{
  "best_global_step": 416,
  "best_metric": 1.697021484375,
  "best_model_checkpoint": "./tinyllama-shakespeare-lora/checkpoint-416",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 520,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.237028416991234,
      "epoch": 0.04819277108433735,
      "grad_norm": 2.990326166152954,
      "learning_rate": 8e-05,
      "loss": 3.3300735473632814,
      "mean_token_accuracy": 0.4388114660978317,
      "num_tokens": 3544.0,
      "step": 5
    },
    {
      "entropy": 2.1227019637823106,
      "epoch": 0.0963855421686747,
      "grad_norm": 3.1520185470581055,
      "learning_rate": 0.00018,
      "loss": 3.040688896179199,
      "mean_token_accuracy": 0.48041139617562295,
      "num_tokens": 6921.0,
      "step": 10
    },
    {
      "entropy": 2.470001167058945,
      "epoch": 0.14457831325301204,
      "grad_norm": 2.7241196632385254,
      "learning_rate": 0.00019843137254901963,
      "loss": 2.6650358200073243,
      "mean_token_accuracy": 0.5012731336057186,
      "num_tokens": 9897.0,
      "step": 15
    },
    {
      "entropy": 2.4187598139047624,
      "epoch": 0.1927710843373494,
      "grad_norm": 1.1179628372192383,
      "learning_rate": 0.00019647058823529413,
      "loss": 2.336894989013672,
      "mean_token_accuracy": 0.544729282706976,
      "num_tokens": 13386.0,
      "step": 20
    },
    {
      "entropy": 2.4283315002918244,
      "epoch": 0.24096385542168675,
      "grad_norm": 1.0476402044296265,
      "learning_rate": 0.00019450980392156863,
      "loss": 2.257316017150879,
      "mean_token_accuracy": 0.5755889780819416,
      "num_tokens": 16770.0,
      "step": 25
    },
    {
      "entropy": 2.09900840818882,
      "epoch": 0.2891566265060241,
      "grad_norm": 1.005602240562439,
      "learning_rate": 0.00019254901960784316,
      "loss": 2.004672813415527,
      "mean_token_accuracy": 0.6035298384726048,
      "num_tokens": 19815.0,
      "step": 30
    },
    {
      "entropy": 2.067019133269787,
      "epoch": 0.3373493975903614,
      "grad_norm": 1.3398667573928833,
      "learning_rate": 0.00019058823529411766,
      "loss": 2.0276206970214843,
      "mean_token_accuracy": 0.5973991125822067,
      "num_tokens": 22956.0,
      "step": 35
    },
    {
      "entropy": 2.0127717584371565,
      "epoch": 0.3855421686746988,
      "grad_norm": 1.3001172542572021,
      "learning_rate": 0.00018862745098039216,
      "loss": 1.989430046081543,
      "mean_token_accuracy": 0.5942102774977684,
      "num_tokens": 26147.0,
      "step": 40
    },
    {
      "entropy": 1.948568905889988,
      "epoch": 0.43373493975903615,
      "grad_norm": 1.463757038116455,
      "learning_rate": 0.0001866666666666667,
      "loss": 2.002850341796875,
      "mean_token_accuracy": 0.6077656075358391,
      "num_tokens": 29290.0,
      "step": 45
    },
    {
      "entropy": 1.9544441923499107,
      "epoch": 0.4819277108433735,
      "grad_norm": 1.0674948692321777,
      "learning_rate": 0.0001847058823529412,
      "loss": 2.0039562225341796,
      "mean_token_accuracy": 0.5964730970561505,
      "num_tokens": 32774.0,
      "step": 50
    },
    {
      "entropy": 1.991774895787239,
      "epoch": 0.5301204819277109,
      "grad_norm": 1.1058229207992554,
      "learning_rate": 0.0001827450980392157,
      "loss": 1.9143487930297851,
      "mean_token_accuracy": 0.5965772472321987,
      "num_tokens": 36361.0,
      "step": 55
    },
    {
      "entropy": 1.9300899803638458,
      "epoch": 0.5783132530120482,
      "grad_norm": 1.3332328796386719,
      "learning_rate": 0.00018078431372549022,
      "loss": 1.989386749267578,
      "mean_token_accuracy": 0.595651201158762,
      "num_tokens": 39883.0,
      "step": 60
    },
    {
      "entropy": 1.8389213621616363,
      "epoch": 0.6265060240963856,
      "grad_norm": 1.600728988647461,
      "learning_rate": 0.00017882352941176472,
      "loss": 1.8484268188476562,
      "mean_token_accuracy": 0.6202847644686699,
      "num_tokens": 43257.0,
      "step": 65
    },
    {
      "entropy": 1.8422008126974105,
      "epoch": 0.6746987951807228,
      "grad_norm": 1.148055911064148,
      "learning_rate": 0.00017686274509803922,
      "loss": 1.846010971069336,
      "mean_token_accuracy": 0.6175518661737442,
      "num_tokens": 46204.0,
      "step": 70
    },
    {
      "entropy": 1.8617600902915001,
      "epoch": 0.7228915662650602,
      "grad_norm": 1.3051718473434448,
      "learning_rate": 0.00017490196078431375,
      "loss": 1.877816390991211,
      "mean_token_accuracy": 0.6046500727534294,
      "num_tokens": 49793.0,
      "step": 75
    },
    {
      "entropy": 1.788547897338867,
      "epoch": 0.7710843373493976,
      "grad_norm": 1.6796694993972778,
      "learning_rate": 0.00017294117647058825,
      "loss": 1.8042177200317382,
      "mean_token_accuracy": 0.626473106443882,
      "num_tokens": 53104.0,
      "step": 80
    },
    {
      "entropy": 1.7356875404715537,
      "epoch": 0.8192771084337349,
      "grad_norm": 1.6048922538757324,
      "learning_rate": 0.00017098039215686275,
      "loss": 1.7470354080200194,
      "mean_token_accuracy": 0.6395883277058602,
      "num_tokens": 56062.0,
      "step": 85
    },
    {
      "entropy": 1.7859284698963165,
      "epoch": 0.8674698795180723,
      "grad_norm": 1.2566527128219604,
      "learning_rate": 0.00016901960784313725,
      "loss": 1.8112169265747071,
      "mean_token_accuracy": 0.6365099921822548,
      "num_tokens": 59161.0,
      "step": 90
    },
    {
      "entropy": 1.8159554556012154,
      "epoch": 0.9156626506024096,
      "grad_norm": 1.3626623153686523,
      "learning_rate": 0.00016705882352941178,
      "loss": 1.831540298461914,
      "mean_token_accuracy": 0.6189694285392762,
      "num_tokens": 62426.0,
      "step": 95
    },
    {
      "entropy": 1.7219859138131142,
      "epoch": 0.963855421686747,
      "grad_norm": 1.4669969081878662,
      "learning_rate": 0.00016509803921568628,
      "loss": 1.708040428161621,
      "mean_token_accuracy": 0.6417586743831635,
      "num_tokens": 65573.0,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_entropy": 1.8488803276648889,
      "eval_loss": 1.8482003211975098,
      "eval_mean_token_accuracy": 0.6005064799235418,
      "eval_num_tokens": 68001.0,
      "eval_runtime": 31.0894,
      "eval_samples_per_second": 6.69,
      "eval_steps_per_second": 0.836,
      "step": 104
    },
    {
      "entropy": 1.703609773987218,
      "epoch": 1.0096385542168675,
      "grad_norm": 1.5622724294662476,
      "learning_rate": 0.00016313725490196078,
      "loss": 1.749994659423828,
      "mean_token_accuracy": 0.6331688251934553,
      "num_tokens": 68613.0,
      "step": 105
    },
    {
      "entropy": 1.841702875494957,
      "epoch": 1.0578313253012048,
      "grad_norm": 1.2482104301452637,
      "learning_rate": 0.0001611764705882353,
      "loss": 1.826130485534668,
      "mean_token_accuracy": 0.618197824060917,
      "num_tokens": 72379.0,
      "step": 110
    },
    {
      "entropy": 1.7080746620893479,
      "epoch": 1.106024096385542,
      "grad_norm": 1.5051069259643555,
      "learning_rate": 0.0001592156862745098,
      "loss": 1.735462760925293,
      "mean_token_accuracy": 0.6424251943826675,
      "num_tokens": 75330.0,
      "step": 115
    },
    {
      "entropy": 1.7118461817502975,
      "epoch": 1.1542168674698796,
      "grad_norm": 1.4837011098861694,
      "learning_rate": 0.0001572549019607843,
      "loss": 1.7383131027221679,
      "mean_token_accuracy": 0.637732881307602,
      "num_tokens": 78552.0,
      "step": 120
    },
    {
      "entropy": 1.7180065914988518,
      "epoch": 1.202409638554217,
      "grad_norm": 1.5369439125061035,
      "learning_rate": 0.00015529411764705884,
      "loss": 1.7133190155029296,
      "mean_token_accuracy": 0.6356611765921116,
      "num_tokens": 81798.0,
      "step": 125
    },
    {
      "entropy": 1.6460102945566177,
      "epoch": 1.2506024096385542,
      "grad_norm": 1.483790636062622,
      "learning_rate": 0.00015333333333333334,
      "loss": 1.598601722717285,
      "mean_token_accuracy": 0.6597890131175518,
      "num_tokens": 85117.0,
      "step": 130
    },
    {
      "entropy": 1.7030983410775662,
      "epoch": 1.2987951807228915,
      "grad_norm": 1.3817803859710693,
      "learning_rate": 0.00015137254901960784,
      "loss": 1.7572540283203124,
      "mean_token_accuracy": 0.6322377510368824,
      "num_tokens": 88742.0,
      "step": 135
    },
    {
      "entropy": 1.7065479427576065,
      "epoch": 1.346987951807229,
      "grad_norm": 1.5301700830459595,
      "learning_rate": 0.00014941176470588237,
      "loss": 1.7399499893188477,
      "mean_token_accuracy": 0.6498943358659744,
      "num_tokens": 92210.0,
      "step": 140
    },
    {
      "entropy": 1.625732569396496,
      "epoch": 1.3951807228915662,
      "grad_norm": 1.7736130952835083,
      "learning_rate": 0.00014745098039215687,
      "loss": 1.5962481498718262,
      "mean_token_accuracy": 0.6658786356449127,
      "num_tokens": 95303.0,
      "step": 145
    },
    {
      "entropy": 1.6358379781246186,
      "epoch": 1.4433734939759035,
      "grad_norm": 1.7948054075241089,
      "learning_rate": 0.00014549019607843137,
      "loss": 1.537781047821045,
      "mean_token_accuracy": 0.6697517029941082,
      "num_tokens": 98384.0,
      "step": 150
    },
    {
      "entropy": 1.5426702469587326,
      "epoch": 1.491566265060241,
      "grad_norm": 1.9344528913497925,
      "learning_rate": 0.0001435294117647059,
      "loss": 1.623773193359375,
      "mean_token_accuracy": 0.6632205478847026,
      "num_tokens": 101490.0,
      "step": 155
    },
    {
      "entropy": 1.52632287889719,
      "epoch": 1.5397590361445783,
      "grad_norm": 1.7781139612197876,
      "learning_rate": 0.0001415686274509804,
      "loss": 1.5219478607177734,
      "mean_token_accuracy": 0.6700925581157208,
      "num_tokens": 104191.0,
      "step": 160
    },
    {
      "entropy": 1.5876709207892419,
      "epoch": 1.5879518072289156,
      "grad_norm": 1.6688432693481445,
      "learning_rate": 0.0001396078431372549,
      "loss": 1.6492792129516602,
      "mean_token_accuracy": 0.6539962455630303,
      "num_tokens": 107261.0,
      "step": 165
    },
    {
      "entropy": 1.670412401854992,
      "epoch": 1.636144578313253,
      "grad_norm": 1.8056694269180298,
      "learning_rate": 0.00013764705882352943,
      "loss": 1.7233877182006836,
      "mean_token_accuracy": 0.6458830818533897,
      "num_tokens": 110608.0,
      "step": 170
    },
    {
      "entropy": 1.6271757155656814,
      "epoch": 1.6843373493975904,
      "grad_norm": 1.6283652782440186,
      "learning_rate": 0.00013568627450980393,
      "loss": 1.6276430130004882,
      "mean_token_accuracy": 0.6651988372206687,
      "num_tokens": 113881.0,
      "step": 175
    },
    {
      "entropy": 1.6719022899866105,
      "epoch": 1.7325301204819277,
      "grad_norm": 1.7494244575500488,
      "learning_rate": 0.00013372549019607843,
      "loss": 1.6668785095214844,
      "mean_token_accuracy": 0.6384630501270294,
      "num_tokens": 117044.0,
      "step": 180
    },
    {
      "entropy": 1.7114125609397888,
      "epoch": 1.7807228915662652,
      "grad_norm": 1.5748646259307861,
      "learning_rate": 0.00013176470588235296,
      "loss": 1.6835250854492188,
      "mean_token_accuracy": 0.6382554531097412,
      "num_tokens": 120606.0,
      "step": 185
    },
    {
      "entropy": 1.575966164469719,
      "epoch": 1.8289156626506025,
      "grad_norm": 1.5379868745803833,
      "learning_rate": 0.00012980392156862746,
      "loss": 1.6510162353515625,
      "mean_token_accuracy": 0.6542762629687786,
      "num_tokens": 123899.0,
      "step": 190
    },
    {
      "entropy": 1.6397091552615166,
      "epoch": 1.8771084337349397,
      "grad_norm": 1.8050676584243774,
      "learning_rate": 0.00012784313725490196,
      "loss": 1.6907238006591796,
      "mean_token_accuracy": 0.6459958925843239,
      "num_tokens": 127473.0,
      "step": 195
    },
    {
      "entropy": 1.6371865630149842,
      "epoch": 1.9253012048192772,
      "grad_norm": 1.7931848764419556,
      "learning_rate": 0.0001258823529411765,
      "loss": 1.680709457397461,
      "mean_token_accuracy": 0.6382987394928932,
      "num_tokens": 130771.0,
      "step": 200
    },
    {
      "entropy": 1.6472558170557021,
      "epoch": 1.9734939759036143,
      "grad_norm": 1.731370210647583,
      "learning_rate": 0.000123921568627451,
      "loss": 1.6797250747680663,
      "mean_token_accuracy": 0.6526046924293041,
      "num_tokens": 134125.0,
      "step": 205
    },
    {
      "epoch": 2.0,
      "eval_entropy": 1.8006169016544635,
      "eval_loss": 1.7452114820480347,
      "eval_mean_token_accuracy": 0.6168445830161755,
      "eval_num_tokens": 136002.0,
      "eval_runtime": 31.081,
      "eval_samples_per_second": 6.692,
      "eval_steps_per_second": 0.837,
      "step": 208
    },
    {
      "entropy": 1.6643306487484981,
      "epoch": 2.019277108433735,
      "grad_norm": 1.6965025663375854,
      "learning_rate": 0.0001219607843137255,
      "loss": 1.658614730834961,
      "mean_token_accuracy": 0.6498823793310868,
      "num_tokens": 137377.0,
      "step": 210
    },
    {
      "entropy": 1.6733600363135337,
      "epoch": 2.067469879518072,
      "grad_norm": 1.8712608814239502,
      "learning_rate": 0.00012,
      "loss": 1.5869686126708984,
      "mean_token_accuracy": 0.656411537528038,
      "num_tokens": 141109.0,
      "step": 215
    },
    {
      "entropy": 1.5299664348363877,
      "epoch": 2.1156626506024097,
      "grad_norm": 2.0042712688446045,
      "learning_rate": 0.00011803921568627452,
      "loss": 1.501875877380371,
      "mean_token_accuracy": 0.6718885555863381,
      "num_tokens": 144100.0,
      "step": 220
    },
    {
      "entropy": 1.5708236396312714,
      "epoch": 2.163855421686747,
      "grad_norm": 2.0190634727478027,
      "learning_rate": 0.00011607843137254903,
      "loss": 1.6497461318969726,
      "mean_token_accuracy": 0.6537894055247306,
      "num_tokens": 147584.0,
      "step": 225
    },
    {
      "entropy": 1.5232929274439813,
      "epoch": 2.212048192771084,
      "grad_norm": 1.9235641956329346,
      "learning_rate": 0.00011411764705882353,
      "loss": 1.6025299072265624,
      "mean_token_accuracy": 0.6778546750545502,
      "num_tokens": 150511.0,
      "step": 230
    },
    {
      "entropy": 1.6279756397008895,
      "epoch": 2.2602409638554217,
      "grad_norm": 1.8622772693634033,
      "learning_rate": 0.00011215686274509805,
      "loss": 1.546085739135742,
      "mean_token_accuracy": 0.6725640386343003,
      "num_tokens": 154053.0,
      "step": 235
    },
    {
      "entropy": 1.5271618023514748,
      "epoch": 2.3084337349397592,
      "grad_norm": 2.066518545150757,
      "learning_rate": 0.00011019607843137256,
      "loss": 1.5498031616210937,
      "mean_token_accuracy": 0.6706199154257775,
      "num_tokens": 157138.0,
      "step": 240
    },
    {
      "entropy": 1.5382411390542985,
      "epoch": 2.3566265060240963,
      "grad_norm": 1.9565099477767944,
      "learning_rate": 0.00010823529411764706,
      "loss": 1.4900429725646973,
      "mean_token_accuracy": 0.6834538638591766,
      "num_tokens": 160436.0,
      "step": 245
    },
    {
      "entropy": 1.5565499082207679,
      "epoch": 2.404819277108434,
      "grad_norm": 1.769296646118164,
      "learning_rate": 0.00010627450980392158,
      "loss": 1.5247212409973145,
      "mean_token_accuracy": 0.6715260729193687,
      "num_tokens": 163761.0,
      "step": 250
    },
    {
      "entropy": 1.560780017077923,
      "epoch": 2.453012048192771,
      "grad_norm": 1.9404278993606567,
      "learning_rate": 0.00010431372549019609,
      "loss": 1.4871397018432617,
      "mean_token_accuracy": 0.6675436168909072,
      "num_tokens": 166898.0,
      "step": 255
    },
    {
      "entropy": 1.4660954281687737,
      "epoch": 2.5012048192771084,
      "grad_norm": 2.1916701793670654,
      "learning_rate": 0.00010235294117647058,
      "loss": 1.5004875183105468,
      "mean_token_accuracy": 0.6763247601687908,
      "num_tokens": 169909.0,
      "step": 260
    },
    {
      "entropy": 1.577199426293373,
      "epoch": 2.549397590361446,
      "grad_norm": 1.8109275102615356,
      "learning_rate": 0.00010039215686274511,
      "loss": 1.5327059745788574,
      "mean_token_accuracy": 0.6665430299937725,
      "num_tokens": 173365.0,
      "step": 265
    },
    {
      "entropy": 1.5943815588951111,
      "epoch": 2.597590361445783,
      "grad_norm": 2.2097110748291016,
      "learning_rate": 9.843137254901961e-05,
      "loss": 1.629846954345703,
      "mean_token_accuracy": 0.663162975013256,
      "num_tokens": 176834.0,
      "step": 270
    },
    {
      "entropy": 1.55985931456089,
      "epoch": 2.6457831325301204,
      "grad_norm": 1.896395206451416,
      "learning_rate": 9.647058823529412e-05,
      "loss": 1.4793538093566894,
      "mean_token_accuracy": 0.6789445534348488,
      "num_tokens": 180153.0,
      "step": 275
    },
    {
      "entropy": 1.6425225555896759,
      "epoch": 2.693975903614458,
      "grad_norm": 1.6390873193740845,
      "learning_rate": 9.450980392156862e-05,
      "loss": 1.5848325729370116,
      "mean_token_accuracy": 0.6619467623531818,
      "num_tokens": 183512.0,
      "step": 280
    },
    {
      "entropy": 1.45964814722538,
      "epoch": 2.742168674698795,
      "grad_norm": 1.9988118410110474,
      "learning_rate": 9.254901960784315e-05,
      "loss": 1.46427001953125,
      "mean_token_accuracy": 0.6810511723160744,
      "num_tokens": 186472.0,
      "step": 285
    },
    {
      "entropy": 1.4157467983663081,
      "epoch": 2.7903614457831325,
      "grad_norm": 2.046511173248291,
      "learning_rate": 9.058823529411765e-05,
      "loss": 1.42630615234375,
      "mean_token_accuracy": 0.7006928496062755,
      "num_tokens": 189517.0,
      "step": 290
    },
    {
      "entropy": 1.476995500922203,
      "epoch": 2.83855421686747,
      "grad_norm": 2.3397531509399414,
      "learning_rate": 8.862745098039215e-05,
      "loss": 1.495310115814209,
      "mean_token_accuracy": 0.6897226825356484,
      "num_tokens": 192652.0,
      "step": 295
    },
    {
      "entropy": 1.4484579890966416,
      "epoch": 2.886746987951807,
      "grad_norm": 2.0256145000457764,
      "learning_rate": 8.666666666666667e-05,
      "loss": 1.5049756050109864,
      "mean_token_accuracy": 0.6935365840792656,
      "num_tokens": 195757.0,
      "step": 300
    },
    {
      "entropy": 1.4375765472650528,
      "epoch": 2.9349397590361446,
      "grad_norm": 1.795584797859192,
      "learning_rate": 8.470588235294118e-05,
      "loss": 1.4198492050170899,
      "mean_token_accuracy": 0.6847325280308724,
      "num_tokens": 198807.0,
      "step": 305
    },
    {
      "entropy": 1.528662195801735,
      "epoch": 2.983132530120482,
      "grad_norm": 2.0084290504455566,
      "learning_rate": 8.274509803921568e-05,
      "loss": 1.565225315093994,
      "mean_token_accuracy": 0.6660942971706391,
      "num_tokens": 202500.0,
      "step": 310
    },
    {
      "epoch": 3.0,
      "eval_entropy": 1.6572299553797796,
      "eval_loss": 1.710593342781067,
      "eval_mean_token_accuracy": 0.6203672221073737,
      "eval_num_tokens": 204003.0,
      "eval_runtime": 31.0505,
      "eval_samples_per_second": 6.699,
      "eval_steps_per_second": 0.837,
      "step": 312
    },
    {
      "entropy": 1.5311788932273263,
      "epoch": 3.0289156626506024,
      "grad_norm": 1.8857117891311646,
      "learning_rate": 8.07843137254902e-05,
      "loss": 1.5346615791320801,
      "mean_token_accuracy": 0.6760246863490657,
      "num_tokens": 205682.0,
      "step": 315
    },
    {
      "entropy": 1.5602496802806853,
      "epoch": 3.07710843373494,
      "grad_norm": 2.1312615871429443,
      "learning_rate": 7.882352941176471e-05,
      "loss": 1.4694817543029786,
      "mean_token_accuracy": 0.6782874375581741,
      "num_tokens": 209067.0,
      "step": 320
    },
    {
      "entropy": 1.5345655620098113,
      "epoch": 3.125301204819277,
      "grad_norm": 1.8577815294265747,
      "learning_rate": 7.686274509803923e-05,
      "loss": 1.4558188438415527,
      "mean_token_accuracy": 0.6776880241930485,
      "num_tokens": 212352.0,
      "step": 325
    },
    {
      "entropy": 1.4676252231001854,
      "epoch": 3.1734939759036145,
      "grad_norm": 2.027308940887451,
      "learning_rate": 7.490196078431373e-05,
      "loss": 1.4121692657470704,
      "mean_token_accuracy": 0.6886831521987915,
      "num_tokens": 215682.0,
      "step": 330
    },
    {
      "entropy": 1.4628461375832558,
      "epoch": 3.221686746987952,
      "grad_norm": 1.9528241157531738,
      "learning_rate": 7.294117647058823e-05,
      "loss": 1.439143466949463,
      "mean_token_accuracy": 0.6888918332755566,
      "num_tokens": 219099.0,
      "step": 335
    },
    {
      "entropy": 1.3313937097787858,
      "epoch": 3.269879518072289,
      "grad_norm": 1.9298614263534546,
      "learning_rate": 7.098039215686276e-05,
      "loss": 1.380887508392334,
      "mean_token_accuracy": 0.7101273387670517,
      "num_tokens": 222041.0,
      "step": 340
    },
    {
      "entropy": 1.5036465510725976,
      "epoch": 3.3180722891566266,
      "grad_norm": 2.5983974933624268,
      "learning_rate": 6.901960784313726e-05,
      "loss": 1.479258918762207,
      "mean_token_accuracy": 0.6855721130967141,
      "num_tokens": 225295.0,
      "step": 345
    },
    {
      "entropy": 1.4219771109521389,
      "epoch": 3.3662650602409636,
      "grad_norm": 2.4156622886657715,
      "learning_rate": 6.705882352941176e-05,
      "loss": 1.4329577445983888,
      "mean_token_accuracy": 0.6899721816182136,
      "num_tokens": 228520.0,
      "step": 350
    },
    {
      "entropy": 1.526054358482361,
      "epoch": 3.414457831325301,
      "grad_norm": 2.509209156036377,
      "learning_rate": 6.509803921568627e-05,
      "loss": 1.4574121475219726,
      "mean_token_accuracy": 0.6843675896525383,
      "num_tokens": 231944.0,
      "step": 355
    },
    {
      "entropy": 1.462461456656456,
      "epoch": 3.4626506024096386,
      "grad_norm": 1.98367178440094,
      "learning_rate": 6.313725490196079e-05,
      "loss": 1.4001091957092284,
      "mean_token_accuracy": 0.6919049903750419,
      "num_tokens": 235338.0,
      "step": 360
    },
    {
      "entropy": 1.4600355744361877,
      "epoch": 3.5108433734939757,
      "grad_norm": 2.3915727138519287,
      "learning_rate": 6.11764705882353e-05,
      "loss": 1.504780101776123,
      "mean_token_accuracy": 0.6843942984938621,
      "num_tokens": 238515.0,
      "step": 365
    },
    {
      "entropy": 1.3431835047900678,
      "epoch": 3.559036144578313,
      "grad_norm": 2.4486849308013916,
      "learning_rate": 5.921568627450981e-05,
      "loss": 1.3256868362426757,
      "mean_token_accuracy": 0.7202241063117981,
      "num_tokens": 241407.0,
      "step": 370
    },
    {
      "entropy": 1.5020532310009003,
      "epoch": 3.6072289156626507,
      "grad_norm": 2.259896993637085,
      "learning_rate": 5.725490196078431e-05,
      "loss": 1.4902771949768066,
      "mean_token_accuracy": 0.6747317329049111,
      "num_tokens": 244810.0,
      "step": 375
    },
    {
      "entropy": 1.449581353366375,
      "epoch": 3.6554216867469878,
      "grad_norm": 2.306088924407959,
      "learning_rate": 5.529411764705883e-05,
      "loss": 1.4183819770812989,
      "mean_token_accuracy": 0.6943565353751182,
      "num_tokens": 247917.0,
      "step": 380
    },
    {
      "entropy": 1.401406754553318,
      "epoch": 3.7036144578313253,
      "grad_norm": 2.6169755458831787,
      "learning_rate": 5.333333333333333e-05,
      "loss": 1.324946975708008,
      "mean_token_accuracy": 0.7123707890510559,
      "num_tokens": 251035.0,
      "step": 385
    },
    {
      "entropy": 1.3621296539902688,
      "epoch": 3.7518072289156628,
      "grad_norm": 2.3834919929504395,
      "learning_rate": 5.137254901960784e-05,
      "loss": 1.4367511749267579,
      "mean_token_accuracy": 0.6939964599907398,
      "num_tokens": 254564.0,
      "step": 390
    },
    {
      "entropy": 1.4182414807379247,
      "epoch": 3.8,
      "grad_norm": 2.1327056884765625,
      "learning_rate": 4.9411764705882355e-05,
      "loss": 1.396888542175293,
      "mean_token_accuracy": 0.6919433787465096,
      "num_tokens": 258001.0,
      "step": 395
    },
    {
      "entropy": 1.4946724489331245,
      "epoch": 3.8481927710843373,
      "grad_norm": 2.383056640625,
      "learning_rate": 4.745098039215686e-05,
      "loss": 1.3979401588439941,
      "mean_token_accuracy": 0.695962718129158,
      "num_tokens": 260974.0,
      "step": 400
    },
    {
      "entropy": 1.449859108030796,
      "epoch": 3.896385542168675,
      "grad_norm": 2.490652084350586,
      "learning_rate": 4.549019607843137e-05,
      "loss": 1.4903193473815919,
      "mean_token_accuracy": 0.686176010966301,
      "num_tokens": 264415.0,
      "step": 405
    },
    {
      "entropy": 1.4792103588581085,
      "epoch": 3.944578313253012,
      "grad_norm": 2.542536973953247,
      "learning_rate": 4.3529411764705885e-05,
      "loss": 1.4710602760314941,
      "mean_token_accuracy": 0.6943801581859589,
      "num_tokens": 267945.0,
      "step": 410
    },
    {
      "entropy": 1.4717445507645608,
      "epoch": 3.9927710843373494,
      "grad_norm": 2.2224795818328857,
      "learning_rate": 4.156862745098039e-05,
      "loss": 1.4788040161132812,
      "mean_token_accuracy": 0.6895710974931717,
      "num_tokens": 271493.0,
      "step": 415
    },
    {
      "epoch": 4.0,
      "eval_entropy": 1.620164128450247,
      "eval_loss": 1.697021484375,
      "eval_mean_token_accuracy": 0.6267852737353399,
      "eval_num_tokens": 272004.0,
      "eval_runtime": 31.0743,
      "eval_samples_per_second": 6.694,
      "eval_steps_per_second": 0.837,
      "step": 416
    },
    {
      "entropy": 1.4130519565783048,
      "epoch": 4.03855421686747,
      "grad_norm": 2.297147274017334,
      "learning_rate": 3.96078431372549e-05,
      "loss": 1.429203701019287,
      "mean_token_accuracy": 0.6895033309334203,
      "num_tokens": 274843.0,
      "step": 420
    },
    {
      "entropy": 1.3913781866431236,
      "epoch": 4.086746987951807,
      "grad_norm": 2.412026882171631,
      "learning_rate": 3.7647058823529415e-05,
      "loss": 1.2258234977722169,
      "mean_token_accuracy": 0.7282624199986458,
      "num_tokens": 277622.0,
      "step": 425
    },
    {
      "entropy": 1.4595662340521813,
      "epoch": 4.134939759036144,
      "grad_norm": 2.2476892471313477,
      "learning_rate": 3.568627450980392e-05,
      "loss": 1.3860394477844238,
      "mean_token_accuracy": 0.6948064118623734,
      "num_tokens": 281115.0,
      "step": 430
    },
    {
      "entropy": 1.4310455292463302,
      "epoch": 4.183132530120482,
      "grad_norm": 1.9430752992630005,
      "learning_rate": 3.372549019607844e-05,
      "loss": 1.4230783462524415,
      "mean_token_accuracy": 0.7034219011664391,
      "num_tokens": 284784.0,
      "step": 435
    },
    {
      "entropy": 1.3278866931796074,
      "epoch": 4.231325301204819,
      "grad_norm": 1.9049440622329712,
      "learning_rate": 3.176470588235294e-05,
      "loss": 1.424240779876709,
      "mean_token_accuracy": 0.7034214176237583,
      "num_tokens": 288256.0,
      "step": 440
    },
    {
      "entropy": 1.3592324391007424,
      "epoch": 4.279518072289156,
      "grad_norm": 2.3773999214172363,
      "learning_rate": 2.9803921568627453e-05,
      "loss": 1.3538729667663574,
      "mean_token_accuracy": 0.7138303920626641,
      "num_tokens": 291712.0,
      "step": 445
    },
    {
      "entropy": 1.4115683421492577,
      "epoch": 4.327710843373494,
      "grad_norm": 2.435506820678711,
      "learning_rate": 2.7843137254901964e-05,
      "loss": 1.3001550674438476,
      "mean_token_accuracy": 0.7063792496919632,
      "num_tokens": 294817.0,
      "step": 450
    },
    {
      "entropy": 1.4070537582039833,
      "epoch": 4.375903614457831,
      "grad_norm": 2.1623501777648926,
      "learning_rate": 2.5882352941176475e-05,
      "loss": 1.396212100982666,
      "mean_token_accuracy": 0.6988388016819954,
      "num_tokens": 298135.0,
      "step": 455
    },
    {
      "entropy": 1.3870512664318084,
      "epoch": 4.424096385542168,
      "grad_norm": 2.5376806259155273,
      "learning_rate": 2.3921568627450983e-05,
      "loss": 1.3213976860046386,
      "mean_token_accuracy": 0.712082713842392,
      "num_tokens": 301380.0,
      "step": 460
    },
    {
      "entropy": 1.412884371727705,
      "epoch": 4.472289156626506,
      "grad_norm": 2.841749668121338,
      "learning_rate": 2.196078431372549e-05,
      "loss": 1.3348329544067383,
      "mean_token_accuracy": 0.6968033984303474,
      "num_tokens": 304449.0,
      "step": 465
    },
    {
      "entropy": 1.4457854568958282,
      "epoch": 4.5204819277108435,
      "grad_norm": 2.5401291847229004,
      "learning_rate": 2e-05,
      "loss": 1.3989011764526367,
      "mean_token_accuracy": 0.6939552336931228,
      "num_tokens": 307631.0,
      "step": 470
    },
    {
      "entropy": 1.3890875995159149,
      "epoch": 4.5686746987951805,
      "grad_norm": 2.5059876441955566,
      "learning_rate": 1.803921568627451e-05,
      "loss": 1.3057866096496582,
      "mean_token_accuracy": 0.7126045942306518,
      "num_tokens": 310718.0,
      "step": 475
    },
    {
      "entropy": 1.4398678675293923,
      "epoch": 4.6168674698795185,
      "grad_norm": 2.388061285018921,
      "learning_rate": 1.607843137254902e-05,
      "loss": 1.3806493759155274,
      "mean_token_accuracy": 0.7140536062419415,
      "num_tokens": 313936.0,
      "step": 480
    },
    {
      "entropy": 1.4656711727380753,
      "epoch": 4.6650602409638555,
      "grad_norm": 2.4591257572174072,
      "learning_rate": 1.411764705882353e-05,
      "loss": 1.328711700439453,
      "mean_token_accuracy": 0.705501814186573,
      "num_tokens": 317225.0,
      "step": 485
    },
    {
      "entropy": 1.4785845473408699,
      "epoch": 4.713253012048193,
      "grad_norm": 2.5279247760772705,
      "learning_rate": 1.215686274509804e-05,
      "loss": 1.5198940277099608,
      "mean_token_accuracy": 0.6828307613730431,
      "num_tokens": 320998.0,
      "step": 490
    },
    {
      "entropy": 1.3545206658542157,
      "epoch": 4.76144578313253,
      "grad_norm": 2.389832019805908,
      "learning_rate": 1.0196078431372549e-05,
      "loss": 1.3498905181884766,
      "mean_token_accuracy": 0.7124156683683396,
      "num_tokens": 324049.0,
      "step": 495
    },
    {
      "entropy": 1.4337334975600242,
      "epoch": 4.809638554216868,
      "grad_norm": 2.5963311195373535,
      "learning_rate": 8.23529411764706e-06,
      "loss": 1.4652009963989259,
      "mean_token_accuracy": 0.6918993666768074,
      "num_tokens": 327578.0,
      "step": 500
    },
    {
      "entropy": 1.3732572302222252,
      "epoch": 4.857831325301205,
      "grad_norm": 2.731126070022583,
      "learning_rate": 6.274509803921569e-06,
      "loss": 1.255812168121338,
      "mean_token_accuracy": 0.7217151954770088,
      "num_tokens": 330740.0,
      "step": 505
    },
    {
      "entropy": 1.3217465229332448,
      "epoch": 4.906024096385542,
      "grad_norm": 2.517321825027466,
      "learning_rate": 4.313725490196079e-06,
      "loss": 1.262193489074707,
      "mean_token_accuracy": 0.7306308448314667,
      "num_tokens": 334020.0,
      "step": 510
    },
    {
      "entropy": 1.4316355839371682,
      "epoch": 4.95421686746988,
      "grad_norm": 2.5220019817352295,
      "learning_rate": 2.3529411764705885e-06,
      "loss": 1.3470137596130372,
      "mean_token_accuracy": 0.7120753511786461,
      "num_tokens": 336979.0,
      "step": 515
    },
    {
      "entropy": 1.4011002747636092,
      "epoch": 5.0,
      "grad_norm": 2.894545555114746,
      "learning_rate": 3.921568627450981e-07,
      "loss": 1.3720563888549804,
      "mean_token_accuracy": 0.7007956771474135,
      "num_tokens": 340005.0,
      "step": 520
    },
    {
      "epoch": 5.0,
      "eval_entropy": 1.568474531173706,
      "eval_loss": 1.7007105350494385,
      "eval_mean_token_accuracy": 0.6262180598882529,
      "eval_num_tokens": 340005.0,
      "eval_runtime": 31.0476,
      "eval_samples_per_second": 6.699,
      "eval_steps_per_second": 0.837,
      "step": 520
    }
  ],
  "logging_steps": 5,
  "max_steps": 520,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2115032078315520.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}