{
  "best_metric": 0.019864432513713837,
  "best_model_checkpoint": "/home/paperspace/Data/models/akash_unifo_757/llm3br256/checkpoint-400",
  "epoch": 4.962406015037594,
  "eval_steps": 5,
  "global_step": 495,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010025062656641603,
      "grad_norm": 0.28569191694259644,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0958,
      "step": 1
    },
    {
      "epoch": 0.020050125313283207,
      "grad_norm": 0.36814188957214355,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0953,
      "step": 2
    },
    {
      "epoch": 0.03007518796992481,
      "grad_norm": 0.2642641067504883,
      "learning_rate": 6e-06,
      "loss": 0.081,
      "step": 3
    },
    {
      "epoch": 0.040100250626566414,
      "grad_norm": 0.2690538763999939,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0846,
      "step": 4
    },
    {
      "epoch": 0.05012531328320802,
      "grad_norm": 0.2873159646987915,
      "learning_rate": 1e-05,
      "loss": 0.0892,
      "step": 5
    },
    {
      "epoch": 0.05012531328320802,
      "eval_loss": 0.09708356112241745,
      "eval_runtime": 10.7924,
      "eval_samples_per_second": 4.633,
      "eval_steps_per_second": 1.205,
      "step": 5
    },
    {
      "epoch": 0.06015037593984962,
      "grad_norm": 0.1856367141008377,
      "learning_rate": 1.2e-05,
      "loss": 0.0689,
      "step": 6
    },
    {
      "epoch": 0.07017543859649122,
      "grad_norm": 0.1307721883058548,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 0.0501,
      "step": 7
    },
    {
      "epoch": 0.08020050125313283,
      "grad_norm": 0.15366266667842865,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.0515,
      "step": 8
    },
    {
      "epoch": 0.09022556390977443,
      "grad_norm": 0.11183061450719833,
      "learning_rate": 1.8e-05,
      "loss": 0.0556,
      "step": 9
    },
    {
      "epoch": 0.10025062656641603,
      "grad_norm": 0.09610645473003387,
      "learning_rate": 2e-05,
      "loss": 0.045,
      "step": 10
    },
    {
      "epoch": 0.10025062656641603,
      "eval_loss": 0.04338453337550163,
      "eval_runtime": 7.8458,
      "eval_samples_per_second": 6.373,
      "eval_steps_per_second": 1.657,
      "step": 10
    },
    {
      "epoch": 0.11027568922305764,
      "grad_norm": 0.09562630206346512,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.0468,
      "step": 11
    },
    {
      "epoch": 0.12030075187969924,
      "grad_norm": 0.08241941034793854,
      "learning_rate": 2.4e-05,
      "loss": 0.0379,
      "step": 12
    },
    {
      "epoch": 0.13032581453634084,
      "grad_norm": 0.08410688489675522,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 0.0364,
      "step": 13
    },
    {
      "epoch": 0.14035087719298245,
      "grad_norm": 0.07379286736249924,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 0.0274,
      "step": 14
    },
    {
      "epoch": 0.15037593984962405,
      "grad_norm": 0.0845954418182373,
      "learning_rate": 3e-05,
      "loss": 0.0496,
      "step": 15
    },
    {
      "epoch": 0.15037593984962405,
      "eval_loss": 0.03607996553182602,
      "eval_runtime": 7.8661,
      "eval_samples_per_second": 6.356,
      "eval_steps_per_second": 1.653,
      "step": 15
    },
    {
      "epoch": 0.16040100250626566,
      "grad_norm": 0.08074897527694702,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.0312,
      "step": 16
    },
    {
      "epoch": 0.17042606516290726,
      "grad_norm": 0.060424063354730606,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 0.03,
      "step": 17
    },
    {
      "epoch": 0.18045112781954886,
      "grad_norm": 0.06651686877012253,
      "learning_rate": 3.6e-05,
      "loss": 0.0296,
      "step": 18
    },
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 0.058752626180648804,
      "learning_rate": 3.8e-05,
      "loss": 0.0272,
      "step": 19
    },
    {
      "epoch": 0.20050125313283207,
      "grad_norm": 0.05913863703608513,
      "learning_rate": 4e-05,
      "loss": 0.0279,
      "step": 20
    },
    {
      "epoch": 0.20050125313283207,
      "eval_loss": 0.03392041102051735,
      "eval_runtime": 7.8762,
      "eval_samples_per_second": 6.348,
      "eval_steps_per_second": 1.651,
      "step": 20
    },
    {
      "epoch": 0.21052631578947367,
      "grad_norm": 0.05515163764357567,
      "learning_rate": 4.2e-05,
      "loss": 0.028,
      "step": 21
    },
    {
      "epoch": 0.22055137844611528,
      "grad_norm": 0.06397461146116257,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0283,
      "step": 22
    },
    {
      "epoch": 0.23057644110275688,
      "grad_norm": 0.055562298744916916,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.0281,
      "step": 23
    },
    {
      "epoch": 0.24060150375939848,
      "grad_norm": 0.07124784588813782,
      "learning_rate": 4.8e-05,
      "loss": 0.0277,
      "step": 24
    },
    {
      "epoch": 0.2506265664160401,
      "grad_norm": 0.04555438458919525,
      "learning_rate": 5e-05,
      "loss": 0.027,
      "step": 25
    },
    {
      "epoch": 0.2506265664160401,
      "eval_loss": 0.033857107162475586,
      "eval_runtime": 7.8415,
      "eval_samples_per_second": 6.376,
      "eval_steps_per_second": 1.658,
      "step": 25
    },
    {
      "epoch": 0.2606516290726817,
      "grad_norm": 0.07086709886789322,
      "learning_rate": 5.2000000000000004e-05,
      "loss": 0.0253,
      "step": 26
    },
    {
      "epoch": 0.2706766917293233,
      "grad_norm": 0.056244127452373505,
      "learning_rate": 5.4000000000000005e-05,
      "loss": 0.0282,
      "step": 27
    },
    {
      "epoch": 0.2807017543859649,
      "grad_norm": 0.04824177920818329,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 0.024,
      "step": 28
    },
    {
      "epoch": 0.2907268170426065,
      "grad_norm": 0.061910007148981094,
      "learning_rate": 5.8e-05,
      "loss": 0.0301,
      "step": 29
    },
    {
      "epoch": 0.3007518796992481,
      "grad_norm": 0.05887613445520401,
      "learning_rate": 6e-05,
      "loss": 0.0265,
      "step": 30
    },
    {
      "epoch": 0.3007518796992481,
      "eval_loss": 0.030988477170467377,
      "eval_runtime": 7.846,
      "eval_samples_per_second": 6.373,
      "eval_steps_per_second": 1.657,
      "step": 30
    },
    {
      "epoch": 0.3107769423558897,
      "grad_norm": 0.06682919710874557,
      "learning_rate": 6.2e-05,
      "loss": 0.0283,
      "step": 31
    },
    {
      "epoch": 0.3208020050125313,
      "grad_norm": 0.04526979848742485,
      "learning_rate": 6.400000000000001e-05,
      "loss": 0.0227,
      "step": 32
    },
    {
      "epoch": 0.3308270676691729,
      "grad_norm": 0.037424199283123016,
      "learning_rate": 6.6e-05,
      "loss": 0.0224,
      "step": 33
    },
    {
      "epoch": 0.3408521303258145,
      "grad_norm": 0.04883978143334389,
      "learning_rate": 6.800000000000001e-05,
      "loss": 0.0251,
      "step": 34
    },
    {
      "epoch": 0.3508771929824561,
      "grad_norm": 0.04064160957932472,
      "learning_rate": 7e-05,
      "loss": 0.0255,
      "step": 35
    },
    {
      "epoch": 0.3508771929824561,
      "eval_loss": 0.029722385108470917,
      "eval_runtime": 7.8878,
      "eval_samples_per_second": 6.339,
      "eval_steps_per_second": 1.648,
      "step": 35
    },
    {
      "epoch": 0.3609022556390977,
      "grad_norm": 0.04703061655163765,
      "learning_rate": 7.2e-05,
      "loss": 0.0251,
      "step": 36
    },
    {
      "epoch": 0.37092731829573933,
      "grad_norm": 0.03412939980626106,
      "learning_rate": 7.4e-05,
      "loss": 0.0221,
      "step": 37
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 0.038730327039957047,
      "learning_rate": 7.6e-05,
      "loss": 0.021,
      "step": 38
    },
    {
      "epoch": 0.39097744360902253,
      "grad_norm": 0.0406007319688797,
      "learning_rate": 7.800000000000001e-05,
      "loss": 0.0196,
      "step": 39
    },
    {
      "epoch": 0.40100250626566414,
      "grad_norm": 0.040442854166030884,
      "learning_rate": 8e-05,
      "loss": 0.0239,
      "step": 40
    },
    {
      "epoch": 0.40100250626566414,
      "eval_loss": 0.027504581958055496,
      "eval_runtime": 7.8471,
      "eval_samples_per_second": 6.372,
      "eval_steps_per_second": 1.657,
      "step": 40
    },
    {
      "epoch": 0.41102756892230574,
      "grad_norm": 0.040039267390966415,
      "learning_rate": 8.2e-05,
      "loss": 0.0192,
      "step": 41
    },
    {
      "epoch": 0.42105263157894735,
      "grad_norm": 0.050538670271635056,
      "learning_rate": 8.4e-05,
      "loss": 0.025,
      "step": 42
    },
    {
      "epoch": 0.43107769423558895,
      "grad_norm": 0.039062079042196274,
      "learning_rate": 8.6e-05,
      "loss": 0.0247,
      "step": 43
    },
    {
      "epoch": 0.44110275689223055,
      "grad_norm": 0.061014942824840546,
      "learning_rate": 8.800000000000001e-05,
      "loss": 0.0173,
      "step": 44
    },
    {
      "epoch": 0.45112781954887216,
      "grad_norm": 0.039824020117521286,
      "learning_rate": 9e-05,
      "loss": 0.019,
      "step": 45
    },
    {
      "epoch": 0.45112781954887216,
      "eval_loss": 0.02625363878905773,
      "eval_runtime": 7.8454,
      "eval_samples_per_second": 6.373,
      "eval_steps_per_second": 1.657,
      "step": 45
    },
    {
      "epoch": 0.46115288220551376,
      "grad_norm": 0.056794699281454086,
      "learning_rate": 9.200000000000001e-05,
      "loss": 0.0202,
      "step": 46
    },
    {
      "epoch": 0.47117794486215536,
      "grad_norm": 0.032318562269210815,
      "learning_rate": 9.4e-05,
      "loss": 0.0191,
      "step": 47
    },
    {
      "epoch": 0.48120300751879697,
      "grad_norm": 0.05849836394190788,
      "learning_rate": 9.6e-05,
      "loss": 0.0247,
      "step": 48
    },
    {
      "epoch": 0.49122807017543857,
      "grad_norm": 0.03626992926001549,
      "learning_rate": 9.8e-05,
      "loss": 0.0211,
      "step": 49
    },
    {
      "epoch": 0.5012531328320802,
      "grad_norm": 0.04883263632655144,
      "learning_rate": 0.0001,
      "loss": 0.0177,
      "step": 50
    },
    {
      "epoch": 0.5012531328320802,
      "eval_loss": 0.025471089407801628,
      "eval_runtime": 7.8509,
      "eval_samples_per_second": 6.369,
      "eval_steps_per_second": 1.656,
      "step": 50
    },
    {
      "epoch": 0.5112781954887218,
      "grad_norm": 0.028212551027536392,
      "learning_rate": 9.999875400032707e-05,
      "loss": 0.0169,
      "step": 51
    },
    {
      "epoch": 0.5213032581453634,
      "grad_norm": 0.03416428342461586,
      "learning_rate": 9.999501606340891e-05,
      "loss": 0.0189,
      "step": 52
    },
    {
      "epoch": 0.531328320802005,
      "grad_norm": 0.04827010631561279,
      "learning_rate": 9.998878637554424e-05,
      "loss": 0.022,
      "step": 53
    },
    {
      "epoch": 0.5413533834586466,
      "grad_norm": 0.044534213840961456,
      "learning_rate": 9.998006524722059e-05,
      "loss": 0.0229,
      "step": 54
    },
    {
      "epoch": 0.5513784461152882,
      "grad_norm": 0.03421458229422569,
      "learning_rate": 9.996885311309891e-05,
      "loss": 0.0178,
      "step": 55
    },
    {
      "epoch": 0.5513784461152882,
      "eval_loss": 0.02499430626630783,
      "eval_runtime": 7.8452,
      "eval_samples_per_second": 6.373,
      "eval_steps_per_second": 1.657,
      "step": 55
    },
    {
      "epoch": 0.5614035087719298,
      "grad_norm": 0.03985392674803734,
      "learning_rate": 9.995515053199182e-05,
      "loss": 0.0213,
      "step": 56
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.037616852670907974,
      "learning_rate": 9.993895818683579e-05,
      "loss": 0.0163,
      "step": 57
    },
    {
      "epoch": 0.581453634085213,
      "grad_norm": 0.03808311000466347,
      "learning_rate": 9.992027688465707e-05,
      "loss": 0.0212,
      "step": 58
    },
    {
      "epoch": 0.5914786967418546,
      "grad_norm": 0.03867008537054062,
      "learning_rate": 9.989910755653154e-05,
      "loss": 0.0205,
      "step": 59
    },
    {
      "epoch": 0.6015037593984962,
      "grad_norm": 0.03990580886602402,
      "learning_rate": 9.987545125753819e-05,
      "loss": 0.0179,
      "step": 60
    },
    {
      "epoch": 0.6015037593984962,
      "eval_loss": 0.02382907271385193,
      "eval_runtime": 7.8913,
      "eval_samples_per_second": 6.336,
      "eval_steps_per_second": 1.647,
      "step": 60
    },
    {
      "epoch": 0.6115288220551378,
      "grad_norm": 0.0392407588660717,
      "learning_rate": 9.98493091667067e-05,
      "loss": 0.0239,
      "step": 61
    },
    {
      "epoch": 0.6215538847117794,
      "grad_norm": 0.030434446409344673,
      "learning_rate": 9.982068258695853e-05,
      "loss": 0.0164,
      "step": 62
    },
    {
      "epoch": 0.631578947368421,
      "grad_norm": 0.042929526418447495,
      "learning_rate": 9.978957294504203e-05,
      "loss": 0.0223,
      "step": 63
    },
    {
      "epoch": 0.6416040100250626,
      "grad_norm": 0.028332175686955452,
      "learning_rate": 9.975598179146133e-05,
      "loss": 0.0185,
      "step": 64
    },
    {
      "epoch": 0.6516290726817042,
      "grad_norm": 0.037013668566942215,
      "learning_rate": 9.97199108003991e-05,
      "loss": 0.0199,
      "step": 65
    },
    {
      "epoch": 0.6516290726817042,
      "eval_loss": 0.023866703733801842,
      "eval_runtime": 7.8611,
      "eval_samples_per_second": 6.36,
      "eval_steps_per_second": 1.654,
      "step": 65
    },
    {
      "epoch": 0.6616541353383458,
      "grad_norm": 0.04389967769384384,
      "learning_rate": 9.968136176963307e-05,
      "loss": 0.0215,
      "step": 66
    },
    {
      "epoch": 0.6716791979949874,
      "grad_norm": 0.03696281462907791,
      "learning_rate": 9.964033662044643e-05,
      "loss": 0.02,
      "step": 67
    },
    {
      "epoch": 0.681704260651629,
      "grad_norm": 0.03329788148403168,
      "learning_rate": 9.959683739753207e-05,
      "loss": 0.0194,
      "step": 68
    },
    {
      "epoch": 0.6917293233082706,
      "grad_norm": 0.0394810289144516,
      "learning_rate": 9.955086626889068e-05,
      "loss": 0.0188,
      "step": 69
    },
    {
      "epoch": 0.7017543859649122,
      "grad_norm": 0.03089299239218235,
      "learning_rate": 9.950242552572271e-05,
      "loss": 0.0165,
      "step": 70
    },
    {
      "epoch": 0.7017543859649122,
      "eval_loss": 0.023690788075327873,
      "eval_runtime": 7.9034,
      "eval_samples_per_second": 6.326,
      "eval_steps_per_second": 1.645,
      "step": 70
    },
    {
      "epoch": 0.7117794486215538,
      "grad_norm": 0.037516698241233826,
      "learning_rate": 9.945151758231421e-05,
      "loss": 0.0181,
      "step": 71
    },
    {
      "epoch": 0.7218045112781954,
      "grad_norm": 0.03217910975217819,
      "learning_rate": 9.939814497591636e-05,
      "loss": 0.0168,
      "step": 72
    },
    {
      "epoch": 0.731829573934837,
      "grad_norm": 0.041075628250837326,
      "learning_rate": 9.934231036661919e-05,
      "loss": 0.02,
      "step": 73
    },
    {
      "epoch": 0.7418546365914787,
      "grad_norm": 0.036599840968847275,
      "learning_rate": 9.928401653721891e-05,
      "loss": 0.0209,
      "step": 74
    },
    {
      "epoch": 0.7518796992481203,
      "grad_norm": 0.03515414148569107,
      "learning_rate": 9.922326639307917e-05,
      "loss": 0.0192,
      "step": 75
    },
    {
      "epoch": 0.7518796992481203,
      "eval_loss": 0.02293418161571026,
      "eval_runtime": 7.8475,
      "eval_samples_per_second": 6.371,
      "eval_steps_per_second": 1.657,
      "step": 75
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 0.030638370662927628,
      "learning_rate": 9.91600629619864e-05,
      "loss": 0.0219,
      "step": 76
    },
    {
      "epoch": 0.7719298245614035,
      "grad_norm": 0.029526930302381516,
      "learning_rate": 9.909440939399876e-05,
      "loss": 0.0188,
      "step": 77
    },
    {
      "epoch": 0.7819548872180451,
      "grad_norm": 0.043333012610673904,
      "learning_rate": 9.902630896128923e-05,
      "loss": 0.0218,
      "step": 78
    },
    {
      "epoch": 0.7919799498746867,
      "grad_norm": 0.04038293659687042,
      "learning_rate": 9.895576505798248e-05,
      "loss": 0.0207,
      "step": 79
    },
    {
      "epoch": 0.8020050125313283,
      "grad_norm": 0.02755335345864296,
      "learning_rate": 9.888278119998573e-05,
      "loss": 0.0158,
      "step": 80
    },
    {
      "epoch": 0.8020050125313283,
      "eval_loss": 0.02307852730154991,
      "eval_runtime": 7.844,
      "eval_samples_per_second": 6.374,
      "eval_steps_per_second": 1.657,
      "step": 80
    },
    {
      "epoch": 0.8120300751879699,
      "grad_norm": 0.036036524921655655,
      "learning_rate": 9.88073610248135e-05,
      "loss": 0.0158,
      "step": 81
    },
    {
      "epoch": 0.8220551378446115,
      "grad_norm": 0.03679794445633888,
      "learning_rate": 9.872950829140633e-05,
      "loss": 0.0276,
      "step": 82
    },
    {
      "epoch": 0.8320802005012531,
      "grad_norm": 0.03850351274013519,
      "learning_rate": 9.864922687994347e-05,
      "loss": 0.0159,
      "step": 83
    },
    {
      "epoch": 0.8421052631578947,
      "grad_norm": 0.03365923836827278,
      "learning_rate": 9.856652079164937e-05,
      "loss": 0.0204,
      "step": 84
    },
    {
      "epoch": 0.8521303258145363,
      "grad_norm": 0.04202227666974068,
      "learning_rate": 9.848139414859441e-05,
      "loss": 0.0202,
      "step": 85
    },
    {
      "epoch": 0.8521303258145363,
      "eval_loss": 0.023334117606282234,
      "eval_runtime": 7.8493,
      "eval_samples_per_second": 6.37,
      "eval_steps_per_second": 1.656,
      "step": 85
    },
    {
      "epoch": 0.8621553884711779,
      "grad_norm": 0.027862610295414925,
      "learning_rate": 9.839385119348937e-05,
      "loss": 0.0162,
      "step": 86
    },
    {
      "epoch": 0.8721804511278195,
      "grad_norm": 0.05110170319676399,
      "learning_rate": 9.830389628947398e-05,
      "loss": 0.0224,
      "step": 87
    },
    {
      "epoch": 0.8822055137844611,
      "grad_norm": 0.039190806448459625,
      "learning_rate": 9.82115339198995e-05,
      "loss": 0.0207,
      "step": 88
    },
    {
      "epoch": 0.8922305764411027,
      "grad_norm": 0.03106129728257656,
      "learning_rate": 9.811676868810517e-05,
      "loss": 0.0217,
      "step": 89
    },
    {
      "epoch": 0.9022556390977443,
      "grad_norm": 0.04018354043364525,
      "learning_rate": 9.801960531718896e-05,
      "loss": 0.0203,
      "step": 90
    },
    {
      "epoch": 0.9022556390977443,
      "eval_loss": 0.023184653371572495,
      "eval_runtime": 7.8695,
      "eval_samples_per_second": 6.354,
      "eval_steps_per_second": 1.652,
      "step": 90
    },
    {
      "epoch": 0.9122807017543859,
      "grad_norm": 0.031660296022892,
      "learning_rate": 9.792004864977198e-05,
      "loss": 0.023,
      "step": 91
    },
    {
      "epoch": 0.9223057644110275,
      "grad_norm": 0.030742425471544266,
      "learning_rate": 9.781810364775722e-05,
      "loss": 0.0166,
      "step": 92
    },
    {
      "epoch": 0.9323308270676691,
      "grad_norm": 0.03069412149488926,
      "learning_rate": 9.771377539208228e-05,
      "loss": 0.018,
      "step": 93
    },
    {
      "epoch": 0.9423558897243107,
      "grad_norm": 0.02346673421561718,
      "learning_rate": 9.760706908246603e-05,
      "loss": 0.0161,
      "step": 94
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 0.0320681668817997,
      "learning_rate": 9.749799003714954e-05,
      "loss": 0.0231,
      "step": 95
    },
    {
      "epoch": 0.9523809523809523,
      "eval_loss": 0.02278592251241207,
      "eval_runtime": 7.8596,
      "eval_samples_per_second": 6.362,
      "eval_steps_per_second": 1.654,
      "step": 95
    },
    {
      "epoch": 0.9624060150375939,
      "grad_norm": 0.03784003481268883,
      "learning_rate": 9.738654369263103e-05,
      "loss": 0.0225,
      "step": 96
    },
    {
      "epoch": 0.9724310776942355,
      "grad_norm": 0.03227786719799042,
      "learning_rate": 9.727273560339483e-05,
      "loss": 0.0145,
      "step": 97
    },
    {
      "epoch": 0.9824561403508771,
      "grad_norm": 0.030028732493519783,
      "learning_rate": 9.715657144163463e-05,
      "loss": 0.0156,
      "step": 98
    },
    {
      "epoch": 0.9924812030075187,
      "grad_norm": 0.03690657019615173,
      "learning_rate": 9.703805699697072e-05,
      "loss": 0.0229,
      "step": 99
    },
    {
      "epoch": 1.0025062656641603,
      "grad_norm": 0.03713948652148247,
      "learning_rate": 9.691719817616147e-05,
      "loss": 0.0175,
      "step": 100
    },
    {
      "epoch": 1.0025062656641603,
      "eval_loss": 0.022534934803843498,
      "eval_runtime": 7.8577,
      "eval_samples_per_second": 6.363,
      "eval_steps_per_second": 1.654,
      "step": 100
    },
    {
      "epoch": 1.012531328320802,
      "grad_norm": 0.03210939094424248,
      "learning_rate": 9.679400100280896e-05,
      "loss": 0.0178,
      "step": 101
    },
    {
      "epoch": 1.0225563909774436,
      "grad_norm": 0.021419484168291092,
      "learning_rate": 9.666847161705867e-05,
      "loss": 0.0147,
      "step": 102
    },
    {
      "epoch": 1.0325814536340852,
      "grad_norm": 0.03585834801197052,
      "learning_rate": 9.654061627529354e-05,
      "loss": 0.02,
      "step": 103
    },
    {
      "epoch": 1.0426065162907268,
      "grad_norm": 0.02969486080110073,
      "learning_rate": 9.641044134982215e-05,
      "loss": 0.0129,
      "step": 104
    },
    {
      "epoch": 1.0526315789473684,
      "grad_norm": 0.023357393220067024,
      "learning_rate": 9.627795332856107e-05,
      "loss": 0.0137,
      "step": 105
    },
    {
      "epoch": 1.0526315789473684,
      "eval_loss": 0.022470790892839432,
      "eval_runtime": 7.8421,
      "eval_samples_per_second": 6.376,
      "eval_steps_per_second": 1.658,
      "step": 105
    },
    {
      "epoch": 1.06265664160401,
      "grad_norm": 0.030119987204670906,
      "learning_rate": 9.614315881471154e-05,
      "loss": 0.0155,
      "step": 106
    },
    {
      "epoch": 1.0726817042606516,
      "grad_norm": 0.023424195125699043,
      "learning_rate": 9.600606452643037e-05,
      "loss": 0.0131,
      "step": 107
    },
    {
      "epoch": 1.0827067669172932,
      "grad_norm": 0.028333192691206932,
      "learning_rate": 9.586667729649513e-05,
      "loss": 0.0169,
      "step": 108
    },
    {
      "epoch": 1.0927318295739348,
      "grad_norm": 0.024621840566396713,
      "learning_rate": 9.572500407196348e-05,
      "loss": 0.0135,
      "step": 109
    },
    {
      "epoch": 1.1027568922305764,
      "grad_norm": 0.0384870246052742,
      "learning_rate": 9.55810519138271e-05,
      "loss": 0.0286,
      "step": 110
    },
    {
      "epoch": 1.1027568922305764,
      "eval_loss": 0.02285490557551384,
      "eval_runtime": 7.8692,
      "eval_samples_per_second": 6.354,
      "eval_steps_per_second": 1.652,
      "step": 110
    },
    {
      "epoch": 1.112781954887218,
      "grad_norm": 0.029268387705087662,
      "learning_rate": 9.543482799665969e-05,
      "loss": 0.0152,
      "step": 111
    },
    {
      "epoch": 1.1228070175438596,
      "grad_norm": 0.02799990586936474,
      "learning_rate": 9.528633960825933e-05,
      "loss": 0.0142,
      "step": 112
    },
    {
      "epoch": 1.1328320802005012,
      "grad_norm": 0.019951699301600456,
      "learning_rate": 9.513559414928538e-05,
      "loss": 0.0118,
      "step": 113
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 0.02334384061396122,
      "learning_rate": 9.498259913288953e-05,
      "loss": 0.0128,
      "step": 114
    },
    {
      "epoch": 1.1528822055137844,
      "grad_norm": 0.028435515239834785,
      "learning_rate": 9.482736218434143e-05,
      "loss": 0.0169,
      "step": 115
    },
    {
      "epoch": 1.1528822055137844,
      "eval_loss": 0.022512206807732582,
      "eval_runtime": 7.8509,
      "eval_samples_per_second": 6.369,
      "eval_steps_per_second": 1.656,
      "step": 115
    },
    {
      "epoch": 1.162907268170426,
      "grad_norm": 0.03522834554314613,
      "learning_rate": 9.466989104064853e-05,
      "loss": 0.0183,
      "step": 116
    },
    {
      "epoch": 1.1729323308270676,
      "grad_norm": 0.025662166997790337,
      "learning_rate": 9.451019355017056e-05,
      "loss": 0.0114,
      "step": 117
    },
    {
      "epoch": 1.1829573934837092,
      "grad_norm": 0.027821151539683342,
      "learning_rate": 9.43482776722284e-05,
      "loss": 0.0111,
      "step": 118
    },
    {
      "epoch": 1.1929824561403508,
      "grad_norm": 0.039541855454444885,
      "learning_rate": 9.418415147670725e-05,
      "loss": 0.0153,
      "step": 119
    },
    {
      "epoch": 1.2030075187969924,
      "grad_norm": 0.02246340923011303,
      "learning_rate": 9.401782314365457e-05,
      "loss": 0.0141,
      "step": 120
    },
    {
      "epoch": 1.2030075187969924,
      "eval_loss": 0.022168081253767014,
      "eval_runtime": 7.868,
      "eval_samples_per_second": 6.355,
      "eval_steps_per_second": 1.652,
      "step": 120
    },
    {
      "epoch": 1.213032581453634,
      "grad_norm": 0.027894971892237663,
      "learning_rate": 9.38493009628723e-05,
      "loss": 0.0166,
      "step": 121
    },
    {
      "epoch": 1.2230576441102756,
      "grad_norm": 0.023307248950004578,
      "learning_rate": 9.36785933335037e-05,
      "loss": 0.0141,
      "step": 122
    },
    {
      "epoch": 1.2330827067669172,
      "grad_norm": 0.03410186246037483,
      "learning_rate": 9.350570876361482e-05,
      "loss": 0.0145,
      "step": 123
    },
    {
      "epoch": 1.2431077694235588,
      "grad_norm": 0.029240259900689125,
      "learning_rate": 9.333065586977035e-05,
      "loss": 0.0102,
      "step": 124
    },
    {
      "epoch": 1.2531328320802004,
      "grad_norm": 0.026229720562696457,
      "learning_rate": 9.315344337660421e-05,
      "loss": 0.0149,
      "step": 125
    },
    {
      "epoch": 1.2531328320802004,
      "eval_loss": 0.02199222519993782,
      "eval_runtime": 7.8428,
      "eval_samples_per_second": 6.375,
      "eval_steps_per_second": 1.658,
      "step": 125
    },
    {
      "epoch": 1.263157894736842,
      "grad_norm": 0.02851460874080658,
      "learning_rate": 9.297408011638477e-05,
      "loss": 0.017,
      "step": 126
    },
    {
      "epoch": 1.2731829573934836,
      "grad_norm": 0.02489631064236164,
      "learning_rate": 9.279257502857455e-05,
      "loss": 0.0106,
      "step": 127
    },
    {
      "epoch": 1.2832080200501252,
      "grad_norm": 0.031498756259679794,
      "learning_rate": 9.260893715938477e-05,
      "loss": 0.0188,
      "step": 128
    },
    {
      "epoch": 1.2932330827067668,
      "grad_norm": 0.026509685441851616,
      "learning_rate": 9.24231756613244e-05,
      "loss": 0.0166,
      "step": 129
    },
    {
      "epoch": 1.3032581453634084,
      "grad_norm": 0.02835785411298275,
      "learning_rate": 9.22352997927441e-05,
      "loss": 0.0123,
      "step": 130
    },
    {
      "epoch": 1.3032581453634084,
      "eval_loss": 0.022584008052945137,
      "eval_runtime": 7.8608,
      "eval_samples_per_second": 6.361,
      "eval_steps_per_second": 1.654,
      "step": 130
    },
    {
      "epoch": 1.31328320802005,
      "grad_norm": 0.03289595618844032,
      "learning_rate": 9.20453189173747e-05,
      "loss": 0.0173,
      "step": 131
    },
    {
      "epoch": 1.3233082706766917,
      "grad_norm": 0.037241410464048386,
      "learning_rate": 9.185324250386054e-05,
      "loss": 0.0205,
      "step": 132
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.02691718004643917,
      "learning_rate": 9.165908012528755e-05,
      "loss": 0.0144,
      "step": 133
    },
    {
      "epoch": 1.3433583959899749,
      "grad_norm": 0.028582507744431496,
      "learning_rate": 9.146284145870614e-05,
      "loss": 0.0182,
      "step": 134
    },
    {
      "epoch": 1.3533834586466165,
      "grad_norm": 0.02578776516020298,
      "learning_rate": 9.126453628464888e-05,
      "loss": 0.0137,
      "step": 135
    },
    {
      "epoch": 1.3533834586466165,
      "eval_loss": 0.022607652470469475,
      "eval_runtime": 7.8746,
      "eval_samples_per_second": 6.35,
      "eval_steps_per_second": 1.651,
      "step": 135
    },
    {
      "epoch": 1.363408521303258,
      "grad_norm": 0.021922461688518524,
      "learning_rate": 9.106417448664306e-05,
      "loss": 0.0124,
      "step": 136
    },
    {
      "epoch": 1.3734335839598997,
      "grad_norm": 0.027213698253035545,
      "learning_rate": 9.086176605071805e-05,
      "loss": 0.0145,
      "step": 137
    },
    {
      "epoch": 1.3834586466165413,
      "grad_norm": 0.03600913658738136,
      "learning_rate": 9.06573210649077e-05,
      "loss": 0.0123,
      "step": 138
    },
    {
      "epoch": 1.3934837092731829,
      "grad_norm": 0.02878659963607788,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0172,
      "step": 139
    },
    {
      "epoch": 1.4035087719298245,
      "grad_norm": 0.022535493597388268,
      "learning_rate": 9.024236230276629e-05,
      "loss": 0.0118,
      "step": 140
    },
    {
      "epoch": 1.4035087719298245,
      "eval_loss": 0.02263771742582321,
      "eval_runtime": 7.8445,
      "eval_samples_per_second": 6.374,
      "eval_steps_per_second": 1.657,
      "step": 140
    },
    {
      "epoch": 1.413533834586466,
      "grad_norm": 0.025746768340468407,
      "learning_rate": 9.003186920797452e-05,
      "loss": 0.0165,
      "step": 141
    },
    {
      "epoch": 1.4235588972431077,
      "grad_norm": 0.017796190455555916,
      "learning_rate": 8.981938092534517e-05,
      "loss": 0.0078,
      "step": 142
    },
    {
      "epoch": 1.4335839598997493,
      "grad_norm": 0.03525475040078163,
      "learning_rate": 8.960490804529144e-05,
      "loss": 0.0205,
      "step": 143
    },
    {
      "epoch": 1.443609022556391,
      "grad_norm": 0.019581295549869537,
      "learning_rate": 8.938846125713891e-05,
      "loss": 0.0091,
      "step": 144
    },
    {
      "epoch": 1.4536340852130325,
      "grad_norm": 0.03014942817389965,
      "learning_rate": 8.917005134859263e-05,
      "loss": 0.015,
      "step": 145
    },
    {
      "epoch": 1.4536340852130325,
      "eval_loss": 0.021907737478613853,
      "eval_runtime": 7.8749,
      "eval_samples_per_second": 6.349,
      "eval_steps_per_second": 1.651,
      "step": 145
    },
    {
      "epoch": 1.463659147869674,
      "grad_norm": 0.028763974085450172,
      "learning_rate": 8.894968920519959e-05,
      "loss": 0.0156,
      "step": 146
    },
    {
      "epoch": 1.4736842105263157,
      "grad_norm": 0.022081634029746056,
      "learning_rate": 8.872738580980615e-05,
      "loss": 0.013,
      "step": 147
    },
    {
      "epoch": 1.4837092731829573,
      "grad_norm": 0.031854048371315,
      "learning_rate": 8.850315224201063e-05,
      "loss": 0.0159,
      "step": 148
    },
    {
      "epoch": 1.493734335839599,
      "grad_norm": 0.03208814561367035,
      "learning_rate": 8.827699967761108e-05,
      "loss": 0.0178,
      "step": 149
    },
    {
      "epoch": 1.5037593984962405,
      "grad_norm": 0.02058330550789833,
      "learning_rate": 8.80489393880484e-05,
      "loss": 0.0059,
      "step": 150
    },
    {
      "epoch": 1.5037593984962405,
      "eval_loss": 0.02317122556269169,
      "eval_runtime": 7.8482,
      "eval_samples_per_second": 6.371,
      "eval_steps_per_second": 1.656,
      "step": 150
    },
    {
      "epoch": 1.5137844611528823,
      "grad_norm": 0.036856647580862045,
      "learning_rate": 8.78189827398444e-05,
      "loss": 0.0165,
      "step": 151
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 0.02781173400580883,
      "learning_rate": 8.758714119403543e-05,
      "loss": 0.0135,
      "step": 152
    },
    {
      "epoch": 1.5338345864661656,
      "grad_norm": 0.020493071526288986,
      "learning_rate": 8.73534263056011e-05,
      "loss": 0.0118,
      "step": 153
    },
    {
      "epoch": 1.543859649122807,
      "grad_norm": 0.031307898461818695,
      "learning_rate": 8.711784972288839e-05,
      "loss": 0.0129,
      "step": 154
    },
    {
      "epoch": 1.5538847117794488,
      "grad_norm": 0.0317239835858345,
      "learning_rate": 8.688042318703111e-05,
      "loss": 0.0155,
      "step": 155
    },
    {
      "epoch": 1.5538847117794488,
      "eval_loss": 0.022369742393493652,
      "eval_runtime": 7.9066,
      "eval_samples_per_second": 6.324,
      "eval_steps_per_second": 1.644,
      "step": 155
    },
    {
      "epoch": 1.5639097744360901,
      "grad_norm": 0.0316733792424202,
      "learning_rate": 8.66411585313647e-05,
      "loss": 0.0146,
      "step": 156
    },
    {
      "epoch": 1.573934837092732,
      "grad_norm": 0.027159877121448517,
      "learning_rate": 8.640006768083647e-05,
      "loss": 0.0124,
      "step": 157
    },
    {
      "epoch": 1.5839598997493733,
      "grad_norm": 0.03147249296307564,
      "learning_rate": 8.615716265141123e-05,
      "loss": 0.017,
      "step": 158
    },
    {
      "epoch": 1.5939849624060152,
      "grad_norm": 0.018718725070357323,
      "learning_rate": 8.59124555494725e-05,
      "loss": 0.0096,
      "step": 159
    },
    {
      "epoch": 1.6040100250626566,
      "grad_norm": 0.025683386251330376,
      "learning_rate": 8.566595857121902e-05,
      "loss": 0.0168,
      "step": 160
    },
    {
      "epoch": 1.6040100250626566,
      "eval_loss": 0.022817488759756088,
      "eval_runtime": 7.8759,
      "eval_samples_per_second": 6.349,
      "eval_steps_per_second": 1.651,
      "step": 160
    },
    {
      "epoch": 1.6140350877192984,
      "grad_norm": 0.02304229326546192,
      "learning_rate": 8.541768400205697e-05,
      "loss": 0.0137,
      "step": 161
    },
    {
      "epoch": 1.6240601503759398,
      "grad_norm": 0.029396969825029373,
      "learning_rate": 8.516764421598762e-05,
      "loss": 0.0177,
      "step": 162
    },
    {
      "epoch": 1.6340852130325816,
      "grad_norm": 0.024926166981458664,
      "learning_rate": 8.491585167499066e-05,
      "loss": 0.0151,
      "step": 163
    },
    {
      "epoch": 1.644110275689223,
      "grad_norm": 0.030115393921732903,
      "learning_rate": 8.4662318928403e-05,
      "loss": 0.0161,
      "step": 164
    },
    {
      "epoch": 1.6541353383458648,
      "grad_norm": 0.024759652093052864,
      "learning_rate": 8.440705861229344e-05,
      "loss": 0.0115,
      "step": 165
    },
    {
      "epoch": 1.6541353383458648,
      "eval_loss": 0.022450853139162064,
      "eval_runtime": 7.8428,
      "eval_samples_per_second": 6.375,
      "eval_steps_per_second": 1.658,
      "step": 165
    },
    {
      "epoch": 1.6641604010025062,
      "grad_norm": 0.02288990467786789,
      "learning_rate": 8.415008344883279e-05,
      "loss": 0.0138,
      "step": 166
    },
    {
      "epoch": 1.674185463659148,
      "grad_norm": 0.031593967229127884,
      "learning_rate": 8.389140624565985e-05,
      "loss": 0.0118,
      "step": 167
    },
    {
      "epoch": 1.6842105263157894,
      "grad_norm": 0.023670073598623276,
      "learning_rate": 8.363103989524302e-05,
      "loss": 0.0138,
      "step": 168
    },
    {
      "epoch": 1.6942355889724312,
      "grad_norm": 0.02136528678238392,
      "learning_rate": 8.33689973742378e-05,
      "loss": 0.0107,
      "step": 169
    },
    {
      "epoch": 1.7042606516290726,
      "grad_norm": 0.025891460478305817,
      "learning_rate": 8.310529174284004e-05,
      "loss": 0.0156,
      "step": 170
    },
    {
      "epoch": 1.7042606516290726,
      "eval_loss": 0.022050362080335617,
      "eval_runtime": 7.8487,
      "eval_samples_per_second": 6.37,
      "eval_steps_per_second": 1.656,
      "step": 170
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 0.02511603571474552,
      "learning_rate": 8.28399361441349e-05,
      "loss": 0.0122,
      "step": 171
    },
    {
      "epoch": 1.7243107769423558,
      "grad_norm": 0.027994632720947266,
      "learning_rate": 8.257294380344198e-05,
      "loss": 0.013,
      "step": 172
    },
    {
      "epoch": 1.7343358395989976,
      "grad_norm": 0.03440297394990921,
      "learning_rate": 8.230432802765607e-05,
      "loss": 0.0174,
      "step": 173
    },
    {
      "epoch": 1.744360902255639,
      "grad_norm": 0.03783298283815384,
      "learning_rate": 8.20341022045839e-05,
      "loss": 0.0107,
      "step": 174
    },
    {
      "epoch": 1.7543859649122808,
      "grad_norm": 0.050754014402627945,
      "learning_rate": 8.176227980227694e-05,
      "loss": 0.0174,
      "step": 175
    },
    {
      "epoch": 1.7543859649122808,
      "eval_loss": 0.021764207631349564,
      "eval_runtime": 7.8631,
      "eval_samples_per_second": 6.359,
      "eval_steps_per_second": 1.653,
      "step": 175
    },
    {
      "epoch": 1.7644110275689222,
      "grad_norm": 0.02387530915439129,
      "learning_rate": 8.148887436836018e-05,
      "loss": 0.0125,
      "step": 176
    },
    {
      "epoch": 1.774436090225564,
      "grad_norm": 0.0533299446105957,
      "learning_rate": 8.121389952935687e-05,
      "loss": 0.015,
      "step": 177
    },
    {
      "epoch": 1.7844611528822054,
      "grad_norm": 0.026894379407167435,
      "learning_rate": 8.093736899000939e-05,
      "loss": 0.0135,
      "step": 178
    },
    {
      "epoch": 1.7944862155388472,
      "grad_norm": 0.032237350940704346,
      "learning_rate": 8.065929653259621e-05,
      "loss": 0.0183,
      "step": 179
    },
    {
      "epoch": 1.8045112781954886,
      "grad_norm": 0.026856914162635803,
      "learning_rate": 8.037969601624495e-05,
      "loss": 0.0147,
      "step": 180
    },
    {
      "epoch": 1.8045112781954886,
      "eval_loss": 0.021373717114329338,
      "eval_runtime": 7.8519,
      "eval_samples_per_second": 6.368,
      "eval_steps_per_second": 1.656,
      "step": 180
    },
    {
      "epoch": 1.8145363408521304,
      "grad_norm": 0.026055332273244858,
      "learning_rate": 8.009858137624171e-05,
      "loss": 0.0141,
      "step": 181
    },
    {
      "epoch": 1.8245614035087718,
      "grad_norm": 0.02705809473991394,
      "learning_rate": 7.981596662333645e-05,
      "loss": 0.0107,
      "step": 182
    },
    {
      "epoch": 1.8345864661654137,
      "grad_norm": 0.025101376697421074,
      "learning_rate": 7.953186584304476e-05,
      "loss": 0.0109,
      "step": 183
    },
    {
      "epoch": 1.844611528822055,
      "grad_norm": 0.027133477851748466,
      "learning_rate": 7.924629319494583e-05,
      "loss": 0.0113,
      "step": 184
    },
    {
      "epoch": 1.8546365914786969,
      "grad_norm": 0.028486840426921844,
      "learning_rate": 7.895926291197667e-05,
      "loss": 0.0113,
      "step": 185
    },
    {
      "epoch": 1.8546365914786969,
      "eval_loss": 0.02112492173910141,
      "eval_runtime": 7.8502,
      "eval_samples_per_second": 6.369,
      "eval_steps_per_second": 1.656,
      "step": 185
    },
    {
      "epoch": 1.8646616541353382,
      "grad_norm": 0.03550581634044647,
      "learning_rate": 7.867078929972287e-05,
      "loss": 0.0131,
      "step": 186
    },
    {
      "epoch": 1.87468671679198,
      "grad_norm": 0.027450090274214745,
      "learning_rate": 7.838088673570547e-05,
      "loss": 0.0114,
      "step": 187
    },
    {
      "epoch": 1.8847117794486214,
      "grad_norm": 0.028570324182510376,
      "learning_rate": 7.808956966866446e-05,
      "loss": 0.0128,
      "step": 188
    },
    {
      "epoch": 1.8947368421052633,
      "grad_norm": 0.03284519538283348,
      "learning_rate": 7.779685261783866e-05,
      "loss": 0.0156,
      "step": 189
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.02620542049407959,
      "learning_rate": 7.750275017224207e-05,
      "loss": 0.0128,
      "step": 190
    },
    {
      "epoch": 1.9047619047619047,
      "eval_loss": 0.021003421396017075,
      "eval_runtime": 7.8602,
      "eval_samples_per_second": 6.361,
      "eval_steps_per_second": 1.654,
      "step": 190
    },
    {
      "epoch": 1.9147869674185465,
      "grad_norm": 0.018250567838549614,
      "learning_rate": 7.72072769899367e-05,
      "loss": 0.0101,
      "step": 191
    },
    {
      "epoch": 1.9248120300751879,
      "grad_norm": 0.02913222834467888,
      "learning_rate": 7.691044779730211e-05,
      "loss": 0.0139,
      "step": 192
    },
    {
      "epoch": 1.9348370927318297,
      "grad_norm": 0.02514669857919216,
      "learning_rate": 7.661227738830137e-05,
      "loss": 0.0133,
      "step": 193
    },
    {
      "epoch": 1.944862155388471,
      "grad_norm": 0.025002580136060715,
      "learning_rate": 7.631278062374376e-05,
      "loss": 0.0146,
      "step": 194
    },
    {
      "epoch": 1.954887218045113,
      "grad_norm": 0.03078017756342888,
      "learning_rate": 7.60119724305441e-05,
      "loss": 0.0158,
      "step": 195
    },
    {
      "epoch": 1.954887218045113,
      "eval_loss": 0.020739883184432983,
      "eval_runtime": 7.8876,
      "eval_samples_per_second": 6.339,
      "eval_steps_per_second": 1.648,
      "step": 195
    },
    {
      "epoch": 1.9649122807017543,
      "grad_norm": 0.019327951595187187,
      "learning_rate": 7.570986780097884e-05,
      "loss": 0.0102,
      "step": 196
    },
    {
      "epoch": 1.974937343358396,
      "grad_norm": 0.023565849289298058,
      "learning_rate": 7.540648179193874e-05,
      "loss": 0.0155,
      "step": 197
    },
    {
      "epoch": 1.9849624060150375,
      "grad_norm": 0.025518329814076424,
      "learning_rate": 7.51018295241785e-05,
      "loss": 0.0095,
      "step": 198
    },
    {
      "epoch": 1.9949874686716793,
      "grad_norm": 0.029831720516085625,
      "learning_rate": 7.479592618156321e-05,
      "loss": 0.0146,
      "step": 199
    },
    {
      "epoch": 2.0050125313283207,
      "grad_norm": 0.03035411797463894,
      "learning_rate": 7.448878701031142e-05,
      "loss": 0.0139,
      "step": 200
    },
    {
      "epoch": 2.0050125313283207,
      "eval_loss": 0.02077317051589489,
      "eval_runtime": 7.8495,
      "eval_samples_per_second": 6.37,
      "eval_steps_per_second": 1.656,
      "step": 200
    },
    {
      "epoch": 2.0150375939849625,
      "grad_norm": 0.02455909363925457,
      "learning_rate": 7.418042731823544e-05,
      "loss": 0.011,
      "step": 201
    },
    {
      "epoch": 2.025062656641604,
      "grad_norm": 0.021201658993959427,
      "learning_rate": 7.387086247397826e-05,
      "loss": 0.0085,
      "step": 202
    },
    {
      "epoch": 2.0350877192982457,
      "grad_norm": 0.024311352521181107,
      "learning_rate": 7.356010790624767e-05,
      "loss": 0.0136,
      "step": 203
    },
    {
      "epoch": 2.045112781954887,
      "grad_norm": 0.024387311190366745,
      "learning_rate": 7.324817910304728e-05,
      "loss": 0.0136,
      "step": 204
    },
    {
      "epoch": 2.055137844611529,
      "grad_norm": 0.021270353347063065,
      "learning_rate": 7.293509161090452e-05,
      "loss": 0.0095,
      "step": 205
    },
    {
      "epoch": 2.055137844611529,
      "eval_loss": 0.02157193422317505,
      "eval_runtime": 7.8654,
      "eval_samples_per_second": 6.357,
      "eval_steps_per_second": 1.653,
      "step": 205
    },
    {
      "epoch": 2.0651629072681703,
      "grad_norm": 0.03074588254094124,
      "learning_rate": 7.262086103409597e-05,
      "loss": 0.0105,
      "step": 206
    },
    {
      "epoch": 2.075187969924812,
      "grad_norm": 0.02580447867512703,
      "learning_rate": 7.230550303386942e-05,
      "loss": 0.01,
      "step": 207
    },
    {
      "epoch": 2.0852130325814535,
      "grad_norm": 0.028543729335069656,
      "learning_rate": 7.198903332766347e-05,
      "loss": 0.0133,
      "step": 208
    },
    {
      "epoch": 2.0952380952380953,
      "grad_norm": 0.030399933457374573,
      "learning_rate": 7.167146768832418e-05,
      "loss": 0.011,
      "step": 209
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 0.03573281690478325,
      "learning_rate": 7.13528219433188e-05,
      "loss": 0.0117,
      "step": 210
    },
    {
      "epoch": 2.1052631578947367,
      "eval_loss": 0.021564830094575882,
      "eval_runtime": 7.8471,
      "eval_samples_per_second": 6.372,
      "eval_steps_per_second": 1.657,
      "step": 210
    },
    {
      "epoch": 2.1152882205513786,
      "grad_norm": 0.031177852302789688,
      "learning_rate": 7.103311197394716e-05,
      "loss": 0.0095,
      "step": 211
    },
    {
      "epoch": 2.12531328320802,
      "grad_norm": 0.030712947249412537,
      "learning_rate": 7.071235371454987e-05,
      "loss": 0.0113,
      "step": 212
    },
    {
      "epoch": 2.1353383458646618,
      "grad_norm": 0.028729625046253204,
      "learning_rate": 7.039056315171445e-05,
      "loss": 0.0103,
      "step": 213
    },
    {
      "epoch": 2.145363408521303,
      "grad_norm": 0.03274873271584511,
      "learning_rate": 7.006775632347831e-05,
      "loss": 0.0103,
      "step": 214
    },
    {
      "epoch": 2.155388471177945,
      "grad_norm": 0.030903812497854233,
      "learning_rate": 6.974394931852956e-05,
      "loss": 0.0117,
      "step": 215
    },
    {
      "epoch": 2.155388471177945,
      "eval_loss": 0.020912937819957733,
      "eval_runtime": 7.8659,
      "eval_samples_per_second": 6.357,
      "eval_steps_per_second": 1.653,
      "step": 215
    },
    {
      "epoch": 2.1654135338345863,
      "grad_norm": 0.02950100228190422,
      "learning_rate": 6.941915827540509e-05,
      "loss": 0.0086,
      "step": 216
    },
    {
      "epoch": 2.175438596491228,
      "grad_norm": 0.0353955514729023,
      "learning_rate": 6.909339938168623e-05,
      "loss": 0.012,
      "step": 217
    },
    {
      "epoch": 2.1854636591478696,
      "grad_norm": 0.020289815962314606,
      "learning_rate": 6.876668887319198e-05,
      "loss": 0.0079,
      "step": 218
    },
    {
      "epoch": 2.1954887218045114,
      "grad_norm": 0.03004349209368229,
      "learning_rate": 6.843904303316983e-05,
      "loss": 0.0109,
      "step": 219
    },
    {
      "epoch": 2.2055137844611528,
      "grad_norm": 0.025885947048664093,
      "learning_rate": 6.811047819148413e-05,
      "loss": 0.0098,
      "step": 220
    },
    {
      "epoch": 2.2055137844611528,
      "eval_loss": 0.02112417295575142,
      "eval_runtime": 7.8566,
      "eval_samples_per_second": 6.364,
      "eval_steps_per_second": 1.655,
      "step": 220
    },
    {
      "epoch": 2.2155388471177946,
      "grad_norm": 0.02360019087791443,
      "learning_rate": 6.77810107238023e-05,
      "loss": 0.0104,
      "step": 221
    },
    {
      "epoch": 2.225563909774436,
      "grad_norm": 0.0300395917147398,
      "learning_rate": 6.745065705077864e-05,
      "loss": 0.008,
      "step": 222
    },
    {
      "epoch": 2.235588972431078,
      "grad_norm": 0.03650784492492676,
      "learning_rate": 6.711943363723588e-05,
      "loss": 0.0146,
      "step": 223
    },
    {
      "epoch": 2.245614035087719,
      "grad_norm": 0.0672539472579956,
      "learning_rate": 6.678735699134459e-05,
      "loss": 0.0153,
      "step": 224
    },
    {
      "epoch": 2.255639097744361,
      "grad_norm": 0.03322110325098038,
      "learning_rate": 6.64544436638005e-05,
      "loss": 0.0116,
      "step": 225
    },
    {
      "epoch": 2.255639097744361,
      "eval_loss": 0.020799538120627403,
      "eval_runtime": 7.858,
      "eval_samples_per_second": 6.363,
      "eval_steps_per_second": 1.654,
      "step": 225
    },
    {
      "epoch": 2.2656641604010024,
      "grad_norm": 0.031463854014873505,
      "learning_rate": 6.612071024699948e-05,
      "loss": 0.0075,
      "step": 226
    },
    {
      "epoch": 2.275689223057644,
      "grad_norm": 0.031025653705000877,
      "learning_rate": 6.578617337421064e-05,
      "loss": 0.0124,
      "step": 227
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 0.03126023709774017,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0119,
      "step": 228
    },
    {
      "epoch": 2.2957393483709274,
      "grad_norm": 0.021827111020684242,
      "learning_rate": 6.511475599313627e-05,
      "loss": 0.0117,
      "step": 229
    },
    {
      "epoch": 2.305764411027569,
      "grad_norm": 0.019596287980675697,
      "learning_rate": 6.477790894828421e-05,
      "loss": 0.0091,
      "step": 230
    },
    {
      "epoch": 2.305764411027569,
      "eval_loss": 0.021051470190286636,
      "eval_runtime": 7.8473,
      "eval_samples_per_second": 6.372,
      "eval_steps_per_second": 1.657,
      "step": 230
    },
    {
      "epoch": 2.3157894736842106,
      "grad_norm": 0.0325470045208931,
      "learning_rate": 6.44403253726435e-05,
      "loss": 0.0144,
      "step": 231
    },
    {
      "epoch": 2.325814536340852,
      "grad_norm": 0.023597577586770058,
      "learning_rate": 6.410202209137515e-05,
      "loss": 0.0114,
      "step": 232
    },
    {
      "epoch": 2.335839598997494,
      "grad_norm": 0.023727184161543846,
      "learning_rate": 6.376301596551025e-05,
      "loss": 0.008,
      "step": 233
    },
    {
      "epoch": 2.345864661654135,
      "grad_norm": 0.024764055386185646,
      "learning_rate": 6.342332389110971e-05,
      "loss": 0.0081,
      "step": 234
    },
    {
      "epoch": 2.355889724310777,
      "grad_norm": 0.033406369388103485,
      "learning_rate": 6.308296279842205e-05,
      "loss": 0.0144,
      "step": 235
    },
    {
      "epoch": 2.355889724310777,
      "eval_loss": 0.020991137251257896,
      "eval_runtime": 7.8508,
      "eval_samples_per_second": 6.369,
      "eval_steps_per_second": 1.656,
      "step": 235
    },
    {
      "epoch": 2.3659147869674184,
      "grad_norm": 0.023914620280265808,
      "learning_rate": 6.274194965103965e-05,
      "loss": 0.0105,
      "step": 236
    },
    {
      "epoch": 2.3759398496240602,
      "grad_norm": 0.033582258969545364,
      "learning_rate": 6.240030144505339e-05,
      "loss": 0.011,
      "step": 237
    },
    {
      "epoch": 2.3859649122807016,
      "grad_norm": 0.02053799293935299,
      "learning_rate": 6.205803520820531e-05,
      "loss": 0.0088,
      "step": 238
    },
    {
      "epoch": 2.3959899749373434,
      "grad_norm": 0.022256596013903618,
      "learning_rate": 6.171516799904022e-05,
      "loss": 0.0116,
      "step": 239
    },
    {
      "epoch": 2.406015037593985,
      "grad_norm": 0.02922326698899269,
      "learning_rate": 6.137171690605533e-05,
      "loss": 0.0128,
      "step": 240
    },
    {
      "epoch": 2.406015037593985,
      "eval_loss": 0.021076854318380356,
      "eval_runtime": 7.8443,
      "eval_samples_per_second": 6.374,
      "eval_steps_per_second": 1.657,
      "step": 240
    },
    {
      "epoch": 2.4160401002506267,
      "grad_norm": 0.02965260110795498,
      "learning_rate": 6.1027699046848595e-05,
      "loss": 0.009,
      "step": 241
    },
    {
      "epoch": 2.426065162907268,
      "grad_norm": 0.02778303064405918,
      "learning_rate": 6.068313156726567e-05,
      "loss": 0.0117,
      "step": 242
    },
    {
      "epoch": 2.43609022556391,
      "grad_norm": 0.017490755766630173,
      "learning_rate": 6.033803164054519e-05,
      "loss": 0.0093,
      "step": 243
    },
    {
      "epoch": 2.4461152882205512,
      "grad_norm": 0.02218572050333023,
      "learning_rate": 5.999241646646301e-05,
      "loss": 0.0115,
      "step": 244
    },
    {
      "epoch": 2.456140350877193,
      "grad_norm": 0.024667710065841675,
      "learning_rate": 5.9646303270474845e-05,
      "loss": 0.0097,
      "step": 245
    },
    {
      "epoch": 2.456140350877193,
      "eval_loss": 0.020879259333014488,
      "eval_runtime": 7.8715,
      "eval_samples_per_second": 6.352,
      "eval_steps_per_second": 1.652,
      "step": 245
    },
    {
      "epoch": 2.4661654135338344,
      "grad_norm": 0.032372280955314636,
      "learning_rate": 5.92997093028579e-05,
      "loss": 0.0121,
      "step": 246
    },
    {
      "epoch": 2.4761904761904763,
      "grad_norm": 0.027126269415020943,
      "learning_rate": 5.895265183785097e-05,
      "loss": 0.0109,
      "step": 247
    },
    {
      "epoch": 2.4862155388471177,
      "grad_norm": 0.017588460817933083,
      "learning_rate": 5.8605148172793546e-05,
      "loss": 0.0062,
      "step": 248
    },
    {
      "epoch": 2.4962406015037595,
      "grad_norm": 0.037390079349279404,
      "learning_rate": 5.8257215627263794e-05,
      "loss": 0.0131,
      "step": 249
    },
    {
      "epoch": 2.506265664160401,
      "grad_norm": 0.03616498038172722,
      "learning_rate": 5.79088715422152e-05,
      "loss": 0.0137,
      "step": 250
    },
    {
      "epoch": 2.506265664160401,
      "eval_loss": 0.02056640014052391,
      "eval_runtime": 7.8501,
      "eval_samples_per_second": 6.369,
      "eval_steps_per_second": 1.656,
      "step": 250
    },
    {
      "epoch": 2.5162907268170427,
      "grad_norm": 0.014069591648876667,
      "learning_rate": 5.7560133279112415e-05,
      "loss": 0.0059,
      "step": 251
    },
    {
      "epoch": 2.526315789473684,
      "grad_norm": 0.017873914912343025,
      "learning_rate": 5.7211018219065916e-05,
      "loss": 0.0069,
      "step": 252
    },
    {
      "epoch": 2.536340852130326,
      "grad_norm": 0.025818752124905586,
      "learning_rate": 5.686154376196572e-05,
      "loss": 0.0115,
      "step": 253
    },
    {
      "epoch": 2.5463659147869673,
      "grad_norm": 0.03378435969352722,
      "learning_rate": 5.651172732561419e-05,
      "loss": 0.0146,
      "step": 254
    },
    {
      "epoch": 2.556390977443609,
      "grad_norm": 0.041657060384750366,
      "learning_rate": 5.616158634485793e-05,
      "loss": 0.0163,
      "step": 255
    },
    {
      "epoch": 2.556390977443609,
      "eval_loss": 0.020469658076763153,
      "eval_runtime": 7.8576,
      "eval_samples_per_second": 6.363,
      "eval_steps_per_second": 1.654,
      "step": 255
    },
    {
      "epoch": 2.5664160401002505,
      "grad_norm": 0.01962040551006794,
      "learning_rate": 5.581113827071889e-05,
      "loss": 0.0073,
      "step": 256
    },
    {
      "epoch": 2.5764411027568923,
      "grad_norm": 0.023728540167212486,
      "learning_rate": 5.546040056952443e-05,
      "loss": 0.012,
      "step": 257
    },
    {
      "epoch": 2.5864661654135337,
      "grad_norm": 0.027505788952112198,
      "learning_rate": 5.510939072203704e-05,
      "loss": 0.0081,
      "step": 258
    },
    {
      "epoch": 2.5964912280701755,
      "grad_norm": 0.028974315151572227,
      "learning_rate": 5.4758126222582916e-05,
      "loss": 0.0102,
      "step": 259
    },
    {
      "epoch": 2.606516290726817,
      "grad_norm": 0.02385525219142437,
      "learning_rate": 5.4406624578180096e-05,
      "loss": 0.0104,
      "step": 260
    },
    {
      "epoch": 2.606516290726817,
      "eval_loss": 0.020320262759923935,
      "eval_runtime": 7.8519,
      "eval_samples_per_second": 6.368,
      "eval_steps_per_second": 1.656,
      "step": 260
    },
    {
      "epoch": 2.6165413533834587,
      "grad_norm": 0.03372911363840103,
      "learning_rate": 5.405490330766595e-05,
      "loss": 0.0132,
      "step": 261
    },
    {
      "epoch": 2.6265664160401,
      "grad_norm": 0.03256263956427574,
      "learning_rate": 5.3702979940824004e-05,
      "loss": 0.0111,
      "step": 262
    },
    {
      "epoch": 2.636591478696742,
      "grad_norm": 0.03408624231815338,
      "learning_rate": 5.335087201751026e-05,
      "loss": 0.0085,
      "step": 263
    },
    {
      "epoch": 2.6466165413533833,
      "grad_norm": 0.030111782252788544,
      "learning_rate": 5.2998597086779e-05,
      "loss": 0.0138,
      "step": 264
    },
    {
      "epoch": 2.656641604010025,
      "grad_norm": 0.03473125770688057,
      "learning_rate": 5.2646172706008156e-05,
      "loss": 0.0124,
      "step": 265
    },
    {
      "epoch": 2.656641604010025,
      "eval_loss": 0.020433049649000168,
      "eval_runtime": 7.8438,
      "eval_samples_per_second": 6.374,
      "eval_steps_per_second": 1.657,
      "step": 265
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.02227642573416233,
      "learning_rate": 5.2293616440024286e-05,
      "loss": 0.0072,
      "step": 266
    },
    {
      "epoch": 2.6766917293233083,
      "grad_norm": 0.02503472939133644,
      "learning_rate": 5.194094586022703e-05,
      "loss": 0.0093,
      "step": 267
    },
    {
      "epoch": 2.6867167919799497,
      "grad_norm": 0.022726213559508324,
      "learning_rate": 5.158817854371349e-05,
      "loss": 0.0081,
      "step": 268
    },
    {
      "epoch": 2.6967418546365916,
      "grad_norm": 0.02940908446907997,
      "learning_rate": 5.123533207240212e-05,
      "loss": 0.0136,
      "step": 269
    },
    {
      "epoch": 2.706766917293233,
      "grad_norm": 0.028102362528443336,
      "learning_rate": 5.088242403215644e-05,
      "loss": 0.0131,
      "step": 270
    },
    {
      "epoch": 2.706766917293233,
      "eval_loss": 0.020840449258685112,
      "eval_runtime": 7.8435,
      "eval_samples_per_second": 6.375,
      "eval_steps_per_second": 1.657,
      "step": 270
    },
    {
      "epoch": 2.7167919799498748,
      "grad_norm": 0.021073520183563232,
      "learning_rate": 5.052947201190852e-05,
      "loss": 0.0093,
      "step": 271
    },
    {
      "epoch": 2.726817042606516,
      "grad_norm": 0.024549121037125587,
      "learning_rate": 5.017649360278246e-05,
      "loss": 0.0094,
      "step": 272
    },
    {
      "epoch": 2.736842105263158,
      "grad_norm": 0.023642798885703087,
      "learning_rate": 4.982350639721754e-05,
      "loss": 0.0105,
      "step": 273
    },
    {
      "epoch": 2.7468671679197993,
      "grad_norm": 0.030647899955511093,
      "learning_rate": 4.947052798809148e-05,
      "loss": 0.0141,
      "step": 274
    },
    {
      "epoch": 2.756892230576441,
      "grad_norm": 0.022092748433351517,
      "learning_rate": 4.911757596784357e-05,
      "loss": 0.0089,
      "step": 275
    },
    {
      "epoch": 2.756892230576441,
      "eval_loss": 0.020474795252084732,
      "eval_runtime": 7.8473,
      "eval_samples_per_second": 6.372,
      "eval_steps_per_second": 1.657,
      "step": 275
    },
    {
      "epoch": 2.7669172932330826,
      "grad_norm": 0.02058694139122963,
      "learning_rate": 4.8764667927597884e-05,
      "loss": 0.0082,
      "step": 276
    },
    {
      "epoch": 2.7769423558897244,
      "grad_norm": 0.026842277497053146,
      "learning_rate": 4.841182145628652e-05,
      "loss": 0.0092,
      "step": 277
    },
    {
      "epoch": 2.7869674185463658,
      "grad_norm": 0.03636394441127777,
      "learning_rate": 4.8059054139772996e-05,
      "loss": 0.0121,
      "step": 278
    },
    {
      "epoch": 2.7969924812030076,
      "grad_norm": 0.02889190800487995,
      "learning_rate": 4.770638355997574e-05,
      "loss": 0.0106,
      "step": 279
    },
    {
      "epoch": 2.807017543859649,
      "grad_norm": 0.027845360338687897,
      "learning_rate": 4.735382729399184e-05,
      "loss": 0.0093,
      "step": 280
    },
    {
      "epoch": 2.807017543859649,
      "eval_loss": 0.020694730803370476,
      "eval_runtime": 7.846,
      "eval_samples_per_second": 6.373,
      "eval_steps_per_second": 1.657,
      "step": 280
    },
    {
      "epoch": 2.817042606516291,
      "grad_norm": 0.04447019472718239,
      "learning_rate": 4.7001402913221016e-05,
      "loss": 0.014,
      "step": 281
    },
    {
      "epoch": 2.827067669172932,
      "grad_norm": 0.02357538230717182,
      "learning_rate": 4.6649127982489746e-05,
      "loss": 0.0052,
      "step": 282
    },
    {
      "epoch": 2.837092731829574,
      "grad_norm": 0.0315440371632576,
      "learning_rate": 4.629702005917601e-05,
      "loss": 0.0161,
      "step": 283
    },
    {
      "epoch": 2.8471177944862154,
      "grad_norm": 0.016042588278651237,
      "learning_rate": 4.594509669233406e-05,
      "loss": 0.0075,
      "step": 284
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.02733796462416649,
      "learning_rate": 4.559337542181993e-05,
      "loss": 0.0139,
      "step": 285
    },
    {
      "epoch": 2.857142857142857,
      "eval_loss": 0.02120617963373661,
      "eval_runtime": 7.8783,
      "eval_samples_per_second": 6.347,
      "eval_steps_per_second": 1.65,
      "step": 285
    },
    {
      "epoch": 2.8671679197994986,
      "grad_norm": 0.023688923567533493,
      "learning_rate": 4.5241873777417096e-05,
      "loss": 0.009,
      "step": 286
    },
    {
      "epoch": 2.8771929824561404,
      "grad_norm": 0.026211708784103394,
      "learning_rate": 4.489060927796297e-05,
      "loss": 0.01,
      "step": 287
    },
    {
      "epoch": 2.887218045112782,
      "grad_norm": 0.020815694704651833,
      "learning_rate": 4.4539599430475575e-05,
      "loss": 0.01,
      "step": 288
    },
    {
      "epoch": 2.8972431077694236,
      "grad_norm": 0.02503124438226223,
      "learning_rate": 4.418886172928113e-05,
      "loss": 0.0082,
      "step": 289
    },
    {
      "epoch": 2.907268170426065,
      "grad_norm": 0.03107696771621704,
      "learning_rate": 4.383841365514208e-05,
      "loss": 0.0121,
      "step": 290
    },
    {
      "epoch": 2.907268170426065,
      "eval_loss": 0.02047637850046158,
      "eval_runtime": 7.8528,
      "eval_samples_per_second": 6.367,
      "eval_steps_per_second": 1.655,
      "step": 290
    },
    {
      "epoch": 2.917293233082707,
      "grad_norm": 0.029050437733530998,
      "learning_rate": 4.3488272674385825e-05,
      "loss": 0.0089,
      "step": 291
    },
    {
      "epoch": 2.927318295739348,
      "grad_norm": 0.020520159974694252,
      "learning_rate": 4.313845623803431e-05,
      "loss": 0.0082,
      "step": 292
    },
    {
      "epoch": 2.93734335839599,
      "grad_norm": 0.02317202091217041,
      "learning_rate": 4.278898178093409e-05,
      "loss": 0.0108,
      "step": 293
    },
    {
      "epoch": 2.9473684210526314,
      "grad_norm": 0.02789182774722576,
      "learning_rate": 4.243986672088758e-05,
      "loss": 0.0131,
      "step": 294
    },
    {
      "epoch": 2.9573934837092732,
      "grad_norm": 0.02339678630232811,
      "learning_rate": 4.209112845778481e-05,
      "loss": 0.0101,
      "step": 295
    },
    {
      "epoch": 2.9573934837092732,
      "eval_loss": 0.02035236358642578,
      "eval_runtime": 7.8472,
      "eval_samples_per_second": 6.372,
      "eval_steps_per_second": 1.657,
      "step": 295
    },
    {
      "epoch": 2.9674185463659146,
      "grad_norm": 0.034076038748025894,
      "learning_rate": 4.174278437273621e-05,
      "loss": 0.0062,
      "step": 296
    },
    {
      "epoch": 2.9774436090225564,
      "grad_norm": 0.026215389370918274,
      "learning_rate": 4.139485182720646e-05,
      "loss": 0.0113,
      "step": 297
    },
    {
      "epoch": 2.987468671679198,
      "grad_norm": 0.021045586094260216,
      "learning_rate": 4.104734816214905e-05,
      "loss": 0.0082,
      "step": 298
    },
    {
      "epoch": 2.9974937343358397,
      "grad_norm": 0.031620148569345474,
      "learning_rate": 4.0700290697142126e-05,
      "loss": 0.0135,
      "step": 299
    },
    {
      "epoch": 3.007518796992481,
      "grad_norm": 0.016665760427713394,
      "learning_rate": 4.035369672952516e-05,
      "loss": 0.0087,
      "step": 300
    },
    {
      "epoch": 3.007518796992481,
      "eval_loss": 0.019945178180933,
      "eval_runtime": 7.8901,
      "eval_samples_per_second": 6.337,
      "eval_steps_per_second": 1.648,
      "step": 300
    },
    {
      "epoch": 3.017543859649123,
      "grad_norm": 0.01904158480465412,
      "learning_rate": 4.000758353353701e-05,
      "loss": 0.0082,
      "step": 301
    },
    {
      "epoch": 3.0275689223057642,
      "grad_norm": 0.019835559651255608,
      "learning_rate": 3.966196835945482e-05,
      "loss": 0.0066,
      "step": 302
    },
    {
      "epoch": 3.037593984962406,
      "grad_norm": 0.02055111713707447,
      "learning_rate": 3.9316868432734335e-05,
      "loss": 0.0083,
      "step": 303
    },
    {
      "epoch": 3.0476190476190474,
      "grad_norm": 0.02135693095624447,
      "learning_rate": 3.897230095315141e-05,
      "loss": 0.0083,
      "step": 304
    },
    {
      "epoch": 3.0576441102756893,
      "grad_norm": 0.01703447662293911,
      "learning_rate": 3.8628283093944686e-05,
      "loss": 0.0079,
      "step": 305
    },
    {
      "epoch": 3.0576441102756893,
      "eval_loss": 0.020382743328809738,
      "eval_runtime": 7.8801,
      "eval_samples_per_second": 6.345,
      "eval_steps_per_second": 1.65,
      "step": 305
    },
    {
      "epoch": 3.0676691729323307,
      "grad_norm": 0.0192066989839077,
      "learning_rate": 3.828483200095978e-05,
      "loss": 0.0086,
      "step": 306
    },
    {
      "epoch": 3.0776942355889725,
      "grad_norm": 0.022239364683628082,
      "learning_rate": 3.79419647917947e-05,
      "loss": 0.0086,
      "step": 307
    },
    {
      "epoch": 3.087719298245614,
      "grad_norm": 0.025497829541563988,
      "learning_rate": 3.7599698554946625e-05,
      "loss": 0.0095,
      "step": 308
    },
    {
      "epoch": 3.0977443609022557,
      "grad_norm": 0.024126721546053886,
      "learning_rate": 3.725805034896035e-05,
      "loss": 0.0076,
      "step": 309
    },
    {
      "epoch": 3.107769423558897,
      "grad_norm": 0.02745652385056019,
      "learning_rate": 3.691703720157798e-05,
      "loss": 0.01,
      "step": 310
    },
    {
      "epoch": 3.107769423558897,
      "eval_loss": 0.020815785974264145,
      "eval_runtime": 7.85,
      "eval_samples_per_second": 6.369,
      "eval_steps_per_second": 1.656,
      "step": 310
    },
    {
      "epoch": 3.117794486215539,
      "grad_norm": 0.032075103372335434,
      "learning_rate": 3.657667610889032e-05,
      "loss": 0.0077,
      "step": 311
    },
    {
      "epoch": 3.1278195488721803,
      "grad_norm": 0.03226759657263756,
      "learning_rate": 3.623698403448976e-05,
      "loss": 0.0093,
      "step": 312
    },
    {
      "epoch": 3.137844611528822,
      "grad_norm": 0.02231232076883316,
      "learning_rate": 3.589797790862485e-05,
      "loss": 0.0067,
      "step": 313
    },
    {
      "epoch": 3.1478696741854635,
      "grad_norm": 0.029100285843014717,
      "learning_rate": 3.555967462735651e-05,
      "loss": 0.0072,
      "step": 314
    },
    {
      "epoch": 3.1578947368421053,
      "grad_norm": 0.030341139063239098,
      "learning_rate": 3.52220910517158e-05,
      "loss": 0.0089,
      "step": 315
    },
    {
      "epoch": 3.1578947368421053,
      "eval_loss": 0.021213330328464508,
      "eval_runtime": 7.8447,
      "eval_samples_per_second": 6.374,
      "eval_steps_per_second": 1.657,
      "step": 315
    },
    {
      "epoch": 3.1679197994987467,
      "grad_norm": 0.024791590869426727,
      "learning_rate": 3.488524400686375e-05,
      "loss": 0.0088,
      "step": 316
    },
    {
      "epoch": 3.1779448621553885,
      "grad_norm": 0.029614470899105072,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.01,
      "step": 317
    },
    {
      "epoch": 3.18796992481203,
      "grad_norm": 0.028596239164471626,
      "learning_rate": 3.421382662578937e-05,
      "loss": 0.0092,
      "step": 318
    },
    {
      "epoch": 3.1979949874686717,
      "grad_norm": 0.038685377687215805,
      "learning_rate": 3.387928975300053e-05,
      "loss": 0.0117,
      "step": 319
    },
    {
      "epoch": 3.208020050125313,
      "grad_norm": 0.03576364368200302,
      "learning_rate": 3.35455563361995e-05,
      "loss": 0.0079,
      "step": 320
    },
    {
      "epoch": 3.208020050125313,
      "eval_loss": 0.020832277834415436,
      "eval_runtime": 7.8486,
      "eval_samples_per_second": 6.371,
      "eval_steps_per_second": 1.656,
      "step": 320
    },
    {
      "epoch": 3.218045112781955,
      "grad_norm": 0.026977384462952614,
      "learning_rate": 3.321264300865541e-05,
      "loss": 0.0073,
      "step": 321
    },
    {
      "epoch": 3.2280701754385963,
      "grad_norm": 0.023794258013367653,
      "learning_rate": 3.288056636276413e-05,
      "loss": 0.0067,
      "step": 322
    },
    {
      "epoch": 3.238095238095238,
      "grad_norm": 0.03268418088555336,
      "learning_rate": 3.2549342949221365e-05,
      "loss": 0.0106,
      "step": 323
    },
    {
      "epoch": 3.2481203007518795,
      "grad_norm": 0.022963467985391617,
      "learning_rate": 3.22189892761977e-05,
      "loss": 0.0082,
      "step": 324
    },
    {
      "epoch": 3.2581453634085213,
      "grad_norm": 0.023204447701573372,
      "learning_rate": 3.188952180851589e-05,
      "loss": 0.006,
      "step": 325
    },
    {
      "epoch": 3.2581453634085213,
      "eval_loss": 0.020569898188114166,
      "eval_runtime": 7.8459,
      "eval_samples_per_second": 6.373,
      "eval_steps_per_second": 1.657,
      "step": 325
    },
    {
      "epoch": 3.2681704260651627,
      "grad_norm": 0.028982898220419884,
      "learning_rate": 3.156095696683018e-05,
      "loss": 0.0085,
      "step": 326
    },
    {
      "epoch": 3.2781954887218046,
      "grad_norm": 0.019690126180648804,
      "learning_rate": 3.123331112680801e-05,
      "loss": 0.0065,
      "step": 327
    },
    {
      "epoch": 3.288220551378446,
      "grad_norm": 0.02583892270922661,
      "learning_rate": 3.090660061831378e-05,
      "loss": 0.0083,
      "step": 328
    },
    {
      "epoch": 3.2982456140350878,
      "grad_norm": 0.025384318083524704,
      "learning_rate": 3.058084172459491e-05,
      "loss": 0.0068,
      "step": 329
    },
    {
      "epoch": 3.308270676691729,
      "grad_norm": 0.030685538426041603,
      "learning_rate": 3.0256050681470444e-05,
      "loss": 0.0094,
      "step": 330
    },
    {
      "epoch": 3.308270676691729,
      "eval_loss": 0.020690646022558212,
      "eval_runtime": 7.8578,
      "eval_samples_per_second": 6.363,
      "eval_steps_per_second": 1.654,
      "step": 330
    },
    {
      "epoch": 3.318295739348371,
      "grad_norm": 0.03488805517554283,
      "learning_rate": 2.9932243676521693e-05,
      "loss": 0.0079,
      "step": 331
    },
    {
      "epoch": 3.3283208020050123,
      "grad_norm": 0.02308155782520771,
      "learning_rate": 2.9609436848285567e-05,
      "loss": 0.0072,
      "step": 332
    },
    {
      "epoch": 3.338345864661654,
      "grad_norm": 0.0337197482585907,
      "learning_rate": 2.9287646285450133e-05,
      "loss": 0.0076,
      "step": 333
    },
    {
      "epoch": 3.3483709273182956,
      "grad_norm": 0.02416362054646015,
      "learning_rate": 2.8966888026052873e-05,
      "loss": 0.0067,
      "step": 334
    },
    {
      "epoch": 3.3583959899749374,
      "grad_norm": 0.026764515787363052,
      "learning_rate": 2.8647178056681194e-05,
      "loss": 0.0091,
      "step": 335
    },
    {
      "epoch": 3.3583959899749374,
      "eval_loss": 0.02051025629043579,
      "eval_runtime": 7.8462,
      "eval_samples_per_second": 6.373,
      "eval_steps_per_second": 1.657,
      "step": 335
    },
    {
      "epoch": 3.3684210526315788,
      "grad_norm": 0.030355772003531456,
      "learning_rate": 2.8328532311675838e-05,
      "loss": 0.0079,
      "step": 336
    },
    {
      "epoch": 3.3784461152882206,
      "grad_norm": 0.02602885104715824,
      "learning_rate": 2.8010966672336536e-05,
      "loss": 0.0083,
      "step": 337
    },
    {
      "epoch": 3.388471177944862,
      "grad_norm": 0.026122696697711945,
      "learning_rate": 2.7694496966130607e-05,
      "loss": 0.0078,
      "step": 338
    },
    {
      "epoch": 3.398496240601504,
      "grad_norm": 0.028853295370936394,
      "learning_rate": 2.7379138965904037e-05,
      "loss": 0.0084,
      "step": 339
    },
    {
      "epoch": 3.408521303258145,
      "grad_norm": 0.020907724276185036,
      "learning_rate": 2.7064908389095468e-05,
      "loss": 0.0077,
      "step": 340
    },
    {
      "epoch": 3.408521303258145,
      "eval_loss": 0.020498408004641533,
      "eval_runtime": 7.8504,
      "eval_samples_per_second": 6.369,
      "eval_steps_per_second": 1.656,
      "step": 340
    },
    {
      "epoch": 3.418546365914787,
      "grad_norm": 0.019861876964569092,
      "learning_rate": 2.675182089695274e-05,
      "loss": 0.0062,
      "step": 341
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 0.028957007452845573,
      "learning_rate": 2.643989209375235e-05,
      "loss": 0.0078,
      "step": 342
    },
    {
      "epoch": 3.43859649122807,
      "grad_norm": 0.02627895213663578,
      "learning_rate": 2.6129137526021773e-05,
      "loss": 0.0073,
      "step": 343
    },
    {
      "epoch": 3.4486215538847116,
      "grad_norm": 0.028672248125076294,
      "learning_rate": 2.581957268176459e-05,
      "loss": 0.0082,
      "step": 344
    },
    {
      "epoch": 3.4586466165413534,
      "grad_norm": 0.021196383982896805,
      "learning_rate": 2.5511212989688586e-05,
      "loss": 0.0074,
      "step": 345
    },
    {
      "epoch": 3.4586466165413534,
      "eval_loss": 0.02022087387740612,
      "eval_runtime": 7.8477,
      "eval_samples_per_second": 6.371,
      "eval_steps_per_second": 1.657,
      "step": 345
    },
    {
      "epoch": 3.468671679197995,
      "grad_norm": 0.024043511599302292,
      "learning_rate": 2.520407381843679e-05,
      "loss": 0.0077,
      "step": 346
    },
    {
      "epoch": 3.4786967418546366,
      "grad_norm": 0.035246286541223526,
      "learning_rate": 2.4898170475821493e-05,
      "loss": 0.0088,
      "step": 347
    },
    {
      "epoch": 3.488721804511278,
      "grad_norm": 0.02104768343269825,
      "learning_rate": 2.4593518208061274e-05,
      "loss": 0.0068,
      "step": 348
    },
    {
      "epoch": 3.49874686716792,
      "grad_norm": 0.025738101452589035,
      "learning_rate": 2.4290132199021176e-05,
      "loss": 0.0068,
      "step": 349
    },
    {
      "epoch": 3.5087719298245617,
      "grad_norm": 0.02881302498281002,
      "learning_rate": 2.3988027569455895e-05,
      "loss": 0.007,
      "step": 350
    },
    {
      "epoch": 3.5087719298245617,
      "eval_loss": 0.020256277173757553,
      "eval_runtime": 7.8561,
      "eval_samples_per_second": 6.364,
      "eval_steps_per_second": 1.655,
      "step": 350
    },
    {
      "epoch": 3.518796992481203,
      "grad_norm": 0.034245338290929794,
      "learning_rate": 2.3687219376256255e-05,
      "loss": 0.0103,
      "step": 351
    },
    {
      "epoch": 3.5288220551378444,
      "grad_norm": 0.021362487226724625,
      "learning_rate": 2.3387722611698653e-05,
      "loss": 0.0063,
      "step": 352
    },
    {
      "epoch": 3.5388471177944862,
      "grad_norm": 0.023916201665997505,
      "learning_rate": 2.30895522026979e-05,
      "loss": 0.0073,
      "step": 353
    },
    {
      "epoch": 3.548872180451128,
      "grad_norm": 0.022715197876095772,
      "learning_rate": 2.2792723010063316e-05,
      "loss": 0.0076,
      "step": 354
    },
    {
      "epoch": 3.5588972431077694,
      "grad_norm": 0.03498687595129013,
      "learning_rate": 2.2497249827757933e-05,
      "loss": 0.0087,
      "step": 355
    },
    {
      "epoch": 3.5588972431077694,
      "eval_loss": 0.020144272595643997,
      "eval_runtime": 7.8441,
      "eval_samples_per_second": 6.374,
      "eval_steps_per_second": 1.657,
      "step": 355
    },
    {
      "epoch": 3.568922305764411,
      "grad_norm": 0.019907698035240173,
      "learning_rate": 2.2203147382161336e-05,
      "loss": 0.0063,
      "step": 356
    },
    {
      "epoch": 3.5789473684210527,
      "grad_norm": 0.02432534098625183,
      "learning_rate": 2.1910430331335553e-05,
      "loss": 0.0075,
      "step": 357
    },
    {
      "epoch": 3.5889724310776945,
      "grad_norm": 0.03905978426337242,
      "learning_rate": 2.1619113264294554e-05,
      "loss": 0.0078,
      "step": 358
    },
    {
      "epoch": 3.598997493734336,
      "grad_norm": 0.024142051115632057,
      "learning_rate": 2.132921070027714e-05,
      "loss": 0.0075,
      "step": 359
    },
    {
      "epoch": 3.6090225563909772,
      "grad_norm": 0.023508962243795395,
      "learning_rate": 2.1040737088023323e-05,
      "loss": 0.0067,
      "step": 360
    },
    {
      "epoch": 3.6090225563909772,
      "eval_loss": 0.02008051984012127,
      "eval_runtime": 7.8442,
      "eval_samples_per_second": 6.374,
      "eval_steps_per_second": 1.657,
      "step": 360
    },
    {
      "epoch": 3.619047619047619,
      "grad_norm": 0.04632345959544182,
      "learning_rate": 2.0753706805054185e-05,
      "loss": 0.0109,
      "step": 361
    },
    {
      "epoch": 3.629072681704261,
      "grad_norm": 0.03106398694217205,
      "learning_rate": 2.0468134156955254e-05,
      "loss": 0.0069,
      "step": 362
    },
    {
      "epoch": 3.6390977443609023,
      "grad_norm": 0.032094817608594894,
      "learning_rate": 2.0184033376663573e-05,
      "loss": 0.0081,
      "step": 363
    },
    {
      "epoch": 3.6491228070175437,
      "grad_norm": 0.025334365665912628,
      "learning_rate": 1.99014186237583e-05,
      "loss": 0.0079,
      "step": 364
    },
    {
      "epoch": 3.6591478696741855,
      "grad_norm": 0.023328186944127083,
      "learning_rate": 1.962030398375506e-05,
      "loss": 0.007,
      "step": 365
    },
    {
      "epoch": 3.6591478696741855,
      "eval_loss": 0.020097460597753525,
      "eval_runtime": 7.8736,
      "eval_samples_per_second": 6.35,
      "eval_steps_per_second": 1.651,
      "step": 365
    },
    {
      "epoch": 3.6691729323308273,
      "grad_norm": 0.02302337810397148,
      "learning_rate": 1.9340703467403793e-05,
      "loss": 0.0056,
      "step": 366
    },
    {
      "epoch": 3.6791979949874687,
      "grad_norm": 0.024331629276275635,
      "learning_rate": 1.9062631009990612e-05,
      "loss": 0.0088,
      "step": 367
    },
    {
      "epoch": 3.68922305764411,
      "grad_norm": 0.02797406166791916,
      "learning_rate": 1.8786100470643143e-05,
      "loss": 0.0067,
      "step": 368
    },
    {
      "epoch": 3.699248120300752,
      "grad_norm": 0.028790263459086418,
      "learning_rate": 1.8511125631639826e-05,
      "loss": 0.0082,
      "step": 369
    },
    {
      "epoch": 3.7092731829573937,
      "grad_norm": 0.02699551172554493,
      "learning_rate": 1.8237720197723075e-05,
      "loss": 0.006,
      "step": 370
    },
    {
      "epoch": 3.7092731829573937,
      "eval_loss": 0.019944829866290092,
      "eval_runtime": 7.875,
      "eval_samples_per_second": 6.349,
      "eval_steps_per_second": 1.651,
      "step": 370
    },
    {
      "epoch": 3.719298245614035,
      "grad_norm": 0.04655482620000839,
      "learning_rate": 1.7965897795416127e-05,
      "loss": 0.0116,
      "step": 371
    },
    {
      "epoch": 3.7293233082706765,
      "grad_norm": 0.023725813254714012,
      "learning_rate": 1.769567197234395e-05,
      "loss": 0.0043,
      "step": 372
    },
    {
      "epoch": 3.7393483709273183,
      "grad_norm": 0.03052056021988392,
      "learning_rate": 1.742705619655802e-05,
      "loss": 0.0077,
      "step": 373
    },
    {
      "epoch": 3.74937343358396,
      "grad_norm": 0.022183619439601898,
      "learning_rate": 1.7160063855865106e-05,
      "loss": 0.0065,
      "step": 374
    },
    {
      "epoch": 3.7593984962406015,
      "grad_norm": 0.025241268798708916,
      "learning_rate": 1.689470825715998e-05,
      "loss": 0.0073,
      "step": 375
    },
    {
      "epoch": 3.7593984962406015,
      "eval_loss": 0.019921263679862022,
      "eval_runtime": 7.8625,
      "eval_samples_per_second": 6.359,
      "eval_steps_per_second": 1.653,
      "step": 375
    },
    {
      "epoch": 3.769423558897243,
      "grad_norm": 0.027392975986003876,
      "learning_rate": 1.6631002625762204e-05,
      "loss": 0.0088,
      "step": 376
    },
    {
      "epoch": 3.7794486215538847,
      "grad_norm": 0.029222341254353523,
      "learning_rate": 1.6368960104757e-05,
      "loss": 0.0078,
      "step": 377
    },
    {
      "epoch": 3.7894736842105265,
      "grad_norm": 0.02577180787920952,
      "learning_rate": 1.6108593754340158e-05,
      "loss": 0.0047,
      "step": 378
    },
    {
      "epoch": 3.799498746867168,
      "grad_norm": 0.037013255059719086,
      "learning_rate": 1.5849916551167198e-05,
      "loss": 0.0106,
      "step": 379
    },
    {
      "epoch": 3.8095238095238093,
      "grad_norm": 0.020461726933717728,
      "learning_rate": 1.559294138770656e-05,
      "loss": 0.0071,
      "step": 380
    },
    {
      "epoch": 3.8095238095238093,
      "eval_loss": 0.019906265661120415,
      "eval_runtime": 7.8481,
      "eval_samples_per_second": 6.371,
      "eval_steps_per_second": 1.656,
      "step": 380
    },
    {
      "epoch": 3.819548872180451,
      "grad_norm": 0.021040087565779686,
      "learning_rate": 1.533768107159701e-05,
      "loss": 0.0064,
      "step": 381
    },
    {
      "epoch": 3.829573934837093,
      "grad_norm": 0.019402172416448593,
      "learning_rate": 1.5084148325009367e-05,
      "loss": 0.0068,
      "step": 382
    },
    {
      "epoch": 3.8395989974937343,
      "grad_norm": 0.02361433021724224,
      "learning_rate": 1.4832355784012386e-05,
      "loss": 0.0074,
      "step": 383
    },
    {
      "epoch": 3.8496240601503757,
      "grad_norm": 0.02696828916668892,
      "learning_rate": 1.4582315997943046e-05,
      "loss": 0.0084,
      "step": 384
    },
    {
      "epoch": 3.8596491228070176,
      "grad_norm": 0.040230002254247665,
      "learning_rate": 1.4334041428781003e-05,
      "loss": 0.01,
      "step": 385
    },
    {
      "epoch": 3.8596491228070176,
      "eval_loss": 0.0194709375500679,
      "eval_runtime": 7.8884,
      "eval_samples_per_second": 6.338,
      "eval_steps_per_second": 1.648,
      "step": 385
    },
    {
      "epoch": 3.8696741854636594,
      "grad_norm": 0.0518641360104084,
      "learning_rate": 1.4087544450527513e-05,
      "loss": 0.0096,
      "step": 386
    },
    {
      "epoch": 3.8796992481203008,
      "grad_norm": 0.025948142632842064,
      "learning_rate": 1.3842837348588783e-05,
      "loss": 0.0073,
      "step": 387
    },
    {
      "epoch": 3.889724310776942,
      "grad_norm": 0.025720641016960144,
      "learning_rate": 1.3599932319163538e-05,
      "loss": 0.0061,
      "step": 388
    },
    {
      "epoch": 3.899749373433584,
      "grad_norm": 0.052690256386995316,
      "learning_rate": 1.3358841468635302e-05,
      "loss": 0.0103,
      "step": 389
    },
    {
      "epoch": 3.909774436090226,
      "grad_norm": 0.02759932167828083,
      "learning_rate": 1.3119576812968892e-05,
      "loss": 0.0081,
      "step": 390
    },
    {
      "epoch": 3.909774436090226,
      "eval_loss": 0.019465075805783272,
      "eval_runtime": 7.8746,
      "eval_samples_per_second": 6.35,
      "eval_steps_per_second": 1.651,
      "step": 390
    },
    {
      "epoch": 3.919799498746867,
      "grad_norm": 0.028103185817599297,
      "learning_rate": 1.2882150277111621e-05,
      "loss": 0.0079,
      "step": 391
    },
    {
      "epoch": 3.9298245614035086,
      "grad_norm": 0.024139197543263435,
      "learning_rate": 1.2646573694398906e-05,
      "loss": 0.0046,
      "step": 392
    },
    {
      "epoch": 3.9398496240601504,
      "grad_norm": 0.024406850337982178,
      "learning_rate": 1.2412858805964573e-05,
      "loss": 0.0084,
      "step": 393
    },
    {
      "epoch": 3.949874686716792,
      "grad_norm": 0.02168712578713894,
      "learning_rate": 1.2181017260155608e-05,
      "loss": 0.0073,
      "step": 394
    },
    {
      "epoch": 3.9598997493734336,
      "grad_norm": 0.020631562918424606,
      "learning_rate": 1.1951060611951615e-05,
      "loss": 0.0077,
      "step": 395
    },
    {
      "epoch": 3.9598997493734336,
      "eval_loss": 0.019751163199543953,
      "eval_runtime": 7.853,
      "eval_samples_per_second": 6.367,
      "eval_steps_per_second": 1.655,
      "step": 395
    },
    {
      "epoch": 3.969924812030075,
      "grad_norm": 0.026922499760985374,
      "learning_rate": 1.1723000322388927e-05,
      "loss": 0.0084,
      "step": 396
    },
    {
      "epoch": 3.979949874686717,
      "grad_norm": 0.03158457204699516,
      "learning_rate": 1.1496847757989381e-05,
      "loss": 0.0105,
      "step": 397
    },
    {
      "epoch": 3.9899749373433586,
      "grad_norm": 0.022796666249632835,
      "learning_rate": 1.1272614190193853e-05,
      "loss": 0.0072,
      "step": 398
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.04201427102088928,
      "learning_rate": 1.1050310794800405e-05,
      "loss": 0.0116,
      "step": 399
    },
    {
      "epoch": 4.010025062656641,
      "grad_norm": 0.02010524645447731,
      "learning_rate": 1.0829948651407374e-05,
      "loss": 0.007,
      "step": 400
    },
    {
      "epoch": 4.010025062656641,
      "eval_loss": 0.019864432513713837,
      "eval_runtime": 7.853,
      "eval_samples_per_second": 6.367,
      "eval_steps_per_second": 1.655,
      "step": 400
    },
    {
      "epoch": 4.020050125313283,
      "grad_norm": 0.017796291038393974,
      "learning_rate": 1.0611538742861099e-05,
      "loss": 0.007,
      "step": 401
    },
    {
      "epoch": 4.030075187969925,
      "grad_norm": 0.02786899544298649,
      "learning_rate": 1.0395091954708546e-05,
      "loss": 0.007,
      "step": 402
    },
    {
      "epoch": 4.040100250626566,
      "grad_norm": 0.025612663477659225,
      "learning_rate": 1.0180619074654835e-05,
      "loss": 0.0057,
      "step": 403
    },
    {
      "epoch": 4.050125313283208,
      "grad_norm": 0.038706522434949875,
      "learning_rate": 9.968130792025487e-06,
      "loss": 0.0076,
      "step": 404
    },
    {
      "epoch": 4.06015037593985,
      "grad_norm": 0.01871400512754917,
      "learning_rate": 9.757637697233723e-06,
      "loss": 0.0052,
      "step": 405
    },
    {
      "epoch": 4.06015037593985,
      "eval_loss": 0.019771499559283257,
      "eval_runtime": 7.8796,
      "eval_samples_per_second": 6.346,
      "eval_steps_per_second": 1.65,
      "step": 405
    },
    {
      "epoch": 4.0701754385964914,
      "grad_norm": 0.0155571224167943,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0054,
      "step": 406
    },
    {
      "epoch": 4.080200501253133,
      "grad_norm": 0.01952575147151947,
      "learning_rate": 9.342678935092303e-06,
      "loss": 0.0062,
      "step": 407
    },
    {
      "epoch": 4.090225563909774,
      "grad_norm": 0.03393295407295227,
      "learning_rate": 9.138233949281944e-06,
      "loss": 0.0064,
      "step": 408
    },
    {
      "epoch": 4.100250626566416,
      "grad_norm": 0.01785857044160366,
      "learning_rate": 8.935825513356954e-06,
      "loss": 0.0059,
      "step": 409
    },
    {
      "epoch": 4.110275689223058,
      "grad_norm": 0.020307229831814766,
      "learning_rate": 8.735463715351139e-06,
      "loss": 0.0068,
      "step": 410
    },
    {
      "epoch": 4.110275689223058,
      "eval_loss": 0.019862821325659752,
      "eval_runtime": 7.8486,
      "eval_samples_per_second": 6.371,
      "eval_steps_per_second": 1.656,
      "step": 410
    },
    {
      "epoch": 4.120300751879699,
      "grad_norm": 0.01746292971074581,
      "learning_rate": 8.537158541293878e-06,
      "loss": 0.0061,
      "step": 411
    },
    {
      "epoch": 4.130325814536341,
      "grad_norm": 0.0255062784999609,
      "learning_rate": 8.340919874712455e-06,
      "loss": 0.0055,
      "step": 412
    },
    {
      "epoch": 4.140350877192983,
      "grad_norm": 0.028920074924826622,
      "learning_rate": 8.146757496139468e-06,
      "loss": 0.0053,
      "step": 413
    },
    {
      "epoch": 4.150375939849624,
      "grad_norm": 0.022704336792230606,
      "learning_rate": 7.954681082625304e-06,
      "loss": 0.0061,
      "step": 414
    },
    {
      "epoch": 4.160401002506266,
      "grad_norm": 0.027427634224295616,
      "learning_rate": 7.764700207255903e-06,
      "loss": 0.007,
      "step": 415
    },
    {
      "epoch": 4.160401002506266,
      "eval_loss": 0.02001427672803402,
      "eval_runtime": 7.9079,
      "eval_samples_per_second": 6.323,
      "eval_steps_per_second": 1.644,
      "step": 415
    },
    {
      "epoch": 4.170426065162907,
      "grad_norm": 0.022619981318712234,
      "learning_rate": 7.576824338675598e-06,
      "loss": 0.0058,
      "step": 416
    },
    {
      "epoch": 4.180451127819548,
      "grad_norm": 0.021383745595812798,
      "learning_rate": 7.3910628406152485e-06,
      "loss": 0.0055,
      "step": 417
    },
    {
      "epoch": 4.190476190476191,
      "grad_norm": 0.023008733987808228,
      "learning_rate": 7.207424971425464e-06,
      "loss": 0.0069,
      "step": 418
    },
    {
      "epoch": 4.200501253132832,
      "grad_norm": 0.028300167992711067,
      "learning_rate": 7.025919883615239e-06,
      "loss": 0.0065,
      "step": 419
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 0.030637523159384727,
      "learning_rate": 6.8465566233957945e-06,
      "loss": 0.0057,
      "step": 420
    },
    {
      "epoch": 4.2105263157894735,
      "eval_loss": 0.020213749259710312,
      "eval_runtime": 7.8454,
      "eval_samples_per_second": 6.373,
      "eval_steps_per_second": 1.657,
      "step": 420
    },
    {
      "epoch": 4.220551378446116,
      "grad_norm": 0.025282373651862144,
      "learning_rate": 6.66934413022966e-06,
      "loss": 0.0064,
      "step": 421
    },
    {
      "epoch": 4.230576441102757,
      "grad_norm": 0.02849014848470688,
      "learning_rate": 6.494291236385186e-06,
      "loss": 0.0067,
      "step": 422
    },
    {
      "epoch": 4.2406015037593985,
      "grad_norm": 0.028334425762295723,
      "learning_rate": 6.321406666496304e-06,
      "loss": 0.0066,
      "step": 423
    },
    {
      "epoch": 4.25062656641604,
      "grad_norm": 0.02328861691057682,
      "learning_rate": 6.150699037127722e-06,
      "loss": 0.0072,
      "step": 424
    },
    {
      "epoch": 4.260651629072681,
      "grad_norm": 0.025907669216394424,
      "learning_rate": 5.982176856345445e-06,
      "loss": 0.0059,
      "step": 425
    },
    {
      "epoch": 4.260651629072681,
      "eval_loss": 0.020313529297709465,
      "eval_runtime": 7.8621,
      "eval_samples_per_second": 6.36,
      "eval_steps_per_second": 1.653,
      "step": 425
    },
    {
      "epoch": 4.2706766917293235,
      "grad_norm": 0.02658798359334469,
      "learning_rate": 5.815848523292755e-06,
      "loss": 0.0058,
      "step": 426
    },
    {
      "epoch": 4.280701754385965,
      "grad_norm": 0.027299784123897552,
      "learning_rate": 5.651722327771603e-06,
      "loss": 0.0055,
      "step": 427
    },
    {
      "epoch": 4.290726817042606,
      "grad_norm": 0.041355252265930176,
      "learning_rate": 5.489806449829438e-06,
      "loss": 0.006,
      "step": 428
    },
    {
      "epoch": 4.3007518796992485,
      "grad_norm": 0.025544259697198868,
      "learning_rate": 5.330108959351482e-06,
      "loss": 0.0059,
      "step": 429
    },
    {
      "epoch": 4.31077694235589,
      "grad_norm": 0.023853585124015808,
      "learning_rate": 5.1726378156585816e-06,
      "loss": 0.005,
      "step": 430
    },
    {
      "epoch": 4.31077694235589,
      "eval_loss": 0.020203066989779472,
      "eval_runtime": 7.9023,
      "eval_samples_per_second": 6.327,
      "eval_steps_per_second": 1.645,
      "step": 430
    },
    {
      "epoch": 4.320802005012531,
      "grad_norm": 0.02388196997344494,
      "learning_rate": 5.017400867110461e-06,
      "loss": 0.0057,
      "step": 431
    },
    {
      "epoch": 4.330827067669173,
      "grad_norm": 0.028964275494217873,
      "learning_rate": 4.864405850714621e-06,
      "loss": 0.0059,
      "step": 432
    },
    {
      "epoch": 4.340852130325814,
      "grad_norm": 0.022113680839538574,
      "learning_rate": 4.713660391740676e-06,
      "loss": 0.0059,
      "step": 433
    },
    {
      "epoch": 4.350877192982456,
      "grad_norm": 0.026654038578271866,
      "learning_rate": 4.565172003340329e-06,
      "loss": 0.0061,
      "step": 434
    },
    {
      "epoch": 4.360902255639098,
      "grad_norm": 0.032207190990448,
      "learning_rate": 4.418948086172914e-06,
      "loss": 0.0062,
      "step": 435
    },
    {
      "epoch": 4.360902255639098,
      "eval_loss": 0.020152073353528976,
      "eval_runtime": 7.8456,
      "eval_samples_per_second": 6.373,
      "eval_steps_per_second": 1.657,
      "step": 435
    },
    {
      "epoch": 4.370927318295739,
      "grad_norm": 0.02554272674024105,
      "learning_rate": 4.2749959280365344e-06,
      "loss": 0.0063,
      "step": 436
    },
    {
      "epoch": 4.380952380952381,
      "grad_norm": 0.028861908242106438,
      "learning_rate": 4.133322703504894e-06,
      "loss": 0.0059,
      "step": 437
    },
    {
      "epoch": 4.390977443609023,
      "grad_norm": 0.025862114503979683,
      "learning_rate": 3.9939354735696375e-06,
      "loss": 0.0056,
      "step": 438
    },
    {
      "epoch": 4.401002506265664,
      "grad_norm": 0.029569407925009727,
      "learning_rate": 3.856841185288473e-06,
      "loss": 0.0051,
      "step": 439
    },
    {
      "epoch": 4.4110275689223055,
      "grad_norm": 0.024914972484111786,
      "learning_rate": 3.72204667143895e-06,
      "loss": 0.0058,
      "step": 440
    },
    {
      "epoch": 4.4110275689223055,
      "eval_loss": 0.020192475989460945,
      "eval_runtime": 7.8522,
      "eval_samples_per_second": 6.368,
      "eval_steps_per_second": 1.656,
      "step": 440
    },
    {
      "epoch": 4.421052631578947,
      "grad_norm": 0.023811664432287216,
      "learning_rate": 3.58955865017786e-06,
      "loss": 0.0061,
      "step": 441
    },
    {
      "epoch": 4.431077694235589,
      "grad_norm": 0.026025721803307533,
      "learning_rate": 3.4593837247064662e-06,
      "loss": 0.0061,
      "step": 442
    },
    {
      "epoch": 4.4411027568922306,
      "grad_norm": 0.03492841124534607,
      "learning_rate": 3.331528382941346e-06,
      "loss": 0.0077,
      "step": 443
    },
    {
      "epoch": 4.451127819548872,
      "grad_norm": 0.03118693083524704,
      "learning_rate": 3.2059989971910508e-06,
      "loss": 0.0056,
      "step": 444
    },
    {
      "epoch": 4.461152882205514,
      "grad_norm": 0.023116862401366234,
      "learning_rate": 3.082801823838527e-06,
      "loss": 0.006,
      "step": 445
    },
    {
      "epoch": 4.461152882205514,
      "eval_loss": 0.02033940888941288,
      "eval_runtime": 7.8753,
      "eval_samples_per_second": 6.349,
      "eval_steps_per_second": 1.651,
      "step": 445
    },
    {
      "epoch": 4.471177944862156,
      "grad_norm": 0.03186218440532684,
      "learning_rate": 2.9619430030292826e-06,
      "loss": 0.0064,
      "step": 446
    },
    {
      "epoch": 4.481203007518797,
      "grad_norm": 0.03249065950512886,
      "learning_rate": 2.8434285583653797e-06,
      "loss": 0.0061,
      "step": 447
    },
    {
      "epoch": 4.491228070175438,
      "grad_norm": 0.024312468245625496,
      "learning_rate": 2.727264396605178e-06,
      "loss": 0.0058,
      "step": 448
    },
    {
      "epoch": 4.50125313283208,
      "grad_norm": 0.03428362309932709,
      "learning_rate": 2.6134563073689778e-06,
      "loss": 0.0064,
      "step": 449
    },
    {
      "epoch": 4.511278195488722,
      "grad_norm": 0.024099471047520638,
      "learning_rate": 2.50200996285046e-06,
      "loss": 0.0057,
      "step": 450
    },
    {
      "epoch": 4.511278195488722,
      "eval_loss": 0.020252428948879242,
      "eval_runtime": 7.8661,
      "eval_samples_per_second": 6.356,
      "eval_steps_per_second": 1.653,
      "step": 450
    },
    {
      "epoch": 4.521303258145363,
      "grad_norm": 0.038607899099588394,
      "learning_rate": 2.392930917533981e-06,
      "loss": 0.0068,
      "step": 451
    },
    {
      "epoch": 4.531328320802005,
      "grad_norm": 0.02548091858625412,
      "learning_rate": 2.2862246079177198e-06,
      "loss": 0.0053,
      "step": 452
    },
    {
      "epoch": 4.541353383458647,
      "grad_norm": 0.047327931970357895,
      "learning_rate": 2.1818963522427737e-06,
      "loss": 0.0063,
      "step": 453
    },
    {
      "epoch": 4.551378446115288,
      "grad_norm": 0.029703263193368912,
      "learning_rate": 2.0799513502280267e-06,
      "loss": 0.0074,
      "step": 454
    },
    {
      "epoch": 4.56140350877193,
      "grad_norm": 0.019005384296178818,
      "learning_rate": 1.9803946828110375e-06,
      "loss": 0.0055,
      "step": 455
    },
    {
      "epoch": 4.56140350877193,
      "eval_loss": 0.020170502364635468,
      "eval_runtime": 7.8694,
      "eval_samples_per_second": 6.354,
      "eval_steps_per_second": 1.652,
      "step": 455
    },
    {
      "epoch": 4.571428571428571,
      "grad_norm": 0.02032121643424034,
      "learning_rate": 1.883231311894823e-06,
      "loss": 0.0051,
      "step": 456
    },
    {
      "epoch": 4.581453634085213,
      "grad_norm": 0.020469170063734055,
      "learning_rate": 1.788466080100515e-06,
      "loss": 0.006,
      "step": 457
    },
    {
      "epoch": 4.591478696741855,
      "grad_norm": 0.030130814760923386,
      "learning_rate": 1.6961037105260202e-06,
      "loss": 0.0053,
      "step": 458
    },
    {
      "epoch": 4.601503759398496,
      "grad_norm": 0.027054738253355026,
      "learning_rate": 1.6061488065106323e-06,
      "loss": 0.0058,
      "step": 459
    },
    {
      "epoch": 4.611528822055138,
      "grad_norm": 0.023562690243124962,
      "learning_rate": 1.5186058514055912e-06,
      "loss": 0.005,
      "step": 460
    },
    {
      "epoch": 4.611528822055138,
      "eval_loss": 0.020171333104372025,
      "eval_runtime": 7.8408,
      "eval_samples_per_second": 6.377,
      "eval_steps_per_second": 1.658,
      "step": 460
    },
    {
      "epoch": 4.62155388471178,
      "grad_norm": 0.028073003515601158,
      "learning_rate": 1.4334792083506355e-06,
      "loss": 0.0054,
      "step": 461
    },
    {
      "epoch": 4.631578947368421,
      "grad_norm": 0.019973045215010643,
      "learning_rate": 1.3507731200565422e-06,
      "loss": 0.0055,
      "step": 462
    },
    {
      "epoch": 4.641604010025063,
      "grad_norm": 0.020437592640519142,
      "learning_rate": 1.2704917085936741e-06,
      "loss": 0.0046,
      "step": 463
    },
    {
      "epoch": 4.651629072681704,
      "grad_norm": 0.01714220456779003,
      "learning_rate": 1.1926389751865108e-06,
      "loss": 0.0051,
      "step": 464
    },
    {
      "epoch": 4.661654135338345,
      "grad_norm": 0.023718336597085,
      "learning_rate": 1.1172188000142802e-06,
      "loss": 0.0061,
      "step": 465
    },
    {
      "epoch": 4.661654135338345,
      "eval_loss": 0.020175131037831306,
      "eval_runtime": 7.867,
      "eval_samples_per_second": 6.356,
      "eval_steps_per_second": 1.652,
      "step": 465
    },
    {
      "epoch": 4.671679197994988,
      "grad_norm": 0.030823465436697006,
      "learning_rate": 1.0442349420175245e-06,
      "loss": 0.0052,
      "step": 466
    },
    {
      "epoch": 4.681704260651629,
      "grad_norm": 0.02299085445702076,
      "learning_rate": 9.736910387107777e-07,
      "loss": 0.0057,
      "step": 467
    },
    {
      "epoch": 4.69172932330827,
      "grad_norm": 0.03316880762577057,
      "learning_rate": 9.055906060012443e-07,
      "loss": 0.0066,
      "step": 468
    },
    {
      "epoch": 4.701754385964913,
      "grad_norm": 0.0231813732534647,
      "learning_rate": 8.399370380136051e-07,
      "loss": 0.0049,
      "step": 469
    },
    {
      "epoch": 4.711779448621554,
      "grad_norm": 0.02695239894092083,
      "learning_rate": 7.76733606920832e-07,
      "loss": 0.0064,
      "step": 470
    },
    {
      "epoch": 4.711779448621554,
      "eval_loss": 0.020140748471021652,
      "eval_runtime": 7.8502,
      "eval_samples_per_second": 6.369,
      "eval_steps_per_second": 1.656,
      "step": 470
    },
    {
      "epoch": 4.7218045112781954,
      "grad_norm": 0.02695208415389061,
      "learning_rate": 7.159834627811058e-07,
      "loss": 0.0057,
      "step": 471
    },
    {
      "epoch": 4.731829573934837,
      "grad_norm": 0.024768374860286713,
      "learning_rate": 6.576896333808091e-07,
      "loss": 0.0065,
      "step": 472
    },
    {
      "epoch": 4.741854636591478,
      "grad_norm": 0.02205987274646759,
      "learning_rate": 6.018550240836418e-07,
      "loss": 0.0052,
      "step": 473
    },
    {
      "epoch": 4.7518796992481205,
      "grad_norm": 0.026055388152599335,
      "learning_rate": 5.484824176857972e-07,
      "loss": 0.0053,
      "step": 474
    },
    {
      "epoch": 4.761904761904762,
      "grad_norm": 0.026830576360225677,
      "learning_rate": 4.975744742772848e-07,
      "loss": 0.0052,
      "step": 475
    },
    {
      "epoch": 4.761904761904762,
      "eval_loss": 0.020160943269729614,
      "eval_runtime": 7.8885,
      "eval_samples_per_second": 6.338,
      "eval_steps_per_second": 1.648,
      "step": 475
    },
    {
      "epoch": 4.771929824561403,
      "grad_norm": 0.03841860592365265,
      "learning_rate": 4.4913373110933023e-07,
      "loss": 0.006,
      "step": 476
    },
    {
      "epoch": 4.7819548872180455,
      "grad_norm": 0.02955353632569313,
      "learning_rate": 4.031626024679491e-07,
      "loss": 0.0049,
      "step": 477
    },
    {
      "epoch": 4.791979949874687,
      "grad_norm": 0.028415614739060402,
      "learning_rate": 3.5966337955357643e-07,
      "loss": 0.0063,
      "step": 478
    },
    {
      "epoch": 4.802005012531328,
      "grad_norm": 0.036227256059646606,
      "learning_rate": 3.1863823036692995e-07,
      "loss": 0.0061,
      "step": 479
    },
    {
      "epoch": 4.81203007518797,
      "grad_norm": 0.02573656104505062,
      "learning_rate": 2.800891996009025e-07,
      "loss": 0.0057,
      "step": 480
    },
    {
      "epoch": 4.81203007518797,
      "eval_loss": 0.020148631185293198,
      "eval_runtime": 7.8749,
      "eval_samples_per_second": 6.349,
      "eval_steps_per_second": 1.651,
      "step": 480
    },
    {
      "epoch": 4.822055137844611,
      "grad_norm": 0.02233925648033619,
      "learning_rate": 2.4401820853867663e-07,
      "loss": 0.0055,
      "step": 481
    },
    {
      "epoch": 4.832080200501253,
      "grad_norm": 0.025654539465904236,
      "learning_rate": 2.1042705495798454e-07,
      "loss": 0.0055,
      "step": 482
    },
    {
      "epoch": 4.842105263157895,
      "grad_norm": 0.021019209176301956,
      "learning_rate": 1.793174130414743e-07,
      "loss": 0.0067,
      "step": 483
    },
    {
      "epoch": 4.852130325814536,
      "grad_norm": 0.023309381678700447,
      "learning_rate": 1.506908332932988e-07,
      "loss": 0.005,
      "step": 484
    },
    {
      "epoch": 4.862155388471178,
      "grad_norm": 0.021895110607147217,
      "learning_rate": 1.245487424618108e-07,
      "loss": 0.0051,
      "step": 485
    },
    {
      "epoch": 4.862155388471178,
      "eval_loss": 0.020108748227357864,
      "eval_runtime": 7.8508,
      "eval_samples_per_second": 6.369,
      "eval_steps_per_second": 1.656,
      "step": 485
    },
    {
      "epoch": 4.87218045112782,
      "grad_norm": 0.02250145561993122,
      "learning_rate": 1.0089244346847548e-07,
      "loss": 0.0061,
      "step": 486
    },
    {
      "epoch": 4.882205513784461,
      "grad_norm": 0.02475419081747532,
      "learning_rate": 7.972311534293342e-08,
      "loss": 0.0061,
      "step": 487
    },
    {
      "epoch": 4.8922305764411025,
      "grad_norm": 0.026803474873304367,
      "learning_rate": 6.104181316421431e-08,
      "loss": 0.0059,
      "step": 488
    },
    {
      "epoch": 4.902255639097744,
      "grad_norm": 0.031020719558000565,
      "learning_rate": 4.484946800817347e-08,
      "loss": 0.0052,
      "step": 489
    },
    {
      "epoch": 4.912280701754386,
      "grad_norm": 0.02696959115564823,
      "learning_rate": 3.1146886901090025e-08,
      "loss": 0.0063,
      "step": 490
    },
    {
      "epoch": 4.912280701754386,
      "eval_loss": 0.020169507712125778,
      "eval_runtime": 7.8484,
      "eval_samples_per_second": 6.371,
      "eval_steps_per_second": 1.656,
      "step": 490
    },
    {
      "epoch": 4.9223057644110275,
      "grad_norm": 0.02662855200469494,
      "learning_rate": 1.993475277941581e-08,
      "loss": 0.0068,
      "step": 491
    },
    {
      "epoch": 4.932330827067669,
      "grad_norm": 0.030369313433766365,
      "learning_rate": 1.1213624455780335e-08,
      "loss": 0.005,
      "step": 492
    },
    {
      "epoch": 4.942355889724311,
      "grad_norm": 0.02946079894900322,
      "learning_rate": 4.983936591090865e-09,
      "loss": 0.0055,
      "step": 493
    },
    {
      "epoch": 4.9523809523809526,
      "grad_norm": 0.023623205721378326,
      "learning_rate": 1.2459996729274892e-09,
      "loss": 0.0058,
      "step": 494
    },
    {
      "epoch": 4.962406015037594,
      "grad_norm": 0.020752649754285812,
      "learning_rate": 0.0,
      "loss": 0.0051,
      "step": 495
    },
    {
      "epoch": 4.962406015037594,
      "eval_loss": 0.02011207491159439,
      "eval_runtime": 7.8991,
      "eval_samples_per_second": 6.33,
      "eval_steps_per_second": 1.646,
      "step": 495
    },
    {
      "epoch": 4.962406015037594,
      "step": 495,
      "total_flos": 9.638341179540111e+17,
      "train_loss": 0.013327934296632355,
      "train_runtime": 7711.8459,
      "train_samples_per_second": 2.069,
      "train_steps_per_second": 0.064
    }
  ],
  "logging_steps": 1,
  "max_steps": 495,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.638341179540111e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}