{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 78,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.3651542961597443,
      "epoch": 0.012944983818770227,
      "grad_norm": 1.3304885625839233,
      "learning_rate": 0.0,
      "loss": 2.1373414993286133,
      "mean_token_accuracy": 0.5150357112288475,
      "num_tokens": 129715.0,
      "step": 1
    },
    {
      "entropy": 1.3770242929458618,
      "epoch": 0.025889967637540454,
      "grad_norm": 1.3239294290542603,
      "learning_rate": 4.166666666666667e-06,
      "loss": 2.154482841491699,
      "mean_token_accuracy": 0.5157716199755669,
      "num_tokens": 259772.0,
      "step": 2
    },
    {
      "entropy": 1.4346312582492828,
      "epoch": 0.038834951456310676,
      "grad_norm": 1.1268572807312012,
      "learning_rate": 8.333333333333334e-06,
      "loss": 2.0695247650146484,
      "mean_token_accuracy": 0.5210058130323887,
      "num_tokens": 389849.0,
      "step": 3
    },
    {
      "entropy": 1.5402012169361115,
      "epoch": 0.05177993527508091,
      "grad_norm": 0.9283472299575806,
      "learning_rate": 1.25e-05,
      "loss": 2.032970905303955,
      "mean_token_accuracy": 0.5240041315555573,
      "num_tokens": 518576.0,
      "step": 4
    },
    {
      "entropy": 1.6357156783342361,
      "epoch": 0.06472491909385113,
      "grad_norm": 0.5993914604187012,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.8982824087142944,
      "mean_token_accuracy": 0.5411873385310173,
      "num_tokens": 648530.0,
      "step": 5
    },
    {
      "entropy": 1.834080547094345,
      "epoch": 0.07766990291262135,
      "grad_norm": 0.41406694054603577,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 1.8447895050048828,
      "mean_token_accuracy": 0.5445547252893448,
      "num_tokens": 778260.0,
      "step": 6
    },
    {
      "entropy": 1.9631124883890152,
      "epoch": 0.09061488673139159,
      "grad_norm": 0.5057855844497681,
      "learning_rate": 2.5e-05,
      "loss": 1.791555404663086,
      "mean_token_accuracy": 0.5538995414972305,
      "num_tokens": 908287.0,
      "step": 7
    },
    {
      "entropy": 2.0287728756666183,
      "epoch": 0.10355987055016182,
      "grad_norm": 0.5796622633934021,
      "learning_rate": 2.916666666666667e-05,
      "loss": 1.7787578105926514,
      "mean_token_accuracy": 0.5525580644607544,
      "num_tokens": 1038626.0,
      "step": 8
    },
    {
      "entropy": 1.8969777673482895,
      "epoch": 0.11650485436893204,
      "grad_norm": 0.49578264355659485,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.6710119247436523,
      "mean_token_accuracy": 0.5709376186132431,
      "num_tokens": 1168231.0,
      "step": 9
    },
    {
      "entropy": 1.7819280922412872,
      "epoch": 0.12944983818770225,
      "grad_norm": 0.39638879895210266,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 1.6355412006378174,
      "mean_token_accuracy": 0.5758759081363678,
      "num_tokens": 1298953.0,
      "step": 10
    },
    {
      "entropy": 1.6452170461416245,
      "epoch": 0.1423948220064725,
      "grad_norm": 0.31017157435417175,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.605017066001892,
      "mean_token_accuracy": 0.5790466815233231,
      "num_tokens": 1429030.0,
      "step": 11
    },
    {
      "entropy": 1.5164579153060913,
      "epoch": 0.1553398058252427,
      "grad_norm": 0.29436805844306946,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 1.5629851818084717,
      "mean_token_accuracy": 0.5837943032383919,
      "num_tokens": 1556810.0,
      "step": 12
    },
    {
      "entropy": 1.4386323243379593,
      "epoch": 0.16828478964401294,
      "grad_norm": 0.2829895317554474,
      "learning_rate": 5e-05,
      "loss": 1.5202760696411133,
      "mean_token_accuracy": 0.5939656496047974,
      "num_tokens": 1687660.0,
      "step": 13
    },
    {
      "entropy": 1.3815916031599045,
      "epoch": 0.18122977346278318,
      "grad_norm": 0.25666117668151855,
      "learning_rate": 4.9997496794168726e-05,
      "loss": 1.4685187339782715,
      "mean_token_accuracy": 0.6037300601601601,
      "num_tokens": 1818622.0,
      "step": 14
    },
    {
      "entropy": 1.4070003181695938,
      "epoch": 0.1941747572815534,
      "grad_norm": 0.2441503256559372,
      "learning_rate": 4.998998767795805e-05,
      "loss": 1.447906732559204,
      "mean_token_accuracy": 0.6062909737229347,
      "num_tokens": 1948742.0,
      "step": 15
    },
    {
      "entropy": 1.4435111433267593,
      "epoch": 0.20711974110032363,
      "grad_norm": 0.26153597235679626,
      "learning_rate": 4.9977474155117045e-05,
      "loss": 1.4333127737045288,
      "mean_token_accuracy": 0.607229471206665,
      "num_tokens": 2078651.0,
      "step": 16
    },
    {
      "entropy": 1.427567034959793,
      "epoch": 0.22006472491909385,
      "grad_norm": 0.23656481504440308,
      "learning_rate": 4.995995873155958e-05,
      "loss": 1.396224021911621,
      "mean_token_accuracy": 0.6134042665362358,
      "num_tokens": 2205557.0,
      "step": 17
    },
    {
      "entropy": 1.433149129152298,
      "epoch": 0.23300970873786409,
      "grad_norm": 0.20005719363689423,
      "learning_rate": 4.99374449148625e-05,
      "loss": 1.392409324645996,
      "mean_token_accuracy": 0.6163519993424416,
      "num_tokens": 2332593.0,
      "step": 18
    },
    {
      "entropy": 1.4219946712255478,
      "epoch": 0.2459546925566343,
      "grad_norm": 0.19785650074481964,
      "learning_rate": 4.9909937213563165e-05,
      "loss": 1.3844692707061768,
      "mean_token_accuracy": 0.6169420257210732,
      "num_tokens": 2461816.0,
      "step": 19
    },
    {
      "entropy": 1.3724359720945358,
      "epoch": 0.2588996763754045,
      "grad_norm": 0.1962433159351349,
      "learning_rate": 4.987744113625665e-05,
      "loss": 1.3383996486663818,
      "mean_token_accuracy": 0.6265854984521866,
      "num_tokens": 2592376.0,
      "step": 20
    },
    {
      "entropy": 1.35466168820858,
      "epoch": 0.27184466019417475,
      "grad_norm": 0.19052749872207642,
      "learning_rate": 4.9839963190492576e-05,
      "loss": 1.326986312866211,
      "mean_token_accuracy": 0.6285954564809799,
      "num_tokens": 2722180.0,
      "step": 21
    },
    {
      "entropy": 1.324196070432663,
      "epoch": 0.284789644012945,
      "grad_norm": 0.1686500757932663,
      "learning_rate": 4.979751088147192e-05,
      "loss": 1.307222604751587,
      "mean_token_accuracy": 0.6329604238271713,
      "num_tokens": 2852915.0,
      "step": 22
    },
    {
      "entropy": 1.3377470970153809,
      "epoch": 0.2977346278317152,
      "grad_norm": 0.16810794174671173,
      "learning_rate": 4.975009271054409e-05,
      "loss": 1.3078091144561768,
      "mean_token_accuracy": 0.6321637183427811,
      "num_tokens": 2983811.0,
      "step": 23
    },
    {
      "entropy": 1.3341724425554276,
      "epoch": 0.3106796116504854,
      "grad_norm": 0.18940500915050507,
      "learning_rate": 4.969771817350445e-05,
      "loss": 1.312269926071167,
      "mean_token_accuracy": 0.6310013681650162,
      "num_tokens": 3114232.0,
      "step": 24
    },
    {
      "entropy": 1.310901552438736,
      "epoch": 0.32362459546925565,
      "grad_norm": 0.1773822158575058,
      "learning_rate": 4.9640397758692715e-05,
      "loss": 1.2985689640045166,
      "mean_token_accuracy": 0.6327972114086151,
      "num_tokens": 3239638.0,
      "step": 25
    },
    {
      "entropy": 1.2651723623275757,
      "epoch": 0.3365695792880259,
      "grad_norm": 0.17164088785648346,
      "learning_rate": 4.957814294489261e-05,
      "loss": 1.2655669450759888,
      "mean_token_accuracy": 0.6391567513346672,
      "num_tokens": 3369129.0,
      "step": 26
    },
    {
      "entropy": 1.2860137075185776,
      "epoch": 0.34951456310679613,
      "grad_norm": 0.16309353709220886,
      "learning_rate": 4.9510966199033174e-05,
      "loss": 1.2871453762054443,
      "mean_token_accuracy": 0.6366170793771744,
      "num_tokens": 3498592.0,
      "step": 27
    },
    {
      "entropy": 1.2693277299404144,
      "epoch": 0.36245954692556637,
      "grad_norm": 0.15370261669158936,
      "learning_rate": 4.943888097369216e-05,
      "loss": 1.264456033706665,
      "mean_token_accuracy": 0.6402129605412483,
      "num_tokens": 3628933.0,
      "step": 28
    },
    {
      "entropy": 1.250020757317543,
      "epoch": 0.37540453074433655,
      "grad_norm": 0.15371794998645782,
      "learning_rate": 4.936190170440208e-05,
      "loss": 1.2349557876586914,
      "mean_token_accuracy": 0.6465005129575729,
      "num_tokens": 3759250.0,
      "step": 29
    },
    {
      "entropy": 1.2473177313804626,
      "epoch": 0.3883495145631068,
      "grad_norm": 0.156179279088974,
      "learning_rate": 4.928004380675941e-05,
      "loss": 1.2247097492218018,
      "mean_token_accuracy": 0.6490338742733002,
      "num_tokens": 3888347.0,
      "step": 30
    },
    {
      "entropy": 1.2503016591072083,
      "epoch": 0.40129449838187703,
      "grad_norm": 0.16324147582054138,
      "learning_rate": 4.9193323673337476e-05,
      "loss": 1.2284612655639648,
      "mean_token_accuracy": 0.6461184471845627,
      "num_tokens": 4018991.0,
      "step": 31
    },
    {
      "entropy": 1.2558909952640533,
      "epoch": 0.41423948220064727,
      "grad_norm": 0.14576396346092224,
      "learning_rate": 4.910175867040377e-05,
      "loss": 1.2321372032165527,
      "mean_token_accuracy": 0.645910307765007,
      "num_tokens": 4149100.0,
      "step": 32
    },
    {
      "entropy": 1.2409392595291138,
      "epoch": 0.42718446601941745,
      "grad_norm": 0.14837919175624847,
      "learning_rate": 4.9005367134442235e-05,
      "loss": 1.2166019678115845,
      "mean_token_accuracy": 0.6500448063015938,
      "num_tokens": 4279686.0,
      "step": 33
    },
    {
      "entropy": 1.2073234170675278,
      "epoch": 0.4401294498381877,
      "grad_norm": 0.1407850831747055,
      "learning_rate": 4.890416836848127e-05,
      "loss": 1.191582441329956,
      "mean_token_accuracy": 0.654162771999836,
      "num_tokens": 4409879.0,
      "step": 34
    },
    {
      "entropy": 1.224804401397705,
      "epoch": 0.45307443365695793,
      "grad_norm": 0.14388902485370636,
      "learning_rate": 4.8798182638228166e-05,
      "loss": 1.2188255786895752,
      "mean_token_accuracy": 0.649294339120388,
      "num_tokens": 4539958.0,
      "step": 35
    },
    {
      "entropy": 1.1996213346719742,
      "epoch": 0.46601941747572817,
      "grad_norm": 0.13835251331329346,
      "learning_rate": 4.868743116801074e-05,
      "loss": 1.1903774738311768,
      "mean_token_accuracy": 0.6544475257396698,
      "num_tokens": 4668863.0,
      "step": 36
    },
    {
      "entropy": 1.166234239935875,
      "epoch": 0.47896440129449835,
      "grad_norm": 0.13608549535274506,
      "learning_rate": 4.857193613652711e-05,
      "loss": 1.1813488006591797,
      "mean_token_accuracy": 0.6577481329441071,
      "num_tokens": 4799874.0,
      "step": 37
    },
    {
      "entropy": 1.197123497724533,
      "epoch": 0.4919093851132686,
      "grad_norm": 0.13925297558307648,
      "learning_rate": 4.845172067240415e-05,
      "loss": 1.2087821960449219,
      "mean_token_accuracy": 0.6513781398534775,
      "num_tokens": 4929372.0,
      "step": 38
    },
    {
      "entropy": 1.1750382333993912,
      "epoch": 0.5048543689320388,
      "grad_norm": 0.13949091732501984,
      "learning_rate": 4.8326808849565936e-05,
      "loss": 1.177015781402588,
      "mean_token_accuracy": 0.6557950600981712,
      "num_tokens": 5059840.0,
      "step": 39
    },
    {
      "entropy": 1.1692189127206802,
      "epoch": 0.517799352750809,
      "grad_norm": 0.14397211372852325,
      "learning_rate": 4.819722568241274e-05,
      "loss": 1.1632399559020996,
      "mean_token_accuracy": 0.6591483578085899,
      "num_tokens": 5189026.0,
      "step": 40
    },
    {
      "entropy": 1.1593063026666641,
      "epoch": 0.5307443365695793,
      "grad_norm": 0.13899798691272736,
      "learning_rate": 4.806299712081172e-05,
      "loss": 1.1517739295959473,
      "mean_token_accuracy": 0.6616862490773201,
      "num_tokens": 5319304.0,
      "step": 41
    },
    {
      "entropy": 1.1965416818857193,
      "epoch": 0.5436893203883495,
      "grad_norm": 0.13963711261749268,
      "learning_rate": 4.792415004490034e-05,
      "loss": 1.1859239339828491,
      "mean_token_accuracy": 0.6556195169687271,
      "num_tokens": 5449076.0,
      "step": 42
    },
    {
      "entropy": 1.1787460893392563,
      "epoch": 0.5566343042071198,
      "grad_norm": 0.13826416432857513,
      "learning_rate": 4.77807122597034e-05,
      "loss": 1.154212236404419,
      "mean_token_accuracy": 0.6633580550551414,
      "num_tokens": 5578972.0,
      "step": 43
    },
    {
      "entropy": 1.1719040125608444,
      "epoch": 0.56957928802589,
      "grad_norm": 0.13135474920272827,
      "learning_rate": 4.7632712489564926e-05,
      "loss": 1.152864694595337,
      "mean_token_accuracy": 0.6626269146800041,
      "num_tokens": 5709351.0,
      "step": 44
    },
    {
      "entropy": 1.1761715561151505,
      "epoch": 0.5825242718446602,
      "grad_norm": 0.1382664293050766,
      "learning_rate": 4.748018037239592e-05,
      "loss": 1.1596863269805908,
      "mean_token_accuracy": 0.66019656509161,
      "num_tokens": 5839241.0,
      "step": 45
    },
    {
      "entropy": 1.1685755103826523,
      "epoch": 0.5954692556634305,
      "grad_norm": 0.1334962695837021,
      "learning_rate": 4.732314645373921e-05,
      "loss": 1.1554309129714966,
      "mean_token_accuracy": 0.6611259058117867,
      "num_tokens": 5969064.0,
      "step": 46
    },
    {
      "entropy": 1.143287107348442,
      "epoch": 0.6084142394822006,
      "grad_norm": 0.13915717601776123,
      "learning_rate": 4.7161642180652464e-05,
      "loss": 1.13901948928833,
      "mean_token_accuracy": 0.6638868674635887,
      "num_tokens": 6099092.0,
      "step": 47
    },
    {
      "entropy": 1.1229832172393799,
      "epoch": 0.6213592233009708,
      "grad_norm": 0.16735908389091492,
      "learning_rate": 4.699569989541074e-05,
      "loss": 1.1253852844238281,
      "mean_token_accuracy": 0.6694925650954247,
      "num_tokens": 6229169.0,
      "step": 48
    },
    {
      "entropy": 1.142636001110077,
      "epoch": 0.6343042071197411,
      "grad_norm": 0.1340433806180954,
      "learning_rate": 4.6825352829029705e-05,
      "loss": 1.1438826322555542,
      "mean_token_accuracy": 0.6639438346028328,
      "num_tokens": 6359167.0,
      "step": 49
    },
    {
      "entropy": 1.1389989107847214,
      "epoch": 0.6472491909385113,
      "grad_norm": 0.1330229490995407,
      "learning_rate": 4.665063509461097e-05,
      "loss": 1.1328400373458862,
      "mean_token_accuracy": 0.668202169239521,
      "num_tokens": 6489483.0,
      "step": 50
    },
    {
      "entropy": 1.1350421458482742,
      "epoch": 0.6601941747572816,
      "grad_norm": 0.13832047581672668,
      "learning_rate": 4.647158168051066e-05,
      "loss": 1.134395956993103,
      "mean_token_accuracy": 0.6662710979580879,
      "num_tokens": 6618304.0,
      "step": 51
    },
    {
      "entropy": 1.1280719488859177,
      "epoch": 0.6731391585760518,
      "grad_norm": 0.13961584866046906,
      "learning_rate": 4.628822844333278e-05,
      "loss": 1.1157816648483276,
      "mean_token_accuracy": 0.6674881353974342,
      "num_tokens": 6748534.0,
      "step": 52
    },
    {
      "entropy": 1.1400533020496368,
      "epoch": 0.686084142394822,
      "grad_norm": 0.13416820764541626,
      "learning_rate": 4.6100612100748765e-05,
      "loss": 1.1322004795074463,
      "mean_token_accuracy": 0.664855606853962,
      "num_tokens": 6878351.0,
      "step": 53
    },
    {
      "entropy": 1.117271676659584,
      "epoch": 0.6990291262135923,
      "grad_norm": 0.13155770301818848,
      "learning_rate": 4.59087702241444e-05,
      "loss": 1.1148383617401123,
      "mean_token_accuracy": 0.6699468269944191,
      "num_tokens": 7008842.0,
      "step": 54
    },
    {
      "entropy": 1.1524481028318405,
      "epoch": 0.7119741100323624,
      "grad_norm": 0.13490092754364014,
      "learning_rate": 4.571274123109606e-05,
      "loss": 1.1433762311935425,
      "mean_token_accuracy": 0.6656594127416611,
      "num_tokens": 7139302.0,
      "step": 55
    },
    {
      "entropy": 1.128203958272934,
      "epoch": 0.7249190938511327,
      "grad_norm": 0.13910333812236786,
      "learning_rate": 4.551256437767719e-05,
      "loss": 1.1198092699050903,
      "mean_token_accuracy": 0.6694299727678299,
      "num_tokens": 7268920.0,
      "step": 56
    },
    {
      "entropy": 1.1252606362104416,
      "epoch": 0.7378640776699029,
      "grad_norm": 0.13880470395088196,
      "learning_rate": 4.530827975059715e-05,
      "loss": 1.1268362998962402,
      "mean_token_accuracy": 0.6689485609531403,
      "num_tokens": 7399537.0,
      "step": 57
    },
    {
      "entropy": 1.1418287009000778,
      "epoch": 0.7508090614886731,
      "grad_norm": 0.1401115506887436,
      "learning_rate": 4.5099928259173516e-05,
      "loss": 1.1322367191314697,
      "mean_token_accuracy": 0.6656085550785065,
      "num_tokens": 7525788.0,
      "step": 58
    },
    {
      "entropy": 1.1480491310358047,
      "epoch": 0.7637540453074434,
      "grad_norm": 0.14310455322265625,
      "learning_rate": 4.488755162713975e-05,
      "loss": 1.1316089630126953,
      "mean_token_accuracy": 0.6664536520838737,
      "num_tokens": 7656150.0,
      "step": 59
    },
    {
      "entropy": 1.1366282999515533,
      "epoch": 0.7766990291262136,
      "grad_norm": 0.12998667359352112,
      "learning_rate": 4.467119238428975e-05,
      "loss": 1.1231651306152344,
      "mean_token_accuracy": 0.6681933254003525,
      "num_tokens": 7786291.0,
      "step": 60
    },
    {
      "entropy": 1.105496659874916,
      "epoch": 0.7896440129449838,
      "grad_norm": 0.1398758739233017,
      "learning_rate": 4.445089385796099e-05,
      "loss": 1.0939688682556152,
      "mean_token_accuracy": 0.6759221702814102,
      "num_tokens": 7916189.0,
      "step": 61
    },
    {
      "entropy": 1.1114656925201416,
      "epoch": 0.8025889967637541,
      "grad_norm": 0.13194960355758667,
      "learning_rate": 4.422670016435792e-05,
      "loss": 1.1039596796035767,
      "mean_token_accuracy": 0.6731015890836716,
      "num_tokens": 8046519.0,
      "step": 62
    },
    {
      "entropy": 1.0888657569885254,
      "epoch": 0.8155339805825242,
      "grad_norm": 0.1397247463464737,
      "learning_rate": 4.3998656199717435e-05,
      "loss": 1.0911369323730469,
      "mean_token_accuracy": 0.6746554225683212,
      "num_tokens": 8176740.0,
      "step": 63
    },
    {
      "entropy": 1.1219657808542252,
      "epoch": 0.8284789644012945,
      "grad_norm": 0.13516487181186676,
      "learning_rate": 4.3766807631318106e-05,
      "loss": 1.1249346733093262,
      "mean_token_accuracy": 0.668804943561554,
      "num_tokens": 8306981.0,
      "step": 64
    },
    {
      "entropy": 1.096419408917427,
      "epoch": 0.8414239482200647,
      "grad_norm": 0.13442112505435944,
      "learning_rate": 4.353120088833501e-05,
      "loss": 1.0880069732666016,
      "mean_token_accuracy": 0.675564743578434,
      "num_tokens": 8437012.0,
      "step": 65
    },
    {
      "entropy": 1.0983364135026932,
      "epoch": 0.8543689320388349,
      "grad_norm": 0.1384594589471817,
      "learning_rate": 4.329188315254196e-05,
      "loss": 1.0896861553192139,
      "mean_token_accuracy": 0.6748483330011368,
      "num_tokens": 8567432.0,
      "step": 66
    },
    {
      "entropy": 1.1061086654663086,
      "epoch": 0.8673139158576052,
      "grad_norm": 0.12975235283374786,
      "learning_rate": 4.3048902348863116e-05,
      "loss": 1.092883825302124,
      "mean_token_accuracy": 0.6738436594605446,
      "num_tokens": 8697482.0,
      "step": 67
    },
    {
      "entropy": 1.1013841927051544,
      "epoch": 0.8802588996763754,
      "grad_norm": 0.13858407735824585,
      "learning_rate": 4.280230713577564e-05,
      "loss": 1.0877230167388916,
      "mean_token_accuracy": 0.6764405071735382,
      "num_tokens": 8826353.0,
      "step": 68
    },
    {
      "entropy": 1.1029959172010422,
      "epoch": 0.8932038834951457,
      "grad_norm": 0.13860292732715607,
      "learning_rate": 4.255214689556557e-05,
      "loss": 1.1019368171691895,
      "mean_token_accuracy": 0.6718069836497307,
      "num_tokens": 8955338.0,
      "step": 69
    },
    {
      "entropy": 1.1007527112960815,
      "epoch": 0.9061488673139159,
      "grad_norm": 0.13615350425243378,
      "learning_rate": 4.229847172443866e-05,
      "loss": 1.1066391468048096,
      "mean_token_accuracy": 0.6710008159279823,
      "num_tokens": 9085706.0,
      "step": 70
    },
    {
      "entropy": 1.0915730148553848,
      "epoch": 0.919093851132686,
      "grad_norm": 0.1343710720539093,
      "learning_rate": 4.204133242248832e-05,
      "loss": 1.0948892831802368,
      "mean_token_accuracy": 0.6748337969183922,
      "num_tokens": 9216028.0,
      "step": 71
    },
    {
      "entropy": 1.1037265211343765,
      "epoch": 0.9320388349514563,
      "grad_norm": 0.13998186588287354,
      "learning_rate": 4.1780780483522575e-05,
      "loss": 1.097463607788086,
      "mean_token_accuracy": 0.6743032932281494,
      "num_tokens": 9346732.0,
      "step": 72
    },
    {
      "entropy": 1.081187054514885,
      "epoch": 0.9449838187702265,
      "grad_norm": 0.13939902186393738,
      "learning_rate": 4.151686808475204e-05,
      "loss": 1.0713167190551758,
      "mean_token_accuracy": 0.6809784397482872,
      "num_tokens": 9477265.0,
      "step": 73
    },
    {
      "entropy": 1.092852994799614,
      "epoch": 0.9579288025889967,
      "grad_norm": 0.13595172762870789,
      "learning_rate": 4.1249648076341165e-05,
      "loss": 1.085597038269043,
      "mean_token_accuracy": 0.6780063956975937,
      "num_tokens": 9606420.0,
      "step": 74
    },
    {
      "entropy": 1.09284046292305,
      "epoch": 0.970873786407767,
      "grad_norm": 0.12815804779529572,
      "learning_rate": 4.0979173970824626e-05,
      "loss": 1.0821534395217896,
      "mean_token_accuracy": 0.6755026131868362,
      "num_tokens": 9737245.0,
      "step": 75
    },
    {
      "entropy": 1.0906607061624527,
      "epoch": 0.9838187702265372,
      "grad_norm": 0.14664161205291748,
      "learning_rate": 4.070549993239106e-05,
      "loss": 1.0785982608795166,
      "mean_token_accuracy": 0.6788846999406815,
      "num_tokens": 9867367.0,
      "step": 76
    },
    {
      "entropy": 1.0882694870233536,
      "epoch": 0.9967637540453075,
      "grad_norm": 0.13629893958568573,
      "learning_rate": 4.0428680766036384e-05,
      "loss": 1.0791587829589844,
      "mean_token_accuracy": 0.6776749864220619,
      "num_tokens": 9998012.0,
      "step": 77
    },
    {
      "entropy": 1.1047788858413696,
      "epoch": 1.0,
      "grad_norm": 0.24593353271484375,
      "learning_rate": 4.0148771906588706e-05,
      "loss": 1.063537359237671,
      "mean_token_accuracy": 0.6715989708900452,
      "num_tokens": 10006202.0,
      "step": 78
    }
  ],
  "logging_steps": 1,
  "max_steps": 234,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0582235533772063e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}