{
  "best_global_step": 1509,
  "best_metric": 1.013381004333496,
  "best_model_checkpoint": "/kaggle/working/models/checkpoints/v4/nosql2doc/checkpoint-1509",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1509,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.8841454654932024,
      "epoch": 0.01990049751243781,
      "grad_norm": 3.4684948921203613,
      "learning_rate": 7.142857142857143e-06,
      "loss": 3.3999290466308594,
      "mean_token_accuracy": 0.36815835647284983,
      "num_tokens": 84042.0,
      "step": 10
    },
    {
      "entropy": 2.871115285158157,
      "epoch": 0.03980099502487562,
      "grad_norm": 3.199768304824829,
      "learning_rate": 1.5079365079365079e-05,
      "loss": 3.1389333724975588,
      "mean_token_accuracy": 0.41857333555817605,
      "num_tokens": 166453.0,
      "step": 20
    },
    {
      "entropy": 2.689709846675396,
      "epoch": 0.05970149253731343,
      "grad_norm": 2.241548776626587,
      "learning_rate": 2.3015873015873015e-05,
      "loss": 2.5403181076049806,
      "mean_token_accuracy": 0.45860434919595716,
      "num_tokens": 248357.0,
      "step": 30
    },
    {
      "entropy": 2.371776354312897,
      "epoch": 0.07960199004975124,
      "grad_norm": 2.2443745136260986,
      "learning_rate": 3.095238095238095e-05,
      "loss": 2.2156551361083983,
      "mean_token_accuracy": 0.534111600741744,
      "num_tokens": 329302.0,
      "step": 40
    },
    {
      "entropy": 1.9646751508116722,
      "epoch": 0.09950248756218906,
      "grad_norm": 2.8479607105255127,
      "learning_rate": 3.888888888888889e-05,
      "loss": 1.9395181655883789,
      "mean_token_accuracy": 0.5719782687723637,
      "num_tokens": 410388.0,
      "step": 50
    },
    {
      "entropy": 1.712027932703495,
      "epoch": 0.11940298507462686,
      "grad_norm": 2.5392374992370605,
      "learning_rate": 4.682539682539683e-05,
      "loss": 1.7669334411621094,
      "mean_token_accuracy": 0.5962413489818573,
      "num_tokens": 491142.0,
      "step": 60
    },
    {
      "entropy": 1.6189243733882903,
      "epoch": 0.13930348258706468,
      "grad_norm": 2.8646080493927,
      "learning_rate": 5.4761904761904766e-05,
      "loss": 1.6198507308959962,
      "mean_token_accuracy": 0.6142528053373099,
      "num_tokens": 573054.0,
      "step": 70
    },
    {
      "entropy": 1.6242131367325783,
      "epoch": 0.15920398009950248,
      "grad_norm": 3.0876874923706055,
      "learning_rate": 6.26984126984127e-05,
      "loss": 1.6246612548828125,
      "mean_token_accuracy": 0.626776534691453,
      "num_tokens": 658316.0,
      "step": 80
    },
    {
      "entropy": 1.5123998656868936,
      "epoch": 0.1791044776119403,
      "grad_norm": 2.827972650527954,
      "learning_rate": 7.063492063492065e-05,
      "loss": 1.5039705276489257,
      "mean_token_accuracy": 0.6451249852776527,
      "num_tokens": 738196.0,
      "step": 90
    },
    {
      "entropy": 1.4389337845146657,
      "epoch": 0.19900497512437812,
      "grad_norm": 2.794956684112549,
      "learning_rate": 7.857142857142858e-05,
      "loss": 1.4707328796386718,
      "mean_token_accuracy": 0.6486270293593407,
      "num_tokens": 818366.0,
      "step": 100
    },
    {
      "entropy": 1.4290485948324203,
      "epoch": 0.21890547263681592,
      "grad_norm": 2.650611639022827,
      "learning_rate": 8.650793650793651e-05,
      "loss": 1.461697483062744,
      "mean_token_accuracy": 0.6452589213848114,
      "num_tokens": 900542.0,
      "step": 110
    },
    {
      "entropy": 1.4537153281271458,
      "epoch": 0.23880597014925373,
      "grad_norm": 3.212292194366455,
      "learning_rate": 9.444444444444444e-05,
      "loss": 1.3799741744995118,
      "mean_token_accuracy": 0.665483495220542,
      "num_tokens": 983783.0,
      "step": 120
    },
    {
      "entropy": 1.3213311471045017,
      "epoch": 0.25870646766169153,
      "grad_norm": 2.367784261703491,
      "learning_rate": 0.00010238095238095237,
      "loss": 1.339876365661621,
      "mean_token_accuracy": 0.6738482430577278,
      "num_tokens": 1065282.0,
      "step": 130
    },
    {
      "entropy": 1.2877162352204323,
      "epoch": 0.27860696517412936,
      "grad_norm": 2.7096056938171387,
      "learning_rate": 0.00011031746031746033,
      "loss": 1.3189430236816406,
      "mean_token_accuracy": 0.6620625633746385,
      "num_tokens": 1147430.0,
      "step": 140
    },
    {
      "entropy": 1.3201022632420063,
      "epoch": 0.29850746268656714,
      "grad_norm": 2.497452974319458,
      "learning_rate": 0.00011825396825396826,
      "loss": 1.3049938201904296,
      "mean_token_accuracy": 0.6833859849721193,
      "num_tokens": 1225840.0,
      "step": 150
    },
    {
      "entropy": 1.2007936105132102,
      "epoch": 0.31840796019900497,
      "grad_norm": 3.0387496948242188,
      "learning_rate": 0.0001261904761904762,
      "loss": 1.231965923309326,
      "mean_token_accuracy": 0.7017441529780626,
      "num_tokens": 1309268.0,
      "step": 160
    },
    {
      "entropy": 1.3119081839919091,
      "epoch": 0.3383084577114428,
      "grad_norm": 2.327012300491333,
      "learning_rate": 0.00013412698412698412,
      "loss": 1.2840826034545898,
      "mean_token_accuracy": 0.6792996764183045,
      "num_tokens": 1389094.0,
      "step": 170
    },
    {
      "entropy": 1.2995639897882938,
      "epoch": 0.3582089552238806,
      "grad_norm": 2.426831007003784,
      "learning_rate": 0.00014206349206349208,
      "loss": 1.2992033958435059,
      "mean_token_accuracy": 0.6910909987986088,
      "num_tokens": 1471243.0,
      "step": 180
    },
    {
      "entropy": 1.1952705256640912,
      "epoch": 0.3781094527363184,
      "grad_norm": 2.783355951309204,
      "learning_rate": 0.00015000000000000001,
      "loss": 1.1649029731750489,
      "mean_token_accuracy": 0.7020247872918844,
      "num_tokens": 1552783.0,
      "step": 190
    },
    {
      "entropy": 1.2681411176919937,
      "epoch": 0.39800995024875624,
      "grad_norm": 2.162163019180298,
      "learning_rate": 0.00015793650793650795,
      "loss": 1.2627052307128905,
      "mean_token_accuracy": 0.6855453528463841,
      "num_tokens": 1635174.0,
      "step": 200
    },
    {
      "entropy": 1.2593041777610778,
      "epoch": 0.417910447761194,
      "grad_norm": 2.3045036792755127,
      "learning_rate": 0.0001658730158730159,
      "loss": 1.237012004852295,
      "mean_token_accuracy": 0.6885611288249492,
      "num_tokens": 1720930.0,
      "step": 210
    },
    {
      "entropy": 1.1904637023806572,
      "epoch": 0.43781094527363185,
      "grad_norm": 2.291395664215088,
      "learning_rate": 0.00017380952380952383,
      "loss": 1.2200750350952148,
      "mean_token_accuracy": 0.6859784577041864,
      "num_tokens": 1803132.0,
      "step": 220
    },
    {
      "entropy": 1.1843627236783505,
      "epoch": 0.4577114427860697,
      "grad_norm": 2.553251266479492,
      "learning_rate": 0.00018174603174603177,
      "loss": 1.2164555549621583,
      "mean_token_accuracy": 0.6943051021546125,
      "num_tokens": 1884825.0,
      "step": 230
    },
    {
      "entropy": 1.1876815035939217,
      "epoch": 0.47761194029850745,
      "grad_norm": 2.403580904006958,
      "learning_rate": 0.0001896825396825397,
      "loss": 1.1573832511901856,
      "mean_token_accuracy": 0.7180207531899214,
      "num_tokens": 1968914.0,
      "step": 240
    },
    {
      "entropy": 1.1310456305742265,
      "epoch": 0.4975124378109453,
      "grad_norm": 2.3557302951812744,
      "learning_rate": 0.00019761904761904763,
      "loss": 1.1767531394958497,
      "mean_token_accuracy": 0.6986724361777306,
      "num_tokens": 2050602.0,
      "step": 250
    },
    {
      "entropy": 1.2115244895219803,
      "epoch": 0.5174129353233831,
      "grad_norm": 2.430337905883789,
      "learning_rate": 0.0001999989408126818,
      "loss": 1.187550926208496,
      "mean_token_accuracy": 0.6943733543157578,
      "num_tokens": 2134774.0,
      "step": 260
    },
    {
      "entropy": 1.1603886000812054,
      "epoch": 0.5373134328358209,
      "grad_norm": 1.848976969718933,
      "learning_rate": 0.0001999937530104439,
      "loss": 1.1683053970336914,
      "mean_token_accuracy": 0.6958795577287674,
      "num_tokens": 2214022.0,
      "step": 270
    },
    {
      "entropy": 1.083036269247532,
      "epoch": 0.5572139303482587,
      "grad_norm": 2.297624111175537,
      "learning_rate": 0.00019998424227267588,
      "loss": 1.087782859802246,
      "mean_token_accuracy": 0.7164598941802979,
      "num_tokens": 2295817.0,
      "step": 280
    },
    {
      "entropy": 1.1467085666954517,
      "epoch": 0.5771144278606966,
      "grad_norm": 2.1926214694976807,
      "learning_rate": 0.00019997040901054646,
      "loss": 1.1336830139160157,
      "mean_token_accuracy": 0.7070459358394146,
      "num_tokens": 2379081.0,
      "step": 290
    },
    {
      "entropy": 1.1092145919799805,
      "epoch": 0.5970149253731343,
      "grad_norm": 2.108574390411377,
      "learning_rate": 0.000199952253822096,
      "loss": 1.104503059387207,
      "mean_token_accuracy": 0.7167623668909073,
      "num_tokens": 2461290.0,
      "step": 300
    },
    {
      "entropy": 1.0762298285961152,
      "epoch": 0.6169154228855721,
      "grad_norm": 1.9235388040542603,
      "learning_rate": 0.00019992977749221064,
      "loss": 1.0758417129516602,
      "mean_token_accuracy": 0.7244217373430729,
      "num_tokens": 2546927.0,
      "step": 310
    },
    {
      "entropy": 1.058315774053335,
      "epoch": 0.6368159203980099,
      "grad_norm": 2.015493154525757,
      "learning_rate": 0.0001999029809925883,
      "loss": 1.0714119911193847,
      "mean_token_accuracy": 0.7185824103653431,
      "num_tokens": 2627151.0,
      "step": 320
    },
    {
      "entropy": 1.040397210419178,
      "epoch": 0.6567164179104478,
      "grad_norm": 1.824608564376831,
      "learning_rate": 0.00019987186548169682,
      "loss": 1.0344210624694825,
      "mean_token_accuracy": 0.7333592250943184,
      "num_tokens": 2711367.0,
      "step": 330
    },
    {
      "entropy": 0.9891116008162498,
      "epoch": 0.6766169154228856,
      "grad_norm": 1.9432896375656128,
      "learning_rate": 0.0001998364323047236,
      "loss": 0.9733485221862793,
      "mean_token_accuracy": 0.7416356436908245,
      "num_tokens": 2793468.0,
      "step": 340
    },
    {
      "entropy": 1.0004291333258153,
      "epoch": 0.6965174129353234,
      "grad_norm": 1.906457781791687,
      "learning_rate": 0.00019979668299351783,
      "loss": 1.0494978904724122,
      "mean_token_accuracy": 0.7176768533885479,
      "num_tokens": 2875166.0,
      "step": 350
    },
    {
      "entropy": 1.0571318455040455,
      "epoch": 0.7164179104477612,
      "grad_norm": 1.7709777355194092,
      "learning_rate": 0.00019975261926652392,
      "loss": 1.0371734619140625,
      "mean_token_accuracy": 0.7288566574454307,
      "num_tokens": 2959671.0,
      "step": 360
    },
    {
      "entropy": 0.9674378834664822,
      "epoch": 0.736318407960199,
      "grad_norm": 1.896713376045227,
      "learning_rate": 0.00019970424302870739,
      "loss": 0.9970813751220703,
      "mean_token_accuracy": 0.7395003981888294,
      "num_tokens": 3041437.0,
      "step": 370
    },
    {
      "entropy": 1.1270258143544196,
      "epoch": 0.7562189054726368,
      "grad_norm": 1.7242449522018433,
      "learning_rate": 0.00019965155637147243,
      "loss": 1.0752775192260742,
      "mean_token_accuracy": 0.7305668473243714,
      "num_tokens": 3123865.0,
      "step": 380
    },
    {
      "entropy": 0.9315242812037468,
      "epoch": 0.7761194029850746,
      "grad_norm": 1.781772255897522,
      "learning_rate": 0.0001995945615725716,
      "loss": 0.9751872062683106,
      "mean_token_accuracy": 0.7497406356036663,
      "num_tokens": 3205032.0,
      "step": 390
    },
    {
      "entropy": 1.076903734356165,
      "epoch": 0.7960199004975125,
      "grad_norm": 1.8422024250030518,
      "learning_rate": 0.00019953326109600728,
      "loss": 1.0796968460083007,
      "mean_token_accuracy": 0.7176821112632752,
      "num_tokens": 3286003.0,
      "step": 400
    },
    {
      "entropy": 0.9634652934968472,
      "epoch": 0.8159203980099502,
      "grad_norm": 1.7082111835479736,
      "learning_rate": 0.00019946765759192497,
      "loss": 0.9697890281677246,
      "mean_token_accuracy": 0.7406083434820175,
      "num_tokens": 3367484.0,
      "step": 410
    },
    {
      "entropy": 0.9321402795612812,
      "epoch": 0.835820895522388,
      "grad_norm": 1.8066761493682861,
      "learning_rate": 0.00019939775389649916,
      "loss": 0.9277518272399903,
      "mean_token_accuracy": 0.7574292354285717,
      "num_tokens": 3453904.0,
      "step": 420
    },
    {
      "entropy": 0.9754443395882845,
      "epoch": 0.8557213930348259,
      "grad_norm": 1.6425416469573975,
      "learning_rate": 0.00019932355303181026,
      "loss": 0.9705208778381348,
      "mean_token_accuracy": 0.7435099229216575,
      "num_tokens": 3536472.0,
      "step": 430
    },
    {
      "entropy": 0.9742012087255716,
      "epoch": 0.8756218905472637,
      "grad_norm": 1.707189679145813,
      "learning_rate": 0.00019924505820571425,
      "loss": 1.036739444732666,
      "mean_token_accuracy": 0.7312861289829016,
      "num_tokens": 3616711.0,
      "step": 440
    },
    {
      "entropy": 1.0537842728197575,
      "epoch": 0.8955223880597015,
      "grad_norm": 1.865692377090454,
      "learning_rate": 0.0001991622728117038,
      "loss": 1.0217846870422362,
      "mean_token_accuracy": 0.7338800489902496,
      "num_tokens": 3701126.0,
      "step": 450
    },
    {
      "entropy": 0.989534855633974,
      "epoch": 0.9154228855721394,
      "grad_norm": 1.7353651523590088,
      "learning_rate": 0.00019907520042876172,
      "loss": 1.0003718376159667,
      "mean_token_accuracy": 0.7480997510254384,
      "num_tokens": 3783495.0,
      "step": 460
    },
    {
      "entropy": 0.9294702004641294,
      "epoch": 0.9353233830845771,
      "grad_norm": 2.007262706756592,
      "learning_rate": 0.00019898384482120614,
      "loss": 1.0100757598876953,
      "mean_token_accuracy": 0.7364842720329762,
      "num_tokens": 3863784.0,
      "step": 470
    },
    {
      "entropy": 0.9523782070726157,
      "epoch": 0.9552238805970149,
      "grad_norm": 1.9396175146102905,
      "learning_rate": 0.0001988882099385278,
      "loss": 0.9432971000671386,
      "mean_token_accuracy": 0.756897260248661,
      "num_tokens": 3947540.0,
      "step": 480
    },
    {
      "entropy": 0.9421051684767008,
      "epoch": 0.9751243781094527,
      "grad_norm": 1.8582719564437866,
      "learning_rate": 0.00019878829991521935,
      "loss": 0.9125359535217286,
      "mean_token_accuracy": 0.7575220972299576,
      "num_tokens": 4031284.0,
      "step": 490
    },
    {
      "entropy": 0.9381607126444578,
      "epoch": 0.9950248756218906,
      "grad_norm": 1.9623899459838867,
      "learning_rate": 0.00019868411907059645,
      "loss": 0.9656861305236817,
      "mean_token_accuracy": 0.7544871643185616,
      "num_tokens": 4112149.0,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_entropy": 0.9874700385516214,
      "eval_loss": 1.063795566558838,
      "eval_mean_token_accuracy": 0.7334998920847551,
      "eval_num_tokens": 4132656.0,
      "eval_runtime": 88.6229,
      "eval_samples_per_second": 11.679,
      "eval_steps_per_second": 5.845,
      "step": 503
    },
    {
      "entropy": 0.8532990099568116,
      "epoch": 1.0139303482587065,
      "grad_norm": 1.6549201011657715,
      "learning_rate": 0.00019857567190861126,
      "loss": 0.7886298179626465,
      "mean_token_accuracy": 0.7860889779893976,
      "num_tokens": 4190723.0,
      "step": 510
    },
    {
      "entropy": 0.7671804554760456,
      "epoch": 1.0338308457711443,
      "grad_norm": 1.6148929595947266,
      "learning_rate": 0.00019846296311765754,
      "loss": 0.777919864654541,
      "mean_token_accuracy": 0.789631099998951,
      "num_tokens": 4273358.0,
      "step": 520
    },
    {
      "entropy": 0.8041674952954054,
      "epoch": 1.053731343283582,
      "grad_norm": 1.765950322151184,
      "learning_rate": 0.00019834599757036803,
      "loss": 0.7967178821563721,
      "mean_token_accuracy": 0.784614659100771,
      "num_tokens": 4356030.0,
      "step": 530
    },
    {
      "entropy": 0.7405303593724966,
      "epoch": 1.07363184079602,
      "grad_norm": 1.7868931293487549,
      "learning_rate": 0.00019822478032340387,
      "loss": 0.7301093578338623,
      "mean_token_accuracy": 0.8020888641476631,
      "num_tokens": 4440157.0,
      "step": 540
    },
    {
      "entropy": 0.7550990927964449,
      "epoch": 1.0935323383084576,
      "grad_norm": 1.991782784461975,
      "learning_rate": 0.0001980993166172358,
      "loss": 0.7884031295776367,
      "mean_token_accuracy": 0.7879343688488006,
      "num_tokens": 4523343.0,
      "step": 550
    },
    {
      "entropy": 0.812371177971363,
      "epoch": 1.1134328358208956,
      "grad_norm": 1.869396686553955,
      "learning_rate": 0.00019796961187591781,
      "loss": 0.8159590721130371,
      "mean_token_accuracy": 0.7711428456008435,
      "num_tokens": 4605454.0,
      "step": 560
    },
    {
      "entropy": 0.7879433136433363,
      "epoch": 1.1333333333333333,
      "grad_norm": 1.9530060291290283,
      "learning_rate": 0.00019783567170685262,
      "loss": 0.8079433441162109,
      "mean_token_accuracy": 0.7875082984566688,
      "num_tokens": 4687318.0,
      "step": 570
    },
    {
      "entropy": 0.8396137218922377,
      "epoch": 1.153233830845771,
      "grad_norm": 1.7895166873931885,
      "learning_rate": 0.00019769750190054905,
      "loss": 0.8188363075256347,
      "mean_token_accuracy": 0.7710263684391976,
      "num_tokens": 4768232.0,
      "step": 580
    },
    {
      "entropy": 0.8406378872692585,
      "epoch": 1.173134328358209,
      "grad_norm": 1.8147255182266235,
      "learning_rate": 0.00019755510843037191,
      "loss": 0.8532954216003418,
      "mean_token_accuracy": 0.7680545568466186,
      "num_tokens": 4850366.0,
      "step": 590
    },
    {
      "entropy": 0.7758986987173557,
      "epoch": 1.1930348258706467,
      "grad_norm": 1.9839149713516235,
      "learning_rate": 0.00019740849745228367,
      "loss": 0.7972042560577393,
      "mean_token_accuracy": 0.77668926268816,
      "num_tokens": 4932940.0,
      "step": 600
    },
    {
      "entropy": 0.825455692410469,
      "epoch": 1.2129353233830846,
      "grad_norm": 1.776328444480896,
      "learning_rate": 0.00019725767530457837,
      "loss": 0.7664913654327392,
      "mean_token_accuracy": 0.7940710753202438,
      "num_tokens": 5012821.0,
      "step": 610
    },
    {
      "entropy": 0.7286036755889654,
      "epoch": 1.2328358208955223,
      "grad_norm": 2.0241360664367676,
      "learning_rate": 0.00019710264850760756,
      "loss": 0.7731894969940185,
      "mean_token_accuracy": 0.7846131704747676,
      "num_tokens": 5095340.0,
      "step": 620
    },
    {
      "entropy": 0.7920811083167791,
      "epoch": 1.25273631840796,
      "grad_norm": 1.7310129404067993,
      "learning_rate": 0.00019694342376349835,
      "loss": 0.7303972244262695,
      "mean_token_accuracy": 0.7983584970235824,
      "num_tokens": 5177547.0,
      "step": 630
    },
    {
      "entropy": 0.7178942065685987,
      "epoch": 1.272636815920398,
      "grad_norm": 1.7608613967895508,
      "learning_rate": 0.00019678000795586386,
      "loss": 0.8018975257873535,
      "mean_token_accuracy": 0.7767767131328582,
      "num_tokens": 5265612.0,
      "step": 640
    },
    {
      "entropy": 0.8001765824854374,
      "epoch": 1.292537313432836,
      "grad_norm": 2.0576155185699463,
      "learning_rate": 0.00019661240814950536,
      "loss": 0.7223423480987549,
      "mean_token_accuracy": 0.7918425254523754,
      "num_tokens": 5348652.0,
      "step": 650
    },
    {
      "entropy": 0.7580439992249012,
      "epoch": 1.3124378109452737,
      "grad_norm": 1.539169430732727,
      "learning_rate": 0.00019644063159010716,
      "loss": 0.8217278480529785,
      "mean_token_accuracy": 0.7793662197887897,
      "num_tokens": 5431834.0,
      "step": 660
    },
    {
      "entropy": 0.8075141672044992,
      "epoch": 1.3323383084577114,
      "grad_norm": 1.8933053016662598,
      "learning_rate": 0.00019626468570392298,
      "loss": 0.7759135723114013,
      "mean_token_accuracy": 0.7861333280801773,
      "num_tokens": 5514228.0,
      "step": 670
    },
    {
      "entropy": 0.7623774103820324,
      "epoch": 1.3522388059701491,
      "grad_norm": 1.7051018476486206,
      "learning_rate": 0.00019608457809745537,
      "loss": 0.8088616371154785,
      "mean_token_accuracy": 0.7746041730046272,
      "num_tokens": 5594702.0,
      "step": 680
    },
    {
      "entropy": 0.8089807592332363,
      "epoch": 1.372139303482587,
      "grad_norm": 1.8061248064041138,
      "learning_rate": 0.00019590031655712631,
      "loss": 0.7946909904479981,
      "mean_token_accuracy": 0.7781320951879025,
      "num_tokens": 5676309.0,
      "step": 690
    },
    {
      "entropy": 0.7627177253365517,
      "epoch": 1.392039800995025,
      "grad_norm": 1.9005513191223145,
      "learning_rate": 0.00019571190904894115,
      "loss": 0.7615675926208496,
      "mean_token_accuracy": 0.788880155980587,
      "num_tokens": 5757314.0,
      "step": 700
    },
    {
      "entropy": 0.7580569207668304,
      "epoch": 1.4119402985074627,
      "grad_norm": 1.9883071184158325,
      "learning_rate": 0.00019551936371814375,
      "loss": 0.7611291408538818,
      "mean_token_accuracy": 0.7861187107861042,
      "num_tokens": 5838877.0,
      "step": 710
    },
    {
      "entropy": 0.7182941000908614,
      "epoch": 1.4318407960199004,
      "grad_norm": 2.187103033065796,
      "learning_rate": 0.0001953226888888647,
      "loss": 0.7204749584197998,
      "mean_token_accuracy": 0.7946790263056756,
      "num_tokens": 5920820.0,
      "step": 720
    },
    {
      "entropy": 0.6957223568111658,
      "epoch": 1.4517412935323384,
      "grad_norm": 1.7006845474243164,
      "learning_rate": 0.00019512189306376118,
      "loss": 0.757196044921875,
      "mean_token_accuracy": 0.7921811021864414,
      "num_tokens": 6004382.0,
      "step": 730
    },
    {
      "entropy": 0.7850441809743642,
      "epoch": 1.471641791044776,
      "grad_norm": 1.6335111856460571,
      "learning_rate": 0.0001949169849236496,
      "loss": 0.7292089462280273,
      "mean_token_accuracy": 0.797366950660944,
      "num_tokens": 6086449.0,
      "step": 740
    },
    {
      "entropy": 0.7639980979263783,
      "epoch": 1.491542288557214,
      "grad_norm": 1.7224810123443604,
      "learning_rate": 0.00019470797332713012,
      "loss": 0.8179457664489747,
      "mean_token_accuracy": 0.7910384394228458,
      "num_tokens": 6168749.0,
      "step": 750
    },
    {
      "entropy": 0.8024922411888837,
      "epoch": 1.5114427860696518,
      "grad_norm": 1.9208451509475708,
      "learning_rate": 0.0001944948673102038,
      "loss": 0.7916707515716552,
      "mean_token_accuracy": 0.7822470977902413,
      "num_tokens": 6248672.0,
      "step": 760
    },
    {
      "entropy": 0.7430106606334448,
      "epoch": 1.5313432835820895,
      "grad_norm": 1.9618816375732422,
      "learning_rate": 0.00019427767608588183,
      "loss": 0.7538708209991455,
      "mean_token_accuracy": 0.7816703900694847,
      "num_tokens": 6331261.0,
      "step": 770
    },
    {
      "entropy": 0.7457086000591516,
      "epoch": 1.5512437810945272,
      "grad_norm": 1.6858505010604858,
      "learning_rate": 0.0001940564090437875,
      "loss": 0.7316685676574707,
      "mean_token_accuracy": 0.8001063913106918,
      "num_tokens": 6415504.0,
      "step": 780
    },
    {
      "entropy": 0.7096565395593644,
      "epoch": 1.5711442786069652,
      "grad_norm": 1.9956564903259277,
      "learning_rate": 0.00019383107574974984,
      "loss": 0.7277198791503906,
      "mean_token_accuracy": 0.7895878478884697,
      "num_tokens": 6497199.0,
      "step": 790
    },
    {
      "entropy": 0.7606754396110773,
      "epoch": 1.591044776119403,
      "grad_norm": 1.7207188606262207,
      "learning_rate": 0.00019360168594539055,
      "loss": 0.7483164310455322,
      "mean_token_accuracy": 0.7934170700609684,
      "num_tokens": 6577494.0,
      "step": 800
    },
    {
      "entropy": 0.7786010228097439,
      "epoch": 1.6109452736318408,
      "grad_norm": 1.7885297536849976,
      "learning_rate": 0.0001933682495477024,
      "loss": 0.7517632007598877,
      "mean_token_accuracy": 0.791710589081049,
      "num_tokens": 6660010.0,
      "step": 810
    },
    {
      "entropy": 0.73126558996737,
      "epoch": 1.6308457711442785,
      "grad_norm": 1.929076910018921,
      "learning_rate": 0.00019313077664862092,
      "loss": 0.7747976303100585,
      "mean_token_accuracy": 0.7895058654248714,
      "num_tokens": 6739669.0,
      "step": 820
    },
    {
      "entropy": 0.7332248043268919,
      "epoch": 1.6507462686567163,
      "grad_norm": 1.6692851781845093,
      "learning_rate": 0.00019288927751458766,
      "loss": 0.6981019020080567,
      "mean_token_accuracy": 0.8055072821676731,
      "num_tokens": 6819838.0,
      "step": 830
    },
    {
      "entropy": 0.712925647944212,
      "epoch": 1.6706467661691542,
      "grad_norm": 1.5395867824554443,
      "learning_rate": 0.0001926437625861068,
      "loss": 0.7524178504943848,
      "mean_token_accuracy": 0.7908283203840256,
      "num_tokens": 6901777.0,
      "step": 840
    },
    {
      "entropy": 0.7585150092840195,
      "epoch": 1.6905472636815921,
      "grad_norm": 1.6641769409179688,
      "learning_rate": 0.00019239424247729345,
      "loss": 0.7306118488311768,
      "mean_token_accuracy": 0.7910356394946575,
      "num_tokens": 6983222.0,
      "step": 850
    },
    {
      "entropy": 0.7012663371860981,
      "epoch": 1.7104477611940299,
      "grad_norm": 1.9537489414215088,
      "learning_rate": 0.0001921407279754149,
      "loss": 0.752435302734375,
      "mean_token_accuracy": 0.8054570883512497,
      "num_tokens": 7067291.0,
      "step": 860
    },
    {
      "entropy": 0.7264402851462364,
      "epoch": 1.7303482587064676,
      "grad_norm": 1.8359386920928955,
      "learning_rate": 0.00019188323004042435,
      "loss": 0.6932102680206299,
      "mean_token_accuracy": 0.802193396538496,
      "num_tokens": 7150268.0,
      "step": 870
    },
    {
      "entropy": 0.6838662784546614,
      "epoch": 1.7502487562189055,
      "grad_norm": 1.9146333932876587,
      "learning_rate": 0.00019162175980448688,
      "loss": 0.7538199424743652,
      "mean_token_accuracy": 0.796929395198822,
      "num_tokens": 7231091.0,
      "step": 880
    },
    {
      "entropy": 0.7262898899614811,
      "epoch": 1.7701492537313432,
      "grad_norm": 2.0510952472686768,
      "learning_rate": 0.0001913563285714984,
      "loss": 0.7104074954986572,
      "mean_token_accuracy": 0.8008449010550975,
      "num_tokens": 7313340.0,
      "step": 890
    },
    {
      "entropy": 0.7481670882552862,
      "epoch": 1.7900497512437812,
      "grad_norm": 2.0180599689483643,
      "learning_rate": 0.0001910869478165969,
      "loss": 0.8105827331542969,
      "mean_token_accuracy": 0.7843330048024655,
      "num_tokens": 7393519.0,
      "step": 900
    },
    {
      "entropy": 0.7768452275544405,
      "epoch": 1.809950248756219,
      "grad_norm": 1.646704912185669,
      "learning_rate": 0.00019081362918566618,
      "loss": 0.7446431636810302,
      "mean_token_accuracy": 0.8008080549538136,
      "num_tokens": 7475016.0,
      "step": 910
    },
    {
      "entropy": 0.683229249343276,
      "epoch": 1.8298507462686566,
      "grad_norm": 1.638253092765808,
      "learning_rate": 0.00019053638449483259,
      "loss": 0.7266605854034424,
      "mean_token_accuracy": 0.7936579927802085,
      "num_tokens": 7559243.0,
      "step": 920
    },
    {
      "entropy": 0.7504678606987,
      "epoch": 1.8497512437810946,
      "grad_norm": 1.6552717685699463,
      "learning_rate": 0.0001902552257299542,
      "loss": 0.738245677947998,
      "mean_token_accuracy": 0.7894984453916549,
      "num_tokens": 7643029.0,
      "step": 930
    },
    {
      "entropy": 0.7069665104150772,
      "epoch": 1.8696517412935323,
      "grad_norm": 1.4619513750076294,
      "learning_rate": 0.00018997016504610246,
      "loss": 0.7275553226470948,
      "mean_token_accuracy": 0.8092396840453148,
      "num_tokens": 7726479.0,
      "step": 940
    },
    {
      "entropy": 0.6909922767430544,
      "epoch": 1.8895522388059702,
      "grad_norm": 1.6785398721694946,
      "learning_rate": 0.00018968121476703678,
      "loss": 0.6958948612213135,
      "mean_token_accuracy": 0.8184511929750442,
      "num_tokens": 7808235.0,
      "step": 950
    },
    {
      "entropy": 0.7206986505538225,
      "epoch": 1.909452736318408,
      "grad_norm": 1.8156358003616333,
      "learning_rate": 0.00018938838738467184,
      "loss": 0.7307656288146973,
      "mean_token_accuracy": 0.795404677093029,
      "num_tokens": 7888802.0,
      "step": 960
    },
    {
      "entropy": 0.7154657423496247,
      "epoch": 1.9293532338308457,
      "grad_norm": 1.9570693969726562,
      "learning_rate": 0.00018909169555853743,
      "loss": 0.6951888084411622,
      "mean_token_accuracy": 0.7979383312165738,
      "num_tokens": 7970204.0,
      "step": 970
    },
    {
      "entropy": 0.6446726582944393,
      "epoch": 1.9492537313432836,
      "grad_norm": 1.5071886777877808,
      "learning_rate": 0.00018879115211523117,
      "loss": 0.6544716358184814,
      "mean_token_accuracy": 0.8045676186680794,
      "num_tokens": 8053307.0,
      "step": 980
    },
    {
      "entropy": 0.699823185056448,
      "epoch": 1.9691542288557216,
      "grad_norm": 1.4266027212142944,
      "learning_rate": 0.0001884867700478641,
      "loss": 0.7055366516113282,
      "mean_token_accuracy": 0.7992964766919612,
      "num_tokens": 8135193.0,
      "step": 990
    },
    {
      "entropy": 0.7140736062079668,
      "epoch": 1.9890547263681593,
      "grad_norm": 2.0378243923187256,
      "learning_rate": 0.00018817856251549867,
      "loss": 0.7184643268585205,
      "mean_token_accuracy": 0.7989312529563903,
      "num_tokens": 8219594.0,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_entropy": 0.8002531397008988,
      "eval_loss": 1.0154445171356201,
      "eval_mean_token_accuracy": 0.7423216365709268,
      "eval_num_tokens": 8265312.0,
      "eval_runtime": 88.2,
      "eval_samples_per_second": 11.735,
      "eval_steps_per_second": 5.873,
      "step": 1006
    },
    {
      "entropy": 0.7038449613671554,
      "epoch": 2.007960199004975,
      "grad_norm": 1.6040326356887817,
      "learning_rate": 0.00018786654284258034,
      "loss": 0.6638185501098632,
      "mean_token_accuracy": 0.8091712115626586,
      "num_tokens": 8297623.0,
      "step": 1010
    },
    {
      "entropy": 0.5252644840627909,
      "epoch": 2.027860696517413,
      "grad_norm": 2.1827661991119385,
      "learning_rate": 0.00018755072451836097,
      "loss": 0.5153174877166748,
      "mean_token_accuracy": 0.8446948520839215,
      "num_tokens": 8379050.0,
      "step": 1020
    },
    {
      "entropy": 0.5121854435652494,
      "epoch": 2.047761194029851,
      "grad_norm": 1.8835452795028687,
      "learning_rate": 0.00018723112119631608,
      "loss": 0.5017234325408936,
      "mean_token_accuracy": 0.8539764225482941,
      "num_tokens": 8462847.0,
      "step": 1030
    },
    {
      "entropy": 0.5651770515367389,
      "epoch": 2.0676616915422885,
      "grad_norm": 1.6592763662338257,
      "learning_rate": 0.00018690774669355442,
      "loss": 0.5333661079406739,
      "mean_token_accuracy": 0.8419696733355522,
      "num_tokens": 8542681.0,
      "step": 1040
    },
    {
      "entropy": 0.47107693143188956,
      "epoch": 2.0875621890547262,
      "grad_norm": 1.8748646974563599,
      "learning_rate": 0.00018658061499022055,
      "loss": 0.4740941047668457,
      "mean_token_accuracy": 0.8567564368247986,
      "num_tokens": 8627432.0,
      "step": 1050
    },
    {
      "entropy": 0.49270418751984835,
      "epoch": 2.107462686567164,
      "grad_norm": 2.027717351913452,
      "learning_rate": 0.0001862497402288906,
      "loss": 0.48979806900024414,
      "mean_token_accuracy": 0.8501696541905404,
      "num_tokens": 8708822.0,
      "step": 1060
    },
    {
      "entropy": 0.4634111661463976,
      "epoch": 2.127363184079602,
      "grad_norm": 1.8452632427215576,
      "learning_rate": 0.0001859151367139608,
      "loss": 0.4521069049835205,
      "mean_token_accuracy": 0.8683336563408375,
      "num_tokens": 8789791.0,
      "step": 1070
    },
    {
      "entropy": 0.46942942440509794,
      "epoch": 2.14726368159204,
      "grad_norm": 1.9295680522918701,
      "learning_rate": 0.000185576818911029,
      "loss": 0.5178594589233398,
      "mean_token_accuracy": 0.8462874710559845,
      "num_tokens": 8870514.0,
      "step": 1080
    },
    {
      "entropy": 0.5393101962283253,
      "epoch": 2.1671641791044776,
      "grad_norm": 1.6308494806289673,
      "learning_rate": 0.00018523480144626948,
      "loss": 0.5196755409240723,
      "mean_token_accuracy": 0.8491615362465381,
      "num_tokens": 8952466.0,
      "step": 1090
    },
    {
      "entropy": 0.491584631241858,
      "epoch": 2.1870646766169153,
      "grad_norm": 1.9126944541931152,
      "learning_rate": 0.00018488909910580037,
      "loss": 0.507460069656372,
      "mean_token_accuracy": 0.8491294652223587,
      "num_tokens": 9034965.0,
      "step": 1100
    },
    {
      "entropy": 0.5460982380434871,
      "epoch": 2.206965174129353,
      "grad_norm": 2.245730400085449,
      "learning_rate": 0.00018453972683504466,
      "loss": 0.5434001445770263,
      "mean_token_accuracy": 0.8417311415076256,
      "num_tokens": 9117636.0,
      "step": 1110
    },
    {
      "entropy": 0.5413472417742013,
      "epoch": 2.226865671641791,
      "grad_norm": 1.7490206956863403,
      "learning_rate": 0.00018418669973808386,
      "loss": 0.5176630973815918,
      "mean_token_accuracy": 0.8437662713229657,
      "num_tokens": 9203719.0,
      "step": 1120
    },
    {
      "entropy": 0.5068975739181042,
      "epoch": 2.246766169154229,
      "grad_norm": 2.32169246673584,
      "learning_rate": 0.00018383003307700525,
      "loss": 0.4966591358184814,
      "mean_token_accuracy": 0.8558630302548409,
      "num_tokens": 9285285.0,
      "step": 1130
    },
    {
      "entropy": 0.472692010179162,
      "epoch": 2.2666666666666666,
      "grad_norm": 2.121119499206543,
      "learning_rate": 0.0001834697422712419,
      "loss": 0.4562082290649414,
      "mean_token_accuracy": 0.861832109093666,
      "num_tokens": 9368724.0,
      "step": 1140
    },
    {
      "entropy": 0.48463920764625074,
      "epoch": 2.2865671641791043,
      "grad_norm": 2.003596305847168,
      "learning_rate": 0.00018310584289690608,
      "loss": 0.5065945148468017,
      "mean_token_accuracy": 0.8411129087209701,
      "num_tokens": 9450345.0,
      "step": 1150
    },
    {
      "entropy": 0.5297361209988594,
      "epoch": 2.306467661691542,
      "grad_norm": 1.9170504808425903,
      "learning_rate": 0.0001827383506861159,
      "loss": 0.5093509674072265,
      "mean_token_accuracy": 0.8439007833600044,
      "num_tokens": 9531914.0,
      "step": 1160
    },
    {
      "entropy": 0.508120141364634,
      "epoch": 2.32636815920398,
      "grad_norm": 1.8182140588760376,
      "learning_rate": 0.00018236728152631526,
      "loss": 0.5065027236938476,
      "mean_token_accuracy": 0.8519511334598064,
      "num_tokens": 9613824.0,
      "step": 1170
    },
    {
      "entropy": 0.5490097716450691,
      "epoch": 2.346268656716418,
      "grad_norm": 1.6451319456100464,
      "learning_rate": 0.00018199265145958678,
      "loss": 0.5736330032348633,
      "mean_token_accuracy": 0.8286732420325279,
      "num_tokens": 9696070.0,
      "step": 1180
    },
    {
      "entropy": 0.5155842589214444,
      "epoch": 2.3661691542288557,
      "grad_norm": 1.8924293518066406,
      "learning_rate": 0.00018161447668195858,
      "loss": 0.5310076236724853,
      "mean_token_accuracy": 0.8431366585195065,
      "num_tokens": 9777568.0,
      "step": 1190
    },
    {
      "entropy": 0.4998965173959732,
      "epoch": 2.3860696517412934,
      "grad_norm": 1.9782637357711792,
      "learning_rate": 0.00018123277354270372,
      "loss": 0.46693716049194334,
      "mean_token_accuracy": 0.8578181132674217,
      "num_tokens": 9858081.0,
      "step": 1200
    },
    {
      "entropy": 0.5187313890084624,
      "epoch": 2.405970149253731,
      "grad_norm": 1.8333914279937744,
      "learning_rate": 0.00018084755854363377,
      "loss": 0.5384686946868896,
      "mean_token_accuracy": 0.8451795615255833,
      "num_tokens": 9940187.0,
      "step": 1210
    },
    {
      "entropy": 0.5085030103102326,
      "epoch": 2.4258706467661693,
      "grad_norm": 2.00839900970459,
      "learning_rate": 0.00018045884833838512,
      "loss": 0.5079335689544677,
      "mean_token_accuracy": 0.8514300569891929,
      "num_tokens": 10018432.0,
      "step": 1220
    },
    {
      "entropy": 0.5382265558466315,
      "epoch": 2.445771144278607,
      "grad_norm": 1.9003617763519287,
      "learning_rate": 0.00018006665973169911,
      "loss": 0.5386298179626465,
      "mean_token_accuracy": 0.8454402416944504,
      "num_tokens": 10100137.0,
      "step": 1230
    },
    {
      "entropy": 0.5438722034916281,
      "epoch": 2.4656716417910447,
      "grad_norm": 2.1470773220062256,
      "learning_rate": 0.0001796710096786956,
      "loss": 0.5553887367248536,
      "mean_token_accuracy": 0.8391343042254448,
      "num_tokens": 10184025.0,
      "step": 1240
    },
    {
      "entropy": 0.5251597924157977,
      "epoch": 2.4855721393034824,
      "grad_norm": 1.992197036743164,
      "learning_rate": 0.0001792719152841398,
      "loss": 0.5232258319854737,
      "mean_token_accuracy": 0.8480332940816879,
      "num_tokens": 10268664.0,
      "step": 1250
    },
    {
      "entropy": 0.5361188564449548,
      "epoch": 2.50547263681592,
      "grad_norm": 1.6610969305038452,
      "learning_rate": 0.0001788693938017029,
      "loss": 0.5130891799926758,
      "mean_token_accuracy": 0.8441365607082844,
      "num_tokens": 10349492.0,
      "step": 1260
    },
    {
      "entropy": 0.4794240856543183,
      "epoch": 2.5253731343283583,
      "grad_norm": 2.0173373222351074,
      "learning_rate": 0.00017846346263321623,
      "loss": 0.5028162956237793,
      "mean_token_accuracy": 0.84928208142519,
      "num_tokens": 10432597.0,
      "step": 1270
    },
    {
      "entropy": 0.5074908941984176,
      "epoch": 2.545273631840796,
      "grad_norm": 2.124582290649414,
      "learning_rate": 0.00017805413932791875,
      "loss": 0.5137399673461914,
      "mean_token_accuracy": 0.8484948351979256,
      "num_tokens": 10513192.0,
      "step": 1280
    },
    {
      "entropy": 0.5202818088233471,
      "epoch": 2.5651741293532337,
      "grad_norm": 1.7741371393203735,
      "learning_rate": 0.00017764144158169856,
      "loss": 0.5218567848205566,
      "mean_token_accuracy": 0.8525593280792236,
      "num_tokens": 10597049.0,
      "step": 1290
    },
    {
      "entropy": 0.5169114828109741,
      "epoch": 2.585074626865672,
      "grad_norm": 2.03674054145813,
      "learning_rate": 0.00017722538723632773,
      "loss": 0.5342438697814942,
      "mean_token_accuracy": 0.8379098229110241,
      "num_tokens": 10679839.0,
      "step": 1300
    },
    {
      "entropy": 0.527920619212091,
      "epoch": 2.604975124378109,
      "grad_norm": 1.7481690645217896,
      "learning_rate": 0.000176805994278691,
      "loss": 0.5268249988555909,
      "mean_token_accuracy": 0.8468565516173839,
      "num_tokens": 10763374.0,
      "step": 1310
    },
    {
      "entropy": 0.5612680882215499,
      "epoch": 2.6248756218905474,
      "grad_norm": 2.044337511062622,
      "learning_rate": 0.0001763832808400082,
      "loss": 0.5637305736541748,
      "mean_token_accuracy": 0.8389740340411663,
      "num_tokens": 10846467.0,
      "step": 1320
    },
    {
      "entropy": 0.526572679169476,
      "epoch": 2.644776119402985,
      "grad_norm": 1.5558581352233887,
      "learning_rate": 0.00017595726519505043,
      "loss": 0.5200609683990478,
      "mean_token_accuracy": 0.848398495465517,
      "num_tokens": 10929729.0,
      "step": 1330
    },
    {
      "entropy": 0.5076324449852109,
      "epoch": 2.664676616915423,
      "grad_norm": 2.233233690261841,
      "learning_rate": 0.00017552796576134985,
      "loss": 0.5190268993377686,
      "mean_token_accuracy": 0.841417095810175,
      "num_tokens": 11012769.0,
      "step": 1340
    },
    {
      "entropy": 0.5502721823751926,
      "epoch": 2.684577114427861,
      "grad_norm": 1.8738890886306763,
      "learning_rate": 0.00017509540109840365,
      "loss": 0.547359848022461,
      "mean_token_accuracy": 0.8380319178104401,
      "num_tokens": 11094716.0,
      "step": 1350
    },
    {
      "entropy": 0.5275072449818253,
      "epoch": 2.7044776119402982,
      "grad_norm": 1.8149367570877075,
      "learning_rate": 0.00017465958990687156,
      "loss": 0.5218305110931396,
      "mean_token_accuracy": 0.8568139627575875,
      "num_tokens": 11175013.0,
      "step": 1360
    },
    {
      "entropy": 0.5119684131816029,
      "epoch": 2.7243781094527364,
      "grad_norm": 1.698498010635376,
      "learning_rate": 0.0001742205510277674,
      "loss": 0.5018157482147216,
      "mean_token_accuracy": 0.8495248600840568,
      "num_tokens": 11256288.0,
      "step": 1370
    },
    {
      "entropy": 0.5109101135283709,
      "epoch": 2.744278606965174,
      "grad_norm": 1.837099552154541,
      "learning_rate": 0.00017377830344164464,
      "loss": 0.5008152484893799,
      "mean_token_accuracy": 0.8569738574326038,
      "num_tokens": 11338498.0,
      "step": 1380
    },
    {
      "entropy": 0.5329740956425667,
      "epoch": 2.764179104477612,
      "grad_norm": 2.1258440017700195,
      "learning_rate": 0.00017333286626777564,
      "loss": 0.5308613777160645,
      "mean_token_accuracy": 0.8408495858311653,
      "num_tokens": 11422122.0,
      "step": 1390
    },
    {
      "entropy": 0.5458354912698269,
      "epoch": 2.78407960199005,
      "grad_norm": 2.0416526794433594,
      "learning_rate": 0.00017288425876332527,
      "loss": 0.5461023807525635,
      "mean_token_accuracy": 0.8418126352131367,
      "num_tokens": 11506227.0,
      "step": 1400
    },
    {
      "entropy": 0.5177356921136379,
      "epoch": 2.8039800995024877,
      "grad_norm": 1.8857133388519287,
      "learning_rate": 0.00017243250032251823,
      "loss": 0.5320337772369385,
      "mean_token_accuracy": 0.8437633819878101,
      "num_tokens": 11586611.0,
      "step": 1410
    },
    {
      "entropy": 0.5068972071632742,
      "epoch": 2.8238805970149254,
      "grad_norm": 2.1604552268981934,
      "learning_rate": 0.00017197761047580082,
      "loss": 0.4947951793670654,
      "mean_token_accuracy": 0.8533644363284111,
      "num_tokens": 11670179.0,
      "step": 1420
    },
    {
      "entropy": 0.491135448589921,
      "epoch": 2.843781094527363,
      "grad_norm": 1.9001272916793823,
      "learning_rate": 0.00017151960888899627,
      "loss": 0.5062966346740723,
      "mean_token_accuracy": 0.8556574188172817,
      "num_tokens": 11750106.0,
      "step": 1430
    },
    {
      "entropy": 0.5211849508807063,
      "epoch": 2.863681592039801,
      "grad_norm": 1.9398213624954224,
      "learning_rate": 0.00017105851536245492,
      "loss": 0.48212461471557616,
      "mean_token_accuracy": 0.8569056294858456,
      "num_tokens": 11831237.0,
      "step": 1440
    },
    {
      "entropy": 0.4800686825066805,
      "epoch": 2.883582089552239,
      "grad_norm": 1.8992606401443481,
      "learning_rate": 0.0001705943498301979,
      "loss": 0.5234591960906982,
      "mean_token_accuracy": 0.8473109625279903,
      "num_tokens": 11915761.0,
      "step": 1450
    },
    {
      "entropy": 0.5110673246905207,
      "epoch": 2.9034825870646768,
      "grad_norm": 1.7360588312149048,
      "learning_rate": 0.00017012713235905547,
      "loss": 0.5000662803649902,
      "mean_token_accuracy": 0.8583845689892768,
      "num_tokens": 11996666.0,
      "step": 1460
    },
    {
      "entropy": 0.49351408518850803,
      "epoch": 2.9233830845771145,
      "grad_norm": 2.0405280590057373,
      "learning_rate": 0.00016965688314779956,
      "loss": 0.5083964824676513,
      "mean_token_accuracy": 0.8502242051064968,
      "num_tokens": 12082260.0,
      "step": 1470
    },
    {
      "entropy": 0.5278301501646638,
      "epoch": 2.943283582089552,
      "grad_norm": 1.9192203283309937,
      "learning_rate": 0.00016918362252627036,
      "loss": 0.5238072872161865,
      "mean_token_accuracy": 0.8439404472708703,
      "num_tokens": 12160973.0,
      "step": 1480
    },
    {
      "entropy": 0.5115002764388918,
      "epoch": 2.96318407960199,
      "grad_norm": 1.6478546857833862,
      "learning_rate": 0.00016870737095449754,
      "loss": 0.5171586036682129,
      "mean_token_accuracy": 0.8549239777028561,
      "num_tokens": 12244781.0,
      "step": 1490
    },
    {
      "entropy": 0.4658357990905643,
      "epoch": 2.983084577114428,
      "grad_norm": 2.152078866958618,
      "learning_rate": 0.00016822814902181583,
      "loss": 0.4724470615386963,
      "mean_token_accuracy": 0.8588747374713421,
      "num_tokens": 12328430.0,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_entropy": 0.6918873670421052,
      "eval_loss": 1.013381004333496,
      "eval_mean_token_accuracy": 0.7516806636537824,
      "eval_num_tokens": 12397968.0,
      "eval_runtime": 89.0725,
      "eval_samples_per_second": 11.62,
      "eval_steps_per_second": 5.815,
      "step": 1509
    }
  ],
  "logging_steps": 10,
  "max_steps": 5030,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.6428765983473664e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}