{
  "best_metric": 1.2004581689834595,
  "best_model_checkpoint": "distilbert_base_train_book_v2/checkpoint-350000",
  "epoch": 25.0,
  "eval_steps": 10000,
  "global_step": 355850,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0351271603203597,
      "grad_norm": 1.9653347730636597,
      "learning_rate": 5e-06,
      "loss": 8.8875,
      "step": 500
    },
    {
      "epoch": 0.0702543206407194,
      "grad_norm": 0.7303386926651001,
      "learning_rate": 1e-05,
      "loss": 6.5828,
      "step": 1000
    },
    {
      "epoch": 0.1053814809610791,
      "grad_norm": 0.6638388633728027,
      "learning_rate": 1.5e-05,
      "loss": 6.0119,
      "step": 1500
    },
    {
      "epoch": 0.1405086412814388,
      "grad_norm": 0.693476140499115,
      "learning_rate": 2e-05,
      "loss": 5.8991,
      "step": 2000
    },
    {
      "epoch": 0.1756358016017985,
      "grad_norm": 0.6988317370414734,
      "learning_rate": 2.5e-05,
      "loss": 5.8401,
      "step": 2500
    },
    {
      "epoch": 0.2107629619221582,
      "grad_norm": 0.5958765149116516,
      "learning_rate": 3e-05,
      "loss": 5.7955,
      "step": 3000
    },
    {
      "epoch": 0.2458901222425179,
      "grad_norm": 0.4958369731903076,
      "learning_rate": 3.5e-05,
      "loss": 5.7665,
      "step": 3500
    },
    {
      "epoch": 0.2810172825628776,
      "grad_norm": 0.4899258613586426,
      "learning_rate": 4e-05,
      "loss": 5.7418,
      "step": 4000
    },
    {
      "epoch": 0.31614444288323734,
      "grad_norm": 0.5407448410987854,
      "learning_rate": 4.5e-05,
      "loss": 5.7212,
      "step": 4500
    },
    {
      "epoch": 0.351271603203597,
      "grad_norm": 0.5828524231910706,
      "learning_rate": 5e-05,
      "loss": 5.7049,
      "step": 5000
    },
    {
      "epoch": 0.38639876352395675,
      "grad_norm": 0.5220353007316589,
      "learning_rate": 5.500000000000001e-05,
      "loss": 5.6856,
      "step": 5500
    },
    {
      "epoch": 0.4215259238443164,
      "grad_norm": 0.4556717574596405,
      "learning_rate": 6e-05,
      "loss": 5.6762,
      "step": 6000
    },
    {
      "epoch": 0.45665308416467615,
      "grad_norm": 0.49513307213783264,
      "learning_rate": 6.500000000000001e-05,
      "loss": 5.6635,
      "step": 6500
    },
    {
      "epoch": 0.4917802444850358,
      "grad_norm": 0.45069605112075806,
      "learning_rate": 7e-05,
      "loss": 5.6566,
      "step": 7000
    },
    {
      "epoch": 0.5269074048053956,
      "grad_norm": 0.5079524517059326,
      "learning_rate": 7.500000000000001e-05,
      "loss": 5.6468,
      "step": 7500
    },
    {
      "epoch": 0.5620345651257552,
      "grad_norm": 0.5765870809555054,
      "learning_rate": 8e-05,
      "loss": 5.6365,
      "step": 8000
    },
    {
      "epoch": 0.5971617254461149,
      "grad_norm": 0.47896265983581543,
      "learning_rate": 8.5e-05,
      "loss": 5.627,
      "step": 8500
    },
    {
      "epoch": 0.6322888857664747,
      "grad_norm": 0.6472557783126831,
      "learning_rate": 9e-05,
      "loss": 5.6161,
      "step": 9000
    },
    {
      "epoch": 0.6674160460868344,
      "grad_norm": 0.43458178639411926,
      "learning_rate": 9.5e-05,
      "loss": 5.6111,
      "step": 9500
    },
    {
      "epoch": 0.702543206407194,
      "grad_norm": 0.4662564694881439,
      "learning_rate": 0.0001,
      "loss": 5.604,
      "step": 10000
    },
    {
      "epoch": 0.702543206407194,
      "eval_accuracy": 0.16497299638082222,
      "eval_loss": 5.45039176940918,
      "eval_runtime": 143.2482,
      "eval_samples_per_second": 842.635,
      "eval_steps_per_second": 5.271,
      "step": 10000
    },
    {
      "epoch": 0.7376703667275537,
      "grad_norm": 0.5193693041801453,
      "learning_rate": 9.985542865404077e-05,
      "loss": 5.5963,
      "step": 10500
    },
    {
      "epoch": 0.7727975270479135,
      "grad_norm": 0.41712281107902527,
      "learning_rate": 9.971085730808155e-05,
      "loss": 5.5891,
      "step": 11000
    },
    {
      "epoch": 0.8079246873682732,
      "grad_norm": 0.4692609906196594,
      "learning_rate": 9.956628596212231e-05,
      "loss": 5.5854,
      "step": 11500
    },
    {
      "epoch": 0.8430518476886328,
      "grad_norm": 0.4751366376876831,
      "learning_rate": 9.942171461616309e-05,
      "loss": 5.5763,
      "step": 12000
    },
    {
      "epoch": 0.8781790080089925,
      "grad_norm": 0.4744866192340851,
      "learning_rate": 9.927714327020385e-05,
      "loss": 5.5709,
      "step": 12500
    },
    {
      "epoch": 0.9133061683293523,
      "grad_norm": 0.4810524582862854,
      "learning_rate": 9.913257192424462e-05,
      "loss": 5.5608,
      "step": 13000
    },
    {
      "epoch": 0.948433328649712,
      "grad_norm": 0.5423319935798645,
      "learning_rate": 9.89880005782854e-05,
      "loss": 5.5541,
      "step": 13500
    },
    {
      "epoch": 0.9835604889700716,
      "grad_norm": 0.4694032371044159,
      "learning_rate": 9.884342923232616e-05,
      "loss": 5.5498,
      "step": 14000
    },
    {
      "epoch": 1.0186876492904313,
      "grad_norm": 0.44750478863716125,
      "learning_rate": 9.869885788636692e-05,
      "loss": 5.5414,
      "step": 14500
    },
    {
      "epoch": 1.053814809610791,
      "grad_norm": 0.4432881772518158,
      "learning_rate": 9.855428654040769e-05,
      "loss": 5.5374,
      "step": 15000
    },
    {
      "epoch": 1.0889419699311507,
      "grad_norm": 0.4728132486343384,
      "learning_rate": 9.840971519444846e-05,
      "loss": 5.5297,
      "step": 15500
    },
    {
      "epoch": 1.1240691302515105,
      "grad_norm": 0.48674553632736206,
      "learning_rate": 9.826514384848924e-05,
      "loss": 5.5254,
      "step": 16000
    },
    {
      "epoch": 1.1591962905718702,
      "grad_norm": 0.5000836849212646,
      "learning_rate": 9.812057250253e-05,
      "loss": 5.52,
      "step": 16500
    },
    {
      "epoch": 1.1943234508922298,
      "grad_norm": 0.43097734451293945,
      "learning_rate": 9.797600115657077e-05,
      "loss": 5.5166,
      "step": 17000
    },
    {
      "epoch": 1.2294506112125896,
      "grad_norm": 0.4628123342990875,
      "learning_rate": 9.783142981061155e-05,
      "loss": 5.5085,
      "step": 17500
    },
    {
      "epoch": 1.2645777715329491,
      "grad_norm": 0.46364983916282654,
      "learning_rate": 9.768685846465231e-05,
      "loss": 5.507,
      "step": 18000
    },
    {
      "epoch": 1.299704931853309,
      "grad_norm": 0.6745333671569824,
      "learning_rate": 9.754228711869307e-05,
      "loss": 5.4862,
      "step": 18500
    },
    {
      "epoch": 1.3348320921736687,
      "grad_norm": 0.9819153547286987,
      "learning_rate": 9.739771577273384e-05,
      "loss": 5.3901,
      "step": 19000
    },
    {
      "epoch": 1.3699592524940285,
      "grad_norm": 2.0434741973876953,
      "learning_rate": 9.725314442677463e-05,
      "loss": 5.1978,
      "step": 19500
    },
    {
      "epoch": 1.405086412814388,
      "grad_norm": 2.197896957397461,
      "learning_rate": 9.71085730808154e-05,
      "loss": 4.6179,
      "step": 20000
    },
    {
      "epoch": 1.405086412814388,
      "eval_accuracy": 0.3758464833297413,
      "eval_loss": 3.827747344970703,
      "eval_runtime": 146.1003,
      "eval_samples_per_second": 826.186,
      "eval_steps_per_second": 5.168,
      "step": 20000
    },
    {
      "epoch": 1.4402135731347478,
      "grad_norm": 1.750232219696045,
      "learning_rate": 9.696400173485616e-05,
      "loss": 4.0652,
      "step": 20500
    },
    {
      "epoch": 1.4753407334551074,
      "grad_norm": 1.4147313833236694,
      "learning_rate": 9.681943038889692e-05,
      "loss": 3.6557,
      "step": 21000
    },
    {
      "epoch": 1.5104678937754672,
      "grad_norm": 1.1749171018600464,
      "learning_rate": 9.66748590429377e-05,
      "loss": 3.3421,
      "step": 21500
    },
    {
      "epoch": 1.545595054095827,
      "grad_norm": 1.1297008991241455,
      "learning_rate": 9.653028769697846e-05,
      "loss": 3.1555,
      "step": 22000
    },
    {
      "epoch": 1.5807222144161865,
      "grad_norm": 1.0993692874908447,
      "learning_rate": 9.638571635101923e-05,
      "loss": 3.0326,
      "step": 22500
    },
    {
      "epoch": 1.6158493747365463,
      "grad_norm": 1.1023796796798706,
      "learning_rate": 9.624114500506e-05,
      "loss": 2.9348,
      "step": 23000
    },
    {
      "epoch": 1.650976535056906,
      "grad_norm": 1.036067008972168,
      "learning_rate": 9.609657365910077e-05,
      "loss": 2.8555,
      "step": 23500
    },
    {
      "epoch": 1.6861036953772657,
      "grad_norm": 1.0167666673660278,
      "learning_rate": 9.595200231314155e-05,
      "loss": 2.7872,
      "step": 24000
    },
    {
      "epoch": 1.7212308556976255,
      "grad_norm": 1.0188016891479492,
      "learning_rate": 9.580743096718231e-05,
      "loss": 2.7329,
      "step": 24500
    },
    {
      "epoch": 1.7563580160179852,
      "grad_norm": 1.0286072492599487,
      "learning_rate": 9.566285962122307e-05,
      "loss": 2.6825,
      "step": 25000
    },
    {
      "epoch": 1.7914851763383448,
      "grad_norm": 0.9929981827735901,
      "learning_rate": 9.551828827526385e-05,
      "loss": 2.6381,
      "step": 25500
    },
    {
      "epoch": 1.8266123366587044,
      "grad_norm": 0.9288387894630432,
      "learning_rate": 9.537371692930462e-05,
      "loss": 2.5937,
      "step": 26000
    },
    {
      "epoch": 1.8617394969790642,
      "grad_norm": 0.9797139763832092,
      "learning_rate": 9.522914558334538e-05,
      "loss": 2.5524,
      "step": 26500
    },
    {
      "epoch": 1.896866657299424,
      "grad_norm": 0.9662619233131409,
      "learning_rate": 9.508457423738616e-05,
      "loss": 2.519,
      "step": 27000
    },
    {
      "epoch": 1.9319938176197837,
      "grad_norm": 0.9594979286193848,
      "learning_rate": 9.494000289142692e-05,
      "loss": 2.4854,
      "step": 27500
    },
    {
      "epoch": 1.9671209779401433,
      "grad_norm": 0.9515316486358643,
      "learning_rate": 9.47954315454677e-05,
      "loss": 2.4515,
      "step": 28000
    },
    {
      "epoch": 2.002248138260503,
      "grad_norm": 0.9741261601448059,
      "learning_rate": 9.465086019950846e-05,
      "loss": 2.4314,
      "step": 28500
    },
    {
      "epoch": 2.0373752985808626,
      "grad_norm": 0.9476631283760071,
      "learning_rate": 9.450628885354923e-05,
      "loss": 2.3975,
      "step": 29000
    },
    {
      "epoch": 2.0725024589012224,
      "grad_norm": 0.9616317749023438,
      "learning_rate": 9.436171750758999e-05,
      "loss": 2.3707,
      "step": 29500
    },
    {
      "epoch": 2.107629619221582,
      "grad_norm": 0.9059496521949768,
      "learning_rate": 9.421714616163077e-05,
      "loss": 2.353,
      "step": 30000
    },
    {
      "epoch": 2.107629619221582,
      "eval_accuracy": 0.5933188090938791,
      "eval_loss": 2.040786027908325,
      "eval_runtime": 146.0422,
      "eval_samples_per_second": 826.515,
      "eval_steps_per_second": 5.17,
      "step": 30000
    },
    {
      "epoch": 2.142756779541942,
      "grad_norm": 0.9514384865760803,
      "learning_rate": 9.407257481567155e-05,
      "loss": 2.3312,
      "step": 30500
    },
    {
      "epoch": 2.1778839398623013,
      "grad_norm": 0.9169853925704956,
      "learning_rate": 9.392800346971231e-05,
      "loss": 2.3091,
      "step": 31000
    },
    {
      "epoch": 2.213011100182661,
      "grad_norm": 0.9202592968940735,
      "learning_rate": 9.378343212375307e-05,
      "loss": 2.2937,
      "step": 31500
    },
    {
      "epoch": 2.248138260503021,
      "grad_norm": 0.9011738300323486,
      "learning_rate": 9.363886077779385e-05,
      "loss": 2.2759,
      "step": 32000
    },
    {
      "epoch": 2.2832654208233807,
      "grad_norm": 0.8867260217666626,
      "learning_rate": 9.349428943183462e-05,
      "loss": 2.2593,
      "step": 32500
    },
    {
      "epoch": 2.3183925811437405,
      "grad_norm": 0.9071447253227234,
      "learning_rate": 9.334971808587538e-05,
      "loss": 2.2402,
      "step": 33000
    },
    {
      "epoch": 2.3535197414641003,
      "grad_norm": 0.9013161063194275,
      "learning_rate": 9.320514673991614e-05,
      "loss": 2.2255,
      "step": 33500
    },
    {
      "epoch": 2.3886469017844596,
      "grad_norm": 0.8916395306587219,
      "learning_rate": 9.306057539395692e-05,
      "loss": 2.2123,
      "step": 34000
    },
    {
      "epoch": 2.4237740621048194,
      "grad_norm": 0.8660832047462463,
      "learning_rate": 9.29160040479977e-05,
      "loss": 2.1955,
      "step": 34500
    },
    {
      "epoch": 2.458901222425179,
      "grad_norm": 0.8876006603240967,
      "learning_rate": 9.277143270203846e-05,
      "loss": 2.1837,
      "step": 35000
    },
    {
      "epoch": 2.494028382745539,
      "grad_norm": 0.8954652547836304,
      "learning_rate": 9.262686135607923e-05,
      "loss": 2.1732,
      "step": 35500
    },
    {
      "epoch": 2.5291555430658983,
      "grad_norm": 0.9105496406555176,
      "learning_rate": 9.248229001012e-05,
      "loss": 2.1599,
      "step": 36000
    },
    {
      "epoch": 2.564282703386258,
      "grad_norm": 0.8876945972442627,
      "learning_rate": 9.233771866416077e-05,
      "loss": 2.1472,
      "step": 36500
    },
    {
      "epoch": 2.599409863706618,
      "grad_norm": 0.8663467168807983,
      "learning_rate": 9.219314731820153e-05,
      "loss": 2.1345,
      "step": 37000
    },
    {
      "epoch": 2.6345370240269776,
      "grad_norm": 0.9176456928253174,
      "learning_rate": 9.20485759722423e-05,
      "loss": 2.1258,
      "step": 37500
    },
    {
      "epoch": 2.6696641843473374,
      "grad_norm": 0.9134395718574524,
      "learning_rate": 9.190400462628307e-05,
      "loss": 2.1125,
      "step": 38000
    },
    {
      "epoch": 2.704791344667697,
      "grad_norm": 0.8617592453956604,
      "learning_rate": 9.175943328032385e-05,
      "loss": 2.1041,
      "step": 38500
    },
    {
      "epoch": 2.739918504988057,
      "grad_norm": 0.8791117668151855,
      "learning_rate": 9.161486193436461e-05,
      "loss": 2.0931,
      "step": 39000
    },
    {
      "epoch": 2.7750456653084163,
      "grad_norm": 0.8784831762313843,
      "learning_rate": 9.147029058840538e-05,
      "loss": 2.0854,
      "step": 39500
    },
    {
      "epoch": 2.810172825628776,
      "grad_norm": 0.8679407238960266,
      "learning_rate": 9.132571924244616e-05,
      "loss": 2.0739,
      "step": 40000
    },
    {
      "epoch": 2.810172825628776,
      "eval_accuracy": 0.6315697031283101,
      "eval_loss": 1.7932097911834717,
      "eval_runtime": 144.409,
      "eval_samples_per_second": 835.862,
      "eval_steps_per_second": 5.228,
      "step": 40000
    },
    {
      "epoch": 2.845299985949136,
      "grad_norm": 0.8974806070327759,
      "learning_rate": 9.118114789648692e-05,
      "loss": 2.0627,
      "step": 40500
    },
    {
      "epoch": 2.8804271462694957,
      "grad_norm": 0.8850300312042236,
      "learning_rate": 9.103657655052768e-05,
      "loss": 2.0544,
      "step": 41000
    },
    {
      "epoch": 2.915554306589855,
      "grad_norm": 0.8605426549911499,
      "learning_rate": 9.089200520456845e-05,
      "loss": 2.0464,
      "step": 41500
    },
    {
      "epoch": 2.950681466910215,
      "grad_norm": 0.8664257526397705,
      "learning_rate": 9.074743385860923e-05,
      "loss": 2.0404,
      "step": 42000
    },
    {
      "epoch": 2.9858086272305746,
      "grad_norm": 0.8507780432701111,
      "learning_rate": 9.060286251265e-05,
      "loss": 2.0325,
      "step": 42500
    },
    {
      "epoch": 3.0209357875509344,
      "grad_norm": 0.8595208525657654,
      "learning_rate": 9.045829116669077e-05,
      "loss": 2.018,
      "step": 43000
    },
    {
      "epoch": 3.056062947871294,
      "grad_norm": 0.8866372108459473,
      "learning_rate": 9.031371982073153e-05,
      "loss": 2.0135,
      "step": 43500
    },
    {
      "epoch": 3.091190108191654,
      "grad_norm": 0.8739855289459229,
      "learning_rate": 9.016914847477231e-05,
      "loss": 2.0101,
      "step": 44000
    },
    {
      "epoch": 3.1263172685120133,
      "grad_norm": 0.8769083619117737,
      "learning_rate": 9.002457712881307e-05,
      "loss": 2.0,
      "step": 44500
    },
    {
      "epoch": 3.161444428832373,
      "grad_norm": 0.8757010102272034,
      "learning_rate": 8.988000578285384e-05,
      "loss": 1.9949,
      "step": 45000
    },
    {
      "epoch": 3.196571589152733,
      "grad_norm": 0.8755298256874084,
      "learning_rate": 8.973543443689461e-05,
      "loss": 1.9849,
      "step": 45500
    },
    {
      "epoch": 3.2316987494730927,
      "grad_norm": 0.8611502647399902,
      "learning_rate": 8.959086309093538e-05,
      "loss": 1.9821,
      "step": 46000
    },
    {
      "epoch": 3.2668259097934524,
      "grad_norm": 0.8552160859107971,
      "learning_rate": 8.944629174497616e-05,
      "loss": 1.9738,
      "step": 46500
    },
    {
      "epoch": 3.301953070113812,
      "grad_norm": 0.8645509481430054,
      "learning_rate": 8.930172039901692e-05,
      "loss": 1.9678,
      "step": 47000
    },
    {
      "epoch": 3.3370802304341716,
      "grad_norm": 0.8320329189300537,
      "learning_rate": 8.915714905305768e-05,
      "loss": 1.9614,
      "step": 47500
    },
    {
      "epoch": 3.3722073907545314,
      "grad_norm": 0.8820323944091797,
      "learning_rate": 8.901257770709845e-05,
      "loss": 1.9578,
      "step": 48000
    },
    {
      "epoch": 3.407334551074891,
      "grad_norm": 0.847457230091095,
      "learning_rate": 8.886800636113923e-05,
      "loss": 1.9494,
      "step": 48500
    },
    {
      "epoch": 3.442461711395251,
      "grad_norm": 0.858734667301178,
      "learning_rate": 8.872343501518e-05,
      "loss": 1.942,
      "step": 49000
    },
    {
      "epoch": 3.4775888717156107,
      "grad_norm": 0.8552329540252686,
      "learning_rate": 8.857886366922077e-05,
      "loss": 1.9367,
      "step": 49500
    },
    {
      "epoch": 3.51271603203597,
      "grad_norm": 0.8536255359649658,
      "learning_rate": 8.843429232326153e-05,
      "loss": 1.9345,
      "step": 50000
    },
    {
      "epoch": 3.51271603203597,
      "eval_accuracy": 0.652718528822783,
      "eval_loss": 1.6618030071258545,
      "eval_runtime": 145.7638,
      "eval_samples_per_second": 828.093,
      "eval_steps_per_second": 5.18,
      "step": 50000
    },
    {
      "epoch": 3.54784319235633,
      "grad_norm": 0.8829108476638794,
      "learning_rate": 8.828972097730231e-05,
      "loss": 1.9303,
      "step": 50500
    },
    {
      "epoch": 3.5829703526766896,
      "grad_norm": 0.8787323236465454,
      "learning_rate": 8.814514963134307e-05,
      "loss": 1.9231,
      "step": 51000
    },
    {
      "epoch": 3.6180975129970494,
      "grad_norm": 0.8392846584320068,
      "learning_rate": 8.800057828538384e-05,
      "loss": 1.9205,
      "step": 51500
    },
    {
      "epoch": 3.6532246733174087,
      "grad_norm": 0.8578284382820129,
      "learning_rate": 8.78560069394246e-05,
      "loss": 1.9147,
      "step": 52000
    },
    {
      "epoch": 3.6883518336377685,
      "grad_norm": 0.8548659086227417,
      "learning_rate": 8.771143559346538e-05,
      "loss": 1.9079,
      "step": 52500
    },
    {
      "epoch": 3.7234789939581283,
      "grad_norm": 0.8397179245948792,
      "learning_rate": 8.756686424750616e-05,
      "loss": 1.9064,
      "step": 53000
    },
    {
      "epoch": 3.758606154278488,
      "grad_norm": 0.886605978012085,
      "learning_rate": 8.742229290154692e-05,
      "loss": 1.9006,
      "step": 53500
    },
    {
      "epoch": 3.793733314598848,
      "grad_norm": 0.8749421834945679,
      "learning_rate": 8.727772155558768e-05,
      "loss": 1.894,
      "step": 54000
    },
    {
      "epoch": 3.8288604749192077,
      "grad_norm": 0.8433464765548706,
      "learning_rate": 8.713315020962846e-05,
      "loss": 1.8917,
      "step": 54500
    },
    {
      "epoch": 3.8639876352395675,
      "grad_norm": 0.8542115092277527,
      "learning_rate": 8.698857886366923e-05,
      "loss": 1.8846,
      "step": 55000
    },
    {
      "epoch": 3.899114795559927,
      "grad_norm": 0.8477907776832581,
      "learning_rate": 8.684400751770999e-05,
      "loss": 1.88,
      "step": 55500
    },
    {
      "epoch": 3.9342419558802866,
      "grad_norm": 0.8515335321426392,
      "learning_rate": 8.669943617175075e-05,
      "loss": 1.8768,
      "step": 56000
    },
    {
      "epoch": 3.9693691162006464,
      "grad_norm": 0.8547754287719727,
      "learning_rate": 8.655486482579153e-05,
      "loss": 1.8742,
      "step": 56500
    },
    {
      "epoch": 4.004496276521006,
      "grad_norm": 0.8512959480285645,
      "learning_rate": 8.641029347983231e-05,
      "loss": 1.868,
      "step": 57000
    },
    {
      "epoch": 4.0396234368413655,
      "grad_norm": 0.8635633587837219,
      "learning_rate": 8.626572213387307e-05,
      "loss": 1.8603,
      "step": 57500
    },
    {
      "epoch": 4.074750597161725,
      "grad_norm": 0.8565309643745422,
      "learning_rate": 8.612115078791384e-05,
      "loss": 1.858,
      "step": 58000
    },
    {
      "epoch": 4.109877757482085,
      "grad_norm": 0.8473580479621887,
      "learning_rate": 8.597657944195461e-05,
      "loss": 1.8543,
      "step": 58500
    },
    {
      "epoch": 4.145004917802445,
      "grad_norm": 0.8602383732795715,
      "learning_rate": 8.583200809599538e-05,
      "loss": 1.8505,
      "step": 59000
    },
    {
      "epoch": 4.180132078122805,
      "grad_norm": 0.8420585989952087,
      "learning_rate": 8.568743675003614e-05,
      "loss": 1.8507,
      "step": 59500
    },
    {
      "epoch": 4.215259238443164,
      "grad_norm": 0.8763858079910278,
      "learning_rate": 8.55428654040769e-05,
      "loss": 1.844,
      "step": 60000
    },
    {
      "epoch": 4.215259238443164,
      "eval_accuracy": 0.6653352124865598,
      "eval_loss": 1.5828626155853271,
      "eval_runtime": 145.5436,
      "eval_samples_per_second": 829.346,
      "eval_steps_per_second": 5.187,
      "step": 60000
    },
    {
      "epoch": 4.250386398763524,
      "grad_norm": 0.8648141026496887,
      "learning_rate": 8.539829405811768e-05,
      "loss": 1.8439,
      "step": 60500
    },
    {
      "epoch": 4.285513559083884,
      "grad_norm": 0.8564465045928955,
      "learning_rate": 8.525372271215846e-05,
      "loss": 1.8372,
      "step": 61000
    },
    {
      "epoch": 4.320640719404244,
      "grad_norm": 0.8347111344337463,
      "learning_rate": 8.510915136619923e-05,
      "loss": 1.8322,
      "step": 61500
    },
    {
      "epoch": 4.355767879724603,
      "grad_norm": 0.8819388747215271,
      "learning_rate": 8.496458002023999e-05,
      "loss": 1.8299,
      "step": 62000
    },
    {
      "epoch": 4.3908950400449624,
      "grad_norm": 0.8561258316040039,
      "learning_rate": 8.482000867428075e-05,
      "loss": 1.8319,
      "step": 62500
    },
    {
      "epoch": 4.426022200365322,
      "grad_norm": 0.8755387663841248,
      "learning_rate": 8.467543732832153e-05,
      "loss": 1.827,
      "step": 63000
    },
    {
      "epoch": 4.461149360685682,
      "grad_norm": 0.8702017068862915,
      "learning_rate": 8.45308659823623e-05,
      "loss": 1.8213,
      "step": 63500
    },
    {
      "epoch": 4.496276521006042,
      "grad_norm": 0.8973068594932556,
      "learning_rate": 8.438629463640307e-05,
      "loss": 1.819,
      "step": 64000
    },
    {
      "epoch": 4.531403681326402,
      "grad_norm": 0.8871073126792908,
      "learning_rate": 8.424172329044384e-05,
      "loss": 1.8166,
      "step": 64500
    },
    {
      "epoch": 4.566530841646761,
      "grad_norm": 0.8684224486351013,
      "learning_rate": 8.409715194448461e-05,
      "loss": 1.8129,
      "step": 65000
    },
    {
      "epoch": 4.601658001967121,
      "grad_norm": 0.8582248687744141,
      "learning_rate": 8.395258059852538e-05,
      "loss": 1.8113,
      "step": 65500
    },
    {
      "epoch": 4.636785162287481,
      "grad_norm": 0.843012809753418,
      "learning_rate": 8.380800925256614e-05,
      "loss": 1.807,
      "step": 66000
    },
    {
      "epoch": 4.671912322607841,
      "grad_norm": 0.8517187237739563,
      "learning_rate": 8.36634379066069e-05,
      "loss": 1.8064,
      "step": 66500
    },
    {
      "epoch": 4.7070394829282005,
      "grad_norm": 0.8724709153175354,
      "learning_rate": 8.351886656064768e-05,
      "loss": 1.8003,
      "step": 67000
    },
    {
      "epoch": 4.742166643248559,
      "grad_norm": 0.8867350220680237,
      "learning_rate": 8.337429521468846e-05,
      "loss": 1.799,
      "step": 67500
    },
    {
      "epoch": 4.777293803568919,
      "grad_norm": 0.8441566824913025,
      "learning_rate": 8.322972386872923e-05,
      "loss": 1.7979,
      "step": 68000
    },
    {
      "epoch": 4.812420963889279,
      "grad_norm": 0.8734049201011658,
      "learning_rate": 8.308515252276999e-05,
      "loss": 1.7957,
      "step": 68500
    },
    {
      "epoch": 4.847548124209639,
      "grad_norm": 0.8671762347221375,
      "learning_rate": 8.294058117681077e-05,
      "loss": 1.789,
      "step": 69000
    },
    {
      "epoch": 4.8826752845299985,
      "grad_norm": 0.8785460591316223,
      "learning_rate": 8.279600983085153e-05,
      "loss": 1.7892,
      "step": 69500
    },
    {
      "epoch": 4.917802444850358,
      "grad_norm": 0.8827086687088013,
      "learning_rate": 8.26514384848923e-05,
      "loss": 1.7874,
      "step": 70000
    },
    {
      "epoch": 4.917802444850358,
      "eval_accuracy": 0.6750020166957967,
      "eval_loss": 1.5247992277145386,
      "eval_runtime": 144.6205,
      "eval_samples_per_second": 834.639,
      "eval_steps_per_second": 5.221,
      "step": 70000
    },
    {
      "epoch": 4.952929605170718,
      "grad_norm": 0.8440523743629456,
      "learning_rate": 8.250686713893306e-05,
      "loss": 1.7851,
      "step": 70500
    },
    {
      "epoch": 4.988056765491078,
      "grad_norm": 0.8763037919998169,
      "learning_rate": 8.236229579297384e-05,
      "loss": 1.7801,
      "step": 71000
    },
    {
      "epoch": 5.023183925811438,
      "grad_norm": 0.88693767786026,
      "learning_rate": 8.221772444701461e-05,
      "loss": 1.7789,
      "step": 71500
    },
    {
      "epoch": 5.0583110861317975,
      "grad_norm": 0.866883397102356,
      "learning_rate": 8.207315310105538e-05,
      "loss": 1.7714,
      "step": 72000
    },
    {
      "epoch": 5.093438246452157,
      "grad_norm": 0.8699513077735901,
      "learning_rate": 8.192858175509614e-05,
      "loss": 1.7682,
      "step": 72500
    },
    {
      "epoch": 5.128565406772516,
      "grad_norm": 0.8703712821006775,
      "learning_rate": 8.178401040913692e-05,
      "loss": 1.7662,
      "step": 73000
    },
    {
      "epoch": 5.163692567092876,
      "grad_norm": 0.854120135307312,
      "learning_rate": 8.163943906317768e-05,
      "loss": 1.7647,
      "step": 73500
    },
    {
      "epoch": 5.198819727413236,
      "grad_norm": 0.8597338795661926,
      "learning_rate": 8.149486771721845e-05,
      "loss": 1.7613,
      "step": 74000
    },
    {
      "epoch": 5.2339468877335955,
      "grad_norm": 0.8737318515777588,
      "learning_rate": 8.135029637125921e-05,
      "loss": 1.7627,
      "step": 74500
    },
    {
      "epoch": 5.269074048053955,
      "grad_norm": 0.8847633600234985,
      "learning_rate": 8.120572502529999e-05,
      "loss": 1.7594,
      "step": 75000
    },
    {
      "epoch": 5.304201208374315,
      "grad_norm": 0.8676943778991699,
      "learning_rate": 8.106115367934077e-05,
      "loss": 1.7574,
      "step": 75500
    },
    {
      "epoch": 5.339328368694675,
      "grad_norm": 0.8717545866966248,
      "learning_rate": 8.091658233338153e-05,
      "loss": 1.7551,
      "step": 76000
    },
    {
      "epoch": 5.374455529015035,
      "grad_norm": 0.8879154920578003,
      "learning_rate": 8.07720109874223e-05,
      "loss": 1.752,
      "step": 76500
    },
    {
      "epoch": 5.409582689335394,
      "grad_norm": 0.861165463924408,
      "learning_rate": 8.062743964146306e-05,
      "loss": 1.748,
      "step": 77000
    },
    {
      "epoch": 5.444709849655754,
      "grad_norm": 0.885162889957428,
      "learning_rate": 8.048286829550384e-05,
      "loss": 1.7444,
      "step": 77500
    },
    {
      "epoch": 5.479837009976113,
      "grad_norm": 0.8831413388252258,
      "learning_rate": 8.03382969495446e-05,
      "loss": 1.7457,
      "step": 78000
    },
    {
      "epoch": 5.514964170296473,
      "grad_norm": 0.8693949580192566,
      "learning_rate": 8.019372560358536e-05,
      "loss": 1.7439,
      "step": 78500
    },
    {
      "epoch": 5.550091330616833,
      "grad_norm": 0.8950614333152771,
      "learning_rate": 8.004915425762614e-05,
      "loss": 1.7414,
      "step": 79000
    },
    {
      "epoch": 5.5852184909371925,
      "grad_norm": 0.8772277235984802,
      "learning_rate": 7.990458291166692e-05,
      "loss": 1.7397,
      "step": 79500
    },
    {
      "epoch": 5.620345651257552,
      "grad_norm": 0.8824251890182495,
      "learning_rate": 7.976001156570768e-05,
      "loss": 1.737,
      "step": 80000
    },
    {
      "epoch": 5.620345651257552,
      "eval_accuracy": 0.6819064890501978,
      "eval_loss": 1.4823895692825317,
      "eval_runtime": 145.2164,
      "eval_samples_per_second": 831.215,
      "eval_steps_per_second": 5.199,
      "step": 80000
    },
    {
      "epoch": 5.655472811577912,
      "grad_norm": 0.9014139175415039,
      "learning_rate": 7.961544021974845e-05,
      "loss": 1.7339,
      "step": 80500
    },
    {
      "epoch": 5.690599971898272,
      "grad_norm": 0.8738614320755005,
      "learning_rate": 7.947086887378921e-05,
      "loss": 1.7395,
      "step": 81000
    },
    {
      "epoch": 5.725727132218632,
      "grad_norm": 0.8764089941978455,
      "learning_rate": 7.932629752782999e-05,
      "loss": 1.7336,
      "step": 81500
    },
    {
      "epoch": 5.760854292538991,
      "grad_norm": 0.9202150106430054,
      "learning_rate": 7.918172618187075e-05,
      "loss": 1.7291,
      "step": 82000
    },
    {
      "epoch": 5.795981452859351,
      "grad_norm": 0.8928285837173462,
      "learning_rate": 7.903715483591153e-05,
      "loss": 1.7302,
      "step": 82500
    },
    {
      "epoch": 5.83110861317971,
      "grad_norm": 0.8655526041984558,
      "learning_rate": 7.88925834899523e-05,
      "loss": 1.7287,
      "step": 83000
    },
    {
      "epoch": 5.866235773500071,
      "grad_norm": 0.8782069087028503,
      "learning_rate": 7.874801214399307e-05,
      "loss": 1.7261,
      "step": 83500
    },
    {
      "epoch": 5.90136293382043,
      "grad_norm": 0.8890528678894043,
      "learning_rate": 7.860344079803384e-05,
      "loss": 1.7237,
      "step": 84000
    },
    {
      "epoch": 5.936490094140789,
      "grad_norm": 0.8649280667304993,
      "learning_rate": 7.84588694520746e-05,
      "loss": 1.7213,
      "step": 84500
    },
    {
      "epoch": 5.971617254461149,
      "grad_norm": 0.9018586277961731,
      "learning_rate": 7.831429810611536e-05,
      "loss": 1.7194,
      "step": 85000
    },
    {
      "epoch": 6.006744414781509,
      "grad_norm": 0.8934980034828186,
      "learning_rate": 7.816972676015614e-05,
      "loss": 1.7184,
      "step": 85500
    },
    {
      "epoch": 6.041871575101869,
      "grad_norm": 0.8839065432548523,
      "learning_rate": 7.802515541419692e-05,
      "loss": 1.7127,
      "step": 86000
    },
    {
      "epoch": 6.076998735422229,
      "grad_norm": 0.8806888461112976,
      "learning_rate": 7.788058406823768e-05,
      "loss": 1.7144,
      "step": 86500
    },
    {
      "epoch": 6.112125895742588,
      "grad_norm": 0.8924936652183533,
      "learning_rate": 7.773601272227845e-05,
      "loss": 1.7084,
      "step": 87000
    },
    {
      "epoch": 6.147253056062948,
      "grad_norm": 0.9053957462310791,
      "learning_rate": 7.759144137631922e-05,
      "loss": 1.7076,
      "step": 87500
    },
    {
      "epoch": 6.182380216383308,
      "grad_norm": 0.8794803023338318,
      "learning_rate": 7.744687003035999e-05,
      "loss": 1.7069,
      "step": 88000
    },
    {
      "epoch": 6.217507376703667,
      "grad_norm": 0.9028088450431824,
      "learning_rate": 7.730229868440075e-05,
      "loss": 1.7067,
      "step": 88500
    },
    {
      "epoch": 6.252634537024027,
      "grad_norm": 0.8775391578674316,
      "learning_rate": 7.715772733844152e-05,
      "loss": 1.7021,
      "step": 89000
    },
    {
      "epoch": 6.287761697344386,
      "grad_norm": 0.9188543558120728,
      "learning_rate": 7.70131559924823e-05,
      "loss": 1.7019,
      "step": 89500
    },
    {
      "epoch": 6.322888857664746,
      "grad_norm": 0.8761645555496216,
      "learning_rate": 7.686858464652307e-05,
      "loss": 1.7017,
      "step": 90000
    },
    {
      "epoch": 6.322888857664746,
      "eval_accuracy": 0.6875799717584923,
      "eval_loss": 1.45062255859375,
      "eval_runtime": 146.6571,
      "eval_samples_per_second": 823.049,
      "eval_steps_per_second": 5.148,
      "step": 90000
    },
    {
      "epoch": 6.358016017985106,
      "grad_norm": 0.8915168642997742,
      "learning_rate": 7.672401330056384e-05,
      "loss": 1.7011,
      "step": 90500
    },
    {
      "epoch": 6.393143178305466,
      "grad_norm": 0.8964366912841797,
      "learning_rate": 7.65794419546046e-05,
      "loss": 1.6985,
      "step": 91000
    },
    {
      "epoch": 6.4282703386258255,
      "grad_norm": 0.8956195712089539,
      "learning_rate": 7.643487060864536e-05,
      "loss": 1.6956,
      "step": 91500
    },
    {
      "epoch": 6.463397498946185,
      "grad_norm": 0.9271071553230286,
      "learning_rate": 7.629029926268614e-05,
      "loss": 1.695,
      "step": 92000
    },
    {
      "epoch": 6.498524659266545,
      "grad_norm": 0.882736086845398,
      "learning_rate": 7.61457279167269e-05,
      "loss": 1.6941,
      "step": 92500
    },
    {
      "epoch": 6.533651819586905,
      "grad_norm": 0.866625964641571,
      "learning_rate": 7.600115657076767e-05,
      "loss": 1.6942,
      "step": 93000
    },
    {
      "epoch": 6.568778979907265,
      "grad_norm": 0.9276591539382935,
      "learning_rate": 7.585658522480845e-05,
      "loss": 1.6915,
      "step": 93500
    },
    {
      "epoch": 6.603906140227624,
      "grad_norm": 0.8902279138565063,
      "learning_rate": 7.571201387884922e-05,
      "loss": 1.6874,
      "step": 94000
    },
    {
      "epoch": 6.639033300547983,
      "grad_norm": 0.8612838983535767,
      "learning_rate": 7.556744253288999e-05,
      "loss": 1.6864,
      "step": 94500
    },
    {
      "epoch": 6.674160460868343,
      "grad_norm": 0.8922297358512878,
      "learning_rate": 7.542287118693075e-05,
      "loss": 1.6851,
      "step": 95000
    },
    {
      "epoch": 6.709287621188703,
      "grad_norm": 0.9003462791442871,
      "learning_rate": 7.527829984097152e-05,
      "loss": 1.6843,
      "step": 95500
    },
    {
      "epoch": 6.744414781509063,
      "grad_norm": 0.8967103958129883,
      "learning_rate": 7.51337284950123e-05,
      "loss": 1.6888,
      "step": 96000
    },
    {
      "epoch": 6.7795419418294225,
      "grad_norm": 0.8979239463806152,
      "learning_rate": 7.498915714905306e-05,
      "loss": 1.6813,
      "step": 96500
    },
    {
      "epoch": 6.814669102149782,
      "grad_norm": 0.8968941569328308,
      "learning_rate": 7.484458580309382e-05,
      "loss": 1.6822,
      "step": 97000
    },
    {
      "epoch": 6.849796262470142,
      "grad_norm": 0.8832077383995056,
      "learning_rate": 7.47000144571346e-05,
      "loss": 1.676,
      "step": 97500
    },
    {
      "epoch": 6.884923422790502,
      "grad_norm": 0.8957480192184448,
      "learning_rate": 7.455544311117538e-05,
      "loss": 1.6772,
      "step": 98000
    },
    {
      "epoch": 6.920050583110862,
      "grad_norm": 0.911115825176239,
      "learning_rate": 7.441087176521614e-05,
      "loss": 1.674,
      "step": 98500
    },
    {
      "epoch": 6.955177743431221,
      "grad_norm": 0.9081796407699585,
      "learning_rate": 7.42663004192569e-05,
      "loss": 1.6746,
      "step": 99000
    },
    {
      "epoch": 6.99030490375158,
      "grad_norm": 0.9057807922363281,
      "learning_rate": 7.412172907329767e-05,
      "loss": 1.6702,
      "step": 99500
    },
    {
      "epoch": 7.02543206407194,
      "grad_norm": 0.9082474708557129,
      "learning_rate": 7.397715772733845e-05,
      "loss": 1.6703,
      "step": 100000
    },
    {
      "epoch": 7.02543206407194,
      "eval_accuracy": 0.6921466981210505,
      "eval_loss": 1.420421838760376,
      "eval_runtime": 146.0221,
      "eval_samples_per_second": 826.628,
      "eval_steps_per_second": 5.17,
      "step": 100000
    },
    {
      "epoch": 7.0605592243923,
      "grad_norm": 0.9067661166191101,
      "learning_rate": 7.383258638137921e-05,
      "loss": 1.6691,
      "step": 100500
    },
    {
      "epoch": 7.09568638471266,
      "grad_norm": 0.9043187499046326,
      "learning_rate": 7.368801503541999e-05,
      "loss": 1.669,
      "step": 101000
    },
    {
      "epoch": 7.1308135450330195,
      "grad_norm": 0.8964481949806213,
      "learning_rate": 7.354344368946075e-05,
      "loss": 1.667,
      "step": 101500
    },
    {
      "epoch": 7.165940705353379,
      "grad_norm": 0.9258145689964294,
      "learning_rate": 7.339887234350153e-05,
      "loss": 1.6653,
      "step": 102000
    },
    {
      "epoch": 7.201067865673739,
      "grad_norm": 0.8769683241844177,
      "learning_rate": 7.32543009975423e-05,
      "loss": 1.664,
      "step": 102500
    },
    {
      "epoch": 7.236195025994099,
      "grad_norm": 0.910929262638092,
      "learning_rate": 7.310972965158306e-05,
      "loss": 1.6624,
      "step": 103000
    },
    {
      "epoch": 7.271322186314459,
      "grad_norm": 0.8838810324668884,
      "learning_rate": 7.296515830562382e-05,
      "loss": 1.6628,
      "step": 103500
    },
    {
      "epoch": 7.306449346634818,
      "grad_norm": 0.8857362270355225,
      "learning_rate": 7.28205869596646e-05,
      "loss": 1.658,
      "step": 104000
    },
    {
      "epoch": 7.341576506955178,
      "grad_norm": 0.8846690654754639,
      "learning_rate": 7.267601561370536e-05,
      "loss": 1.6543,
      "step": 104500
    },
    {
      "epoch": 7.376703667275537,
      "grad_norm": 0.8960644006729126,
      "learning_rate": 7.253144426774614e-05,
      "loss": 1.6574,
      "step": 105000
    },
    {
      "epoch": 7.411830827595897,
      "grad_norm": 0.9093925356864929,
      "learning_rate": 7.23868729217869e-05,
      "loss": 1.6602,
      "step": 105500
    },
    {
      "epoch": 7.446957987916257,
      "grad_norm": 0.8878596425056458,
      "learning_rate": 7.224230157582768e-05,
      "loss": 1.6583,
      "step": 106000
    },
    {
      "epoch": 7.482085148236616,
      "grad_norm": 0.8982939124107361,
      "learning_rate": 7.209773022986845e-05,
      "loss": 1.6542,
      "step": 106500
    },
    {
      "epoch": 7.517212308556976,
      "grad_norm": 0.931877613067627,
      "learning_rate": 7.195315888390921e-05,
      "loss": 1.6513,
      "step": 107000
    },
    {
      "epoch": 7.552339468877336,
      "grad_norm": 0.9102097749710083,
      "learning_rate": 7.180858753794997e-05,
      "loss": 1.6505,
      "step": 107500
    },
    {
      "epoch": 7.587466629197696,
      "grad_norm": 0.8833571076393127,
      "learning_rate": 7.166401619199075e-05,
      "loss": 1.6506,
      "step": 108000
    },
    {
      "epoch": 7.6225937895180556,
      "grad_norm": 0.8974440693855286,
      "learning_rate": 7.151944484603153e-05,
      "loss": 1.6529,
      "step": 108500
    },
    {
      "epoch": 7.657720949838415,
      "grad_norm": 0.8859068155288696,
      "learning_rate": 7.13748735000723e-05,
      "loss": 1.6483,
      "step": 109000
    },
    {
      "epoch": 7.692848110158775,
      "grad_norm": 0.9290302991867065,
      "learning_rate": 7.123030215411306e-05,
      "loss": 1.649,
      "step": 109500
    },
    {
      "epoch": 7.727975270479135,
      "grad_norm": 0.912346601486206,
      "learning_rate": 7.108573080815382e-05,
      "loss": 1.6497,
      "step": 110000
    },
    {
      "epoch": 7.727975270479135,
      "eval_accuracy": 0.696107465901776,
      "eval_loss": 1.3987531661987305,
      "eval_runtime": 145.5941,
      "eval_samples_per_second": 829.058,
      "eval_steps_per_second": 5.186,
      "step": 110000
    },
    {
      "epoch": 7.763102430799494,
      "grad_norm": 0.9191212058067322,
      "learning_rate": 7.09411594621946e-05,
      "loss": 1.6456,
      "step": 110500
    },
    {
      "epoch": 7.798229591119854,
      "grad_norm": 0.9030919075012207,
      "learning_rate": 7.079658811623536e-05,
      "loss": 1.6428,
      "step": 111000
    },
    {
      "epoch": 7.833356751440213,
      "grad_norm": 0.9317150115966797,
      "learning_rate": 7.065201677027613e-05,
      "loss": 1.6428,
      "step": 111500
    },
    {
      "epoch": 7.868483911760573,
      "grad_norm": 0.9374363422393799,
      "learning_rate": 7.05074454243169e-05,
      "loss": 1.6444,
      "step": 112000
    },
    {
      "epoch": 7.903611072080933,
      "grad_norm": 0.8871532678604126,
      "learning_rate": 7.036287407835768e-05,
      "loss": 1.6428,
      "step": 112500
    },
    {
      "epoch": 7.938738232401293,
      "grad_norm": 0.947810709476471,
      "learning_rate": 7.021830273239845e-05,
      "loss": 1.6366,
      "step": 113000
    },
    {
      "epoch": 7.9738653927216525,
      "grad_norm": 0.8954718708992004,
      "learning_rate": 7.007373138643921e-05,
      "loss": 1.6379,
      "step": 113500
    },
    {
      "epoch": 8.008992553042011,
      "grad_norm": 0.8930262327194214,
      "learning_rate": 6.992916004047997e-05,
      "loss": 1.6387,
      "step": 114000
    },
    {
      "epoch": 8.044119713362372,
      "grad_norm": 0.9441146850585938,
      "learning_rate": 6.978458869452075e-05,
      "loss": 1.6332,
      "step": 114500
    },
    {
      "epoch": 8.079246873682731,
      "grad_norm": 0.9339563846588135,
      "learning_rate": 6.964001734856152e-05,
      "loss": 1.6326,
      "step": 115000
    },
    {
      "epoch": 8.114374034003092,
      "grad_norm": 0.917434811592102,
      "learning_rate": 6.949544600260228e-05,
      "loss": 1.6311,
      "step": 115500
    },
    {
      "epoch": 8.14950119432345,
      "grad_norm": 0.9269524216651917,
      "learning_rate": 6.935087465664306e-05,
      "loss": 1.6327,
      "step": 116000
    },
    {
      "epoch": 8.184628354643811,
      "grad_norm": 0.9198341965675354,
      "learning_rate": 6.920630331068383e-05,
      "loss": 1.6323,
      "step": 116500
    },
    {
      "epoch": 8.21975551496417,
      "grad_norm": 0.9190070629119873,
      "learning_rate": 6.90617319647246e-05,
      "loss": 1.6286,
      "step": 117000
    },
    {
      "epoch": 8.25488267528453,
      "grad_norm": 0.9343723654747009,
      "learning_rate": 6.891716061876536e-05,
      "loss": 1.6291,
      "step": 117500
    },
    {
      "epoch": 8.29000983560489,
      "grad_norm": 0.9313236474990845,
      "learning_rate": 6.877258927280613e-05,
      "loss": 1.6253,
      "step": 118000
    },
    {
      "epoch": 8.325136995925249,
      "grad_norm": 0.9170568585395813,
      "learning_rate": 6.86280179268469e-05,
      "loss": 1.6244,
      "step": 118500
    },
    {
      "epoch": 8.36026415624561,
      "grad_norm": 0.9194092154502869,
      "learning_rate": 6.848344658088767e-05,
      "loss": 1.6251,
      "step": 119000
    },
    {
      "epoch": 8.395391316565968,
      "grad_norm": 0.903550922870636,
      "learning_rate": 6.833887523492845e-05,
      "loss": 1.6245,
      "step": 119500
    },
    {
      "epoch": 8.430518476886329,
      "grad_norm": 0.9155687689781189,
      "learning_rate": 6.819430388896921e-05,
      "loss": 1.6245,
      "step": 120000
    },
    {
      "epoch": 8.430518476886329,
      "eval_accuracy": 0.699622936687572,
      "eval_loss": 1.376625895500183,
      "eval_runtime": 146.4286,
      "eval_samples_per_second": 824.334,
      "eval_steps_per_second": 5.156,
      "step": 120000
    },
    {
      "epoch": 8.465645637206688,
      "grad_norm": 0.9170736074447632,
      "learning_rate": 6.804973254300999e-05,
      "loss": 1.6241,
      "step": 120500
    },
    {
      "epoch": 8.500772797527048,
      "grad_norm": 0.918855607509613,
      "learning_rate": 6.790516119705075e-05,
      "loss": 1.6238,
      "step": 121000
    },
    {
      "epoch": 8.535899957847407,
      "grad_norm": 0.9043271541595459,
      "learning_rate": 6.776058985109152e-05,
      "loss": 1.6253,
      "step": 121500
    },
    {
      "epoch": 8.571027118167768,
      "grad_norm": 0.9378098845481873,
      "learning_rate": 6.761601850513228e-05,
      "loss": 1.6207,
      "step": 122000
    },
    {
      "epoch": 8.606154278488127,
      "grad_norm": 0.9023655652999878,
      "learning_rate": 6.747144715917306e-05,
      "loss": 1.6192,
      "step": 122500
    },
    {
      "epoch": 8.641281438808488,
      "grad_norm": 0.90032958984375,
      "learning_rate": 6.732687581321382e-05,
      "loss": 1.6188,
      "step": 123000
    },
    {
      "epoch": 8.676408599128846,
      "grad_norm": 0.9296461939811707,
      "learning_rate": 6.71823044672546e-05,
      "loss": 1.6147,
      "step": 123500
    },
    {
      "epoch": 8.711535759449205,
      "grad_norm": 0.917468249797821,
      "learning_rate": 6.703773312129536e-05,
      "loss": 1.614,
      "step": 124000
    },
    {
      "epoch": 8.746662919769566,
      "grad_norm": 0.9283198714256287,
      "learning_rate": 6.689316177533613e-05,
      "loss": 1.6088,
      "step": 124500
    },
    {
      "epoch": 8.781790080089925,
      "grad_norm": 0.9109067916870117,
      "learning_rate": 6.67485904293769e-05,
      "loss": 1.6148,
      "step": 125000
    },
    {
      "epoch": 8.816917240410286,
      "grad_norm": 0.9425482153892517,
      "learning_rate": 6.660401908341767e-05,
      "loss": 1.6125,
      "step": 125500
    },
    {
      "epoch": 8.852044400730644,
      "grad_norm": 0.9108548164367676,
      "learning_rate": 6.645944773745843e-05,
      "loss": 1.6118,
      "step": 126000
    },
    {
      "epoch": 8.887171561051005,
      "grad_norm": 0.9401180744171143,
      "learning_rate": 6.631487639149921e-05,
      "loss": 1.6121,
      "step": 126500
    },
    {
      "epoch": 8.922298721371364,
      "grad_norm": 0.9434899091720581,
      "learning_rate": 6.617030504553999e-05,
      "loss": 1.6099,
      "step": 127000
    },
    {
      "epoch": 8.957425881691725,
      "grad_norm": 0.9705188870429993,
      "learning_rate": 6.602573369958075e-05,
      "loss": 1.6143,
      "step": 127500
    },
    {
      "epoch": 8.992553042012084,
      "grad_norm": 0.9429416060447693,
      "learning_rate": 6.588116235362152e-05,
      "loss": 1.6131,
      "step": 128000
    },
    {
      "epoch": 9.027680202332444,
      "grad_norm": 0.9325101971626282,
      "learning_rate": 6.573659100766228e-05,
      "loss": 1.6054,
      "step": 128500
    },
    {
      "epoch": 9.062807362652803,
      "grad_norm": 0.9009438157081604,
      "learning_rate": 6.559201966170306e-05,
      "loss": 1.6049,
      "step": 129000
    },
    {
      "epoch": 9.097934522973162,
      "grad_norm": 0.9559112191200256,
      "learning_rate": 6.544744831574382e-05,
      "loss": 1.6065,
      "step": 129500
    },
    {
      "epoch": 9.133061683293523,
      "grad_norm": 0.9075998067855835,
      "learning_rate": 6.530287696978458e-05,
      "loss": 1.6015,
      "step": 130000
    },
    {
      "epoch": 9.133061683293523,
      "eval_accuracy": 0.701941754455614,
      "eval_loss": 1.362774133682251,
      "eval_runtime": 146.6899,
      "eval_samples_per_second": 822.865,
      "eval_steps_per_second": 5.147,
      "step": 130000
    },
    {
      "epoch": 9.168188843613882,
      "grad_norm": 0.9487557411193848,
      "learning_rate": 6.515830562382536e-05,
      "loss": 1.603,
      "step": 130500
    },
    {
      "epoch": 9.203316003934242,
      "grad_norm": 0.9400453567504883,
      "learning_rate": 6.501373427786614e-05,
      "loss": 1.6042,
      "step": 131000
    },
    {
      "epoch": 9.238443164254601,
      "grad_norm": 0.9347612261772156,
      "learning_rate": 6.48691629319069e-05,
      "loss": 1.6033,
      "step": 131500
    },
    {
      "epoch": 9.273570324574962,
      "grad_norm": 0.9069722890853882,
      "learning_rate": 6.472459158594767e-05,
      "loss": 1.6038,
      "step": 132000
    },
    {
      "epoch": 9.30869748489532,
      "grad_norm": 0.9513418674468994,
      "learning_rate": 6.458002023998843e-05,
      "loss": 1.6023,
      "step": 132500
    },
    {
      "epoch": 9.343824645215681,
      "grad_norm": 0.9263851642608643,
      "learning_rate": 6.443544889402921e-05,
      "loss": 1.5997,
      "step": 133000
    },
    {
      "epoch": 9.37895180553604,
      "grad_norm": 0.9571240544319153,
      "learning_rate": 6.429087754806997e-05,
      "loss": 1.5987,
      "step": 133500
    },
    {
      "epoch": 9.414078965856401,
      "grad_norm": 0.9037933349609375,
      "learning_rate": 6.414630620211074e-05,
      "loss": 1.5979,
      "step": 134000
    },
    {
      "epoch": 9.44920612617676,
      "grad_norm": 0.9799155592918396,
      "learning_rate": 6.400173485615151e-05,
      "loss": 1.5952,
      "step": 134500
    },
    {
      "epoch": 9.484333286497119,
      "grad_norm": 0.9275627732276917,
      "learning_rate": 6.385716351019229e-05,
      "loss": 1.595,
      "step": 135000
    },
    {
      "epoch": 9.51946044681748,
      "grad_norm": 0.9277747273445129,
      "learning_rate": 6.371259216423306e-05,
      "loss": 1.5961,
      "step": 135500
    },
    {
      "epoch": 9.554587607137838,
      "grad_norm": 0.9405021071434021,
      "learning_rate": 6.356802081827382e-05,
      "loss": 1.5955,
      "step": 136000
    },
    {
      "epoch": 9.589714767458199,
      "grad_norm": 0.9006831049919128,
      "learning_rate": 6.342344947231458e-05,
      "loss": 1.5967,
      "step": 136500
    },
    {
      "epoch": 9.624841927778558,
      "grad_norm": 0.9479206800460815,
      "learning_rate": 6.327887812635536e-05,
      "loss": 1.5923,
      "step": 137000
    },
    {
      "epoch": 9.659969088098919,
      "grad_norm": 0.9280245304107666,
      "learning_rate": 6.313430678039613e-05,
      "loss": 1.5932,
      "step": 137500
    },
    {
      "epoch": 9.695096248419278,
      "grad_norm": 0.9124751091003418,
      "learning_rate": 6.29897354344369e-05,
      "loss": 1.5882,
      "step": 138000
    },
    {
      "epoch": 9.730223408739638,
      "grad_norm": 0.9291515946388245,
      "learning_rate": 6.284516408847767e-05,
      "loss": 1.5948,
      "step": 138500
    },
    {
      "epoch": 9.765350569059997,
      "grad_norm": 0.9335107207298279,
      "learning_rate": 6.270059274251843e-05,
      "loss": 1.5896,
      "step": 139000
    },
    {
      "epoch": 9.800477729380358,
      "grad_norm": 0.9176722764968872,
      "learning_rate": 6.255602139655921e-05,
      "loss": 1.5901,
      "step": 139500
    },
    {
      "epoch": 9.835604889700717,
      "grad_norm": 0.9138041138648987,
      "learning_rate": 6.241145005059997e-05,
      "loss": 1.5882,
      "step": 140000
    },
    {
      "epoch": 9.835604889700717,
      "eval_accuracy": 0.7051619721997767,
      "eval_loss": 1.3451346158981323,
      "eval_runtime": 145.0544,
      "eval_samples_per_second": 832.143,
      "eval_steps_per_second": 5.205,
      "step": 140000
    },
    {
      "epoch": 9.870732050021076,
      "grad_norm": 0.9771543741226196,
      "learning_rate": 6.226687870464074e-05,
      "loss": 1.5882,
      "step": 140500
    },
    {
      "epoch": 9.905859210341436,
      "grad_norm": 0.9482930302619934,
      "learning_rate": 6.212230735868151e-05,
      "loss": 1.5866,
      "step": 141000
    },
    {
      "epoch": 9.940986370661795,
      "grad_norm": 0.9545276761054993,
      "learning_rate": 6.197773601272228e-05,
      "loss": 1.5868,
      "step": 141500
    },
    {
      "epoch": 9.976113530982156,
      "grad_norm": 0.962986171245575,
      "learning_rate": 6.183316466676306e-05,
      "loss": 1.5833,
      "step": 142000
    },
    {
      "epoch": 10.011240691302515,
      "grad_norm": 0.9559778571128845,
      "learning_rate": 6.168859332080382e-05,
      "loss": 1.5858,
      "step": 142500
    },
    {
      "epoch": 10.046367851622875,
      "grad_norm": 0.9056123495101929,
      "learning_rate": 6.154402197484458e-05,
      "loss": 1.5855,
      "step": 143000
    },
    {
      "epoch": 10.081495011943234,
      "grad_norm": 0.9433762431144714,
      "learning_rate": 6.139945062888536e-05,
      "loss": 1.581,
      "step": 143500
    },
    {
      "epoch": 10.116622172263595,
      "grad_norm": 0.9369404315948486,
      "learning_rate": 6.125487928292613e-05,
      "loss": 1.5795,
      "step": 144000
    },
    {
      "epoch": 10.151749332583954,
      "grad_norm": 1.009413480758667,
      "learning_rate": 6.111030793696689e-05,
      "loss": 1.5793,
      "step": 144500
    },
    {
      "epoch": 10.186876492904315,
      "grad_norm": 0.9554888606071472,
      "learning_rate": 6.096573659100766e-05,
      "loss": 1.5797,
      "step": 145000
    },
    {
      "epoch": 10.222003653224673,
      "grad_norm": 0.9255361557006836,
      "learning_rate": 6.082116524504844e-05,
      "loss": 1.5809,
      "step": 145500
    },
    {
      "epoch": 10.257130813545032,
      "grad_norm": 0.9112059473991394,
      "learning_rate": 6.067659389908921e-05,
      "loss": 1.5813,
      "step": 146000
    },
    {
      "epoch": 10.292257973865393,
      "grad_norm": 0.9442374110221863,
      "learning_rate": 6.053202255312997e-05,
      "loss": 1.5757,
      "step": 146500
    },
    {
      "epoch": 10.327385134185752,
      "grad_norm": 0.9561610221862793,
      "learning_rate": 6.0387451207170744e-05,
      "loss": 1.5765,
      "step": 147000
    },
    {
      "epoch": 10.362512294506113,
      "grad_norm": 0.9480382800102234,
      "learning_rate": 6.024287986121151e-05,
      "loss": 1.5769,
      "step": 147500
    },
    {
      "epoch": 10.397639454826471,
      "grad_norm": 0.9375914931297302,
      "learning_rate": 6.009830851525228e-05,
      "loss": 1.5744,
      "step": 148000
    },
    {
      "epoch": 10.432766615146832,
      "grad_norm": 0.9849138259887695,
      "learning_rate": 5.995373716929304e-05,
      "loss": 1.5778,
      "step": 148500
    },
    {
      "epoch": 10.467893775467191,
      "grad_norm": 0.9578582048416138,
      "learning_rate": 5.980916582333381e-05,
      "loss": 1.5729,
      "step": 149000
    },
    {
      "epoch": 10.503020935787552,
      "grad_norm": 0.9367204308509827,
      "learning_rate": 5.966459447737459e-05,
      "loss": 1.5714,
      "step": 149500
    },
    {
      "epoch": 10.53814809610791,
      "grad_norm": 0.9462071061134338,
      "learning_rate": 5.952002313141536e-05,
      "loss": 1.5738,
      "step": 150000
    },
    {
      "epoch": 10.53814809610791,
      "eval_accuracy": 0.7075515620989917,
      "eval_loss": 1.331017255783081,
      "eval_runtime": 146.1625,
      "eval_samples_per_second": 825.834,
      "eval_steps_per_second": 5.165,
      "step": 150000
    },
    {
      "epoch": 10.573275256428271,
      "grad_norm": 0.9778929948806763,
      "learning_rate": 5.9375451785456126e-05,
      "loss": 1.5759,
      "step": 150500
    },
    {
      "epoch": 10.60840241674863,
      "grad_norm": 0.9483457803726196,
      "learning_rate": 5.9230880439496896e-05,
      "loss": 1.5711,
      "step": 151000
    },
    {
      "epoch": 10.643529577068989,
      "grad_norm": 0.9661710262298584,
      "learning_rate": 5.908630909353766e-05,
      "loss": 1.5713,
      "step": 151500
    },
    {
      "epoch": 10.67865673738935,
      "grad_norm": 0.926715612411499,
      "learning_rate": 5.894173774757843e-05,
      "loss": 1.5706,
      "step": 152000
    },
    {
      "epoch": 10.713783897709709,
      "grad_norm": 0.9615681767463684,
      "learning_rate": 5.8797166401619195e-05,
      "loss": 1.5685,
      "step": 152500
    },
    {
      "epoch": 10.74891105803007,
      "grad_norm": 0.9441356658935547,
      "learning_rate": 5.865259505565997e-05,
      "loss": 1.5689,
      "step": 153000
    },
    {
      "epoch": 10.784038218350428,
      "grad_norm": 0.9553456902503967,
      "learning_rate": 5.8508023709700743e-05,
      "loss": 1.5708,
      "step": 153500
    },
    {
      "epoch": 10.819165378670789,
      "grad_norm": 0.9203099012374878,
      "learning_rate": 5.8363452363741514e-05,
      "loss": 1.5691,
      "step": 154000
    },
    {
      "epoch": 10.854292538991148,
      "grad_norm": 0.9613307118415833,
      "learning_rate": 5.821888101778228e-05,
      "loss": 1.5691,
      "step": 154500
    },
    {
      "epoch": 10.889419699311508,
      "grad_norm": 0.93576580286026,
      "learning_rate": 5.807430967182305e-05,
      "loss": 1.5683,
      "step": 155000
    },
    {
      "epoch": 10.924546859631867,
      "grad_norm": 0.959653913974762,
      "learning_rate": 5.792973832586381e-05,
      "loss": 1.5685,
      "step": 155500
    },
    {
      "epoch": 10.959674019952226,
      "grad_norm": 0.9653539657592773,
      "learning_rate": 5.7785166979904584e-05,
      "loss": 1.5662,
      "step": 156000
    },
    {
      "epoch": 10.994801180272587,
      "grad_norm": 0.9286296367645264,
      "learning_rate": 5.764059563394536e-05,
      "loss": 1.5656,
      "step": 156500
    },
    {
      "epoch": 11.029928340592946,
      "grad_norm": 0.9834336042404175,
      "learning_rate": 5.7496024287986125e-05,
      "loss": 1.5588,
      "step": 157000
    },
    {
      "epoch": 11.065055500913306,
      "grad_norm": 0.9424375295639038,
      "learning_rate": 5.7351452942026896e-05,
      "loss": 1.562,
      "step": 157500
    },
    {
      "epoch": 11.100182661233665,
      "grad_norm": 0.9713384509086609,
      "learning_rate": 5.720688159606766e-05,
      "loss": 1.562,
      "step": 158000
    },
    {
      "epoch": 11.135309821554026,
      "grad_norm": 0.9827730655670166,
      "learning_rate": 5.706231025010843e-05,
      "loss": 1.5595,
      "step": 158500
    },
    {
      "epoch": 11.170436981874385,
      "grad_norm": 0.969085693359375,
      "learning_rate": 5.6917738904149195e-05,
      "loss": 1.5606,
      "step": 159000
    },
    {
      "epoch": 11.205564142194746,
      "grad_norm": 0.9821890592575073,
      "learning_rate": 5.6773167558189966e-05,
      "loss": 1.5607,
      "step": 159500
    },
    {
      "epoch": 11.240691302515105,
      "grad_norm": 0.9560614824295044,
      "learning_rate": 5.6628596212230736e-05,
      "loss": 1.563,
      "step": 160000
    },
    {
      "epoch": 11.240691302515105,
      "eval_accuracy": 0.7090796716396742,
      "eval_loss": 1.3214398622512817,
      "eval_runtime": 147.4505,
      "eval_samples_per_second": 818.621,
      "eval_steps_per_second": 5.12,
      "step": 160000
    },
    {
      "epoch": 11.275818462835465,
      "grad_norm": 0.9682839512825012,
      "learning_rate": 5.6484024866271514e-05,
      "loss": 1.558,
      "step": 160500
    },
    {
      "epoch": 11.310945623155824,
      "grad_norm": 0.958651065826416,
      "learning_rate": 5.633945352031228e-05,
      "loss": 1.5618,
      "step": 161000
    },
    {
      "epoch": 11.346072783476183,
      "grad_norm": 0.9705203771591187,
      "learning_rate": 5.619488217435305e-05,
      "loss": 1.5576,
      "step": 161500
    },
    {
      "epoch": 11.381199943796544,
      "grad_norm": 0.9865082502365112,
      "learning_rate": 5.605031082839381e-05,
      "loss": 1.5583,
      "step": 162000
    },
    {
      "epoch": 11.416327104116903,
      "grad_norm": 0.9645934700965881,
      "learning_rate": 5.5905739482434584e-05,
      "loss": 1.5551,
      "step": 162500
    },
    {
      "epoch": 11.451454264437263,
      "grad_norm": 0.9389051795005798,
      "learning_rate": 5.576116813647535e-05,
      "loss": 1.5521,
      "step": 163000
    },
    {
      "epoch": 11.486581424757622,
      "grad_norm": 0.9918972849845886,
      "learning_rate": 5.561659679051612e-05,
      "loss": 1.5551,
      "step": 163500
    },
    {
      "epoch": 11.521708585077983,
      "grad_norm": 0.96946120262146,
      "learning_rate": 5.5472025444556896e-05,
      "loss": 1.5554,
      "step": 164000
    },
    {
      "epoch": 11.556835745398342,
      "grad_norm": 0.9327898621559143,
      "learning_rate": 5.532745409859767e-05,
      "loss": 1.5536,
      "step": 164500
    },
    {
      "epoch": 11.591962905718702,
      "grad_norm": 0.9520469307899475,
      "learning_rate": 5.518288275263843e-05,
      "loss": 1.554,
      "step": 165000
    },
    {
      "epoch": 11.627090066039061,
      "grad_norm": 0.9468493461608887,
      "learning_rate": 5.50383114066792e-05,
      "loss": 1.5541,
      "step": 165500
    },
    {
      "epoch": 11.662217226359422,
      "grad_norm": 0.9519227147102356,
      "learning_rate": 5.4893740060719965e-05,
      "loss": 1.551,
      "step": 166000
    },
    {
      "epoch": 11.69734438667978,
      "grad_norm": 0.9576557278633118,
      "learning_rate": 5.4749168714760736e-05,
      "loss": 1.5519,
      "step": 166500
    },
    {
      "epoch": 11.73247154700014,
      "grad_norm": 0.9346196055412292,
      "learning_rate": 5.46045973688015e-05,
      "loss": 1.549,
      "step": 167000
    },
    {
      "epoch": 11.7675987073205,
      "grad_norm": 0.964485764503479,
      "learning_rate": 5.446002602284227e-05,
      "loss": 1.5494,
      "step": 167500
    },
    {
      "epoch": 11.80272586764086,
      "grad_norm": 0.9761068820953369,
      "learning_rate": 5.431545467688305e-05,
      "loss": 1.551,
      "step": 168000
    },
    {
      "epoch": 11.83785302796122,
      "grad_norm": 0.9633313417434692,
      "learning_rate": 5.417088333092382e-05,
      "loss": 1.5475,
      "step": 168500
    },
    {
      "epoch": 11.872980188281579,
      "grad_norm": 0.9491254091262817,
      "learning_rate": 5.402631198496458e-05,
      "loss": 1.5468,
      "step": 169000
    },
    {
      "epoch": 11.90810734860194,
      "grad_norm": 0.9572640061378479,
      "learning_rate": 5.3881740639005354e-05,
      "loss": 1.5463,
      "step": 169500
    },
    {
      "epoch": 11.943234508922298,
      "grad_norm": 0.9447980523109436,
      "learning_rate": 5.373716929304612e-05,
      "loss": 1.5473,
      "step": 170000
    },
    {
      "epoch": 11.943234508922298,
      "eval_accuracy": 0.7113065905004013,
      "eval_loss": 1.308741807937622,
      "eval_runtime": 147.0083,
      "eval_samples_per_second": 821.083,
      "eval_steps_per_second": 5.136,
      "step": 170000
    },
    {
      "epoch": 11.978361669242659,
      "grad_norm": 0.9711387753486633,
      "learning_rate": 5.359259794708689e-05,
      "loss": 1.5469,
      "step": 170500
    },
    {
      "epoch": 12.013488829563018,
      "grad_norm": 0.9664705395698547,
      "learning_rate": 5.344802660112765e-05,
      "loss": 1.547,
      "step": 171000
    },
    {
      "epoch": 12.048615989883379,
      "grad_norm": 0.9575341939926147,
      "learning_rate": 5.330345525516843e-05,
      "loss": 1.5439,
      "step": 171500
    },
    {
      "epoch": 12.083743150203738,
      "grad_norm": 0.9776381850242615,
      "learning_rate": 5.31588839092092e-05,
      "loss": 1.5416,
      "step": 172000
    },
    {
      "epoch": 12.118870310524096,
      "grad_norm": 0.9849351048469543,
      "learning_rate": 5.3014312563249965e-05,
      "loss": 1.544,
      "step": 172500
    },
    {
      "epoch": 12.153997470844457,
      "grad_norm": 1.0078338384628296,
      "learning_rate": 5.2869741217290736e-05,
      "loss": 1.5463,
      "step": 173000
    },
    {
      "epoch": 12.189124631164816,
      "grad_norm": 0.9730892777442932,
      "learning_rate": 5.272516987133151e-05,
      "loss": 1.5424,
      "step": 173500
    },
    {
      "epoch": 12.224251791485177,
      "grad_norm": 1.0085258483886719,
      "learning_rate": 5.258059852537227e-05,
      "loss": 1.5384,
      "step": 174000
    },
    {
      "epoch": 12.259378951805536,
      "grad_norm": 0.9556441307067871,
      "learning_rate": 5.243602717941304e-05,
      "loss": 1.5427,
      "step": 174500
    },
    {
      "epoch": 12.294506112125896,
      "grad_norm": 0.9524454474449158,
      "learning_rate": 5.229145583345382e-05,
      "loss": 1.5373,
      "step": 175000
    },
    {
      "epoch": 12.329633272446255,
      "grad_norm": 0.9806123375892639,
      "learning_rate": 5.214688448749458e-05,
      "loss": 1.5425,
      "step": 175500
    },
    {
      "epoch": 12.364760432766616,
      "grad_norm": 0.9411491751670837,
      "learning_rate": 5.2002313141535354e-05,
      "loss": 1.5394,
      "step": 176000
    },
    {
      "epoch": 12.399887593086975,
      "grad_norm": 1.0032098293304443,
      "learning_rate": 5.185774179557612e-05,
      "loss": 1.5388,
      "step": 176500
    },
    {
      "epoch": 12.435014753407334,
      "grad_norm": 0.9663287401199341,
      "learning_rate": 5.171317044961689e-05,
      "loss": 1.5389,
      "step": 177000
    },
    {
      "epoch": 12.470141913727694,
      "grad_norm": 0.9497302770614624,
      "learning_rate": 5.156859910365765e-05,
      "loss": 1.537,
      "step": 177500
    },
    {
      "epoch": 12.505269074048053,
      "grad_norm": 0.9728620648384094,
      "learning_rate": 5.1424027757698423e-05,
      "loss": 1.5397,
      "step": 178000
    },
    {
      "epoch": 12.540396234368414,
      "grad_norm": 0.9789900779724121,
      "learning_rate": 5.127945641173919e-05,
      "loss": 1.5405,
      "step": 178500
    },
    {
      "epoch": 12.575523394688773,
      "grad_norm": 0.9774718284606934,
      "learning_rate": 5.113488506577997e-05,
      "loss": 1.5368,
      "step": 179000
    },
    {
      "epoch": 12.610650555009133,
      "grad_norm": 0.9627470970153809,
      "learning_rate": 5.0990313719820736e-05,
      "loss": 1.5365,
      "step": 179500
    },
    {
      "epoch": 12.645777715329492,
      "grad_norm": 0.9805966019630432,
      "learning_rate": 5.0845742373861507e-05,
      "loss": 1.5364,
      "step": 180000
    },
    {
      "epoch": 12.645777715329492,
      "eval_accuracy": 0.7135348973126568,
      "eval_loss": 1.2944310903549194,
      "eval_runtime": 147.0212,
      "eval_samples_per_second": 821.011,
      "eval_steps_per_second": 5.135,
      "step": 180000
    },
    {
      "epoch": 12.680904875649853,
      "grad_norm": 1.0078229904174805,
      "learning_rate": 5.070117102790227e-05,
      "loss": 1.5344,
      "step": 180500
    },
    {
      "epoch": 12.716032035970212,
      "grad_norm": 1.0073761940002441,
      "learning_rate": 5.055659968194304e-05,
      "loss": 1.5357,
      "step": 181000
    },
    {
      "epoch": 12.751159196290573,
      "grad_norm": 0.9848378300666809,
      "learning_rate": 5.0412028335983805e-05,
      "loss": 1.5352,
      "step": 181500
    },
    {
      "epoch": 12.786286356610931,
      "grad_norm": 0.9649278521537781,
      "learning_rate": 5.0267456990024576e-05,
      "loss": 1.5356,
      "step": 182000
    },
    {
      "epoch": 12.82141351693129,
      "grad_norm": 0.9573400616645813,
      "learning_rate": 5.0122885644065354e-05,
      "loss": 1.5363,
      "step": 182500
    },
    {
      "epoch": 12.856540677251651,
      "grad_norm": 0.9978417754173279,
      "learning_rate": 4.997831429810612e-05,
      "loss": 1.5338,
      "step": 183000
    },
    {
      "epoch": 12.89166783757201,
      "grad_norm": 0.99289870262146,
      "learning_rate": 4.983374295214689e-05,
      "loss": 1.5328,
      "step": 183500
    },
    {
      "epoch": 12.92679499789237,
      "grad_norm": 0.974976122379303,
      "learning_rate": 4.968917160618766e-05,
      "loss": 1.5298,
      "step": 184000
    },
    {
      "epoch": 12.96192215821273,
      "grad_norm": 0.9644068479537964,
      "learning_rate": 4.954460026022842e-05,
      "loss": 1.5341,
      "step": 184500
    },
    {
      "epoch": 12.99704931853309,
      "grad_norm": 0.9787358641624451,
      "learning_rate": 4.9400028914269194e-05,
      "loss": 1.532,
      "step": 185000
    },
    {
      "epoch": 13.032176478853449,
      "grad_norm": 0.9854114651679993,
      "learning_rate": 4.9255457568309965e-05,
      "loss": 1.5277,
      "step": 185500
    },
    {
      "epoch": 13.06730363917381,
      "grad_norm": 0.9944697618484497,
      "learning_rate": 4.9110886222350736e-05,
      "loss": 1.5284,
      "step": 186000
    },
    {
      "epoch": 13.102430799494169,
      "grad_norm": 0.9880611300468445,
      "learning_rate": 4.89663148763915e-05,
      "loss": 1.5284,
      "step": 186500
    },
    {
      "epoch": 13.13755795981453,
      "grad_norm": 0.9830440282821655,
      "learning_rate": 4.882174353043227e-05,
      "loss": 1.5251,
      "step": 187000
    },
    {
      "epoch": 13.172685120134888,
      "grad_norm": 0.992473304271698,
      "learning_rate": 4.867717218447304e-05,
      "loss": 1.5256,
      "step": 187500
    },
    {
      "epoch": 13.207812280455247,
      "grad_norm": 0.948597252368927,
      "learning_rate": 4.853260083851381e-05,
      "loss": 1.5271,
      "step": 188000
    },
    {
      "epoch": 13.242939440775608,
      "grad_norm": 0.9829763174057007,
      "learning_rate": 4.8388029492554576e-05,
      "loss": 1.5259,
      "step": 188500
    },
    {
      "epoch": 13.278066601095967,
      "grad_norm": 0.9461808800697327,
      "learning_rate": 4.824345814659535e-05,
      "loss": 1.525,
      "step": 189000
    },
    {
      "epoch": 13.313193761416327,
      "grad_norm": 0.9874758720397949,
      "learning_rate": 4.809888680063612e-05,
      "loss": 1.5286,
      "step": 189500
    },
    {
      "epoch": 13.348320921736686,
      "grad_norm": 0.9801004528999329,
      "learning_rate": 4.795431545467688e-05,
      "loss": 1.5257,
      "step": 190000
    },
    {
      "epoch": 13.348320921736686,
      "eval_accuracy": 0.7146395658479376,
      "eval_loss": 1.2904942035675049,
      "eval_runtime": 148.7015,
      "eval_samples_per_second": 811.734,
      "eval_steps_per_second": 5.077,
      "step": 190000
    },
    {
      "epoch": 13.383448082057047,
      "grad_norm": 1.0048809051513672,
      "learning_rate": 4.780974410871765e-05,
      "loss": 1.5222,
      "step": 190500
    },
    {
      "epoch": 13.418575242377406,
      "grad_norm": 0.9752638339996338,
      "learning_rate": 4.766517276275842e-05,
      "loss": 1.5242,
      "step": 191000
    },
    {
      "epoch": 13.453702402697767,
      "grad_norm": 0.9492470622062683,
      "learning_rate": 4.7520601416799194e-05,
      "loss": 1.5219,
      "step": 191500
    },
    {
      "epoch": 13.488829563018125,
      "grad_norm": 0.973146915435791,
      "learning_rate": 4.737603007083996e-05,
      "loss": 1.5234,
      "step": 192000
    },
    {
      "epoch": 13.523956723338486,
      "grad_norm": 0.9545627236366272,
      "learning_rate": 4.723145872488073e-05,
      "loss": 1.5252,
      "step": 192500
    },
    {
      "epoch": 13.559083883658845,
      "grad_norm": 0.9910150170326233,
      "learning_rate": 4.70868873789215e-05,
      "loss": 1.5229,
      "step": 193000
    },
    {
      "epoch": 13.594211043979204,
      "grad_norm": 1.0169930458068848,
      "learning_rate": 4.694231603296227e-05,
      "loss": 1.5233,
      "step": 193500
    },
    {
      "epoch": 13.629338204299565,
      "grad_norm": 1.0121934413909912,
      "learning_rate": 4.6797744687003034e-05,
      "loss": 1.5201,
      "step": 194000
    },
    {
      "epoch": 13.664465364619923,
      "grad_norm": 0.9987900853157043,
      "learning_rate": 4.665317334104381e-05,
      "loss": 1.5187,
      "step": 194500
    },
    {
      "epoch": 13.699592524940284,
      "grad_norm": 1.0241477489471436,
      "learning_rate": 4.6508601995084576e-05,
      "loss": 1.5206,
      "step": 195000
    },
    {
      "epoch": 13.734719685260643,
      "grad_norm": 0.9895103573799133,
      "learning_rate": 4.6364030649125346e-05,
      "loss": 1.5201,
      "step": 195500
    },
    {
      "epoch": 13.769846845581004,
      "grad_norm": 0.9821479916572571,
      "learning_rate": 4.621945930316611e-05,
      "loss": 1.5201,
      "step": 196000
    },
    {
      "epoch": 13.804974005901363,
      "grad_norm": 1.005090594291687,
      "learning_rate": 4.607488795720689e-05,
      "loss": 1.5182,
      "step": 196500
    },
    {
      "epoch": 13.840101166221723,
      "grad_norm": 0.9969967603683472,
      "learning_rate": 4.593031661124765e-05,
      "loss": 1.5195,
      "step": 197000
    },
    {
      "epoch": 13.875228326542082,
      "grad_norm": 0.9817072153091431,
      "learning_rate": 4.578574526528842e-05,
      "loss": 1.5175,
      "step": 197500
    },
    {
      "epoch": 13.910355486862443,
      "grad_norm": 0.9905675053596497,
      "learning_rate": 4.564117391932919e-05,
      "loss": 1.5182,
      "step": 198000
    },
    {
      "epoch": 13.945482647182802,
      "grad_norm": 0.9947062134742737,
      "learning_rate": 4.5496602573369964e-05,
      "loss": 1.5162,
      "step": 198500
    },
    {
      "epoch": 13.98060980750316,
      "grad_norm": 0.97999107837677,
      "learning_rate": 4.535203122741073e-05,
      "loss": 1.5216,
      "step": 199000
    },
    {
      "epoch": 14.015736967823521,
      "grad_norm": 0.9755523204803467,
      "learning_rate": 4.52074598814515e-05,
      "loss": 1.5161,
      "step": 199500
    },
    {
      "epoch": 14.05086412814388,
      "grad_norm": 0.9752248525619507,
      "learning_rate": 4.506288853549227e-05,
      "loss": 1.5164,
      "step": 200000
    },
    {
      "epoch": 14.05086412814388,
      "eval_accuracy": 0.716148456990617,
      "eval_loss": 1.2788549661636353,
      "eval_runtime": 145.5914,
      "eval_samples_per_second": 829.074,
      "eval_steps_per_second": 5.186,
      "step": 200000
    },
    {
      "epoch": 14.08599128846424,
      "grad_norm": 0.984459638595581,
      "learning_rate": 4.491831718953304e-05,
      "loss": 1.5128,
      "step": 200500
    },
    {
      "epoch": 14.1211184487846,
      "grad_norm": 0.9917986392974854,
      "learning_rate": 4.4773745843573805e-05,
      "loss": 1.5142,
      "step": 201000
    },
    {
      "epoch": 14.15624560910496,
      "grad_norm": 0.9588636159896851,
      "learning_rate": 4.4629174497614575e-05,
      "loss": 1.5122,
      "step": 201500
    },
    {
      "epoch": 14.19137276942532,
      "grad_norm": 0.9958001971244812,
      "learning_rate": 4.4484603151655346e-05,
      "loss": 1.5105,
      "step": 202000
    },
    {
      "epoch": 14.22649992974568,
      "grad_norm": 1.012403130531311,
      "learning_rate": 4.434003180569612e-05,
      "loss": 1.5129,
      "step": 202500
    },
    {
      "epoch": 14.261627090066039,
      "grad_norm": 1.0305534601211548,
      "learning_rate": 4.419546045973688e-05,
      "loss": 1.5133,
      "step": 203000
    },
    {
      "epoch": 14.2967542503864,
      "grad_norm": 0.9992120265960693,
      "learning_rate": 4.405088911377765e-05,
      "loss": 1.5114,
      "step": 203500
    },
    {
      "epoch": 14.331881410706758,
      "grad_norm": 1.008920669555664,
      "learning_rate": 4.390631776781842e-05,
      "loss": 1.5139,
      "step": 204000
    },
    {
      "epoch": 14.367008571027117,
      "grad_norm": 1.0051392316818237,
      "learning_rate": 4.3761746421859187e-05,
      "loss": 1.5112,
      "step": 204500
    },
    {
      "epoch": 14.402135731347478,
      "grad_norm": 0.9850152730941772,
      "learning_rate": 4.361717507589996e-05,
      "loss": 1.5111,
      "step": 205000
    },
    {
      "epoch": 14.437262891667837,
      "grad_norm": 0.9894864559173584,
      "learning_rate": 4.347260372994073e-05,
      "loss": 1.5119,
      "step": 205500
    },
    {
      "epoch": 14.472390051988198,
      "grad_norm": 1.0255316495895386,
      "learning_rate": 4.33280323839815e-05,
      "loss": 1.51,
      "step": 206000
    },
    {
      "epoch": 14.507517212308557,
      "grad_norm": 1.044765830039978,
      "learning_rate": 4.318346103802226e-05,
      "loss": 1.5082,
      "step": 206500
    },
    {
      "epoch": 14.542644372628917,
      "grad_norm": 1.0084095001220703,
      "learning_rate": 4.3038889692063034e-05,
      "loss": 1.5073,
      "step": 207000
    },
    {
      "epoch": 14.577771532949276,
      "grad_norm": 0.9992097616195679,
      "learning_rate": 4.2894318346103804e-05,
      "loss": 1.5085,
      "step": 207500
    },
    {
      "epoch": 14.612898693269637,
      "grad_norm": 1.016776204109192,
      "learning_rate": 4.2749747000144575e-05,
      "loss": 1.5092,
      "step": 208000
    },
    {
      "epoch": 14.648025853589996,
      "grad_norm": 1.0093616247177124,
      "learning_rate": 4.260517565418534e-05,
      "loss": 1.5074,
      "step": 208500
    },
    {
      "epoch": 14.683153013910356,
      "grad_norm": 1.0291818380355835,
      "learning_rate": 4.246060430822611e-05,
      "loss": 1.5079,
      "step": 209000
    },
    {
      "epoch": 14.718280174230715,
      "grad_norm": 0.9756091833114624,
      "learning_rate": 4.231603296226688e-05,
      "loss": 1.5071,
      "step": 209500
    },
    {
      "epoch": 14.753407334551074,
      "grad_norm": 1.0051275491714478,
      "learning_rate": 4.217146161630765e-05,
      "loss": 1.5071,
      "step": 210000
    },
    {
      "epoch": 14.753407334551074,
      "eval_accuracy": 0.7175581891473035,
      "eval_loss": 1.2701668739318848,
      "eval_runtime": 145.5115,
      "eval_samples_per_second": 829.529,
      "eval_steps_per_second": 5.189,
      "step": 210000
    },
    {
      "epoch": 14.788534494871435,
      "grad_norm": 1.011240005493164,
      "learning_rate": 4.2026890270348416e-05,
      "loss": 1.5067,
      "step": 210500
    },
    {
      "epoch": 14.823661655191794,
      "grad_norm": 1.0316245555877686,
      "learning_rate": 4.1882318924389186e-05,
      "loss": 1.5059,
      "step": 211000
    },
    {
      "epoch": 14.858788815512154,
      "grad_norm": 0.9921131730079651,
      "learning_rate": 4.173774757842996e-05,
      "loss": 1.5065,
      "step": 211500
    },
    {
      "epoch": 14.893915975832513,
      "grad_norm": 1.0233694314956665,
      "learning_rate": 4.159317623247073e-05,
      "loss": 1.5039,
      "step": 212000
    },
    {
      "epoch": 14.929043136152874,
      "grad_norm": 1.0409730672836304,
      "learning_rate": 4.144860488651149e-05,
      "loss": 1.507,
      "step": 212500
    },
    {
      "epoch": 14.964170296473233,
      "grad_norm": 1.0045446157455444,
      "learning_rate": 4.130403354055227e-05,
      "loss": 1.5061,
      "step": 213000
    },
    {
      "epoch": 14.999297456793593,
      "grad_norm": 0.9876782298088074,
      "learning_rate": 4.1159462194593033e-05,
      "loss": 1.5067,
      "step": 213500
    },
    {
      "epoch": 15.034424617113952,
      "grad_norm": 1.0133213996887207,
      "learning_rate": 4.1014890848633804e-05,
      "loss": 1.5002,
      "step": 214000
    },
    {
      "epoch": 15.069551777434313,
      "grad_norm": 0.9735503792762756,
      "learning_rate": 4.087031950267457e-05,
      "loss": 1.5016,
      "step": 214500
    },
    {
      "epoch": 15.104678937754672,
      "grad_norm": 0.985035240650177,
      "learning_rate": 4.0725748156715346e-05,
      "loss": 1.5021,
      "step": 215000
    },
    {
      "epoch": 15.13980609807503,
      "grad_norm": 1.0019500255584717,
      "learning_rate": 4.058117681075611e-05,
      "loss": 1.4975,
      "step": 215500
    },
    {
      "epoch": 15.174933258395392,
      "grad_norm": 1.0004901885986328,
      "learning_rate": 4.043660546479688e-05,
      "loss": 1.5017,
      "step": 216000
    },
    {
      "epoch": 15.21006041871575,
      "grad_norm": 0.9935667514801025,
      "learning_rate": 4.0292034118837645e-05,
      "loss": 1.5015,
      "step": 216500
    },
    {
      "epoch": 15.245187579036111,
      "grad_norm": 1.0111830234527588,
      "learning_rate": 4.014746277287842e-05,
      "loss": 1.5004,
      "step": 217000
    },
    {
      "epoch": 15.28031473935647,
      "grad_norm": 1.0003619194030762,
      "learning_rate": 4.0002891426919186e-05,
      "loss": 1.4991,
      "step": 217500
    },
    {
      "epoch": 15.31544189967683,
      "grad_norm": 1.0029890537261963,
      "learning_rate": 3.985832008095996e-05,
      "loss": 1.4992,
      "step": 218000
    },
    {
      "epoch": 15.35056905999719,
      "grad_norm": 1.0191558599472046,
      "learning_rate": 3.971374873500072e-05,
      "loss": 1.5008,
      "step": 218500
    },
    {
      "epoch": 15.38569622031755,
      "grad_norm": 1.01618492603302,
      "learning_rate": 3.95691773890415e-05,
      "loss": 1.4989,
      "step": 219000
    },
    {
      "epoch": 15.42082338063791,
      "grad_norm": 1.020931601524353,
      "learning_rate": 3.942460604308226e-05,
      "loss": 1.4985,
      "step": 219500
    },
    {
      "epoch": 15.45595054095827,
      "grad_norm": 1.0255357027053833,
      "learning_rate": 3.928003469712303e-05,
      "loss": 1.4972,
      "step": 220000
    },
    {
      "epoch": 15.45595054095827,
      "eval_accuracy": 0.7193424504315581,
      "eval_loss": 1.2618342638015747,
      "eval_runtime": 145.885,
      "eval_samples_per_second": 827.405,
      "eval_steps_per_second": 5.175,
      "step": 220000
    },
    {
      "epoch": 15.491077701278629,
      "grad_norm": 1.0047690868377686,
      "learning_rate": 3.9135463351163804e-05,
      "loss": 1.4989,
      "step": 220500
    },
    {
      "epoch": 15.526204861598988,
      "grad_norm": 1.040840983390808,
      "learning_rate": 3.899089200520457e-05,
      "loss": 1.5001,
      "step": 221000
    },
    {
      "epoch": 15.561332021919348,
      "grad_norm": 0.9770060777664185,
      "learning_rate": 3.884632065924534e-05,
      "loss": 1.4976,
      "step": 221500
    },
    {
      "epoch": 15.596459182239707,
      "grad_norm": 0.9793146848678589,
      "learning_rate": 3.870174931328611e-05,
      "loss": 1.4986,
      "step": 222000
    },
    {
      "epoch": 15.631586342560068,
      "grad_norm": 0.9713142514228821,
      "learning_rate": 3.855717796732688e-05,
      "loss": 1.4947,
      "step": 222500
    },
    {
      "epoch": 15.666713502880427,
      "grad_norm": 1.0131899118423462,
      "learning_rate": 3.8412606621367644e-05,
      "loss": 1.4965,
      "step": 223000
    },
    {
      "epoch": 15.701840663200787,
      "grad_norm": 1.0238277912139893,
      "learning_rate": 3.8268035275408415e-05,
      "loss": 1.4961,
      "step": 223500
    },
    {
      "epoch": 15.736967823521146,
      "grad_norm": 1.0393719673156738,
      "learning_rate": 3.812346392944918e-05,
      "loss": 1.4944,
      "step": 224000
    },
    {
      "epoch": 15.772094983841507,
      "grad_norm": 0.993442952632904,
      "learning_rate": 3.797889258348996e-05,
      "loss": 1.4932,
      "step": 224500
    },
    {
      "epoch": 15.807222144161866,
      "grad_norm": 1.010707974433899,
      "learning_rate": 3.783432123753072e-05,
      "loss": 1.4919,
      "step": 225000
    },
    {
      "epoch": 15.842349304482227,
      "grad_norm": 1.0285263061523438,
      "learning_rate": 3.768974989157149e-05,
      "loss": 1.4922,
      "step": 225500
    },
    {
      "epoch": 15.877476464802585,
      "grad_norm": 0.9992517232894897,
      "learning_rate": 3.754517854561226e-05,
      "loss": 1.496,
      "step": 226000
    },
    {
      "epoch": 15.912603625122944,
      "grad_norm": 1.0487091541290283,
      "learning_rate": 3.740060719965303e-05,
      "loss": 1.496,
      "step": 226500
    },
    {
      "epoch": 15.947730785443305,
      "grad_norm": 0.9960684776306152,
      "learning_rate": 3.72560358536938e-05,
      "loss": 1.4916,
      "step": 227000
    },
    {
      "epoch": 15.982857945763664,
      "grad_norm": 1.0461766719818115,
      "learning_rate": 3.711146450773457e-05,
      "loss": 1.4945,
      "step": 227500
    },
    {
      "epoch": 16.017985106084023,
      "grad_norm": 1.058475375175476,
      "learning_rate": 3.696689316177534e-05,
      "loss": 1.4899,
      "step": 228000
    },
    {
      "epoch": 16.053112266404383,
      "grad_norm": 1.0232276916503906,
      "learning_rate": 3.682232181581611e-05,
      "loss": 1.4884,
      "step": 228500
    },
    {
      "epoch": 16.088239426724744,
      "grad_norm": 1.0691484212875366,
      "learning_rate": 3.667775046985687e-05,
      "loss": 1.4886,
      "step": 229000
    },
    {
      "epoch": 16.123366587045105,
      "grad_norm": 1.003165602684021,
      "learning_rate": 3.6533179123897644e-05,
      "loss": 1.4894,
      "step": 229500
    },
    {
      "epoch": 16.158493747365462,
      "grad_norm": 1.010514736175537,
      "learning_rate": 3.6388607777938415e-05,
      "loss": 1.4915,
      "step": 230000
    },
    {
      "epoch": 16.158493747365462,
      "eval_accuracy": 0.7201344726261764,
      "eval_loss": 1.2572591304779053,
      "eval_runtime": 146.9469,
      "eval_samples_per_second": 821.426,
      "eval_steps_per_second": 5.138,
      "step": 230000
    },
    {
      "epoch": 16.193620907685823,
      "grad_norm": 1.0277786254882812,
      "learning_rate": 3.6244036431979186e-05,
      "loss": 1.4867,
      "step": 230500
    },
    {
      "epoch": 16.228748068006183,
      "grad_norm": 1.0123547315597534,
      "learning_rate": 3.609946508601995e-05,
      "loss": 1.4901,
      "step": 231000
    },
    {
      "epoch": 16.26387522832654,
      "grad_norm": 1.0431201457977295,
      "learning_rate": 3.595489374006073e-05,
      "loss": 1.485,
      "step": 231500
    },
    {
      "epoch": 16.2990023886469,
      "grad_norm": 0.9797715544700623,
      "learning_rate": 3.581032239410149e-05,
      "loss": 1.4898,
      "step": 232000
    },
    {
      "epoch": 16.33412954896726,
      "grad_norm": 1.0554850101470947,
      "learning_rate": 3.566575104814226e-05,
      "loss": 1.4867,
      "step": 232500
    },
    {
      "epoch": 16.369256709287622,
      "grad_norm": 1.0041229724884033,
      "learning_rate": 3.5521179702183026e-05,
      "loss": 1.4876,
      "step": 233000
    },
    {
      "epoch": 16.40438386960798,
      "grad_norm": 1.004367470741272,
      "learning_rate": 3.5376608356223804e-05,
      "loss": 1.4866,
      "step": 233500
    },
    {
      "epoch": 16.43951102992834,
      "grad_norm": 0.9991381764411926,
      "learning_rate": 3.523203701026457e-05,
      "loss": 1.4866,
      "step": 234000
    },
    {
      "epoch": 16.4746381902487,
      "grad_norm": 1.0062518119812012,
      "learning_rate": 3.508746566430534e-05,
      "loss": 1.4869,
      "step": 234500
    },
    {
      "epoch": 16.50976535056906,
      "grad_norm": 1.0116221904754639,
      "learning_rate": 3.49428943183461e-05,
      "loss": 1.4867,
      "step": 235000
    },
    {
      "epoch": 16.54489251088942,
      "grad_norm": 1.0057295560836792,
      "learning_rate": 3.479832297238687e-05,
      "loss": 1.4836,
      "step": 235500
    },
    {
      "epoch": 16.58001967120978,
      "grad_norm": 0.9946209788322449,
      "learning_rate": 3.4653751626427644e-05,
      "loss": 1.4871,
      "step": 236000
    },
    {
      "epoch": 16.61514683153014,
      "grad_norm": 1.052956223487854,
      "learning_rate": 3.4509180280468415e-05,
      "loss": 1.4836,
      "step": 236500
    },
    {
      "epoch": 16.650273991850497,
      "grad_norm": 1.0504807233810425,
      "learning_rate": 3.436460893450918e-05,
      "loss": 1.4898,
      "step": 237000
    },
    {
      "epoch": 16.685401152170858,
      "grad_norm": 1.0513533353805542,
      "learning_rate": 3.422003758854995e-05,
      "loss": 1.4857,
      "step": 237500
    },
    {
      "epoch": 16.72052831249122,
      "grad_norm": 1.0254963636398315,
      "learning_rate": 3.407546624259072e-05,
      "loss": 1.484,
      "step": 238000
    },
    {
      "epoch": 16.75565547281158,
      "grad_norm": 1.0302786827087402,
      "learning_rate": 3.393089489663149e-05,
      "loss": 1.4862,
      "step": 238500
    },
    {
      "epoch": 16.790782633131936,
      "grad_norm": 1.0228885412216187,
      "learning_rate": 3.378632355067226e-05,
      "loss": 1.4836,
      "step": 239000
    },
    {
      "epoch": 16.825909793452297,
      "grad_norm": 1.018209457397461,
      "learning_rate": 3.3641752204713026e-05,
      "loss": 1.4827,
      "step": 239500
    },
    {
      "epoch": 16.861036953772658,
      "grad_norm": 0.989748477935791,
      "learning_rate": 3.3497180858753797e-05,
      "loss": 1.4824,
      "step": 240000
    },
    {
      "epoch": 16.861036953772658,
      "eval_accuracy": 0.7210892050293545,
      "eval_loss": 1.2514859437942505,
      "eval_runtime": 145.8765,
      "eval_samples_per_second": 827.453,
      "eval_steps_per_second": 5.176,
      "step": 240000
    },
    {
      "epoch": 16.89616411409302,
      "grad_norm": 1.0294135808944702,
      "learning_rate": 3.335260951279456e-05,
      "loss": 1.4826,
      "step": 240500
    },
    {
      "epoch": 16.931291274413375,
      "grad_norm": 1.06317138671875,
      "learning_rate": 3.320803816683534e-05,
      "loss": 1.4851,
      "step": 241000
    },
    {
      "epoch": 16.966418434733736,
      "grad_norm": 1.072068452835083,
      "learning_rate": 3.30634668208761e-05,
      "loss": 1.4808,
      "step": 241500
    },
    {
      "epoch": 17.001545595054097,
      "grad_norm": 1.0590496063232422,
      "learning_rate": 3.291889547491687e-05,
      "loss": 1.4849,
      "step": 242000
    },
    {
      "epoch": 17.036672755374454,
      "grad_norm": 1.0390037298202515,
      "learning_rate": 3.277432412895764e-05,
      "loss": 1.4826,
      "step": 242500
    },
    {
      "epoch": 17.071799915694815,
      "grad_norm": 1.0264649391174316,
      "learning_rate": 3.2629752782998414e-05,
      "loss": 1.4784,
      "step": 243000
    },
    {
      "epoch": 17.106927076015175,
      "grad_norm": 1.0615930557250977,
      "learning_rate": 3.248518143703918e-05,
      "loss": 1.4799,
      "step": 243500
    },
    {
      "epoch": 17.142054236335536,
      "grad_norm": 1.0263277292251587,
      "learning_rate": 3.234061009107995e-05,
      "loss": 1.4773,
      "step": 244000
    },
    {
      "epoch": 17.177181396655893,
      "grad_norm": 1.016964077949524,
      "learning_rate": 3.219603874512072e-05,
      "loss": 1.4776,
      "step": 244500
    },
    {
      "epoch": 17.212308556976254,
      "grad_norm": 1.0330182313919067,
      "learning_rate": 3.205146739916149e-05,
      "loss": 1.48,
      "step": 245000
    },
    {
      "epoch": 17.247435717296614,
      "grad_norm": 1.0279370546340942,
      "learning_rate": 3.1906896053202255e-05,
      "loss": 1.4767,
      "step": 245500
    },
    {
      "epoch": 17.282562877616975,
      "grad_norm": 1.053545594215393,
      "learning_rate": 3.1762324707243026e-05,
      "loss": 1.4807,
      "step": 246000
    },
    {
      "epoch": 17.317690037937332,
      "grad_norm": 1.0184204578399658,
      "learning_rate": 3.1617753361283796e-05,
      "loss": 1.4776,
      "step": 246500
    },
    {
      "epoch": 17.352817198257693,
      "grad_norm": 1.0180692672729492,
      "learning_rate": 3.147318201532457e-05,
      "loss": 1.4765,
      "step": 247000
    },
    {
      "epoch": 17.387944358578054,
      "grad_norm": 1.03151535987854,
      "learning_rate": 3.132861066936533e-05,
      "loss": 1.4784,
      "step": 247500
    },
    {
      "epoch": 17.42307151889841,
      "grad_norm": 1.0360081195831299,
      "learning_rate": 3.11840393234061e-05,
      "loss": 1.4725,
      "step": 248000
    },
    {
      "epoch": 17.45819867921877,
      "grad_norm": 1.0606461763381958,
      "learning_rate": 3.103946797744687e-05,
      "loss": 1.4753,
      "step": 248500
    },
    {
      "epoch": 17.493325839539132,
      "grad_norm": 1.0544891357421875,
      "learning_rate": 3.0894896631487643e-05,
      "loss": 1.474,
      "step": 249000
    },
    {
      "epoch": 17.528452999859493,
      "grad_norm": 1.0830744504928589,
      "learning_rate": 3.075032528552841e-05,
      "loss": 1.4728,
      "step": 249500
    },
    {
      "epoch": 17.56358016017985,
      "grad_norm": 1.0082223415374756,
      "learning_rate": 3.060575393956918e-05,
      "loss": 1.4748,
      "step": 250000
    },
    {
      "epoch": 17.56358016017985,
      "eval_accuracy": 0.7222705826647003,
      "eval_loss": 1.245027780532837,
      "eval_runtime": 145.2394,
      "eval_samples_per_second": 831.083,
      "eval_steps_per_second": 5.198,
      "step": 250000
    },
    {
      "epoch": 17.59870732050021,
      "grad_norm": 1.0504813194274902,
      "learning_rate": 3.046118259360995e-05,
      "loss": 1.4783,
      "step": 250500
    },
    {
      "epoch": 17.63383448082057,
      "grad_norm": 1.0163605213165283,
      "learning_rate": 3.0316611247650716e-05,
      "loss": 1.4719,
      "step": 251000
    },
    {
      "epoch": 17.668961641140932,
      "grad_norm": 1.0471776723861694,
      "learning_rate": 3.0172039901691484e-05,
      "loss": 1.4745,
      "step": 251500
    },
    {
      "epoch": 17.70408880146129,
      "grad_norm": 1.050743818283081,
      "learning_rate": 3.0027468555732258e-05,
      "loss": 1.4754,
      "step": 252000
    },
    {
      "epoch": 17.73921596178165,
      "grad_norm": 1.0446287393569946,
      "learning_rate": 2.9882897209773025e-05,
      "loss": 1.4742,
      "step": 252500
    },
    {
      "epoch": 17.77434312210201,
      "grad_norm": 1.0279436111450195,
      "learning_rate": 2.9738325863813793e-05,
      "loss": 1.4728,
      "step": 253000
    },
    {
      "epoch": 17.809470282422367,
      "grad_norm": 1.0320532321929932,
      "learning_rate": 2.959375451785456e-05,
      "loss": 1.4751,
      "step": 253500
    },
    {
      "epoch": 17.844597442742728,
      "grad_norm": 1.0268456935882568,
      "learning_rate": 2.9449183171895334e-05,
      "loss": 1.472,
      "step": 254000
    },
    {
      "epoch": 17.87972460306309,
      "grad_norm": 1.0411797761917114,
      "learning_rate": 2.93046118259361e-05,
      "loss": 1.4757,
      "step": 254500
    },
    {
      "epoch": 17.91485176338345,
      "grad_norm": 1.0576931238174438,
      "learning_rate": 2.916004047997687e-05,
      "loss": 1.4738,
      "step": 255000
    },
    {
      "epoch": 17.949978923703807,
      "grad_norm": 1.0698353052139282,
      "learning_rate": 2.9015469134017636e-05,
      "loss": 1.4749,
      "step": 255500
    },
    {
      "epoch": 17.985106084024167,
      "grad_norm": 1.0517598390579224,
      "learning_rate": 2.887089778805841e-05,
      "loss": 1.4719,
      "step": 256000
    },
    {
      "epoch": 18.020233244344528,
      "grad_norm": 1.030639886856079,
      "learning_rate": 2.8726326442099178e-05,
      "loss": 1.4722,
      "step": 256500
    },
    {
      "epoch": 18.05536040466489,
      "grad_norm": 1.038255214691162,
      "learning_rate": 2.8581755096139945e-05,
      "loss": 1.4713,
      "step": 257000
    },
    {
      "epoch": 18.090487564985246,
      "grad_norm": 1.0348975658416748,
      "learning_rate": 2.843718375018072e-05,
      "loss": 1.4678,
      "step": 257500
    },
    {
      "epoch": 18.125614725305606,
      "grad_norm": 1.0541919469833374,
      "learning_rate": 2.8292612404221487e-05,
      "loss": 1.4711,
      "step": 258000
    },
    {
      "epoch": 18.160741885625967,
      "grad_norm": 1.0226538181304932,
      "learning_rate": 2.8148041058262254e-05,
      "loss": 1.4693,
      "step": 258500
    },
    {
      "epoch": 18.195869045946324,
      "grad_norm": 1.0200276374816895,
      "learning_rate": 2.800346971230302e-05,
      "loss": 1.4705,
      "step": 259000
    },
    {
      "epoch": 18.230996206266685,
      "grad_norm": 1.0429800748825073,
      "learning_rate": 2.7858898366343792e-05,
      "loss": 1.4704,
      "step": 259500
    },
    {
      "epoch": 18.266123366587045,
      "grad_norm": 1.0760138034820557,
      "learning_rate": 2.7714327020384563e-05,
      "loss": 1.4686,
      "step": 260000
    },
    {
      "epoch": 18.266123366587045,
      "eval_accuracy": 0.7233516160401716,
      "eval_loss": 1.238907814025879,
      "eval_runtime": 142.7876,
      "eval_samples_per_second": 845.353,
      "eval_steps_per_second": 5.288,
      "step": 260000
    },
    {
      "epoch": 18.301250526907406,
      "grad_norm": 1.0434221029281616,
      "learning_rate": 2.756975567442533e-05,
      "loss": 1.468,
      "step": 260500
    },
    {
      "epoch": 18.336377687227763,
      "grad_norm": 1.0621415376663208,
      "learning_rate": 2.7425184328466098e-05,
      "loss": 1.4665,
      "step": 261000
    },
    {
      "epoch": 18.371504847548124,
      "grad_norm": 1.0416877269744873,
      "learning_rate": 2.728061298250687e-05,
      "loss": 1.4703,
      "step": 261500
    },
    {
      "epoch": 18.406632007868485,
      "grad_norm": 1.0364477634429932,
      "learning_rate": 2.7136041636547636e-05,
      "loss": 1.4679,
      "step": 262000
    },
    {
      "epoch": 18.441759168188845,
      "grad_norm": 1.0509207248687744,
      "learning_rate": 2.6991470290588407e-05,
      "loss": 1.4645,
      "step": 262500
    },
    {
      "epoch": 18.476886328509202,
      "grad_norm": 1.0550258159637451,
      "learning_rate": 2.6846898944629178e-05,
      "loss": 1.465,
      "step": 263000
    },
    {
      "epoch": 18.512013488829563,
      "grad_norm": 1.045780897140503,
      "learning_rate": 2.6702327598669945e-05,
      "loss": 1.4645,
      "step": 263500
    },
    {
      "epoch": 18.547140649149924,
      "grad_norm": 1.033547043800354,
      "learning_rate": 2.6557756252710713e-05,
      "loss": 1.464,
      "step": 264000
    },
    {
      "epoch": 18.58226780947028,
      "grad_norm": 1.0473381280899048,
      "learning_rate": 2.641318490675148e-05,
      "loss": 1.4647,
      "step": 264500
    },
    {
      "epoch": 18.61739496979064,
      "grad_norm": 1.0651229619979858,
      "learning_rate": 2.6268613560792254e-05,
      "loss": 1.4683,
      "step": 265000
    },
    {
      "epoch": 18.652522130111002,
      "grad_norm": 1.0227559804916382,
      "learning_rate": 2.612404221483302e-05,
      "loss": 1.4653,
      "step": 265500
    },
    {
      "epoch": 18.687649290431363,
      "grad_norm": 1.0518014430999756,
      "learning_rate": 2.597947086887379e-05,
      "loss": 1.4634,
      "step": 266000
    },
    {
      "epoch": 18.72277645075172,
      "grad_norm": 1.0625571012496948,
      "learning_rate": 2.5834899522914556e-05,
      "loss": 1.4663,
      "step": 266500
    },
    {
      "epoch": 18.75790361107208,
      "grad_norm": 1.0320320129394531,
      "learning_rate": 2.569032817695533e-05,
      "loss": 1.4658,
      "step": 267000
    },
    {
      "epoch": 18.79303077139244,
      "grad_norm": 1.0507465600967407,
      "learning_rate": 2.5545756830996098e-05,
      "loss": 1.4667,
      "step": 267500
    },
    {
      "epoch": 18.828157931712802,
      "grad_norm": 1.0624668598175049,
      "learning_rate": 2.5401185485036865e-05,
      "loss": 1.465,
      "step": 268000
    },
    {
      "epoch": 18.86328509203316,
      "grad_norm": 1.1198619604110718,
      "learning_rate": 2.5256614139077633e-05,
      "loss": 1.4639,
      "step": 268500
    },
    {
      "epoch": 18.89841225235352,
      "grad_norm": 1.0421173572540283,
      "learning_rate": 2.5112042793118407e-05,
      "loss": 1.4641,
      "step": 269000
    },
    {
      "epoch": 18.93353941267388,
      "grad_norm": 1.0857394933700562,
      "learning_rate": 2.4967471447159174e-05,
      "loss": 1.4633,
      "step": 269500
    },
    {
      "epoch": 18.968666572994238,
      "grad_norm": 1.0285334587097168,
      "learning_rate": 2.4822900101199945e-05,
      "loss": 1.4649,
      "step": 270000
    },
    {
      "epoch": 18.968666572994238,
      "eval_accuracy": 0.7242654194082216,
      "eval_loss": 1.2332816123962402,
      "eval_runtime": 141.3716,
      "eval_samples_per_second": 853.821,
      "eval_steps_per_second": 5.341,
      "step": 270000
    },
    {
      "epoch": 19.0037937333146,
      "grad_norm": 1.0661959648132324,
      "learning_rate": 2.4678328755240712e-05,
      "loss": 1.464,
      "step": 270500
    },
    {
      "epoch": 19.03892089363496,
      "grad_norm": 1.0679501295089722,
      "learning_rate": 2.4533757409281483e-05,
      "loss": 1.4604,
      "step": 271000
    },
    {
      "epoch": 19.07404805395532,
      "grad_norm": 1.0748372077941895,
      "learning_rate": 2.438918606332225e-05,
      "loss": 1.4611,
      "step": 271500
    },
    {
      "epoch": 19.109175214275677,
      "grad_norm": 1.0354524850845337,
      "learning_rate": 2.424461471736302e-05,
      "loss": 1.4604,
      "step": 272000
    },
    {
      "epoch": 19.144302374596037,
      "grad_norm": 1.0497136116027832,
      "learning_rate": 2.410004337140379e-05,
      "loss": 1.4616,
      "step": 272500
    },
    {
      "epoch": 19.179429534916398,
      "grad_norm": 1.0558326244354248,
      "learning_rate": 2.395547202544456e-05,
      "loss": 1.4611,
      "step": 273000
    },
    {
      "epoch": 19.21455669523676,
      "grad_norm": 1.0470489263534546,
      "learning_rate": 2.3810900679485327e-05,
      "loss": 1.4575,
      "step": 273500
    },
    {
      "epoch": 19.249683855557116,
      "grad_norm": 1.0182609558105469,
      "learning_rate": 2.3666329333526098e-05,
      "loss": 1.4594,
      "step": 274000
    },
    {
      "epoch": 19.284811015877477,
      "grad_norm": 1.0594130754470825,
      "learning_rate": 2.3521757987566865e-05,
      "loss": 1.4582,
      "step": 274500
    },
    {
      "epoch": 19.319938176197837,
      "grad_norm": 1.0967395305633545,
      "learning_rate": 2.3377186641607636e-05,
      "loss": 1.4607,
      "step": 275000
    },
    {
      "epoch": 19.355065336518194,
      "grad_norm": 1.0475187301635742,
      "learning_rate": 2.3232615295648403e-05,
      "loss": 1.4617,
      "step": 275500
    },
    {
      "epoch": 19.390192496838555,
      "grad_norm": 1.0457967519760132,
      "learning_rate": 2.3088043949689174e-05,
      "loss": 1.4569,
      "step": 276000
    },
    {
      "epoch": 19.425319657158916,
      "grad_norm": 1.0595502853393555,
      "learning_rate": 2.2943472603729945e-05,
      "loss": 1.4609,
      "step": 276500
    },
    {
      "epoch": 19.460446817479276,
      "grad_norm": 1.0646470785140991,
      "learning_rate": 2.2798901257770712e-05,
      "loss": 1.4572,
      "step": 277000
    },
    {
      "epoch": 19.495573977799634,
      "grad_norm": 1.0434619188308716,
      "learning_rate": 2.265432991181148e-05,
      "loss": 1.4597,
      "step": 277500
    },
    {
      "epoch": 19.530701138119994,
      "grad_norm": 1.039019227027893,
      "learning_rate": 2.250975856585225e-05,
      "loss": 1.4576,
      "step": 278000
    },
    {
      "epoch": 19.565828298440355,
      "grad_norm": 1.0945369005203247,
      "learning_rate": 2.2365187219893018e-05,
      "loss": 1.459,
      "step": 278500
    },
    {
      "epoch": 19.600955458760716,
      "grad_norm": 1.0497277975082397,
      "learning_rate": 2.2220615873933785e-05,
      "loss": 1.4555,
      "step": 279000
    },
    {
      "epoch": 19.636082619081073,
      "grad_norm": 1.0733200311660767,
      "learning_rate": 2.2076044527974556e-05,
      "loss": 1.4586,
      "step": 279500
    },
    {
      "epoch": 19.671209779401433,
      "grad_norm": 1.0626968145370483,
      "learning_rate": 2.1931473182015323e-05,
      "loss": 1.4566,
      "step": 280000
    },
    {
      "epoch": 19.671209779401433,
      "eval_accuracy": 0.7253028435871818,
      "eval_loss": 1.2285022735595703,
      "eval_runtime": 141.7716,
      "eval_samples_per_second": 851.412,
      "eval_steps_per_second": 5.325,
      "step": 280000
    },
    {
      "epoch": 19.706336939721794,
      "grad_norm": 1.0224528312683105,
      "learning_rate": 2.1786901836056094e-05,
      "loss": 1.4583,
      "step": 280500
    },
    {
      "epoch": 19.74146410004215,
      "grad_norm": 1.0466312170028687,
      "learning_rate": 2.164233049009686e-05,
      "loss": 1.457,
      "step": 281000
    },
    {
      "epoch": 19.776591260362512,
      "grad_norm": 1.0637938976287842,
      "learning_rate": 2.1497759144137632e-05,
      "loss": 1.4577,
      "step": 281500
    },
    {
      "epoch": 19.811718420682872,
      "grad_norm": 1.0693141222000122,
      "learning_rate": 2.1353187798178403e-05,
      "loss": 1.4566,
      "step": 282000
    },
    {
      "epoch": 19.846845581003233,
      "grad_norm": 1.0720393657684326,
      "learning_rate": 2.120861645221917e-05,
      "loss": 1.456,
      "step": 282500
    },
    {
      "epoch": 19.88197274132359,
      "grad_norm": 1.071040153503418,
      "learning_rate": 2.106404510625994e-05,
      "loss": 1.4546,
      "step": 283000
    },
    {
      "epoch": 19.91709990164395,
      "grad_norm": 1.0675835609436035,
      "learning_rate": 2.091947376030071e-05,
      "loss": 1.4582,
      "step": 283500
    },
    {
      "epoch": 19.95222706196431,
      "grad_norm": 1.0603220462799072,
      "learning_rate": 2.077490241434148e-05,
      "loss": 1.4547,
      "step": 284000
    },
    {
      "epoch": 19.987354222284672,
      "grad_norm": 1.0802963972091675,
      "learning_rate": 2.0630331068382247e-05,
      "loss": 1.4546,
      "step": 284500
    },
    {
      "epoch": 20.02248138260503,
      "grad_norm": 1.0588197708129883,
      "learning_rate": 2.0485759722423017e-05,
      "loss": 1.4575,
      "step": 285000
    },
    {
      "epoch": 20.05760854292539,
      "grad_norm": 1.0602234601974487,
      "learning_rate": 2.0341188376463785e-05,
      "loss": 1.4521,
      "step": 285500
    },
    {
      "epoch": 20.09273570324575,
      "grad_norm": 1.1232300996780396,
      "learning_rate": 2.0196617030504556e-05,
      "loss": 1.4517,
      "step": 286000
    },
    {
      "epoch": 20.127862863566108,
      "grad_norm": 1.0660282373428345,
      "learning_rate": 2.0052045684545323e-05,
      "loss": 1.4523,
      "step": 286500
    },
    {
      "epoch": 20.16299002388647,
      "grad_norm": 1.0167709589004517,
      "learning_rate": 1.9907474338586094e-05,
      "loss": 1.4519,
      "step": 287000
    },
    {
      "epoch": 20.19811718420683,
      "grad_norm": 1.047339677810669,
      "learning_rate": 1.976290299262686e-05,
      "loss": 1.4502,
      "step": 287500
    },
    {
      "epoch": 20.23324434452719,
      "grad_norm": 1.0285905599594116,
      "learning_rate": 1.9618331646667632e-05,
      "loss": 1.4514,
      "step": 288000
    },
    {
      "epoch": 20.268371504847547,
      "grad_norm": 1.05581533908844,
      "learning_rate": 1.9473760300708403e-05,
      "loss": 1.449,
      "step": 288500
    },
    {
      "epoch": 20.303498665167908,
      "grad_norm": 1.0714962482452393,
      "learning_rate": 1.932918895474917e-05,
      "loss": 1.4521,
      "step": 289000
    },
    {
      "epoch": 20.33862582548827,
      "grad_norm": 1.0655773878097534,
      "learning_rate": 1.918461760878994e-05,
      "loss": 1.4508,
      "step": 289500
    },
    {
      "epoch": 20.37375298580863,
      "grad_norm": 1.0571376085281372,
      "learning_rate": 1.9040046262830708e-05,
      "loss": 1.4529,
      "step": 290000
    },
    {
      "epoch": 20.37375298580863,
      "eval_accuracy": 0.7261482132790039,
      "eval_loss": 1.223044753074646,
      "eval_runtime": 141.5948,
      "eval_samples_per_second": 852.475,
      "eval_steps_per_second": 5.332,
      "step": 290000
    },
    {
      "epoch": 20.408880146128986,
      "grad_norm": 1.0565483570098877,
      "learning_rate": 1.889547491687148e-05,
      "loss": 1.4504,
      "step": 290500
    },
    {
      "epoch": 20.444007306449347,
      "grad_norm": 1.0659668445587158,
      "learning_rate": 1.8750903570912246e-05,
      "loss": 1.4525,
      "step": 291000
    },
    {
      "epoch": 20.479134466769708,
      "grad_norm": 1.122076153755188,
      "learning_rate": 1.8606332224953017e-05,
      "loss": 1.4485,
      "step": 291500
    },
    {
      "epoch": 20.514261627090065,
      "grad_norm": 1.0631004571914673,
      "learning_rate": 1.8461760878993785e-05,
      "loss": 1.4513,
      "step": 292000
    },
    {
      "epoch": 20.549388787410425,
      "grad_norm": 1.0630178451538086,
      "learning_rate": 1.8317189533034555e-05,
      "loss": 1.4486,
      "step": 292500
    },
    {
      "epoch": 20.584515947730786,
      "grad_norm": 1.0705844163894653,
      "learning_rate": 1.8172618187075323e-05,
      "loss": 1.4503,
      "step": 293000
    },
    {
      "epoch": 20.619643108051147,
      "grad_norm": 1.0725988149642944,
      "learning_rate": 1.8028046841116093e-05,
      "loss": 1.4515,
      "step": 293500
    },
    {
      "epoch": 20.654770268371504,
      "grad_norm": 1.0657289028167725,
      "learning_rate": 1.788347549515686e-05,
      "loss": 1.4475,
      "step": 294000
    },
    {
      "epoch": 20.689897428691864,
      "grad_norm": 1.062153697013855,
      "learning_rate": 1.7738904149197628e-05,
      "loss": 1.4503,
      "step": 294500
    },
    {
      "epoch": 20.725024589012225,
      "grad_norm": 1.0483520030975342,
      "learning_rate": 1.75943328032384e-05,
      "loss": 1.4492,
      "step": 295000
    },
    {
      "epoch": 20.760151749332586,
      "grad_norm": 1.07576322555542,
      "learning_rate": 1.7449761457279166e-05,
      "loss": 1.4514,
      "step": 295500
    },
    {
      "epoch": 20.795278909652943,
      "grad_norm": 1.0874416828155518,
      "learning_rate": 1.7305190111319937e-05,
      "loss": 1.4463,
      "step": 296000
    },
    {
      "epoch": 20.830406069973304,
      "grad_norm": 1.0873123407363892,
      "learning_rate": 1.7160618765360705e-05,
      "loss": 1.4483,
      "step": 296500
    },
    {
      "epoch": 20.865533230293664,
      "grad_norm": 1.0750373601913452,
      "learning_rate": 1.7016047419401475e-05,
      "loss": 1.4461,
      "step": 297000
    },
    {
      "epoch": 20.90066039061402,
      "grad_norm": 1.0832915306091309,
      "learning_rate": 1.6871476073442243e-05,
      "loss": 1.45,
      "step": 297500
    },
    {
      "epoch": 20.935787550934382,
      "grad_norm": 1.0408779382705688,
      "learning_rate": 1.6726904727483014e-05,
      "loss": 1.4483,
      "step": 298000
    },
    {
      "epoch": 20.970914711254743,
      "grad_norm": 1.0541013479232788,
      "learning_rate": 1.658233338152378e-05,
      "loss": 1.4492,
      "step": 298500
    },
    {
      "epoch": 21.006041871575103,
      "grad_norm": 1.0725327730178833,
      "learning_rate": 1.6437762035564552e-05,
      "loss": 1.4475,
      "step": 299000
    },
    {
      "epoch": 21.04116903189546,
      "grad_norm": 1.0820538997650146,
      "learning_rate": 1.629319068960532e-05,
      "loss": 1.4481,
      "step": 299500
    },
    {
      "epoch": 21.07629619221582,
      "grad_norm": 1.0279037952423096,
      "learning_rate": 1.614861934364609e-05,
      "loss": 1.4451,
      "step": 300000
    },
    {
      "epoch": 21.07629619221582,
      "eval_accuracy": 0.7268696778575053,
      "eval_loss": 1.2188650369644165,
      "eval_runtime": 142.0191,
      "eval_samples_per_second": 849.928,
      "eval_steps_per_second": 5.316,
      "step": 300000
    },
    {
      "epoch": 21.111423352536182,
      "grad_norm": 1.034382700920105,
      "learning_rate": 1.6004047997686857e-05,
      "loss": 1.4447,
      "step": 300500
    },
    {
      "epoch": 21.14655051285654,
      "grad_norm": 1.0651057958602905,
      "learning_rate": 1.5859476651727628e-05,
      "loss": 1.4467,
      "step": 301000
    },
    {
      "epoch": 21.1816776731769,
      "grad_norm": 1.086431622505188,
      "learning_rate": 1.57149053057684e-05,
      "loss": 1.4452,
      "step": 301500
    },
    {
      "epoch": 21.21680483349726,
      "grad_norm": 1.0573840141296387,
      "learning_rate": 1.5570333959809166e-05,
      "loss": 1.4487,
      "step": 302000
    },
    {
      "epoch": 21.25193199381762,
      "grad_norm": 1.0548052787780762,
      "learning_rate": 1.5425762613849937e-05,
      "loss": 1.4424,
      "step": 302500
    },
    {
      "epoch": 21.287059154137978,
      "grad_norm": 1.0701725482940674,
      "learning_rate": 1.5281191267890704e-05,
      "loss": 1.4441,
      "step": 303000
    },
    {
      "epoch": 21.32218631445834,
      "grad_norm": 1.0664572715759277,
      "learning_rate": 1.5136619921931475e-05,
      "loss": 1.4449,
      "step": 303500
    },
    {
      "epoch": 21.3573134747787,
      "grad_norm": 1.0336108207702637,
      "learning_rate": 1.4992048575972243e-05,
      "loss": 1.4415,
      "step": 304000
    },
    {
      "epoch": 21.39244063509906,
      "grad_norm": 1.098109245300293,
      "learning_rate": 1.4847477230013013e-05,
      "loss": 1.4453,
      "step": 304500
    },
    {
      "epoch": 21.427567795419417,
      "grad_norm": 1.0706530809402466,
      "learning_rate": 1.470290588405378e-05,
      "loss": 1.4433,
      "step": 305000
    },
    {
      "epoch": 21.462694955739778,
      "grad_norm": 1.086916208267212,
      "learning_rate": 1.4558334538094551e-05,
      "loss": 1.4441,
      "step": 305500
    },
    {
      "epoch": 21.49782211606014,
      "grad_norm": 1.1005213260650635,
      "learning_rate": 1.4413763192135319e-05,
      "loss": 1.4429,
      "step": 306000
    },
    {
      "epoch": 21.532949276380496,
      "grad_norm": 1.1273337602615356,
      "learning_rate": 1.4269191846176088e-05,
      "loss": 1.4429,
      "step": 306500
    },
    {
      "epoch": 21.568076436700856,
      "grad_norm": 1.092872142791748,
      "learning_rate": 1.4124620500216857e-05,
      "loss": 1.4417,
      "step": 307000
    },
    {
      "epoch": 21.603203597021217,
      "grad_norm": 1.0761197805404663,
      "learning_rate": 1.3980049154257626e-05,
      "loss": 1.4427,
      "step": 307500
    },
    {
      "epoch": 21.638330757341578,
      "grad_norm": 1.1086857318878174,
      "learning_rate": 1.3835477808298397e-05,
      "loss": 1.447,
      "step": 308000
    },
    {
      "epoch": 21.673457917661935,
      "grad_norm": 1.091773271560669,
      "learning_rate": 1.3690906462339164e-05,
      "loss": 1.4437,
      "step": 308500
    },
    {
      "epoch": 21.708585077982296,
      "grad_norm": 1.0596261024475098,
      "learning_rate": 1.3546335116379935e-05,
      "loss": 1.4421,
      "step": 309000
    },
    {
      "epoch": 21.743712238302656,
      "grad_norm": 1.0684653520584106,
      "learning_rate": 1.3401763770420702e-05,
      "loss": 1.4407,
      "step": 309500
    },
    {
      "epoch": 21.778839398623017,
      "grad_norm": 1.0676227807998657,
      "learning_rate": 1.3257192424461473e-05,
      "loss": 1.443,
      "step": 310000
    },
    {
      "epoch": 21.778839398623017,
      "eval_accuracy": 0.7277766831400472,
      "eval_loss": 1.2135990858078003,
      "eval_runtime": 142.0609,
      "eval_samples_per_second": 849.678,
      "eval_steps_per_second": 5.315,
      "step": 310000
    },
    {
      "epoch": 21.813966558943374,
      "grad_norm": 1.0764987468719482,
      "learning_rate": 1.311262107850224e-05,
      "loss": 1.4392,
      "step": 310500
    },
    {
      "epoch": 21.849093719263735,
      "grad_norm": 1.075104832649231,
      "learning_rate": 1.2968049732543011e-05,
      "loss": 1.439,
      "step": 311000
    },
    {
      "epoch": 21.884220879584095,
      "grad_norm": 1.0967333316802979,
      "learning_rate": 1.2823478386583779e-05,
      "loss": 1.4428,
      "step": 311500
    },
    {
      "epoch": 21.919348039904452,
      "grad_norm": 1.0619230270385742,
      "learning_rate": 1.267890704062455e-05,
      "loss": 1.4394,
      "step": 312000
    },
    {
      "epoch": 21.954475200224813,
      "grad_norm": 1.074622631072998,
      "learning_rate": 1.2534335694665317e-05,
      "loss": 1.4416,
      "step": 312500
    },
    {
      "epoch": 21.989602360545174,
      "grad_norm": 1.1134517192840576,
      "learning_rate": 1.2389764348706088e-05,
      "loss": 1.4424,
      "step": 313000
    },
    {
      "epoch": 22.024729520865534,
      "grad_norm": 1.0811127424240112,
      "learning_rate": 1.2245193002746857e-05,
      "loss": 1.4396,
      "step": 313500
    },
    {
      "epoch": 22.05985668118589,
      "grad_norm": 1.0637929439544678,
      "learning_rate": 1.2100621656787626e-05,
      "loss": 1.4368,
      "step": 314000
    },
    {
      "epoch": 22.094983841506252,
      "grad_norm": 1.0761417150497437,
      "learning_rate": 1.1956050310828395e-05,
      "loss": 1.4403,
      "step": 314500
    },
    {
      "epoch": 22.130111001826613,
      "grad_norm": 1.099725604057312,
      "learning_rate": 1.1811478964869162e-05,
      "loss": 1.4378,
      "step": 315000
    },
    {
      "epoch": 22.165238162146974,
      "grad_norm": 1.0871621370315552,
      "learning_rate": 1.1666907618909931e-05,
      "loss": 1.4362,
      "step": 315500
    },
    {
      "epoch": 22.20036532246733,
      "grad_norm": 1.057110071182251,
      "learning_rate": 1.15223362729507e-05,
      "loss": 1.4365,
      "step": 316000
    },
    {
      "epoch": 22.23549248278769,
      "grad_norm": 1.1010171175003052,
      "learning_rate": 1.137776492699147e-05,
      "loss": 1.4387,
      "step": 316500
    },
    {
      "epoch": 22.270619643108052,
      "grad_norm": 1.058934211730957,
      "learning_rate": 1.123319358103224e-05,
      "loss": 1.4394,
      "step": 317000
    },
    {
      "epoch": 22.30574680342841,
      "grad_norm": 1.0834625959396362,
      "learning_rate": 1.108862223507301e-05,
      "loss": 1.4376,
      "step": 317500
    },
    {
      "epoch": 22.34087396374877,
      "grad_norm": 1.0423885583877563,
      "learning_rate": 1.0944050889113779e-05,
      "loss": 1.4372,
      "step": 318000
    },
    {
      "epoch": 22.37600112406913,
      "grad_norm": 1.0966408252716064,
      "learning_rate": 1.0799479543154548e-05,
      "loss": 1.4394,
      "step": 318500
    },
    {
      "epoch": 22.41112828438949,
      "grad_norm": 1.0695067644119263,
      "learning_rate": 1.0654908197195317e-05,
      "loss": 1.4362,
      "step": 319000
    },
    {
      "epoch": 22.44625544470985,
      "grad_norm": 1.134260892868042,
      "learning_rate": 1.0510336851236086e-05,
      "loss": 1.435,
      "step": 319500
    },
    {
      "epoch": 22.48138260503021,
      "grad_norm": 1.1000449657440186,
      "learning_rate": 1.0365765505276855e-05,
      "loss": 1.4357,
      "step": 320000
    },
    {
      "epoch": 22.48138260503021,
      "eval_accuracy": 0.7284053903950559,
      "eval_loss": 1.210018277168274,
      "eval_runtime": 142.872,
      "eval_samples_per_second": 844.854,
      "eval_steps_per_second": 5.284,
      "step": 320000
    },
    {
      "epoch": 22.51650976535057,
      "grad_norm": 1.0945639610290527,
      "learning_rate": 1.0221194159317624e-05,
      "loss": 1.4366,
      "step": 320500
    },
    {
      "epoch": 22.55163692567093,
      "grad_norm": 1.1207478046417236,
      "learning_rate": 1.0076622813358393e-05,
      "loss": 1.4399,
      "step": 321000
    },
    {
      "epoch": 22.586764085991287,
      "grad_norm": 1.0734148025512695,
      "learning_rate": 9.932051467399162e-06,
      "loss": 1.4385,
      "step": 321500
    },
    {
      "epoch": 22.621891246311648,
      "grad_norm": 1.0629650354385376,
      "learning_rate": 9.787480121439931e-06,
      "loss": 1.4352,
      "step": 322000
    },
    {
      "epoch": 22.65701840663201,
      "grad_norm": 1.0973799228668213,
      "learning_rate": 9.6429087754807e-06,
      "loss": 1.4377,
      "step": 322500
    },
    {
      "epoch": 22.692145566952366,
      "grad_norm": 1.0496876239776611,
      "learning_rate": 9.49833742952147e-06,
      "loss": 1.4372,
      "step": 323000
    },
    {
      "epoch": 22.727272727272727,
      "grad_norm": 1.0563302040100098,
      "learning_rate": 9.353766083562238e-06,
      "loss": 1.4352,
      "step": 323500
    },
    {
      "epoch": 22.762399887593087,
      "grad_norm": 1.1020653247833252,
      "learning_rate": 9.209194737603008e-06,
      "loss": 1.4331,
      "step": 324000
    },
    {
      "epoch": 22.797527047913448,
      "grad_norm": 1.0854381322860718,
      "learning_rate": 9.064623391643777e-06,
      "loss": 1.4371,
      "step": 324500
    },
    {
      "epoch": 22.832654208233805,
      "grad_norm": 1.0770697593688965,
      "learning_rate": 8.920052045684546e-06,
      "loss": 1.4361,
      "step": 325000
    },
    {
      "epoch": 22.867781368554166,
      "grad_norm": 1.0986183881759644,
      "learning_rate": 8.775480699725315e-06,
      "loss": 1.4354,
      "step": 325500
    },
    {
      "epoch": 22.902908528874526,
      "grad_norm": 1.1258665323257446,
      "learning_rate": 8.630909353766084e-06,
      "loss": 1.4376,
      "step": 326000
    },
    {
      "epoch": 22.938035689194887,
      "grad_norm": 1.0938217639923096,
      "learning_rate": 8.486338007806853e-06,
      "loss": 1.4346,
      "step": 326500
    },
    {
      "epoch": 22.973162849515244,
      "grad_norm": 1.1074005365371704,
      "learning_rate": 8.341766661847622e-06,
      "loss": 1.4306,
      "step": 327000
    },
    {
      "epoch": 23.008290009835605,
      "grad_norm": 1.0893038511276245,
      "learning_rate": 8.197195315888391e-06,
      "loss": 1.4346,
      "step": 327500
    },
    {
      "epoch": 23.043417170155966,
      "grad_norm": 1.1042568683624268,
      "learning_rate": 8.05262396992916e-06,
      "loss": 1.4335,
      "step": 328000
    },
    {
      "epoch": 23.078544330476323,
      "grad_norm": 1.0980304479599,
      "learning_rate": 7.90805262396993e-06,
      "loss": 1.43,
      "step": 328500
    },
    {
      "epoch": 23.113671490796683,
      "grad_norm": 1.0918561220169067,
      "learning_rate": 7.763481278010698e-06,
      "loss": 1.4333,
      "step": 329000
    },
    {
      "epoch": 23.148798651117044,
      "grad_norm": 1.1151840686798096,
      "learning_rate": 7.618909932051468e-06,
      "loss": 1.4346,
      "step": 329500
    },
    {
      "epoch": 23.183925811437405,
      "grad_norm": 1.0970309972763062,
      "learning_rate": 7.474338586092237e-06,
      "loss": 1.4327,
      "step": 330000
    },
    {
      "epoch": 23.183925811437405,
      "eval_accuracy": 0.7289820707456471,
      "eval_loss": 1.2067663669586182,
      "eval_runtime": 141.4523,
      "eval_samples_per_second": 853.333,
      "eval_steps_per_second": 5.337,
      "step": 330000
    },
    {
      "epoch": 23.219052971757762,
      "grad_norm": 1.1229480504989624,
      "learning_rate": 7.3297672401330065e-06,
      "loss": 1.429,
      "step": 330500
    },
    {
      "epoch": 23.254180132078123,
      "grad_norm": 1.103352665901184,
      "learning_rate": 7.1851958941737756e-06,
      "loss": 1.4332,
      "step": 331000
    },
    {
      "epoch": 23.289307292398483,
      "grad_norm": 1.0836817026138306,
      "learning_rate": 7.040624548214545e-06,
      "loss": 1.4331,
      "step": 331500
    },
    {
      "epoch": 23.324434452718844,
      "grad_norm": 1.07485830783844,
      "learning_rate": 6.896053202255314e-06,
      "loss": 1.4351,
      "step": 332000
    },
    {
      "epoch": 23.3595616130392,
      "grad_norm": 1.0870476961135864,
      "learning_rate": 6.751481856296083e-06,
      "loss": 1.4328,
      "step": 332500
    },
    {
      "epoch": 23.39468877335956,
      "grad_norm": 1.0797423124313354,
      "learning_rate": 6.606910510336851e-06,
      "loss": 1.4297,
      "step": 333000
    },
    {
      "epoch": 23.429815933679922,
      "grad_norm": 1.072199821472168,
      "learning_rate": 6.46233916437762e-06,
      "loss": 1.431,
      "step": 333500
    },
    {
      "epoch": 23.46494309400028,
      "grad_norm": 1.0913589000701904,
      "learning_rate": 6.317767818418389e-06,
      "loss": 1.4317,
      "step": 334000
    },
    {
      "epoch": 23.50007025432064,
      "grad_norm": 1.1012929677963257,
      "learning_rate": 6.173196472459159e-06,
      "loss": 1.4333,
      "step": 334500
    },
    {
      "epoch": 23.535197414641,
      "grad_norm": 1.083678960800171,
      "learning_rate": 6.028625126499928e-06,
      "loss": 1.4331,
      "step": 335000
    },
    {
      "epoch": 23.57032457496136,
      "grad_norm": 1.1238749027252197,
      "learning_rate": 5.884053780540697e-06,
      "loss": 1.4321,
      "step": 335500
    },
    {
      "epoch": 23.60545173528172,
      "grad_norm": 1.0786747932434082,
      "learning_rate": 5.739482434581466e-06,
      "loss": 1.4292,
      "step": 336000
    },
    {
      "epoch": 23.64057889560208,
      "grad_norm": 1.1055099964141846,
      "learning_rate": 5.5949110886222355e-06,
      "loss": 1.4271,
      "step": 336500
    },
    {
      "epoch": 23.67570605592244,
      "grad_norm": 1.1266461610794067,
      "learning_rate": 5.4503397426630046e-06,
      "loss": 1.4317,
      "step": 337000
    },
    {
      "epoch": 23.7108332162428,
      "grad_norm": 1.0842081308364868,
      "learning_rate": 5.305768396703774e-06,
      "loss": 1.4254,
      "step": 337500
    },
    {
      "epoch": 23.745960376563158,
      "grad_norm": 1.0907618999481201,
      "learning_rate": 5.161197050744543e-06,
      "loss": 1.4276,
      "step": 338000
    },
    {
      "epoch": 23.78108753688352,
      "grad_norm": 1.0893789529800415,
      "learning_rate": 5.016625704785312e-06,
      "loss": 1.4274,
      "step": 338500
    },
    {
      "epoch": 23.81621469720388,
      "grad_norm": 1.108290195465088,
      "learning_rate": 4.872054358826081e-06,
      "loss": 1.4303,
      "step": 339000
    },
    {
      "epoch": 23.851341857524236,
      "grad_norm": 1.1075944900512695,
      "learning_rate": 4.72748301286685e-06,
      "loss": 1.43,
      "step": 339500
    },
    {
      "epoch": 23.886469017844597,
      "grad_norm": 1.0970890522003174,
      "learning_rate": 4.582911666907619e-06,
      "loss": 1.4309,
      "step": 340000
    },
    {
      "epoch": 23.886469017844597,
      "eval_accuracy": 0.7294783881654981,
      "eval_loss": 1.2039755582809448,
      "eval_runtime": 142.3189,
      "eval_samples_per_second": 848.137,
      "eval_steps_per_second": 5.305,
      "step": 340000
    },
    {
      "epoch": 23.921596178164958,
      "grad_norm": 1.102753758430481,
      "learning_rate": 4.438340320948388e-06,
      "loss": 1.4285,
      "step": 340500
    },
    {
      "epoch": 23.956723338485318,
      "grad_norm": 1.0833710432052612,
      "learning_rate": 4.293768974989157e-06,
      "loss": 1.4296,
      "step": 341000
    },
    {
      "epoch": 23.991850498805675,
      "grad_norm": 1.0727189779281616,
      "learning_rate": 4.149197629029926e-06,
      "loss": 1.4298,
      "step": 341500
    },
    {
      "epoch": 24.026977659126036,
      "grad_norm": 1.0745258331298828,
      "learning_rate": 4.004626283070695e-06,
      "loss": 1.4306,
      "step": 342000
    },
    {
      "epoch": 24.062104819446397,
      "grad_norm": 1.1169886589050293,
      "learning_rate": 3.8600549371114645e-06,
      "loss": 1.4281,
      "step": 342500
    },
    {
      "epoch": 24.097231979766757,
      "grad_norm": 1.1066441535949707,
      "learning_rate": 3.7154835911522336e-06,
      "loss": 1.4271,
      "step": 343000
    },
    {
      "epoch": 24.132359140087114,
      "grad_norm": 1.0809811353683472,
      "learning_rate": 3.5709122451930026e-06,
      "loss": 1.4288,
      "step": 343500
    },
    {
      "epoch": 24.167486300407475,
      "grad_norm": 1.0997700691223145,
      "learning_rate": 3.426340899233772e-06,
      "loss": 1.4311,
      "step": 344000
    },
    {
      "epoch": 24.202613460727836,
      "grad_norm": 1.0773580074310303,
      "learning_rate": 3.2817695532745412e-06,
      "loss": 1.4286,
      "step": 344500
    },
    {
      "epoch": 24.237740621048193,
      "grad_norm": 1.1262527704238892,
      "learning_rate": 3.1371982073153103e-06,
      "loss": 1.4309,
      "step": 345000
    },
    {
      "epoch": 24.272867781368554,
      "grad_norm": 1.123502254486084,
      "learning_rate": 2.9926268613560794e-06,
      "loss": 1.4282,
      "step": 345500
    },
    {
      "epoch": 24.307994941688914,
      "grad_norm": 1.1002156734466553,
      "learning_rate": 2.8480555153968485e-06,
      "loss": 1.4275,
      "step": 346000
    },
    {
      "epoch": 24.343122102009275,
      "grad_norm": 1.07537043094635,
      "learning_rate": 2.7034841694376176e-06,
      "loss": 1.4253,
      "step": 346500
    },
    {
      "epoch": 24.378249262329632,
      "grad_norm": 1.1036957502365112,
      "learning_rate": 2.5589128234783866e-06,
      "loss": 1.4279,
      "step": 347000
    },
    {
      "epoch": 24.413376422649993,
      "grad_norm": 1.1081724166870117,
      "learning_rate": 2.414341477519156e-06,
      "loss": 1.427,
      "step": 347500
    },
    {
      "epoch": 24.448503582970353,
      "grad_norm": 1.0847169160842896,
      "learning_rate": 2.269770131559925e-06,
      "loss": 1.4282,
      "step": 348000
    },
    {
      "epoch": 24.483630743290714,
      "grad_norm": 1.0844043493270874,
      "learning_rate": 2.125198785600694e-06,
      "loss": 1.4258,
      "step": 348500
    },
    {
      "epoch": 24.51875790361107,
      "grad_norm": 1.0586968660354614,
      "learning_rate": 1.980627439641463e-06,
      "loss": 1.4246,
      "step": 349000
    },
    {
      "epoch": 24.553885063931432,
      "grad_norm": 1.0943015813827515,
      "learning_rate": 1.8360560936822323e-06,
      "loss": 1.425,
      "step": 349500
    },
    {
      "epoch": 24.589012224251793,
      "grad_norm": 1.080601453781128,
      "learning_rate": 1.6914847477230013e-06,
      "loss": 1.4281,
      "step": 350000
    },
    {
      "epoch": 24.589012224251793,
      "eval_accuracy": 0.7300085771105469,
      "eval_loss": 1.2004581689834595,
      "eval_runtime": 142.3807,
      "eval_samples_per_second": 847.769,
      "eval_steps_per_second": 5.303,
      "step": 350000
    },
    {
      "epoch": 24.62413938457215,
      "grad_norm": 1.1096524000167847,
      "learning_rate": 1.5469134017637704e-06,
      "loss": 1.4259,
      "step": 350500
    },
    {
      "epoch": 24.65926654489251,
      "grad_norm": 1.1088886260986328,
      "learning_rate": 1.4023420558045395e-06,
      "loss": 1.4247,
      "step": 351000
    },
    {
      "epoch": 24.69439370521287,
      "grad_norm": 1.0978398323059082,
      "learning_rate": 1.2577707098453088e-06,
      "loss": 1.4257,
      "step": 351500
    },
    {
      "epoch": 24.72952086553323,
      "grad_norm": 1.131113886833191,
      "learning_rate": 1.1131993638860779e-06,
      "loss": 1.4275,
      "step": 352000
    },
    {
      "epoch": 24.76464802585359,
      "grad_norm": 1.1421183347702026,
      "learning_rate": 9.68628017926847e-07,
      "loss": 1.4258,
      "step": 352500
    },
    {
      "epoch": 24.79977518617395,
      "grad_norm": 1.0884921550750732,
      "learning_rate": 8.240566719676161e-07,
      "loss": 1.4264,
      "step": 353000
    },
    {
      "epoch": 24.83490234649431,
      "grad_norm": 1.0992377996444702,
      "learning_rate": 6.794853260083851e-07,
      "loss": 1.4257,
      "step": 353500
    },
    {
      "epoch": 24.870029506814667,
      "grad_norm": 1.0836498737335205,
      "learning_rate": 5.349139800491543e-07,
      "loss": 1.4275,
      "step": 354000
    },
    {
      "epoch": 24.905156667135028,
      "grad_norm": 1.1038559675216675,
      "learning_rate": 3.903426340899234e-07,
      "loss": 1.4262,
      "step": 354500
    },
    {
      "epoch": 24.94028382745539,
      "grad_norm": 1.1141635179519653,
      "learning_rate": 2.457712881306925e-07,
      "loss": 1.4226,
      "step": 355000
    },
    {
      "epoch": 24.97541098777575,
      "grad_norm": 1.0651347637176514,
      "learning_rate": 1.0119994217146162e-07,
      "loss": 1.4269,
      "step": 355500
    },
    {
      "epoch": 25.0,
      "step": 355850,
      "total_flos": 7.547177208649421e+18,
      "train_loss": 1.849292092869351,
      "train_runtime": 122993.5623,
      "train_samples_per_second": 462.899,
      "train_steps_per_second": 2.893
    }
  ],
  "logging_steps": 500,
  "max_steps": 355850,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 10000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.547177208649421e+18,
  "train_batch_size": 160,
  "trial_name": null,
  "trial_params": null
}