 0.0001,
      "loss": 2.4286,
      "step": 2750
    },
    {
      "epoch": 1.966690290574061,
      "grad_norm": 0.35452115535736084,
      "learning_rate": 0.0001,
      "loss": 2.4265,
      "step": 2775
    },
    {
      "epoch": 1.9844082211197733,
      "grad_norm": 0.7978025674819946,
      "learning_rate": 0.0001,
      "loss": 2.493,
      "step": 2800
    },
    {
      "epoch": 1.9844082211197733,
      "eval_loss": 2.444624423980713,
      "eval_runtime": 32.6679,
      "eval_samples_per_second": 9.122,
      "eval_steps_per_second": 2.296,
      "step": 2800
    },
    {
      "epoch": 2.0021261516654856,
      "grad_norm": 0.46374988555908203,
      "learning_rate": 0.0001,
      "loss": 2.5357,
      "step": 2825
    },
    {
      "epoch": 2.019844082211198,
      "grad_norm": 0.4631684422492981,
      "learning_rate": 0.0001,
      "loss": 2.4015,
      "step": 2850
    },
    {
      "epoch": 2.03756201275691,
      "grad_norm": 0.4475260376930237,
      "learning_rate": 0.0001,
      "loss": 2.3658,
      "step": 2875
    },
    {
      "epoch": 2.0552799433026223,
      "grad_norm": 0.47790655493736267,
      "learning_rate": 0.0001,
      "loss": 2.3527,
      "step": 2900
    },
    {
      "epoch": 2.0552799433026223,
      "eval_loss": 2.441364049911499,
      "eval_runtime": 32.6599,
      "eval_samples_per_second": 9.124,
      "eval_steps_per_second": 2.296,
      "step": 2900
    },
    {
      "epoch": 2.0729978738483346,
      "grad_norm": 0.5602151155471802,
      "learning_rate": 0.0001,
      "loss": 2.4763,
      "step": 2925
    },
    {
      "epoch": 2.090715804394047,
      "grad_norm": 0.37178730964660645,
      "learning_rate": 0.0001,
      "loss": 2.4431,
      "step": 2950
    },
    {
      "epoch": 2.108433734939759,
      "grad_norm": 0.47269827127456665,
      "learning_rate": 0.0001,
      "loss": 2.5528,
      "step": 2975
    },
    {
      "epoch": 2.1261516654854713,
      "grad_norm": 0.5636725425720215,
      "learning_rate": 0.0001,
      "loss": 2.5243,
      "step": 3000
    },
    {
      "epoch": 2.1261516654854713,
      "eval_loss": 2.4375791549682617,
      "eval_runtime": 32.6693,
      "eval_samples_per_second": 9.122,
      "eval_steps_per_second": 2.296,
      "step": 3000
    },
    {
      "epoch": 2.1438695960311835,
      "grad_norm": 0.5602971315383911,
      "learning_rate": 0.0001,
      "loss": 2.4726,
      "step": 3025
    },
    {
      "epoch": 2.161587526576896,
      "grad_norm": 0.7102957367897034,
      "learning_rate": 0.0001,
      "loss": 2.4513,
      "step": 3050
    },
    {
      "epoch": 2.179305457122608,
      "grad_norm": 0.5028663277626038,
      "learning_rate": 0.0001,
      "loss": 2.4038,
      "step": 3075
    },
    {
      "epoch": 2.1970233876683203,
      "grad_norm": 0.5358246564865112,
      "learning_rate": 0.0001,
      "loss": 2.4644,
      "step": 3100
    },
    {
      "epoch": 2.1970233876683203,
      "eval_loss": 2.433030605316162,
      "eval_runtime": 32.6486,
      "eval_samples_per_second": 9.128,
      "eval_steps_per_second": 2.297,
      "step": 3100
    },
    {
      "epoch": 2.2147413182140325,
      "grad_norm": 0.5380859971046448,
      "learning_rate": 0.0001,
      "loss": 2.4342,
      "step": 3125
    },
    {
      "epoch": 2.2324592487597448,
      "grad_norm": 0.8703417181968689,
      "learning_rate": 0.0001,
      "loss": 2.4462,
      "step": 3150
    },
    {
      "epoch": 2.250177179305457,
      "grad_norm": 0.44465309381484985,
      "learning_rate": 0.0001,
      "loss": 2.4428,
      "step": 3175
    },
    {
      "epoch": 2.2678951098511693,
      "grad_norm": 0.4541110396385193,
      "learning_rate": 0.0001,
      "loss": 2.386,
      "step": 3200
    },
    {
      "epoch": 2.2678951098511693,
      "eval_loss": 2.4308454990386963,
      "eval_runtime": 32.6678,
      "eval_samples_per_second": 9.122,
      "eval_steps_per_second": 2.296,
      "step": 3200
    },
    {
      "epoch": 2.2856130403968815,
      "grad_norm": 0.6527560949325562,
      "learning_rate": 0.0001,
      "loss": 2.3964,
      "step": 3225
    },
    {
      "epoch": 2.3033309709425938,
      "grad_norm": 0.5541362762451172,
      "learning_rate": 0.0001,
      "loss": 2.4817,
      "step": 3250
    },
    {
      "epoch": 2.321048901488306,
      "grad_norm": 0.5997689366340637,
      "learning_rate": 0.0001,
      "loss": 2.496,
      "step": 3275
    },
    {
      "epoch": 2.3387668320340183,
      "grad_norm": 0.5446316003799438,
      "learning_rate": 0.0001,
      "loss": 2.3762,
      "step": 3300
    },
    {
      "epoch": 2.3387668320340183,
      "eval_loss": 2.428109645843506,
      "eval_runtime": 32.6878,
      "eval_samples_per_second": 9.117,
      "eval_steps_per_second": 2.294,
      "step": 3300
    },
    {
      "epoch": 2.3564847625797305,
      "grad_norm": 0.3934761881828308,
      "learning_rate": 0.0001,
      "loss": 2.5195,
      "step": 3325
    },
    {
      "epoch": 2.3742026931254427,
      "grad_norm": 0.47348254919052124,
      "learning_rate": 0.0001,
      "loss": 2.5055,
      "step": 3350
    },
    {
      "epoch": 2.3919206236711554,
      "grad_norm": 0.5292345881462097,
      "learning_rate": 0.0001,
      "loss": 2.3635,
      "step": 3375
    },
    {
      "epoch": 2.4096385542168672,
      "grad_norm": 0.6056796312332153,
      "learning_rate": 0.0001,
      "loss": 2.3827,
      "step": 3400
    },
    {
      "epoch": 2.4096385542168672,
      "eval_loss": 2.4244818687438965,
      "eval_runtime": 32.5864,
      "eval_samples_per_second": 9.145,
      "eval_steps_per_second": 2.302,
      "step": 3400
    },
    {
      "epoch": 2.42735648476258,
      "grad_norm": 0.2907162010669708,
      "learning_rate": 0.0001,
      "loss": 2.3354,
      "step": 3425
    },
    {
      "epoch": 2.445074415308292,
      "grad_norm": 0.43741077184677124,
      "learning_rate": 0.0001,
      "loss": 2.4199,
      "step": 3450
    },
    {
      "epoch": 2.4627923458540044,
      "grad_norm": 0.36141782999038696,
      "learning_rate": 0.0001,
      "loss": 2.4165,
      "step": 3475
    },
    {
      "epoch": 2.4805102763997167,
      "grad_norm": 0.5461854338645935,
      "learning_rate": 0.0001,
      "loss": 2.3487,
      "step": 3500
    },
    {
      "epoch": 2.4805102763997167,
      "eval_loss": 2.4221482276916504,
      "eval_runtime": 32.6681,
      "eval_samples_per_second": 9.122,
      "eval_steps_per_second": 2.296,
      "step": 3500
    },
    {
      "epoch": 2.498228206945429,
      "grad_norm": 0.61762934923172,
      "learning_rate": 0.0001,
      "loss": 2.4136,
      "step": 3525
    },
    {
      "epoch": 2.515946137491141,
      "grad_norm": 0.41114169359207153,
      "learning_rate": 0.0001,
      "loss": 2.3587,
      "step": 3550
    },
    {
      "epoch": 2.5336640680368534,
      "grad_norm": 0.5726279020309448,
      "learning_rate": 0.0001,
      "loss": 2.5009,
      "step": 3575
    },
    {
      "epoch": 2.5513819985825656,
      "grad_norm": 0.4807787239551544,
      "learning_rate": 0.0001,
      "loss": 2.4737,
      "step": 3600
    },
    {
      "epoch": 2.5513819985825656,
      "eval_loss": 2.4191701412200928,
      "eval_runtime": 32.6743,
      "eval_samples_per_second": 9.12,
      "eval_steps_per_second": 2.295,
      "step": 3600
    },
    {
      "epoch": 2.569099929128278,
      "grad_norm": 0.5931722521781921,
      "learning_rate": 0.0001,
      "loss": 2.4178,
      "step": 3625
    },
    {
      "epoch": 2.58681785967399,
      "grad_norm": 0.4658395051956177,
      "learning_rate": 0.0001,
      "loss": 2.5162,
      "step": 3650
    },
    {
      "epoch": 2.6045357902197024,
      "grad_norm": 0.5829235315322876,
      "learning_rate": 0.0001,
      "loss": 2.3402,
      "step": 3675
    },
    {
      "epoch": 2.6222537207654146,
      "grad_norm": 0.6382436156272888,
      "learning_rate": 0.0001,
      "loss": 2.4907,
      "step": 3700
    },
    {
      "epoch": 2.6222537207654146,
      "eval_loss": 2.417147397994995,
      "eval_runtime": 32.6259,
      "eval_samples_per_second": 9.134,
      "eval_steps_per_second": 2.299,
      "step": 3700
    },
    {
      "epoch": 2.639971651311127,
      "grad_norm": 0.578823983669281,
      "learning_rate": 0.0001,
      "loss": 2.4281,
      "step": 3725
    },
    {
      "epoch": 2.657689581856839,
      "grad_norm": 0.5311617255210876,
      "learning_rate": 0.0001,
      "loss": 2.4315,
      "step": 3750
    },
    {
      "epoch": 2.6754075124025514,
      "grad_norm": 0.4713846743106842,
      "learning_rate": 0.0001,
      "loss": 2.4268,
      "step": 3775
    },
    {
      "epoch": 2.6931254429482636,
      "grad_norm": 0.7472273111343384,
      "learning_rate": 0.0001,
      "loss": 2.3967,
      "step": 3800
    },
    {
      "epoch": 2.6931254429482636,
      "eval_loss": 2.415893077850342,
      "eval_runtime": 32.6303,
      "eval_samples_per_second": 9.133,
      "eval_steps_per_second": 2.298,
      "step": 3800
    },
    {
      "epoch": 2.710843373493976,
      "grad_norm": 0.5875506401062012,
      "learning_rate": 0.0001,
      "loss": 2.3541,
      "step": 3825
    },
    {
      "epoch": 2.728561304039688,
      "grad_norm": 0.38152602314949036,
      "learning_rate": 0.0001,
      "loss": 2.4623,
      "step": 3850
    },
    {
      "epoch": 2.7462792345854004,
      "grad_norm": 0.35034802556037903,
      "learning_rate": 0.0001,
      "loss": 2.4392,
      "step": 3875
    },
    {
      "epoch": 2.7639971651311126,
      "grad_norm": 0.3683781027793884,
      "learning_rate": 0.0001,
      "loss": 2.4772,
      "step": 3900
    },
    {
      "epoch": 2.7639971651311126,
      "eval_loss": 2.414635181427002,
      "eval_runtime": 32.6546,
      "eval_samples_per_second": 9.126,
      "eval_steps_per_second": 2.297,
      "step": 3900
    },
    {
      "epoch": 2.781715095676825,
      "grad_norm": 0.632203221321106,
      "learning_rate": 0.0001,
      "loss": 2.48,
      "step": 3925
    },
    {
      "epoch": 2.799433026222537,
      "grad_norm": 0.4688514173030853,
      "learning_rate": 0.0001,
      "loss": 2.4058,
      "step": 3950
    },
    {
      "epoch": 2.8171509567682493,
      "grad_norm": 0.3703823685646057,
      "learning_rate": 0.0001,
      "loss": 2.3802,
      "step": 3975
    },
    {
      "epoch": 2.8348688873139616,
      "grad_norm": 0.4395906329154968,
      "learning_rate": 0.0001,
      "loss": 2.4114,
      "step": 4000
    },
    {
      "epoch": 2.8348688873139616,
      "eval_loss": 2.4105727672576904,
      "eval_runtime": 32.6241,
      "eval_samples_per_second": 9.134,
      "eval_steps_per_second": 2.299,
      "step": 4000
    },
    {
      "epoch": 2.852586817859674,
      "grad_norm": 0.3975163996219635,
      "learning_rate": 0.0001,
      "loss": 2.4189,
      "step": 4025
    },
    {
      "epoch": 2.8703047484053865,
      "grad_norm": 0.37457334995269775,
      "learning_rate": 0.0001,
      "loss": 2.3946,
      "step": 4050
    },
    {
      "epoch": 2.8880226789510983,
      "grad_norm": 0.3786819279193878,
      "learning_rate": 0.0001,
      "loss": 2.4683,
      "step": 4075
    },
    {
      "epoch": 2.905740609496811,
      "grad_norm": 0.633921205997467,
      "learning_rate": 0.0001,
      "loss": 2.4017,
      "step": 4100
    },
    {
      "epoch": 2.905740609496811,
      "eval_loss": 2.406451463699341,
      "eval_runtime": 32.6727,
      "eval_samples_per_second": 9.121,
      "eval_steps_per_second": 2.295,
      "step": 4100
    },
    {
      "epoch": 2.923458540042523,
      "grad_norm": 0.9005404710769653,
      "learning_rate": 0.0001,
      "loss": 2.4099,
      "step": 4125
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 0.5802463293075562,
      "learning_rate": 0.0001,
      "loss": 2.4068,
      "step": 4150
    },
    {
      "epoch": 2.9588944011339473,
      "grad_norm": 0.3155713975429535,
      "learning_rate": 0.0001,
      "loss": 2.3682,
      "step": 4175
    },
    {
      "epoch": 2.97661233167966,
      "grad_norm": 0.4876560568809509,
      "learning_rate": 0.0001,
      "loss": 2.3477,
      "step": 4200
    },
    {
      "epoch": 2.97661233167966,
      "eval_loss": 2.405850648880005,
      "eval_runtime": 32.6084,
      "eval_samples_per_second": 9.139,
      "eval_steps_per_second": 2.3,
      "step": 4200
    },
    {
      "epoch": 2.9943302622253722,
      "grad_norm": 0.49624720215797424,
      "learning_rate": 0.0001,
      "loss": 2.445,
      "step": 4225
    },
    {
      "epoch": 3.0,
      "step": 4233,
      "total_flos": 9152314711474176.0,
      "train_loss": 2.5398848939386913,
      "train_runtime": 5263.4986,
      "train_samples_per_second": 3.217,
      "train_steps_per_second": 0.804
    }
  ],
  "logging_steps": 25,
  "max_steps": 4233,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9152314711474176.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}