{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9972932091393998,
  "eval_steps": 100,
  "global_step": 784,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.012737839344001274,
      "grad_norm": 7.091875076293945,
      "learning_rate": 3.75e-05,
      "loss": 37.744,
      "step": 5
    },
    {
      "epoch": 0.02547567868800255,
      "grad_norm": 2.930401563644409,
      "learning_rate": 7.5e-05,
      "loss": 34.0864,
      "step": 10
    },
    {
      "epoch": 0.03821351803200382,
      "grad_norm": 1.8477588891983032,
      "learning_rate": 0.0001125,
      "loss": 31.2726,
      "step": 15
    },
    {
      "epoch": 0.0509513573760051,
      "grad_norm": 1.3455390930175781,
      "learning_rate": 0.00015,
      "loss": 28.3763,
      "step": 20
    },
    {
      "epoch": 0.06368919672000636,
      "grad_norm": 1.138717532157898,
      "learning_rate": 0.00018749999999999998,
      "loss": 26.957,
      "step": 25
    },
    {
      "epoch": 0.07642703606400764,
      "grad_norm": 0.9747544527053833,
      "learning_rate": 0.000225,
      "loss": 24.4616,
      "step": 30
    },
    {
      "epoch": 0.08916487540800892,
      "grad_norm": 0.9035225510597229,
      "learning_rate": 0.0002625,
      "loss": 22.5748,
      "step": 35
    },
    {
      "epoch": 0.1019027147520102,
      "grad_norm": 0.7786006927490234,
      "learning_rate": 0.0003,
      "loss": 20.6574,
      "step": 40
    },
    {
      "epoch": 0.11464055409601147,
      "grad_norm": 0.7649045586585999,
      "learning_rate": 0.0003,
      "loss": 18.9346,
      "step": 45
    },
    {
      "epoch": 0.12737839344001273,
      "grad_norm": 0.6415356993675232,
      "learning_rate": 0.0003,
      "loss": 17.8129,
      "step": 50
    },
    {
      "epoch": 0.140116232784014,
      "grad_norm": 0.5701594948768616,
      "learning_rate": 0.0003,
      "loss": 16.881,
      "step": 55
    },
    {
      "epoch": 0.15285407212801527,
      "grad_norm": 0.49638187885284424,
      "learning_rate": 0.0003,
      "loss": 16.2049,
      "step": 60
    },
    {
      "epoch": 0.16559191147201657,
      "grad_norm": 0.44346606731414795,
      "learning_rate": 0.0003,
      "loss": 15.9336,
      "step": 65
    },
    {
      "epoch": 0.17832975081601785,
      "grad_norm": 0.4194740355014801,
      "learning_rate": 0.0003,
      "loss": 15.2473,
      "step": 70
    },
    {
      "epoch": 0.19106759016001912,
      "grad_norm": 0.4130041301250458,
      "learning_rate": 0.0003,
      "loss": 15.1218,
      "step": 75
    },
    {
      "epoch": 0.2038054295040204,
      "grad_norm": 0.40480196475982666,
      "learning_rate": 0.0003,
      "loss": 14.7839,
      "step": 80
    },
    {
      "epoch": 0.21654326884802166,
      "grad_norm": 0.394378662109375,
      "learning_rate": 0.0003,
      "loss": 14.2312,
      "step": 85
    },
    {
      "epoch": 0.22928110819202294,
      "grad_norm": 0.39825204014778137,
      "learning_rate": 0.0003,
      "loss": 13.9441,
      "step": 90
    },
    {
      "epoch": 0.2420189475360242,
      "grad_norm": 0.38816991448402405,
      "learning_rate": 0.0003,
      "loss": 13.4799,
      "step": 95
    },
    {
      "epoch": 0.25475678688002545,
      "grad_norm": 0.36586159467697144,
      "learning_rate": 0.0003,
      "loss": 13.3276,
      "step": 100
    },
    {
      "epoch": 0.25475678688002545,
      "eval_accuracy": 0.013133919843597262,
      "eval_loss": 12.040165901184082,
      "eval_runtime": 14.4617,
      "eval_samples_per_second": 17.287,
      "eval_steps_per_second": 4.356,
      "step": 100
    },
    {
      "epoch": 0.26749462622402675,
      "grad_norm": 0.40571218729019165,
      "learning_rate": 0.0003,
      "loss": 13.1015,
      "step": 105
    },
    {
      "epoch": 0.280232465568028,
      "grad_norm": 0.3502795696258545,
      "learning_rate": 0.0003,
      "loss": 12.614,
      "step": 110
    },
    {
      "epoch": 0.2929703049120293,
      "grad_norm": 0.33776018023490906,
      "learning_rate": 0.0003,
      "loss": 12.488,
      "step": 115
    },
    {
      "epoch": 0.30570814425603055,
      "grad_norm": 0.3277961015701294,
      "learning_rate": 0.0003,
      "loss": 12.2282,
      "step": 120
    },
    {
      "epoch": 0.31844598360003185,
      "grad_norm": 0.3399854898452759,
      "learning_rate": 0.0003,
      "loss": 12.0168,
      "step": 125
    },
    {
      "epoch": 0.33118382294403315,
      "grad_norm": 0.31557145714759827,
      "learning_rate": 0.0003,
      "loss": 11.832,
      "step": 130
    },
    {
      "epoch": 0.3439216622880344,
      "grad_norm": 0.32902857661247253,
      "learning_rate": 0.0003,
      "loss": 11.4818,
      "step": 135
    },
    {
      "epoch": 0.3566595016320357,
      "grad_norm": 0.34518980979919434,
      "learning_rate": 0.0003,
      "loss": 11.3197,
      "step": 140
    },
    {
      "epoch": 0.36939734097603694,
      "grad_norm": 0.32530176639556885,
      "learning_rate": 0.0003,
      "loss": 11.0346,
      "step": 145
    },
    {
      "epoch": 0.38213518032003824,
      "grad_norm": 0.3253624141216278,
      "learning_rate": 0.0003,
      "loss": 10.6717,
      "step": 150
    },
    {
      "epoch": 0.3948730196640395,
      "grad_norm": 0.33527347445487976,
      "learning_rate": 0.0003,
      "loss": 10.5302,
      "step": 155
    },
    {
      "epoch": 0.4076108590080408,
      "grad_norm": 0.3164774477481842,
      "learning_rate": 0.0003,
      "loss": 10.2009,
      "step": 160
    },
    {
      "epoch": 0.420348698352042,
      "grad_norm": 0.3047502934932709,
      "learning_rate": 0.0003,
      "loss": 10.1689,
      "step": 165
    },
    {
      "epoch": 0.4330865376960433,
      "grad_norm": 0.31613191962242126,
      "learning_rate": 0.0003,
      "loss": 9.85,
      "step": 170
    },
    {
      "epoch": 0.4458243770400446,
      "grad_norm": 0.3114412724971771,
      "learning_rate": 0.0003,
      "loss": 9.6662,
      "step": 175
    },
    {
      "epoch": 0.4585622163840459,
      "grad_norm": 0.31863468885421753,
      "learning_rate": 0.0003,
      "loss": 9.4857,
      "step": 180
    },
    {
      "epoch": 0.4713000557280471,
      "grad_norm": 0.3024883568286896,
      "learning_rate": 0.0003,
      "loss": 9.2409,
      "step": 185
    },
    {
      "epoch": 0.4840378950720484,
      "grad_norm": 0.3118532598018646,
      "learning_rate": 0.0003,
      "loss": 9.156,
      "step": 190
    },
    {
      "epoch": 0.49677573441604966,
      "grad_norm": 0.3026701807975769,
      "learning_rate": 0.0003,
      "loss": 9.0273,
      "step": 195
    },
    {
      "epoch": 0.5095135737600509,
      "grad_norm": 0.3058376908302307,
      "learning_rate": 0.0003,
      "loss": 8.9207,
      "step": 200
    },
    {
      "epoch": 0.5095135737600509,
      "eval_accuracy": 0.03601173020527859,
      "eval_loss": 8.031224250793457,
      "eval_runtime": 14.6886,
      "eval_samples_per_second": 17.02,
      "eval_steps_per_second": 4.289,
      "step": 200
    },
    {
      "epoch": 0.5222514131040522,
      "grad_norm": 0.31776145100593567,
      "learning_rate": 0.0003,
      "loss": 8.819,
      "step": 205
    },
    {
      "epoch": 0.5349892524480535,
      "grad_norm": 0.3050650656223297,
      "learning_rate": 0.0003,
      "loss": 8.7563,
      "step": 210
    },
    {
      "epoch": 0.5477270917920548,
      "grad_norm": 0.31346216797828674,
      "learning_rate": 0.0003,
      "loss": 8.4781,
      "step": 215
    },
    {
      "epoch": 0.560464931136056,
      "grad_norm": 0.3162192404270172,
      "learning_rate": 0.0003,
      "loss": 8.49,
      "step": 220
    },
    {
      "epoch": 0.5732027704800573,
      "grad_norm": 0.2908290922641754,
      "learning_rate": 0.0003,
      "loss": 8.1487,
      "step": 225
    },
    {
      "epoch": 0.5859406098240586,
      "grad_norm": 0.29553738236427307,
      "learning_rate": 0.0003,
      "loss": 8.2668,
      "step": 230
    },
    {
      "epoch": 0.5986784491680599,
      "grad_norm": 0.288335919380188,
      "learning_rate": 0.0003,
      "loss": 8.1061,
      "step": 235
    },
    {
      "epoch": 0.6114162885120611,
      "grad_norm": 0.30966615676879883,
      "learning_rate": 0.0003,
      "loss": 8.1297,
      "step": 240
    },
    {
      "epoch": 0.6241541278560624,
      "grad_norm": 0.29941117763519287,
      "learning_rate": 0.0003,
      "loss": 7.8082,
      "step": 245
    },
    {
      "epoch": 0.6368919672000637,
      "grad_norm": 0.29136765003204346,
      "learning_rate": 0.0003,
      "loss": 7.937,
      "step": 250
    },
    {
      "epoch": 0.649629806544065,
      "grad_norm": 0.30150941014289856,
      "learning_rate": 0.0003,
      "loss": 7.7454,
      "step": 255
    },
    {
      "epoch": 0.6623676458880663,
      "grad_norm": 0.28709036111831665,
      "learning_rate": 0.0003,
      "loss": 7.8069,
      "step": 260
    },
    {
      "epoch": 0.6751054852320675,
      "grad_norm": 0.31939393281936646,
      "learning_rate": 0.0003,
      "loss": 7.631,
      "step": 265
    },
    {
      "epoch": 0.6878433245760688,
      "grad_norm": 0.29692211747169495,
      "learning_rate": 0.0003,
      "loss": 7.6632,
      "step": 270
    },
    {
      "epoch": 0.7005811639200701,
      "grad_norm": 0.3304164409637451,
      "learning_rate": 0.0003,
      "loss": 7.4727,
      "step": 275
    },
    {
      "epoch": 0.7133190032640714,
      "grad_norm": 0.28332462906837463,
      "learning_rate": 0.0003,
      "loss": 7.4796,
      "step": 280
    },
    {
      "epoch": 0.7260568426080726,
      "grad_norm": 0.2897827625274658,
      "learning_rate": 0.0003,
      "loss": 7.5389,
      "step": 285
    },
    {
      "epoch": 0.7387946819520739,
      "grad_norm": 0.2887686491012573,
      "learning_rate": 0.0003,
      "loss": 7.382,
      "step": 290
    },
    {
      "epoch": 0.7515325212960752,
      "grad_norm": 0.3093564212322235,
      "learning_rate": 0.0003,
      "loss": 7.2586,
      "step": 295
    },
    {
      "epoch": 0.7642703606400765,
      "grad_norm": 0.2902717590332031,
      "learning_rate": 0.0003,
      "loss": 7.2681,
      "step": 300
    },
    {
      "epoch": 0.7642703606400765,
      "eval_accuracy": 0.050643206256109484,
      "eval_loss": 6.477533340454102,
      "eval_runtime": 14.6327,
      "eval_samples_per_second": 17.085,
      "eval_steps_per_second": 4.305,
      "step": 300
    },
    {
      "epoch": 0.7770081999840777,
      "grad_norm": 0.2867899239063263,
      "learning_rate": 0.0003,
      "loss": 7.0712,
      "step": 305
    },
    {
      "epoch": 0.789746039328079,
      "grad_norm": 0.27321040630340576,
      "learning_rate": 0.0003,
      "loss": 7.0524,
      "step": 310
    },
    {
      "epoch": 0.8024838786720803,
      "grad_norm": 0.3487064242362976,
      "learning_rate": 0.0003,
      "loss": 7.0939,
      "step": 315
    },
    {
      "epoch": 0.8152217180160816,
      "grad_norm": 0.329608291387558,
      "learning_rate": 0.0003,
      "loss": 6.9997,
      "step": 320
    },
    {
      "epoch": 0.8279595573600828,
      "grad_norm": 0.3154338300228119,
      "learning_rate": 0.0003,
      "loss": 6.9663,
      "step": 325
    },
    {
      "epoch": 0.840697396704084,
      "grad_norm": 0.31021803617477417,
      "learning_rate": 0.0003,
      "loss": 6.7821,
      "step": 330
    },
    {
      "epoch": 0.8534352360480854,
      "grad_norm": 0.388336181640625,
      "learning_rate": 0.0003,
      "loss": 6.7751,
      "step": 335
    },
    {
      "epoch": 0.8661730753920867,
      "grad_norm": 0.31887954473495483,
      "learning_rate": 0.0003,
      "loss": 6.702,
      "step": 340
    },
    {
      "epoch": 0.8789109147360878,
      "grad_norm": 0.31558957695961,
      "learning_rate": 0.0003,
      "loss": 6.6206,
      "step": 345
    },
    {
      "epoch": 0.8916487540800891,
      "grad_norm": 0.30751529335975647,
      "learning_rate": 0.0003,
      "loss": 6.7077,
      "step": 350
    },
    {
      "epoch": 0.9043865934240904,
      "grad_norm": 0.33058232069015503,
      "learning_rate": 0.0003,
      "loss": 6.557,
      "step": 355
    },
    {
      "epoch": 0.9171244327680917,
      "grad_norm": 0.3375111222267151,
      "learning_rate": 0.0003,
      "loss": 6.6369,
      "step": 360
    },
    {
      "epoch": 0.9298622721120929,
      "grad_norm": 0.3047392964363098,
      "learning_rate": 0.0003,
      "loss": 6.5796,
      "step": 365
    },
    {
      "epoch": 0.9426001114560942,
      "grad_norm": 0.430053174495697,
      "learning_rate": 0.0003,
      "loss": 6.5548,
      "step": 370
    },
    {
      "epoch": 0.9553379508000955,
      "grad_norm": 0.3610515296459198,
      "learning_rate": 0.0003,
      "loss": 6.4576,
      "step": 375
    },
    {
      "epoch": 0.9680757901440968,
      "grad_norm": 0.32095110416412354,
      "learning_rate": 0.0003,
      "loss": 6.4266,
      "step": 380
    },
    {
      "epoch": 0.980813629488098,
      "grad_norm": 0.32170969247817993,
      "learning_rate": 0.0003,
      "loss": 6.5597,
      "step": 385
    },
    {
      "epoch": 0.9935514688320993,
      "grad_norm": 0.29942792654037476,
      "learning_rate": 0.0003,
      "loss": 6.3873,
      "step": 390
    },
    {
      "epoch": 1.0062893081761006,
      "grad_norm": 0.2971299886703491,
      "learning_rate": 0.0003,
      "loss": 6.3915,
      "step": 395
    },
    {
      "epoch": 1.0190271475201018,
      "grad_norm": 0.2800815999507904,
      "learning_rate": 0.0003,
      "loss": 6.3187,
      "step": 400
    },
    {
      "epoch": 1.0190271475201018,
      "eval_accuracy": 0.0433822091886608,
      "eval_loss": 5.622740268707275,
      "eval_runtime": 14.4103,
      "eval_samples_per_second": 17.349,
      "eval_steps_per_second": 4.372,
      "step": 400
    },
    {
      "epoch": 1.0317649868641032,
      "grad_norm": 0.28819501399993896,
      "learning_rate": 0.0003,
      "loss": 6.328,
      "step": 405
    },
    {
      "epoch": 1.0445028262081044,
      "grad_norm": 0.3983236849308014,
      "learning_rate": 0.0003,
      "loss": 6.3988,
      "step": 410
    },
    {
      "epoch": 1.0572406655521058,
      "grad_norm": 0.2969406545162201,
      "learning_rate": 0.0003,
      "loss": 6.2509,
      "step": 415
    },
    {
      "epoch": 1.069978504896107,
      "grad_norm": 0.2973212003707886,
      "learning_rate": 0.0003,
      "loss": 6.1234,
      "step": 420
    },
    {
      "epoch": 1.0827163442401082,
      "grad_norm": 0.3298945426940918,
      "learning_rate": 0.0003,
      "loss": 6.3219,
      "step": 425
    },
    {
      "epoch": 1.0954541835841096,
      "grad_norm": 0.3493943214416504,
      "learning_rate": 0.0003,
      "loss": 6.0888,
      "step": 430
    },
    {
      "epoch": 1.1081920229281108,
      "grad_norm": 0.3639209270477295,
      "learning_rate": 0.0003,
      "loss": 6.2226,
      "step": 435
    },
    {
      "epoch": 1.120929862272112,
      "grad_norm": 0.43913957476615906,
      "learning_rate": 0.0003,
      "loss": 6.0308,
      "step": 440
    },
    {
      "epoch": 1.1336677016161134,
      "grad_norm": 0.43267834186553955,
      "learning_rate": 0.0003,
      "loss": 6.0806,
      "step": 445
    },
    {
      "epoch": 1.1464055409601146,
      "grad_norm": 0.4563148021697998,
      "learning_rate": 0.0003,
      "loss": 5.9703,
      "step": 450
    },
    {
      "epoch": 1.159143380304116,
      "grad_norm": 0.4002761244773865,
      "learning_rate": 0.0003,
      "loss": 5.9163,
      "step": 455
    },
    {
      "epoch": 1.1718812196481172,
      "grad_norm": 0.4359826147556305,
      "learning_rate": 0.0003,
      "loss": 5.8285,
      "step": 460
    },
    {
      "epoch": 1.1846190589921184,
      "grad_norm": 0.5450247526168823,
      "learning_rate": 0.0003,
      "loss": 5.8063,
      "step": 465
    },
    {
      "epoch": 1.1973568983361198,
      "grad_norm": 0.3597274422645569,
      "learning_rate": 0.0003,
      "loss": 5.6978,
      "step": 470
    },
    {
      "epoch": 1.210094737680121,
      "grad_norm": 0.4141215980052948,
      "learning_rate": 0.0003,
      "loss": 5.6078,
      "step": 475
    },
    {
      "epoch": 1.2228325770241222,
      "grad_norm": 0.3695543110370636,
      "learning_rate": 0.0003,
      "loss": 5.6728,
      "step": 480
    },
    {
      "epoch": 1.2355704163681236,
      "grad_norm": 0.5060051083564758,
      "learning_rate": 0.0003,
      "loss": 5.6049,
      "step": 485
    },
    {
      "epoch": 1.2483082557121248,
      "grad_norm": 0.5355808138847351,
      "learning_rate": 0.0003,
      "loss": 5.6564,
      "step": 490
    },
    {
      "epoch": 1.261046095056126,
      "grad_norm": 0.4578459858894348,
      "learning_rate": 0.0003,
      "loss": 5.5758,
      "step": 495
    },
    {
      "epoch": 1.2737839344001274,
      "grad_norm": 0.4868403673171997,
      "learning_rate": 0.0003,
      "loss": 5.5695,
      "step": 500
    },
    {
      "epoch": 1.2737839344001274,
      "eval_accuracy": 0.36348778103616813,
      "eval_loss": 4.77961540222168,
      "eval_runtime": 14.5581,
      "eval_samples_per_second": 17.173,
      "eval_steps_per_second": 4.328,
      "step": 500
    },
    {
      "epoch": 1.2865217737441286,
      "grad_norm": 0.550255298614502,
      "learning_rate": 0.0003,
      "loss": 5.5591,
      "step": 505
    },
    {
      "epoch": 1.29925961308813,
      "grad_norm": 0.5515110492706299,
      "learning_rate": 0.0003,
      "loss": 5.4588,
      "step": 510
    },
    {
      "epoch": 1.3119974524321312,
      "grad_norm": 0.44656914472579956,
      "learning_rate": 0.0003,
      "loss": 5.4336,
      "step": 515
    },
    {
      "epoch": 1.3247352917761326,
      "grad_norm": 0.5925999283790588,
      "learning_rate": 0.0003,
      "loss": 5.5185,
      "step": 520
    },
    {
      "epoch": 1.3374731311201338,
      "grad_norm": 0.632453203201294,
      "learning_rate": 0.0003,
      "loss": 5.325,
      "step": 525
    },
    {
      "epoch": 1.350210970464135,
      "grad_norm": 0.5380024909973145,
      "learning_rate": 0.0003,
      "loss": 5.4005,
      "step": 530
    },
    {
      "epoch": 1.3629488098081364,
      "grad_norm": 0.5659191012382507,
      "learning_rate": 0.0003,
      "loss": 5.3564,
      "step": 535
    },
    {
      "epoch": 1.3756866491521376,
      "grad_norm": 0.8913821578025818,
      "learning_rate": 0.0003,
      "loss": 5.2763,
      "step": 540
    },
    {
      "epoch": 1.3884244884961388,
      "grad_norm": 0.9271002411842346,
      "learning_rate": 0.0003,
      "loss": 5.4129,
      "step": 545
    },
    {
      "epoch": 1.4011623278401402,
      "grad_norm": 0.7141408324241638,
      "learning_rate": 0.0003,
      "loss": 5.4437,
      "step": 550
    },
    {
      "epoch": 1.4139001671841414,
      "grad_norm": 0.5360827445983887,
      "learning_rate": 0.0003,
      "loss": 5.3523,
      "step": 555
    },
    {
      "epoch": 1.4266380065281425,
      "grad_norm": 0.6563194990158081,
      "learning_rate": 0.0003,
      "loss": 5.1103,
      "step": 560
    },
    {
      "epoch": 1.439375845872144,
      "grad_norm": 0.6325790882110596,
      "learning_rate": 0.0003,
      "loss": 5.4026,
      "step": 565
    },
    {
      "epoch": 1.4521136852161451,
      "grad_norm": 0.8463213443756104,
      "learning_rate": 0.0003,
      "loss": 5.3129,
      "step": 570
    },
    {
      "epoch": 1.4648515245601466,
      "grad_norm": 0.8394812345504761,
      "learning_rate": 0.0003,
      "loss": 5.3415,
      "step": 575
    },
    {
      "epoch": 1.4775893639041477,
      "grad_norm": 0.692244291305542,
      "learning_rate": 0.0003,
      "loss": 5.2649,
      "step": 580
    },
    {
      "epoch": 1.4903272032481492,
      "grad_norm": 0.6197806000709534,
      "learning_rate": 0.0003,
      "loss": 5.112,
      "step": 585
    },
    {
      "epoch": 1.5030650425921503,
      "grad_norm": 0.6573797464370728,
      "learning_rate": 0.0003,
      "loss": 5.1669,
      "step": 590
    },
    {
      "epoch": 1.5158028819361515,
      "grad_norm": 0.795892059803009,
      "learning_rate": 0.0003,
      "loss": 5.1693,
      "step": 595
    },
    {
      "epoch": 1.528540721280153,
      "grad_norm": 0.6279253363609314,
      "learning_rate": 0.0003,
      "loss": 5.2926,
      "step": 600
    },
    {
      "epoch": 1.528540721280153,
      "eval_accuracy": 0.3952492668621701,
      "eval_loss": 4.392324447631836,
      "eval_runtime": 14.409,
      "eval_samples_per_second": 17.35,
      "eval_steps_per_second": 4.372,
      "step": 600
    },
    {
      "epoch": 1.5412785606241541,
      "grad_norm": 0.5762287378311157,
      "learning_rate": 0.0003,
      "loss": 5.0475,
      "step": 605
    },
    {
      "epoch": 1.5540163999681553,
      "grad_norm": 0.5149503350257874,
      "learning_rate": 0.0003,
      "loss": 5.1185,
      "step": 610
    },
    {
      "epoch": 1.5667542393121567,
      "grad_norm": 0.581633985042572,
      "learning_rate": 0.0003,
      "loss": 5.1166,
      "step": 615
    },
    {
      "epoch": 1.579492078656158,
      "grad_norm": 0.5910624861717224,
      "learning_rate": 0.0003,
      "loss": 4.9907,
      "step": 620
    },
    {
      "epoch": 1.5922299180001591,
      "grad_norm": 0.8280585408210754,
      "learning_rate": 0.0003,
      "loss": 5.0748,
      "step": 625
    },
    {
      "epoch": 1.6049677573441605,
      "grad_norm": 0.5128599405288696,
      "learning_rate": 0.0003,
      "loss": 4.9768,
      "step": 630
    },
    {
      "epoch": 1.6177055966881617,
      "grad_norm": 0.7540919184684753,
      "learning_rate": 0.0003,
      "loss": 5.0806,
      "step": 635
    },
    {
      "epoch": 1.630443436032163,
      "grad_norm": 0.6239334940910339,
      "learning_rate": 0.0003,
      "loss": 5.1277,
      "step": 640
    },
    {
      "epoch": 1.6431812753761643,
      "grad_norm": 0.7787991166114807,
      "learning_rate": 0.0003,
      "loss": 5.0778,
      "step": 645
    },
    {
      "epoch": 1.6559191147201657,
      "grad_norm": 0.6328299641609192,
      "learning_rate": 0.0003,
      "loss": 4.9763,
      "step": 650
    },
    {
      "epoch": 1.668656954064167,
      "grad_norm": 0.5455794334411621,
      "learning_rate": 0.0003,
      "loss": 5.0049,
      "step": 655
    },
    {
      "epoch": 1.681394793408168,
      "grad_norm": 0.7078703045845032,
      "learning_rate": 0.0003,
      "loss": 5.0258,
      "step": 660
    },
    {
      "epoch": 1.6941326327521695,
      "grad_norm": 0.6339858770370483,
      "learning_rate": 0.0003,
      "loss": 5.1028,
      "step": 665
    },
    {
      "epoch": 1.7068704720961707,
      "grad_norm": 0.6060242652893066,
      "learning_rate": 0.0003,
      "loss": 5.0428,
      "step": 670
    },
    {
      "epoch": 1.719608311440172,
      "grad_norm": 0.9218889474868774,
      "learning_rate": 0.0003,
      "loss": 4.9891,
      "step": 675
    },
    {
      "epoch": 1.7323461507841733,
      "grad_norm": 0.6890697479248047,
      "learning_rate": 0.0003,
      "loss": 4.8921,
      "step": 680
    },
    {
      "epoch": 1.7450839901281745,
      "grad_norm": 0.9093934297561646,
      "learning_rate": 0.0003,
      "loss": 4.9385,
      "step": 685
    },
    {
      "epoch": 1.7578218294721757,
      "grad_norm": 0.5929202437400818,
      "learning_rate": 0.0003,
      "loss": 4.9376,
      "step": 690
    },
    {
      "epoch": 1.770559668816177,
      "grad_norm": 0.6317362189292908,
      "learning_rate": 0.0003,
      "loss": 4.9681,
      "step": 695
    },
    {
      "epoch": 1.7832975081601783,
      "grad_norm": 0.5537763237953186,
      "learning_rate": 0.0003,
      "loss": 4.878,
      "step": 700
    },
    {
      "epoch": 1.7832975081601783,
      "eval_accuracy": 0.40849266862170086,
      "eval_loss": 4.187656402587891,
      "eval_runtime": 14.4745,
      "eval_samples_per_second": 17.272,
      "eval_steps_per_second": 4.352,
      "step": 700
    },
    {
      "epoch": 1.7960353475041795,
      "grad_norm": 0.5984592437744141,
      "learning_rate": 0.0003,
      "loss": 4.9092,
      "step": 705
    },
    {
      "epoch": 1.808773186848181,
      "grad_norm": 0.5060558915138245,
      "learning_rate": 0.0003,
      "loss": 4.989,
      "step": 710
    },
    {
      "epoch": 1.8215110261921823,
      "grad_norm": 0.8713288903236389,
      "learning_rate": 0.0003,
      "loss": 4.8114,
      "step": 715
    },
    {
      "epoch": 1.8342488655361833,
      "grad_norm": 0.8011664748191833,
      "learning_rate": 0.0003,
      "loss": 4.8468,
      "step": 720
    },
    {
      "epoch": 1.8469867048801847,
      "grad_norm": 0.6774628758430481,
      "learning_rate": 0.0003,
      "loss": 4.8899,
      "step": 725
    },
    {
      "epoch": 1.859724544224186,
      "grad_norm": 1.05668044090271,
      "learning_rate": 0.0003,
      "loss": 4.8676,
      "step": 730
    },
    {
      "epoch": 1.8724623835681873,
      "grad_norm": 0.8638430237770081,
      "learning_rate": 0.0003,
      "loss": 4.8515,
      "step": 735
    },
    {
      "epoch": 1.8852002229121885,
      "grad_norm": 0.8210180997848511,
      "learning_rate": 0.0003,
      "loss": 4.9094,
      "step": 740
    },
    {
      "epoch": 1.8979380622561899,
      "grad_norm": 0.6894564032554626,
      "learning_rate": 0.0003,
      "loss": 4.8406,
      "step": 745
    },
    {
      "epoch": 1.910675901600191,
      "grad_norm": 0.7244303822517395,
      "learning_rate": 0.0003,
      "loss": 4.8299,
      "step": 750
    },
    {
      "epoch": 1.9234137409441923,
      "grad_norm": 0.5788025856018066,
      "learning_rate": 0.0003,
      "loss": 4.8843,
      "step": 755
    },
    {
      "epoch": 1.9361515802881937,
      "grad_norm": 0.5082942843437195,
      "learning_rate": 0.0003,
      "loss": 4.7624,
      "step": 760
    },
    {
      "epoch": 1.9488894196321949,
      "grad_norm": 0.6290297508239746,
      "learning_rate": 0.0003,
      "loss": 4.7709,
      "step": 765
    },
    {
      "epoch": 1.961627258976196,
      "grad_norm": 0.5582670569419861,
      "learning_rate": 0.0003,
      "loss": 4.7169,
      "step": 770
    },
    {
      "epoch": 1.9743650983201975,
      "grad_norm": 0.6051950454711914,
      "learning_rate": 0.0003,
      "loss": 4.7701,
      "step": 775
    },
    {
      "epoch": 1.9871029376641989,
      "grad_norm": 0.6427810788154602,
      "learning_rate": 0.0003,
      "loss": 4.7729,
      "step": 780
    },
    {
      "epoch": 1.9972932091393998,
      "step": 784,
      "total_flos": 6.247688798679859e+16,
      "train_loss": 8.457253451250038,
      "train_runtime": 65851.1603,
      "train_samples_per_second": 1.526,
      "train_steps_per_second": 0.012
    }
  ],
  "logging_steps": 5,
  "max_steps": 784,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "total_flos": 6.247688798679859e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}