{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8021390374331551,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.3211480975151062,
      "epoch": 0.0005347593582887701,
      "grad_norm": 1.212019681930542,
      "learning_rate": 0.0,
      "loss": 1.8108,
      "mean_token_accuracy": 0.6068548411130905,
      "num_tokens": 32768.0,
      "step": 1
    },
    {
      "entropy": 1.319978803396225,
      "epoch": 0.0010695187165775401,
      "grad_norm": 1.2323760986328125,
      "learning_rate": 3.5087719298245615e-06,
      "loss": 1.8242,
      "mean_token_accuracy": 0.6032210290431976,
      "num_tokens": 65399.0,
      "step": 2
    },
    {
      "entropy": 1.3422840237617493,
      "epoch": 0.0016042780748663102,
      "grad_norm": 1.2383126020431519,
      "learning_rate": 7.017543859649123e-06,
      "loss": 1.8527,
      "mean_token_accuracy": 0.5934058427810669,
      "num_tokens": 97907.0,
      "step": 3
    },
    {
      "entropy": 1.342028558254242,
      "epoch": 0.0021390374331550803,
      "grad_norm": 1.2295197248458862,
      "learning_rate": 1.0526315789473684e-05,
      "loss": 1.8324,
      "mean_token_accuracy": 0.6023643612861633,
      "num_tokens": 130675.0,
      "step": 4
    },
    {
      "entropy": 1.36601322889328,
      "epoch": 0.00267379679144385,
      "grad_norm": 1.1857184171676636,
      "learning_rate": 1.4035087719298246e-05,
      "loss": 1.8561,
      "mean_token_accuracy": 0.5968963950872421,
      "num_tokens": 163443.0,
      "step": 5
    },
    {
      "entropy": 1.3593934178352356,
      "epoch": 0.0032085561497326204,
      "grad_norm": 1.140973448753357,
      "learning_rate": 1.7543859649122806e-05,
      "loss": 1.8346,
      "mean_token_accuracy": 0.5957307517528534,
      "num_tokens": 195506.0,
      "step": 6
    },
    {
      "entropy": 1.4132839739322662,
      "epoch": 0.0037433155080213902,
      "grad_norm": 1.1102650165557861,
      "learning_rate": 2.105263157894737e-05,
      "loss": 1.8617,
      "mean_token_accuracy": 0.5902331173419952,
      "num_tokens": 228061.0,
      "step": 7
    },
    {
      "entropy": 1.331324964761734,
      "epoch": 0.0042780748663101605,
      "grad_norm": 1.0002244710922241,
      "learning_rate": 2.456140350877193e-05,
      "loss": 1.741,
      "mean_token_accuracy": 0.6074766218662262,
      "num_tokens": 260804.0,
      "step": 8
    },
    {
      "entropy": 1.426033079624176,
      "epoch": 0.004812834224598931,
      "grad_norm": 0.8240267038345337,
      "learning_rate": 2.8070175438596492e-05,
      "loss": 1.7741,
      "mean_token_accuracy": 0.6042277812957764,
      "num_tokens": 293572.0,
      "step": 9
    },
    {
      "entropy": 1.4304762780666351,
      "epoch": 0.0053475935828877,
      "grad_norm": 0.6928900480270386,
      "learning_rate": 3.157894736842105e-05,
      "loss": 1.7112,
      "mean_token_accuracy": 0.6084738671779633,
      "num_tokens": 326340.0,
      "step": 10
    },
    {
      "entropy": 1.4698238670825958,
      "epoch": 0.0058823529411764705,
      "grad_norm": 0.6097795367240906,
      "learning_rate": 3.508771929824561e-05,
      "loss": 1.6587,
      "mean_token_accuracy": 0.6164537668228149,
      "num_tokens": 358594.0,
      "step": 11
    },
    {
      "entropy": 1.5248451828956604,
      "epoch": 0.006417112299465241,
      "grad_norm": 0.5526272058486938,
      "learning_rate": 3.859649122807018e-05,
      "loss": 1.6538,
      "mean_token_accuracy": 0.6143695116043091,
      "num_tokens": 391362.0,
      "step": 12
    },
    {
      "entropy": 1.5860177278518677,
      "epoch": 0.006951871657754011,
      "grad_norm": 0.5539830923080444,
      "learning_rate": 4.210526315789474e-05,
      "loss": 1.67,
      "mean_token_accuracy": 0.6128115952014923,
      "num_tokens": 424130.0,
      "step": 13
    },
    {
      "entropy": 1.5437233448028564,
      "epoch": 0.0074866310160427805,
      "grad_norm": 0.6331645846366882,
      "learning_rate": 4.56140350877193e-05,
      "loss": 1.5714,
      "mean_token_accuracy": 0.6330629885196686,
      "num_tokens": 456485.0,
      "step": 14
    },
    {
      "entropy": 1.6081304848194122,
      "epoch": 0.008021390374331552,
      "grad_norm": 0.6115171909332275,
      "learning_rate": 4.912280701754386e-05,
      "loss": 1.5736,
      "mean_token_accuracy": 0.6276161074638367,
      "num_tokens": 489052.0,
      "step": 15
    },
    {
      "entropy": 1.547630399465561,
      "epoch": 0.008556149732620321,
      "grad_norm": 0.5796217918395996,
      "learning_rate": 5.2631578947368424e-05,
      "loss": 1.475,
      "mean_token_accuracy": 0.6473007500171661,
      "num_tokens": 521691.0,
      "step": 16
    },
    {
      "entropy": 1.6032908260822296,
      "epoch": 0.00909090909090909,
      "grad_norm": 0.5352026224136353,
      "learning_rate": 5.6140350877192984e-05,
      "loss": 1.475,
      "mean_token_accuracy": 0.6525703221559525,
      "num_tokens": 554357.0,
      "step": 17
    },
    {
      "entropy": 1.5571948289871216,
      "epoch": 0.009625668449197862,
      "grad_norm": 0.49015113711357117,
      "learning_rate": 5.9649122807017544e-05,
      "loss": 1.4005,
      "mean_token_accuracy": 0.669808566570282,
      "num_tokens": 586990.0,
      "step": 18
    },
    {
      "entropy": 1.5882920324802399,
      "epoch": 0.010160427807486631,
      "grad_norm": 0.5017333030700684,
      "learning_rate": 6.31578947368421e-05,
      "loss": 1.408,
      "mean_token_accuracy": 0.6676928251981735,
      "num_tokens": 619449.0,
      "step": 19
    },
    {
      "entropy": 1.5626413524150848,
      "epoch": 0.0106951871657754,
      "grad_norm": 0.5140109062194824,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.3925,
      "mean_token_accuracy": 0.6779542714357376,
      "num_tokens": 652165.0,
      "step": 20
    },
    {
      "entropy": 1.4453319311141968,
      "epoch": 0.011229946524064172,
      "grad_norm": 0.5080186128616333,
      "learning_rate": 7.017543859649122e-05,
      "loss": 1.301,
      "mean_token_accuracy": 0.6922348439693451,
      "num_tokens": 684933.0,
      "step": 21
    },
    {
      "entropy": 1.4204711019992828,
      "epoch": 0.011764705882352941,
      "grad_norm": 0.5037912130355835,
      "learning_rate": 7.368421052631579e-05,
      "loss": 1.3096,
      "mean_token_accuracy": 0.690493643283844,
      "num_tokens": 717701.0,
      "step": 22
    },
    {
      "entropy": 1.3163996636867523,
      "epoch": 0.01229946524064171,
      "grad_norm": 0.47926607728004456,
      "learning_rate": 7.719298245614036e-05,
      "loss": 1.2587,
      "mean_token_accuracy": 0.7055350989103317,
      "num_tokens": 750036.0,
      "step": 23
    },
    {
      "entropy": 1.1758967638015747,
      "epoch": 0.012834224598930482,
      "grad_norm": 0.4284074306488037,
      "learning_rate": 8.070175438596491e-05,
      "loss": 1.1893,
      "mean_token_accuracy": 0.7212548851966858,
      "num_tokens": 782804.0,
      "step": 24
    },
    {
      "entropy": 1.1575067341327667,
      "epoch": 0.013368983957219251,
      "grad_norm": 0.4281991720199585,
      "learning_rate": 8.421052631578948e-05,
      "loss": 1.2208,
      "mean_token_accuracy": 0.7123541980981827,
      "num_tokens": 815474.0,
      "step": 25
    },
    {
      "entropy": 1.0813260674476624,
      "epoch": 0.013903743315508022,
      "grad_norm": 0.38478702306747437,
      "learning_rate": 8.771929824561403e-05,
      "loss": 1.1507,
      "mean_token_accuracy": 0.7251185774803162,
      "num_tokens": 848059.0,
      "step": 26
    },
    {
      "entropy": 1.0400638580322266,
      "epoch": 0.014438502673796792,
      "grad_norm": 0.36128902435302734,
      "learning_rate": 9.12280701754386e-05,
      "loss": 1.1269,
      "mean_token_accuracy": 0.7298998236656189,
      "num_tokens": 880827.0,
      "step": 27
    },
    {
      "entropy": 1.0808460414409637,
      "epoch": 0.014973262032085561,
      "grad_norm": 0.3295935094356537,
      "learning_rate": 9.473684210526316e-05,
      "loss": 1.1262,
      "mean_token_accuracy": 0.7256537228822708,
      "num_tokens": 913595.0,
      "step": 28
    },
    {
      "entropy": 1.1501913368701935,
      "epoch": 0.015508021390374332,
      "grad_norm": 0.29811128973960876,
      "learning_rate": 9.824561403508771e-05,
      "loss": 1.1599,
      "mean_token_accuracy": 0.7209208011627197,
      "num_tokens": 945634.0,
      "step": 29
    },
    {
      "entropy": 1.173025131225586,
      "epoch": 0.016042780748663103,
      "grad_norm": 0.2845797836780548,
      "learning_rate": 0.0001017543859649123,
      "loss": 1.135,
      "mean_token_accuracy": 0.721837192773819,
      "num_tokens": 978214.0,
      "step": 30
    },
    {
      "entropy": 1.1669634282588959,
      "epoch": 0.016577540106951873,
      "grad_norm": 0.2711884379386902,
      "learning_rate": 0.00010526315789473685,
      "loss": 1.1043,
      "mean_token_accuracy": 0.7299549430608749,
      "num_tokens": 1010857.0,
      "step": 31
    },
    {
      "entropy": 1.206926554441452,
      "epoch": 0.017112299465240642,
      "grad_norm": 0.25197407603263855,
      "learning_rate": 0.00010877192982456141,
      "loss": 1.1569,
      "mean_token_accuracy": 0.7189331501722336,
      "num_tokens": 1043571.0,
      "step": 32
    },
    {
      "entropy": 1.169577717781067,
      "epoch": 0.01764705882352941,
      "grad_norm": 0.21474692225456238,
      "learning_rate": 0.00011228070175438597,
      "loss": 1.1459,
      "mean_token_accuracy": 0.7232099175453186,
      "num_tokens": 1076339.0,
      "step": 33
    },
    {
      "entropy": 1.1237863898277283,
      "epoch": 0.01818181818181818,
      "grad_norm": 0.18793237209320068,
      "learning_rate": 0.00011578947368421053,
      "loss": 1.1102,
      "mean_token_accuracy": 0.724839448928833,
      "num_tokens": 1109022.0,
      "step": 34
    },
    {
      "entropy": 1.1527383029460907,
      "epoch": 0.01871657754010695,
      "grad_norm": 0.17025186121463776,
      "learning_rate": 0.00011929824561403509,
      "loss": 1.1621,
      "mean_token_accuracy": 0.7170107066631317,
      "num_tokens": 1141346.0,
      "step": 35
    },
    {
      "entropy": 1.1023530662059784,
      "epoch": 0.019251336898395723,
      "grad_norm": 0.16425541043281555,
      "learning_rate": 0.00012280701754385965,
      "loss": 1.1279,
      "mean_token_accuracy": 0.7268872112035751,
      "num_tokens": 1174021.0,
      "step": 36
    },
    {
      "entropy": 1.1330838203430176,
      "epoch": 0.019786096256684493,
      "grad_norm": 0.15079551935195923,
      "learning_rate": 0.0001263157894736842,
      "loss": 1.1532,
      "mean_token_accuracy": 0.7206839919090271,
      "num_tokens": 1206463.0,
      "step": 37
    },
    {
      "entropy": 1.099076896905899,
      "epoch": 0.020320855614973262,
      "grad_norm": 0.1482747495174408,
      "learning_rate": 0.0001298245614035088,
      "loss": 1.1045,
      "mean_token_accuracy": 0.7323130667209625,
      "num_tokens": 1239231.0,
      "step": 38
    },
    {
      "entropy": 1.1178455352783203,
      "epoch": 0.02085561497326203,
      "grad_norm": 0.14701837301254272,
      "learning_rate": 0.00013333333333333334,
      "loss": 1.126,
      "mean_token_accuracy": 0.724346861243248,
      "num_tokens": 1271872.0,
      "step": 39
    },
    {
      "entropy": 1.105290025472641,
      "epoch": 0.0213903743315508,
      "grad_norm": 0.13479559123516083,
      "learning_rate": 0.0001368421052631579,
      "loss": 1.0997,
      "mean_token_accuracy": 0.7327084541320801,
      "num_tokens": 1304443.0,
      "step": 40
    },
    {
      "entropy": 1.0889496803283691,
      "epoch": 0.021925133689839574,
      "grad_norm": 0.13660141825675964,
      "learning_rate": 0.00014035087719298245,
      "loss": 1.0839,
      "mean_token_accuracy": 0.7350012362003326,
      "num_tokens": 1337211.0,
      "step": 41
    },
    {
      "entropy": 1.0803407430648804,
      "epoch": 0.022459893048128343,
      "grad_norm": 0.13740068674087524,
      "learning_rate": 0.00014385964912280703,
      "loss": 1.0622,
      "mean_token_accuracy": 0.7330767512321472,
      "num_tokens": 1369979.0,
      "step": 42
    },
    {
      "entropy": 1.1109019815921783,
      "epoch": 0.022994652406417113,
      "grad_norm": 0.13445861637592316,
      "learning_rate": 0.00014736842105263158,
      "loss": 1.098,
      "mean_token_accuracy": 0.7316307574510574,
      "num_tokens": 1402446.0,
      "step": 43
    },
    {
      "entropy": 1.0604596138000488,
      "epoch": 0.023529411764705882,
      "grad_norm": 0.13360567390918732,
      "learning_rate": 0.00015087719298245616,
      "loss": 1.0433,
      "mean_token_accuracy": 0.740469217300415,
      "num_tokens": 1435214.0,
      "step": 44
    },
    {
      "entropy": 1.0373520702123642,
      "epoch": 0.02406417112299465,
      "grad_norm": 0.135879784822464,
      "learning_rate": 0.0001543859649122807,
      "loss": 1.0364,
      "mean_token_accuracy": 0.7431574016809464,
      "num_tokens": 1467242.0,
      "step": 45
    },
    {
      "entropy": 1.0861902832984924,
      "epoch": 0.02459893048128342,
      "grad_norm": 0.12583845853805542,
      "learning_rate": 0.00015789473684210527,
      "loss": 1.0982,
      "mean_token_accuracy": 0.7279447764158249,
      "num_tokens": 1500010.0,
      "step": 46
    },
    {
      "entropy": 1.1126323342323303,
      "epoch": 0.025133689839572194,
      "grad_norm": 0.13058513402938843,
      "learning_rate": 0.00016140350877192982,
      "loss": 1.1172,
      "mean_token_accuracy": 0.7277615070343018,
      "num_tokens": 1532778.0,
      "step": 47
    },
    {
      "entropy": 1.0311926305294037,
      "epoch": 0.025668449197860963,
      "grad_norm": 0.13519766926765442,
      "learning_rate": 0.0001649122807017544,
      "loss": 1.0401,
      "mean_token_accuracy": 0.74251589179039,
      "num_tokens": 1565546.0,
      "step": 48
    },
    {
      "entropy": 1.047630786895752,
      "epoch": 0.026203208556149733,
      "grad_norm": 0.12377266585826874,
      "learning_rate": 0.00016842105263157895,
      "loss": 1.0609,
      "mean_token_accuracy": 0.7348484992980957,
      "num_tokens": 1598314.0,
      "step": 49
    },
    {
      "entropy": 1.092830628156662,
      "epoch": 0.026737967914438502,
      "grad_norm": 0.14160823822021484,
      "learning_rate": 0.00017192982456140353,
      "loss": 1.1167,
      "mean_token_accuracy": 0.7267261147499084,
      "num_tokens": 1630989.0,
      "step": 50
    },
    {
      "entropy": 1.0829817354679108,
      "epoch": 0.02727272727272727,
      "grad_norm": 0.1272098571062088,
      "learning_rate": 0.00017543859649122806,
      "loss": 1.0993,
      "mean_token_accuracy": 0.7326164692640305,
      "num_tokens": 1663449.0,
      "step": 51
    },
    {
      "entropy": 1.1042539179325104,
      "epoch": 0.027807486631016044,
      "grad_norm": 0.13002848625183105,
      "learning_rate": 0.00017894736842105264,
      "loss": 1.1247,
      "mean_token_accuracy": 0.7220796793699265,
      "num_tokens": 1696217.0,
      "step": 52
    },
    {
      "entropy": 0.9782739132642746,
      "epoch": 0.028342245989304814,
      "grad_norm": 0.1287028193473816,
      "learning_rate": 0.0001824561403508772,
      "loss": 0.9925,
      "mean_token_accuracy": 0.7520161271095276,
      "num_tokens": 1728985.0,
      "step": 53
    },
    {
      "entropy": 1.0866808593273163,
      "epoch": 0.028877005347593583,
      "grad_norm": 0.12313037365674973,
      "learning_rate": 0.00018596491228070177,
      "loss": 1.0956,
      "mean_token_accuracy": 0.72519551217556,
      "num_tokens": 1761753.0,
      "step": 54
    },
    {
      "entropy": 1.0856190919876099,
      "epoch": 0.029411764705882353,
      "grad_norm": 0.129723459482193,
      "learning_rate": 0.00018947368421052632,
      "loss": 1.0936,
      "mean_token_accuracy": 0.7340177297592163,
      "num_tokens": 1794424.0,
      "step": 55
    },
    {
      "entropy": 1.059571549296379,
      "epoch": 0.029946524064171122,
      "grad_norm": 0.12538288533687592,
      "learning_rate": 0.00019298245614035088,
      "loss": 1.0644,
      "mean_token_accuracy": 0.7346231341362,
      "num_tokens": 1827133.0,
      "step": 56
    },
    {
      "entropy": 1.0712441504001617,
      "epoch": 0.03048128342245989,
      "grad_norm": 0.12101280689239502,
      "learning_rate": 0.00019649122807017543,
      "loss": 1.0796,
      "mean_token_accuracy": 0.7299673855304718,
      "num_tokens": 1859852.0,
      "step": 57
    },
    {
      "entropy": 1.0577708184719086,
      "epoch": 0.031016042780748664,
      "grad_norm": 0.12647317349910736,
      "learning_rate": 0.0002,
      "loss": 1.0632,
      "mean_token_accuracy": 0.7361465245485306,
      "num_tokens": 1892494.0,
      "step": 58
    },
    {
      "entropy": 1.0639779716730118,
      "epoch": 0.03155080213903743,
      "grad_norm": 0.22237052023410797,
      "learning_rate": 0.0001999998648809627,
      "loss": 1.0602,
      "mean_token_accuracy": 0.7368035316467285,
      "num_tokens": 1925262.0,
      "step": 59
    },
    {
      "entropy": 1.0609510242938995,
      "epoch": 0.03208556149732621,
      "grad_norm": 0.3154584765434265,
      "learning_rate": 0.00019999945952425653,
      "loss": 1.0658,
      "mean_token_accuracy": 0.7347353398799896,
      "num_tokens": 1957815.0,
      "step": 60
    },
    {
      "entropy": 1.1127621531486511,
      "epoch": 0.032620320855614976,
      "grad_norm": 0.15080197155475616,
      "learning_rate": 0.0001999987839310986,
      "loss": 1.1205,
      "mean_token_accuracy": 0.7253787964582443,
      "num_tokens": 1990583.0,
      "step": 61
    },
    {
      "entropy": 1.1040469706058502,
      "epoch": 0.033155080213903745,
      "grad_norm": 0.16909120976924896,
      "learning_rate": 0.00019999783810351746,
      "loss": 1.0834,
      "mean_token_accuracy": 0.7315352112054825,
      "num_tokens": 2023225.0,
      "step": 62
    },
    {
      "entropy": 1.0955560803413391,
      "epoch": 0.033689839572192515,
      "grad_norm": 0.16623055934906006,
      "learning_rate": 0.00019999662204435317,
      "loss": 1.0766,
      "mean_token_accuracy": 0.7339774519205093,
      "num_tokens": 2055864.0,
      "step": 63
    },
    {
      "entropy": 1.1271432042121887,
      "epoch": 0.034224598930481284,
      "grad_norm": 0.12925198674201965,
      "learning_rate": 0.00019999513575725706,
      "loss": 1.1286,
      "mean_token_accuracy": 0.724672481417656,
      "num_tokens": 2088299.0,
      "step": 64
    },
    {
      "entropy": 1.0197514593601227,
      "epoch": 0.034759358288770054,
      "grad_norm": 0.1556098461151123,
      "learning_rate": 0.00019999337924669195,
      "loss": 1.0207,
      "mean_token_accuracy": 0.7488697469234467,
      "num_tokens": 2121067.0,
      "step": 65
    },
    {
      "entropy": 1.029851257801056,
      "epoch": 0.03529411764705882,
      "grad_norm": 0.15798597037792206,
      "learning_rate": 0.00019999135251793203,
      "loss": 1.0195,
      "mean_token_accuracy": 0.7466137111186981,
      "num_tokens": 2153662.0,
      "step": 66
    },
    {
      "entropy": 1.049491971731186,
      "epoch": 0.03582887700534759,
      "grad_norm": 0.11463714390993118,
      "learning_rate": 0.0001999890555770628,
      "loss": 1.05,
      "mean_token_accuracy": 0.7364568263292313,
      "num_tokens": 2186282.0,
      "step": 67
    },
    {
      "entropy": 1.0518030226230621,
      "epoch": 0.03636363636363636,
      "grad_norm": 0.14380760490894318,
      "learning_rate": 0.0001999864884309812,
      "loss": 1.0533,
      "mean_token_accuracy": 0.7368646264076233,
      "num_tokens": 2219050.0,
      "step": 68
    },
    {
      "entropy": 1.043479710817337,
      "epoch": 0.03689839572192513,
      "grad_norm": 0.14671902358531952,
      "learning_rate": 0.00019998365108739544,
      "loss": 1.0646,
      "mean_token_accuracy": 0.740808829665184,
      "num_tokens": 2250998.0,
      "step": 69
    },
    {
      "entropy": 1.0794665217399597,
      "epoch": 0.0374331550802139,
      "grad_norm": 0.12887804210186005,
      "learning_rate": 0.00019998054355482508,
      "loss": 1.1062,
      "mean_token_accuracy": 0.729044497013092,
      "num_tokens": 2283766.0,
      "step": 70
    },
    {
      "entropy": 1.1021603047847748,
      "epoch": 0.03796791443850268,
      "grad_norm": 0.13224957883358002,
      "learning_rate": 0.00019997716584260088,
      "loss": 1.121,
      "mean_token_accuracy": 0.7261119335889816,
      "num_tokens": 2316534.0,
      "step": 71
    },
    {
      "entropy": 1.0996226370334625,
      "epoch": 0.038502673796791446,
      "grad_norm": 0.12894891202449799,
      "learning_rate": 0.000199973517960865,
      "loss": 1.0928,
      "mean_token_accuracy": 0.7264660596847534,
      "num_tokens": 2349140.0,
      "step": 72
    },
    {
      "entropy": 1.1015254855155945,
      "epoch": 0.039037433155080216,
      "grad_norm": 0.11442919820547104,
      "learning_rate": 0.00019996959992057067,
      "loss": 1.0973,
      "mean_token_accuracy": 0.7288917452096939,
      "num_tokens": 2381908.0,
      "step": 73
    },
    {
      "entropy": 1.0996100306510925,
      "epoch": 0.039572192513368985,
      "grad_norm": 0.13185158371925354,
      "learning_rate": 0.00019996541173348238,
      "loss": 1.0737,
      "mean_token_accuracy": 0.73190838098526,
      "num_tokens": 2414346.0,
      "step": 74
    },
    {
      "entropy": 1.0455727726221085,
      "epoch": 0.040106951871657755,
      "grad_norm": 0.11635924875736237,
      "learning_rate": 0.0001999609534121758,
      "loss": 1.0108,
      "mean_token_accuracy": 0.7491058707237244,
      "num_tokens": 2446922.0,
      "step": 75
    },
    {
      "entropy": 1.0418742299079895,
      "epoch": 0.040641711229946524,
      "grad_norm": 0.11113014817237854,
      "learning_rate": 0.00019995622497003768,
      "loss": 1.023,
      "mean_token_accuracy": 0.7431905567646027,
      "num_tokens": 2479505.0,
      "step": 76
    },
    {
      "entropy": 1.0230610966682434,
      "epoch": 0.041176470588235294,
      "grad_norm": 0.10827413946390152,
      "learning_rate": 0.00019995122642126582,
      "loss": 1.0158,
      "mean_token_accuracy": 0.7468015551567078,
      "num_tokens": 2512204.0,
      "step": 77
    },
    {
      "entropy": 1.037041187286377,
      "epoch": 0.04171122994652406,
      "grad_norm": 0.12525281310081482,
      "learning_rate": 0.00019994595778086917,
      "loss": 1.0359,
      "mean_token_accuracy": 0.7424547970294952,
      "num_tokens": 2544972.0,
      "step": 78
    },
    {
      "entropy": 1.0368339121341705,
      "epoch": 0.04224598930481283,
      "grad_norm": 0.11969482898712158,
      "learning_rate": 0.00019994041906466745,
      "loss": 1.0618,
      "mean_token_accuracy": 0.7377283573150635,
      "num_tokens": 2577602.0,
      "step": 79
    },
    {
      "entropy": 1.0786939859390259,
      "epoch": 0.0427807486631016,
      "grad_norm": 0.12571953237056732,
      "learning_rate": 0.00019993461028929157,
      "loss": 1.0884,
      "mean_token_accuracy": 0.7278703451156616,
      "num_tokens": 2609992.0,
      "step": 80
    },
    {
      "entropy": 1.0542809069156647,
      "epoch": 0.04331550802139037,
      "grad_norm": 0.10666581988334656,
      "learning_rate": 0.0001999285314721832,
      "loss": 1.0673,
      "mean_token_accuracy": 0.7366433292627335,
      "num_tokens": 2642590.0,
      "step": 81
    },
    {
      "entropy": 1.073119729757309,
      "epoch": 0.04385026737967915,
      "grad_norm": 0.12644757330417633,
      "learning_rate": 0.00019992218263159485,
      "loss": 1.0759,
      "mean_token_accuracy": 0.733117088675499,
      "num_tokens": 2674660.0,
      "step": 82
    },
    {
      "entropy": 1.0488686561584473,
      "epoch": 0.04438502673796792,
      "grad_norm": 0.1137339323759079,
      "learning_rate": 0.00019991556378658992,
      "loss": 1.0481,
      "mean_token_accuracy": 0.7383919954299927,
      "num_tokens": 2707428.0,
      "step": 83
    },
    {
      "entropy": 1.0880440175533295,
      "epoch": 0.044919786096256686,
      "grad_norm": 0.11253127455711365,
      "learning_rate": 0.0001999086749570423,
      "loss": 1.0823,
      "mean_token_accuracy": 0.7314790040254593,
      "num_tokens": 2740103.0,
      "step": 84
    },
    {
      "entropy": 1.0373365581035614,
      "epoch": 0.045454545454545456,
      "grad_norm": 0.10806500911712646,
      "learning_rate": 0.00019990151616363684,
      "loss": 1.0379,
      "mean_token_accuracy": 0.7419739663600922,
      "num_tokens": 2772727.0,
      "step": 85
    },
    {
      "entropy": 1.0969046205282211,
      "epoch": 0.045989304812834225,
      "grad_norm": 0.1056627705693245,
      "learning_rate": 0.00019989408742786876,
      "loss": 1.0811,
      "mean_token_accuracy": 0.7304458320140839,
      "num_tokens": 2805247.0,
      "step": 86
    },
    {
      "entropy": 1.06920325756073,
      "epoch": 0.046524064171122995,
      "grad_norm": 0.11184051632881165,
      "learning_rate": 0.00019988638877204402,
      "loss": 1.0608,
      "mean_token_accuracy": 0.7345244139432907,
      "num_tokens": 2837817.0,
      "step": 87
    },
    {
      "entropy": 1.0635611712932587,
      "epoch": 0.047058823529411764,
      "grad_norm": 0.10178792476654053,
      "learning_rate": 0.00019987842021927887,
      "loss": 1.0573,
      "mean_token_accuracy": 0.7352456152439117,
      "num_tokens": 2870585.0,
      "step": 88
    },
    {
      "entropy": 1.0648300647735596,
      "epoch": 0.04759358288770053,
      "grad_norm": 0.10483819246292114,
      "learning_rate": 0.00019987018179350006,
      "loss": 1.0668,
      "mean_token_accuracy": 0.7302337884902954,
      "num_tokens": 2903134.0,
      "step": 89
    },
    {
      "entropy": 1.0471554398536682,
      "epoch": 0.0481283422459893,
      "grad_norm": 0.10856444388628006,
      "learning_rate": 0.00019986167351944466,
      "loss": 1.0554,
      "mean_token_accuracy": 0.73912513256073,
      "num_tokens": 2935902.0,
      "step": 90
    },
    {
      "entropy": 1.0166204869747162,
      "epoch": 0.04866310160427807,
      "grad_norm": 0.11259638518095016,
      "learning_rate": 0.00019985289542266002,
      "loss": 1.0351,
      "mean_token_accuracy": 0.7440738081932068,
      "num_tokens": 2968670.0,
      "step": 91
    },
    {
      "entropy": 1.046201303601265,
      "epoch": 0.04919786096256684,
      "grad_norm": 0.18939243257045746,
      "learning_rate": 0.00019984384752950364,
      "loss": 1.0577,
      "mean_token_accuracy": 0.7357452511787415,
      "num_tokens": 3000903.0,
      "step": 92
    },
    {
      "entropy": 1.0445843040943146,
      "epoch": 0.04973262032085562,
      "grad_norm": 0.11610482633113861,
      "learning_rate": 0.00019983452986714316,
      "loss": 1.0697,
      "mean_token_accuracy": 0.7375977635383606,
      "num_tokens": 3033671.0,
      "step": 93
    },
    {
      "entropy": 1.1029761135578156,
      "epoch": 0.05026737967914439,
      "grad_norm": 0.11175025254487991,
      "learning_rate": 0.00019982494246355617,
      "loss": 1.1181,
      "mean_token_accuracy": 0.725279837846756,
      "num_tokens": 3066347.0,
      "step": 94
    },
    {
      "entropy": 1.0046235769987106,
      "epoch": 0.05080213903743316,
      "grad_norm": 0.10787836462259293,
      "learning_rate": 0.00019981508534753028,
      "loss": 0.986,
      "mean_token_accuracy": 0.751044511795044,
      "num_tokens": 3098766.0,
      "step": 95
    },
    {
      "entropy": 1.0180357545614243,
      "epoch": 0.051336898395721926,
      "grad_norm": 0.10719222575426102,
      "learning_rate": 0.00019980495854866287,
      "loss": 1.0098,
      "mean_token_accuracy": 0.7471800893545151,
      "num_tokens": 3131448.0,
      "step": 96
    },
    {
      "entropy": 1.086255043745041,
      "epoch": 0.051871657754010696,
      "grad_norm": 0.1420079469680786,
      "learning_rate": 0.0001997945620973612,
      "loss": 1.0721,
      "mean_token_accuracy": 0.7333822250366211,
      "num_tokens": 3164216.0,
      "step": 97
    },
    {
      "entropy": 1.0434383749961853,
      "epoch": 0.052406417112299465,
      "grad_norm": 0.11976777762174606,
      "learning_rate": 0.00019978389602484203,
      "loss": 1.0372,
      "mean_token_accuracy": 0.7404081225395203,
      "num_tokens": 3196984.0,
      "step": 98
    },
    {
      "entropy": 1.0948996245861053,
      "epoch": 0.052941176470588235,
      "grad_norm": 0.11928348988294601,
      "learning_rate": 0.00019977296036313182,
      "loss": 1.119,
      "mean_token_accuracy": 0.7248346358537674,
      "num_tokens": 3229387.0,
      "step": 99
    },
    {
      "entropy": 1.0682441890239716,
      "epoch": 0.053475935828877004,
      "grad_norm": 0.11096571385860443,
      "learning_rate": 0.00019976175514506655,
      "loss": 1.0658,
      "mean_token_accuracy": 0.7372749745845795,
      "num_tokens": 3262145.0,
      "step": 100
    },
    {
      "entropy": 1.0616617947816849,
      "epoch": 0.05401069518716577,
      "grad_norm": 0.1181572750210762,
      "learning_rate": 0.00019975028040429145,
      "loss": 1.0604,
      "mean_token_accuracy": 0.7366300523281097,
      "num_tokens": 3294504.0,
      "step": 101
    },
    {
      "entropy": 1.0692074596881866,
      "epoch": 0.05454545454545454,
      "grad_norm": 0.11676756292581558,
      "learning_rate": 0.0001997385361752611,
      "loss": 1.069,
      "mean_token_accuracy": 0.7343024760484695,
      "num_tokens": 3326611.0,
      "step": 102
    },
    {
      "entropy": 1.0682894587516785,
      "epoch": 0.05508021390374331,
      "grad_norm": 0.11502093076705933,
      "learning_rate": 0.00019972652249323924,
      "loss": 1.0762,
      "mean_token_accuracy": 0.7318246513605118,
      "num_tokens": 3358661.0,
      "step": 103
    },
    {
      "entropy": 1.0630333721637726,
      "epoch": 0.05561497326203209,
      "grad_norm": 0.11878052353858948,
      "learning_rate": 0.00019971423939429877,
      "loss": 1.0523,
      "mean_token_accuracy": 0.7361778318881989,
      "num_tokens": 3391416.0,
      "step": 104
    },
    {
      "entropy": 1.0106346756219864,
      "epoch": 0.05614973262032086,
      "grad_norm": 0.10611041635274887,
      "learning_rate": 0.0001997016869153214,
      "loss": 1.0027,
      "mean_token_accuracy": 0.747006356716156,
      "num_tokens": 3424184.0,
      "step": 105
    },
    {
      "entropy": 1.0402246713638306,
      "epoch": 0.05668449197860963,
      "grad_norm": 0.11598736047744751,
      "learning_rate": 0.00019968886509399785,
      "loss": 1.051,
      "mean_token_accuracy": 0.7395623475313187,
      "num_tokens": 3456920.0,
      "step": 106
    },
    {
      "entropy": 1.0482510924339294,
      "epoch": 0.0572192513368984,
      "grad_norm": 0.11055859923362732,
      "learning_rate": 0.00019967577396882738,
      "loss": 1.0407,
      "mean_token_accuracy": 0.7408681511878967,
      "num_tokens": 3489392.0,
      "step": 107
    },
    {
      "entropy": 1.081942230463028,
      "epoch": 0.057754010695187166,
      "grad_norm": 0.11166244745254517,
      "learning_rate": 0.0001996624135791181,
      "loss": 1.0967,
      "mean_token_accuracy": 0.7317631989717484,
      "num_tokens": 3522160.0,
      "step": 108
    },
    {
      "entropy": 1.0274831652641296,
      "epoch": 0.058288770053475936,
      "grad_norm": 0.12437647581100464,
      "learning_rate": 0.0001996487839649865,
      "loss": 1.038,
      "mean_token_accuracy": 0.7414470911026001,
      "num_tokens": 3554175.0,
      "step": 109
    },
    {
      "entropy": 1.0478383004665375,
      "epoch": 0.058823529411764705,
      "grad_norm": 0.10964363068342209,
      "learning_rate": 0.00019963488516735745,
      "loss": 1.0471,
      "mean_token_accuracy": 0.7395887672901154,
      "num_tokens": 3586689.0,
      "step": 110
    },
    {
      "entropy": 1.0617424249649048,
      "epoch": 0.059358288770053474,
      "grad_norm": 0.12272202223539352,
      "learning_rate": 0.00019962071722796416,
      "loss": 1.0658,
      "mean_token_accuracy": 0.7380151003599167,
      "num_tokens": 3618950.0,
      "step": 111
    },
    {
      "entropy": 1.0538257956504822,
      "epoch": 0.059893048128342244,
      "grad_norm": 0.1245708093047142,
      "learning_rate": 0.00019960628018934786,
      "loss": 1.0585,
      "mean_token_accuracy": 0.7368472963571548,
      "num_tokens": 3651324.0,
      "step": 112
    },
    {
      "entropy": 1.0768160820007324,
      "epoch": 0.06042780748663101,
      "grad_norm": 0.1088668629527092,
      "learning_rate": 0.00019959157409485788,
      "loss": 1.0647,
      "mean_token_accuracy": 0.7375608086585999,
      "num_tokens": 3683822.0,
      "step": 113
    },
    {
      "entropy": 1.0708918273448944,
      "epoch": 0.06096256684491978,
      "grad_norm": 0.11890657991170883,
      "learning_rate": 0.00019957659898865143,
      "loss": 1.0666,
      "mean_token_accuracy": 0.7333291918039322,
      "num_tokens": 3716478.0,
      "step": 114
    },
    {
      "entropy": 1.071320116519928,
      "epoch": 0.06149732620320856,
      "grad_norm": 0.12597399950027466,
      "learning_rate": 0.00019956135491569344,
      "loss": 1.0616,
      "mean_token_accuracy": 0.7387379854917526,
      "num_tokens": 3748896.0,
      "step": 115
    },
    {
      "entropy": 1.071579933166504,
      "epoch": 0.06203208556149733,
      "grad_norm": 0.10608608275651932,
      "learning_rate": 0.0001995458419217564,
      "loss": 1.062,
      "mean_token_accuracy": 0.7356106638908386,
      "num_tokens": 3781466.0,
      "step": 116
    },
    {
      "entropy": 1.0784811079502106,
      "epoch": 0.06256684491978609,
      "grad_norm": 0.11198359727859497,
      "learning_rate": 0.00019953006005342036,
      "loss": 1.0794,
      "mean_token_accuracy": 0.7334589809179306,
      "num_tokens": 3813869.0,
      "step": 117
    },
    {
      "entropy": 1.073084980249405,
      "epoch": 0.06310160427807486,
      "grad_norm": 0.11027467995882034,
      "learning_rate": 0.0001995140093580727,
      "loss": 1.0646,
      "mean_token_accuracy": 0.7342070043087006,
      "num_tokens": 3846637.0,
      "step": 118
    },
    {
      "entropy": 1.0337157994508743,
      "epoch": 0.06363636363636363,
      "grad_norm": 0.1102069839835167,
      "learning_rate": 0.00019949768988390795,
      "loss": 1.0444,
      "mean_token_accuracy": 0.7443217188119888,
      "num_tokens": 3879176.0,
      "step": 119
    },
    {
      "entropy": 1.0600185543298721,
      "epoch": 0.06417112299465241,
      "grad_norm": 0.11147669702768326,
      "learning_rate": 0.00019948110167992768,
      "loss": 1.0755,
      "mean_token_accuracy": 0.7351716160774231,
      "num_tokens": 3911359.0,
      "step": 120
    },
    {
      "entropy": 1.0136645436286926,
      "epoch": 0.06470588235294118,
      "grad_norm": 0.11211156100034714,
      "learning_rate": 0.0001994642447959404,
      "loss": 1.0102,
      "mean_token_accuracy": 0.7462002784013748,
      "num_tokens": 3943619.0,
      "step": 121
    },
    {
      "entropy": 1.0536611080169678,
      "epoch": 0.06524064171122995,
      "grad_norm": 0.11316773295402527,
      "learning_rate": 0.00019944711928256134,
      "loss": 1.0641,
      "mean_token_accuracy": 0.7334738671779633,
      "num_tokens": 3976387.0,
      "step": 122
    },
    {
      "entropy": 1.0741864740848541,
      "epoch": 0.06577540106951872,
      "grad_norm": 0.11843208223581314,
      "learning_rate": 0.00019942972519121236,
      "loss": 1.0816,
      "mean_token_accuracy": 0.7349095940589905,
      "num_tokens": 4009155.0,
      "step": 123
    },
    {
      "entropy": 1.052805870771408,
      "epoch": 0.06631016042780749,
      "grad_norm": 0.10992563515901566,
      "learning_rate": 0.00019941206257412174,
      "loss": 1.0584,
      "mean_token_accuracy": 0.7372617423534393,
      "num_tokens": 4041923.0,
      "step": 124
    },
    {
      "entropy": 1.030860036611557,
      "epoch": 0.06684491978609626,
      "grad_norm": 0.10842890292406082,
      "learning_rate": 0.00019939413148432406,
      "loss": 1.0267,
      "mean_token_accuracy": 0.744562566280365,
      "num_tokens": 4074691.0,
      "step": 125
    },
    {
      "entropy": 1.0848415195941925,
      "epoch": 0.06737967914438503,
      "grad_norm": 0.11439650505781174,
      "learning_rate": 0.00019937593197566,
      "loss": 1.0704,
      "mean_token_accuracy": 0.7342680990695953,
      "num_tokens": 4107459.0,
      "step": 126
    },
    {
      "entropy": 1.0568927228450775,
      "epoch": 0.0679144385026738,
      "grad_norm": 0.11309509724378586,
      "learning_rate": 0.00019935746410277625,
      "loss": 1.042,
      "mean_token_accuracy": 0.738794133067131,
      "num_tokens": 4139840.0,
      "step": 127
    },
    {
      "entropy": 1.0811631679534912,
      "epoch": 0.06844919786096257,
      "grad_norm": 0.10690435022115707,
      "learning_rate": 0.00019933872792112525,
      "loss": 1.0668,
      "mean_token_accuracy": 0.7340264767408371,
      "num_tokens": 4172477.0,
      "step": 128
    },
    {
      "entropy": 1.0234280824661255,
      "epoch": 0.06898395721925134,
      "grad_norm": 0.10712087899446487,
      "learning_rate": 0.00019931972348696516,
      "loss": 1.0132,
      "mean_token_accuracy": 0.7456015348434448,
      "num_tokens": 4205042.0,
      "step": 129
    },
    {
      "entropy": 1.0305497348308563,
      "epoch": 0.06951871657754011,
      "grad_norm": 0.11901942640542984,
      "learning_rate": 0.00019930045085735952,
      "loss": 1.0288,
      "mean_token_accuracy": 0.741858035326004,
      "num_tokens": 4237786.0,
      "step": 130
    },
    {
      "entropy": 1.0717166364192963,
      "epoch": 0.07005347593582888,
      "grad_norm": 0.12446736544370651,
      "learning_rate": 0.00019928091009017719,
      "loss": 1.0685,
      "mean_token_accuracy": 0.7327694445848465,
      "num_tokens": 4270369.0,
      "step": 131
    },
    {
      "entropy": 1.0448037981987,
      "epoch": 0.07058823529411765,
      "grad_norm": 0.11024358868598938,
      "learning_rate": 0.00019926110124409216,
      "loss": 1.0584,
      "mean_token_accuracy": 0.7348179519176483,
      "num_tokens": 4303137.0,
      "step": 132
    },
    {
      "entropy": 1.0688583254814148,
      "epoch": 0.07112299465240642,
      "grad_norm": 0.11428970843553543,
      "learning_rate": 0.00019924102437858342,
      "loss": 1.0915,
      "mean_token_accuracy": 0.7348983436822891,
      "num_tokens": 4335600.0,
      "step": 133
    },
    {
      "entropy": 1.0266284346580505,
      "epoch": 0.07165775401069518,
      "grad_norm": 0.11974222213029861,
      "learning_rate": 0.00019922067955393462,
      "loss": 1.0375,
      "mean_token_accuracy": 0.7398888170719147,
      "num_tokens": 4368368.0,
      "step": 134
    },
    {
      "entropy": 1.040811687707901,
      "epoch": 0.07219251336898395,
      "grad_norm": 0.11608098447322845,
      "learning_rate": 0.00019920006683123406,
      "loss": 1.0513,
      "mean_token_accuracy": 0.7392490655183792,
      "num_tokens": 4400868.0,
      "step": 135
    },
    {
      "entropy": 1.0473428666591644,
      "epoch": 0.07272727272727272,
      "grad_norm": 0.10295426100492477,
      "learning_rate": 0.00019917918627237446,
      "loss": 1.0444,
      "mean_token_accuracy": 0.7357507795095444,
      "num_tokens": 4433511.0,
      "step": 136
    },
    {
      "entropy": 1.0344346463680267,
      "epoch": 0.0732620320855615,
      "grad_norm": 0.11141891032457352,
      "learning_rate": 0.00019915803794005274,
      "loss": 1.015,
      "mean_token_accuracy": 0.7442876398563385,
      "num_tokens": 4466279.0,
      "step": 137
    },
    {
      "entropy": 1.0611892938613892,
      "epoch": 0.07379679144385026,
      "grad_norm": 0.11746582388877869,
      "learning_rate": 0.00019913662189776978,
      "loss": 1.0577,
      "mean_token_accuracy": 0.7358400076627731,
      "num_tokens": 4499045.0,
      "step": 138
    },
    {
      "entropy": 1.0629407465457916,
      "epoch": 0.07433155080213903,
      "grad_norm": 0.10986413061618805,
      "learning_rate": 0.00019911493820983043,
      "loss": 1.0569,
      "mean_token_accuracy": 0.7403775751590729,
      "num_tokens": 4531813.0,
      "step": 139
    },
    {
      "entropy": 1.0572038888931274,
      "epoch": 0.0748663101604278,
      "grad_norm": 0.10804872959852219,
      "learning_rate": 0.00019909298694134307,
      "loss": 1.0489,
      "mean_token_accuracy": 0.7369809001684189,
      "num_tokens": 4564407.0,
      "step": 140
    },
    {
      "entropy": 1.04163858294487,
      "epoch": 0.07540106951871657,
      "grad_norm": 0.11756208539009094,
      "learning_rate": 0.0001990707681582196,
      "loss": 1.0528,
      "mean_token_accuracy": 0.739686906337738,
      "num_tokens": 4596642.0,
      "step": 141
    },
    {
      "entropy": 1.102006733417511,
      "epoch": 0.07593582887700535,
      "grad_norm": 0.14406466484069824,
      "learning_rate": 0.00019904828192717506,
      "loss": 1.1079,
      "mean_token_accuracy": 0.7283953577280045,
      "num_tokens": 4629322.0,
      "step": 142
    },
    {
      "entropy": 1.056261733174324,
      "epoch": 0.07647058823529412,
      "grad_norm": 0.10974816232919693,
      "learning_rate": 0.00019902552831572773,
      "loss": 1.0593,
      "mean_token_accuracy": 0.740469217300415,
      "num_tokens": 4662090.0,
      "step": 143
    },
    {
      "entropy": 1.0207159221172333,
      "epoch": 0.07700534759358289,
      "grad_norm": 0.11418622732162476,
      "learning_rate": 0.0001990025073921986,
      "loss": 1.0214,
      "mean_token_accuracy": 0.7460290491580963,
      "num_tokens": 4694445.0,
      "step": 144
    },
    {
      "entropy": 1.0311139822006226,
      "epoch": 0.07754010695187166,
      "grad_norm": 0.11726009100675583,
      "learning_rate": 0.00019897921922571135,
      "loss": 1.0314,
      "mean_token_accuracy": 0.7410428076982498,
      "num_tokens": 4727111.0,
      "step": 145
    },
    {
      "entropy": 1.0391079187393188,
      "epoch": 0.07807486631016043,
      "grad_norm": 0.10857763886451721,
      "learning_rate": 0.00019895566388619207,
      "loss": 1.0379,
      "mean_token_accuracy": 0.7357028722763062,
      "num_tokens": 4759610.0,
      "step": 146
    },
    {
      "entropy": 1.0950137078762054,
      "epoch": 0.0786096256684492,
      "grad_norm": 0.10514643788337708,
      "learning_rate": 0.00019893184144436903,
      "loss": 1.0951,
      "mean_token_accuracy": 0.7290331572294235,
      "num_tokens": 4792024.0,
      "step": 147
    },
    {
      "entropy": 1.0630358755588531,
      "epoch": 0.07914438502673797,
      "grad_norm": 0.11535201221704483,
      "learning_rate": 0.00019890775197177265,
      "loss": 1.0638,
      "mean_token_accuracy": 0.7384617477655411,
      "num_tokens": 4824569.0,
      "step": 148
    },
    {
      "entropy": 1.1177841126918793,
      "epoch": 0.07967914438502674,
      "grad_norm": 0.12657026946544647,
      "learning_rate": 0.00019888339554073505,
      "loss": 1.1044,
      "mean_token_accuracy": 0.7259286493062973,
      "num_tokens": 4857337.0,
      "step": 149
    },
    {
      "entropy": 1.0614158511161804,
      "epoch": 0.08021390374331551,
      "grad_norm": 0.11264298111200333,
      "learning_rate": 0.00019885877222438996,
      "loss": 1.0535,
      "mean_token_accuracy": 0.7389418482780457,
      "num_tokens": 4890105.0,
      "step": 150
    },
    {
      "entropy": 1.0805165469646454,
      "epoch": 0.08074866310160428,
      "grad_norm": 0.11258802562952042,
      "learning_rate": 0.00019883388209667246,
      "loss": 1.0775,
      "mean_token_accuracy": 0.7345370054244995,
      "num_tokens": 4922865.0,
      "step": 151
    },
    {
      "entropy": 1.0690855234861374,
      "epoch": 0.08128342245989305,
      "grad_norm": 0.13155648112297058,
      "learning_rate": 0.00019880872523231875,
      "loss": 1.0856,
      "mean_token_accuracy": 0.7277711480855942,
      "num_tokens": 4955260.0,
      "step": 152
    },
    {
      "entropy": 1.0855732262134552,
      "epoch": 0.08181818181818182,
      "grad_norm": 0.1189059242606163,
      "learning_rate": 0.00019878330170686602,
      "loss": 1.0923,
      "mean_token_accuracy": 0.7301136553287506,
      "num_tokens": 4988028.0,
      "step": 153
    },
    {
      "entropy": 1.0443801879882812,
      "epoch": 0.08235294117647059,
      "grad_norm": 0.11662495881319046,
      "learning_rate": 0.00019875761159665202,
      "loss": 1.0544,
      "mean_token_accuracy": 0.740158811211586,
      "num_tokens": 5020587.0,
      "step": 154
    },
    {
      "entropy": 1.0272859036922455,
      "epoch": 0.08288770053475936,
      "grad_norm": 0.12569770216941833,
      "learning_rate": 0.0001987316549788151,
      "loss": 1.0285,
      "mean_token_accuracy": 0.7470496594905853,
      "num_tokens": 5053027.0,
      "step": 155
    },
    {
      "entropy": 1.0688476860523224,
      "epoch": 0.08342245989304813,
      "grad_norm": 0.11243730783462524,
      "learning_rate": 0.00019870543193129377,
      "loss": 1.0608,
      "mean_token_accuracy": 0.738514170050621,
      "num_tokens": 5085795.0,
      "step": 156
    },
    {
      "entropy": 1.042402446269989,
      "epoch": 0.0839572192513369,
      "grad_norm": 0.11054984480142593,
      "learning_rate": 0.00019867894253282654,
      "loss": 1.0362,
      "mean_token_accuracy": 0.7432184815406799,
      "num_tokens": 5118563.0,
      "step": 157
    },
    {
      "entropy": 1.053287386894226,
      "epoch": 0.08449197860962566,
      "grad_norm": 0.1238577663898468,
      "learning_rate": 0.00019865218686295163,
      "loss": 1.046,
      "mean_token_accuracy": 0.7390121221542358,
      "num_tokens": 5150708.0,
      "step": 158
    },
    {
      "entropy": 1.0408558249473572,
      "epoch": 0.08502673796791443,
      "grad_norm": 0.11221247911453247,
      "learning_rate": 0.0001986251650020069,
      "loss": 1.0432,
      "mean_token_accuracy": 0.7450513243675232,
      "num_tokens": 5183476.0,
      "step": 159
    },
    {
      "entropy": 1.0343620330095291,
      "epoch": 0.0855614973262032,
      "grad_norm": 0.10995996743440628,
      "learning_rate": 0.00019859787703112937,
      "loss": 1.0217,
      "mean_token_accuracy": 0.7431268393993378,
      "num_tokens": 5216244.0,
      "step": 160
    },
    {
      "entropy": 1.0292404145002365,
      "epoch": 0.08609625668449197,
      "grad_norm": 0.11850112676620483,
      "learning_rate": 0.00019857032303225512,
      "loss": 1.037,
      "mean_token_accuracy": 0.7418358474969864,
      "num_tokens": 5249011.0,
      "step": 161
    },
    {
      "entropy": 1.0845747888088226,
      "epoch": 0.08663101604278074,
      "grad_norm": 0.11430878937244415,
      "learning_rate": 0.0001985425030881191,
      "loss": 1.0887,
      "mean_token_accuracy": 0.7280058711767197,
      "num_tokens": 5281779.0,
      "step": 162
    },
    {
      "entropy": 1.0065880417823792,
      "epoch": 0.08716577540106951,
      "grad_norm": 0.1066722646355629,
      "learning_rate": 0.00019851441728225465,
      "loss": 0.9979,
      "mean_token_accuracy": 0.7491233050823212,
      "num_tokens": 5314105.0,
      "step": 163
    },
    {
      "entropy": 1.0722036361694336,
      "epoch": 0.0877005347593583,
      "grad_norm": 0.11814260482788086,
      "learning_rate": 0.00019848606569899358,
      "loss": 1.0843,
      "mean_token_accuracy": 0.7380961626768112,
      "num_tokens": 5346493.0,
      "step": 164
    },
    {
      "entropy": 1.0834770202636719,
      "epoch": 0.08823529411764706,
      "grad_norm": 0.11347079277038574,
      "learning_rate": 0.00019845744842346554,
      "loss": 1.0796,
      "mean_token_accuracy": 0.7332694083452225,
      "num_tokens": 5379196.0,
      "step": 165
    },
    {
      "entropy": 1.0517558157444,
      "epoch": 0.08877005347593583,
      "grad_norm": 0.11098195612430573,
      "learning_rate": 0.0001984285655415981,
      "loss": 1.0576,
      "mean_token_accuracy": 0.7342869937419891,
      "num_tokens": 5411955.0,
      "step": 166
    },
    {
      "entropy": 1.0513665974140167,
      "epoch": 0.0893048128342246,
      "grad_norm": 0.11459293216466904,
      "learning_rate": 0.00019839941714011627,
      "loss": 1.0619,
      "mean_token_accuracy": 0.7377733439207077,
      "num_tokens": 5444628.0,
      "step": 167
    },
    {
      "entropy": 1.0738406777381897,
      "epoch": 0.08983957219251337,
      "grad_norm": 0.10587628185749054,
      "learning_rate": 0.00019837000330654244,
      "loss": 1.0757,
      "mean_token_accuracy": 0.7364008128643036,
      "num_tokens": 5477105.0,
      "step": 168
    },
    {
      "entropy": 1.0741328001022339,
      "epoch": 0.09037433155080214,
      "grad_norm": 0.10436589270830154,
      "learning_rate": 0.0001983403241291959,
      "loss": 1.0673,
      "mean_token_accuracy": 0.7362231314182281,
      "num_tokens": 5509873.0,
      "step": 169
    },
    {
      "entropy": 1.0744796991348267,
      "epoch": 0.09090909090909091,
      "grad_norm": 0.11066372692584991,
      "learning_rate": 0.00019831037969719262,
      "loss": 1.0661,
      "mean_token_accuracy": 0.7409461587667465,
      "num_tokens": 5542172.0,
      "step": 170
    },
    {
      "entropy": 1.0207346379756927,
      "epoch": 0.09144385026737968,
      "grad_norm": 0.10659626126289368,
      "learning_rate": 0.00019828017010044518,
      "loss": 1.014,
      "mean_token_accuracy": 0.7454484403133392,
      "num_tokens": 5574940.0,
      "step": 171
    },
    {
      "entropy": 1.0747233629226685,
      "epoch": 0.09197860962566845,
      "grad_norm": 0.10680273920297623,
      "learning_rate": 0.00019824969542966227,
      "loss": 1.086,
      "mean_token_accuracy": 0.735552579164505,
      "num_tokens": 5607622.0,
      "step": 172
    },
    {
      "entropy": 1.082386240363121,
      "epoch": 0.09251336898395722,
      "grad_norm": 0.10763442516326904,
      "learning_rate": 0.00019821895577634848,
      "loss": 1.0902,
      "mean_token_accuracy": 0.7295721918344498,
      "num_tokens": 5640310.0,
      "step": 173
    },
    {
      "entropy": 1.0541490614414215,
      "epoch": 0.09304812834224599,
      "grad_norm": 0.11163617670536041,
      "learning_rate": 0.0001981879512328041,
      "loss": 1.0566,
      "mean_token_accuracy": 0.7395776063203812,
      "num_tokens": 5672750.0,
      "step": 174
    },
    {
      "entropy": 1.0431351363658905,
      "epoch": 0.09358288770053476,
      "grad_norm": 0.1073818951845169,
      "learning_rate": 0.00019815668189212477,
      "loss": 1.0562,
      "mean_token_accuracy": 0.7366202473640442,
      "num_tokens": 5705518.0,
      "step": 175
    },
    {
      "entropy": 1.0744110643863678,
      "epoch": 0.09411764705882353,
      "grad_norm": 0.1086694523692131,
      "learning_rate": 0.0001981251478482013,
      "loss": 1.0858,
      "mean_token_accuracy": 0.729690670967102,
      "num_tokens": 5738149.0,
      "step": 176
    },
    {
      "entropy": 1.066655308008194,
      "epoch": 0.0946524064171123,
      "grad_norm": 0.10837189108133316,
      "learning_rate": 0.00019809334919571912,
      "loss": 1.0546,
      "mean_token_accuracy": 0.7364211529493332,
      "num_tokens": 5770470.0,
      "step": 177
    },
    {
      "entropy": 1.0561938136816025,
      "epoch": 0.09518716577540107,
      "grad_norm": 0.12487713992595673,
      "learning_rate": 0.00019806128603015838,
      "loss": 1.0408,
      "mean_token_accuracy": 0.7411956340074539,
      "num_tokens": 5802994.0,
      "step": 178
    },
    {
      "entropy": 1.051276072859764,
      "epoch": 0.09572192513368984,
      "grad_norm": 0.12799374759197235,
      "learning_rate": 0.0001980289584477934,
      "loss": 1.047,
      "mean_token_accuracy": 0.740519255399704,
      "num_tokens": 5835383.0,
      "step": 179
    },
    {
      "entropy": 1.124489665031433,
      "epoch": 0.0962566844919786,
      "grad_norm": 0.1299794316291809,
      "learning_rate": 0.0001979963665456924,
      "loss": 1.1152,
      "mean_token_accuracy": 0.731091171503067,
      "num_tokens": 5868151.0,
      "step": 180
    },
    {
      "entropy": 1.019109919667244,
      "epoch": 0.09679144385026738,
      "grad_norm": 0.1263756901025772,
      "learning_rate": 0.00019796351042171735,
      "loss": 1.0133,
      "mean_token_accuracy": 0.7450751811265945,
      "num_tokens": 5900659.0,
      "step": 181
    },
    {
      "entropy": 1.0870310068130493,
      "epoch": 0.09732620320855614,
      "grad_norm": 0.11404166370630264,
      "learning_rate": 0.00019793039017452344,
      "loss": 1.0762,
      "mean_token_accuracy": 0.7330767512321472,
      "num_tokens": 5933427.0,
      "step": 182
    },
    {
      "entropy": 1.0590984374284744,
      "epoch": 0.09786096256684491,
      "grad_norm": 0.1062210351228714,
      "learning_rate": 0.00019789700590355916,
      "loss": 1.0502,
      "mean_token_accuracy": 0.7384225428104401,
      "num_tokens": 5966195.0,
      "step": 183
    },
    {
      "entropy": 1.062455177307129,
      "epoch": 0.09839572192513368,
      "grad_norm": 0.11313077062368393,
      "learning_rate": 0.00019786335770906552,
      "loss": 1.066,
      "mean_token_accuracy": 0.7376256883144379,
      "num_tokens": 5998575.0,
      "step": 184
    },
    {
      "entropy": 1.1060601472854614,
      "epoch": 0.09893048128342247,
      "grad_norm": 0.11265819519758224,
      "learning_rate": 0.00019782944569207616,
      "loss": 1.1055,
      "mean_token_accuracy": 0.7272727489471436,
      "num_tokens": 6031343.0,
      "step": 185
    },
    {
      "entropy": 1.030540555715561,
      "epoch": 0.09946524064171124,
      "grad_norm": 0.12320833653211594,
      "learning_rate": 0.00019779526995441685,
      "loss": 1.0115,
      "mean_token_accuracy": 0.7461815774440765,
      "num_tokens": 6064111.0,
      "step": 186
    },
    {
      "entropy": 1.0417098999023438,
      "epoch": 0.1,
      "grad_norm": 0.12461020052433014,
      "learning_rate": 0.00019776083059870513,
      "loss": 1.0605,
      "mean_token_accuracy": 0.7379734069108963,
      "num_tokens": 6096712.0,
      "step": 187
    },
    {
      "entropy": 1.046680524945259,
      "epoch": 0.10053475935828877,
      "grad_norm": 0.12361578643321991,
      "learning_rate": 0.0001977261277283503,
      "loss": 1.0588,
      "mean_token_accuracy": 0.7382086962461472,
      "num_tokens": 6129480.0,
      "step": 188
    },
    {
      "entropy": 1.0391662418842316,
      "epoch": 0.10106951871657754,
      "grad_norm": 0.12591002881526947,
      "learning_rate": 0.00019769116144755268,
      "loss": 1.0587,
      "mean_token_accuracy": 0.7403542697429657,
      "num_tokens": 6162164.0,
      "step": 189
    },
    {
      "entropy": 1.092732846736908,
      "epoch": 0.10160427807486631,
      "grad_norm": 0.11028733849525452,
      "learning_rate": 0.0001976559318613036,
      "loss": 1.1019,
      "mean_token_accuracy": 0.7272328734397888,
      "num_tokens": 6194327.0,
      "step": 190
    },
    {
      "entropy": 1.090900957584381,
      "epoch": 0.10213903743315508,
      "grad_norm": 0.10880324244499207,
      "learning_rate": 0.00019762043907538504,
      "loss": 1.079,
      "mean_token_accuracy": 0.7331654280424118,
      "num_tokens": 6227073.0,
      "step": 191
    },
    {
      "entropy": 1.1134484112262726,
      "epoch": 0.10267379679144385,
      "grad_norm": 0.1132679283618927,
      "learning_rate": 0.00019758468319636924,
      "loss": 1.1013,
      "mean_token_accuracy": 0.7296248972415924,
      "num_tokens": 6259841.0,
      "step": 192
    },
    {
      "entropy": 1.0712683200836182,
      "epoch": 0.10320855614973262,
      "grad_norm": 0.1124231144785881,
      "learning_rate": 0.00019754866433161842,
      "loss": 1.0755,
      "mean_token_accuracy": 0.734634667634964,
      "num_tokens": 6292609.0,
      "step": 193
    },
    {
      "entropy": 1.0586844384670258,
      "epoch": 0.10374331550802139,
      "grad_norm": 0.11450853943824768,
      "learning_rate": 0.00019751238258928447,
      "loss": 1.0439,
      "mean_token_accuracy": 0.7394875437021255,
      "num_tokens": 6325114.0,
      "step": 194
    },
    {
      "entropy": 1.0706075578927994,
      "epoch": 0.10427807486631016,
      "grad_norm": 0.12188895791769028,
      "learning_rate": 0.00019747583807830856,
      "loss": 1.0577,
      "mean_token_accuracy": 0.7379983961582184,
      "num_tokens": 6357514.0,
      "step": 195
    },
    {
      "entropy": 1.0366948395967484,
      "epoch": 0.10481283422459893,
      "grad_norm": 0.11499596387147903,
      "learning_rate": 0.00019743903090842094,
      "loss": 1.0249,
      "mean_token_accuracy": 0.7436766922473907,
      "num_tokens": 6390282.0,
      "step": 196
    },
    {
      "entropy": 1.0938681066036224,
      "epoch": 0.1053475935828877,
      "grad_norm": 0.10795161128044128,
      "learning_rate": 0.00019740196119014044,
      "loss": 1.0955,
      "mean_token_accuracy": 0.7291055768728256,
      "num_tokens": 6423050.0,
      "step": 197
    },
    {
      "entropy": 1.0152499079704285,
      "epoch": 0.10588235294117647,
      "grad_norm": 0.12174174934625626,
      "learning_rate": 0.00019736462903477428,
      "loss": 1.0136,
      "mean_token_accuracy": 0.7461740970611572,
      "num_tokens": 6455035.0,
      "step": 198
    },
    {
      "entropy": 1.0703981518745422,
      "epoch": 0.10641711229946524,
      "grad_norm": 0.10990523546934128,
      "learning_rate": 0.0001973270345544177,
      "loss": 1.0787,
      "mean_token_accuracy": 0.7326185256242752,
      "num_tokens": 6487803.0,
      "step": 199
    },
    {
      "entropy": 1.1004179418087006,
      "epoch": 0.10695187165775401,
      "grad_norm": 0.11005926132202148,
      "learning_rate": 0.0001972891778619536,
      "loss": 1.104,
      "mean_token_accuracy": 0.7294110655784607,
      "num_tokens": 6520571.0,
      "step": 200
    },
    {
      "entropy": 1.0091599971055984,
      "epoch": 0.10748663101604278,
      "grad_norm": 0.11055716127157211,
      "learning_rate": 0.00019725105907105215,
      "loss": 1.0109,
      "mean_token_accuracy": 0.7463954091072083,
      "num_tokens": 6553339.0,
      "step": 201
    },
    {
      "entropy": 1.0590536296367645,
      "epoch": 0.10802139037433155,
      "grad_norm": 0.11450552195310593,
      "learning_rate": 0.0001972126782961706,
      "loss": 1.0717,
      "mean_token_accuracy": 0.735278457403183,
      "num_tokens": 6585876.0,
      "step": 202
    },
    {
      "entropy": 1.049636334180832,
      "epoch": 0.10855614973262032,
      "grad_norm": 0.10723718255758286,
      "learning_rate": 0.00019717403565255276,
      "loss": 1.0435,
      "mean_token_accuracy": 0.7375023663043976,
      "num_tokens": 6618534.0,
      "step": 203
    },
    {
      "entropy": 1.0268239080905914,
      "epoch": 0.10909090909090909,
      "grad_norm": 0.110516756772995,
      "learning_rate": 0.00019713513125622882,
      "loss": 1.0285,
      "mean_token_accuracy": 0.7447504550218582,
      "num_tokens": 6651106.0,
      "step": 204
    },
    {
      "entropy": 1.0767436623573303,
      "epoch": 0.10962566844919786,
      "grad_norm": 0.1162244901061058,
      "learning_rate": 0.00019709596522401491,
      "loss": 1.0834,
      "mean_token_accuracy": 0.733895018696785,
      "num_tokens": 6683662.0,
      "step": 205
    },
    {
      "entropy": 1.0388650298118591,
      "epoch": 0.11016042780748662,
      "grad_norm": 0.12721914052963257,
      "learning_rate": 0.00019705653767351265,
      "loss": 1.0488,
      "mean_token_accuracy": 0.7420195192098618,
      "num_tokens": 6716238.0,
      "step": 206
    },
    {
      "entropy": 1.0942329466342926,
      "epoch": 0.11069518716577541,
      "grad_norm": 0.14424873888492584,
      "learning_rate": 0.00019701684872310904,
      "loss": 1.1077,
      "mean_token_accuracy": 0.7278364151716232,
      "num_tokens": 6748830.0,
      "step": 207
    },
    {
      "entropy": 1.0293540060520172,
      "epoch": 0.11122994652406418,
      "grad_norm": 0.14370134472846985,
      "learning_rate": 0.0001969768984919759,
      "loss": 1.037,
      "mean_token_accuracy": 0.7417389005422592,
      "num_tokens": 6781247.0,
      "step": 208
    },
    {
      "entropy": 1.0413859337568283,
      "epoch": 0.11176470588235295,
      "grad_norm": 0.13019073009490967,
      "learning_rate": 0.00019693668710006964,
      "loss": 1.0304,
      "mean_token_accuracy": 0.7385945320129395,
      "num_tokens": 6813848.0,
      "step": 209
    },
    {
      "entropy": 1.0882729589939117,
      "epoch": 0.11229946524064172,
      "grad_norm": 0.11451397836208344,
      "learning_rate": 0.00019689621466813077,
      "loss": 1.0789,
      "mean_token_accuracy": 0.7339456081390381,
      "num_tokens": 6846614.0,
      "step": 210
    },
    {
      "entropy": 1.0684372186660767,
      "epoch": 0.11283422459893049,
      "grad_norm": 0.11447669565677643,
      "learning_rate": 0.00019685548131768364,
      "loss": 1.0549,
      "mean_token_accuracy": 0.7408982962369919,
      "num_tokens": 6878989.0,
      "step": 211
    },
    {
      "entropy": 1.0850865244865417,
      "epoch": 0.11336898395721925,
      "grad_norm": 0.13719269633293152,
      "learning_rate": 0.00019681448717103609,
      "loss": 1.0645,
      "mean_token_accuracy": 0.7355644553899765,
      "num_tokens": 6911286.0,
      "step": 212
    },
    {
      "entropy": 1.0667450577020645,
      "epoch": 0.11390374331550802,
      "grad_norm": 0.14587928354740143,
      "learning_rate": 0.00019677323235127895,
      "loss": 1.0689,
      "mean_token_accuracy": 0.7330211699008942,
      "num_tokens": 6943962.0,
      "step": 213
    },
    {
      "entropy": 1.0296750366687775,
      "epoch": 0.1144385026737968,
      "grad_norm": 0.1201789379119873,
      "learning_rate": 0.00019673171698228584,
      "loss": 1.0215,
      "mean_token_accuracy": 0.7414021492004395,
      "num_tokens": 6976661.0,
      "step": 214
    },
    {
      "entropy": 1.042615681886673,
      "epoch": 0.11497326203208556,
      "grad_norm": 0.10887690633535385,
      "learning_rate": 0.0001966899411887127,
      "loss": 1.0306,
      "mean_token_accuracy": 0.7439039051532745,
      "num_tokens": 7009306.0,
      "step": 215
    },
    {
      "entropy": 1.0513212084770203,
      "epoch": 0.11550802139037433,
      "grad_norm": 0.127265065908432,
      "learning_rate": 0.00019664790509599738,
      "loss": 1.052,
      "mean_token_accuracy": 0.7402859330177307,
      "num_tokens": 7042074.0,
      "step": 216
    },
    {
      "entropy": 1.0337154269218445,
      "epoch": 0.1160427807486631,
      "grad_norm": 0.11888471245765686,
      "learning_rate": 0.0001966056088303594,
      "loss": 1.0388,
      "mean_token_accuracy": 0.7428157329559326,
      "num_tokens": 7074612.0,
      "step": 217
    },
    {
      "entropy": 1.02349191904068,
      "epoch": 0.11657754010695187,
      "grad_norm": 0.11285922676324844,
      "learning_rate": 0.0001965630525187994,
      "loss": 1.0401,
      "mean_token_accuracy": 0.7402248382568359,
      "num_tokens": 7107380.0,
      "step": 218
    },
    {
      "entropy": 1.0236869156360626,
      "epoch": 0.11711229946524064,
      "grad_norm": 0.12390273064374924,
      "learning_rate": 0.00019652023628909893,
      "loss": 1.0392,
      "mean_token_accuracy": 0.735905334353447,
      "num_tokens": 7139747.0,
      "step": 219
    },
    {
      "entropy": 1.031931921839714,
      "epoch": 0.11764705882352941,
      "grad_norm": 0.11192239075899124,
      "learning_rate": 0.00019647716026981987,
      "loss": 1.0439,
      "mean_token_accuracy": 0.7395527958869934,
      "num_tokens": 7172515.0,
      "step": 220
    },
    {
      "entropy": 1.0469720661640167,
      "epoch": 0.11818181818181818,
      "grad_norm": 0.11783554404973984,
      "learning_rate": 0.00019643382459030433,
      "loss": 1.0647,
      "mean_token_accuracy": 0.7378115952014923,
      "num_tokens": 7205283.0,
      "step": 221
    },
    {
      "entropy": 1.0731564462184906,
      "epoch": 0.11871657754010695,
      "grad_norm": 0.12124251574277878,
      "learning_rate": 0.0001963902293806739,
      "loss": 1.0743,
      "mean_token_accuracy": 0.734205812215805,
      "num_tokens": 7237546.0,
      "step": 222
    },
    {
      "entropy": 1.07041397690773,
      "epoch": 0.11925133689839572,
      "grad_norm": 0.114452064037323,
      "learning_rate": 0.00019634637477182953,
      "loss": 1.0555,
      "mean_token_accuracy": 0.7356832921504974,
      "num_tokens": 7270131.0,
      "step": 223
    },
    {
      "entropy": 1.0573597848415375,
      "epoch": 0.11978609625668449,
      "grad_norm": 0.11932668089866638,
      "learning_rate": 0.0001963022608954511,
      "loss": 1.0612,
      "mean_token_accuracy": 0.7345124781131744,
      "num_tokens": 7302899.0,
      "step": 224
    },
    {
      "entropy": 1.0339885354042053,
      "epoch": 0.12032085561497326,
      "grad_norm": 0.11648670583963394,
      "learning_rate": 0.00019625788788399696,
      "loss": 1.0235,
      "mean_token_accuracy": 0.7396539747714996,
      "num_tokens": 7335362.0,
      "step": 225
    },
    {
      "entropy": 1.0512037575244904,
      "epoch": 0.12085561497326203,
      "grad_norm": 0.10895999521017075,
      "learning_rate": 0.00019621325587070344,
      "loss": 1.0434,
      "mean_token_accuracy": 0.7386669218540192,
      "num_tokens": 7368130.0,
      "step": 226
    },
    {
      "entropy": 1.0894387364387512,
      "epoch": 0.1213903743315508,
      "grad_norm": 0.11288847029209137,
      "learning_rate": 0.00019616836498958465,
      "loss": 1.0788,
      "mean_token_accuracy": 0.7326047420501709,
      "num_tokens": 7400605.0,
      "step": 227
    },
    {
      "entropy": 1.0398660898208618,
      "epoch": 0.12192513368983957,
      "grad_norm": 0.11348799616098404,
      "learning_rate": 0.00019612321537543204,
      "loss": 1.0378,
      "mean_token_accuracy": 0.7441670298576355,
      "num_tokens": 7433152.0,
      "step": 228
    },
    {
      "entropy": 1.0839430689811707,
      "epoch": 0.12245989304812835,
      "grad_norm": 0.11604870110750198,
      "learning_rate": 0.00019607780716381388,
      "loss": 1.0873,
      "mean_token_accuracy": 0.7298715263605118,
      "num_tokens": 7465724.0,
      "step": 229
    },
    {
      "entropy": 1.0554787516593933,
      "epoch": 0.12299465240641712,
      "grad_norm": 0.11123524606227875,
      "learning_rate": 0.00019603214049107487,
      "loss": 1.0564,
      "mean_token_accuracy": 0.7370816320180893,
      "num_tokens": 7498434.0,
      "step": 230
    },
    {
      "entropy": 1.0759001970291138,
      "epoch": 0.12352941176470589,
      "grad_norm": 0.11666589230298996,
      "learning_rate": 0.00019598621549433585,
      "loss": 1.0862,
      "mean_token_accuracy": 0.7339395433664322,
      "num_tokens": 7531071.0,
      "step": 231
    },
    {
      "entropy": 1.061589390039444,
      "epoch": 0.12406417112299466,
      "grad_norm": 0.11102268844842911,
      "learning_rate": 0.00019594003231149332,
      "loss": 1.0663,
      "mean_token_accuracy": 0.7356441020965576,
      "num_tokens": 7563290.0,
      "step": 232
    },
    {
      "entropy": 1.0681027621030807,
      "epoch": 0.12459893048128343,
      "grad_norm": 0.11785712093114853,
      "learning_rate": 0.00019589359108121895,
      "loss": 1.0725,
      "mean_token_accuracy": 0.7335348129272461,
      "num_tokens": 7595961.0,
      "step": 233
    },
    {
      "entropy": 1.0394710004329681,
      "epoch": 0.12513368983957218,
      "grad_norm": 0.11143148690462112,
      "learning_rate": 0.00019584689194295936,
      "loss": 1.0345,
      "mean_token_accuracy": 0.7412939965724945,
      "num_tokens": 7628729.0,
      "step": 234
    },
    {
      "entropy": 1.0309483706951141,
      "epoch": 0.12566844919786097,
      "grad_norm": 0.1119936853647232,
      "learning_rate": 0.00019579993503693538,
      "loss": 1.0204,
      "mean_token_accuracy": 0.7455660849809647,
      "num_tokens": 7661131.0,
      "step": 235
    },
    {
      "entropy": 1.0677326619625092,
      "epoch": 0.12620320855614972,
      "grad_norm": 0.11484680324792862,
      "learning_rate": 0.00019575272050414207,
      "loss": 1.0735,
      "mean_token_accuracy": 0.7364104688167572,
      "num_tokens": 7693705.0,
      "step": 236
    },
    {
      "entropy": 1.0587773025035858,
      "epoch": 0.1267379679144385,
      "grad_norm": 0.11827201396226883,
      "learning_rate": 0.00019570524848634783,
      "loss": 1.0684,
      "mean_token_accuracy": 0.7397666275501251,
      "num_tokens": 7726473.0,
      "step": 237
    },
    {
      "entropy": 1.0162574499845505,
      "epoch": 0.12727272727272726,
      "grad_norm": 0.11548800766468048,
      "learning_rate": 0.0001956575191260943,
      "loss": 1.0103,
      "mean_token_accuracy": 0.7485337257385254,
      "num_tokens": 7759241.0,
      "step": 238
    },
    {
      "entropy": 1.032597839832306,
      "epoch": 0.12780748663101604,
      "grad_norm": 0.11080261319875717,
      "learning_rate": 0.00019560953256669587,
      "loss": 1.0355,
      "mean_token_accuracy": 0.738309696316719,
      "num_tokens": 7791975.0,
      "step": 239
    },
    {
      "entropy": 1.0555288195610046,
      "epoch": 0.12834224598930483,
      "grad_norm": 0.11107134819030762,
      "learning_rate": 0.00019556128895223914,
      "loss": 1.0604,
      "mean_token_accuracy": 0.7379528433084488,
      "num_tokens": 7824469.0,
      "step": 240
    },
    {
      "entropy": 1.1027131378650665,
      "epoch": 0.12887700534759358,
      "grad_norm": 0.11281178146600723,
      "learning_rate": 0.0001955127884275825,
      "loss": 1.1042,
      "mean_token_accuracy": 0.728361964225769,
      "num_tokens": 7856992.0,
      "step": 241
    },
    {
      "entropy": 1.067112773656845,
      "epoch": 0.12941176470588237,
      "grad_norm": 0.10883133858442307,
      "learning_rate": 0.00019546403113835593,
      "loss": 1.0644,
      "mean_token_accuracy": 0.7364980578422546,
      "num_tokens": 7889760.0,
      "step": 242
    },
    {
      "entropy": 1.014612227678299,
      "epoch": 0.12994652406417112,
      "grad_norm": 0.11845094710588455,
      "learning_rate": 0.00019541501723096017,
      "loss": 1.0173,
      "mean_token_accuracy": 0.7426800578832626,
      "num_tokens": 7922118.0,
      "step": 243
    },
    {
      "entropy": 1.0526921898126602,
      "epoch": 0.1304812834224599,
      "grad_norm": 0.1122470572590828,
      "learning_rate": 0.0001953657468525667,
      "loss": 1.0454,
      "mean_token_accuracy": 0.7415078282356262,
      "num_tokens": 7954886.0,
      "step": 244
    },
    {
      "entropy": 1.066915899515152,
      "epoch": 0.13101604278074866,
      "grad_norm": 0.11756544560194016,
      "learning_rate": 0.0001953162201511169,
      "loss": 1.0708,
      "mean_token_accuracy": 0.7362254858016968,
      "num_tokens": 7987371.0,
      "step": 245
    },
    {
      "entropy": 1.0463462471961975,
      "epoch": 0.13155080213903744,
      "grad_norm": 0.11988546699285507,
      "learning_rate": 0.00019526643727532197,
      "loss": 1.0468,
      "mean_token_accuracy": 0.7412023544311523,
      "num_tokens": 8020139.0,
      "step": 246
    },
    {
      "entropy": 1.0550406724214554,
      "epoch": 0.1320855614973262,
      "grad_norm": 0.13351471722126007,
      "learning_rate": 0.00019521639837466213,
      "loss": 1.0703,
      "mean_token_accuracy": 0.7336876839399338,
      "num_tokens": 8052907.0,
      "step": 247
    },
    {
      "entropy": 1.042212188243866,
      "epoch": 0.13262032085561498,
      "grad_norm": 0.125226229429245,
      "learning_rate": 0.00019516610359938648,
      "loss": 1.0332,
      "mean_token_accuracy": 0.7394306063652039,
      "num_tokens": 8085675.0,
      "step": 248
    },
    {
      "entropy": 1.0784787833690643,
      "epoch": 0.13315508021390374,
      "grad_norm": 0.11965163797140121,
      "learning_rate": 0.00019511555310051246,
      "loss": 1.0729,
      "mean_token_accuracy": 0.7330728024244308,
      "num_tokens": 8118247.0,
      "step": 249
    },
    {
      "entropy": 1.0723508298397064,
      "epoch": 0.13368983957219252,
      "grad_norm": 0.12876088917255402,
      "learning_rate": 0.00019506474702982523,
      "loss": 1.0664,
      "mean_token_accuracy": 0.7342070043087006,
      "num_tokens": 8151015.0,
      "step": 250
    },
    {
      "entropy": 1.0198566615581512,
      "epoch": 0.13422459893048128,
      "grad_norm": 0.12335485965013504,
      "learning_rate": 0.00019501368553987737,
      "loss": 1.0205,
      "mean_token_accuracy": 0.7444403767585754,
      "num_tokens": 8183783.0,
      "step": 251
    },
    {
      "entropy": 1.0937196016311646,
      "epoch": 0.13475935828877006,
      "grad_norm": 0.11913212388753891,
      "learning_rate": 0.00019496236878398848,
      "loss": 1.0806,
      "mean_token_accuracy": 0.7335934787988663,
      "num_tokens": 8216273.0,
      "step": 252
    },
    {
      "entropy": 1.0336759388446808,
      "epoch": 0.13529411764705881,
      "grad_norm": 0.11281498521566391,
      "learning_rate": 0.00019491079691624454,
      "loss": 1.0276,
      "mean_token_accuracy": 0.7413501590490341,
      "num_tokens": 8248701.0,
      "step": 253
    },
    {
      "entropy": 0.9990439265966415,
      "epoch": 0.1358288770053476,
      "grad_norm": 0.11167177557945251,
      "learning_rate": 0.0001948589700914976,
      "loss": 1.0042,
      "mean_token_accuracy": 0.7490535080432892,
      "num_tokens": 8280796.0,
      "step": 254
    },
    {
      "entropy": 1.0568778216838837,
      "epoch": 0.13636363636363635,
      "grad_norm": 0.11134691536426544,
      "learning_rate": 0.0001948068884653652,
      "loss": 1.0593,
      "mean_token_accuracy": 0.7315690666437149,
      "num_tokens": 8313379.0,
      "step": 255
    },
    {
      "entropy": 1.03424671292305,
      "epoch": 0.13689839572192514,
      "grad_norm": 0.11792710423469543,
      "learning_rate": 0.00019475455219422998,
      "loss": 1.0441,
      "mean_token_accuracy": 0.740011528134346,
      "num_tokens": 8345856.0,
      "step": 256
    },
    {
      "entropy": 1.0480815172195435,
      "epoch": 0.1374331550802139,
      "grad_norm": 0.12474671006202698,
      "learning_rate": 0.00019470196143523923,
      "loss": 1.047,
      "mean_token_accuracy": 0.7390372902154922,
      "num_tokens": 8378372.0,
      "step": 257
    },
    {
      "entropy": 1.0477408915758133,
      "epoch": 0.13796791443850268,
      "grad_norm": 0.1266440451145172,
      "learning_rate": 0.00019464911634630433,
      "loss": 1.043,
      "mean_token_accuracy": 0.7413789927959442,
      "num_tokens": 8410900.0,
      "step": 258
    },
    {
      "entropy": 1.0490031838417053,
      "epoch": 0.13850267379679143,
      "grad_norm": 0.14051200449466705,
      "learning_rate": 0.00019459601708610036,
      "loss": 1.0402,
      "mean_token_accuracy": 0.7361215502023697,
      "num_tokens": 8443641.0,
      "step": 259
    },
    {
      "entropy": 1.0794505178928375,
      "epoch": 0.13903743315508021,
      "grad_norm": 0.158530592918396,
      "learning_rate": 0.0001945426638140655,
      "loss": 1.0739,
      "mean_token_accuracy": 0.7352101802825928,
      "num_tokens": 8476266.0,
      "step": 260
    },
    {
      "entropy": 1.0522571057081223,
      "epoch": 0.139572192513369,
      "grad_norm": 0.1495906412601471,
      "learning_rate": 0.00019448905669040077,
      "loss": 1.0482,
      "mean_token_accuracy": 0.7383309006690979,
      "num_tokens": 8509034.0,
      "step": 261
    },
    {
      "entropy": 1.0746977925300598,
      "epoch": 0.14010695187165775,
      "grad_norm": 0.12712834775447845,
      "learning_rate": 0.00019443519587606934,
      "loss": 1.083,
      "mean_token_accuracy": 0.7317632138729095,
      "num_tokens": 8541802.0,
      "step": 262
    },
    {
      "entropy": 1.03669573366642,
      "epoch": 0.14064171122994654,
      "grad_norm": 0.11646346002817154,
      "learning_rate": 0.0001943810815327961,
      "loss": 1.043,
      "mean_token_accuracy": 0.740713581442833,
      "num_tokens": 8574570.0,
      "step": 263
    },
    {
      "entropy": 1.0398886501789093,
      "epoch": 0.1411764705882353,
      "grad_norm": 0.13226716220378876,
      "learning_rate": 0.0001943267138230672,
      "loss": 1.0401,
      "mean_token_accuracy": 0.7394306063652039,
      "num_tokens": 8607338.0,
      "step": 264
    },
    {
      "entropy": 1.0552105903625488,
      "epoch": 0.14171122994652408,
      "grad_norm": 0.14164507389068604,
      "learning_rate": 0.00019427209291012965,
      "loss": 1.0592,
      "mean_token_accuracy": 0.7373839169740677,
      "num_tokens": 8640106.0,
      "step": 265
    },
    {
      "entropy": 1.0696653425693512,
      "epoch": 0.14224598930481283,
      "grad_norm": 0.11818902939558029,
      "learning_rate": 0.00019421721895799063,
      "loss": 1.067,
      "mean_token_accuracy": 0.7361311912536621,
      "num_tokens": 8672636.0,
      "step": 266
    },
    {
      "entropy": 1.0401161164045334,
      "epoch": 0.14278074866310161,
      "grad_norm": 0.1131942868232727,
      "learning_rate": 0.0001941620921314172,
      "loss": 1.0391,
      "mean_token_accuracy": 0.7399394363164902,
      "num_tokens": 8705347.0,
      "step": 267
    },
    {
      "entropy": 1.0111385881900787,
      "epoch": 0.14331550802139037,
      "grad_norm": 0.13214008510112762,
      "learning_rate": 0.00019410671259593562,
      "loss": 1.0127,
      "mean_token_accuracy": 0.7480546683073044,
      "num_tokens": 8737150.0,
      "step": 268
    },
    {
      "entropy": 1.02693110704422,
      "epoch": 0.14385026737967915,
      "grad_norm": 0.11571832001209259,
      "learning_rate": 0.000194051080517831,
      "loss": 1.0207,
      "mean_token_accuracy": 0.7417827546596527,
      "num_tokens": 8769918.0,
      "step": 269
    },
    {
      "entropy": 1.0444089770317078,
      "epoch": 0.1443850267379679,
      "grad_norm": 0.11474455893039703,
      "learning_rate": 0.00019399519606414674,
      "loss": 1.0607,
      "mean_token_accuracy": 0.7389113008975983,
      "num_tokens": 8802686.0,
      "step": 270
    },
    {
      "entropy": 1.0395529866218567,
      "epoch": 0.1449197860962567,
      "grad_norm": 0.12115021049976349,
      "learning_rate": 0.00019393905940268404,
      "loss": 1.0433,
      "mean_token_accuracy": 0.735184520483017,
      "num_tokens": 8835454.0,
      "step": 271
    },
    {
      "entropy": 1.0157050043344498,
      "epoch": 0.14545454545454545,
      "grad_norm": 0.11376800388097763,
      "learning_rate": 0.00019388267070200135,
      "loss": 1.0048,
      "mean_token_accuracy": 0.7480817437171936,
      "num_tokens": 8867817.0,
      "step": 272
    },
    {
      "entropy": 1.0406829416751862,
      "epoch": 0.14598930481283423,
      "grad_norm": 0.12187753617763519,
      "learning_rate": 0.00019382603013141402,
      "loss": 1.0516,
      "mean_token_accuracy": 0.7399662286043167,
      "num_tokens": 8900363.0,
      "step": 273
    },
    {
      "entropy": 1.0107262432575226,
      "epoch": 0.146524064171123,
      "grad_norm": 0.11920078098773956,
      "learning_rate": 0.00019376913786099348,
      "loss": 1.0196,
      "mean_token_accuracy": 0.7423813045024872,
      "num_tokens": 8932749.0,
      "step": 274
    },
    {
      "entropy": 1.0222293138504028,
      "epoch": 0.14705882352941177,
      "grad_norm": 0.11740428954362869,
      "learning_rate": 0.0001937119940615671,
      "loss": 1.0285,
      "mean_token_accuracy": 0.7422104030847549,
      "num_tokens": 8965517.0,
      "step": 275
    },
    {
      "entropy": 1.0424260348081589,
      "epoch": 0.14759358288770053,
      "grad_norm": 0.12438567727804184,
      "learning_rate": 0.00019365459890471746,
      "loss": 1.0362,
      "mean_token_accuracy": 0.7398019582033157,
      "num_tokens": 8997876.0,
      "step": 276
    },
    {
      "entropy": 1.0288828313350677,
      "epoch": 0.1481283422459893,
      "grad_norm": 0.11123205721378326,
      "learning_rate": 0.00019359695256278178,
      "loss": 1.0216,
      "mean_token_accuracy": 0.7422104179859161,
      "num_tokens": 9030644.0,
      "step": 277
    },
    {
      "entropy": 1.020155742764473,
      "epoch": 0.14866310160427806,
      "grad_norm": 0.11991836130619049,
      "learning_rate": 0.00019353905520885166,
      "loss": 1.0231,
      "mean_token_accuracy": 0.7435545027256012,
      "num_tokens": 9063412.0,
      "step": 278
    },
    {
      "entropy": 1.0389754474163055,
      "epoch": 0.14919786096256685,
      "grad_norm": 0.1264876425266266,
      "learning_rate": 0.00019348090701677223,
      "loss": 1.0405,
      "mean_token_accuracy": 0.7415297776460648,
      "num_tokens": 9096021.0,
      "step": 279
    },
    {
      "entropy": 0.9926124960184097,
      "epoch": 0.1497326203208556,
      "grad_norm": 0.11023156344890594,
      "learning_rate": 0.00019342250816114197,
      "loss": 0.9886,
      "mean_token_accuracy": 0.7518328428268433,
      "num_tokens": 9128789.0,
      "step": 280
    },
    {
      "entropy": 0.9978072941303253,
      "epoch": 0.15026737967914439,
      "grad_norm": 0.12432621419429779,
      "learning_rate": 0.00019336385881731186,
      "loss": 0.9991,
      "mean_token_accuracy": 0.7462638318538666,
      "num_tokens": 9161515.0,
      "step": 281
    },
    {
      "entropy": 1.0047390162944794,
      "epoch": 0.15080213903743314,
      "grad_norm": 0.12100157886743546,
      "learning_rate": 0.00019330495916138513,
      "loss": 0.9988,
      "mean_token_accuracy": 0.7465151995420456,
      "num_tokens": 9193969.0,
      "step": 282
    },
    {
      "entropy": 1.051861196756363,
      "epoch": 0.15133689839572192,
      "grad_norm": 0.11842017620801926,
      "learning_rate": 0.00019324580937021653,
      "loss": 1.0475,
      "mean_token_accuracy": 0.7398037314414978,
      "num_tokens": 9226307.0,
      "step": 283
    },
    {
      "entropy": 1.1205312013626099,
      "epoch": 0.1518716577540107,
      "grad_norm": 0.12409147620201111,
      "learning_rate": 0.00019318640962141188,
      "loss": 1.1388,
      "mean_token_accuracy": 0.7234237641096115,
      "num_tokens": 9259075.0,
      "step": 284
    },
    {
      "entropy": 1.0474530458450317,
      "epoch": 0.15240641711229946,
      "grad_norm": 0.12057706713676453,
      "learning_rate": 0.0001931267600933276,
      "loss": 1.041,
      "mean_token_accuracy": 0.7406096160411835,
      "num_tokens": 9291580.0,
      "step": 285
    },
    {
      "entropy": 1.071075975894928,
      "epoch": 0.15294117647058825,
      "grad_norm": 0.11215157806873322,
      "learning_rate": 0.00019306686096506997,
      "loss": 1.0546,
      "mean_token_accuracy": 0.7369257062673569,
      "num_tokens": 9324348.0,
      "step": 286
    },
    {
      "entropy": 1.0828769505023956,
      "epoch": 0.153475935828877,
      "grad_norm": 0.12447725981473923,
      "learning_rate": 0.00019300671241649495,
      "loss": 1.0681,
      "mean_token_accuracy": 0.7351312190294266,
      "num_tokens": 9356912.0,
      "step": 287
    },
    {
      "entropy": 1.0940111577510834,
      "epoch": 0.15401069518716579,
      "grad_norm": 0.118400439620018,
      "learning_rate": 0.00019294631462820721,
      "loss": 1.0894,
      "mean_token_accuracy": 0.7328831255435944,
      "num_tokens": 9389393.0,
      "step": 288
    },
    {
      "entropy": 1.068321093916893,
      "epoch": 0.15454545454545454,
      "grad_norm": 0.11101941764354706,
      "learning_rate": 0.00019288566778155993,
      "loss": 1.0553,
      "mean_token_accuracy": 0.7367358803749084,
      "num_tokens": 9422119.0,
      "step": 289
    },
    {
      "entropy": 1.0740725696086884,
      "epoch": 0.15508021390374332,
      "grad_norm": 0.1184263750910759,
      "learning_rate": 0.00019282477205865406,
      "loss": 1.0899,
      "mean_token_accuracy": 0.7305536270141602,
      "num_tokens": 9454682.0,
      "step": 290
    },
    {
      "entropy": 1.0920554399490356,
      "epoch": 0.15561497326203208,
      "grad_norm": 0.12500989437103271,
      "learning_rate": 0.00019276362764233783,
      "loss": 1.0918,
      "mean_token_accuracy": 0.7287611365318298,
      "num_tokens": 9487286.0,
      "step": 291
    },
    {
      "entropy": 0.9881408363580704,
      "epoch": 0.15614973262032086,
      "grad_norm": 0.12167588621377945,
      "learning_rate": 0.00019270223471620624,
      "loss": 1.0068,
      "mean_token_accuracy": 0.7483971863985062,
      "num_tokens": 9520021.0,
      "step": 292
    },
    {
      "entropy": 1.069222331047058,
      "epoch": 0.15668449197860962,
      "grad_norm": 0.11795824766159058,
      "learning_rate": 0.0001926405934646005,
      "loss": 1.0925,
      "mean_token_accuracy": 0.7297313064336777,
      "num_tokens": 9552590.0,
      "step": 293
    },
    {
      "entropy": 1.0657239258289337,
      "epoch": 0.1572192513368984,
      "grad_norm": 0.12100639939308167,
      "learning_rate": 0.00019257870407260742,
      "loss": 1.053,
      "mean_token_accuracy": 0.7422927170991898,
      "num_tokens": 9585309.0,
      "step": 294
    },
    {
      "entropy": 1.0440329909324646,
      "epoch": 0.15775401069518716,
      "grad_norm": 0.11367267370223999,
      "learning_rate": 0.00019251656672605887,
      "loss": 1.0335,
      "mean_token_accuracy": 0.7432643920183182,
      "num_tokens": 9617995.0,
      "step": 295
    },
    {
      "entropy": 1.0245574712753296,
      "epoch": 0.15828877005347594,
      "grad_norm": 0.11685089766979218,
      "learning_rate": 0.00019245418161153126,
      "loss": 1.0271,
      "mean_token_accuracy": 0.7453873455524445,
      "num_tokens": 9650763.0,
      "step": 296
    },
    {
      "entropy": 1.0340452790260315,
      "epoch": 0.1588235294117647,
      "grad_norm": 0.11204793304204941,
      "learning_rate": 0.00019239154891634498,
      "loss": 1.0265,
      "mean_token_accuracy": 0.7396213859319687,
      "num_tokens": 9683363.0,
      "step": 297
    },
    {
      "entropy": 0.9923126399517059,
      "epoch": 0.15935828877005348,
      "grad_norm": 0.1120043396949768,
      "learning_rate": 0.00019232866882856376,
      "loss": 0.9909,
      "mean_token_accuracy": 0.7515579164028168,
      "num_tokens": 9716131.0,
      "step": 298
    },
    {
      "entropy": 1.0405441522598267,
      "epoch": 0.15989304812834224,
      "grad_norm": 0.11795374006032944,
      "learning_rate": 0.00019226554153699423,
      "loss": 1.0353,
      "mean_token_accuracy": 0.7428170144557953,
      "num_tokens": 9748696.0,
      "step": 299
    },
    {
      "entropy": 1.0656865388154984,
      "epoch": 0.16042780748663102,
      "grad_norm": 0.11844547092914581,
      "learning_rate": 0.00019220216723118527,
      "loss": 1.063,
      "mean_token_accuracy": 0.7360092848539352,
      "num_tokens": 9781464.0,
      "step": 300
    },
    {
      "entropy": 1.0744484663009644,
      "epoch": 0.16096256684491977,
      "grad_norm": 0.1192609965801239,
      "learning_rate": 0.00019213854610142738,
      "loss": 1.0867,
      "mean_token_accuracy": 0.7393917888402939,
      "num_tokens": 9814042.0,
      "step": 301
    },
    {
      "entropy": 0.9988169819116592,
      "epoch": 0.16149732620320856,
      "grad_norm": 0.1213647648692131,
      "learning_rate": 0.0001920746783387523,
      "loss": 1.0146,
      "mean_token_accuracy": 0.7446236610412598,
      "num_tokens": 9846810.0,
      "step": 302
    },
    {
      "entropy": 0.9821544885635376,
      "epoch": 0.1620320855614973,
      "grad_norm": 0.1157236248254776,
      "learning_rate": 0.00019201056413493224,
      "loss": 0.9847,
      "mean_token_accuracy": 0.7526452541351318,
      "num_tokens": 9879232.0,
      "step": 303
    },
    {
      "entropy": 1.0186610370874405,
      "epoch": 0.1625668449197861,
      "grad_norm": 0.1273307055234909,
      "learning_rate": 0.00019194620368247938,
      "loss": 1.0323,
      "mean_token_accuracy": 0.7344208359718323,
      "num_tokens": 9912000.0,
      "step": 304
    },
    {
      "entropy": 0.9831012636423111,
      "epoch": 0.16310160427807488,
      "grad_norm": 0.1286727786064148,
      "learning_rate": 0.0001918815971746454,
      "loss": 0.9839,
      "mean_token_accuracy": 0.7520772218704224,
      "num_tokens": 9944768.0,
      "step": 305
    },
    {
      "entropy": 1.0608522593975067,
      "epoch": 0.16363636363636364,
      "grad_norm": 0.12535078823566437,
      "learning_rate": 0.0001918167448054207,
      "loss": 1.0415,
      "mean_token_accuracy": 0.7414893060922623,
      "num_tokens": 9977158.0,
      "step": 306
    },
    {
      "entropy": 1.082308292388916,
      "epoch": 0.16417112299465242,
      "grad_norm": 0.13904598355293274,
      "learning_rate": 0.00019175164676953389,
      "loss": 1.0945,
      "mean_token_accuracy": 0.7317221611738205,
      "num_tokens": 10009789.0,
      "step": 307
    },
    {
      "entropy": 1.021088719367981,
      "epoch": 0.16470588235294117,
      "grad_norm": 0.15948276221752167,
      "learning_rate": 0.00019168630326245133,
      "loss": 1.0002,
      "mean_token_accuracy": 0.7490701675415039,
      "num_tokens": 10042367.0,
      "step": 308
    },
    {
      "entropy": 1.0862839818000793,
      "epoch": 0.16524064171122996,
      "grad_norm": 0.16437754034996033,
      "learning_rate": 0.0001916207144803764,
      "loss": 1.0709,
      "mean_token_accuracy": 0.7315387725830078,
      "num_tokens": 10074834.0,
      "step": 309
    },
    {
      "entropy": 1.0433639287948608,
      "epoch": 0.1657754010695187,
      "grad_norm": 0.14935532212257385,
      "learning_rate": 0.0001915548806202489,
      "loss": 1.0226,
      "mean_token_accuracy": 0.744517520070076,
      "num_tokens": 10107112.0,
      "step": 310
    },
    {
      "entropy": 1.0890674889087677,
      "epoch": 0.1663101604278075,
      "grad_norm": 0.11999236047267914,
      "learning_rate": 0.00019148880187974462,
      "loss": 1.0863,
      "mean_token_accuracy": 0.7288990914821625,
      "num_tokens": 10139343.0,
      "step": 311
    },
    {
      "entropy": 1.0622902810573578,
      "epoch": 0.16684491978609625,
      "grad_norm": 0.11667732894420624,
      "learning_rate": 0.00019142247845727452,
      "loss": 1.0516,
      "mean_token_accuracy": 0.7366507947444916,
      "num_tokens": 10172111.0,
      "step": 312
    },
    {
      "entropy": 1.0628046095371246,
      "epoch": 0.16737967914438504,
      "grad_norm": 0.12990324199199677,
      "learning_rate": 0.00019135591055198433,
      "loss": 1.0667,
      "mean_token_accuracy": 0.7355510890483856,
      "num_tokens": 10204879.0,
      "step": 313
    },
    {
      "entropy": 1.0701580345630646,
      "epoch": 0.1679144385026738,
      "grad_norm": 0.12336275726556778,
      "learning_rate": 0.00019128909836375391,
      "loss": 1.0886,
      "mean_token_accuracy": 0.7303274720907211,
      "num_tokens": 10237647.0,
      "step": 314
    },
    {
      "entropy": 1.0163717865943909,
      "epoch": 0.16844919786096257,
      "grad_norm": 0.12419279664754868,
      "learning_rate": 0.00019122204209319648,
      "loss": 1.0398,
      "mean_token_accuracy": 0.7439745962619781,
      "num_tokens": 10270290.0,
      "step": 315
    },
    {
      "entropy": 0.9949931651353836,
      "epoch": 0.16898395721925133,
      "grad_norm": 0.12302132695913315,
      "learning_rate": 0.00019115474194165822,
      "loss": 1.002,
      "mean_token_accuracy": 0.7492363154888153,
      "num_tokens": 10303058.0,
      "step": 316
    },
    {
      "entropy": 1.0401550829410553,
      "epoch": 0.1695187165775401,
      "grad_norm": 0.13105161488056183,
      "learning_rate": 0.00019108719811121772,
      "loss": 1.0611,
      "mean_token_accuracy": 0.7385796755552292,
      "num_tokens": 10335660.0,
      "step": 317
    },
    {
      "entropy": 1.046198695898056,
      "epoch": 0.17005347593582887,
      "grad_norm": 0.11633765697479248,
      "learning_rate": 0.0001910194108046851,
      "loss": 1.0372,
      "mean_token_accuracy": 0.7395435571670532,
      "num_tokens": 10368347.0,
      "step": 318
    },
    {
      "entropy": 1.1130651533603668,
      "epoch": 0.17058823529411765,
      "grad_norm": 0.12539620697498322,
      "learning_rate": 0.00019095138022560154,
      "loss": 1.1134,
      "mean_token_accuracy": 0.724764883518219,
      "num_tokens": 10401040.0,
      "step": 319
    },
    {
      "entropy": 1.0333769619464874,
      "epoch": 0.1711229946524064,
      "grad_norm": 0.12401431053876877,
      "learning_rate": 0.00019088310657823876,
      "loss": 1.0277,
      "mean_token_accuracy": 0.7436408698558807,
      "num_tokens": 10433536.0,
      "step": 320
    },
    {
      "entropy": 1.0465964674949646,
      "epoch": 0.1716577540106952,
      "grad_norm": 0.12771785259246826,
      "learning_rate": 0.00019081459006759833,
      "loss": 1.0341,
      "mean_token_accuracy": 0.7408052384853363,
      "num_tokens": 10466304.0,
      "step": 321
    },
    {
      "entropy": 1.0823104679584503,
      "epoch": 0.17219251336898395,
      "grad_norm": 0.1222015768289566,
      "learning_rate": 0.000190745830899411,
      "loss": 1.0656,
      "mean_token_accuracy": 0.7383493930101395,
      "num_tokens": 10498790.0,
      "step": 322
    },
    {
      "entropy": 1.1021881699562073,
      "epoch": 0.17272727272727273,
      "grad_norm": 0.1238536611199379,
      "learning_rate": 0.0001906768292801362,
      "loss": 1.0898,
      "mean_token_accuracy": 0.7280010730028152,
      "num_tokens": 10531248.0,
      "step": 323
    },
    {
      "entropy": 1.0325301140546799,
      "epoch": 0.17326203208556148,
      "grad_norm": 0.115191251039505,
      "learning_rate": 0.0001906075854169613,
      "loss": 1.0249,
      "mean_token_accuracy": 0.7412202507257462,
      "num_tokens": 10563888.0,
      "step": 324
    },
    {
      "entropy": 1.032944694161415,
      "epoch": 0.17379679144385027,
      "grad_norm": 0.12355396151542664,
      "learning_rate": 0.00019053809951780106,
      "loss": 1.0509,
      "mean_token_accuracy": 0.7413963824510574,
      "num_tokens": 10596639.0,
      "step": 325
    },
    {
      "entropy": 1.0835492014884949,
      "epoch": 0.17433155080213902,
      "grad_norm": 0.12105121463537216,
      "learning_rate": 0.00019046837179129695,
      "loss": 1.0933,
      "mean_token_accuracy": 0.7261119484901428,
      "num_tokens": 10629407.0,
      "step": 326
    },
    {
      "entropy": 1.0530675500631332,
      "epoch": 0.1748663101604278,
      "grad_norm": 0.1176065057516098,
      "learning_rate": 0.00019039840244681664,
      "loss": 1.0537,
      "mean_token_accuracy": 0.7409262210130692,
      "num_tokens": 10661368.0,
      "step": 327
    },
    {
      "entropy": 1.0586835443973541,
      "epoch": 0.1754010695187166,
      "grad_norm": 0.11797241121530533,
      "learning_rate": 0.00019032819169445322,
      "loss": 1.0729,
      "mean_token_accuracy": 0.73880934715271,
      "num_tokens": 10694127.0,
      "step": 328
    },
    {
      "entropy": 1.0565836429595947,
      "epoch": 0.17593582887700535,
      "grad_norm": 0.12271396815776825,
      "learning_rate": 0.00019025773974502472,
      "loss": 1.0577,
      "mean_token_accuracy": 0.7349401414394379,
      "num_tokens": 10726895.0,
      "step": 329
    },
    {
      "entropy": 1.055390864610672,
      "epoch": 0.17647058823529413,
      "grad_norm": 0.12326763570308685,
      "learning_rate": 0.0001901870468100733,
      "loss": 1.057,
      "mean_token_accuracy": 0.7380874454975128,
      "num_tokens": 10759200.0,
      "step": 330
    },
    {
      "entropy": 1.0646967589855194,
      "epoch": 0.17700534759358288,
      "grad_norm": 0.11958156526088715,
      "learning_rate": 0.00019011611310186478,
      "loss": 1.0731,
      "mean_token_accuracy": 0.7317423224449158,
      "num_tokens": 10791603.0,
      "step": 331
    },
    {
      "entropy": 1.0165407955646515,
      "epoch": 0.17754010695187167,
      "grad_norm": 0.1222132071852684,
      "learning_rate": 0.00019004493883338796,
      "loss": 1.0066,
      "mean_token_accuracy": 0.7471896409988403,
      "num_tokens": 10824371.0,
      "step": 332
    },
    {
      "entropy": 1.048438012599945,
      "epoch": 0.17807486631016042,
      "grad_norm": 0.12711326777935028,
      "learning_rate": 0.0001899735242183539,
      "loss": 1.0211,
      "mean_token_accuracy": 0.7453262507915497,
      "num_tokens": 10857139.0,
      "step": 333
    },
    {
      "entropy": 1.016156643629074,
      "epoch": 0.1786096256684492,
      "grad_norm": 0.11839782446622849,
      "learning_rate": 0.00018990186947119534,
      "loss": 1.0209,
      "mean_token_accuracy": 0.741521418094635,
      "num_tokens": 10889784.0,
      "step": 334
    },
    {
      "entropy": 1.0444038808345795,
      "epoch": 0.17914438502673796,
      "grad_norm": 0.11765804886817932,
      "learning_rate": 0.0001898299748070661,
      "loss": 1.0497,
      "mean_token_accuracy": 0.7371090054512024,
      "num_tokens": 10922552.0,
      "step": 335
    },
    {
      "entropy": 1.0168912708759308,
      "epoch": 0.17967914438502675,
      "grad_norm": 0.11785600334405899,
      "learning_rate": 0.00018975784044184031,
      "loss": 1.0209,
      "mean_token_accuracy": 0.7431573867797852,
      "num_tokens": 10955320.0,
      "step": 336
    },
    {
      "entropy": 1.0603578388690948,
      "epoch": 0.1802139037433155,
      "grad_norm": 0.343005895614624,
      "learning_rate": 0.00018968546659211194,
      "loss": 1.0687,
      "mean_token_accuracy": 0.7360398471355438,
      "num_tokens": 10988088.0,
      "step": 337
    },
    {
      "entropy": 1.022160217165947,
      "epoch": 0.18074866310160428,
      "grad_norm": 0.12493393570184708,
      "learning_rate": 0.00018961285347519392,
      "loss": 1.028,
      "mean_token_accuracy": 0.7398343086242676,
      "num_tokens": 11020497.0,
      "step": 338
    },
    {
      "entropy": 0.9765367656946182,
      "epoch": 0.18128342245989304,
      "grad_norm": 0.12323349714279175,
      "learning_rate": 0.00018954000130911772,
      "loss": 0.9741,
      "mean_token_accuracy": 0.7540017068386078,
      "num_tokens": 11053265.0,
      "step": 339
    },
    {
      "entropy": 1.0765683948993683,
      "epoch": 0.18181818181818182,
      "grad_norm": 0.12275473773479462,
      "learning_rate": 0.00018946691031263252,
      "loss": 1.084,
      "mean_token_accuracy": 0.7321603298187256,
      "num_tokens": 11086033.0,
      "step": 340
    },
    {
      "entropy": 1.035182625055313,
      "epoch": 0.18235294117647058,
      "grad_norm": 0.11736733466386795,
      "learning_rate": 0.0001893935807052047,
      "loss": 1.0236,
      "mean_token_accuracy": 0.7407746911048889,
      "num_tokens": 11118801.0,
      "step": 341
    },
    {
      "entropy": 1.0754371583461761,
      "epoch": 0.18288770053475936,
      "grad_norm": 0.11700764298439026,
      "learning_rate": 0.00018932001270701706,
      "loss": 1.0737,
      "mean_token_accuracy": 0.7352294772863388,
      "num_tokens": 11151427.0,
      "step": 342
    },
    {
      "entropy": 1.0477129220962524,
      "epoch": 0.18342245989304812,
      "grad_norm": 0.12123466283082962,
      "learning_rate": 0.00018924620653896815,
      "loss": 1.0226,
      "mean_token_accuracy": 0.7426380813121796,
      "num_tokens": 11184195.0,
      "step": 343
    },
    {
      "entropy": 1.0352286100387573,
      "epoch": 0.1839572192513369,
      "grad_norm": 0.13648684322834015,
      "learning_rate": 0.00018917216242267179,
      "loss": 1.0256,
      "mean_token_accuracy": 0.7435887604951859,
      "num_tokens": 11216545.0,
      "step": 344
    },
    {
      "entropy": 0.979494497179985,
      "epoch": 0.18449197860962566,
      "grad_norm": 0.16731642186641693,
      "learning_rate": 0.00018909788058045612,
      "loss": 0.9789,
      "mean_token_accuracy": 0.7541547864675522,
      "num_tokens": 11248991.0,
      "step": 345
    },
    {
      "entropy": 0.9837481826543808,
      "epoch": 0.18502673796791444,
      "grad_norm": 0.19374457001686096,
      "learning_rate": 0.0001890233612353632,
      "loss": 0.9886,
      "mean_token_accuracy": 0.7531463801860809,
      "num_tokens": 11281759.0,
      "step": 346
    },
    {
      "entropy": 1.0341242849826813,
      "epoch": 0.1855614973262032,
      "grad_norm": 0.18110328912734985,
      "learning_rate": 0.00018894860461114818,
      "loss": 1.0489,
      "mean_token_accuracy": 0.7438734620809555,
      "num_tokens": 11314486.0,
      "step": 347
    },
    {
      "entropy": 0.9882454127073288,
      "epoch": 0.18609625668449198,
      "grad_norm": 0.12236490100622177,
      "learning_rate": 0.00018887361093227872,
      "loss": 1.0048,
      "mean_token_accuracy": 0.7507801651954651,
      "num_tokens": 11347165.0,
      "step": 348
    },
    {
      "entropy": 1.0009735822677612,
      "epoch": 0.18663101604278076,
      "grad_norm": 0.13956494629383087,
      "learning_rate": 0.00018879838042393414,
      "loss": 1.0179,
      "mean_token_accuracy": 0.7439083456993103,
      "num_tokens": 11379742.0,
      "step": 349
    },
    {
      "entropy": 1.0519947409629822,
      "epoch": 0.18716577540106952,
      "grad_norm": 0.15381091833114624,
      "learning_rate": 0.00018872291331200504,
      "loss": 1.0515,
      "mean_token_accuracy": 0.7423379421234131,
      "num_tokens": 11412483.0,
      "step": 350
    },
    {
      "entropy": 1.053426742553711,
      "epoch": 0.1877005347593583,
      "grad_norm": 0.11703289300203323,
      "learning_rate": 0.00018864720982309242,
      "loss": 1.0532,
      "mean_token_accuracy": 0.7338098883628845,
      "num_tokens": 11445251.0,
      "step": 351
    },
    {
      "entropy": 1.0194750875234604,
      "epoch": 0.18823529411764706,
      "grad_norm": 0.1410088986158371,
      "learning_rate": 0.00018857127018450694,
      "loss": 1.02,
      "mean_token_accuracy": 0.7453415095806122,
      "num_tokens": 11477668.0,
      "step": 352
    },
    {
      "entropy": 1.0677756518125534,
      "epoch": 0.18877005347593584,
      "grad_norm": 0.13086667656898499,
      "learning_rate": 0.00018849509462426844,
      "loss": 1.0452,
      "mean_token_accuracy": 0.7393670529127121,
      "num_tokens": 11510162.0,
      "step": 353
    },
    {
      "entropy": 1.0423657298088074,
      "epoch": 0.1893048128342246,
      "grad_norm": 0.12789122760295868,
      "learning_rate": 0.00018841868337110508,
      "loss": 1.0239,
      "mean_token_accuracy": 0.7400415390729904,
      "num_tokens": 11542930.0,
      "step": 354
    },
    {
      "entropy": 1.0276905447244644,
      "epoch": 0.18983957219251338,
      "grad_norm": 0.1383594572544098,
      "learning_rate": 0.00018834203665445273,
      "loss": 1.0199,
      "mean_token_accuracy": 0.7431428879499435,
      "num_tokens": 11575513.0,
      "step": 355
    },
    {
      "entropy": 1.017955631017685,
      "epoch": 0.19037433155080213,
      "grad_norm": 0.12443074584007263,
      "learning_rate": 0.0001882651547044543,
      "loss": 1.0202,
      "mean_token_accuracy": 0.7448633909225464,
      "num_tokens": 11607784.0,
      "step": 356
    },
    {
      "entropy": 1.047084480524063,
      "epoch": 0.19090909090909092,
      "grad_norm": 0.13764646649360657,
      "learning_rate": 0.00018818803775195902,
      "loss": 1.044,
      "mean_token_accuracy": 0.7388728559017181,
      "num_tokens": 11640511.0,
      "step": 357
    },
    {
      "entropy": 1.0577262043952942,
      "epoch": 0.19144385026737967,
      "grad_norm": 0.12104304879903793,
      "learning_rate": 0.00018811068602852182,
      "loss": 1.0679,
      "mean_token_accuracy": 0.7354218810796738,
      "num_tokens": 11673124.0,
      "step": 358
    },
    {
      "entropy": 1.0496059656143188,
      "epoch": 0.19197860962566846,
      "grad_norm": 0.1261080801486969,
      "learning_rate": 0.00018803309976640238,
      "loss": 1.066,
      "mean_token_accuracy": 0.7340228408575058,
      "num_tokens": 11705792.0,
      "step": 359
    },
    {
      "entropy": 1.0624986588954926,
      "epoch": 0.1925133689839572,
      "grad_norm": 0.12073995172977448,
      "learning_rate": 0.00018795527919856478,
      "loss": 1.0651,
      "mean_token_accuracy": 0.7365228086709976,
      "num_tokens": 11738238.0,
      "step": 360
    },
    {
      "entropy": 1.0668971538543701,
      "epoch": 0.193048128342246,
      "grad_norm": 0.12623706459999084,
      "learning_rate": 0.00018787722455867653,
      "loss": 1.0783,
      "mean_token_accuracy": 0.7339484840631485,
      "num_tokens": 11770819.0,
      "step": 361
    },
    {
      "entropy": 0.9814553409814835,
      "epoch": 0.19358288770053475,
      "grad_norm": 0.11412834376096725,
      "learning_rate": 0.0001877989360811081,
      "loss": 0.9742,
      "mean_token_accuracy": 0.7551625072956085,
      "num_tokens": 11803587.0,
      "step": 362
    },
    {
      "entropy": 1.0311852097511292,
      "epoch": 0.19411764705882353,
      "grad_norm": 0.12380577623844147,
      "learning_rate": 0.00018772041400093192,
      "loss": 1.0344,
      "mean_token_accuracy": 0.7427579909563065,
      "num_tokens": 11836073.0,
      "step": 363
    },
    {
      "entropy": 1.0336099565029144,
      "epoch": 0.1946524064171123,
      "grad_norm": 0.11791545897722244,
      "learning_rate": 0.000187641658553922,
      "loss": 1.0344,
      "mean_token_accuracy": 0.7441349029541016,
      "num_tokens": 11868841.0,
      "step": 364
    },
    {
      "entropy": 1.012706995010376,
      "epoch": 0.19518716577540107,
      "grad_norm": 0.13155929744243622,
      "learning_rate": 0.00018756266997655305,
      "loss": 1.004,
      "mean_token_accuracy": 0.7522120326757431,
      "num_tokens": 11901481.0,
      "step": 365
    },
    {
      "entropy": 0.9872090071439743,
      "epoch": 0.19572192513368983,
      "grad_norm": 0.12945513427257538,
      "learning_rate": 0.00018748344850599968,
      "loss": 0.9943,
      "mean_token_accuracy": 0.7532074749469757,
      "num_tokens": 11934249.0,
      "step": 366
    },
    {
      "entropy": 1.0130062401294708,
      "epoch": 0.1962566844919786,
      "grad_norm": 0.1238793134689331,
      "learning_rate": 0.0001874039943801359,
      "loss": 1.0095,
      "mean_token_accuracy": 0.7486846446990967,
      "num_tokens": 11966973.0,
      "step": 367
    },
    {
      "entropy": 1.037500947713852,
      "epoch": 0.19679144385026737,
      "grad_norm": 0.11912263929843903,
      "learning_rate": 0.00018732430783753424,
      "loss": 1.0407,
      "mean_token_accuracy": 0.7420326620340347,
      "num_tokens": 11999419.0,
      "step": 368
    },
    {
      "entropy": 0.9854730367660522,
      "epoch": 0.19732620320855615,
      "grad_norm": 0.12523306906223297,
      "learning_rate": 0.00018724438911746512,
      "loss": 0.9861,
      "mean_token_accuracy": 0.7542155385017395,
      "num_tokens": 12032187.0,
      "step": 369
    },
    {
      "entropy": 1.0310074090957642,
      "epoch": 0.19786096256684493,
      "grad_norm": 0.13841532170772552,
      "learning_rate": 0.0001871642384598962,
      "loss": 1.0353,
      "mean_token_accuracy": 0.7444813102483749,
      "num_tokens": 12064494.0,
      "step": 370
    },
    {
      "entropy": 1.0406521409749985,
      "epoch": 0.1983957219251337,
      "grad_norm": 0.12111449986696243,
      "learning_rate": 0.00018708385610549132,
      "loss": 1.0325,
      "mean_token_accuracy": 0.7387104332447052,
      "num_tokens": 12097191.0,
      "step": 371
    },
    {
      "entropy": 1.0652667582035065,
      "epoch": 0.19893048128342247,
      "grad_norm": 0.12144946306943893,
      "learning_rate": 0.0001870032422956103,
      "loss": 1.0514,
      "mean_token_accuracy": 0.7392802536487579,
      "num_tokens": 12129721.0,
      "step": 372
    },
    {
      "entropy": 1.081633985042572,
      "epoch": 0.19946524064171123,
      "grad_norm": 0.11997553706169128,
      "learning_rate": 0.00018692239727230775,
      "loss": 1.0706,
      "mean_token_accuracy": 0.7328018248081207,
      "num_tokens": 12162489.0,
      "step": 373
    },
    {
      "entropy": 1.019247144460678,
      "epoch": 0.2,
      "grad_norm": 0.12147719413042068,
      "learning_rate": 0.00018684132127833261,
      "loss": 1.0006,
      "mean_token_accuracy": 0.7459330856800079,
      "num_tokens": 12194555.0,
      "step": 374
    },
    {
      "entropy": 1.0349364280700684,
      "epoch": 0.20053475935828877,
      "grad_norm": 0.1279066652059555,
      "learning_rate": 0.00018676001455712735,
      "loss": 1.044,
      "mean_token_accuracy": 0.743842676281929,
      "num_tokens": 12227219.0,
      "step": 375
    },
    {
      "entropy": 1.0251546055078506,
      "epoch": 0.20106951871657755,
      "grad_norm": 0.12099071592092514,
      "learning_rate": 0.00018667847735282718,
      "loss": 1.0499,
      "mean_token_accuracy": 0.7395733743906021,
      "num_tokens": 12259788.0,
      "step": 376
    },
    {
      "entropy": 1.0475497543811798,
      "epoch": 0.2016042780748663,
      "grad_norm": 0.12799841165542603,
      "learning_rate": 0.0001865967099102594,
      "loss": 1.0465,
      "mean_token_accuracy": 0.7413245439529419,
      "num_tokens": 12292556.0,
      "step": 377
    },
    {
      "entropy": 1.039279356598854,
      "epoch": 0.2021390374331551,
      "grad_norm": 0.12248227000236511,
      "learning_rate": 0.00018651471247494271,
      "loss": 1.0659,
      "mean_token_accuracy": 0.7357649058103561,
      "num_tokens": 12325324.0,
      "step": 378
    },
    {
      "entropy": 1.0145466774702072,
      "epoch": 0.20267379679144384,
      "grad_norm": 0.12063663452863693,
      "learning_rate": 0.0001864324852930862,
      "loss": 1.0182,
      "mean_token_accuracy": 0.7449902296066284,
      "num_tokens": 12358092.0,
      "step": 379
    },
    {
      "entropy": 1.0798068642616272,
      "epoch": 0.20320855614973263,
      "grad_norm": 0.12489446997642517,
      "learning_rate": 0.00018635002861158907,
      "loss": 1.0682,
      "mean_token_accuracy": 0.7357427775859833,
      "num_tokens": 12390537.0,
      "step": 380
    },
    {
      "entropy": 1.076945185661316,
      "epoch": 0.20374331550802138,
      "grad_norm": 0.11916907131671906,
      "learning_rate": 0.00018626734267803937,
      "loss": 1.0765,
      "mean_token_accuracy": 0.7331989258527756,
      "num_tokens": 12423305.0,
      "step": 381
    },
    {
      "entropy": 1.0346795320510864,
      "epoch": 0.20427807486631017,
      "grad_norm": 0.1225687712430954,
      "learning_rate": 0.00018618442774071364,
      "loss": 1.0329,
      "mean_token_accuracy": 0.7417216598987579,
      "num_tokens": 12456073.0,
      "step": 382
    },
    {
      "entropy": 1.0055777728557587,
      "epoch": 0.20481283422459892,
      "grad_norm": 0.12688912451267242,
      "learning_rate": 0.00018610128404857604,
      "loss": 0.9866,
      "mean_token_accuracy": 0.750934898853302,
      "num_tokens": 12488140.0,
      "step": 383
    },
    {
      "entropy": 1.0724992752075195,
      "epoch": 0.2053475935828877,
      "grad_norm": 0.11506830900907516,
      "learning_rate": 0.00018601791185127766,
      "loss": 1.0648,
      "mean_token_accuracy": 0.7341200113296509,
      "num_tokens": 12520340.0,
      "step": 384
    },
    {
      "entropy": 1.0266041606664658,
      "epoch": 0.20588235294117646,
      "grad_norm": 0.12300301343202591,
      "learning_rate": 0.00018593431139915552,
      "loss": 1.0227,
      "mean_token_accuracy": 0.7410801649093628,
      "num_tokens": 12553108.0,
      "step": 385
    },
    {
      "entropy": 1.0123431384563446,
      "epoch": 0.20641711229946524,
      "grad_norm": 0.1202920526266098,
      "learning_rate": 0.0001858504829432322,
      "loss": 0.9988,
      "mean_token_accuracy": 0.7493287026882172,
      "num_tokens": 12585557.0,
      "step": 386
    },
    {
      "entropy": 1.0406652837991714,
      "epoch": 0.206951871657754,
      "grad_norm": 0.11811408400535583,
      "learning_rate": 0.00018576642673521482,
      "loss": 1.0461,
      "mean_token_accuracy": 0.736681342124939,
      "num_tokens": 12618325.0,
      "step": 387
    },
    {
      "entropy": 1.0288068503141403,
      "epoch": 0.20748663101604278,
      "grad_norm": 0.11991453170776367,
      "learning_rate": 0.00018568214302749428,
      "loss": 1.0299,
      "mean_token_accuracy": 0.7442842423915863,
      "num_tokens": 12650986.0,
      "step": 388
    },
    {
      "entropy": 1.0087891221046448,
      "epoch": 0.20802139037433154,
      "grad_norm": 0.11459293961524963,
      "learning_rate": 0.00018559763207314468,
      "loss": 1.0041,
      "mean_token_accuracy": 0.7455913424491882,
      "num_tokens": 12683661.0,
      "step": 389
    },
    {
      "entropy": 0.9773480743169785,
      "epoch": 0.20855614973262032,
      "grad_norm": 0.11871340870857239,
      "learning_rate": 0.00018551289412592248,
      "loss": 0.9853,
      "mean_token_accuracy": 0.7529376000165939,
      "num_tokens": 12716352.0,
      "step": 390
    },
    {
      "entropy": 0.9953966587781906,
      "epoch": 0.20909090909090908,
      "grad_norm": 0.1180933490395546,
      "learning_rate": 0.00018542792944026566,
      "loss": 1.0069,
      "mean_token_accuracy": 0.7465786933898926,
      "num_tokens": 12749120.0,
      "step": 391
    },
    {
      "entropy": 1.0300866663455963,
      "epoch": 0.20962566844919786,
      "grad_norm": 0.12579035758972168,
      "learning_rate": 0.00018534273827129305,
      "loss": 1.0327,
      "mean_token_accuracy": 0.740877166390419,
      "num_tokens": 12781794.0,
      "step": 392
    },
    {
      "entropy": 0.9939381182193756,
      "epoch": 0.21016042780748664,
      "grad_norm": 0.11927133053541183,
      "learning_rate": 0.00018525732087480346,
      "loss": 0.9842,
      "mean_token_accuracy": 0.7507331371307373,
      "num_tokens": 12814562.0,
      "step": 393
    },
    {
      "entropy": 1.0085733830928802,
      "epoch": 0.2106951871657754,
      "grad_norm": 0.12323582172393799,
      "learning_rate": 0.00018517167750727506,
      "loss": 1.0029,
      "mean_token_accuracy": 0.7487475574016571,
      "num_tokens": 12847330.0,
      "step": 394
    },
    {
      "entropy": 1.0676177442073822,
      "epoch": 0.21122994652406418,
      "grad_norm": 0.12364771217107773,
      "learning_rate": 0.0001850858084258645,
      "loss": 1.0592,
      "mean_token_accuracy": 0.7352042198181152,
      "num_tokens": 12879999.0,
      "step": 395
    },
    {
      "entropy": 1.002898469567299,
      "epoch": 0.21176470588235294,
      "grad_norm": 0.12208950519561768,
      "learning_rate": 0.0001849997138884062,
      "loss": 1.0057,
      "mean_token_accuracy": 0.7455271631479263,
      "num_tokens": 12912638.0,
      "step": 396
    },
    {
      "entropy": 1.0296027660369873,
      "epoch": 0.21229946524064172,
      "grad_norm": 0.12361408025026321,
      "learning_rate": 0.00018491339415341147,
      "loss": 1.0372,
      "mean_token_accuracy": 0.7383919954299927,
      "num_tokens": 12945406.0,
      "step": 397
    },
    {
      "entropy": 1.0023837834596634,
      "epoch": 0.21283422459893048,
      "grad_norm": 0.11596493422985077,
      "learning_rate": 0.00018482684948006788,
      "loss": 0.9916,
      "mean_token_accuracy": 0.748087540268898,
      "num_tokens": 12978152.0,
      "step": 398
    },
    {
      "entropy": 1.0242990553379059,
      "epoch": 0.21336898395721926,
      "grad_norm": 0.12375675141811371,
      "learning_rate": 0.00018474008012823835,
      "loss": 1.0141,
      "mean_token_accuracy": 0.7455989420413971,
      "num_tokens": 13010174.0,
      "step": 399
    },
    {
      "entropy": 1.0101054310798645,
      "epoch": 0.21390374331550802,
      "grad_norm": 0.13966000080108643,
      "learning_rate": 0.0001846530863584605,
      "loss": 1.0308,
      "mean_token_accuracy": 0.7396749705076218,
      "num_tokens": 13042942.0,
      "step": 400
    },
    {
      "entropy": 1.0114232301712036,
      "epoch": 0.2144385026737968,
      "grad_norm": 0.1396958827972412,
      "learning_rate": 0.0001845658684319458,
      "loss": 0.9996,
      "mean_token_accuracy": 0.7465191334486008,
      "num_tokens": 13075481.0,
      "step": 401
    },
    {
      "entropy": 0.9742856174707413,
      "epoch": 0.21497326203208555,
      "grad_norm": 0.14045993983745575,
      "learning_rate": 0.00018447842661057867,
      "loss": 0.9846,
      "mean_token_accuracy": 0.750946968793869,
      "num_tokens": 13108249.0,
      "step": 402
    },
    {
      "entropy": 1.0806566923856735,
      "epoch": 0.21550802139037434,
      "grad_norm": 0.13654112815856934,
      "learning_rate": 0.0001843907611569159,
      "loss": 1.0752,
      "mean_token_accuracy": 0.7342375367879868,
      "num_tokens": 13141017.0,
      "step": 403
    },
    {
      "entropy": 0.9732632040977478,
      "epoch": 0.2160427807486631,
      "grad_norm": 0.12816397845745087,
      "learning_rate": 0.0001843028723341858,
      "loss": 0.9799,
      "mean_token_accuracy": 0.7495112419128418,
      "num_tokens": 13173785.0,
      "step": 404
    },
    {
      "entropy": 1.0426502525806427,
      "epoch": 0.21657754010695188,
      "grad_norm": 0.1289195716381073,
      "learning_rate": 0.00018421476040628723,
      "loss": 1.0596,
      "mean_token_accuracy": 0.7384530901908875,
      "num_tokens": 13206553.0,
      "step": 405
    },
    {
      "entropy": 1.0064526945352554,
      "epoch": 0.21711229946524063,
      "grad_norm": 0.1307249516248703,
      "learning_rate": 0.00018412642563778918,
      "loss": 0.9978,
      "mean_token_accuracy": 0.7462714910507202,
      "num_tokens": 13239020.0,
      "step": 406
    },
    {
      "entropy": 1.0445029735565186,
      "epoch": 0.21764705882352942,
      "grad_norm": 0.135030135512352,
      "learning_rate": 0.0001840378682939295,
      "loss": 1.0335,
      "mean_token_accuracy": 0.7428824603557587,
      "num_tokens": 13271788.0,
      "step": 407
    },
    {
      "entropy": 1.0235690474510193,
      "epoch": 0.21818181818181817,
      "grad_norm": 0.13958315551280975,
      "learning_rate": 0.00018394908864061458,
      "loss": 1.0282,
      "mean_token_accuracy": 0.7400415539741516,
      "num_tokens": 13304556.0,
      "step": 408
    },
    {
      "entropy": 1.0385311245918274,
      "epoch": 0.21871657754010695,
      "grad_norm": 0.14284038543701172,
      "learning_rate": 0.00018386008694441815,
      "loss": 1.0238,
      "mean_token_accuracy": 0.7407746911048889,
      "num_tokens": 13337324.0,
      "step": 409
    },
    {
      "entropy": 1.0151693224906921,
      "epoch": 0.2192513368983957,
      "grad_norm": 0.1381206065416336,
      "learning_rate": 0.0001837708634725808,
      "loss": 1.0196,
      "mean_token_accuracy": 0.7443615198135376,
      "num_tokens": 13370035.0,
      "step": 410
    },
    {
      "entropy": 1.029556393623352,
      "epoch": 0.2197860962566845,
      "grad_norm": 0.12878087162971497,
      "learning_rate": 0.0001836814184930089,
      "loss": 1.0254,
      "mean_token_accuracy": 0.74590665102005,
      "num_tokens": 13402803.0,
      "step": 411
    },
    {
      "entropy": 1.0323573648929596,
      "epoch": 0.22032085561497325,
      "grad_norm": 0.12288200855255127,
      "learning_rate": 0.00018359175227427405,
      "loss": 1.0255,
      "mean_token_accuracy": 0.7426676005125046,
      "num_tokens": 13435252.0,
      "step": 412
    },
    {
      "entropy": 1.0320729613304138,
      "epoch": 0.22085561497326203,
      "grad_norm": 0.1420479416847229,
      "learning_rate": 0.00018350186508561206,
      "loss": 1.0396,
      "mean_token_accuracy": 0.7392666786909103,
      "num_tokens": 13467366.0,
      "step": 413
    },
    {
      "entropy": 1.0164061188697815,
      "epoch": 0.22139037433155082,
      "grad_norm": 0.13489247858524323,
      "learning_rate": 0.0001834117571969223,
      "loss": 1.0241,
      "mean_token_accuracy": 0.7414467334747314,
      "num_tokens": 13500134.0,
      "step": 414
    },
    {
      "entropy": 0.9836405515670776,
      "epoch": 0.22192513368983957,
      "grad_norm": 0.12505358457565308,
      "learning_rate": 0.0001833214288787668,
      "loss": 0.9955,
      "mean_token_accuracy": 0.7502505928277969,
      "num_tokens": 13532484.0,
      "step": 415
    },
    {
      "entropy": 1.0373257994651794,
      "epoch": 0.22245989304812835,
      "grad_norm": 0.13148537278175354,
      "learning_rate": 0.0001832308804023694,
      "loss": 1.0423,
      "mean_token_accuracy": 0.74172542989254,
      "num_tokens": 13565187.0,
      "step": 416
    },
    {
      "entropy": 1.025197684764862,
      "epoch": 0.2229946524064171,
      "grad_norm": 0.1361037790775299,
      "learning_rate": 0.00018314011203961515,
      "loss": 1.0128,
      "mean_token_accuracy": 0.7460659295320511,
      "num_tokens": 13597697.0,
      "step": 417
    },
    {
      "entropy": 1.0798514187335968,
      "epoch": 0.2235294117647059,
      "grad_norm": 0.12265218794345856,
      "learning_rate": 0.00018304912406304916,
      "loss": 1.0693,
      "mean_token_accuracy": 0.7381899803876877,
      "num_tokens": 13630215.0,
      "step": 418
    },
    {
      "entropy": 1.101951003074646,
      "epoch": 0.22406417112299465,
      "grad_norm": 0.12851333618164062,
      "learning_rate": 0.00018295791674587605,
      "loss": 1.1005,
      "mean_token_accuracy": 0.7265847474336624,
      "num_tokens": 13662970.0,
      "step": 419
    },
    {
      "entropy": 1.0959776639938354,
      "epoch": 0.22459893048128343,
      "grad_norm": 0.14182980358600616,
      "learning_rate": 0.0001828664903619591,
      "loss": 1.0953,
      "mean_token_accuracy": 0.7329133898019791,
      "num_tokens": 13695397.0,
      "step": 420
    },
    {
      "entropy": 1.0631342232227325,
      "epoch": 0.2251336898395722,
      "grad_norm": 0.12244658917188644,
      "learning_rate": 0.00018277484518581928,
      "loss": 1.0564,
      "mean_token_accuracy": 0.7375366538763046,
      "num_tokens": 13728165.0,
      "step": 421
    },
    {
      "entropy": 1.0424781888723373,
      "epoch": 0.22566844919786097,
      "grad_norm": 0.1308673918247223,
      "learning_rate": 0.00018268298149263455,
      "loss": 1.0384,
      "mean_token_accuracy": 0.7420882135629654,
      "num_tokens": 13760933.0,
      "step": 422
    },
    {
      "entropy": 1.0469609051942825,
      "epoch": 0.22620320855614973,
      "grad_norm": 0.14152593910694122,
      "learning_rate": 0.00018259089955823899,
      "loss": 1.0395,
      "mean_token_accuracy": 0.7439210712909698,
      "num_tokens": 13793701.0,
      "step": 423
    },
    {
      "entropy": 1.0544581413269043,
      "epoch": 0.2267379679144385,
      "grad_norm": 0.12065031379461288,
      "learning_rate": 0.00018249859965912196,
      "loss": 1.0531,
      "mean_token_accuracy": 0.7355510890483856,
      "num_tokens": 13826469.0,
      "step": 424
    },
    {
      "entropy": 1.080516904592514,
      "epoch": 0.22727272727272727,
      "grad_norm": 0.1398177295923233,
      "learning_rate": 0.00018240608207242736,
      "loss": 1.0895,
      "mean_token_accuracy": 0.7328049540519714,
      "num_tokens": 13859107.0,
      "step": 425
    },
    {
      "entropy": 1.0148239135742188,
      "epoch": 0.22780748663101605,
      "grad_norm": 0.1272018402814865,
      "learning_rate": 0.00018231334707595267,
      "loss": 1.0362,
      "mean_token_accuracy": 0.7434576749801636,
      "num_tokens": 13891780.0,
      "step": 426
    },
    {
      "entropy": 1.0019170939922333,
      "epoch": 0.2283422459893048,
      "grad_norm": 0.12834444642066956,
      "learning_rate": 0.00018222039494814813,
      "loss": 1.0023,
      "mean_token_accuracy": 0.7483878880739212,
      "num_tokens": 13924267.0,
      "step": 427
    },
    {
      "entropy": 1.0581309497356415,
      "epoch": 0.2288770053475936,
      "grad_norm": 0.12580765783786774,
      "learning_rate": 0.0001821272259681161,
      "loss": 1.0591,
      "mean_token_accuracy": 0.7372151464223862,
      "num_tokens": 13956221.0,
      "step": 428
    },
    {
      "entropy": 1.0421131998300552,
      "epoch": 0.22941176470588234,
      "grad_norm": 0.12588627636432648,
      "learning_rate": 0.00018203384041560994,
      "loss": 1.042,
      "mean_token_accuracy": 0.7395138293504715,
      "num_tokens": 13988988.0,
      "step": 429
    },
    {
      "entropy": 1.053133800625801,
      "epoch": 0.22994652406417113,
      "grad_norm": 0.13137686252593994,
      "learning_rate": 0.00018194023857103333,
      "loss": 1.0433,
      "mean_token_accuracy": 0.7409885227680206,
      "num_tokens": 14021756.0,
      "step": 430
    },
    {
      "entropy": 1.0503493249416351,
      "epoch": 0.23048128342245988,
      "grad_norm": 0.11900833249092102,
      "learning_rate": 0.0001818464207154394,
      "loss": 1.041,
      "mean_token_accuracy": 0.7390946745872498,
      "num_tokens": 14054281.0,
      "step": 431
    },
    {
      "entropy": 1.018021285533905,
      "epoch": 0.23101604278074866,
      "grad_norm": 0.1321375072002411,
      "learning_rate": 0.00018175238713052992,
      "loss": 1.016,
      "mean_token_accuracy": 0.7537267804145813,
      "num_tokens": 14087049.0,
      "step": 432
    },
    {
      "entropy": 1.079847276210785,
      "epoch": 0.23155080213903742,
      "grad_norm": 0.12485760450363159,
      "learning_rate": 0.00018165813809865435,
      "loss": 1.0716,
      "mean_token_accuracy": 0.7340243756771088,
      "num_tokens": 14119639.0,
      "step": 433
    },
    {
      "entropy": 1.0371785908937454,
      "epoch": 0.2320855614973262,
      "grad_norm": 0.12414313852787018,
      "learning_rate": 0.00018156367390280914,
      "loss": 1.0331,
      "mean_token_accuracy": 0.7424853444099426,
      "num_tokens": 14152407.0,
      "step": 434
    },
    {
      "entropy": 1.0535239577293396,
      "epoch": 0.232620320855615,
      "grad_norm": 0.13701845705509186,
      "learning_rate": 0.00018146899482663674,
      "loss": 1.0642,
      "mean_token_accuracy": 0.7345901429653168,
      "num_tokens": 14184691.0,
      "step": 435
    },
    {
      "entropy": 1.014063835144043,
      "epoch": 0.23315508021390374,
      "grad_norm": 0.12825563549995422,
      "learning_rate": 0.0001813741011544248,
      "loss": 1.0217,
      "mean_token_accuracy": 0.7455740571022034,
      "num_tokens": 14217452.0,
      "step": 436
    },
    {
      "entropy": 1.0205018222332,
      "epoch": 0.23368983957219253,
      "grad_norm": 0.13586899638175964,
      "learning_rate": 0.0001812789931711054,
      "loss": 1.0306,
      "mean_token_accuracy": 0.7433712184429169,
      "num_tokens": 14250220.0,
      "step": 437
    },
    {
      "entropy": 0.995821937918663,
      "epoch": 0.23422459893048128,
      "grad_norm": 0.12893128395080566,
      "learning_rate": 0.0001811836711622541,
      "loss": 0.9992,
      "mean_token_accuracy": 0.7476478517055511,
      "num_tokens": 14282988.0,
      "step": 438
    },
    {
      "entropy": 0.996499165892601,
      "epoch": 0.23475935828877006,
      "grad_norm": 0.12627406418323517,
      "learning_rate": 0.00018108813541408905,
      "loss": 0.9886,
      "mean_token_accuracy": 0.7468492090702057,
      "num_tokens": 14315721.0,
      "step": 439
    },
    {
      "entropy": 1.0268193781375885,
      "epoch": 0.23529411764705882,
      "grad_norm": 0.12785205245018005,
      "learning_rate": 0.00018099238621347013,
      "loss": 1.0178,
      "mean_token_accuracy": 0.7465786933898926,
      "num_tokens": 14348489.0,
      "step": 440
    },
    {
      "entropy": 1.0284498482942581,
      "epoch": 0.2358288770053476,
      "grad_norm": 0.12388309091329575,
      "learning_rate": 0.00018089642384789837,
      "loss": 1.0158,
      "mean_token_accuracy": 0.7436087131500244,
      "num_tokens": 14381033.0,
      "step": 441
    },
    {
      "entropy": 1.014558106660843,
      "epoch": 0.23636363636363636,
      "grad_norm": 0.13476227223873138,
      "learning_rate": 0.00018080024860551462,
      "loss": 1.0128,
      "mean_token_accuracy": 0.7464248687028885,
      "num_tokens": 14413724.0,
      "step": 442
    },
    {
      "entropy": 1.0255836695432663,
      "epoch": 0.23689839572192514,
      "grad_norm": 0.1396826058626175,
      "learning_rate": 0.00018070386077509906,
      "loss": 1.0065,
      "mean_token_accuracy": 0.7451001703739166,
      "num_tokens": 14446293.0,
      "step": 443
    },
    {
      "entropy": 1.0523072481155396,
      "epoch": 0.2374331550802139,
      "grad_norm": 0.13807937502861023,
      "learning_rate": 0.00018060726064607013,
      "loss": 1.0719,
      "mean_token_accuracy": 0.7356121689081192,
      "num_tokens": 14479061.0,
      "step": 444
    },
    {
      "entropy": 1.0252759754657745,
      "epoch": 0.23796791443850268,
      "grad_norm": 0.12841564416885376,
      "learning_rate": 0.00018051044850848376,
      "loss": 1.043,
      "mean_token_accuracy": 0.7401412576436996,
      "num_tokens": 14511633.0,
      "step": 445
    },
    {
      "entropy": 0.9956532716751099,
      "epoch": 0.23850267379679144,
      "grad_norm": 0.1594788134098053,
      "learning_rate": 0.00018041342465303245,
      "loss": 1.0014,
      "mean_token_accuracy": 0.7484622746706009,
      "num_tokens": 14544241.0,
      "step": 446
    },
    {
      "entropy": 1.054281085729599,
      "epoch": 0.23903743315508022,
      "grad_norm": 0.13689427077770233,
      "learning_rate": 0.00018031618937104443,
      "loss": 1.0645,
      "mean_token_accuracy": 0.7383919954299927,
      "num_tokens": 14577009.0,
      "step": 447
    },
    {
      "entropy": 1.0149857997894287,
      "epoch": 0.23957219251336898,
      "grad_norm": 0.12950752675533295,
      "learning_rate": 0.00018021874295448274,
      "loss": 1.0095,
      "mean_token_accuracy": 0.7468536198139191,
      "num_tokens": 14609777.0,
      "step": 448
    },
    {
      "entropy": 1.0120531618595123,
      "epoch": 0.24010695187165776,
      "grad_norm": 0.14452537894248962,
      "learning_rate": 0.00018012108569594442,
      "loss": 1.0125,
      "mean_token_accuracy": 0.7444403767585754,
      "num_tokens": 14642545.0,
      "step": 449
    },
    {
      "entropy": 1.069253385066986,
      "epoch": 0.24064171122994651,
      "grad_norm": 0.12331288307905197,
      "learning_rate": 0.00018002321788865956,
      "loss": 1.0612,
      "mean_token_accuracy": 0.7359224706888199,
      "num_tokens": 14674929.0,
      "step": 450
    },
    {
      "entropy": 1.0392256081104279,
      "epoch": 0.2411764705882353,
      "grad_norm": 0.1363597810268402,
      "learning_rate": 0.0001799251398264905,
      "loss": 1.0422,
      "mean_token_accuracy": 0.7407746911048889,
      "num_tokens": 14707697.0,
      "step": 451
    },
    {
      "entropy": 1.0564654767513275,
      "epoch": 0.24171122994652405,
      "grad_norm": 0.12315265089273453,
      "learning_rate": 0.00017982685180393084,
      "loss": 1.0578,
      "mean_token_accuracy": 0.7412023544311523,
      "num_tokens": 14740465.0,
      "step": 452
    },
    {
      "entropy": 0.997405618429184,
      "epoch": 0.24224598930481284,
      "grad_norm": 0.1231595054268837,
      "learning_rate": 0.00017972835411610467,
      "loss": 0.9876,
      "mean_token_accuracy": 0.7521994113922119,
      "num_tokens": 14773233.0,
      "step": 453
    },
    {
      "entropy": 1.0320776253938675,
      "epoch": 0.2427807486631016,
      "grad_norm": 0.13027168810367584,
      "learning_rate": 0.00017962964705876562,
      "loss": 1.0324,
      "mean_token_accuracy": 0.7407857179641724,
      "num_tokens": 14805547.0,
      "step": 454
    },
    {
      "entropy": 1.0264951586723328,
      "epoch": 0.24331550802139038,
      "grad_norm": 0.123665951192379,
      "learning_rate": 0.0001795307309282959,
      "loss": 1.0368,
      "mean_token_accuracy": 0.740406945347786,
      "num_tokens": 14838168.0,
      "step": 455
    },
    {
      "entropy": 1.039396345615387,
      "epoch": 0.24385026737967913,
      "grad_norm": 0.1296997368335724,
      "learning_rate": 0.0001794316060217056,
      "loss": 1.0429,
      "mean_token_accuracy": 0.7405395954847336,
      "num_tokens": 14870827.0,
      "step": 456
    },
    {
      "entropy": 0.9936301857233047,
      "epoch": 0.24438502673796791,
      "grad_norm": 0.12982916831970215,
      "learning_rate": 0.00017933227263663168,
      "loss": 0.9777,
      "mean_token_accuracy": 0.7492057681083679,
      "num_tokens": 14903595.0,
      "step": 457
    },
    {
      "entropy": 1.0279450565576553,
      "epoch": 0.2449197860962567,
      "grad_norm": 0.13376674056053162,
      "learning_rate": 0.00017923273107133702,
      "loss": 1.0339,
      "mean_token_accuracy": 0.7414161860942841,
      "num_tokens": 14936363.0,
      "step": 458
    },
    {
      "entropy": 1.0354578495025635,
      "epoch": 0.24545454545454545,
      "grad_norm": 0.12655210494995117,
      "learning_rate": 0.0001791329816247097,
      "loss": 1.0215,
      "mean_token_accuracy": 0.7418133020401001,
      "num_tokens": 14969131.0,
      "step": 459
    },
    {
      "entropy": 1.0363547652959824,
      "epoch": 0.24598930481283424,
      "grad_norm": 0.12433706969022751,
      "learning_rate": 0.00017903302459626181,
      "loss": 1.0429,
      "mean_token_accuracy": 0.7416990548372269,
      "num_tokens": 15001726.0,
      "step": 460
    },
    {
      "entropy": 1.0166066586971283,
      "epoch": 0.246524064171123,
      "grad_norm": 0.13178230822086334,
      "learning_rate": 0.00017893286028612893,
      "loss": 1.0154,
      "mean_token_accuracy": 0.7441807091236115,
      "num_tokens": 15034238.0,
      "step": 461
    },
    {
      "entropy": 1.0380278527736664,
      "epoch": 0.24705882352941178,
      "grad_norm": 0.12242249399423599,
      "learning_rate": 0.00017883248899506894,
      "loss": 1.0187,
      "mean_token_accuracy": 0.7434017658233643,
      "num_tokens": 15067006.0,
      "step": 462
    },
    {
      "entropy": 1.0094977915287018,
      "epoch": 0.24759358288770053,
      "grad_norm": 0.1288090944290161,
      "learning_rate": 0.00017873191102446123,
      "loss": 1.0005,
      "mean_token_accuracy": 0.7474645674228668,
      "num_tokens": 15099774.0,
      "step": 463
    },
    {
      "entropy": 1.0057529509067535,
      "epoch": 0.24812834224598931,
      "grad_norm": 0.127349853515625,
      "learning_rate": 0.0001786311266763057,
      "loss": 1.017,
      "mean_token_accuracy": 0.7436461448669434,
      "num_tokens": 15132542.0,
      "step": 464
    },
    {
      "entropy": 1.0217185318470001,
      "epoch": 0.24866310160427807,
      "grad_norm": 0.12888741493225098,
      "learning_rate": 0.0001785301362532221,
      "loss": 1.0314,
      "mean_token_accuracy": 0.7411412596702576,
      "num_tokens": 15165310.0,
      "step": 465
    },
    {
      "entropy": 1.0755583345890045,
      "epoch": 0.24919786096256685,
      "grad_norm": 0.12444429099559784,
      "learning_rate": 0.00017842894005844876,
      "loss": 1.0886,
      "mean_token_accuracy": 0.7273490577936172,
      "num_tokens": 15197781.0,
      "step": 466
    },
    {
      "entropy": 0.9956681579351425,
      "epoch": 0.2497326203208556,
      "grad_norm": 0.13689233362674713,
      "learning_rate": 0.00017832753839584206,
      "loss": 1.0204,
      "mean_token_accuracy": 0.7466397881507874,
      "num_tokens": 15230549.0,
      "step": 467
    },
    {
      "entropy": 1.092822641134262,
      "epoch": 0.25026737967914436,
      "grad_norm": 0.128508061170578,
      "learning_rate": 0.0001782259315698751,
      "loss": 1.0894,
      "mean_token_accuracy": 0.7267534136772156,
      "num_tokens": 15263317.0,
      "step": 468
    },
    {
      "entropy": 1.0412489771842957,
      "epoch": 0.25080213903743315,
      "grad_norm": 0.12759077548980713,
      "learning_rate": 0.0001781241198856372,
      "loss": 1.0309,
      "mean_token_accuracy": 0.7388196587562561,
      "num_tokens": 15296085.0,
      "step": 469
    },
    {
      "entropy": 1.0570957958698273,
      "epoch": 0.25133689839572193,
      "grad_norm": 0.13412855565547943,
      "learning_rate": 0.00017802210364883272,
      "loss": 1.0521,
      "mean_token_accuracy": 0.7335044145584106,
      "num_tokens": 15328853.0,
      "step": 470
    },
    {
      "entropy": 1.072775959968567,
      "epoch": 0.2518716577540107,
      "grad_norm": 0.12720152735710144,
      "learning_rate": 0.00017791988316578024,
      "loss": 1.0691,
      "mean_token_accuracy": 0.7365253865718842,
      "num_tokens": 15360830.0,
      "step": 471
    },
    {
      "entropy": 1.0713469088077545,
      "epoch": 0.25240641711229944,
      "grad_norm": 0.16397106647491455,
      "learning_rate": 0.00017781745874341162,
      "loss": 1.0732,
      "mean_token_accuracy": 0.7350176274776459,
      "num_tokens": 15393480.0,
      "step": 472
    },
    {
      "entropy": 1.0052785575389862,
      "epoch": 0.2529411764705882,
      "grad_norm": 0.14997598528862,
      "learning_rate": 0.00017771483068927105,
      "loss": 1.009,
      "mean_token_accuracy": 0.7479838728904724,
      "num_tokens": 15426248.0,
      "step": 473
    },
    {
      "entropy": 1.0340908020734787,
      "epoch": 0.253475935828877,
      "grad_norm": 0.13000135123729706,
      "learning_rate": 0.00017761199931151414,
      "loss": 1.0149,
      "mean_token_accuracy": 0.7482282519340515,
      "num_tokens": 15459016.0,
      "step": 474
    },
    {
      "entropy": 1.0392605364322662,
      "epoch": 0.2540106951871658,
      "grad_norm": 0.12135802954435349,
      "learning_rate": 0.00017750896491890711,
      "loss": 1.0381,
      "mean_token_accuracy": 0.7424401193857193,
      "num_tokens": 15491690.0,
      "step": 475
    },
    {
      "entropy": 0.9842015951871872,
      "epoch": 0.2545454545454545,
      "grad_norm": 0.1403506100177765,
      "learning_rate": 0.00017740572782082566,
      "loss": 0.984,
      "mean_token_accuracy": 0.7504887580871582,
      "num_tokens": 15524458.0,
      "step": 476
    },
    {
      "entropy": 1.0196618139743805,
      "epoch": 0.2550802139037433,
      "grad_norm": 0.13045115768909454,
      "learning_rate": 0.00017730228832725414,
      "loss": 1.0212,
      "mean_token_accuracy": 0.7442428767681122,
      "num_tokens": 15557040.0,
      "step": 477
    },
    {
      "entropy": 1.0410612225532532,
      "epoch": 0.2556149732620321,
      "grad_norm": 0.12354070693254471,
      "learning_rate": 0.00017719864674878473,
      "loss": 1.0511,
      "mean_token_accuracy": 0.7382087111473083,
      "num_tokens": 15589808.0,
      "step": 478
    },
    {
      "entropy": 1.0096575170755386,
      "epoch": 0.25614973262032087,
      "grad_norm": 0.12986738979816437,
      "learning_rate": 0.00017709480339661625,
      "loss": 1.0118,
      "mean_token_accuracy": 0.7461607456207275,
      "num_tokens": 15622509.0,
      "step": 479
    },
    {
      "entropy": 1.0030359774827957,
      "epoch": 0.25668449197860965,
      "grad_norm": 0.13224437832832336,
      "learning_rate": 0.00017699075858255351,
      "loss": 1.0026,
      "mean_token_accuracy": 0.7482029497623444,
      "num_tokens": 15654978.0,
      "step": 480
    },
    {
      "entropy": 1.0419865548610687,
      "epoch": 0.2572192513368984,
      "grad_norm": 0.12151559442281723,
      "learning_rate": 0.00017688651261900615,
      "loss": 1.0244,
      "mean_token_accuracy": 0.7438905239105225,
      "num_tokens": 15687746.0,
      "step": 481
    },
    {
      "entropy": 0.993943452835083,
      "epoch": 0.25775401069518716,
      "grad_norm": 0.12632091343402863,
      "learning_rate": 0.0001767820658189878,
      "loss": 0.988,
      "mean_token_accuracy": 0.7510421872138977,
      "num_tokens": 15720109.0,
      "step": 482
    },
    {
      "entropy": 1.048466607928276,
      "epoch": 0.25828877005347595,
      "grad_norm": 0.12814119458198547,
      "learning_rate": 0.00017667741849611518,
      "loss": 1.0317,
      "mean_token_accuracy": 0.742913007736206,
      "num_tokens": 15752877.0,
      "step": 483
    },
    {
      "entropy": 0.9965784847736359,
      "epoch": 0.25882352941176473,
      "grad_norm": 0.1250571310520172,
      "learning_rate": 0.00017657257096460709,
      "loss": 0.9938,
      "mean_token_accuracy": 0.7524084597826004,
      "num_tokens": 15785581.0,
      "step": 484
    },
    {
      "entropy": 1.0023357719182968,
      "epoch": 0.25935828877005346,
      "grad_norm": 0.14221987128257751,
      "learning_rate": 0.00017646752353928343,
      "loss": 0.9945,
      "mean_token_accuracy": 0.7445931136608124,
      "num_tokens": 15818349.0,
      "step": 485
    },
    {
      "entropy": 1.0235685110092163,
      "epoch": 0.25989304812834224,
      "grad_norm": 0.12448383867740631,
      "learning_rate": 0.00017636227653556444,
      "loss": 1.0278,
      "mean_token_accuracy": 0.7433406561613083,
      "num_tokens": 15851117.0,
      "step": 486
    },
    {
      "entropy": 1.0168971568346024,
      "epoch": 0.260427807486631,
      "grad_norm": 0.14768727123737335,
      "learning_rate": 0.00017625683026946943,
      "loss": 1.0336,
      "mean_token_accuracy": 0.7419965863227844,
      "num_tokens": 15883885.0,
      "step": 487
    },
    {
      "entropy": 1.0200735032558441,
      "epoch": 0.2609625668449198,
      "grad_norm": 0.1485704928636551,
      "learning_rate": 0.0001761511850576162,
      "loss": 1.0421,
      "mean_token_accuracy": 0.7370060086250305,
      "num_tokens": 15916589.0,
      "step": 488
    },
    {
      "entropy": 1.0495162308216095,
      "epoch": 0.26149732620320854,
      "grad_norm": 0.12823961675167084,
      "learning_rate": 0.00017604534121721982,
      "loss": 1.0479,
      "mean_token_accuracy": 0.735199049115181,
      "num_tokens": 15949249.0,
      "step": 489
    },
    {
      "entropy": 1.003669187426567,
      "epoch": 0.2620320855614973,
      "grad_norm": 0.1225598081946373,
      "learning_rate": 0.00017593929906609183,
      "loss": 0.9972,
      "mean_token_accuracy": 0.751557931303978,
      "num_tokens": 15982017.0,
      "step": 490
    },
    {
      "entropy": 1.0210746675729752,
      "epoch": 0.2625668449197861,
      "grad_norm": 0.14348438382148743,
      "learning_rate": 0.00017583305892263912,
      "loss": 1.0116,
      "mean_token_accuracy": 0.7477700412273407,
      "num_tokens": 16014785.0,
      "step": 491
    },
    {
      "entropy": 1.0653307139873505,
      "epoch": 0.2631016042780749,
      "grad_norm": 0.1344357132911682,
      "learning_rate": 0.0001757266211058633,
      "loss": 1.0628,
      "mean_token_accuracy": 0.7340181171894073,
      "num_tokens": 16047491.0,
      "step": 492
    },
    {
      "entropy": 1.0467574447393417,
      "epoch": 0.2636363636363636,
      "grad_norm": 0.12318356335163116,
      "learning_rate": 0.0001756199859353592,
      "loss": 1.0468,
      "mean_token_accuracy": 0.7361839115619659,
      "num_tokens": 16079723.0,
      "step": 493
    },
    {
      "entropy": 1.0086955726146698,
      "epoch": 0.2641711229946524,
      "grad_norm": 0.1444452404975891,
      "learning_rate": 0.0001755131537313145,
      "loss": 1.0057,
      "mean_token_accuracy": 0.7454583793878555,
      "num_tokens": 16112352.0,
      "step": 494
    },
    {
      "entropy": 1.0184801816940308,
      "epoch": 0.2647058823529412,
      "grad_norm": 0.15575626492500305,
      "learning_rate": 0.0001754061248145084,
      "loss": 1.0077,
      "mean_token_accuracy": 0.7482426911592484,
      "num_tokens": 16145107.0,
      "step": 495
    },
    {
      "entropy": 0.9972895234823227,
      "epoch": 0.26524064171122996,
      "grad_norm": 0.14014114439487457,
      "learning_rate": 0.00017529889950631083,
      "loss": 1.0077,
      "mean_token_accuracy": 0.7456616163253784,
      "num_tokens": 16177485.0,
      "step": 496
    },
    {
      "entropy": 0.9907697141170502,
      "epoch": 0.2657754010695187,
      "grad_norm": 0.12591008841991425,
      "learning_rate": 0.00017519147812868128,
      "loss": 1.002,
      "mean_token_accuracy": 0.7483922243118286,
      "num_tokens": 16209917.0,
      "step": 497
    },
    {
      "entropy": 1.0470885783433914,
      "epoch": 0.2663101604278075,
      "grad_norm": 0.1289779245853424,
      "learning_rate": 0.00017508386100416806,
      "loss": 1.0299,
      "mean_token_accuracy": 0.7467584758996964,
      "num_tokens": 16242673.0,
      "step": 498
    },
    {
      "entropy": 1.0440703928470612,
      "epoch": 0.26684491978609626,
      "grad_norm": 0.1313711255788803,
      "learning_rate": 0.0001749760484559072,
      "loss": 1.036,
      "mean_token_accuracy": 0.7399499118328094,
      "num_tokens": 16275441.0,
      "step": 499
    },
    {
      "entropy": 1.0163945257663727,
      "epoch": 0.26737967914438504,
      "grad_norm": 0.1350429356098175,
      "learning_rate": 0.0001748680408076216,
      "loss": 1.0155,
      "mean_token_accuracy": 0.7431573867797852,
      "num_tokens": 16308209.0,
      "step": 500
    },
    {
      "entropy": 1.0228415131568909,
      "epoch": 0.2679144385026738,
      "grad_norm": 0.1291084736585617,
      "learning_rate": 0.00017475983838361985,
      "loss": 1.0213,
      "mean_token_accuracy": 0.7467547506093979,
      "num_tokens": 16340628.0,
      "step": 501
    },
    {
      "entropy": 1.0208053439855576,
      "epoch": 0.26844919786096255,
      "grad_norm": 0.13432392477989197,
      "learning_rate": 0.00017465144150879548,
      "loss": 1.0301,
      "mean_token_accuracy": 0.7436855286359787,
      "num_tokens": 16373004.0,
      "step": 502
    },
    {
      "entropy": 1.0301097184419632,
      "epoch": 0.26898395721925134,
      "grad_norm": 0.12891647219657898,
      "learning_rate": 0.00017454285050862587,
      "loss": 1.0238,
      "mean_token_accuracy": 0.7430192977190018,
      "num_tokens": 16405632.0,
      "step": 503
    },
    {
      "entropy": 1.0221526473760605,
      "epoch": 0.2695187165775401,
      "grad_norm": 0.12521539628505707,
      "learning_rate": 0.00017443406570917124,
      "loss": 1.0217,
      "mean_token_accuracy": 0.7419375777244568,
      "num_tokens": 16438246.0,
      "step": 504
    },
    {
      "entropy": 1.0305202156305313,
      "epoch": 0.2700534759358289,
      "grad_norm": 0.13040553033351898,
      "learning_rate": 0.00017432508743707374,
      "loss": 1.0377,
      "mean_token_accuracy": 0.7410676926374435,
      "num_tokens": 16470508.0,
      "step": 505
    },
    {
      "entropy": 0.9664936810731888,
      "epoch": 0.27058823529411763,
      "grad_norm": 0.14031217992305756,
      "learning_rate": 0.0001742159160195565,
      "loss": 0.9722,
      "mean_token_accuracy": 0.7550433725118637,
      "num_tokens": 16503037.0,
      "step": 506
    },
    {
      "entropy": 0.9803265780210495,
      "epoch": 0.2711229946524064,
      "grad_norm": 0.1306711584329605,
      "learning_rate": 0.00017410655178442259,
      "loss": 0.9881,
      "mean_token_accuracy": 0.747403472661972,
      "num_tokens": 16535805.0,
      "step": 507
    },
    {
      "entropy": 0.9995748400688171,
      "epoch": 0.2716577540106952,
      "grad_norm": 0.13603125512599945,
      "learning_rate": 0.00017399699506005393,
      "loss": 1.0104,
      "mean_token_accuracy": 0.7487271130084991,
      "num_tokens": 16567934.0,
      "step": 508
    },
    {
      "entropy": 1.0516096651554108,
      "epoch": 0.272192513368984,
      "grad_norm": 0.13807128369808197,
      "learning_rate": 0.00017388724617541058,
      "loss": 1.0551,
      "mean_token_accuracy": 0.7384747862815857,
      "num_tokens": 16600228.0,
      "step": 509
    },
    {
      "entropy": 1.0359112918376923,
      "epoch": 0.2727272727272727,
      "grad_norm": 0.1329445093870163,
      "learning_rate": 0.00017377730546002944,
      "loss": 1.0464,
      "mean_token_accuracy": 0.7396444380283356,
      "num_tokens": 16632996.0,
      "step": 510
    },
    {
      "entropy": 1.0623237937688828,
      "epoch": 0.2732620320855615,
      "grad_norm": 0.12603521347045898,
      "learning_rate": 0.00017366717324402353,
      "loss": 1.0525,
      "mean_token_accuracy": 0.7370479106903076,
      "num_tokens": 16665764.0,
      "step": 511
    },
    {
      "entropy": 1.060817837715149,
      "epoch": 0.2737967914438503,
      "grad_norm": 0.13725413382053375,
      "learning_rate": 0.00017355684985808078,
      "loss": 1.0642,
      "mean_token_accuracy": 0.7357343584299088,
      "num_tokens": 16698532.0,
      "step": 512
    },
    {
      "entropy": 1.0357243567705154,
      "epoch": 0.27433155080213906,
      "grad_norm": 0.15183106064796448,
      "learning_rate": 0.00017344633563346325,
      "loss": 1.0332,
      "mean_token_accuracy": 0.7371090054512024,
      "num_tokens": 16731300.0,
      "step": 513
    },
    {
      "entropy": 0.9966153800487518,
      "epoch": 0.2748663101604278,
      "grad_norm": 0.13324125111103058,
      "learning_rate": 0.0001733356309020059,
      "loss": 0.9744,
      "mean_token_accuracy": 0.7511016726493835,
      "num_tokens": 16763882.0,
      "step": 514
    },
    {
      "entropy": 1.0532770156860352,
      "epoch": 0.27540106951871657,
      "grad_norm": 0.12529823184013367,
      "learning_rate": 0.00017322473599611579,
      "loss": 1.0507,
      "mean_token_accuracy": 0.7382698059082031,
      "num_tokens": 16796650.0,
      "step": 515
    },
    {
      "entropy": 1.0636248290538788,
      "epoch": 0.27593582887700535,
      "grad_norm": 0.13705985248088837,
      "learning_rate": 0.000173113651248771,
      "loss": 1.0659,
      "mean_token_accuracy": 0.738916352391243,
      "num_tokens": 16829259.0,
      "step": 516
    },
    {
      "entropy": 1.0040618181228638,
      "epoch": 0.27647058823529413,
      "grad_norm": 0.1455538123846054,
      "learning_rate": 0.00017300237699351958,
      "loss": 0.992,
      "mean_token_accuracy": 0.7514193654060364,
      "num_tokens": 16861664.0,
      "step": 517
    },
    {
      "entropy": 1.0494634211063385,
      "epoch": 0.27700534759358286,
      "grad_norm": 0.13522738218307495,
      "learning_rate": 0.00017289091356447868,
      "loss": 1.0542,
      "mean_token_accuracy": 0.7430696189403534,
      "num_tokens": 16894217.0,
      "step": 518
    },
    {
      "entropy": 1.0269292891025543,
      "epoch": 0.27754010695187165,
      "grad_norm": 0.1228233054280281,
      "learning_rate": 0.00017277926129633341,
      "loss": 1.0023,
      "mean_token_accuracy": 0.7457484006881714,
      "num_tokens": 16926573.0,
      "step": 519
    },
    {
      "entropy": 1.0406086146831512,
      "epoch": 0.27807486631016043,
      "grad_norm": 0.14273546636104584,
      "learning_rate": 0.000172667420524336,
      "loss": 1.0352,
      "mean_token_accuracy": 0.7418330907821655,
      "num_tokens": 16959322.0,
      "step": 520
    },
    {
      "entropy": 1.0543610453605652,
      "epoch": 0.2786096256684492,
      "grad_norm": 0.17359653115272522,
      "learning_rate": 0.00017255539158430452,
      "loss": 1.0485,
      "mean_token_accuracy": 0.7373014837503433,
      "num_tokens": 16991838.0,
      "step": 521
    },
    {
      "entropy": 1.009380429983139,
      "epoch": 0.279144385026738,
      "grad_norm": 0.12541234493255615,
      "learning_rate": 0.00017244317481262218,
      "loss": 1.007,
      "mean_token_accuracy": 0.7480654567480087,
      "num_tokens": 17024409.0,
      "step": 522
    },
    {
      "entropy": 1.0351165235042572,
      "epoch": 0.2796791443850267,
      "grad_norm": 0.1301652491092682,
      "learning_rate": 0.0001723307705462361,
      "loss": 1.0375,
      "mean_token_accuracy": 0.739235907793045,
      "num_tokens": 17056854.0,
      "step": 523
    },
    {
      "entropy": 0.9839812368154526,
      "epoch": 0.2802139037433155,
      "grad_norm": 0.13403740525245667,
      "learning_rate": 0.0001722181791226565,
      "loss": 1.0012,
      "mean_token_accuracy": 0.7473640590906143,
      "num_tokens": 17089577.0,
      "step": 524
    },
    {
      "entropy": 1.013493835926056,
      "epoch": 0.2807486631016043,
      "grad_norm": 0.13362367451190948,
      "learning_rate": 0.00017210540087995546,
      "loss": 1.0236,
      "mean_token_accuracy": 0.7451429665088654,
      "num_tokens": 17122345.0,
      "step": 525
    },
    {
      "entropy": 1.0553540736436844,
      "epoch": 0.2812834224598931,
      "grad_norm": 0.12962518632411957,
      "learning_rate": 0.00017199243615676597,
      "loss": 1.0545,
      "mean_token_accuracy": 0.7424853444099426,
      "num_tokens": 17155113.0,
      "step": 526
    },
    {
      "entropy": 1.0302038192749023,
      "epoch": 0.2818181818181818,
      "grad_norm": 0.13777880370616913,
      "learning_rate": 0.0001718792852922811,
      "loss": 1.0216,
      "mean_token_accuracy": 0.7463139593601227,
      "num_tokens": 17187808.0,
      "step": 527
    },
    {
      "entropy": 1.0302510261535645,
      "epoch": 0.2823529411764706,
      "grad_norm": 0.13522762060165405,
      "learning_rate": 0.00017176594862625274,
      "loss": 1.0233,
      "mean_token_accuracy": 0.7411107122898102,
      "num_tokens": 17220576.0,
      "step": 528
    },
    {
      "entropy": 1.0147441774606705,
      "epoch": 0.28288770053475937,
      "grad_norm": 0.1387215256690979,
      "learning_rate": 0.00017165242649899061,
      "loss": 1.0277,
      "mean_token_accuracy": 0.7405515313148499,
      "num_tokens": 17253033.0,
      "step": 529
    },
    {
      "entropy": 1.0255239754915237,
      "epoch": 0.28342245989304815,
      "grad_norm": 0.13352422416210175,
      "learning_rate": 0.00017153871925136154,
      "loss": 1.0302,
      "mean_token_accuracy": 0.7430898249149323,
      "num_tokens": 17285505.0,
      "step": 530
    },
    {
      "entropy": 1.0094399154186249,
      "epoch": 0.2839572192513369,
      "grad_norm": 0.12503620982170105,
      "learning_rate": 0.00017142482722478795,
      "loss": 0.9919,
      "mean_token_accuracy": 0.7527617067098618,
      "num_tokens": 17318210.0,
      "step": 531
    },
    {
      "entropy": 1.0132300108671188,
      "epoch": 0.28449197860962566,
      "grad_norm": 0.12262210994958878,
      "learning_rate": 0.00017131075076124724,
      "loss": 1.0109,
      "mean_token_accuracy": 0.7407647520303726,
      "num_tokens": 17350844.0,
      "step": 532
    },
    {
      "entropy": 0.9696976691484451,
      "epoch": 0.28502673796791445,
      "grad_norm": 0.12632109224796295,
      "learning_rate": 0.00017119649020327053,
      "loss": 0.9535,
      "mean_token_accuracy": 0.7574230134487152,
      "num_tokens": 17383612.0,
      "step": 533
    },
    {
      "entropy": 0.989849328994751,
      "epoch": 0.28556149732620323,
      "grad_norm": 0.12456396967172623,
      "learning_rate": 0.00017108204589394178,
      "loss": 0.9767,
      "mean_token_accuracy": 0.7528417408466339,
      "num_tokens": 17416080.0,
      "step": 534
    },
    {
      "entropy": 1.013772889971733,
      "epoch": 0.28609625668449196,
      "grad_norm": 0.12125220149755478,
      "learning_rate": 0.00017096741817689663,
      "loss": 1.0128,
      "mean_token_accuracy": 0.7480128556489944,
      "num_tokens": 17448632.0,
      "step": 535
    },
    {
      "entropy": 0.9724691212177277,
      "epoch": 0.28663101604278074,
      "grad_norm": 0.13034920394420624,
      "learning_rate": 0.00017085260739632148,
      "loss": 0.9904,
      "mean_token_accuracy": 0.7504276633262634,
      "num_tokens": 17481400.0,
      "step": 536
    },
    {
      "entropy": 0.9728181064128876,
      "epoch": 0.2871657754010695,
      "grad_norm": 0.12571309506893158,
      "learning_rate": 0.00017073761389695233,
      "loss": 0.9688,
      "mean_token_accuracy": 0.7547041773796082,
      "num_tokens": 17514069.0,
      "step": 537
    },
    {
      "entropy": 1.0064800530672073,
      "epoch": 0.2877005347593583,
      "grad_norm": 0.1352233588695526,
      "learning_rate": 0.00017062243802407393,
      "loss": 1.0153,
      "mean_token_accuracy": 0.7431587725877762,
      "num_tokens": 17546810.0,
      "step": 538
    },
    {
      "entropy": 0.9872759431600571,
      "epoch": 0.28823529411764703,
      "grad_norm": 0.13819102942943573,
      "learning_rate": 0.00017050708012351852,
      "loss": 1.0031,
      "mean_token_accuracy": 0.7485620528459549,
      "num_tokens": 17579265.0,
      "step": 539
    },
    {
      "entropy": 1.0200807750225067,
      "epoch": 0.2887700534759358,
      "grad_norm": 0.13746803998947144,
      "learning_rate": 0.000170391540541665,
      "loss": 1.0361,
      "mean_token_accuracy": 0.7412613034248352,
      "num_tokens": 17611772.0,
      "step": 540
    },
    {
      "entropy": 1.0252193808555603,
      "epoch": 0.2893048128342246,
      "grad_norm": 0.12763537466526031,
      "learning_rate": 0.00017027581962543768,
      "loss": 1.0328,
      "mean_token_accuracy": 0.742118775844574,
      "num_tokens": 17644540.0,
      "step": 541
    },
    {
      "entropy": 1.0053461492061615,
      "epoch": 0.2898395721925134,
      "grad_norm": 0.12436962127685547,
      "learning_rate": 0.00017015991772230545,
      "loss": 0.9979,
      "mean_token_accuracy": 0.742668628692627,
      "num_tokens": 17677308.0,
      "step": 542
    },
    {
      "entropy": 1.030677169561386,
      "epoch": 0.2903743315508021,
      "grad_norm": 0.13748657703399658,
      "learning_rate": 0.00017004383518028069,
      "loss": 1.0105,
      "mean_token_accuracy": 0.7427105903625488,
      "num_tokens": 17710034.0,
      "step": 543
    },
    {
      "entropy": 1.0635046660900116,
      "epoch": 0.2909090909090909,
      "grad_norm": 0.12991686165332794,
      "learning_rate": 0.000169927572347918,
      "loss": 1.0535,
      "mean_token_accuracy": 0.7350575774908066,
      "num_tokens": 17742638.0,
      "step": 544
    },
    {
      "entropy": 0.9904083460569382,
      "epoch": 0.2914438502673797,
      "grad_norm": 0.1312393695116043,
      "learning_rate": 0.00016981112957431345,
      "loss": 0.9696,
      "mean_token_accuracy": 0.7539711594581604,
      "num_tokens": 17775406.0,
      "step": 545
    },
    {
      "entropy": 1.080278903245926,
      "epoch": 0.29197860962566846,
      "grad_norm": 0.13253837823867798,
      "learning_rate": 0.0001696945072091034,
      "loss": 1.0676,
      "mean_token_accuracy": 0.735523447394371,
      "num_tokens": 17807650.0,
      "step": 546
    },
    {
      "entropy": 0.9899517744779587,
      "epoch": 0.29251336898395724,
      "grad_norm": 0.12947142124176025,
      "learning_rate": 0.00016957770560246344,
      "loss": 0.9899,
      "mean_token_accuracy": 0.75409334897995,
      "num_tokens": 17840418.0,
      "step": 547
    },
    {
      "entropy": 0.9981743395328522,
      "epoch": 0.293048128342246,
      "grad_norm": 0.1302729994058609,
      "learning_rate": 0.00016946072510510733,
      "loss": 1.0118,
      "mean_token_accuracy": 0.747682735323906,
      "num_tokens": 17873143.0,
      "step": 548
    },
    {
      "entropy": 0.9922465980052948,
      "epoch": 0.29358288770053476,
      "grad_norm": 0.13240595161914825,
      "learning_rate": 0.00016934356606828604,
      "loss": 1.0037,
      "mean_token_accuracy": 0.7481169998645782,
      "num_tokens": 17905764.0,
      "step": 549
    },
    {
      "entropy": 1.0052950233221054,
      "epoch": 0.29411764705882354,
      "grad_norm": 0.13417880237102509,
      "learning_rate": 0.0001692262288437866,
      "loss": 1.0208,
      "mean_token_accuracy": 0.7440185844898224,
      "num_tokens": 17938223.0,
      "step": 550
    },
    {
      "entropy": 1.037492722272873,
      "epoch": 0.2946524064171123,
      "grad_norm": 0.23050248622894287,
      "learning_rate": 0.00016910871378393108,
      "loss": 1.0561,
      "mean_token_accuracy": 0.7388520389795303,
      "num_tokens": 17970901.0,
      "step": 551
    },
    {
      "entropy": 1.0313400328159332,
      "epoch": 0.29518716577540105,
      "grad_norm": 0.13766978681087494,
      "learning_rate": 0.00016899102124157548,
      "loss": 1.0446,
      "mean_token_accuracy": 0.7397971600294113,
      "num_tokens": 18003669.0,
      "step": 552
    },
    {
      "entropy": 1.0375900715589523,
      "epoch": 0.29572192513368983,
      "grad_norm": 0.13091754913330078,
      "learning_rate": 0.00016887315157010878,
      "loss": 1.038,
      "mean_token_accuracy": 0.7395756989717484,
      "num_tokens": 18036436.0,
      "step": 553
    },
    {
      "entropy": 1.039006382226944,
      "epoch": 0.2962566844919786,
      "grad_norm": 0.1290474236011505,
      "learning_rate": 0.0001687551051234518,
      "loss": 1.0344,
      "mean_token_accuracy": 0.7386363744735718,
      "num_tokens": 18069204.0,
      "step": 554
    },
    {
      "entropy": 0.9995266646146774,
      "epoch": 0.2967914438502674,
      "grad_norm": 0.13149632513523102,
      "learning_rate": 0.00016863688225605614,
      "loss": 0.9852,
      "mean_token_accuracy": 0.7554068863391876,
      "num_tokens": 18101972.0,
      "step": 555
    },
    {
      "entropy": 1.0221814066171646,
      "epoch": 0.29732620320855613,
      "grad_norm": 0.15010221302509308,
      "learning_rate": 0.00016851848332290312,
      "loss": 1.0073,
      "mean_token_accuracy": 0.7423326075077057,
      "num_tokens": 18134740.0,
      "step": 556
    },
    {
      "entropy": 1.008228451013565,
      "epoch": 0.2978609625668449,
      "grad_norm": 0.1910933405160904,
      "learning_rate": 0.00016839990867950275,
      "loss": 1.0087,
      "mean_token_accuracy": 0.7485948204994202,
      "num_tokens": 18167508.0,
      "step": 557
    },
    {
      "entropy": 0.9887708127498627,
      "epoch": 0.2983957219251337,
      "grad_norm": 0.23146788775920868,
      "learning_rate": 0.0001682811586818926,
      "loss": 0.9947,
      "mean_token_accuracy": 0.7519334852695465,
      "num_tokens": 18200202.0,
      "step": 558
    },
    {
      "entropy": 1.0437268614768982,
      "epoch": 0.2989304812834225,
      "grad_norm": 0.222130686044693,
      "learning_rate": 0.00016816223368663682,
      "loss": 1.0384,
      "mean_token_accuracy": 0.7401331961154938,
      "num_tokens": 18232970.0,
      "step": 559
    },
    {
      "entropy": 1.046816885471344,
      "epoch": 0.2994652406417112,
      "grad_norm": 0.163495272397995,
      "learning_rate": 0.00016804313405082498,
      "loss": 1.0507,
      "mean_token_accuracy": 0.7394961565732956,
      "num_tokens": 18265492.0,
      "step": 560
    },
    {
      "entropy": 1.0264595448970795,
      "epoch": 0.3,
      "grad_norm": 0.1313883364200592,
      "learning_rate": 0.000167923860132071,
      "loss": 1.0323,
      "mean_token_accuracy": 0.7444403767585754,
      "num_tokens": 18298260.0,
      "step": 561
    },
    {
      "entropy": 1.0883614718914032,
      "epoch": 0.30053475935828877,
      "grad_norm": 0.16005565226078033,
      "learning_rate": 0.00016780441228851224,
      "loss": 1.0815,
      "mean_token_accuracy": 0.7313090562820435,
      "num_tokens": 18330777.0,
      "step": 562
    },
    {
      "entropy": 1.0424848198890686,
      "epoch": 0.30106951871657756,
      "grad_norm": 0.14987064898014069,
      "learning_rate": 0.00016768479087880814,
      "loss": 1.034,
      "mean_token_accuracy": 0.7385043501853943,
      "num_tokens": 18363464.0,
      "step": 563
    },
    {
      "entropy": 1.0108812600374222,
      "epoch": 0.3016042780748663,
      "grad_norm": 0.13205383718013763,
      "learning_rate": 0.00016756499626213934,
      "loss": 0.9954,
      "mean_token_accuracy": 0.7490835785865784,
      "num_tokens": 18396232.0,
      "step": 564
    },
    {
      "entropy": 1.0051104724407196,
      "epoch": 0.30213903743315507,
      "grad_norm": 0.1460045427083969,
      "learning_rate": 0.00016744502879820658,
      "loss": 0.9975,
      "mean_token_accuracy": 0.7545073330402374,
      "num_tokens": 18428843.0,
      "step": 565
    },
    {
      "entropy": 1.0304394662380219,
      "epoch": 0.30267379679144385,
      "grad_norm": 0.13122116029262543,
      "learning_rate": 0.00016732488884722962,
      "loss": 1.028,
      "mean_token_accuracy": 0.7366521507501602,
      "num_tokens": 18461495.0,
      "step": 566
    },
    {
      "entropy": 0.990816205739975,
      "epoch": 0.30320855614973263,
      "grad_norm": 0.13677483797073364,
      "learning_rate": 0.00016720457676994616,
      "loss": 0.9934,
      "mean_token_accuracy": 0.749927431344986,
      "num_tokens": 18494227.0,
      "step": 567
    },
    {
      "entropy": 1.0301152765750885,
      "epoch": 0.3037433155080214,
      "grad_norm": 0.13701874017715454,
      "learning_rate": 0.0001670840929276106,
      "loss": 1.0299,
      "mean_token_accuracy": 0.7401285469532013,
      "num_tokens": 18526964.0,
      "step": 568
    },
    {
      "entropy": 0.9530722349882126,
      "epoch": 0.30427807486631014,
      "grad_norm": 0.1365976631641388,
      "learning_rate": 0.00016696343768199328,
      "loss": 0.9612,
      "mean_token_accuracy": 0.7563553899526596,
      "num_tokens": 18559139.0,
      "step": 569
    },
    {
      "entropy": 0.97348852455616,
      "epoch": 0.3048128342245989,
      "grad_norm": 0.150132954120636,
      "learning_rate": 0.000166842611395379,
      "loss": 0.9953,
      "mean_token_accuracy": 0.7487475574016571,
      "num_tokens": 18591907.0,
      "step": 570
    },
    {
      "entropy": 1.022206872701645,
      "epoch": 0.3053475935828877,
      "grad_norm": 0.133019357919693,
      "learning_rate": 0.00016672161443056634,
      "loss": 1.0462,
      "mean_token_accuracy": 0.7444709241390228,
      "num_tokens": 18624675.0,
      "step": 571
    },
    {
      "entropy": 1.0235802233219147,
      "epoch": 0.3058823529411765,
      "grad_norm": 0.14455930888652802,
      "learning_rate": 0.00016660044715086615,
      "loss": 1.0421,
      "mean_token_accuracy": 0.7430094182491302,
      "num_tokens": 18656296.0,
      "step": 572
    },
    {
      "entropy": 1.021704226732254,
      "epoch": 0.3064171122994652,
      "grad_norm": 0.13559989631175995,
      "learning_rate": 0.00016647910992010088,
      "loss": 1.0196,
      "mean_token_accuracy": 0.7471896409988403,
      "num_tokens": 18689064.0,
      "step": 573
    },
    {
      "entropy": 1.0069758594036102,
      "epoch": 0.306951871657754,
      "grad_norm": 0.1325547844171524,
      "learning_rate": 0.00016635760310260318,
      "loss": 1.0036,
      "mean_token_accuracy": 0.7480007261037827,
      "num_tokens": 18721775.0,
      "step": 574
    },
    {
      "entropy": 1.0201173424720764,
      "epoch": 0.3074866310160428,
      "grad_norm": 0.14399488270282745,
      "learning_rate": 0.00016623592706321491,
      "loss": 1.0093,
      "mean_token_accuracy": 0.7490224838256836,
      "num_tokens": 18754543.0,
      "step": 575
    },
    {
      "entropy": 1.0243518650531769,
      "epoch": 0.30802139037433157,
      "grad_norm": 0.12975279986858368,
      "learning_rate": 0.00016611408216728603,
      "loss": 1.0159,
      "mean_token_accuracy": 0.7417757958173752,
      "num_tokens": 18787081.0,
      "step": 576
    },
    {
      "entropy": 0.952204629778862,
      "epoch": 0.3085561497326203,
      "grad_norm": 0.13181157410144806,
      "learning_rate": 0.00016599206878067355,
      "loss": 0.9344,
      "mean_token_accuracy": 0.7643770426511765,
      "num_tokens": 18819834.0,
      "step": 577
    },
    {
      "entropy": 0.9844817519187927,
      "epoch": 0.3090909090909091,
      "grad_norm": 0.1284974217414856,
      "learning_rate": 0.00016586988726974042,
      "loss": 0.9761,
      "mean_token_accuracy": 0.7529546320438385,
      "num_tokens": 18852582.0,
      "step": 578
    },
    {
      "entropy": 1.0278932452201843,
      "epoch": 0.30962566844919787,
      "grad_norm": 0.1554325520992279,
      "learning_rate": 0.00016574753800135437,
      "loss": 1.0216,
      "mean_token_accuracy": 0.7442286312580109,
      "num_tokens": 18885047.0,
      "step": 579
    },
    {
      "entropy": 1.012410581111908,
      "epoch": 0.31016042780748665,
      "grad_norm": 0.12814141809940338,
      "learning_rate": 0.00016562502134288685,
      "loss": 1.0062,
      "mean_token_accuracy": 0.7505649775266647,
      "num_tokens": 18917677.0,
      "step": 580
    },
    {
      "entropy": 1.0278237909078598,
      "epoch": 0.3106951871657754,
      "grad_norm": 0.12905891239643097,
      "learning_rate": 0.00016550233766221185,
      "loss": 1.0384,
      "mean_token_accuracy": 0.7401637434959412,
      "num_tokens": 18950445.0,
      "step": 581
    },
    {
      "entropy": 1.0323994159698486,
      "epoch": 0.31122994652406416,
      "grad_norm": 0.13156746327877045,
      "learning_rate": 0.00016537948732770508,
      "loss": 1.0529,
      "mean_token_accuracy": 0.7383771538734436,
      "num_tokens": 18983096.0,
      "step": 582
    },
    {
      "entropy": 1.0405890196561813,
      "epoch": 0.31176470588235294,
      "grad_norm": 0.13030070066452026,
      "learning_rate": 0.0001652564707082424,
      "loss": 1.0401,
      "mean_token_accuracy": 0.7420965284109116,
      "num_tokens": 19015394.0,
      "step": 583
    },
    {
      "entropy": 1.0359513461589813,
      "epoch": 0.3122994652406417,
      "grad_norm": 0.12979918718338013,
      "learning_rate": 0.00016513328817319915,
      "loss": 1.0324,
      "mean_token_accuracy": 0.7427784353494644,
      "num_tokens": 19048014.0,
      "step": 584
    },
    {
      "entropy": 1.0870505571365356,
      "epoch": 0.31283422459893045,
      "grad_norm": 0.13175484538078308,
      "learning_rate": 0.00016500994009244877,
      "loss": 1.0817,
      "mean_token_accuracy": 0.7303274869918823,
      "num_tokens": 19080782.0,
      "step": 585
    },
    {
      "entropy": 1.0574510097503662,
      "epoch": 0.31336898395721924,
      "grad_norm": 0.12824955582618713,
      "learning_rate": 0.00016488642683636174,
      "loss": 1.0582,
      "mean_token_accuracy": 0.7358871102333069,
      "num_tokens": 19113550.0,
      "step": 586
    },
    {
      "entropy": 1.059159204363823,
      "epoch": 0.313903743315508,
      "grad_norm": 0.13426367938518524,
      "learning_rate": 0.00016476274877580462,
      "loss": 1.0562,
      "mean_token_accuracy": 0.7356574982404709,
      "num_tokens": 19145703.0,
      "step": 587
    },
    {
      "entropy": 1.0259163677692413,
      "epoch": 0.3144385026737968,
      "grad_norm": 0.13646824657917023,
      "learning_rate": 0.00016463890628213874,
      "loss": 1.0073,
      "mean_token_accuracy": 0.74399633705616,
      "num_tokens": 19178333.0,
      "step": 588
    },
    {
      "entropy": 1.020667016506195,
      "epoch": 0.3149732620320856,
      "grad_norm": 0.12998659908771515,
      "learning_rate": 0.00016451489972721915,
      "loss": 1.007,
      "mean_token_accuracy": 0.7440432608127594,
      "num_tokens": 19211101.0,
      "step": 589
    },
    {
      "entropy": 1.0039927810430527,
      "epoch": 0.3155080213903743,
      "grad_norm": 0.13191555440425873,
      "learning_rate": 0.00016439072948339358,
      "loss": 1.0113,
      "mean_token_accuracy": 0.7398888170719147,
      "num_tokens": 19243869.0,
      "step": 590
    },
    {
      "entropy": 1.0305038839578629,
      "epoch": 0.3160427807486631,
      "grad_norm": 0.13551065325737,
      "learning_rate": 0.00016426639592350122,
      "loss": 1.0333,
      "mean_token_accuracy": 0.7388502061367035,
      "num_tokens": 19276637.0,
      "step": 591
    },
    {
      "entropy": 0.9789949357509613,
      "epoch": 0.3165775401069519,
      "grad_norm": 0.1322057992219925,
      "learning_rate": 0.00016414189942087163,
      "loss": 0.9724,
      "mean_token_accuracy": 0.7544249892234802,
      "num_tokens": 19308982.0,
      "step": 592
    },
    {
      "entropy": 1.0381294786930084,
      "epoch": 0.31711229946524067,
      "grad_norm": 0.13753201067447662,
      "learning_rate": 0.0001640172403493237,
      "loss": 1.0481,
      "mean_token_accuracy": 0.7402539104223251,
      "num_tokens": 19341114.0,
      "step": 593
    },
    {
      "entropy": 1.008999764919281,
      "epoch": 0.3176470588235294,
      "grad_norm": 0.148467019200325,
      "learning_rate": 0.0001638924190831643,
      "loss": 1.0246,
      "mean_token_accuracy": 0.7437377870082855,
      "num_tokens": 19373882.0,
      "step": 594
    },
    {
      "entropy": 0.9896555691957474,
      "epoch": 0.3181818181818182,
      "grad_norm": 0.1336461752653122,
      "learning_rate": 0.00016376743599718748,
      "loss": 0.9853,
      "mean_token_accuracy": 0.7501374632120132,
      "num_tokens": 19406386.0,
      "step": 595
    },
    {
      "entropy": 1.028982162475586,
      "epoch": 0.31871657754010696,
      "grad_norm": 0.13734780251979828,
      "learning_rate": 0.00016364229146667312,
      "loss": 1.0014,
      "mean_token_accuracy": 0.7460211962461472,
      "num_tokens": 19439068.0,
      "step": 596
    },
    {
      "entropy": 1.0380202233791351,
      "epoch": 0.31925133689839574,
      "grad_norm": 0.15408451855182648,
      "learning_rate": 0.00016351698586738583,
      "loss": 1.0402,
      "mean_token_accuracy": 0.7399239391088486,
      "num_tokens": 19471403.0,
      "step": 597
    },
    {
      "entropy": 0.9874198585748672,
      "epoch": 0.31978609625668447,
      "grad_norm": 0.14568470418453217,
      "learning_rate": 0.00016339151957557384,
      "loss": 0.9877,
      "mean_token_accuracy": 0.7511913478374481,
      "num_tokens": 19504171.0,
      "step": 598
    },
    {
      "entropy": 1.002236232161522,
      "epoch": 0.32032085561497325,
      "grad_norm": 0.13177983462810516,
      "learning_rate": 0.00016326589296796802,
      "loss": 1.0073,
      "mean_token_accuracy": 0.7449902296066284,
      "num_tokens": 19536939.0,
      "step": 599
    },
    {
      "entropy": 0.982133075594902,
      "epoch": 0.32085561497326204,
      "grad_norm": 0.14566829800605774,
      "learning_rate": 0.00016314010642178042,
      "loss": 0.9991,
      "mean_token_accuracy": 0.751496821641922,
      "num_tokens": 19569707.0,
      "step": 600
    },
    {
      "entropy": 1.0116736739873886,
      "epoch": 0.3213903743315508,
      "grad_norm": 0.15575391054153442,
      "learning_rate": 0.00016301416031470342,
      "loss": 1.0168,
      "mean_token_accuracy": 0.7445014715194702,
      "num_tokens": 19602475.0,
      "step": 601
    },
    {
      "entropy": 1.0237414687871933,
      "epoch": 0.32192513368983955,
      "grad_norm": 0.1417372077703476,
      "learning_rate": 0.00016288805502490857,
      "loss": 1.024,
      "mean_token_accuracy": 0.7416205853223801,
      "num_tokens": 19635118.0,
      "step": 602
    },
    {
      "entropy": 1.002694696187973,
      "epoch": 0.32245989304812833,
      "grad_norm": 0.13840502500534058,
      "learning_rate": 0.00016276179093104528,
      "loss": 1.0029,
      "mean_token_accuracy": 0.7493388056755066,
      "num_tokens": 19667570.0,
      "step": 603
    },
    {
      "entropy": 1.0062010288238525,
      "epoch": 0.3229946524064171,
      "grad_norm": 0.1401420682668686,
      "learning_rate": 0.00016263536841223982,
      "loss": 1.0024,
      "mean_token_accuracy": 0.7453262507915497,
      "num_tokens": 19700338.0,
      "step": 604
    },
    {
      "entropy": 1.0327490121126175,
      "epoch": 0.3235294117647059,
      "grad_norm": 0.1464383602142334,
      "learning_rate": 0.0001625087878480942,
      "loss": 1.0259,
      "mean_token_accuracy": 0.7398741096258163,
      "num_tokens": 19731972.0,
      "step": 605
    },
    {
      "entropy": 0.9721281826496124,
      "epoch": 0.3240641711229946,
      "grad_norm": 0.14435307681560516,
      "learning_rate": 0.00016238204961868493,
      "loss": 0.9685,
      "mean_token_accuracy": 0.7569648027420044,
      "num_tokens": 19764740.0,
      "step": 606
    },
    {
      "entropy": 1.028033286333084,
      "epoch": 0.3245989304812834,
      "grad_norm": 0.13649606704711914,
      "learning_rate": 0.00016225515410456197,
      "loss": 1.0172,
      "mean_token_accuracy": 0.7414901256561279,
      "num_tokens": 19797454.0,
      "step": 607
    },
    {
      "entropy": 1.0394579023122787,
      "epoch": 0.3251336898395722,
      "grad_norm": 0.12738116085529327,
      "learning_rate": 0.00016212810168674753,
      "loss": 1.0294,
      "mean_token_accuracy": 0.7402873635292053,
      "num_tokens": 19830177.0,
      "step": 608
    },
    {
      "entropy": 0.9974191337823868,
      "epoch": 0.325668449197861,
      "grad_norm": 0.13835933804512024,
      "learning_rate": 0.00016200089274673492,
      "loss": 0.984,
      "mean_token_accuracy": 0.748787984251976,
      "num_tokens": 19862661.0,
      "step": 609
    },
    {
      "entropy": 0.9926651865243912,
      "epoch": 0.32620320855614976,
      "grad_norm": 0.1357496827840805,
      "learning_rate": 0.00016187352766648753,
      "loss": 0.9856,
      "mean_token_accuracy": 0.752595990896225,
      "num_tokens": 19895402.0,
      "step": 610
    },
    {
      "entropy": 0.9659290164709091,
      "epoch": 0.3267379679144385,
      "grad_norm": 0.1416723132133484,
      "learning_rate": 0.00016174600682843747,
      "loss": 0.9723,
      "mean_token_accuracy": 0.7547358125448227,
      "num_tokens": 19927869.0,
      "step": 611
    },
    {
      "entropy": 1.0146066099405289,
      "epoch": 0.32727272727272727,
      "grad_norm": 0.15567292273044586,
      "learning_rate": 0.00016161833061548463,
      "loss": 1.024,
      "mean_token_accuracy": 0.7476064115762711,
      "num_tokens": 19960378.0,
      "step": 612
    },
    {
      "entropy": 0.9735582172870636,
      "epoch": 0.32780748663101605,
      "grad_norm": 0.16228708624839783,
      "learning_rate": 0.00016149049941099528,
      "loss": 0.9995,
      "mean_token_accuracy": 0.7474645674228668,
      "num_tokens": 19993146.0,
      "step": 613
    },
    {
      "entropy": 0.9890165776014328,
      "epoch": 0.32834224598930484,
      "grad_norm": 0.14233818650245667,
      "learning_rate": 0.00016136251359880128,
      "loss": 1.0019,
      "mean_token_accuracy": 0.7476700097322464,
      "num_tokens": 20025356.0,
      "step": 614
    },
    {
      "entropy": 1.0183548629283905,
      "epoch": 0.32887700534759357,
      "grad_norm": 0.1370004415512085,
      "learning_rate": 0.00016123437356319859,
      "loss": 1.0333,
      "mean_token_accuracy": 0.7452346086502075,
      "num_tokens": 20058124.0,
      "step": 615
    },
    {
      "entropy": 0.9772834926843643,
      "epoch": 0.32941176470588235,
      "grad_norm": 0.13862505555152893,
      "learning_rate": 0.0001611060796889463,
      "loss": 0.9591,
      "mean_token_accuracy": 0.7551965862512589,
      "num_tokens": 20090631.0,
      "step": 616
    },
    {
      "entropy": 1.0807088017463684,
      "epoch": 0.32994652406417113,
      "grad_norm": 0.13603933155536652,
      "learning_rate": 0.0001609776323612654,
      "loss": 1.0786,
      "mean_token_accuracy": 0.7374144792556763,
      "num_tokens": 20123399.0,
      "step": 617
    },
    {
      "entropy": 1.0449609160423279,
      "epoch": 0.3304812834224599,
      "grad_norm": 0.13282941281795502,
      "learning_rate": 0.0001608490319658376,
      "loss": 1.0362,
      "mean_token_accuracy": 0.7421600371599197,
      "num_tokens": 20155570.0,
      "step": 618
    },
    {
      "entropy": 1.0110503286123276,
      "epoch": 0.33101604278074864,
      "grad_norm": 0.13819116353988647,
      "learning_rate": 0.00016072027888880438,
      "loss": 0.9944,
      "mean_token_accuracy": 0.750076025724411,
      "num_tokens": 20187984.0,
      "step": 619
    },
    {
      "entropy": 1.037804901599884,
      "epoch": 0.3315508021390374,
      "grad_norm": 0.13222205638885498,
      "learning_rate": 0.00016059137351676542,
      "loss": 1.0153,
      "mean_token_accuracy": 0.7441959977149963,
      "num_tokens": 20220752.0,
      "step": 620
    },
    {
      "entropy": 1.0193375647068024,
      "epoch": 0.3320855614973262,
      "grad_norm": 0.13431870937347412,
      "learning_rate": 0.00016046231623677797,
      "loss": 1.0021,
      "mean_token_accuracy": 0.7499389052391052,
      "num_tokens": 20253520.0,
      "step": 621
    },
    {
      "entropy": 1.0201116502285004,
      "epoch": 0.332620320855615,
      "grad_norm": 0.13786272704601288,
      "learning_rate": 0.00016033310743635514,
      "loss": 1.0337,
      "mean_token_accuracy": 0.7389914989471436,
      "num_tokens": 20286123.0,
      "step": 622
    },
    {
      "entropy": 1.0212906897068024,
      "epoch": 0.3331550802139037,
      "grad_norm": 0.1344262808561325,
      "learning_rate": 0.0001602037475034652,
      "loss": 1.0276,
      "mean_token_accuracy": 0.7460288405418396,
      "num_tokens": 20318891.0,
      "step": 623
    },
    {
      "entropy": 0.9826373159885406,
      "epoch": 0.3336898395721925,
      "grad_norm": 0.1419702172279358,
      "learning_rate": 0.00016007423682653008,
      "loss": 0.9947,
      "mean_token_accuracy": 0.7482587993144989,
      "num_tokens": 20351659.0,
      "step": 624
    },
    {
      "entropy": 1.0362474918365479,
      "epoch": 0.3342245989304813,
      "grad_norm": 0.14024557173252106,
      "learning_rate": 0.00015994457579442443,
      "loss": 1.0406,
      "mean_token_accuracy": 0.7413856387138367,
      "num_tokens": 20384427.0,
      "step": 625
    },
    {
      "entropy": 0.9735687077045441,
      "epoch": 0.33475935828877007,
      "grad_norm": 0.13640111684799194,
      "learning_rate": 0.00015981476479647437,
      "loss": 0.9725,
      "mean_token_accuracy": 0.755733072757721,
      "num_tokens": 20417153.0,
      "step": 626
    },
    {
      "entropy": 1.0228614211082458,
      "epoch": 0.3352941176470588,
      "grad_norm": 0.14213134348392487,
      "learning_rate": 0.00015968480422245626,
      "loss": 1.0324,
      "mean_token_accuracy": 0.7432184815406799,
      "num_tokens": 20449921.0,
      "step": 627
    },
    {
      "entropy": 1.0295179784297943,
      "epoch": 0.3358288770053476,
      "grad_norm": 0.1468447744846344,
      "learning_rate": 0.00015955469446259557,
      "loss": 1.0245,
      "mean_token_accuracy": 0.7425564378499985,
      "num_tokens": 20482396.0,
      "step": 628
    },
    {
      "entropy": 1.0403719544410706,
      "epoch": 0.33636363636363636,
      "grad_norm": 0.12881560623645782,
      "learning_rate": 0.0001594244359075658,
      "loss": 1.0434,
      "mean_token_accuracy": 0.7414287030696869,
      "num_tokens": 20515037.0,
      "step": 629
    },
    {
      "entropy": 0.9976829290390015,
      "epoch": 0.33689839572192515,
      "grad_norm": 0.15784220397472382,
      "learning_rate": 0.00015929402894848714,
      "loss": 0.9934,
      "mean_token_accuracy": 0.7500075548887253,
      "num_tokens": 20547082.0,
      "step": 630
    },
    {
      "entropy": 1.0174478143453598,
      "epoch": 0.33743315508021393,
      "grad_norm": 0.16386115550994873,
      "learning_rate": 0.00015916347397692548,
      "loss": 1.0229,
      "mean_token_accuracy": 0.7399440705776215,
      "num_tokens": 20579826.0,
      "step": 631
    },
    {
      "entropy": 0.9906077682971954,
      "epoch": 0.33796791443850266,
      "grad_norm": 0.13278624415397644,
      "learning_rate": 0.00015903277138489105,
      "loss": 0.9912,
      "mean_token_accuracy": 0.7534213066101074,
      "num_tokens": 20612594.0,
      "step": 632
    },
    {
      "entropy": 1.0336619019508362,
      "epoch": 0.33850267379679144,
      "grad_norm": 0.16966547071933746,
      "learning_rate": 0.00015890192156483738,
      "loss": 1.0395,
      "mean_token_accuracy": 0.7384225428104401,
      "num_tokens": 20645362.0,
      "step": 633
    },
    {
      "entropy": 1.0511770248413086,
      "epoch": 0.3390374331550802,
      "grad_norm": 0.16399900615215302,
      "learning_rate": 0.00015877092490966006,
      "loss": 1.0482,
      "mean_token_accuracy": 0.7364983707666397,
      "num_tokens": 20677920.0,
      "step": 634
    },
    {
      "entropy": 1.0212816298007965,
      "epoch": 0.339572192513369,
      "grad_norm": 0.13357484340667725,
      "learning_rate": 0.00015863978181269558,
      "loss": 1.009,
      "mean_token_accuracy": 0.7460338771343231,
      "num_tokens": 20710466.0,
      "step": 635
    },
    {
      "entropy": 1.0000340044498444,
      "epoch": 0.34010695187165774,
      "grad_norm": 0.15471495687961578,
      "learning_rate": 0.00015850849266772017,
      "loss": 0.9948,
      "mean_token_accuracy": 0.7509164214134216,
      "num_tokens": 20743234.0,
      "step": 636
    },
    {
      "entropy": 1.0232605040073395,
      "epoch": 0.3406417112299465,
      "grad_norm": 0.1373874694108963,
      "learning_rate": 0.0001583770578689485,
      "loss": 1.0167,
      "mean_token_accuracy": 0.7430351972579956,
      "num_tokens": 20776002.0,
      "step": 637
    },
    {
      "entropy": 1.0646322965621948,
      "epoch": 0.3411764705882353,
      "grad_norm": 0.14697957038879395,
      "learning_rate": 0.00015824547781103268,
      "loss": 1.072,
      "mean_token_accuracy": 0.7312438935041428,
      "num_tokens": 20808770.0,
      "step": 638
    },
    {
      "entropy": 1.0304863154888153,
      "epoch": 0.3417112299465241,
      "grad_norm": 0.16963070631027222,
      "learning_rate": 0.00015811375288906097,
      "loss": 1.036,
      "mean_token_accuracy": 0.7408357858657837,
      "num_tokens": 20841538.0,
      "step": 639
    },
    {
      "entropy": 0.9575765281915665,
      "epoch": 0.3422459893048128,
      "grad_norm": 0.14465726912021637,
      "learning_rate": 0.00015798188349855659,
      "loss": 0.954,
      "mean_token_accuracy": 0.7573956996202469,
      "num_tokens": 20874027.0,
      "step": 640
    },
    {
      "entropy": 0.9817063212394714,
      "epoch": 0.3427807486631016,
      "grad_norm": 0.13434670865535736,
      "learning_rate": 0.0001578498700354765,
      "loss": 0.9651,
      "mean_token_accuracy": 0.7546056061983109,
      "num_tokens": 20906619.0,
      "step": 641
    },
    {
      "entropy": 1.0032513439655304,
      "epoch": 0.3433155080213904,
      "grad_norm": 0.15078270435333252,
      "learning_rate": 0.00015771771289621028,
      "loss": 0.997,
      "mean_token_accuracy": 0.7474170327186584,
      "num_tokens": 20938677.0,
      "step": 642
    },
    {
      "entropy": 1.0587093532085419,
      "epoch": 0.34385026737967916,
      "grad_norm": 0.1376420557498932,
      "learning_rate": 0.00015758541247757902,
      "loss": 1.0776,
      "mean_token_accuracy": 0.7357954680919647,
      "num_tokens": 20971445.0,
      "step": 643
    },
    {
      "entropy": 1.0180152505636215,
      "epoch": 0.3443850267379679,
      "grad_norm": 0.147924542427063,
      "learning_rate": 0.00015745296917683388,
      "loss": 1.0215,
      "mean_token_accuracy": 0.7416360229253769,
      "num_tokens": 21003936.0,
      "step": 644
    },
    {
      "entropy": 0.9809147417545319,
      "epoch": 0.3449197860962567,
      "grad_norm": 0.14757901430130005,
      "learning_rate": 0.0001573203833916551,
      "loss": 0.9837,
      "mean_token_accuracy": 0.7514662742614746,
      "num_tokens": 21036704.0,
      "step": 645
    },
    {
      "entropy": 1.0416185408830643,
      "epoch": 0.34545454545454546,
      "grad_norm": 0.14264920353889465,
      "learning_rate": 0.00015718765552015087,
      "loss": 1.0426,
      "mean_token_accuracy": 0.7363147735595703,
      "num_tokens": 21069472.0,
      "step": 646
    },
    {
      "entropy": 0.9874245077371597,
      "epoch": 0.34598930481283424,
      "grad_norm": 0.14516086876392365,
      "learning_rate": 0.00015705478596085578,
      "loss": 0.9906,
      "mean_token_accuracy": 0.7513746321201324,
      "num_tokens": 21102240.0,
      "step": 647
    },
    {
      "entropy": 1.0033420026302338,
      "epoch": 0.34652406417112297,
      "grad_norm": 0.12710031867027283,
      "learning_rate": 0.00015692177511273005,
      "loss": 1.0026,
      "mean_token_accuracy": 0.7465786933898926,
      "num_tokens": 21135008.0,
      "step": 648
    },
    {
      "entropy": 1.0446359515190125,
      "epoch": 0.34705882352941175,
      "grad_norm": 0.13629251718521118,
      "learning_rate": 0.00015678862337515807,
      "loss": 1.0441,
      "mean_token_accuracy": 0.7418743968009949,
      "num_tokens": 21167776.0,
      "step": 649
    },
    {
      "entropy": 0.9714336842298508,
      "epoch": 0.34759358288770054,
      "grad_norm": 0.1313450187444687,
      "learning_rate": 0.0001566553311479473,
      "loss": 0.9537,
      "mean_token_accuracy": 0.7604472041130066,
      "num_tokens": 21200544.0,
      "step": 650
    },
    {
      "entropy": 1.0382590889930725,
      "epoch": 0.3481283422459893,
      "grad_norm": 0.141065776348114,
      "learning_rate": 0.0001565218988313269,
      "loss": 1.0306,
      "mean_token_accuracy": 0.7439709454774857,
      "num_tokens": 21233284.0,
      "step": 651
    },
    {
      "entropy": 1.055101215839386,
      "epoch": 0.34866310160427805,
      "grad_norm": 0.1360592395067215,
      "learning_rate": 0.00015638832682594693,
      "loss": 1.065,
      "mean_token_accuracy": 0.7342474460601807,
      "num_tokens": 21265631.0,
      "step": 652
    },
    {
      "entropy": 0.9873165488243103,
      "epoch": 0.34919786096256683,
      "grad_norm": 0.13543623685836792,
      "learning_rate": 0.0001562546155328766,
      "loss": 0.9769,
      "mean_token_accuracy": 0.7527990341186523,
      "num_tokens": 21297750.0,
      "step": 653
    },
    {
      "entropy": 1.0083692371845245,
      "epoch": 0.3497326203208556,
      "grad_norm": 0.13810250163078308,
      "learning_rate": 0.00015612076535360364,
      "loss": 1.0123,
      "mean_token_accuracy": 0.7467314302921295,
      "num_tokens": 21330518.0,
      "step": 654
    },
    {
      "entropy": 0.9826506972312927,
      "epoch": 0.3502673796791444,
      "grad_norm": 0.1295788437128067,
      "learning_rate": 0.00015598677669003262,
      "loss": 0.9971,
      "mean_token_accuracy": 0.7486559152603149,
      "num_tokens": 21363286.0,
      "step": 655
    },
    {
      "entropy": 1.007062405347824,
      "epoch": 0.3508021390374332,
      "grad_norm": 0.14045320451259613,
      "learning_rate": 0.00015585264994448386,
      "loss": 1.0114,
      "mean_token_accuracy": 0.7455706298351288,
      "num_tokens": 21396054.0,
      "step": 656
    },
    {
      "entropy": 0.9903112798929214,
      "epoch": 0.3513368983957219,
      "grad_norm": 0.13183598220348358,
      "learning_rate": 0.00015571838551969248,
      "loss": 0.9828,
      "mean_token_accuracy": 0.7507942318916321,
      "num_tokens": 21428822.0,
      "step": 657
    },
    {
      "entropy": 0.9780847728252411,
      "epoch": 0.3518716577540107,
      "grad_norm": 0.13663578033447266,
      "learning_rate": 0.00015558398381880693,
      "loss": 0.9698,
      "mean_token_accuracy": 0.7527776211500168,
      "num_tokens": 21460828.0,
      "step": 658
    },
    {
      "entropy": 1.0236111879348755,
      "epoch": 0.3524064171122995,
      "grad_norm": 0.14290697872638702,
      "learning_rate": 0.00015544944524538784,
      "loss": 1.028,
      "mean_token_accuracy": 0.7419543862342834,
      "num_tokens": 21493501.0,
      "step": 659
    },
    {
      "entropy": 1.0137494504451752,
      "epoch": 0.35294117647058826,
      "grad_norm": 0.13214318454265594,
      "learning_rate": 0.0001553147702034068,
      "loss": 1.0105,
      "mean_token_accuracy": 0.7438588291406631,
      "num_tokens": 21525788.0,
      "step": 660
    },
    {
      "entropy": 1.0354048311710358,
      "epoch": 0.353475935828877,
      "grad_norm": 0.13967375457286835,
      "learning_rate": 0.0001551799590972452,
      "loss": 1.0365,
      "mean_token_accuracy": 0.7424995005130768,
      "num_tokens": 21558272.0,
      "step": 661
    },
    {
      "entropy": 0.9711843430995941,
      "epoch": 0.35401069518716577,
      "grad_norm": 0.13694122433662415,
      "learning_rate": 0.00015504501233169296,
      "loss": 0.9785,
      "mean_token_accuracy": 0.7561705857515335,
      "num_tokens": 21591040.0,
      "step": 662
    },
    {
      "entropy": 1.0343670547008514,
      "epoch": 0.35454545454545455,
      "grad_norm": 0.13821734488010406,
      "learning_rate": 0.00015490993031194735,
      "loss": 1.0318,
      "mean_token_accuracy": 0.7411490827798843,
      "num_tokens": 21623615.0,
      "step": 663
    },
    {
      "entropy": 0.9731929302215576,
      "epoch": 0.35508021390374334,
      "grad_norm": 0.14156140387058258,
      "learning_rate": 0.00015477471344361177,
      "loss": 0.9659,
      "mean_token_accuracy": 0.7541355192661285,
      "num_tokens": 21656050.0,
      "step": 664
    },
    {
      "entropy": 1.0352194905281067,
      "epoch": 0.35561497326203206,
      "grad_norm": 0.13904334604740143,
      "learning_rate": 0.0001546393621326945,
      "loss": 1.0232,
      "mean_token_accuracy": 0.7438294291496277,
      "num_tokens": 21688818.0,
      "step": 665
    },
    {
      "entropy": 1.0565296709537506,
      "epoch": 0.35614973262032085,
      "grad_norm": 0.155804842710495,
      "learning_rate": 0.0001545038767856075,
      "loss": 1.0565,
      "mean_token_accuracy": 0.7359127998352051,
      "num_tokens": 21721418.0,
      "step": 666
    },
    {
      "entropy": 1.0352788120508194,
      "epoch": 0.35668449197860963,
      "grad_norm": 0.15668074786663055,
      "learning_rate": 0.00015436825780916523,
      "loss": 1.0307,
      "mean_token_accuracy": 0.7426380813121796,
      "num_tokens": 21754186.0,
      "step": 667
    },
    {
      "entropy": 0.9863010793924332,
      "epoch": 0.3572192513368984,
      "grad_norm": 0.13319331407546997,
      "learning_rate": 0.00015423250561058337,
      "loss": 0.9948,
      "mean_token_accuracy": 0.7499694526195526,
      "num_tokens": 21786954.0,
      "step": 668
    },
    {
      "entropy": 1.0410586893558502,
      "epoch": 0.35775401069518714,
      "grad_norm": 0.15658292174339294,
      "learning_rate": 0.00015409662059747762,
      "loss": 1.0425,
      "mean_token_accuracy": 0.7384836375713348,
      "num_tokens": 21819722.0,
      "step": 669
    },
    {
      "entropy": 1.0100665241479874,
      "epoch": 0.3582887700534759,
      "grad_norm": 0.17908963561058044,
      "learning_rate": 0.00015396060317786247,
      "loss": 1.0263,
      "mean_token_accuracy": 0.7435024678707123,
      "num_tokens": 21852173.0,
      "step": 670
    },
    {
      "entropy": 1.04761204123497,
      "epoch": 0.3588235294117647,
      "grad_norm": 0.1655554324388504,
      "learning_rate": 0.00015382445376014995,
      "loss": 1.0588,
      "mean_token_accuracy": 0.7338538020849228,
      "num_tokens": 21884869.0,
      "step": 671
    },
    {
      "entropy": 0.9861435741186142,
      "epoch": 0.3593582887700535,
      "grad_norm": 0.13408905267715454,
      "learning_rate": 0.00015368817275314848,
      "loss": 0.9848,
      "mean_token_accuracy": 0.7526270747184753,
      "num_tokens": 21917637.0,
      "step": 672
    },
    {
      "entropy": 1.0289635956287384,
      "epoch": 0.3598930481283422,
      "grad_norm": 0.14396777749061584,
      "learning_rate": 0.0001535517605660616,
      "loss": 1.0321,
      "mean_token_accuracy": 0.7429722100496292,
      "num_tokens": 21950393.0,
      "step": 673
    },
    {
      "entropy": 1.0139815360307693,
      "epoch": 0.360427807486631,
      "grad_norm": 0.47777441143989563,
      "learning_rate": 0.00015341521760848674,
      "loss": 0.9951,
      "mean_token_accuracy": 0.7478394359350204,
      "num_tokens": 21982768.0,
      "step": 674
    },
    {
      "entropy": 1.0153920203447342,
      "epoch": 0.3609625668449198,
      "grad_norm": 0.16766510903835297,
      "learning_rate": 0.0001532785442904139,
      "loss": 1.0223,
      "mean_token_accuracy": 0.7463888227939606,
      "num_tokens": 22015316.0,
      "step": 675
    },
    {
      "entropy": 1.0040026307106018,
      "epoch": 0.36149732620320857,
      "grad_norm": 0.13487398624420166,
      "learning_rate": 0.00015314174102222462,
      "loss": 0.9979,
      "mean_token_accuracy": 0.75,
      "num_tokens": 22048084.0,
      "step": 676
    },
    {
      "entropy": 0.9781406968832016,
      "epoch": 0.36203208556149735,
      "grad_norm": 0.1550796926021576,
      "learning_rate": 0.00015300480821469058,
      "loss": 0.9712,
      "mean_token_accuracy": 0.755834549665451,
      "num_tokens": 22080852.0,
      "step": 677
    },
    {
      "entropy": 1.0409707725048065,
      "epoch": 0.3625668449197861,
      "grad_norm": 0.1624067723751068,
      "learning_rate": 0.00015286774627897246,
      "loss": 1.0409,
      "mean_token_accuracy": 0.7391927093267441,
      "num_tokens": 22113316.0,
      "step": 678
    },
    {
      "entropy": 1.0326741486787796,
      "epoch": 0.36310160427807486,
      "grad_norm": 0.13654468953609467,
      "learning_rate": 0.0001527305556266185,
      "loss": 1.0325,
      "mean_token_accuracy": 0.7431894689798355,
      "num_tokens": 22145955.0,
      "step": 679
    },
    {
      "entropy": 0.9604055285453796,
      "epoch": 0.36363636363636365,
      "grad_norm": 0.15321598947048187,
      "learning_rate": 0.00015259323666956374,
      "loss": 0.9473,
      "mean_token_accuracy": 0.7591336667537689,
      "num_tokens": 22178723.0,
      "step": 680
    },
    {
      "entropy": 0.9818018078804016,
      "epoch": 0.36417112299465243,
      "grad_norm": 0.14409194886684418,
      "learning_rate": 0.0001524557898201282,
      "loss": 0.9948,
      "mean_token_accuracy": 0.7462469041347504,
      "num_tokens": 22210814.0,
      "step": 681
    },
    {
      "entropy": 0.971877321600914,
      "epoch": 0.36470588235294116,
      "grad_norm": 0.13896019756793976,
      "learning_rate": 0.000152318215491016,
      "loss": 0.9756,
      "mean_token_accuracy": 0.7509926110506058,
      "num_tokens": 22243493.0,
      "step": 682
    },
    {
      "entropy": 1.0054988861083984,
      "epoch": 0.36524064171122994,
      "grad_norm": 0.1456078439950943,
      "learning_rate": 0.0001521805140953141,
      "loss": 1.005,
      "mean_token_accuracy": 0.7481652200222015,
      "num_tokens": 22275840.0,
      "step": 683
    },
    {
      "entropy": 1.0024132430553436,
      "epoch": 0.3657754010695187,
      "grad_norm": 0.1406199187040329,
      "learning_rate": 0.00015204268604649096,
      "loss": 1.0161,
      "mean_token_accuracy": 0.7452756762504578,
      "num_tokens": 22308573.0,
      "step": 684
    },
    {
      "entropy": 1.0235384851694107,
      "epoch": 0.3663101604278075,
      "grad_norm": 0.13761594891548157,
      "learning_rate": 0.00015190473175839524,
      "loss": 1.0248,
      "mean_token_accuracy": 0.7436275780200958,
      "num_tokens": 22341051.0,
      "step": 685
    },
    {
      "entropy": 0.9671759307384491,
      "epoch": 0.36684491978609624,
      "grad_norm": 0.13828864693641663,
      "learning_rate": 0.00015176665164525478,
      "loss": 0.9566,
      "mean_token_accuracy": 0.7577615976333618,
      "num_tokens": 22373581.0,
      "step": 686
    },
    {
      "entropy": 1.0064850896596909,
      "epoch": 0.367379679144385,
      "grad_norm": 0.13809967041015625,
      "learning_rate": 0.0001516284461216752,
      "loss": 0.9868,
      "mean_token_accuracy": 0.7506383806467056,
      "num_tokens": 22406345.0,
      "step": 687
    },
    {
      "entropy": 0.9803296625614166,
      "epoch": 0.3679144385026738,
      "grad_norm": 0.1352642923593521,
      "learning_rate": 0.0001514901156026386,
      "loss": 0.9801,
      "mean_token_accuracy": 0.7542366534471512,
      "num_tokens": 22438701.0,
      "step": 688
    },
    {
      "entropy": 1.0375657081604004,
      "epoch": 0.3684491978609626,
      "grad_norm": 0.1408119946718216,
      "learning_rate": 0.00015135166050350249,
      "loss": 1.0349,
      "mean_token_accuracy": 0.7384836375713348,
      "num_tokens": 22471469.0,
      "step": 689
    },
    {
      "entropy": 1.0136317312717438,
      "epoch": 0.3689839572192513,
      "grad_norm": 0.1566014438867569,
      "learning_rate": 0.00015121308123999846,
      "loss": 1.0118,
      "mean_token_accuracy": 0.745280846953392,
      "num_tokens": 22504023.0,
      "step": 690
    },
    {
      "entropy": 0.9935726672410965,
      "epoch": 0.3695187165775401,
      "grad_norm": 0.14247219264507294,
      "learning_rate": 0.00015107437822823086,
      "loss": 1.0105,
      "mean_token_accuracy": 0.7484749108552933,
      "num_tokens": 22536412.0,
      "step": 691
    },
    {
      "entropy": 1.0056940615177155,
      "epoch": 0.3700534759358289,
      "grad_norm": 0.15261436998844147,
      "learning_rate": 0.00015093555188467556,
      "loss": 0.9915,
      "mean_token_accuracy": 0.7517784386873245,
      "num_tokens": 22569101.0,
      "step": 692
    },
    {
      "entropy": 1.0365577340126038,
      "epoch": 0.37058823529411766,
      "grad_norm": 0.15125557780265808,
      "learning_rate": 0.00015079660262617888,
      "loss": 1.0414,
      "mean_token_accuracy": 0.7398726642131805,
      "num_tokens": 22601254.0,
      "step": 693
    },
    {
      "entropy": 0.9999520629644394,
      "epoch": 0.3711229946524064,
      "grad_norm": 0.14125481247901917,
      "learning_rate": 0.0001506575308699562,
      "loss": 1.0168,
      "mean_token_accuracy": 0.7414161711931229,
      "num_tokens": 22634022.0,
      "step": 694
    },
    {
      "entropy": 0.9776995927095413,
      "epoch": 0.3716577540106952,
      "grad_norm": 0.14051266014575958,
      "learning_rate": 0.00015051833703359058,
      "loss": 0.978,
      "mean_token_accuracy": 0.7500337809324265,
      "num_tokens": 22666733.0,
      "step": 695
    },
    {
      "entropy": 1.013497769832611,
      "epoch": 0.37219251336898396,
      "grad_norm": 0.1376378983259201,
      "learning_rate": 0.00015037902153503183,
      "loss": 1.0246,
      "mean_token_accuracy": 0.7472812831401825,
      "num_tokens": 22699501.0,
      "step": 696
    },
    {
      "entropy": 0.9666542112827301,
      "epoch": 0.37272727272727274,
      "grad_norm": 0.14466111361980438,
      "learning_rate": 0.00015023958479259492,
      "loss": 0.969,
      "mean_token_accuracy": 0.754028245806694,
      "num_tokens": 22732026.0,
      "step": 697
    },
    {
      "entropy": 1.0464845448732376,
      "epoch": 0.3732620320855615,
      "grad_norm": 0.13540534675121307,
      "learning_rate": 0.0001501000272249589,
      "loss": 1.049,
      "mean_token_accuracy": 0.7369167357683182,
      "num_tokens": 22764683.0,
      "step": 698
    },
    {
      "entropy": 0.9937262833118439,
      "epoch": 0.37379679144385025,
      "grad_norm": 0.1344163864850998,
      "learning_rate": 0.0001499603492511657,
      "loss": 0.9841,
      "mean_token_accuracy": 0.7498625665903091,
      "num_tokens": 22796797.0,
      "step": 699
    },
    {
      "entropy": 1.0619724094867706,
      "epoch": 0.37433155080213903,
      "grad_norm": 0.13974133133888245,
      "learning_rate": 0.00014982055129061868,
      "loss": 1.0481,
      "mean_token_accuracy": 0.7397315353155136,
      "num_tokens": 22829462.0,
      "step": 700
    },
    {
      "entropy": 0.986870288848877,
      "epoch": 0.3748663101604278,
      "grad_norm": 0.13458530604839325,
      "learning_rate": 0.00014968063376308156,
      "loss": 0.9795,
      "mean_token_accuracy": 0.7523122578859329,
      "num_tokens": 22861955.0,
      "step": 701
    },
    {
      "entropy": 1.0136100053787231,
      "epoch": 0.3754010695187166,
      "grad_norm": 0.13495780527591705,
      "learning_rate": 0.000149540597088677,
      "loss": 1.0055,
      "mean_token_accuracy": 0.7504582107067108,
      "num_tokens": 22894723.0,
      "step": 702
    },
    {
      "entropy": 1.0152283012866974,
      "epoch": 0.37593582887700533,
      "grad_norm": 0.1380801647901535,
      "learning_rate": 0.0001494004416878855,
      "loss": 1.0009,
      "mean_token_accuracy": 0.7458455562591553,
      "num_tokens": 22927491.0,
      "step": 703
    },
    {
      "entropy": 0.9783022105693817,
      "epoch": 0.3764705882352941,
      "grad_norm": 0.1381734162569046,
      "learning_rate": 0.00014926016798154403,
      "loss": 0.9774,
      "mean_token_accuracy": 0.752027615904808,
      "num_tokens": 22960110.0,
      "step": 704
    },
    {
      "entropy": 1.0288219153881073,
      "epoch": 0.3770053475935829,
      "grad_norm": 0.1360618621110916,
      "learning_rate": 0.0001491197763908447,
      "loss": 1.033,
      "mean_token_accuracy": 0.7418133020401001,
      "num_tokens": 22992878.0,
      "step": 705
    },
    {
      "entropy": 0.9818412363529205,
      "epoch": 0.3775401069518717,
      "grad_norm": 0.13849791884422302,
      "learning_rate": 0.0001489792673373337,
      "loss": 0.9687,
      "mean_token_accuracy": 0.7564186900854111,
      "num_tokens": 23025584.0,
      "step": 706
    },
    {
      "entropy": 0.9837158620357513,
      "epoch": 0.3780748663101604,
      "grad_norm": 0.14172674715518951,
      "learning_rate": 0.0001488386412429099,
      "loss": 0.9944,
      "mean_token_accuracy": 0.750454768538475,
      "num_tokens": 23058056.0,
      "step": 707
    },
    {
      "entropy": 1.0224552154541016,
      "epoch": 0.3786096256684492,
      "grad_norm": 0.14361903071403503,
      "learning_rate": 0.0001486978985298235,
      "loss": 1.0279,
      "mean_token_accuracy": 0.7439707517623901,
      "num_tokens": 23090319.0,
      "step": 708
    },
    {
      "entropy": 1.022420734167099,
      "epoch": 0.379144385026738,
      "grad_norm": 0.14148057997226715,
      "learning_rate": 0.00014855703962067502,
      "loss": 1.0485,
      "mean_token_accuracy": 0.7371670007705688,
      "num_tokens": 23123075.0,
      "step": 709
    },
    {
      "entropy": 1.0484061241149902,
      "epoch": 0.37967914438502676,
      "grad_norm": 0.14547275006771088,
      "learning_rate": 0.0001484160649384138,
      "loss": 1.0597,
      "mean_token_accuracy": 0.7371066063642502,
      "num_tokens": 23155609.0,
      "step": 710
    },
    {
      "entropy": 0.9728060364723206,
      "epoch": 0.3802139037433155,
      "grad_norm": 0.1438276469707489,
      "learning_rate": 0.0001482749749063367,
      "loss": 0.9747,
      "mean_token_accuracy": 0.7522335052490234,
      "num_tokens": 23188155.0,
      "step": 711
    },
    {
      "entropy": 1.0211612433195114,
      "epoch": 0.38074866310160427,
      "grad_norm": 0.1322088986635208,
      "learning_rate": 0.00014813376994808717,
      "loss": 1.0085,
      "mean_token_accuracy": 0.745509535074234,
      "num_tokens": 23220923.0,
      "step": 712
    },
    {
      "entropy": 1.035618707537651,
      "epoch": 0.38128342245989305,
      "grad_norm": 0.13468991219997406,
      "learning_rate": 0.0001479924504876535,
      "loss": 1.0273,
      "mean_token_accuracy": 0.7402165532112122,
      "num_tokens": 23253650.0,
      "step": 713
    },
    {
      "entropy": 1.0208526104688644,
      "epoch": 0.38181818181818183,
      "grad_norm": 0.1391787976026535,
      "learning_rate": 0.000147851016949368,
      "loss": 1.0081,
      "mean_token_accuracy": 0.7423631548881531,
      "num_tokens": 23286418.0,
      "step": 714
    },
    {
      "entropy": 0.9913294613361359,
      "epoch": 0.38235294117647056,
      "grad_norm": 0.13534633815288544,
      "learning_rate": 0.00014770946975790534,
      "loss": 0.9803,
      "mean_token_accuracy": 0.7509450316429138,
      "num_tokens": 23318480.0,
      "step": 715
    },
    {
      "entropy": 1.0779812783002853,
      "epoch": 0.38288770053475935,
      "grad_norm": 0.1359453648328781,
      "learning_rate": 0.00014756780933828157,
      "loss": 1.0667,
      "mean_token_accuracy": 0.7307667434215546,
      "num_tokens": 23351175.0,
      "step": 716
    },
    {
      "entropy": 0.9769474864006042,
      "epoch": 0.38342245989304813,
      "grad_norm": 0.14193221926689148,
      "learning_rate": 0.0001474260361158526,
      "loss": 0.9711,
      "mean_token_accuracy": 0.7550708651542664,
      "num_tokens": 23383943.0,
      "step": 717
    },
    {
      "entropy": 0.9842025190591812,
      "epoch": 0.3839572192513369,
      "grad_norm": 0.13903889060020447,
      "learning_rate": 0.00014728415051631325,
      "loss": 0.9939,
      "mean_token_accuracy": 0.7491141259670258,
      "num_tokens": 23416711.0,
      "step": 718
    },
    {
      "entropy": 1.0313686281442642,
      "epoch": 0.3844919786096257,
      "grad_norm": 0.15327425301074982,
      "learning_rate": 0.0001471421529656956,
      "loss": 1.0396,
      "mean_token_accuracy": 0.7405342608690262,
      "num_tokens": 23449223.0,
      "step": 719
    },
    {
      "entropy": 0.9921634346246719,
      "epoch": 0.3850267379679144,
      "grad_norm": 0.1574993133544922,
      "learning_rate": 0.00014700004389036784,
      "loss": 1.0025,
      "mean_token_accuracy": 0.7462220937013626,
      "num_tokens": 23481947.0,
      "step": 720
    },
    {
      "entropy": 0.9802977293729782,
      "epoch": 0.3855614973262032,
      "grad_norm": 0.17303505539894104,
      "learning_rate": 0.00014685782371703322,
      "loss": 0.9906,
      "mean_token_accuracy": 0.7480792254209518,
      "num_tokens": 23513871.0,
      "step": 721
    },
    {
      "entropy": 1.0380671918392181,
      "epoch": 0.386096256684492,
      "grad_norm": 0.1449204385280609,
      "learning_rate": 0.0001467154928727284,
      "loss": 1.0363,
      "mean_token_accuracy": 0.7386363595724106,
      "num_tokens": 23546639.0,
      "step": 722
    },
    {
      "entropy": 1.0341337323188782,
      "epoch": 0.3866310160427808,
      "grad_norm": 0.14379708468914032,
      "learning_rate": 0.00014657305178482246,
      "loss": 1.0431,
      "mean_token_accuracy": 0.741865485906601,
      "num_tokens": 23578906.0,
      "step": 723
    },
    {
      "entropy": 0.9877164959907532,
      "epoch": 0.3871657754010695,
      "grad_norm": 0.15079493820667267,
      "learning_rate": 0.00014643050088101545,
      "loss": 0.9661,
      "mean_token_accuracy": 0.7591642141342163,
      "num_tokens": 23611674.0,
      "step": 724
    },
    {
      "entropy": 0.9995661824941635,
      "epoch": 0.3877005347593583,
      "grad_norm": 0.16325317323207855,
      "learning_rate": 0.00014628784058933717,
      "loss": 0.9907,
      "mean_token_accuracy": 0.7485337257385254,
      "num_tokens": 23644442.0,
      "step": 725
    },
    {
      "entropy": 0.9665198773145676,
      "epoch": 0.38823529411764707,
      "grad_norm": 0.14961887896060944,
      "learning_rate": 0.00014614507133814584,
      "loss": 0.9425,
      "mean_token_accuracy": 0.7581816762685776,
      "num_tokens": 23676432.0,
      "step": 726
    },
    {
      "entropy": 1.0407246053218842,
      "epoch": 0.38877005347593585,
      "grad_norm": 0.14445669949054718,
      "learning_rate": 0.00014600219355612694,
      "loss": 1.0202,
      "mean_token_accuracy": 0.7398888170719147,
      "num_tokens": 23709200.0,
      "step": 727
    },
    {
      "entropy": 0.9938240647315979,
      "epoch": 0.3893048128342246,
      "grad_norm": 0.15048936009407043,
      "learning_rate": 0.0001458592076722917,
      "loss": 0.9936,
      "mean_token_accuracy": 0.7492668628692627,
      "num_tokens": 23741968.0,
      "step": 728
    },
    {
      "entropy": 0.9772266745567322,
      "epoch": 0.38983957219251336,
      "grad_norm": 0.1531447023153305,
      "learning_rate": 0.00014571611411597605,
      "loss": 0.9777,
      "mean_token_accuracy": 0.7505871802568436,
      "num_tokens": 23774679.0,
      "step": 729
    },
    {
      "entropy": 0.9953857660293579,
      "epoch": 0.39037433155080214,
      "grad_norm": 0.14569969475269318,
      "learning_rate": 0.00014557291331683916,
      "loss": 0.9935,
      "mean_token_accuracy": 0.7505297958850861,
      "num_tokens": 23807385.0,
      "step": 730
    },
    {
      "entropy": 0.9788328260183334,
      "epoch": 0.39090909090909093,
      "grad_norm": 0.1406850516796112,
      "learning_rate": 0.00014542960570486222,
      "loss": 0.9856,
      "mean_token_accuracy": 0.7500044405460358,
      "num_tokens": 23839661.0,
      "step": 731
    },
    {
      "entropy": 1.0059930682182312,
      "epoch": 0.39144385026737966,
      "grad_norm": 0.14730219542980194,
      "learning_rate": 0.00014528619171034717,
      "loss": 0.9887,
      "mean_token_accuracy": 0.7483198940753937,
      "num_tokens": 23872429.0,
      "step": 732
    },
    {
      "entropy": 1.0020950585603714,
      "epoch": 0.39197860962566844,
      "grad_norm": 0.14256757497787476,
      "learning_rate": 0.00014514267176391542,
      "loss": 1.0101,
      "mean_token_accuracy": 0.7468536198139191,
      "num_tokens": 23905197.0,
      "step": 733
    },
    {
      "entropy": 0.9817915260791779,
      "epoch": 0.3925133689839572,
      "grad_norm": 0.1441333293914795,
      "learning_rate": 0.00014499904629650638,
      "loss": 1.0013,
      "mean_token_accuracy": 0.7450421899557114,
      "num_tokens": 23937585.0,
      "step": 734
    },
    {
      "entropy": 1.0192552655935287,
      "epoch": 0.393048128342246,
      "grad_norm": 0.15254747867584229,
      "learning_rate": 0.0001448553157393764,
      "loss": 1.0264,
      "mean_token_accuracy": 0.7418911159038544,
      "num_tokens": 23970285.0,
      "step": 735
    },
    {
      "entropy": 0.996012270450592,
      "epoch": 0.39358288770053473,
      "grad_norm": 0.1529400646686554,
      "learning_rate": 0.00014471148052409732,
      "loss": 1.004,
      "mean_token_accuracy": 0.7430134862661362,
      "num_tokens": 24003021.0,
      "step": 736
    },
    {
      "entropy": 0.9864199012517929,
      "epoch": 0.3941176470588235,
      "grad_norm": 0.1369430273771286,
      "learning_rate": 0.0001445675410825554,
      "loss": 0.9948,
      "mean_token_accuracy": 0.7503091543912888,
      "num_tokens": 24035141.0,
      "step": 737
    },
    {
      "entropy": 0.977002888917923,
      "epoch": 0.3946524064171123,
      "grad_norm": 0.16152888536453247,
      "learning_rate": 0.00014442349784694956,
      "loss": 0.9748,
      "mean_token_accuracy": 0.7555102407932281,
      "num_tokens": 24067417.0,
      "step": 738
    },
    {
      "entropy": 1.027063712477684,
      "epoch": 0.3951871657754011,
      "grad_norm": 0.15807777643203735,
      "learning_rate": 0.00014427935124979068,
      "loss": 1.0293,
      "mean_token_accuracy": 0.7409990280866623,
      "num_tokens": 24099845.0,
      "step": 739
    },
    {
      "entropy": 0.9792217314243317,
      "epoch": 0.39572192513368987,
      "grad_norm": 0.13410022854804993,
      "learning_rate": 0.00014413510172389977,
      "loss": 0.9764,
      "mean_token_accuracy": 0.7549799382686615,
      "num_tokens": 24132073.0,
      "step": 740
    },
    {
      "entropy": 1.0001242756843567,
      "epoch": 0.3962566844919786,
      "grad_norm": 0.17179086804389954,
      "learning_rate": 0.00014399074970240702,
      "loss": 0.986,
      "mean_token_accuracy": 0.7496117204427719,
      "num_tokens": 24164170.0,
      "step": 741
    },
    {
      "entropy": 0.9808478057384491,
      "epoch": 0.3967914438502674,
      "grad_norm": 0.17493928968906403,
      "learning_rate": 0.0001438462956187503,
      "loss": 0.9799,
      "mean_token_accuracy": 0.7514575868844986,
      "num_tokens": 24196887.0,
      "step": 742
    },
    {
      "entropy": 0.9944099485874176,
      "epoch": 0.39732620320855616,
      "grad_norm": 0.14331574738025665,
      "learning_rate": 0.00014370173990667406,
      "loss": 0.9931,
      "mean_token_accuracy": 0.7517488449811935,
      "num_tokens": 24229346.0,
      "step": 743
    },
    {
      "entropy": 1.0133265405893326,
      "epoch": 0.39786096256684494,
      "grad_norm": 0.15669451653957367,
      "learning_rate": 0.00014355708300022778,
      "loss": 1.0169,
      "mean_token_accuracy": 0.7478005886077881,
      "num_tokens": 24262114.0,
      "step": 744
    },
    {
      "entropy": 0.9847605526447296,
      "epoch": 0.3983957219251337,
      "grad_norm": 0.16532684862613678,
      "learning_rate": 0.00014341232533376483,
      "loss": 0.9982,
      "mean_token_accuracy": 0.7511332482099533,
      "num_tokens": 24294705.0,
      "step": 745
    },
    {
      "entropy": 1.0536640584468842,
      "epoch": 0.39893048128342246,
      "grad_norm": 0.13670744001865387,
      "learning_rate": 0.00014326746734194114,
      "loss": 1.0482,
      "mean_token_accuracy": 0.737246185541153,
      "num_tokens": 24327241.0,
      "step": 746
    },
    {
      "entropy": 0.9804176688194275,
      "epoch": 0.39946524064171124,
      "grad_norm": 0.1547093689441681,
      "learning_rate": 0.00014312250945971392,
      "loss": 1.0004,
      "mean_token_accuracy": 0.7477412968873978,
      "num_tokens": 24359934.0,
      "step": 747
    },
    {
      "entropy": 0.9940600544214249,
      "epoch": 0.4,
      "grad_norm": 0.15820637345314026,
      "learning_rate": 0.0001429774521223402,
      "loss": 1.0192,
      "mean_token_accuracy": 0.7446351796388626,
      "num_tokens": 24392082.0,
      "step": 748
    },
    {
      "entropy": 0.9975368827581406,
      "epoch": 0.40053475935828875,
      "grad_norm": 0.14259840548038483,
      "learning_rate": 0.00014283229576537572,
      "loss": 0.9917,
      "mean_token_accuracy": 0.7497696876525879,
      "num_tokens": 24424720.0,
      "step": 749
    },
    {
      "entropy": 1.000609129667282,
      "epoch": 0.40106951871657753,
      "grad_norm": 0.13693387806415558,
      "learning_rate": 0.00014268704082467357,
      "loss": 0.9917,
      "mean_token_accuracy": 0.7509403973817825,
      "num_tokens": 24457378.0,
      "step": 750
    },
    {
      "entropy": 1.0345368534326553,
      "epoch": 0.4016042780748663,
      "grad_norm": 0.14850850403308868,
      "learning_rate": 0.00014254168773638278,
      "loss": 1.0117,
      "mean_token_accuracy": 0.7432490140199661,
      "num_tokens": 24490146.0,
      "step": 751
    },
    {
      "entropy": 1.0088206380605698,
      "epoch": 0.4021390374331551,
      "grad_norm": 0.13687895238399506,
      "learning_rate": 0.00014239623693694712,
      "loss": 0.9965,
      "mean_token_accuracy": 0.7491752207279205,
      "num_tokens": 24522914.0,
      "step": 752
    },
    {
      "entropy": 1.0217233449220657,
      "epoch": 0.4026737967914438,
      "grad_norm": 0.13860391080379486,
      "learning_rate": 0.00014225068886310375,
      "loss": 1.0085,
      "mean_token_accuracy": 0.7483198940753937,
      "num_tokens": 24555682.0,
      "step": 753
    },
    {
      "entropy": 0.9968471378087997,
      "epoch": 0.4032085561497326,
      "grad_norm": 0.14066515862941742,
      "learning_rate": 0.0001421050439518819,
      "loss": 0.9831,
      "mean_token_accuracy": 0.7497556209564209,
      "num_tokens": 24588450.0,
      "step": 754
    },
    {
      "entropy": 0.9938175976276398,
      "epoch": 0.4037433155080214,
      "grad_norm": 0.13824208080768585,
      "learning_rate": 0.00014195930264060159,
      "loss": 0.9947,
      "mean_token_accuracy": 0.7470369040966034,
      "num_tokens": 24621218.0,
      "step": 755
    },
    {
      "entropy": 1.0178113877773285,
      "epoch": 0.4042780748663102,
      "grad_norm": 0.13856786489486694,
      "learning_rate": 0.00014181346536687225,
      "loss": 1.012,
      "mean_token_accuracy": 0.7461252063512802,
      "num_tokens": 24653641.0,
      "step": 756
    },
    {
      "entropy": 0.997627466917038,
      "epoch": 0.4048128342245989,
      "grad_norm": 0.14167548716068268,
      "learning_rate": 0.00014166753256859152,
      "loss": 1.0032,
      "mean_token_accuracy": 0.7465094774961472,
      "num_tokens": 24686245.0,
      "step": 757
    },
    {
      "entropy": 0.9874056875705719,
      "epoch": 0.4053475935828877,
      "grad_norm": 0.1428968608379364,
      "learning_rate": 0.00014152150468394378,
      "loss": 0.9719,
      "mean_token_accuracy": 0.7541117370128632,
      "num_tokens": 24718792.0,
      "step": 758
    },
    {
      "entropy": 0.9934220016002655,
      "epoch": 0.40588235294117647,
      "grad_norm": 0.139199897646904,
      "learning_rate": 0.000141375382151399,
      "loss": 0.994,
      "mean_token_accuracy": 0.7510247379541397,
      "num_tokens": 24751530.0,
      "step": 759
    },
    {
      "entropy": 0.9941307455301285,
      "epoch": 0.40641711229946526,
      "grad_norm": 0.14097781479358673,
      "learning_rate": 0.0001412291654097113,
      "loss": 1.0032,
      "mean_token_accuracy": 0.7473887205123901,
      "num_tokens": 24784094.0,
      "step": 760
    },
    {
      "entropy": 1.0188378989696503,
      "epoch": 0.406951871657754,
      "grad_norm": 0.13915149867534637,
      "learning_rate": 0.00014108285489791768,
      "loss": 1.0281,
      "mean_token_accuracy": 0.7360606342554092,
      "num_tokens": 24816746.0,
      "step": 761
    },
    {
      "entropy": 1.0229030549526215,
      "epoch": 0.40748663101604277,
      "grad_norm": 0.14130914211273193,
      "learning_rate": 0.00014093645105533675,
      "loss": 1.0192,
      "mean_token_accuracy": 0.7431410849094391,
      "num_tokens": 24849356.0,
      "step": 762
    },
    {
      "entropy": 1.0327558815479279,
      "epoch": 0.40802139037433155,
      "grad_norm": 0.14207275211811066,
      "learning_rate": 0.00014078995432156724,
      "loss": 1.0427,
      "mean_token_accuracy": 0.7328323721885681,
      "num_tokens": 24882124.0,
      "step": 763
    },
    {
      "entropy": 0.997843861579895,
      "epoch": 0.40855614973262033,
      "grad_norm": 0.13978444039821625,
      "learning_rate": 0.00014064336513648697,
      "loss": 0.9981,
      "mean_token_accuracy": 0.7477700412273407,
      "num_tokens": 24914892.0,
      "step": 764
    },
    {
      "entropy": 0.9983325004577637,
      "epoch": 0.4090909090909091,
      "grad_norm": 0.14260488748550415,
      "learning_rate": 0.00014049668394025119,
      "loss": 0.9921,
      "mean_token_accuracy": 0.7530656903982162,
      "num_tokens": 24947529.0,
      "step": 765
    },
    {
      "entropy": 1.0412585139274597,
      "epoch": 0.40962566844919784,
      "grad_norm": 0.14486606419086456,
      "learning_rate": 0.00014034991117329156,
      "loss": 1.0233,
      "mean_token_accuracy": 0.7403512895107269,
      "num_tokens": 24979735.0,
      "step": 766
    },
    {
      "entropy": 0.9993823319673538,
      "epoch": 0.4101604278074866,
      "grad_norm": 0.15249890089035034,
      "learning_rate": 0.00014020304727631452,
      "loss": 0.9798,
      "mean_token_accuracy": 0.7502091825008392,
      "num_tokens": 25012457.0,
      "step": 767
    },
    {
      "entropy": 1.007173702120781,
      "epoch": 0.4106951871657754,
      "grad_norm": 0.16284959018230438,
      "learning_rate": 0.00014005609269030037,
      "loss": 1.009,
      "mean_token_accuracy": 0.7418438494205475,
      "num_tokens": 25045225.0,
      "step": 768
    },
    {
      "entropy": 1.006159782409668,
      "epoch": 0.4112299465240642,
      "grad_norm": 0.14431804418563843,
      "learning_rate": 0.00013990904785650158,
      "loss": 1.0189,
      "mean_token_accuracy": 0.746059387922287,
      "num_tokens": 25077993.0,
      "step": 769
    },
    {
      "entropy": 0.8995373249053955,
      "epoch": 0.4117647058823529,
      "grad_norm": 0.1489420235157013,
      "learning_rate": 0.00013976191321644161,
      "loss": 0.9115,
      "mean_token_accuracy": 0.7671065330505371,
      "num_tokens": 25110761.0,
      "step": 770
    },
    {
      "entropy": 1.0053518116474152,
      "epoch": 0.4122994652406417,
      "grad_norm": 0.1627763956785202,
      "learning_rate": 0.0001396146892119136,
      "loss": 1.0225,
      "mean_token_accuracy": 0.7457539141178131,
      "num_tokens": 25143529.0,
      "step": 771
    },
    {
      "entropy": 0.9979801774024963,
      "epoch": 0.4128342245989305,
      "grad_norm": 0.1462087333202362,
      "learning_rate": 0.000139467376284979,
      "loss": 1.017,
      "mean_token_accuracy": 0.7471773624420166,
      "num_tokens": 25176281.0,
      "step": 772
    },
    {
      "entropy": 1.0216413140296936,
      "epoch": 0.41336898395721927,
      "grad_norm": 0.14449453353881836,
      "learning_rate": 0.00013931997487796628,
      "loss": 1.0358,
      "mean_token_accuracy": 0.741568922996521,
      "num_tokens": 25209049.0,
      "step": 773
    },
    {
      "entropy": 0.9999716579914093,
      "epoch": 0.413903743315508,
      "grad_norm": 0.15826645493507385,
      "learning_rate": 0.00013917248543346956,
      "loss": 1.0139,
      "mean_token_accuracy": 0.7501832842826843,
      "num_tokens": 25241817.0,
      "step": 774
    },
    {
      "entropy": 1.0087449997663498,
      "epoch": 0.4144385026737968,
      "grad_norm": 0.14455418288707733,
      "learning_rate": 0.0001390249083943474,
      "loss": 0.9827,
      "mean_token_accuracy": 0.7504959404468536,
      "num_tokens": 25274306.0,
      "step": 775
    },
    {
      "entropy": 1.0708818435668945,
      "epoch": 0.41497326203208557,
      "grad_norm": 0.14365294575691223,
      "learning_rate": 0.0001388772442037212,
      "loss": 1.0512,
      "mean_token_accuracy": 0.7327451109886169,
      "num_tokens": 25306807.0,
      "step": 776
    },
    {
      "entropy": 1.000595673918724,
      "epoch": 0.41550802139037435,
      "grad_norm": 0.1471114158630371,
      "learning_rate": 0.00013872949330497416,
      "loss": 0.9777,
      "mean_token_accuracy": 0.7492363154888153,
      "num_tokens": 25339575.0,
      "step": 777
    },
    {
      "entropy": 1.0157749354839325,
      "epoch": 0.4160427807486631,
      "grad_norm": 0.13333575427532196,
      "learning_rate": 0.00013858165614174985,
      "loss": 0.9994,
      "mean_token_accuracy": 0.7480064779520035,
      "num_tokens": 25372306.0,
      "step": 778
    },
    {
      "entropy": 0.9878341257572174,
      "epoch": 0.41657754010695186,
      "grad_norm": 0.1525413691997528,
      "learning_rate": 0.00013843373315795086,
      "loss": 0.9778,
      "mean_token_accuracy": 0.7529974728822708,
      "num_tokens": 25404641.0,
      "step": 779
    },
    {
      "entropy": 1.0254554450511932,
      "epoch": 0.41711229946524064,
      "grad_norm": 0.17594583332538605,
      "learning_rate": 0.00013828572479773735,
      "loss": 1.0232,
      "mean_token_accuracy": 0.7407135963439941,
      "num_tokens": 25437409.0,
      "step": 780
    },
    {
      "entropy": 0.9847807586193085,
      "epoch": 0.4176470588235294,
      "grad_norm": 0.15318170189857483,
      "learning_rate": 0.00013813763150552602,
      "loss": 0.9991,
      "mean_token_accuracy": 0.7479104548692703,
      "num_tokens": 25469642.0,
      "step": 781
    },
    {
      "entropy": 0.9743617326021194,
      "epoch": 0.41818181818181815,
      "grad_norm": 0.13956089317798615,
      "learning_rate": 0.0001379894537259884,
      "loss": 0.9872,
      "mean_token_accuracy": 0.7450818568468094,
      "num_tokens": 25502410.0,
      "step": 782
    },
    {
      "entropy": 1.0024569183588028,
      "epoch": 0.41871657754010694,
      "grad_norm": 0.17088918387889862,
      "learning_rate": 0.0001378411919040499,
      "loss": 1.0059,
      "mean_token_accuracy": 0.743884950876236,
      "num_tokens": 25534753.0,
      "step": 783
    },
    {
      "entropy": 0.9946104288101196,
      "epoch": 0.4192513368983957,
      "grad_norm": 0.15087063610553741,
      "learning_rate": 0.00013769284648488807,
      "loss": 0.9915,
      "mean_token_accuracy": 0.7477570027112961,
      "num_tokens": 25567224.0,
      "step": 784
    },
    {
      "entropy": 1.005013570189476,
      "epoch": 0.4197860962566845,
      "grad_norm": 0.14750632643699646,
      "learning_rate": 0.00013754441791393167,
      "loss": 1.0233,
      "mean_token_accuracy": 0.7427409589290619,
      "num_tokens": 25599950.0,
      "step": 785
    },
    {
      "entropy": 1.0433073788881302,
      "epoch": 0.4203208556149733,
      "grad_norm": 0.158527672290802,
      "learning_rate": 0.00013739590663685902,
      "loss": 1.052,
      "mean_token_accuracy": 0.7364682704210281,
      "num_tokens": 25632429.0,
      "step": 786
    },
    {
      "entropy": 0.9824782460927963,
      "epoch": 0.420855614973262,
      "grad_norm": 0.1468294858932495,
      "learning_rate": 0.00013724731309959677,
      "loss": 1.0058,
      "mean_token_accuracy": 0.7483760863542557,
      "num_tokens": 25664811.0,
      "step": 787
    },
    {
      "entropy": 1.0482485890388489,
      "epoch": 0.4213903743315508,
      "grad_norm": 0.14201639592647552,
      "learning_rate": 0.00013709863774831864,
      "loss": 1.0458,
      "mean_token_accuracy": 0.7343597412109375,
      "num_tokens": 25697579.0,
      "step": 788
    },
    {
      "entropy": 0.9861636310815811,
      "epoch": 0.4219251336898396,
      "grad_norm": 0.1537942737340927,
      "learning_rate": 0.00013694988102944396,
      "loss": 0.9716,
      "mean_token_accuracy": 0.7534565180540085,
      "num_tokens": 25729923.0,
      "step": 789
    },
    {
      "entropy": 1.0346594899892807,
      "epoch": 0.42245989304812837,
      "grad_norm": 0.14477504789829254,
      "learning_rate": 0.00013680104338963643,
      "loss": 1.011,
      "mean_token_accuracy": 0.7462104111909866,
      "num_tokens": 25762085.0,
      "step": 790
    },
    {
      "entropy": 0.9832995384931564,
      "epoch": 0.4229946524064171,
      "grad_norm": 0.14770746231079102,
      "learning_rate": 0.00013665212527580258,
      "loss": 0.973,
      "mean_token_accuracy": 0.7535790205001831,
      "num_tokens": 25794308.0,
      "step": 791
    },
    {
      "entropy": 0.9789574444293976,
      "epoch": 0.4235294117647059,
      "grad_norm": 0.1468288004398346,
      "learning_rate": 0.00013650312713509076,
      "loss": 0.9663,
      "mean_token_accuracy": 0.757819414138794,
      "num_tokens": 25826744.0,
      "step": 792
    },
    {
      "entropy": 0.9459770172834396,
      "epoch": 0.42406417112299466,
      "grad_norm": 0.14010605216026306,
      "learning_rate": 0.00013635404941488954,
      "loss": 0.9378,
      "mean_token_accuracy": 0.7616658359766006,
      "num_tokens": 25859274.0,
      "step": 793
    },
    {
      "entropy": 0.9862398356199265,
      "epoch": 0.42459893048128344,
      "grad_norm": 0.15382592380046844,
      "learning_rate": 0.00013620489256282642,
      "loss": 1.0046,
      "mean_token_accuracy": 0.7458150088787079,
      "num_tokens": 25892042.0,
      "step": 794
    },
    {
      "entropy": 1.00739586353302,
      "epoch": 0.42513368983957217,
      "grad_norm": 0.1552584171295166,
      "learning_rate": 0.0001360556570267665,
      "loss": 1.0101,
      "mean_token_accuracy": 0.7430644780397415,
      "num_tokens": 25924536.0,
      "step": 795
    },
    {
      "entropy": 0.9220166951417923,
      "epoch": 0.42566844919786095,
      "grad_norm": 0.14095531404018402,
      "learning_rate": 0.0001359063432548111,
      "loss": 0.9278,
      "mean_token_accuracy": 0.7619739770889282,
      "num_tokens": 25956828.0,
      "step": 796
    },
    {
      "entropy": 1.0195706486701965,
      "epoch": 0.42620320855614974,
      "grad_norm": 0.15685130655765533,
      "learning_rate": 0.00013575695169529669,
      "loss": 1.0378,
      "mean_token_accuracy": 0.7380559742450714,
      "num_tokens": 25989596.0,
      "step": 797
    },
    {
      "entropy": 0.9775845110416412,
      "epoch": 0.4267379679144385,
      "grad_norm": 0.141723170876503,
      "learning_rate": 0.0001356074827967929,
      "loss": 0.9914,
      "mean_token_accuracy": 0.7492057681083679,
      "num_tokens": 26022364.0,
      "step": 798
    },
    {
      "entropy": 0.9945722669363022,
      "epoch": 0.42727272727272725,
      "grad_norm": 0.14415983855724335,
      "learning_rate": 0.00013545793700810204,
      "loss": 1.0042,
      "mean_token_accuracy": 0.7473423779010773,
      "num_tokens": 26055132.0,
      "step": 799
    },
    {
      "entropy": 0.9478647261857986,
      "epoch": 0.42780748663101603,
      "grad_norm": 0.1493312567472458,
      "learning_rate": 0.00013530831477825687,
      "loss": 0.9379,
      "mean_token_accuracy": 0.7591336667537689,
      "num_tokens": 26087900.0,
      "step": 800
    },
    {
      "entropy": 1.0211884379386902,
      "epoch": 0.4283422459893048,
      "grad_norm": 0.13690948486328125,
      "learning_rate": 0.00013515861655652,
      "loss": 1.0188,
      "mean_token_accuracy": 0.744165450334549,
      "num_tokens": 26120668.0,
      "step": 801
    },
    {
      "entropy": 1.0712903141975403,
      "epoch": 0.4288770053475936,
      "grad_norm": 0.15197446942329407,
      "learning_rate": 0.00013500884279238202,
      "loss": 1.0723,
      "mean_token_accuracy": 0.7357724756002426,
      "num_tokens": 26153171.0,
      "step": 802
    },
    {
      "entropy": 0.9801386594772339,
      "epoch": 0.4294117647058823,
      "grad_norm": 0.14681871235370636,
      "learning_rate": 0.00013485899393556042,
      "loss": 0.9791,
      "mean_token_accuracy": 0.750549852848053,
      "num_tokens": 26185939.0,
      "step": 803
    },
    {
      "entropy": 0.9867227226495743,
      "epoch": 0.4299465240641711,
      "grad_norm": 0.14551116526126862,
      "learning_rate": 0.0001347090704359982,
      "loss": 0.9803,
      "mean_token_accuracy": 0.7532755136489868,
      "num_tokens": 26218088.0,
      "step": 804
    },
    {
      "entropy": 1.046304628252983,
      "epoch": 0.4304812834224599,
      "grad_norm": 0.1528748720884323,
      "learning_rate": 0.00013455907274386244,
      "loss": 1.043,
      "mean_token_accuracy": 0.7424242496490479,
      "num_tokens": 26250856.0,
      "step": 805
    },
    {
      "entropy": 0.9992060363292694,
      "epoch": 0.4310160427807487,
      "grad_norm": 0.14023494720458984,
      "learning_rate": 0.00013440900130954296,
      "loss": 0.9985,
      "mean_token_accuracy": 0.7464259564876556,
      "num_tokens": 26283624.0,
      "step": 806
    },
    {
      "entropy": 1.0101826190948486,
      "epoch": 0.43155080213903746,
      "grad_norm": 0.15695171058177948,
      "learning_rate": 0.0001342588565836511,
      "loss": 1.0149,
      "mean_token_accuracy": 0.7469891905784607,
      "num_tokens": 26315876.0,
      "step": 807
    },
    {
      "entropy": 1.025041863322258,
      "epoch": 0.4320855614973262,
      "grad_norm": 0.14769817888736725,
      "learning_rate": 0.00013410863901701817,
      "loss": 1.031,
      "mean_token_accuracy": 0.7399892508983612,
      "num_tokens": 26348516.0,
      "step": 808
    },
    {
      "entropy": 1.0198047310113907,
      "epoch": 0.43262032085561497,
      "grad_norm": 0.14085803925991058,
      "learning_rate": 0.00013395834906069424,
      "loss": 1.0278,
      "mean_token_accuracy": 0.7437599748373032,
      "num_tokens": 26381254.0,
      "step": 809
    },
    {
      "entropy": 1.055108368396759,
      "epoch": 0.43315508021390375,
      "grad_norm": 0.14935609698295593,
      "learning_rate": 0.00013380798716594675,
      "loss": 1.0716,
      "mean_token_accuracy": 0.7361881732940674,
      "num_tokens": 26413887.0,
      "step": 810
    },
    {
      "entropy": 0.9968017041683197,
      "epoch": 0.43368983957219254,
      "grad_norm": 0.13966552913188934,
      "learning_rate": 0.00013365755378425914,
      "loss": 0.9842,
      "mean_token_accuracy": 0.7523777335882187,
      "num_tokens": 26446328.0,
      "step": 811
    },
    {
      "entropy": 1.0102295130491257,
      "epoch": 0.43422459893048126,
      "grad_norm": 0.1441984623670578,
      "learning_rate": 0.00013350704936732947,
      "loss": 0.9868,
      "mean_token_accuracy": 0.750946968793869,
      "num_tokens": 26479096.0,
      "step": 812
    },
    {
      "entropy": 0.9585129767656326,
      "epoch": 0.43475935828877005,
      "grad_norm": 0.14310646057128906,
      "learning_rate": 0.00013335647436706912,
      "loss": 0.9537,
      "mean_token_accuracy": 0.759378045797348,
      "num_tokens": 26511864.0,
      "step": 813
    },
    {
      "entropy": 0.9890850186347961,
      "epoch": 0.43529411764705883,
      "grad_norm": 0.14099349081516266,
      "learning_rate": 0.00013320582923560142,
      "loss": 0.9891,
      "mean_token_accuracy": 0.7498167157173157,
      "num_tokens": 26544632.0,
      "step": 814
    },
    {
      "entropy": 0.9900194853544235,
      "epoch": 0.4358288770053476,
      "grad_norm": 0.16107913851737976,
      "learning_rate": 0.00013305511442526026,
      "loss": 0.9974,
      "mean_token_accuracy": 0.7471900135278702,
      "num_tokens": 26577280.0,
      "step": 815
    },
    {
      "entropy": 0.9850942343473434,
      "epoch": 0.43636363636363634,
      "grad_norm": 0.14523376524448395,
      "learning_rate": 0.0001329043303885887,
      "loss": 1.0001,
      "mean_token_accuracy": 0.7438905239105225,
      "num_tokens": 26610048.0,
      "step": 816
    },
    {
      "entropy": 1.0441436767578125,
      "epoch": 0.4368983957219251,
      "grad_norm": 0.15303680300712585,
      "learning_rate": 0.00013275347757833782,
      "loss": 1.0364,
      "mean_token_accuracy": 0.7361928671598434,
      "num_tokens": 26642730.0,
      "step": 817
    },
    {
      "entropy": 1.006567895412445,
      "epoch": 0.4374331550802139,
      "grad_norm": 0.14957407116889954,
      "learning_rate": 0.000132602556447465,
      "loss": 1.0212,
      "mean_token_accuracy": 0.7475256621837616,
      "num_tokens": 26675498.0,
      "step": 818
    },
    {
      "entropy": 0.9686824828386307,
      "epoch": 0.4379679144385027,
      "grad_norm": 0.1421763002872467,
      "learning_rate": 0.0001324515674491329,
      "loss": 0.9715,
      "mean_token_accuracy": 0.7550385147333145,
      "num_tokens": 26707685.0,
      "step": 819
    },
    {
      "entropy": 0.9856327921152115,
      "epoch": 0.4385026737967914,
      "grad_norm": 0.14262866973876953,
      "learning_rate": 0.00013230051103670792,
      "loss": 0.9634,
      "mean_token_accuracy": 0.7571175396442413,
      "num_tokens": 26740453.0,
      "step": 820
    },
    {
      "entropy": 1.0294437855482101,
      "epoch": 0.4390374331550802,
      "grad_norm": 0.13774727284908295,
      "learning_rate": 0.0001321493876637589,
      "loss": 1.007,
      "mean_token_accuracy": 0.7464000433683395,
      "num_tokens": 26773101.0,
      "step": 821
    },
    {
      "entropy": 1.0105653703212738,
      "epoch": 0.439572192513369,
      "grad_norm": 0.14089708030223846,
      "learning_rate": 0.00013199819778405569,
      "loss": 0.9933,
      "mean_token_accuracy": 0.7512582391500473,
      "num_tokens": 26805686.0,
      "step": 822
    },
    {
      "entropy": 1.0349905490875244,
      "epoch": 0.44010695187165777,
      "grad_norm": 0.1354954093694687,
      "learning_rate": 0.0001318469418515679,
      "loss": 1.0276,
      "mean_token_accuracy": 0.7418626844882965,
      "num_tokens": 26838265.0,
      "step": 823
    },
    {
      "entropy": 0.9851645231246948,
      "epoch": 0.4406417112299465,
      "grad_norm": 0.14308570325374603,
      "learning_rate": 0.0001316956203204634,
      "loss": 0.9748,
      "mean_token_accuracy": 0.7493155598640442,
      "num_tokens": 26870521.0,
      "step": 824
    },
    {
      "entropy": 1.0248220413923264,
      "epoch": 0.4411764705882353,
      "grad_norm": 0.15384997427463531,
      "learning_rate": 0.0001315442336451071,
      "loss": 1.0232,
      "mean_token_accuracy": 0.74516561627388,
      "num_tokens": 26903232.0,
      "step": 825
    },
    {
      "entropy": 0.9613745510578156,
      "epoch": 0.44171122994652406,
      "grad_norm": 0.14238059520721436,
      "learning_rate": 0.00013139278228005947,
      "loss": 0.9549,
      "mean_token_accuracy": 0.7580950558185577,
      "num_tokens": 26936000.0,
      "step": 826
    },
    {
      "entropy": 0.9351648837327957,
      "epoch": 0.44224598930481285,
      "grad_norm": 0.16106504201889038,
      "learning_rate": 0.00013124126668007523,
      "loss": 0.9674,
      "mean_token_accuracy": 0.7571175396442413,
      "num_tokens": 26968768.0,
      "step": 827
    },
    {
      "entropy": 0.9881116300821304,
      "epoch": 0.44278074866310163,
      "grad_norm": 0.15062279999256134,
      "learning_rate": 0.00013108968730010204,
      "loss": 0.9876,
      "mean_token_accuracy": 0.7498167902231216,
      "num_tokens": 27001383.0,
      "step": 828
    },
    {
      "entropy": 0.9285563677549362,
      "epoch": 0.44331550802139036,
      "grad_norm": 0.14957569539546967,
      "learning_rate": 0.00013093804459527893,
      "loss": 0.9313,
      "mean_token_accuracy": 0.7575452029705048,
      "num_tokens": 27034151.0,
      "step": 829
    },
    {
      "entropy": 0.9487057030200958,
      "epoch": 0.44385026737967914,
      "grad_norm": 0.15658754110336304,
      "learning_rate": 0.0001307863390209352,
      "loss": 0.9504,
      "mean_token_accuracy": 0.7548264861106873,
      "num_tokens": 27066919.0,
      "step": 830
    },
    {
      "entropy": 1.002123698592186,
      "epoch": 0.4443850267379679,
      "grad_norm": 0.1467520296573639,
      "learning_rate": 0.0001306345710325888,
      "loss": 1.0041,
      "mean_token_accuracy": 0.7428213655948639,
      "num_tokens": 27099687.0,
      "step": 831
    },
    {
      "entropy": 0.9975673407316208,
      "epoch": 0.4449197860962567,
      "grad_norm": 0.15837520360946655,
      "learning_rate": 0.0001304827410859453,
      "loss": 1.0,
      "mean_token_accuracy": 0.7424547970294952,
      "num_tokens": 27132455.0,
      "step": 832
    },
    {
      "entropy": 0.9692659825086594,
      "epoch": 0.44545454545454544,
      "grad_norm": 0.14246657490730286,
      "learning_rate": 0.00013033084963689603,
      "loss": 0.9597,
      "mean_token_accuracy": 0.7527073472738266,
      "num_tokens": 27165037.0,
      "step": 833
    },
    {
      "entropy": 1.0118547677993774,
      "epoch": 0.4459893048128342,
      "grad_norm": 0.1440034955739975,
      "learning_rate": 0.00013017889714151724,
      "loss": 0.9975,
      "mean_token_accuracy": 0.7490766048431396,
      "num_tokens": 27197774.0,
      "step": 834
    },
    {
      "entropy": 0.9778098613023758,
      "epoch": 0.446524064171123,
      "grad_norm": 0.15016303956508636,
      "learning_rate": 0.00013002688405606828,
      "loss": 0.9959,
      "mean_token_accuracy": 0.7487170100212097,
      "num_tokens": 27230542.0,
      "step": 835
    },
    {
      "entropy": 0.9894908368587494,
      "epoch": 0.4470588235294118,
      "grad_norm": 0.13868270814418793,
      "learning_rate": 0.00012987481083699063,
      "loss": 0.9977,
      "mean_token_accuracy": 0.7494760751724243,
      "num_tokens": 27263147.0,
      "step": 836
    },
    {
      "entropy": 0.9909839183092117,
      "epoch": 0.4475935828877005,
      "grad_norm": 0.14243672788143158,
      "learning_rate": 0.00012972267794090613,
      "loss": 0.9888,
      "mean_token_accuracy": 0.7501832842826843,
      "num_tokens": 27295915.0,
      "step": 837
    },
    {
      "entropy": 0.9736573696136475,
      "epoch": 0.4481283422459893,
      "grad_norm": 0.1431356966495514,
      "learning_rate": 0.0001295704858246159,
      "loss": 0.9723,
      "mean_token_accuracy": 0.7547959387302399,
      "num_tokens": 27328683.0,
      "step": 838
    },
    {
      "entropy": 0.9292502403259277,
      "epoch": 0.4486631016042781,
      "grad_norm": 0.1411665827035904,
      "learning_rate": 0.00012941823494509898,
      "loss": 0.9282,
      "mean_token_accuracy": 0.7612108886241913,
      "num_tokens": 27361451.0,
      "step": 839
    },
    {
      "entropy": 1.0365600287914276,
      "epoch": 0.44919786096256686,
      "grad_norm": 0.13857612013816833,
      "learning_rate": 0.00012926592575951065,
      "loss": 1.0261,
      "mean_token_accuracy": 0.7461815774440765,
      "num_tokens": 27394219.0,
      "step": 840
    },
    {
      "entropy": 0.9583263546228409,
      "epoch": 0.4497326203208556,
      "grad_norm": 0.1433681845664978,
      "learning_rate": 0.00012911355872518142,
      "loss": 0.9393,
      "mean_token_accuracy": 0.7587670981884003,
      "num_tokens": 27426987.0,
      "step": 841
    },
    {
      "entropy": 1.020112156867981,
      "epoch": 0.4502673796791444,
      "grad_norm": 0.14346186816692352,
      "learning_rate": 0.00012896113429961535,
      "loss": 1.0132,
      "mean_token_accuracy": 0.7436766922473907,
      "num_tokens": 27459755.0,
      "step": 842
    },
    {
      "entropy": 1.013671264052391,
      "epoch": 0.45080213903743316,
      "grad_norm": 0.14903905987739563,
      "learning_rate": 0.0001288086529404891,
      "loss": 1.001,
      "mean_token_accuracy": 0.7490814626216888,
      "num_tokens": 27492481.0,
      "step": 843
    },
    {
      "entropy": 0.9899970591068268,
      "epoch": 0.45133689839572194,
      "grad_norm": 0.1441020518541336,
      "learning_rate": 0.00012865611510564994,
      "loss": 0.9945,
      "mean_token_accuracy": 0.7511302530765533,
      "num_tokens": 27525249.0,
      "step": 844
    },
    {
      "entropy": 0.9621583074331284,
      "epoch": 0.45187165775401067,
      "grad_norm": 0.15059061348438263,
      "learning_rate": 0.00012850352125311494,
      "loss": 0.9645,
      "mean_token_accuracy": 0.7551609575748444,
      "num_tokens": 27557373.0,
      "step": 845
    },
    {
      "entropy": 0.9534913152456284,
      "epoch": 0.45240641711229945,
      "grad_norm": 0.1444634199142456,
      "learning_rate": 0.00012835087184106934,
      "loss": 0.9622,
      "mean_token_accuracy": 0.7565398961305618,
      "num_tokens": 27589801.0,
      "step": 846
    },
    {
      "entropy": 0.9190541058778763,
      "epoch": 0.45294117647058824,
      "grad_norm": 0.15014390647411346,
      "learning_rate": 0.00012819816732786513,
      "loss": 0.9335,
      "mean_token_accuracy": 0.7609359622001648,
      "num_tokens": 27622569.0,
      "step": 847
    },
    {
      "entropy": 0.9643352031707764,
      "epoch": 0.453475935828877,
      "grad_norm": 0.1724693328142166,
      "learning_rate": 0.0001280454081720198,
      "loss": 0.9732,
      "mean_token_accuracy": 0.7566371709108353,
      "num_tokens": 27655309.0,
      "step": 848
    },
    {
      "entropy": 1.0093938112258911,
      "epoch": 0.4540106951871658,
      "grad_norm": 0.15161548554897308,
      "learning_rate": 0.00012789259483221493,
      "loss": 1.0115,
      "mean_token_accuracy": 0.7467968165874481,
      "num_tokens": 27687970.0,
      "step": 849
    },
    {
      "entropy": 0.9799563735723495,
      "epoch": 0.45454545454545453,
      "grad_norm": 0.14368638396263123,
      "learning_rate": 0.00012773972776729476,
      "loss": 0.9727,
      "mean_token_accuracy": 0.7515579164028168,
      "num_tokens": 27720738.0,
      "step": 850
    },
    {
      "entropy": 1.012539565563202,
      "epoch": 0.4550802139037433,
      "grad_norm": 0.15072520077228546,
      "learning_rate": 0.00012758680743626483,
      "loss": 1.0122,
      "mean_token_accuracy": 0.74500472843647,
      "num_tokens": 27753417.0,
      "step": 851
    },
    {
      "entropy": 0.9671606421470642,
      "epoch": 0.4556149732620321,
      "grad_norm": 0.1450292468070984,
      "learning_rate": 0.00012743383429829073,
      "loss": 0.9603,
      "mean_token_accuracy": 0.7547348439693451,
      "num_tokens": 27786185.0,
      "step": 852
    },
    {
      "entropy": 0.966804251074791,
      "epoch": 0.4561497326203209,
      "grad_norm": 0.1427607387304306,
      "learning_rate": 0.0001272808088126964,
      "loss": 0.9497,
      "mean_token_accuracy": 0.7570531070232391,
      "num_tokens": 27818856.0,
      "step": 853
    },
    {
      "entropy": 1.018742486834526,
      "epoch": 0.4566844919786096,
      "grad_norm": 0.1700286716222763,
      "learning_rate": 0.00012712773143896326,
      "loss": 1.0105,
      "mean_token_accuracy": 0.7449291348457336,
      "num_tokens": 27851624.0,
      "step": 854
    },
    {
      "entropy": 1.0010544210672379,
      "epoch": 0.4572192513368984,
      "grad_norm": 0.15574532747268677,
      "learning_rate": 0.00012697460263672822,
      "loss": 1.0087,
      "mean_token_accuracy": 0.7441959977149963,
      "num_tokens": 27884392.0,
      "step": 855
    },
    {
      "entropy": 0.9586425423622131,
      "epoch": 0.4577540106951872,
      "grad_norm": 0.14490839838981628,
      "learning_rate": 0.0001268214228657828,
      "loss": 0.9636,
      "mean_token_accuracy": 0.7560789287090302,
      "num_tokens": 27917160.0,
      "step": 856
    },
    {
      "entropy": 0.9910256862640381,
      "epoch": 0.45828877005347596,
      "grad_norm": 0.16581670939922333,
      "learning_rate": 0.00012666819258607158,
      "loss": 1.0053,
      "mean_token_accuracy": 0.7519976049661636,
      "num_tokens": 27949828.0,
      "step": 857
    },
    {
      "entropy": 0.9686436355113983,
      "epoch": 0.4588235294117647,
      "grad_norm": 0.15091009438037872,
      "learning_rate": 0.0001265149122576906,
      "loss": 0.9813,
      "mean_token_accuracy": 0.7518328428268433,
      "num_tokens": 27982596.0,
      "step": 858
    },
    {
      "entropy": 0.9729765504598618,
      "epoch": 0.45935828877005347,
      "grad_norm": 0.15223021805286407,
      "learning_rate": 0.0001263615823408864,
      "loss": 0.9776,
      "mean_token_accuracy": 0.7539650201797485,
      "num_tokens": 28014505.0,
      "step": 859
    },
    {
      "entropy": 0.9859947264194489,
      "epoch": 0.45989304812834225,
      "grad_norm": 0.1753186583518982,
      "learning_rate": 0.00012620820329605433,
      "loss": 0.9838,
      "mean_token_accuracy": 0.7529747784137726,
      "num_tokens": 28046971.0,
      "step": 860
    },
    {
      "entropy": 1.02519690990448,
      "epoch": 0.46042780748663104,
      "grad_norm": 0.15426649153232574,
      "learning_rate": 0.00012605477558373727,
      "loss": 1.0202,
      "mean_token_accuracy": 0.7468536198139191,
      "num_tokens": 28079739.0,
      "step": 861
    },
    {
      "entropy": 0.9805189371109009,
      "epoch": 0.46096256684491976,
      "grad_norm": 0.15347301959991455,
      "learning_rate": 0.00012590129966462418,
      "loss": 0.9741,
      "mean_token_accuracy": 0.7568426132202148,
      "num_tokens": 28112507.0,
      "step": 862
    },
    {
      "entropy": 0.9748925268650055,
      "epoch": 0.46149732620320855,
      "grad_norm": 0.16348502039909363,
      "learning_rate": 0.0001257477759995488,
      "loss": 0.9668,
      "mean_token_accuracy": 0.7535220235586166,
      "num_tokens": 28144969.0,
      "step": 863
    },
    {
      "entropy": 1.0007185190916061,
      "epoch": 0.46203208556149733,
      "grad_norm": 0.13907505571842194,
      "learning_rate": 0.00012559420504948827,
      "loss": 1.0023,
      "mean_token_accuracy": 0.7492653876543045,
      "num_tokens": 28177395.0,
      "step": 864
    },
    {
      "entropy": 0.97891004383564,
      "epoch": 0.4625668449197861,
      "grad_norm": 0.16052573919296265,
      "learning_rate": 0.00012544058727556161,
      "loss": 0.9777,
      "mean_token_accuracy": 0.7531695663928986,
      "num_tokens": 28209795.0,
      "step": 865
    },
    {
      "entropy": 0.9599785953760147,
      "epoch": 0.46310160427807484,
      "grad_norm": 0.1533554643392563,
      "learning_rate": 0.00012528692313902858,
      "loss": 0.9597,
      "mean_token_accuracy": 0.7586449086666107,
      "num_tokens": 28242563.0,
      "step": 866
    },
    {
      "entropy": 0.9907165765762329,
      "epoch": 0.4636363636363636,
      "grad_norm": 0.14271271228790283,
      "learning_rate": 0.00012513321310128804,
      "loss": 0.983,
      "mean_token_accuracy": 0.750079333782196,
      "num_tokens": 28274945.0,
      "step": 867
    },
    {
      "entropy": 1.0088491588830948,
      "epoch": 0.4641711229946524,
      "grad_norm": 0.1458507478237152,
      "learning_rate": 0.00012497945762387672,
      "loss": 1.0089,
      "mean_token_accuracy": 0.7421345859766006,
      "num_tokens": 28307272.0,
      "step": 868
    },
    {
      "entropy": 1.0230762511491776,
      "epoch": 0.4647058823529412,
      "grad_norm": 0.14179329574108124,
      "learning_rate": 0.00012482565716846774,
      "loss": 1.0319,
      "mean_token_accuracy": 0.743267372250557,
      "num_tokens": 28339540.0,
      "step": 869
    },
    {
      "entropy": 1.0096471905708313,
      "epoch": 0.46524064171123,
      "grad_norm": 0.14528143405914307,
      "learning_rate": 0.00012467181219686936,
      "loss": 0.9995,
      "mean_token_accuracy": 0.7422715127468109,
      "num_tokens": 28372308.0,
      "step": 870
    },
    {
      "entropy": 1.0115074217319489,
      "epoch": 0.4657754010695187,
      "grad_norm": 0.14259186387062073,
      "learning_rate": 0.00012451792317102345,
      "loss": 1.0178,
      "mean_token_accuracy": 0.7431919872760773,
      "num_tokens": 28405028.0,
      "step": 871
    },
    {
      "entropy": 1.0354815274477005,
      "epoch": 0.4663101604278075,
      "grad_norm": 0.15655957162380219,
      "learning_rate": 0.00012436399055300415,
      "loss": 1.0393,
      "mean_token_accuracy": 0.739911362528801,
      "num_tokens": 28437386.0,
      "step": 872
    },
    {
      "entropy": 1.028480052947998,
      "epoch": 0.46684491978609627,
      "grad_norm": 0.14866073429584503,
      "learning_rate": 0.00012421001480501648,
      "loss": 1.0383,
      "mean_token_accuracy": 0.7398970276117325,
      "num_tokens": 28469968.0,
      "step": 873
    },
    {
      "entropy": 0.972516804933548,
      "epoch": 0.46737967914438505,
      "grad_norm": 0.1423788070678711,
      "learning_rate": 0.00012405599638939501,
      "loss": 0.9601,
      "mean_token_accuracy": 0.7564070969820023,
      "num_tokens": 28502628.0,
      "step": 874
    },
    {
      "entropy": 1.001707598567009,
      "epoch": 0.4679144385026738,
      "grad_norm": 0.15020547807216644,
      "learning_rate": 0.0001239019357686025,
      "loss": 0.9869,
      "mean_token_accuracy": 0.7530111223459244,
      "num_tokens": 28534961.0,
      "step": 875
    },
    {
      "entropy": 1.0281101614236832,
      "epoch": 0.46844919786096256,
      "grad_norm": 0.1444360762834549,
      "learning_rate": 0.00012374783340522816,
      "loss": 1.0178,
      "mean_token_accuracy": 0.7450513243675232,
      "num_tokens": 28567729.0,
      "step": 876
    },
    {
      "entropy": 0.9586435705423355,
      "epoch": 0.46898395721925135,
      "grad_norm": 0.14336393773555756,
      "learning_rate": 0.0001235936897619869,
      "loss": 0.95,
      "mean_token_accuracy": 0.7565875798463821,
      "num_tokens": 28600231.0,
      "step": 877
    },
    {
      "entropy": 0.9826674312353134,
      "epoch": 0.46951871657754013,
      "grad_norm": 0.14744499325752258,
      "learning_rate": 0.0001234395053017173,
      "loss": 0.975,
      "mean_token_accuracy": 0.7538526058197021,
      "num_tokens": 28632909.0,
      "step": 878
    },
    {
      "entropy": 0.9591492861509323,
      "epoch": 0.47005347593582886,
      "grad_norm": 0.14284496009349823,
      "learning_rate": 0.00012328528048738063,
      "loss": 0.9697,
      "mean_token_accuracy": 0.7575547695159912,
      "num_tokens": 28665573.0,
      "step": 879
    },
    {
      "entropy": 1.0010185837745667,
      "epoch": 0.47058823529411764,
      "grad_norm": 0.14766953885555267,
      "learning_rate": 0.00012313101578205932,
      "loss": 1.0112,
      "mean_token_accuracy": 0.7468580156564713,
      "num_tokens": 28697136.0,
      "step": 880
    },
    {
      "entropy": 0.9989901632070541,
      "epoch": 0.4711229946524064,
      "grad_norm": 0.14356346428394318,
      "learning_rate": 0.0001229767116489555,
      "loss": 0.9938,
      "mean_token_accuracy": 0.7492057681083679,
      "num_tokens": 28729904.0,
      "step": 881
    },
    {
      "entropy": 0.9695397615432739,
      "epoch": 0.4716577540106952,
      "grad_norm": 0.14482583105564117,
      "learning_rate": 0.00012282236855138982,
      "loss": 0.9823,
      "mean_token_accuracy": 0.7534518539905548,
      "num_tokens": 28762672.0,
      "step": 882
    },
    {
      "entropy": 0.9446674883365631,
      "epoch": 0.47219251336898393,
      "grad_norm": 0.15149655938148499,
      "learning_rate": 0.00012266798695279978,
      "loss": 0.9578,
      "mean_token_accuracy": 0.7553152441978455,
      "num_tokens": 28795440.0,
      "step": 883
    },
    {
      "entropy": 0.9982113689184189,
      "epoch": 0.4727272727272727,
      "grad_norm": 0.149799183011055,
      "learning_rate": 0.00012251356731673866,
      "loss": 0.9681,
      "mean_token_accuracy": 0.7516845017671585,
      "num_tokens": 28827741.0,
      "step": 884
    },
    {
      "entropy": 1.0306490808725357,
      "epoch": 0.4732620320855615,
      "grad_norm": 0.15076981484889984,
      "learning_rate": 0.00012235911010687373,
      "loss": 1.039,
      "mean_token_accuracy": 0.7403458207845688,
      "num_tokens": 28859822.0,
      "step": 885
    },
    {
      "entropy": 1.0157879292964935,
      "epoch": 0.4737967914438503,
      "grad_norm": 0.14644651114940643,
      "learning_rate": 0.00012220461578698526,
      "loss": 1.0187,
      "mean_token_accuracy": 0.7441051155328751,
      "num_tokens": 28892138.0,
      "step": 886
    },
    {
      "entropy": 1.0000844895839691,
      "epoch": 0.474331550802139,
      "grad_norm": 0.14986386895179749,
      "learning_rate": 0.00012205008482096487,
      "loss": 1.0087,
      "mean_token_accuracy": 0.7471285462379456,
      "num_tokens": 28924906.0,
      "step": 887
    },
    {
      "entropy": 0.9762436598539352,
      "epoch": 0.4748663101604278,
      "grad_norm": 0.15048319101333618,
      "learning_rate": 0.00012189551767281427,
      "loss": 0.9881,
      "mean_token_accuracy": 0.74590665102005,
      "num_tokens": 28957674.0,
      "step": 888
    },
    {
      "entropy": 0.9624285250902176,
      "epoch": 0.4754010695187166,
      "grad_norm": 0.1492486149072647,
      "learning_rate": 0.00012174091480664376,
      "loss": 0.9728,
      "mean_token_accuracy": 0.7519520968198776,
      "num_tokens": 28990390.0,
      "step": 889
    },
    {
      "entropy": 1.0277507901191711,
      "epoch": 0.47593582887700536,
      "grad_norm": 0.1422109156847,
      "learning_rate": 0.00012158627668667091,
      "loss": 1.0284,
      "mean_token_accuracy": 0.7427111566066742,
      "num_tokens": 29023106.0,
      "step": 890
    },
    {
      "entropy": 1.029417023062706,
      "epoch": 0.4764705882352941,
      "grad_norm": 0.14125901460647583,
      "learning_rate": 0.0001214316037772191,
      "loss": 1.0008,
      "mean_token_accuracy": 0.7472001910209656,
      "num_tokens": 29055779.0,
      "step": 891
    },
    {
      "entropy": 0.9816141426563263,
      "epoch": 0.4770053475935829,
      "grad_norm": 0.1472245454788208,
      "learning_rate": 0.00012127689654271626,
      "loss": 0.9673,
      "mean_token_accuracy": 0.7555326372385025,
      "num_tokens": 29088386.0,
      "step": 892
    },
    {
      "entropy": 0.9743991792201996,
      "epoch": 0.47754010695187166,
      "grad_norm": 0.14505396783351898,
      "learning_rate": 0.00012112215544769332,
      "loss": 0.9704,
      "mean_token_accuracy": 0.75552998483181,
      "num_tokens": 29121084.0,
      "step": 893
    },
    {
      "entropy": 0.984330803155899,
      "epoch": 0.47807486631016044,
      "grad_norm": 0.14463602006435394,
      "learning_rate": 0.00012096738095678287,
      "loss": 0.9661,
      "mean_token_accuracy": 0.7528714537620544,
      "num_tokens": 29153852.0,
      "step": 894
    },
    {
      "entropy": 0.9667259305715561,
      "epoch": 0.4786096256684492,
      "grad_norm": 0.14450709521770477,
      "learning_rate": 0.00012081257353471784,
      "loss": 0.9738,
      "mean_token_accuracy": 0.7543333023786545,
      "num_tokens": 29186546.0,
      "step": 895
    },
    {
      "entropy": 0.9456564784049988,
      "epoch": 0.47914438502673795,
      "grad_norm": 0.14816603064537048,
      "learning_rate": 0.00012065773364633001,
      "loss": 0.9543,
      "mean_token_accuracy": 0.7598668038845062,
      "num_tokens": 29219314.0,
      "step": 896
    },
    {
      "entropy": 0.9788111746311188,
      "epoch": 0.47967914438502673,
      "grad_norm": 0.17237494885921478,
      "learning_rate": 0.00012050286175654861,
      "loss": 0.9996,
      "mean_token_accuracy": 0.7479639053344727,
      "num_tokens": 29251749.0,
      "step": 897
    },
    {
      "entropy": 0.9980193078517914,
      "epoch": 0.4802139037433155,
      "grad_norm": 0.1541370004415512,
      "learning_rate": 0.000120347958330399,
      "loss": 1.0037,
      "mean_token_accuracy": 0.7468230724334717,
      "num_tokens": 29284517.0,
      "step": 898
    },
    {
      "entropy": 0.9955044686794281,
      "epoch": 0.4807486631016043,
      "grad_norm": 0.15029774606227875,
      "learning_rate": 0.00012019302383300121,
      "loss": 0.9974,
      "mean_token_accuracy": 0.7520772218704224,
      "num_tokens": 29317285.0,
      "step": 899
    },
    {
      "entropy": 0.9913774579763412,
      "epoch": 0.48128342245989303,
      "grad_norm": 0.1621326059103012,
      "learning_rate": 0.00012003805872956863,
      "loss": 1.0074,
      "mean_token_accuracy": 0.7462121248245239,
      "num_tokens": 29350053.0,
      "step": 900
    },
    {
      "entropy": 0.9967214167118073,
      "epoch": 0.4818181818181818,
      "grad_norm": 0.17531584203243256,
      "learning_rate": 0.00011988306348540642,
      "loss": 1.0158,
      "mean_token_accuracy": 0.7427083104848862,
      "num_tokens": 29382226.0,
      "step": 901
    },
    {
      "entropy": 1.0519507229328156,
      "epoch": 0.4823529411764706,
      "grad_norm": 0.16274113953113556,
      "learning_rate": 0.00011972803856591036,
      "loss": 1.069,
      "mean_token_accuracy": 0.7385945022106171,
      "num_tokens": 29414591.0,
      "step": 902
    },
    {
      "entropy": 1.022826910018921,
      "epoch": 0.4828877005347594,
      "grad_norm": 0.14667034149169922,
      "learning_rate": 0.00011957298443656528,
      "loss": 1.0044,
      "mean_token_accuracy": 0.746406689286232,
      "num_tokens": 29447242.0,
      "step": 903
    },
    {
      "entropy": 1.0191742479801178,
      "epoch": 0.4834224598930481,
      "grad_norm": 0.1471247673034668,
      "learning_rate": 0.0001194179015629437,
      "loss": 1.0026,
      "mean_token_accuracy": 0.7468249201774597,
      "num_tokens": 29479897.0,
      "step": 904
    },
    {
      "entropy": 1.0249872356653214,
      "epoch": 0.4839572192513369,
      "grad_norm": 0.14551971852779388,
      "learning_rate": 0.00011926279041070452,
      "loss": 1.0099,
      "mean_token_accuracy": 0.7473421692848206,
      "num_tokens": 29512280.0,
      "step": 905
    },
    {
      "entropy": 0.9875836968421936,
      "epoch": 0.4844919786096257,
      "grad_norm": 0.14124327898025513,
      "learning_rate": 0.0001191076514455915,
      "loss": 0.9656,
      "mean_token_accuracy": 0.7548264861106873,
      "num_tokens": 29545048.0,
      "step": 906
    },
    {
      "entropy": 1.0192315578460693,
      "epoch": 0.48502673796791446,
      "grad_norm": 0.14510181546211243,
      "learning_rate": 0.00011895248513343193,
      "loss": 1.0005,
      "mean_token_accuracy": 0.7453512996435165,
      "num_tokens": 29577383.0,
      "step": 907
    },
    {
      "entropy": 0.9869286417961121,
      "epoch": 0.4855614973262032,
      "grad_norm": 0.16035196185112,
      "learning_rate": 0.0001187972919401352,
      "loss": 0.9887,
      "mean_token_accuracy": 0.7483401149511337,
      "num_tokens": 29609831.0,
      "step": 908
    },
    {
      "entropy": 0.9757640361785889,
      "epoch": 0.48609625668449197,
      "grad_norm": 0.17180685698986053,
      "learning_rate": 0.00011864207233169136,
      "loss": 0.9699,
      "mean_token_accuracy": 0.7536656856536865,
      "num_tokens": 29642599.0,
      "step": 909
    },
    {
      "entropy": 1.001309648156166,
      "epoch": 0.48663101604278075,
      "grad_norm": 0.15699726343154907,
      "learning_rate": 0.00011848682677416994,
      "loss": 0.9882,
      "mean_token_accuracy": 0.75,
      "num_tokens": 29675367.0,
      "step": 910
    },
    {
      "entropy": 0.9576734304428101,
      "epoch": 0.48716577540106953,
      "grad_norm": 0.1499813348054886,
      "learning_rate": 0.00011833155573371815,
      "loss": 0.9598,
      "mean_token_accuracy": 0.7557627856731415,
      "num_tokens": 29707958.0,
      "step": 911
    },
    {
      "entropy": 1.0111593008041382,
      "epoch": 0.48770053475935826,
      "grad_norm": 0.14601005613803864,
      "learning_rate": 0.00011817625967655996,
      "loss": 1.0223,
      "mean_token_accuracy": 0.7433928549289703,
      "num_tokens": 29740374.0,
      "step": 912
    },
    {
      "entropy": 0.9894819855690002,
      "epoch": 0.48823529411764705,
      "grad_norm": 0.1447693109512329,
      "learning_rate": 0.00011802093906899432,
      "loss": 1.0035,
      "mean_token_accuracy": 0.7454178929328918,
      "num_tokens": 29773142.0,
      "step": 913
    },
    {
      "entropy": 1.0526953637599945,
      "epoch": 0.48877005347593583,
      "grad_norm": 0.15337249636650085,
      "learning_rate": 0.00011786559437739386,
      "loss": 1.0696,
      "mean_token_accuracy": 0.7307856976985931,
      "num_tokens": 29805910.0,
      "step": 914
    },
    {
      "entropy": 1.005323514342308,
      "epoch": 0.4893048128342246,
      "grad_norm": 0.1650882065296173,
      "learning_rate": 0.00011771022606820364,
      "loss": 1.0051,
      "mean_token_accuracy": 0.7460218220949173,
      "num_tokens": 29838478.0,
      "step": 915
    },
    {
      "entropy": 1.0439465939998627,
      "epoch": 0.4898395721925134,
      "grad_norm": 0.15761952102184296,
      "learning_rate": 0.00011755483460793954,
      "loss": 1.0571,
      "mean_token_accuracy": 0.7354590445756912,
      "num_tokens": 29871196.0,
      "step": 916
    },
    {
      "entropy": 1.0409222543239594,
      "epoch": 0.4903743315508021,
      "grad_norm": 0.14647220075130463,
      "learning_rate": 0.00011739942046318703,
      "loss": 1.0447,
      "mean_token_accuracy": 0.7362842261791229,
      "num_tokens": 29903964.0,
      "step": 917
    },
    {
      "entropy": 0.9852460622787476,
      "epoch": 0.4909090909090909,
      "grad_norm": 0.15946930646896362,
      "learning_rate": 0.00011724398410059962,
      "loss": 0.9707,
      "mean_token_accuracy": 0.7496487200260162,
      "num_tokens": 29936309.0,
      "step": 918
    },
    {
      "entropy": 1.020201325416565,
      "epoch": 0.4914438502673797,
      "grad_norm": 0.17865873873233795,
      "learning_rate": 0.0001170885259868976,
      "loss": 0.9944,
      "mean_token_accuracy": 0.747556209564209,
      "num_tokens": 29969077.0,
      "step": 919
    },
    {
      "entropy": 0.9647395312786102,
      "epoch": 0.4919786096256685,
      "grad_norm": 0.14664915204048157,
      "learning_rate": 0.00011693304658886654,
      "loss": 0.955,
      "mean_token_accuracy": 0.759252518415451,
      "num_tokens": 30001668.0,
      "step": 920
    },
    {
      "entropy": 1.0127326548099518,
      "epoch": 0.4925133689839572,
      "grad_norm": 0.16004888713359833,
      "learning_rate": 0.00011677754637335591,
      "loss": 1.0018,
      "mean_token_accuracy": 0.7477956712245941,
      "num_tokens": 30034131.0,
      "step": 921
    },
    {
      "entropy": 0.998745933175087,
      "epoch": 0.493048128342246,
      "grad_norm": 0.17857477068901062,
      "learning_rate": 0.0001166220258072777,
      "loss": 1.0134,
      "mean_token_accuracy": 0.7447153031826019,
      "num_tokens": 30066899.0,
      "step": 922
    },
    {
      "entropy": 0.9973787516355515,
      "epoch": 0.49358288770053477,
      "grad_norm": 0.14939211308956146,
      "learning_rate": 0.00011646648535760498,
      "loss": 1.0064,
      "mean_token_accuracy": 0.7462234795093536,
      "num_tokens": 30099318.0,
      "step": 923
    },
    {
      "entropy": 1.0201772600412369,
      "epoch": 0.49411764705882355,
      "grad_norm": 0.19249583780765533,
      "learning_rate": 0.00011631092549137057,
      "loss": 1.0283,
      "mean_token_accuracy": 0.7428824603557587,
      "num_tokens": 30132086.0,
      "step": 924
    },
    {
      "entropy": 1.0128138363361359,
      "epoch": 0.4946524064171123,
      "grad_norm": 0.1913338303565979,
      "learning_rate": 0.0001161553466756656,
      "loss": 1.0195,
      "mean_token_accuracy": 0.7421518564224243,
      "num_tokens": 30164631.0,
      "step": 925
    },
    {
      "entropy": 0.9628503024578094,
      "epoch": 0.49518716577540106,
      "grad_norm": 0.15156152844429016,
      "learning_rate": 0.00011599974937763799,
      "loss": 0.9489,
      "mean_token_accuracy": 0.7556818127632141,
      "num_tokens": 30197399.0,
      "step": 926
    },
    {
      "entropy": 1.0384571850299835,
      "epoch": 0.49572192513368984,
      "grad_norm": 0.20053574442863464,
      "learning_rate": 0.00011584413406449128,
      "loss": 1.0471,
      "mean_token_accuracy": 0.7360357940196991,
      "num_tokens": 30230029.0,
      "step": 927
    },
    {
      "entropy": 0.9797585010528564,
      "epoch": 0.49625668449197863,
      "grad_norm": 0.14603500068187714,
      "learning_rate": 0.00011568850120348306,
      "loss": 0.9685,
      "mean_token_accuracy": 0.7514859288930893,
      "num_tokens": 30262449.0,
      "step": 928
    },
    {
      "entropy": 0.9807269275188446,
      "epoch": 0.49679144385026736,
      "grad_norm": 0.1834476888179779,
      "learning_rate": 0.00011553285126192355,
      "loss": 0.9852,
      "mean_token_accuracy": 0.7513086497783661,
      "num_tokens": 30294775.0,
      "step": 929
    },
    {
      "entropy": 1.0138742625713348,
      "epoch": 0.49732620320855614,
      "grad_norm": 0.1444700062274933,
      "learning_rate": 0.00011537718470717438,
      "loss": 0.9956,
      "mean_token_accuracy": 0.7464201748371124,
      "num_tokens": 30327479.0,
      "step": 930
    },
    {
      "entropy": 0.9629910737276077,
      "epoch": 0.4978609625668449,
      "grad_norm": 0.18388797342777252,
      "learning_rate": 0.00011522150200664695,
      "loss": 0.9479,
      "mean_token_accuracy": 0.7568120658397675,
      "num_tokens": 30360247.0,
      "step": 931
    },
    {
      "entropy": 0.987065315246582,
      "epoch": 0.4983957219251337,
      "grad_norm": 0.14784592390060425,
      "learning_rate": 0.00011506580362780121,
      "loss": 0.9769,
      "mean_token_accuracy": 0.7503300607204437,
      "num_tokens": 30392876.0,
      "step": 932
    },
    {
      "entropy": 0.9994514882564545,
      "epoch": 0.49893048128342243,
      "grad_norm": 0.18403013050556183,
      "learning_rate": 0.00011491009003814417,
      "loss": 0.9884,
      "mean_token_accuracy": 0.7486559152603149,
      "num_tokens": 30425644.0,
      "step": 933
    },
    {
      "entropy": 0.9785473197698593,
      "epoch": 0.4994652406417112,
      "grad_norm": 0.163219153881073,
      "learning_rate": 0.00011475436170522842,
      "loss": 0.9865,
      "mean_token_accuracy": 0.7483198940753937,
      "num_tokens": 30458412.0,
      "step": 934
    },
    {
      "entropy": 0.9718093425035477,
      "epoch": 0.5,
      "grad_norm": 0.1547645777463913,
      "learning_rate": 0.00011459861909665106,
      "loss": 0.9693,
      "mean_token_accuracy": 0.7519071251153946,
      "num_tokens": 30491058.0,
      "step": 935
    },
    {
      "entropy": 0.9814210534095764,
      "epoch": 0.5005347593582887,
      "grad_norm": 0.1539863497018814,
      "learning_rate": 0.00011444286268005173,
      "loss": 0.9738,
      "mean_token_accuracy": 0.751496821641922,
      "num_tokens": 30523826.0,
      "step": 936
    },
    {
      "entropy": 0.924762949347496,
      "epoch": 0.5010695187165776,
      "grad_norm": 0.15415582060813904,
      "learning_rate": 0.00011428709292311179,
      "loss": 0.9331,
      "mean_token_accuracy": 0.7617912888526917,
      "num_tokens": 30556594.0,
      "step": 937
    },
    {
      "entropy": 0.9639741778373718,
      "epoch": 0.5016042780748663,
      "grad_norm": 0.16413013637065887,
      "learning_rate": 0.0001141313102935526,
      "loss": 0.9578,
      "mean_token_accuracy": 0.7549447417259216,
      "num_tokens": 30588411.0,
      "step": 938
    },
    {
      "entropy": 0.9839192181825638,
      "epoch": 0.5021390374331551,
      "grad_norm": 0.15187539160251617,
      "learning_rate": 0.00011397551525913403,
      "loss": 0.9943,
      "mean_token_accuracy": 0.7456000745296478,
      "num_tokens": 30621116.0,
      "step": 939
    },
    {
      "entropy": 0.9420085400342941,
      "epoch": 0.5026737967914439,
      "grad_norm": 0.1593165397644043,
      "learning_rate": 0.00011381970828765336,
      "loss": 0.9376,
      "mean_token_accuracy": 0.7606199234724045,
      "num_tokens": 30653222.0,
      "step": 940
    },
    {
      "entropy": 0.9681316465139389,
      "epoch": 0.5032085561497326,
      "grad_norm": 0.1515263170003891,
      "learning_rate": 0.00011366388984694365,
      "loss": 0.9952,
      "mean_token_accuracy": 0.7483191341161728,
      "num_tokens": 30685809.0,
      "step": 941
    },
    {
      "entropy": 0.9541009217500687,
      "epoch": 0.5037433155080214,
      "grad_norm": 0.15999698638916016,
      "learning_rate": 0.00011350806040487245,
      "loss": 0.9562,
      "mean_token_accuracy": 0.7580645084381104,
      "num_tokens": 30718577.0,
      "step": 942
    },
    {
      "entropy": 0.9750600904226303,
      "epoch": 0.5042780748663102,
      "grad_norm": 0.14765378832817078,
      "learning_rate": 0.00011335222042934022,
      "loss": 0.9757,
      "mean_token_accuracy": 0.7525048851966858,
      "num_tokens": 30751345.0,
      "step": 943
    },
    {
      "entropy": 0.9981716424226761,
      "epoch": 0.5048128342245989,
      "grad_norm": 0.15309534966945648,
      "learning_rate": 0.00011319637038827918,
      "loss": 0.9948,
      "mean_token_accuracy": 0.7471605241298676,
      "num_tokens": 30784097.0,
      "step": 944
    },
    {
      "entropy": 0.9929149597883224,
      "epoch": 0.5053475935828877,
      "grad_norm": 0.14857131242752075,
      "learning_rate": 0.0001130405107496517,
      "loss": 0.9993,
      "mean_token_accuracy": 0.7473356425762177,
      "num_tokens": 30816631.0,
      "step": 945
    },
    {
      "entropy": 0.9805185496807098,
      "epoch": 0.5058823529411764,
      "grad_norm": 0.15281623601913452,
      "learning_rate": 0.000112884641981449,
      "loss": 0.9783,
      "mean_token_accuracy": 0.7508858740329742,
      "num_tokens": 30849399.0,
      "step": 946
    },
    {
      "entropy": 1.0177381336688995,
      "epoch": 0.5064171122994653,
      "grad_norm": 0.14827761054039001,
      "learning_rate": 0.00011272876455168965,
      "loss": 1.0084,
      "mean_token_accuracy": 0.7458373308181763,
      "num_tokens": 30882166.0,
      "step": 947
    },
    {
      "entropy": 0.9999857991933823,
      "epoch": 0.506951871657754,
      "grad_norm": 0.14711898565292358,
      "learning_rate": 0.00011257287892841835,
      "loss": 0.9881,
      "mean_token_accuracy": 0.7506720423698425,
      "num_tokens": 30914934.0,
      "step": 948
    },
    {
      "entropy": 1.0082168877124786,
      "epoch": 0.5074866310160427,
      "grad_norm": 0.14489367604255676,
      "learning_rate": 0.00011241698557970428,
      "loss": 1.008,
      "mean_token_accuracy": 0.7478005886077881,
      "num_tokens": 30947702.0,
      "step": 949
    },
    {
      "entropy": 0.9544709473848343,
      "epoch": 0.5080213903743316,
      "grad_norm": 0.1495532989501953,
      "learning_rate": 0.0001122610849736399,
      "loss": 0.9539,
      "mean_token_accuracy": 0.7564558535814285,
      "num_tokens": 30980308.0,
      "step": 950
    },
    {
      "entropy": 0.9981829226016998,
      "epoch": 0.5085561497326203,
      "grad_norm": 0.15191273391246796,
      "learning_rate": 0.0001121051775783394,
      "loss": 1.0066,
      "mean_token_accuracy": 0.7489893287420273,
      "num_tokens": 31012771.0,
      "step": 951
    },
    {
      "entropy": 0.9858329743146896,
      "epoch": 0.509090909090909,
      "grad_norm": 0.15609891712665558,
      "learning_rate": 0.00011194926386193738,
      "loss": 0.9947,
      "mean_token_accuracy": 0.7483371645212173,
      "num_tokens": 31045494.0,
      "step": 952
    },
    {
      "entropy": 0.951867401599884,
      "epoch": 0.5096256684491979,
      "grad_norm": 0.150090754032135,
      "learning_rate": 0.00011179334429258747,
      "loss": 0.9614,
      "mean_token_accuracy": 0.7558956444263458,
      "num_tokens": 31078262.0,
      "step": 953
    },
    {
      "entropy": 0.9978301674127579,
      "epoch": 0.5101604278074866,
      "grad_norm": 0.14746372401714325,
      "learning_rate": 0.00011163741933846077,
      "loss": 0.9908,
      "mean_token_accuracy": 0.7496818602085114,
      "num_tokens": 31110967.0,
      "step": 954
    },
    {
      "entropy": 1.0230722427368164,
      "epoch": 0.5106951871657754,
      "grad_norm": 0.14543987810611725,
      "learning_rate": 0.00011148148946774465,
      "loss": 1.0243,
      "mean_token_accuracy": 0.7438846528530121,
      "num_tokens": 31143570.0,
      "step": 955
    },
    {
      "entropy": 0.9597311317920685,
      "epoch": 0.5112299465240642,
      "grad_norm": 0.1489652544260025,
      "learning_rate": 0.00011132555514864118,
      "loss": 0.9572,
      "mean_token_accuracy": 0.7564149498939514,
      "num_tokens": 31176338.0,
      "step": 956
    },
    {
      "entropy": 1.0460785180330276,
      "epoch": 0.5117647058823529,
      "grad_norm": 0.1502375602722168,
      "learning_rate": 0.00011116961684936582,
      "loss": 1.0382,
      "mean_token_accuracy": 0.7419738918542862,
      "num_tokens": 31208888.0,
      "step": 957
    },
    {
      "entropy": 1.0283286571502686,
      "epoch": 0.5122994652406417,
      "grad_norm": 0.15182124078273773,
      "learning_rate": 0.00011101367503814591,
      "loss": 1.0118,
      "mean_token_accuracy": 0.7439312040805817,
      "num_tokens": 31241523.0,
      "step": 958
    },
    {
      "entropy": 1.035444438457489,
      "epoch": 0.5128342245989305,
      "grad_norm": 0.14735203981399536,
      "learning_rate": 0.00011085773018321947,
      "loss": 1.0272,
      "mean_token_accuracy": 0.744809091091156,
      "num_tokens": 31273776.0,
      "step": 959
    },
    {
      "entropy": 0.9777446836233139,
      "epoch": 0.5133689839572193,
      "grad_norm": 0.16726234555244446,
      "learning_rate": 0.00011070178275283359,
      "loss": 0.97,
      "mean_token_accuracy": 0.7573237866163254,
      "num_tokens": 31306310.0,
      "step": 960
    },
    {
      "entropy": 0.9591176509857178,
      "epoch": 0.513903743315508,
      "grad_norm": 0.16735734045505524,
      "learning_rate": 0.00011054583321524298,
      "loss": 0.9704,
      "mean_token_accuracy": 0.7537226974964142,
      "num_tokens": 31338881.0,
      "step": 961
    },
    {
      "entropy": 0.9389115124940872,
      "epoch": 0.5144385026737968,
      "grad_norm": 0.1522175818681717,
      "learning_rate": 0.00011038988203870888,
      "loss": 0.9487,
      "mean_token_accuracy": 0.7594985961914062,
      "num_tokens": 31371418.0,
      "step": 962
    },
    {
      "entropy": 0.9493906497955322,
      "epoch": 0.5149732620320856,
      "grad_norm": 0.16043366491794586,
      "learning_rate": 0.0001102339296914973,
      "loss": 0.9398,
      "mean_token_accuracy": 0.7587670981884003,
      "num_tokens": 31404186.0,
      "step": 963
    },
    {
      "entropy": 1.0073987543582916,
      "epoch": 0.5155080213903743,
      "grad_norm": 0.15452541410923004,
      "learning_rate": 0.00011007797664187784,
      "loss": 0.9966,
      "mean_token_accuracy": 0.7497439980506897,
      "num_tokens": 31436823.0,
      "step": 964
    },
    {
      "entropy": 0.9574171453714371,
      "epoch": 0.516042780748663,
      "grad_norm": 0.14766879379749298,
      "learning_rate": 0.00010992202335812218,
      "loss": 0.9608,
      "mean_token_accuracy": 0.7573619186878204,
      "num_tokens": 31469591.0,
      "step": 965
    },
    {
      "entropy": 1.042914018034935,
      "epoch": 0.5165775401069519,
      "grad_norm": 0.1512092649936676,
      "learning_rate": 0.00010976607030850275,
      "loss": 1.034,
      "mean_token_accuracy": 0.7408221811056137,
      "num_tokens": 31502350.0,
      "step": 966
    },
    {
      "entropy": 0.9637761563062668,
      "epoch": 0.5171122994652406,
      "grad_norm": 0.15066123008728027,
      "learning_rate": 0.00010961011796129117,
      "loss": 0.9702,
      "mean_token_accuracy": 0.755041629076004,
      "num_tokens": 31535071.0,
      "step": 967
    },
    {
      "entropy": 1.0045630931854248,
      "epoch": 0.5176470588235295,
      "grad_norm": 0.14877843856811523,
      "learning_rate": 0.00010945416678475707,
      "loss": 1.0027,
      "mean_token_accuracy": 0.7474340200424194,
      "num_tokens": 31567839.0,
      "step": 968
    },
    {
      "entropy": 0.9725369513034821,
      "epoch": 0.5181818181818182,
      "grad_norm": 0.15289999544620514,
      "learning_rate": 0.00010929821724716645,
      "loss": 0.9667,
      "mean_token_accuracy": 0.7536260038614273,
      "num_tokens": 31600299.0,
      "step": 969
    },
    {
      "entropy": 0.9684153497219086,
      "epoch": 0.5187165775401069,
      "grad_norm": 0.14860281348228455,
      "learning_rate": 0.00010914226981678055,
      "loss": 0.9442,
      "mean_token_accuracy": 0.7581672668457031,
      "num_tokens": 31632916.0,
      "step": 970
    },
    {
      "entropy": 0.9741877317428589,
      "epoch": 0.5192513368983958,
      "grad_norm": 0.14804121851921082,
      "learning_rate": 0.00010898632496185413,
      "loss": 0.974,
      "mean_token_accuracy": 0.7507331371307373,
      "num_tokens": 31665684.0,
      "step": 971
    },
    {
      "entropy": 0.9431410431861877,
      "epoch": 0.5197860962566845,
      "grad_norm": 0.1512025147676468,
      "learning_rate": 0.00010883038315063421,
      "loss": 0.9523,
      "mean_token_accuracy": 0.7577895820140839,
      "num_tokens": 31698452.0,
      "step": 972
    },
    {
      "entropy": 0.9640598893165588,
      "epoch": 0.5203208556149732,
      "grad_norm": 0.158531054854393,
      "learning_rate": 0.00010867444485135884,
      "loss": 0.9787,
      "mean_token_accuracy": 0.7556378543376923,
      "num_tokens": 31731149.0,
      "step": 973
    },
    {
      "entropy": 0.9679831564426422,
      "epoch": 0.520855614973262,
      "grad_norm": 0.14988689124584198,
      "learning_rate": 0.0001085185105322554,
      "loss": 0.9634,
      "mean_token_accuracy": 0.7536378353834152,
      "num_tokens": 31763506.0,
      "step": 974
    },
    {
      "entropy": 0.9720947295427322,
      "epoch": 0.5213903743315508,
      "grad_norm": 0.14716538786888123,
      "learning_rate": 0.00010836258066153925,
      "loss": 0.974,
      "mean_token_accuracy": 0.7534502297639847,
      "num_tokens": 31796162.0,
      "step": 975
    },
    {
      "entropy": 1.0021846890449524,
      "epoch": 0.5219251336898396,
      "grad_norm": 0.14980396628379822,
      "learning_rate": 0.00010820665570741255,
      "loss": 1.0027,
      "mean_token_accuracy": 0.7444467395544052,
      "num_tokens": 31828756.0,
      "step": 976
    },
    {
      "entropy": 1.0210732370615005,
      "epoch": 0.5224598930481283,
      "grad_norm": 0.14784151315689087,
      "learning_rate": 0.00010805073613806263,
      "loss": 1.0369,
      "mean_token_accuracy": 0.7449291348457336,
      "num_tokens": 31861524.0,
      "step": 977
    },
    {
      "entropy": 1.0197188556194305,
      "epoch": 0.5229946524064171,
      "grad_norm": 0.14898277819156647,
      "learning_rate": 0.00010789482242166065,
      "loss": 1.0113,
      "mean_token_accuracy": 0.7475687712430954,
      "num_tokens": 31893923.0,
      "step": 978
    },
    {
      "entropy": 0.9855650812387466,
      "epoch": 0.5235294117647059,
      "grad_norm": 0.14652405679225922,
      "learning_rate": 0.00010773891502636013,
      "loss": 0.9902,
      "mean_token_accuracy": 0.7477983981370926,
      "num_tokens": 31926465.0,
      "step": 979
    },
    {
      "entropy": 1.0192323327064514,
      "epoch": 0.5240641711229946,
      "grad_norm": 0.15139682590961456,
      "learning_rate": 0.00010758301442029573,
      "loss": 1.0169,
      "mean_token_accuracy": 0.7415623664855957,
      "num_tokens": 31958894.0,
      "step": 980
    },
    {
      "entropy": 1.0423056483268738,
      "epoch": 0.5245989304812835,
      "grad_norm": 0.15091025829315186,
      "learning_rate": 0.00010742712107158169,
      "loss": 1.048,
      "mean_token_accuracy": 0.737475574016571,
      "num_tokens": 31991662.0,
      "step": 981
    },
    {
      "entropy": 1.0017990171909332,
      "epoch": 0.5251336898395722,
      "grad_norm": 0.14979922771453857,
      "learning_rate": 0.0001072712354483104,
      "loss": 0.9875,
      "mean_token_accuracy": 0.7491593360900879,
      "num_tokens": 32024339.0,
      "step": 982
    },
    {
      "entropy": 0.9572014361619949,
      "epoch": 0.5256684491978609,
      "grad_norm": 0.16085804998874664,
      "learning_rate": 0.00010711535801855104,
      "loss": 0.952,
      "mean_token_accuracy": 0.7552261650562286,
      "num_tokens": 32056882.0,
      "step": 983
    },
    {
      "entropy": 0.9995868802070618,
      "epoch": 0.5262032085561498,
      "grad_norm": 0.15602126717567444,
      "learning_rate": 0.00010695948925034833,
      "loss": 1.0072,
      "mean_token_accuracy": 0.7474450767040253,
      "num_tokens": 32089485.0,
      "step": 984
    },
    {
      "entropy": 1.0375087559223175,
      "epoch": 0.5267379679144385,
      "grad_norm": 0.14877036213874817,
      "learning_rate": 0.00010680362961172086,
      "loss": 1.0437,
      "mean_token_accuracy": 0.7392778694629669,
      "num_tokens": 32122253.0,
      "step": 985
    },
    {
      "entropy": 0.9701367020606995,
      "epoch": 0.5272727272727272,
      "grad_norm": 0.15291082859039307,
      "learning_rate": 0.00010664777957065981,
      "loss": 0.969,
      "mean_token_accuracy": 0.7528199255466461,
      "num_tokens": 32154959.0,
      "step": 986
    },
    {
      "entropy": 0.9755905866622925,
      "epoch": 0.5278074866310161,
      "grad_norm": 0.15403801202774048,
      "learning_rate": 0.00010649193959512758,
      "loss": 0.9582,
      "mean_token_accuracy": 0.7578248232603073,
      "num_tokens": 32187679.0,
      "step": 987
    },
    {
      "entropy": 0.9252783358097076,
      "epoch": 0.5283422459893048,
      "grad_norm": 0.15754017233848572,
      "learning_rate": 0.00010633611015305636,
      "loss": 0.9317,
      "mean_token_accuracy": 0.7646315395832062,
      "num_tokens": 32220116.0,
      "step": 988
    },
    {
      "entropy": 0.9623700231313705,
      "epoch": 0.5288770053475936,
      "grad_norm": 0.16240595281124115,
      "learning_rate": 0.00010618029171234669,
      "loss": 0.9608,
      "mean_token_accuracy": 0.7552236020565033,
      "num_tokens": 32252884.0,
      "step": 989
    },
    {
      "entropy": 0.9905337542295456,
      "epoch": 0.5294117647058824,
      "grad_norm": 0.19256940484046936,
      "learning_rate": 0.00010602448474086602,
      "loss": 0.99,
      "mean_token_accuracy": 0.7485212236642838,
      "num_tokens": 32285038.0,
      "step": 990
    },
    {
      "entropy": 1.048939511179924,
      "epoch": 0.5299465240641711,
      "grad_norm": 0.15117883682250977,
      "learning_rate": 0.00010586868970644745,
      "loss": 1.0577,
      "mean_token_accuracy": 0.7386181652545929,
      "num_tokens": 32317800.0,
      "step": 991
    },
    {
      "entropy": 0.953388124704361,
      "epoch": 0.5304812834224599,
      "grad_norm": 0.1687232106924057,
      "learning_rate": 0.00010571290707688823,
      "loss": 0.9412,
      "mean_token_accuracy": 0.7602576017379761,
      "num_tokens": 32349982.0,
      "step": 992
    },
    {
      "entropy": 0.9880169332027435,
      "epoch": 0.5310160427807487,
      "grad_norm": 0.15983946621418,
      "learning_rate": 0.00010555713731994829,
      "loss": 0.9676,
      "mean_token_accuracy": 0.7497250735759735,
      "num_tokens": 32382750.0,
      "step": 993
    },
    {
      "entropy": 1.0043831318616867,
      "epoch": 0.5315508021390374,
      "grad_norm": 0.14981651306152344,
      "learning_rate": 0.00010540138090334897,
      "loss": 1.0026,
      "mean_token_accuracy": 0.7486239373683929,
      "num_tokens": 32415099.0,
      "step": 994
    },
    {
      "entropy": 1.0175066590309143,
      "epoch": 0.5320855614973262,
      "grad_norm": 0.16312158107757568,
      "learning_rate": 0.00010524563829477157,
      "loss": 1.025,
      "mean_token_accuracy": 0.7404081225395203,
      "num_tokens": 32447867.0,
      "step": 995
    },
    {
      "entropy": 0.9783085435628891,
      "epoch": 0.532620320855615,
      "grad_norm": 0.1860574185848236,
      "learning_rate": 0.0001050899099618559,
      "loss": 0.9823,
      "mean_token_accuracy": 0.7511302530765533,
      "num_tokens": 32480635.0,
      "step": 996
    },
    {
      "entropy": 1.0055098831653595,
      "epoch": 0.5331550802139038,
      "grad_norm": 0.1509351283311844,
      "learning_rate": 0.00010493419637219881,
      "loss": 1.0153,
      "mean_token_accuracy": 0.7382027506828308,
      "num_tokens": 32513121.0,
      "step": 997
    },
    {
      "entropy": 0.9564041495323181,
      "epoch": 0.5336898395721925,
      "grad_norm": 0.19098837673664093,
      "learning_rate": 0.00010477849799335306,
      "loss": 0.9433,
      "mean_token_accuracy": 0.7583567798137665,
      "num_tokens": 32545807.0,
      "step": 998
    },
    {
      "entropy": 0.9641523063182831,
      "epoch": 0.5342245989304812,
      "grad_norm": 0.1565517634153366,
      "learning_rate": 0.00010462281529282567,
      "loss": 0.9567,
      "mean_token_accuracy": 0.7590420246124268,
      "num_tokens": 32578575.0,
      "step": 999
    },
    {
      "entropy": 0.9831237494945526,
      "epoch": 0.5347593582887701,
      "grad_norm": 0.1680804342031479,
      "learning_rate": 0.00010446714873807648,
      "loss": 0.974,
      "mean_token_accuracy": 0.7528853267431259,
      "num_tokens": 32610897.0,
      "step": 1000
    },
    {
      "entropy": 0.9834153801202774,
      "epoch": 0.5352941176470588,
      "grad_norm": 0.17859195172786713,
      "learning_rate": 0.00010431149879651699,
      "loss": 0.9688,
      "mean_token_accuracy": 0.7543960511684418,
      "num_tokens": 32643225.0,
      "step": 1001
    },
    {
      "entropy": 0.9904739707708359,
      "epoch": 0.5358288770053476,
      "grad_norm": 0.15401509404182434,
      "learning_rate": 0.00010415586593550874,
      "loss": 0.9977,
      "mean_token_accuracy": 0.752443790435791,
      "num_tokens": 32675993.0,
      "step": 1002
    },
    {
      "entropy": 1.036316692829132,
      "epoch": 0.5363636363636364,
      "grad_norm": 0.18634629249572754,
      "learning_rate": 0.00010400025062236205,
      "loss": 1.0428,
      "mean_token_accuracy": 0.7439382523298264,
      "num_tokens": 32708720.0,
      "step": 1003
    },
    {
      "entropy": 0.9291573613882065,
      "epoch": 0.5368983957219251,
      "grad_norm": 0.16239655017852783,
      "learning_rate": 0.00010384465332433444,
      "loss": 0.9462,
      "mean_token_accuracy": 0.7608358263969421,
      "num_tokens": 32741274.0,
      "step": 1004
    },
    {
      "entropy": 0.9835901707410812,
      "epoch": 0.5374331550802139,
      "grad_norm": 0.15718552470207214,
      "learning_rate": 0.00010368907450862944,
      "loss": 0.994,
      "mean_token_accuracy": 0.7442265450954437,
      "num_tokens": 32774042.0,
      "step": 1005
    },
    {
      "entropy": 0.9875447750091553,
      "epoch": 0.5379679144385027,
      "grad_norm": 0.14904296398162842,
      "learning_rate": 0.00010353351464239502,
      "loss": 0.9828,
      "mean_token_accuracy": 0.7503054738044739,
      "num_tokens": 32806810.0,
      "step": 1006
    },
    {
      "entropy": 0.9224865585565567,
      "epoch": 0.5385026737967914,
      "grad_norm": 0.1498638242483139,
      "learning_rate": 0.00010337797419272232,
      "loss": 0.9247,
      "mean_token_accuracy": 0.7628298997879028,
      "num_tokens": 32839578.0,
      "step": 1007
    },
    {
      "entropy": 0.9844998568296432,
      "epoch": 0.5390374331550802,
      "grad_norm": 0.14565370976924896,
      "learning_rate": 0.00010322245362664411,
      "loss": 0.9869,
      "mean_token_accuracy": 0.7503665685653687,
      "num_tokens": 32872346.0,
      "step": 1008
    },
    {
      "entropy": 0.9935092180967331,
      "epoch": 0.539572192513369,
      "grad_norm": 0.14800257980823517,
      "learning_rate": 0.00010306695341113348,
      "loss": 0.9729,
      "mean_token_accuracy": 0.7524667382240295,
      "num_tokens": 32904790.0,
      "step": 1009
    },
    {
      "entropy": 0.9884016215801239,
      "epoch": 0.5401069518716578,
      "grad_norm": 0.1749376505613327,
      "learning_rate": 0.00010291147401310244,
      "loss": 0.9811,
      "mean_token_accuracy": 0.7519508004188538,
      "num_tokens": 32937098.0,
      "step": 1010
    },
    {
      "entropy": 0.9943345189094543,
      "epoch": 0.5406417112299465,
      "grad_norm": 0.16141724586486816,
      "learning_rate": 0.0001027560158994004,
      "loss": 0.9946,
      "mean_token_accuracy": 0.7510824054479599,
      "num_tokens": 32969322.0,
      "step": 1011
    },
    {
      "entropy": 0.9604488164186478,
      "epoch": 0.5411764705882353,
      "grad_norm": 0.1509343534708023,
      "learning_rate": 0.00010260057953681299,
      "loss": 0.9533,
      "mean_token_accuracy": 0.7595346719026566,
      "num_tokens": 33002009.0,
      "step": 1012
    },
    {
      "entropy": 1.0147838592529297,
      "epoch": 0.5417112299465241,
      "grad_norm": 0.1797001212835312,
      "learning_rate": 0.00010244516539206048,
      "loss": 1.0128,
      "mean_token_accuracy": 0.7456928193569183,
      "num_tokens": 33034777.0,
      "step": 1013
    },
    {
      "entropy": 0.9897986203432083,
      "epoch": 0.5422459893048128,
      "grad_norm": 0.15886810421943665,
      "learning_rate": 0.00010228977393179641,
      "loss": 0.995,
      "mean_token_accuracy": 0.7470195144414902,
      "num_tokens": 33067285.0,
      "step": 1014
    },
    {
      "entropy": 0.9278879165649414,
      "epoch": 0.5427807486631016,
      "grad_norm": 0.15260817110538483,
      "learning_rate": 0.00010213440562260616,
      "loss": 0.9347,
      "mean_token_accuracy": 0.7599593251943588,
      "num_tokens": 33099583.0,
      "step": 1015
    },
    {
      "entropy": 0.9589514136314392,
      "epoch": 0.5433155080213904,
      "grad_norm": 0.16330361366271973,
      "learning_rate": 0.0001019790609310057,
      "loss": 0.9709,
      "mean_token_accuracy": 0.7548265010118484,
      "num_tokens": 33132351.0,
      "step": 1016
    },
    {
      "entropy": 0.9311648309230804,
      "epoch": 0.5438502673796791,
      "grad_norm": 0.15833255648612976,
      "learning_rate": 0.00010182374032344005,
      "loss": 0.9228,
      "mean_token_accuracy": 0.7587970197200775,
      "num_tokens": 33165070.0,
      "step": 1017
    },
    {
      "entropy": 0.9727670103311539,
      "epoch": 0.544385026737968,
      "grad_norm": 0.15479743480682373,
      "learning_rate": 0.00010166844426628186,
      "loss": 0.9725,
      "mean_token_accuracy": 0.7509775161743164,
      "num_tokens": 33197838.0,
      "step": 1018
    },
    {
      "entropy": 0.9768361747264862,
      "epoch": 0.5449197860962567,
      "grad_norm": 0.18230433762073517,
      "learning_rate": 0.0001015131732258301,
      "loss": 0.9838,
      "mean_token_accuracy": 0.7502443790435791,
      "num_tokens": 33230606.0,
      "step": 1019
    },
    {
      "entropy": 1.0158773809671402,
      "epoch": 0.5454545454545454,
      "grad_norm": 0.17845484614372253,
      "learning_rate": 0.00010135792766830865,
      "loss": 1.0109,
      "mean_token_accuracy": 0.7454035729169846,
      "num_tokens": 33263126.0,
      "step": 1020
    },
    {
      "entropy": 1.011165291070938,
      "epoch": 0.5459893048128343,
      "grad_norm": 0.15542946755886078,
      "learning_rate": 0.00010120270805986487,
      "loss": 1.0111,
      "mean_token_accuracy": 0.7446542084217072,
      "num_tokens": 33295894.0,
      "step": 1021
    },
    {
      "entropy": 0.9618840366601944,
      "epoch": 0.546524064171123,
      "grad_norm": 0.19388443231582642,
      "learning_rate": 0.00010104751486656809,
      "loss": 0.9531,
      "mean_token_accuracy": 0.7557734698057175,
      "num_tokens": 33328662.0,
      "step": 1022
    },
    {
      "entropy": 0.9803057909011841,
      "epoch": 0.5470588235294118,
      "grad_norm": 0.1536754071712494,
      "learning_rate": 0.00010089234855440852,
      "loss": 0.9603,
      "mean_token_accuracy": 0.7563233077526093,
      "num_tokens": 33361430.0,
      "step": 1023
    },
    {
      "entropy": 1.0168762803077698,
      "epoch": 0.5475935828877005,
      "grad_norm": 0.2355695366859436,
      "learning_rate": 0.00010073720958929548,
      "loss": 1.0217,
      "mean_token_accuracy": 0.7425296157598495,
      "num_tokens": 33393519.0,
      "step": 1024
    },
    {
      "entropy": 1.0100078284740448,
      "epoch": 0.5481283422459893,
      "grad_norm": 0.17015263438224792,
      "learning_rate": 0.00010058209843705632,
      "loss": 1.0212,
      "mean_token_accuracy": 0.7407746911048889,
      "num_tokens": 33426287.0,
      "step": 1025
    },
    {
      "entropy": 0.978960245847702,
      "epoch": 0.5486631016042781,
      "grad_norm": 0.16321249306201935,
      "learning_rate": 0.00010042701556343477,
      "loss": 0.9714,
      "mean_token_accuracy": 0.7537573277950287,
      "num_tokens": 33459055.0,
      "step": 1026
    },
    {
      "entropy": 1.0137168616056442,
      "epoch": 0.5491978609625668,
      "grad_norm": 0.16961227357387543,
      "learning_rate": 0.00010027196143408968,
      "loss": 1.0129,
      "mean_token_accuracy": 0.7448566257953644,
      "num_tokens": 33491781.0,
      "step": 1027
    },
    {
      "entropy": 0.9298945963382721,
      "epoch": 0.5497326203208556,
      "grad_norm": 0.14900726079940796,
      "learning_rate": 0.00010011693651459362,
      "loss": 0.9123,
      "mean_token_accuracy": 0.7624548673629761,
      "num_tokens": 33524362.0,
      "step": 1028
    },
    {
      "entropy": 0.9306894838809967,
      "epoch": 0.5502673796791444,
      "grad_norm": 0.16008970141410828,
      "learning_rate": 9.996194127043142e-05,
      "loss": 0.9353,
      "mean_token_accuracy": 0.759422555565834,
      "num_tokens": 33556506.0,
      "step": 1029
    },
    {
      "entropy": 0.973898783326149,
      "epoch": 0.5508021390374331,
      "grad_norm": 0.1580318957567215,
      "learning_rate": 9.980697616699878e-05,
      "loss": 0.9867,
      "mean_token_accuracy": 0.7452651560306549,
      "num_tokens": 33589274.0,
      "step": 1030
    },
    {
      "entropy": 0.9807162135839462,
      "epoch": 0.551336898395722,
      "grad_norm": 0.1588246375322342,
      "learning_rate": 9.9652041669601e-05,
      "loss": 0.972,
      "mean_token_accuracy": 0.7508553266525269,
      "num_tokens": 33622042.0,
      "step": 1031
    },
    {
      "entropy": 0.9518315494060516,
      "epoch": 0.5518716577540107,
      "grad_norm": 0.16469627618789673,
      "learning_rate": 9.949713824345142e-05,
      "loss": 0.9637,
      "mean_token_accuracy": 0.7573671042919159,
      "num_tokens": 33654776.0,
      "step": 1032
    },
    {
      "entropy": 1.003676250576973,
      "epoch": 0.5524064171122994,
      "grad_norm": 0.16015905141830444,
      "learning_rate": 9.934226635367001e-05,
      "loss": 1.0174,
      "mean_token_accuracy": 0.7440948337316513,
      "num_tokens": 33687412.0,
      "step": 1033
    },
    {
      "entropy": 0.9762967377901077,
      "epoch": 0.5529411764705883,
      "grad_norm": 0.16960027813911438,
      "learning_rate": 9.918742646528217e-05,
      "loss": 0.9947,
      "mean_token_accuracy": 0.7487170100212097,
      "num_tokens": 33720180.0,
      "step": 1034
    },
    {
      "entropy": 1.021599069237709,
      "epoch": 0.553475935828877,
      "grad_norm": 0.15136435627937317,
      "learning_rate": 9.903261904321713e-05,
      "loss": 1.0245,
      "mean_token_accuracy": 0.7440814822912216,
      "num_tokens": 33752595.0,
      "step": 1035
    },
    {
      "entropy": 0.9720664918422699,
      "epoch": 0.5540106951871657,
      "grad_norm": 0.15822596848011017,
      "learning_rate": 9.887784455230673e-05,
      "loss": 0.9758,
      "mean_token_accuracy": 0.7544098794460297,
      "num_tokens": 33785146.0,
      "step": 1036
    },
    {
      "entropy": 1.010760858654976,
      "epoch": 0.5545454545454546,
      "grad_norm": 0.15759292244911194,
      "learning_rate": 9.872310345728375e-05,
      "loss": 0.999,
      "mean_token_accuracy": 0.744012713432312,
      "num_tokens": 33817914.0,
      "step": 1037
    },
    {
      "entropy": 0.9856355041265488,
      "epoch": 0.5550802139037433,
      "grad_norm": 0.15385037660598755,
      "learning_rate": 9.856839622278091e-05,
      "loss": 0.9708,
      "mean_token_accuracy": 0.752427726984024,
      "num_tokens": 33850642.0,
      "step": 1038
    },
    {
      "entropy": 0.9753337502479553,
      "epoch": 0.5556149732620321,
      "grad_norm": 0.15805310010910034,
      "learning_rate": 9.841372331332915e-05,
      "loss": 0.9631,
      "mean_token_accuracy": 0.758213609457016,
      "num_tokens": 33883110.0,
      "step": 1039
    },
    {
      "entropy": 0.9520341902971268,
      "epoch": 0.5561497326203209,
      "grad_norm": 0.15785223245620728,
      "learning_rate": 9.825908519335624e-05,
      "loss": 0.9572,
      "mean_token_accuracy": 0.7571136206388474,
      "num_tokens": 33915493.0,
      "step": 1040
    },
    {
      "entropy": 1.0030668526887894,
      "epoch": 0.5566844919786096,
      "grad_norm": 0.18076303601264954,
      "learning_rate": 9.810448232718574e-05,
      "loss": 0.9865,
      "mean_token_accuracy": 0.7505433708429337,
      "num_tokens": 33948105.0,
      "step": 1041
    },
    {
      "entropy": 1.00014129281044,
      "epoch": 0.5572192513368984,
      "grad_norm": 0.15228135883808136,
      "learning_rate": 9.794991517903513e-05,
      "loss": 1.0084,
      "mean_token_accuracy": 0.7415904402732849,
      "num_tokens": 33980868.0,
      "step": 1042
    },
    {
      "entropy": 0.9635073244571686,
      "epoch": 0.5577540106951872,
      "grad_norm": 0.17229047417640686,
      "learning_rate": 9.779538421301478e-05,
      "loss": 0.9665,
      "mean_token_accuracy": 0.7538184225559235,
      "num_tokens": 34013636.0,
      "step": 1043
    },
    {
      "entropy": 0.9166641980409622,
      "epoch": 0.558288770053476,
      "grad_norm": 0.15340466797351837,
      "learning_rate": 9.76408898931263e-05,
      "loss": 0.9141,
      "mean_token_accuracy": 0.7659208625555038,
      "num_tokens": 34046231.0,
      "step": 1044
    },
    {
      "entropy": 0.991406574845314,
      "epoch": 0.5588235294117647,
      "grad_norm": 0.1854625940322876,
      "learning_rate": 9.748643268326138e-05,
      "loss": 0.9961,
      "mean_token_accuracy": 0.7477089464664459,
      "num_tokens": 34078999.0,
      "step": 1045
    },
    {
      "entropy": 0.9557927846908569,
      "epoch": 0.5593582887700534,
      "grad_norm": 0.14812085032463074,
      "learning_rate": 9.733201304720018e-05,
      "loss": 0.9597,
      "mean_token_accuracy": 0.753896564245224,
      "num_tokens": 34111761.0,
      "step": 1046
    },
    {
      "entropy": 0.9458355009555817,
      "epoch": 0.5598930481283423,
      "grad_norm": 0.21885894238948822,
      "learning_rate": 9.717763144861018e-05,
      "loss": 0.9483,
      "mean_token_accuracy": 0.7567524760961533,
      "num_tokens": 34144427.0,
      "step": 1047
    },
    {
      "entropy": 0.9695868790149689,
      "epoch": 0.560427807486631,
      "grad_norm": 0.174393892288208,
      "learning_rate": 9.702328835104451e-05,
      "loss": 0.9599,
      "mean_token_accuracy": 0.7538753300905228,
      "num_tokens": 34177075.0,
      "step": 1048
    },
    {
      "entropy": 0.9795443713665009,
      "epoch": 0.5609625668449197,
      "grad_norm": 0.19289985299110413,
      "learning_rate": 9.68689842179407e-05,
      "loss": 0.969,
      "mean_token_accuracy": 0.7516495585441589,
      "num_tokens": 34209843.0,
      "step": 1049
    },
    {
      "entropy": 0.9906342476606369,
      "epoch": 0.5614973262032086,
      "grad_norm": 0.18270674347877502,
      "learning_rate": 9.671471951261939e-05,
      "loss": 0.992,
      "mean_token_accuracy": 0.7464437782764435,
      "num_tokens": 34241903.0,
      "step": 1050
    },
    {
      "entropy": 0.9662675708532333,
      "epoch": 0.5620320855614973,
      "grad_norm": 0.19764238595962524,
      "learning_rate": 9.65604946982827e-05,
      "loss": 0.9758,
      "mean_token_accuracy": 0.7566316574811935,
      "num_tokens": 34274573.0,
      "step": 1051
    },
    {
      "entropy": 0.9911040216684341,
      "epoch": 0.5625668449197861,
      "grad_norm": 0.18982456624507904,
      "learning_rate": 9.640631023801312e-05,
      "loss": 0.9882,
      "mean_token_accuracy": 0.7486253678798676,
      "num_tokens": 34307341.0,
      "step": 1052
    },
    {
      "entropy": 0.9605313539505005,
      "epoch": 0.5631016042780749,
      "grad_norm": 0.1607426106929779,
      "learning_rate": 9.62521665947718e-05,
      "loss": 0.946,
      "mean_token_accuracy": 0.7591642141342163,
      "num_tokens": 34340109.0,
      "step": 1053
    },
    {
      "entropy": 1.0086674988269806,
      "epoch": 0.5636363636363636,
      "grad_norm": 0.16547992825508118,
      "learning_rate": 9.609806423139756e-05,
      "loss": 1.0119,
      "mean_token_accuracy": 0.7409510314464569,
      "num_tokens": 34372763.0,
      "step": 1054
    },
    {
      "entropy": 0.9168087691068649,
      "epoch": 0.5641711229946524,
      "grad_norm": 0.16736042499542236,
      "learning_rate": 9.594400361060501e-05,
      "loss": 0.9146,
      "mean_token_accuracy": 0.7686299085617065,
      "num_tokens": 34404772.0,
      "step": 1055
    },
    {
      "entropy": 1.0090791881084442,
      "epoch": 0.5647058823529412,
      "grad_norm": 0.17390309274196625,
      "learning_rate": 9.578998519498352e-05,
      "loss": 1.01,
      "mean_token_accuracy": 0.7473423779010773,
      "num_tokens": 34437540.0,
      "step": 1056
    },
    {
      "entropy": 0.9466338604688644,
      "epoch": 0.5652406417112299,
      "grad_norm": 0.15810315310955048,
      "learning_rate": 9.563600944699588e-05,
      "loss": 0.9524,
      "mean_token_accuracy": 0.7534824013710022,
      "num_tokens": 34470308.0,
      "step": 1057
    },
    {
      "entropy": 0.98748779296875,
      "epoch": 0.5657754010695187,
      "grad_norm": 0.167922243475914,
      "learning_rate": 9.548207682897656e-05,
      "loss": 0.9992,
      "mean_token_accuracy": 0.7505677789449692,
      "num_tokens": 34502945.0,
      "step": 1058
    },
    {
      "entropy": 0.9745074212551117,
      "epoch": 0.5663101604278075,
      "grad_norm": 0.16281603276729584,
      "learning_rate": 9.532818780313064e-05,
      "loss": 0.9697,
      "mean_token_accuracy": 0.7572364807128906,
      "num_tokens": 34535692.0,
      "step": 1059
    },
    {
      "entropy": 0.9525927603244781,
      "epoch": 0.5668449197860963,
      "grad_norm": 0.1510886698961258,
      "learning_rate": 9.517434283153224e-05,
      "loss": 0.9686,
      "mean_token_accuracy": 0.7546126544475555,
      "num_tokens": 34568460.0,
      "step": 1060
    },
    {
      "entropy": 0.9627717286348343,
      "epoch": 0.567379679144385,
      "grad_norm": 0.2195834219455719,
      "learning_rate": 9.502054237612332e-05,
      "loss": 0.9488,
      "mean_token_accuracy": 0.7562622129917145,
      "num_tokens": 34601228.0,
      "step": 1061
    },
    {
      "entropy": 0.9885230660438538,
      "epoch": 0.5679144385026738,
      "grad_norm": 0.16450177133083344,
      "learning_rate": 9.486678689871199e-05,
      "loss": 0.9812,
      "mean_token_accuracy": 0.7506720423698425,
      "num_tokens": 34633996.0,
      "step": 1062
    },
    {
      "entropy": 0.9659951478242874,
      "epoch": 0.5684491978609626,
      "grad_norm": 0.18741483986377716,
      "learning_rate": 9.471307686097142e-05,
      "loss": 0.9614,
      "mean_token_accuracy": 0.7537059783935547,
      "num_tokens": 34666458.0,
      "step": 1063
    },
    {
      "entropy": 1.0185732692480087,
      "epoch": 0.5689839572192513,
      "grad_norm": 0.16106286644935608,
      "learning_rate": 9.455941272443838e-05,
      "loss": 1.0255,
      "mean_token_accuracy": 0.7418881952762604,
      "num_tokens": 34698844.0,
      "step": 1064
    },
    {
      "entropy": 0.9865347892045975,
      "epoch": 0.56951871657754,
      "grad_norm": 0.1745121031999588,
      "learning_rate": 9.440579495051178e-05,
      "loss": 0.9701,
      "mean_token_accuracy": 0.7532685697078705,
      "num_tokens": 34731612.0,
      "step": 1065
    },
    {
      "entropy": 0.9505593776702881,
      "epoch": 0.5700534759358289,
      "grad_norm": 0.14933449029922485,
      "learning_rate": 9.425222400045124e-05,
      "loss": 0.9413,
      "mean_token_accuracy": 0.7585378736257553,
      "num_tokens": 34764209.0,
      "step": 1066
    },
    {
      "entropy": 0.996794581413269,
      "epoch": 0.5705882352941176,
      "grad_norm": 0.18804962933063507,
      "learning_rate": 9.409870033537584e-05,
      "loss": 0.9973,
      "mean_token_accuracy": 0.7489432394504547,
      "num_tokens": 34796870.0,
      "step": 1067
    },
    {
      "entropy": 0.9904070794582367,
      "epoch": 0.5711229946524065,
      "grad_norm": 0.1591782420873642,
      "learning_rate": 9.394522441626276e-05,
      "loss": 0.9844,
      "mean_token_accuracy": 0.7517068386077881,
      "num_tokens": 34829618.0,
      "step": 1068
    },
    {
      "entropy": 0.9387254118919373,
      "epoch": 0.5716577540106952,
      "grad_norm": 0.19616670906543732,
      "learning_rate": 9.379179670394566e-05,
      "loss": 0.9328,
      "mean_token_accuracy": 0.7617285847663879,
      "num_tokens": 34862260.0,
      "step": 1069
    },
    {
      "entropy": 0.9555445462465286,
      "epoch": 0.5721925133689839,
      "grad_norm": 0.1653156578540802,
      "learning_rate": 9.36384176591136e-05,
      "loss": 0.9705,
      "mean_token_accuracy": 0.7522734850645065,
      "num_tokens": 34895013.0,
      "step": 1070
    },
    {
      "entropy": 0.9247808158397675,
      "epoch": 0.5727272727272728,
      "grad_norm": 0.1534564346075058,
      "learning_rate": 9.348508774230942e-05,
      "loss": 0.9272,
      "mean_token_accuracy": 0.7621528804302216,
      "num_tokens": 34927693.0,
      "step": 1071
    },
    {
      "entropy": 0.9609176367521286,
      "epoch": 0.5732620320855615,
      "grad_norm": 0.18071511387825012,
      "learning_rate": 9.333180741392849e-05,
      "loss": 0.9683,
      "mean_token_accuracy": 0.7553821057081223,
      "num_tokens": 34959731.0,
      "step": 1072
    },
    {
      "entropy": 0.9573362022638321,
      "epoch": 0.5737967914438503,
      "grad_norm": 0.15922591090202332,
      "learning_rate": 9.317857713421723e-05,
      "loss": 0.9664,
      "mean_token_accuracy": 0.7546432018280029,
      "num_tokens": 34992499.0,
      "step": 1073
    },
    {
      "entropy": 0.9590311795473099,
      "epoch": 0.574331550802139,
      "grad_norm": 0.21154066920280457,
      "learning_rate": 9.30253973632718e-05,
      "loss": 0.9667,
      "mean_token_accuracy": 0.7547042965888977,
      "num_tokens": 35025267.0,
      "step": 1074
    },
    {
      "entropy": 0.9398636966943741,
      "epoch": 0.5748663101604278,
      "grad_norm": 0.15687710046768188,
      "learning_rate": 9.287226856103677e-05,
      "loss": 0.955,
      "mean_token_accuracy": 0.756537139415741,
      "num_tokens": 35058035.0,
      "step": 1075
    },
    {
      "entropy": 1.0047118067741394,
      "epoch": 0.5754010695187166,
      "grad_norm": 0.17438744008541107,
      "learning_rate": 9.271919118730358e-05,
      "loss": 1.0077,
      "mean_token_accuracy": 0.7462121248245239,
      "num_tokens": 35090803.0,
      "step": 1076
    },
    {
      "entropy": 0.9554581344127655,
      "epoch": 0.5759358288770053,
      "grad_norm": 0.17973192036151886,
      "learning_rate": 9.25661657017093e-05,
      "loss": 0.9418,
      "mean_token_accuracy": 0.7558650970458984,
      "num_tokens": 35123571.0,
      "step": 1077
    },
    {
      "entropy": 0.9667383879423141,
      "epoch": 0.5764705882352941,
      "grad_norm": 0.18665874004364014,
      "learning_rate": 9.241319256373518e-05,
      "loss": 0.938,
      "mean_token_accuracy": 0.7555632442235947,
      "num_tokens": 35156255.0,
      "step": 1078
    },
    {
      "entropy": 1.0221573561429977,
      "epoch": 0.5770053475935829,
      "grad_norm": 0.1763426810503006,
      "learning_rate": 9.226027223270528e-05,
      "loss": 1.0139,
      "mean_token_accuracy": 0.7463648617267609,
      "num_tokens": 35189023.0,
      "step": 1079
    },
    {
      "entropy": 1.031988486647606,
      "epoch": 0.5775401069518716,
      "grad_norm": 0.16598355770111084,
      "learning_rate": 9.210740516778511e-05,
      "loss": 1.0285,
      "mean_token_accuracy": 0.7438599765300751,
      "num_tokens": 35221791.0,
      "step": 1080
    },
    {
      "entropy": 0.9958008080720901,
      "epoch": 0.5780748663101605,
      "grad_norm": 0.16901369392871857,
      "learning_rate": 9.195459182798021e-05,
      "loss": 1.0063,
      "mean_token_accuracy": 0.7480481863021851,
      "num_tokens": 35254386.0,
      "step": 1081
    },
    {
      "entropy": 0.9488922357559204,
      "epoch": 0.5786096256684492,
      "grad_norm": 0.17280098795890808,
      "learning_rate": 9.180183267213487e-05,
      "loss": 0.9483,
      "mean_token_accuracy": 0.7611047327518463,
      "num_tokens": 35286675.0,
      "step": 1082
    },
    {
      "entropy": 0.9606172740459442,
      "epoch": 0.5791443850267379,
      "grad_norm": 0.17144496738910675,
      "learning_rate": 9.16491281589307e-05,
      "loss": 0.9474,
      "mean_token_accuracy": 0.7588587403297424,
      "num_tokens": 35319443.0,
      "step": 1083
    },
    {
      "entropy": 0.9573392570018768,
      "epoch": 0.5796791443850268,
      "grad_norm": 0.18120399117469788,
      "learning_rate": 9.149647874688508e-05,
      "loss": 0.9548,
      "mean_token_accuracy": 0.7587325572967529,
      "num_tokens": 35352193.0,
      "step": 1084
    },
    {
      "entropy": 1.0083431005477905,
      "epoch": 0.5802139037433155,
      "grad_norm": 0.16116108000278473,
      "learning_rate": 9.134388489435008e-05,
      "loss": 1.0083,
      "mean_token_accuracy": 0.7453295439481735,
      "num_tokens": 35384500.0,
      "step": 1085
    },
    {
      "entropy": 0.9728332459926605,
      "epoch": 0.5807486631016042,
      "grad_norm": 0.18752528727054596,
      "learning_rate": 9.119134705951096e-05,
      "loss": 0.983,
      "mean_token_accuracy": 0.7519244849681854,
      "num_tokens": 35417268.0,
      "step": 1086
    },
    {
      "entropy": 0.9972249269485474,
      "epoch": 0.5812834224598931,
      "grad_norm": 0.1591930240392685,
      "learning_rate": 9.103886570038467e-05,
      "loss": 1.0113,
      "mean_token_accuracy": 0.7478616833686829,
      "num_tokens": 35450036.0,
      "step": 1087
    },
    {
      "entropy": 0.9868172854185104,
      "epoch": 0.5818181818181818,
      "grad_norm": 0.18106745183467865,
      "learning_rate": 9.088644127481863e-05,
      "loss": 0.9828,
      "mean_token_accuracy": 0.7541590183973312,
      "num_tokens": 35482512.0,
      "step": 1088
    },
    {
      "entropy": 1.004557117819786,
      "epoch": 0.5823529411764706,
      "grad_norm": 0.15707138180732727,
      "learning_rate": 9.073407424048937e-05,
      "loss": 1.0098,
      "mean_token_accuracy": 0.7464142739772797,
      "num_tokens": 35514811.0,
      "step": 1089
    },
    {
      "entropy": 0.9354948103427887,
      "epoch": 0.5828877005347594,
      "grad_norm": 0.16020675003528595,
      "learning_rate": 9.058176505490106e-05,
      "loss": 0.9314,
      "mean_token_accuracy": 0.7615575194358826,
      "num_tokens": 35547361.0,
      "step": 1090
    },
    {
      "entropy": 0.9814960062503815,
      "epoch": 0.5834224598930481,
      "grad_norm": 0.15799638628959656,
      "learning_rate": 9.04295141753841e-05,
      "loss": 0.9946,
      "mean_token_accuracy": 0.7523322403430939,
      "num_tokens": 35580081.0,
      "step": 1091
    },
    {
      "entropy": 1.014263927936554,
      "epoch": 0.5839572192513369,
      "grad_norm": 0.17104533314704895,
      "learning_rate": 9.027732205909389e-05,
      "loss": 1.0043,
      "mean_token_accuracy": 0.7447763979434967,
      "num_tokens": 35612849.0,
      "step": 1092
    },
    {
      "entropy": 0.9643675088882446,
      "epoch": 0.5844919786096257,
      "grad_norm": 0.15428514778614044,
      "learning_rate": 9.012518916300939e-05,
      "loss": 0.939,
      "mean_token_accuracy": 0.7590114772319794,
      "num_tokens": 35645617.0,
      "step": 1093
    },
    {
      "entropy": 0.9955264329910278,
      "epoch": 0.5850267379679145,
      "grad_norm": 0.16135382652282715,
      "learning_rate": 8.997311594393172e-05,
      "loss": 0.9938,
      "mean_token_accuracy": 0.7491723150014877,
      "num_tokens": 35678042.0,
      "step": 1094
    },
    {
      "entropy": 0.9898623675107956,
      "epoch": 0.5855614973262032,
      "grad_norm": 0.1541261523962021,
      "learning_rate": 8.982110285848278e-05,
      "loss": 0.9866,
      "mean_token_accuracy": 0.7481366097927094,
      "num_tokens": 35710810.0,
      "step": 1095
    },
    {
      "entropy": 0.954077735543251,
      "epoch": 0.586096256684492,
      "grad_norm": 0.1534184366464615,
      "learning_rate": 8.966915036310396e-05,
      "loss": 0.9543,
      "mean_token_accuracy": 0.7587708383798599,
      "num_tokens": 35743567.0,
      "step": 1096
    },
    {
      "entropy": 0.956783264875412,
      "epoch": 0.5866310160427808,
      "grad_norm": 0.16506008803844452,
      "learning_rate": 8.951725891405476e-05,
      "loss": 0.9501,
      "mean_token_accuracy": 0.755631148815155,
      "num_tokens": 35776045.0,
      "step": 1097
    },
    {
      "entropy": 1.0246059000492096,
      "epoch": 0.5871657754010695,
      "grad_norm": 0.1616167426109314,
      "learning_rate": 8.936542896741123e-05,
      "loss": 1.0158,
      "mean_token_accuracy": 0.7454789876937866,
      "num_tokens": 35808813.0,
      "step": 1098
    },
    {
      "entropy": 0.9628048241138458,
      "epoch": 0.5877005347593582,
      "grad_norm": 0.16270913183689117,
      "learning_rate": 8.921366097906484e-05,
      "loss": 0.9509,
      "mean_token_accuracy": 0.7559598535299301,
      "num_tokens": 35841233.0,
      "step": 1099
    },
    {
      "entropy": 0.9961771219968796,
      "epoch": 0.5882352941176471,
      "grad_norm": 0.16457895934581757,
      "learning_rate": 8.906195540472108e-05,
      "loss": 0.972,
      "mean_token_accuracy": 0.7511082738637924,
      "num_tokens": 35873894.0,
      "step": 1100
    },
    {
      "entropy": 0.9758996218442917,
      "epoch": 0.5887700534759358,
      "grad_norm": 0.17612816393375397,
      "learning_rate": 8.8910312699898e-05,
      "loss": 0.9898,
      "mean_token_accuracy": 0.7528423368930817,
      "num_tokens": 35906576.0,
      "step": 1101
    },
    {
      "entropy": 0.9655701667070389,
      "epoch": 0.5893048128342246,
      "grad_norm": 0.15732590854167938,
      "learning_rate": 8.875873331992477e-05,
      "loss": 0.9654,
      "mean_token_accuracy": 0.7554679811000824,
      "num_tokens": 35939344.0,
      "step": 1102
    },
    {
      "entropy": 0.9655955731868744,
      "epoch": 0.5898395721925134,
      "grad_norm": 0.19670598208904266,
      "learning_rate": 8.860721771994053e-05,
      "loss": 0.9602,
      "mean_token_accuracy": 0.7536127120256424,
      "num_tokens": 35971865.0,
      "step": 1103
    },
    {
      "entropy": 1.0124208182096481,
      "epoch": 0.5903743315508021,
      "grad_norm": 0.1736741065979004,
      "learning_rate": 8.845576635489292e-05,
      "loss": 1.0143,
      "mean_token_accuracy": 0.7426514476537704,
      "num_tokens": 36004347.0,
      "step": 1104
    },
    {
      "entropy": 0.9384248703718185,
      "epoch": 0.5909090909090909,
      "grad_norm": 0.17268306016921997,
      "learning_rate": 8.830437967953664e-05,
      "loss": 0.945,
      "mean_token_accuracy": 0.7619134783744812,
      "num_tokens": 36037115.0,
      "step": 1105
    },
    {
      "entropy": 0.9208074063062668,
      "epoch": 0.5914438502673797,
      "grad_norm": 0.20721158385276794,
      "learning_rate": 8.815305814843214e-05,
      "loss": 0.9171,
      "mean_token_accuracy": 0.7619870901107788,
      "num_tokens": 36069843.0,
      "step": 1106
    },
    {
      "entropy": 0.9940818250179291,
      "epoch": 0.5919786096256684,
      "grad_norm": 0.18215195834636688,
      "learning_rate": 8.80018022159443e-05,
      "loss": 1.008,
      "mean_token_accuracy": 0.7472860366106033,
      "num_tokens": 36102079.0,
      "step": 1107
    },
    {
      "entropy": 0.9343003779649734,
      "epoch": 0.5925133689839572,
      "grad_norm": 0.20239266753196716,
      "learning_rate": 8.785061233624113e-05,
      "loss": 0.9514,
      "mean_token_accuracy": 0.7605693936347961,
      "num_tokens": 36134847.0,
      "step": 1108
    },
    {
      "entropy": 0.9898380935192108,
      "epoch": 0.593048128342246,
      "grad_norm": 0.15859541296958923,
      "learning_rate": 8.76994889632921e-05,
      "loss": 0.9993,
      "mean_token_accuracy": 0.7506109476089478,
      "num_tokens": 36167615.0,
      "step": 1109
    },
    {
      "entropy": 1.002129703760147,
      "epoch": 0.5935828877005348,
      "grad_norm": 0.1886453926563263,
      "learning_rate": 8.754843255086711e-05,
      "loss": 1.0126,
      "mean_token_accuracy": 0.7439516186714172,
      "num_tokens": 36200383.0,
      "step": 1110
    },
    {
      "entropy": 0.9945786744356155,
      "epoch": 0.5941176470588235,
      "grad_norm": 0.15921494364738464,
      "learning_rate": 8.739744355253501e-05,
      "loss": 0.9889,
      "mean_token_accuracy": 0.7528126686811447,
      "num_tokens": 36232975.0,
      "step": 1111
    },
    {
      "entropy": 0.9854760468006134,
      "epoch": 0.5946524064171123,
      "grad_norm": 0.16590659320354462,
      "learning_rate": 8.724652242166223e-05,
      "loss": 0.9882,
      "mean_token_accuracy": 0.7484716773033142,
      "num_tokens": 36265471.0,
      "step": 1112
    },
    {
      "entropy": 0.9827838242053986,
      "epoch": 0.5951871657754011,
      "grad_norm": 0.1649196296930313,
      "learning_rate": 8.709566961141129e-05,
      "loss": 0.9743,
      "mean_token_accuracy": 0.751588299870491,
      "num_tokens": 36298105.0,
      "step": 1113
    },
    {
      "entropy": 0.9589742720127106,
      "epoch": 0.5957219251336898,
      "grad_norm": 0.15379104018211365,
      "learning_rate": 8.694488557473976e-05,
      "loss": 0.9509,
      "mean_token_accuracy": 0.7559805363416672,
      "num_tokens": 36330617.0,
      "step": 1114
    },
    {
      "entropy": 0.9678229987621307,
      "epoch": 0.5962566844919787,
      "grad_norm": 0.1546468585729599,
      "learning_rate": 8.679417076439862e-05,
      "loss": 0.9535,
      "mean_token_accuracy": 0.7550341635942459,
      "num_tokens": 36362990.0,
      "step": 1115
    },
    {
      "entropy": 1.0166302472352982,
      "epoch": 0.5967914438502674,
      "grad_norm": 0.16416624188423157,
      "learning_rate": 8.66435256329309e-05,
      "loss": 1.0267,
      "mean_token_accuracy": 0.7411920875310898,
      "num_tokens": 36395335.0,
      "step": 1116
    },
    {
      "entropy": 0.9371855705976486,
      "epoch": 0.5973262032085561,
      "grad_norm": 0.15952086448669434,
      "learning_rate": 8.649295063267056e-05,
      "loss": 0.9313,
      "mean_token_accuracy": 0.7642102837562561,
      "num_tokens": 36427288.0,
      "step": 1117
    },
    {
      "entropy": 0.994782492518425,
      "epoch": 0.597860962566845,
      "grad_norm": 0.15828780829906464,
      "learning_rate": 8.634244621574088e-05,
      "loss": 0.9903,
      "mean_token_accuracy": 0.7489925920963287,
      "num_tokens": 36459884.0,
      "step": 1118
    },
    {
      "entropy": 0.9942251890897751,
      "epoch": 0.5983957219251337,
      "grad_norm": 0.16231967508792877,
      "learning_rate": 8.619201283405327e-05,
      "loss": 0.987,
      "mean_token_accuracy": 0.7529947757720947,
      "num_tokens": 36492227.0,
      "step": 1119
    },
    {
      "entropy": 0.979880079627037,
      "epoch": 0.5989304812834224,
      "grad_norm": 0.15880082547664642,
      "learning_rate": 8.604165093930578e-05,
      "loss": 0.9734,
      "mean_token_accuracy": 0.7522595971822739,
      "num_tokens": 36524228.0,
      "step": 1120
    },
    {
      "entropy": 0.9602563828229904,
      "epoch": 0.5994652406417113,
      "grad_norm": 0.1610957682132721,
      "learning_rate": 8.589136098298185e-05,
      "loss": 0.9491,
      "mean_token_accuracy": 0.7566593289375305,
      "num_tokens": 36556996.0,
      "step": 1121
    },
    {
      "entropy": 0.9304652363061905,
      "epoch": 0.6,
      "grad_norm": 0.16996730864048004,
      "learning_rate": 8.574114341634891e-05,
      "loss": 0.9285,
      "mean_token_accuracy": 0.7644611895084381,
      "num_tokens": 36589742.0,
      "step": 1122
    },
    {
      "entropy": 0.9693291038274765,
      "epoch": 0.6005347593582888,
      "grad_norm": 0.15727782249450684,
      "learning_rate": 8.559099869045706e-05,
      "loss": 0.9728,
      "mean_token_accuracy": 0.7531582415103912,
      "num_tokens": 36622299.0,
      "step": 1123
    },
    {
      "entropy": 0.9516191184520721,
      "epoch": 0.6010695187165775,
      "grad_norm": 0.1869746893644333,
      "learning_rate": 8.544092725613759e-05,
      "loss": 0.9525,
      "mean_token_accuracy": 0.7592806667089462,
      "num_tokens": 36654933.0,
      "step": 1124
    },
    {
      "entropy": 0.9548157751560211,
      "epoch": 0.6016042780748663,
      "grad_norm": 0.1933574229478836,
      "learning_rate": 8.52909295640018e-05,
      "loss": 0.9587,
      "mean_token_accuracy": 0.75409334897995,
      "num_tokens": 36687701.0,
      "step": 1125
    },
    {
      "entropy": 0.9195386320352554,
      "epoch": 0.6021390374331551,
      "grad_norm": 0.16644708812236786,
      "learning_rate": 8.51410060644396e-05,
      "loss": 0.9236,
      "mean_token_accuracy": 0.7597140669822693,
      "num_tokens": 36720469.0,
      "step": 1126
    },
    {
      "entropy": 0.9431671649217606,
      "epoch": 0.6026737967914438,
      "grad_norm": 0.1637318879365921,
      "learning_rate": 8.499115720761802e-05,
      "loss": 0.9534,
      "mean_token_accuracy": 0.758542999625206,
      "num_tokens": 36752930.0,
      "step": 1127
    },
    {
      "entropy": 0.9756807684898376,
      "epoch": 0.6032085561497326,
      "grad_norm": 0.15840820968151093,
      "learning_rate": 8.484138344348002e-05,
      "loss": 0.9918,
      "mean_token_accuracy": 0.7510746717453003,
      "num_tokens": 36785566.0,
      "step": 1128
    },
    {
      "entropy": 0.9955893605947495,
      "epoch": 0.6037433155080214,
      "grad_norm": 0.16000495851039886,
      "learning_rate": 8.469168522174315e-05,
      "loss": 0.9939,
      "mean_token_accuracy": 0.7508969306945801,
      "num_tokens": 36817123.0,
      "step": 1129
    },
    {
      "entropy": 0.9839041531085968,
      "epoch": 0.6042780748663101,
      "grad_norm": 0.16458529233932495,
      "learning_rate": 8.454206299189804e-05,
      "loss": 0.9871,
      "mean_token_accuracy": 0.7512162774801254,
      "num_tokens": 36849688.0,
      "step": 1130
    },
    {
      "entropy": 0.9587539881467819,
      "epoch": 0.604812834224599,
      "grad_norm": 0.1586272120475769,
      "learning_rate": 8.439251720320709e-05,
      "loss": 0.9582,
      "mean_token_accuracy": 0.7575050890445709,
      "num_tokens": 36882114.0,
      "step": 1131
    },
    {
      "entropy": 0.9999642670154572,
      "epoch": 0.6053475935828877,
      "grad_norm": 0.18612682819366455,
      "learning_rate": 8.424304830470335e-05,
      "loss": 0.9911,
      "mean_token_accuracy": 0.7502443790435791,
      "num_tokens": 36914882.0,
      "step": 1132
    },
    {
      "entropy": 0.9600541740655899,
      "epoch": 0.6058823529411764,
      "grad_norm": 0.16042885184288025,
      "learning_rate": 8.409365674518888e-05,
      "loss": 0.9422,
      "mean_token_accuracy": 0.7583836913108826,
      "num_tokens": 36947505.0,
      "step": 1133
    },
    {
      "entropy": 1.0059276521205902,
      "epoch": 0.6064171122994653,
      "grad_norm": 0.17258723080158234,
      "learning_rate": 8.394434297323355e-05,
      "loss": 1.0066,
      "mean_token_accuracy": 0.7461808770895004,
      "num_tokens": 36979947.0,
      "step": 1134
    },
    {
      "entropy": 0.9567270874977112,
      "epoch": 0.606951871657754,
      "grad_norm": 0.17004330456256866,
      "learning_rate": 8.37951074371736e-05,
      "loss": 0.9575,
      "mean_token_accuracy": 0.7579072117805481,
      "num_tokens": 37012622.0,
      "step": 1135
    },
    {
      "entropy": 1.0197716355323792,
      "epoch": 0.6074866310160428,
      "grad_norm": 0.16168364882469177,
      "learning_rate": 8.36459505851105e-05,
      "loss": 1.0093,
      "mean_token_accuracy": 0.7469758093357086,
      "num_tokens": 37045390.0,
      "step": 1136
    },
    {
      "entropy": 1.0205204039812088,
      "epoch": 0.6080213903743316,
      "grad_norm": 0.16087409853935242,
      "learning_rate": 8.349687286490926e-05,
      "loss": 1.013,
      "mean_token_accuracy": 0.7450649440288544,
      "num_tokens": 37078051.0,
      "step": 1137
    },
    {
      "entropy": 0.9923149198293686,
      "epoch": 0.6085561497326203,
      "grad_norm": 0.15231212973594666,
      "learning_rate": 8.334787472419745e-05,
      "loss": 0.9837,
      "mean_token_accuracy": 0.7486864626407623,
      "num_tokens": 37110819.0,
      "step": 1138
    },
    {
      "entropy": 0.9868474304676056,
      "epoch": 0.6090909090909091,
      "grad_norm": 0.16185684502124786,
      "learning_rate": 8.319895661036361e-05,
      "loss": 0.994,
      "mean_token_accuracy": 0.7506148368120193,
      "num_tokens": 37143572.0,
      "step": 1139
    },
    {
      "entropy": 0.9274433553218842,
      "epoch": 0.6096256684491979,
      "grad_norm": 0.16928084194660187,
      "learning_rate": 8.305011897055606e-05,
      "loss": 0.9355,
      "mean_token_accuracy": 0.759330689907074,
      "num_tokens": 37175360.0,
      "step": 1140
    },
    {
      "entropy": 0.9518332779407501,
      "epoch": 0.6101604278074866,
      "grad_norm": 0.16280804574489594,
      "learning_rate": 8.29013622516814e-05,
      "loss": 0.9511,
      "mean_token_accuracy": 0.75792196393013,
      "num_tokens": 37207756.0,
      "step": 1141
    },
    {
      "entropy": 0.9957929700613022,
      "epoch": 0.6106951871657754,
      "grad_norm": 0.16850176453590393,
      "learning_rate": 8.275268690040325e-05,
      "loss": 0.99,
      "mean_token_accuracy": 0.7500610947608948,
      "num_tokens": 37240524.0,
      "step": 1142
    },
    {
      "entropy": 0.9780972301959991,
      "epoch": 0.6112299465240641,
      "grad_norm": 0.1570751816034317,
      "learning_rate": 8.260409336314101e-05,
      "loss": 0.9725,
      "mean_token_accuracy": 0.7514051795005798,
      "num_tokens": 37273292.0,
      "step": 1143
    },
    {
      "entropy": 0.9441059827804565,
      "epoch": 0.611764705882353,
      "grad_norm": 0.16775594651699066,
      "learning_rate": 8.245558208606836e-05,
      "loss": 0.9442,
      "mean_token_accuracy": 0.7611830681562424,
      "num_tokens": 37305117.0,
      "step": 1144
    },
    {
      "entropy": 0.9759251922369003,
      "epoch": 0.6122994652406417,
      "grad_norm": 0.16318322718143463,
      "learning_rate": 8.230715351511193e-05,
      "loss": 0.9728,
      "mean_token_accuracy": 0.7533602118492126,
      "num_tokens": 37337885.0,
      "step": 1145
    },
    {
      "entropy": 1.004443645477295,
      "epoch": 0.6128342245989304,
      "grad_norm": 0.16552051901817322,
      "learning_rate": 8.215880809595014e-05,
      "loss": 1.0108,
      "mean_token_accuracy": 0.7459371984004974,
      "num_tokens": 37370653.0,
      "step": 1146
    },
    {
      "entropy": 0.9759458899497986,
      "epoch": 0.6133689839572193,
      "grad_norm": 0.1621161550283432,
      "learning_rate": 8.201054627401159e-05,
      "loss": 0.9805,
      "mean_token_accuracy": 0.7463954091072083,
      "num_tokens": 37403421.0,
      "step": 1147
    },
    {
      "entropy": 0.9828044027090073,
      "epoch": 0.613903743315508,
      "grad_norm": 0.1711197793483734,
      "learning_rate": 8.186236849447401e-05,
      "loss": 0.9985,
      "mean_token_accuracy": 0.749053031206131,
      "num_tokens": 37436189.0,
      "step": 1148
    },
    {
      "entropy": 0.9460033178329468,
      "epoch": 0.6144385026737967,
      "grad_norm": 0.1649860441684723,
      "learning_rate": 8.171427520226267e-05,
      "loss": 0.9542,
      "mean_token_accuracy": 0.7548500895500183,
      "num_tokens": 37468471.0,
      "step": 1149
    },
    {
      "entropy": 0.9906985908746719,
      "epoch": 0.6149732620320856,
      "grad_norm": 0.16224952042102814,
      "learning_rate": 8.156626684204916e-05,
      "loss": 1.0017,
      "mean_token_accuracy": 0.7461031377315521,
      "num_tokens": 37501131.0,
      "step": 1150
    },
    {
      "entropy": 0.9174692332744598,
      "epoch": 0.6155080213903743,
      "grad_norm": 0.17507107555866241,
      "learning_rate": 8.141834385825014e-05,
      "loss": 0.921,
      "mean_token_accuracy": 0.765705406665802,
      "num_tokens": 37533434.0,
      "step": 1151
    },
    {
      "entropy": 0.9863606989383698,
      "epoch": 0.6160427807486631,
      "grad_norm": 0.1530178338289261,
      "learning_rate": 8.127050669502588e-05,
      "loss": 0.9774,
      "mean_token_accuracy": 0.7490722090005875,
      "num_tokens": 37566153.0,
      "step": 1152
    },
    {
      "entropy": 0.9695703238248825,
      "epoch": 0.6165775401069519,
      "grad_norm": 0.1572701632976532,
      "learning_rate": 8.112275579627883e-05,
      "loss": 0.9645,
      "mean_token_accuracy": 0.7534518539905548,
      "num_tokens": 37598921.0,
      "step": 1153
    },
    {
      "entropy": 0.9552498608827591,
      "epoch": 0.6171122994652406,
      "grad_norm": 0.1595485359430313,
      "learning_rate": 8.097509160565265e-05,
      "loss": 0.9425,
      "mean_token_accuracy": 0.7607965171337128,
      "num_tokens": 37631010.0,
      "step": 1154
    },
    {
      "entropy": 0.9826492518186569,
      "epoch": 0.6176470588235294,
      "grad_norm": 0.16088756918907166,
      "learning_rate": 8.082751456653046e-05,
      "loss": 0.9722,
      "mean_token_accuracy": 0.7502389848232269,
      "num_tokens": 37663680.0,
      "step": 1155
    },
    {
      "entropy": 0.9433806240558624,
      "epoch": 0.6181818181818182,
      "grad_norm": 0.1699216216802597,
      "learning_rate": 8.068002512203375e-05,
      "loss": 0.9358,
      "mean_token_accuracy": 0.7620356678962708,
      "num_tokens": 37696448.0,
      "step": 1156
    },
    {
      "entropy": 0.9715216606855392,
      "epoch": 0.618716577540107,
      "grad_norm": 0.18060708045959473,
      "learning_rate": 8.053262371502102e-05,
      "loss": 0.9635,
      "mean_token_accuracy": 0.7556838393211365,
      "num_tokens": 37729169.0,
      "step": 1157
    },
    {
      "entropy": 1.0064241141080856,
      "epoch": 0.6192513368983957,
      "grad_norm": 0.16485314071178436,
      "learning_rate": 8.03853107880864e-05,
      "loss": 1.0032,
      "mean_token_accuracy": 0.749094694852829,
      "num_tokens": 37761445.0,
      "step": 1158
    },
    {
      "entropy": 0.9281845837831497,
      "epoch": 0.6197860962566845,
      "grad_norm": 0.16989584267139435,
      "learning_rate": 8.023808678355842e-05,
      "loss": 0.9421,
      "mean_token_accuracy": 0.7602249681949615,
      "num_tokens": 37793148.0,
      "step": 1159
    },
    {
      "entropy": 0.9493147134780884,
      "epoch": 0.6203208556149733,
      "grad_norm": 0.15808895230293274,
      "learning_rate": 8.009095214349842e-05,
      "loss": 0.951,
      "mean_token_accuracy": 0.7550097703933716,
      "num_tokens": 37825916.0,
      "step": 1160
    },
    {
      "entropy": 0.9858775585889816,
      "epoch": 0.620855614973262,
      "grad_norm": 0.1600574553012848,
      "learning_rate": 7.994390730969962e-05,
      "loss": 0.993,
      "mean_token_accuracy": 0.747996062040329,
      "num_tokens": 37858567.0,
      "step": 1161
    },
    {
      "entropy": 0.9468472003936768,
      "epoch": 0.6213903743315508,
      "grad_norm": 0.18879085779190063,
      "learning_rate": 7.979695272368549e-05,
      "loss": 0.9594,
      "mean_token_accuracy": 0.7577645629644394,
      "num_tokens": 37890859.0,
      "step": 1162
    },
    {
      "entropy": 0.9671633839607239,
      "epoch": 0.6219251336898396,
      "grad_norm": 0.1572384238243103,
      "learning_rate": 7.96500888267085e-05,
      "loss": 0.9652,
      "mean_token_accuracy": 0.7520887106657028,
      "num_tokens": 37923190.0,
      "step": 1163
    },
    {
      "entropy": 0.9662671089172363,
      "epoch": 0.6224598930481283,
      "grad_norm": 0.17573218047618866,
      "learning_rate": 7.950331605974884e-05,
      "loss": 0.9652,
      "mean_token_accuracy": 0.7542543113231659,
      "num_tokens": 37955830.0,
      "step": 1164
    },
    {
      "entropy": 0.9456570893526077,
      "epoch": 0.6229946524064172,
      "grad_norm": 0.15757232904434204,
      "learning_rate": 7.935663486351306e-05,
      "loss": 0.9404,
      "mean_token_accuracy": 0.7586449086666107,
      "num_tokens": 37988598.0,
      "step": 1165
    },
    {
      "entropy": 0.9794042408466339,
      "epoch": 0.6235294117647059,
      "grad_norm": 0.18899321556091309,
      "learning_rate": 7.921004567843278e-05,
      "loss": 0.969,
      "mean_token_accuracy": 0.7550794333219528,
      "num_tokens": 38020828.0,
      "step": 1166
    },
    {
      "entropy": 0.9964238554239273,
      "epoch": 0.6240641711229946,
      "grad_norm": 0.16320250928401947,
      "learning_rate": 7.906354894466328e-05,
      "loss": 0.9957,
      "mean_token_accuracy": 0.7484726309776306,
      "num_tokens": 38053596.0,
      "step": 1167
    },
    {
      "entropy": 0.9406930059194565,
      "epoch": 0.6245989304812835,
      "grad_norm": 0.16028763353824615,
      "learning_rate": 7.891714510208232e-05,
      "loss": 0.9464,
      "mean_token_accuracy": 0.7546432018280029,
      "num_tokens": 38086364.0,
      "step": 1168
    },
    {
      "entropy": 0.9705571532249451,
      "epoch": 0.6251336898395722,
      "grad_norm": 0.1809701919555664,
      "learning_rate": 7.877083459028872e-05,
      "loss": 0.9651,
      "mean_token_accuracy": 0.7501221895217896,
      "num_tokens": 38119132.0,
      "step": 1169
    },
    {
      "entropy": 0.9514707177877426,
      "epoch": 0.6256684491978609,
      "grad_norm": 0.17112714052200317,
      "learning_rate": 7.862461784860103e-05,
      "loss": 0.9769,
      "mean_token_accuracy": 0.7516950070858002,
      "num_tokens": 38151834.0,
      "step": 1170
    },
    {
      "entropy": 0.9061668813228607,
      "epoch": 0.6262032085561497,
      "grad_norm": 0.16375577449798584,
      "learning_rate": 7.847849531605624e-05,
      "loss": 0.9032,
      "mean_token_accuracy": 0.7696725130081177,
      "num_tokens": 38184602.0,
      "step": 1171
    },
    {
      "entropy": 0.9816283136606216,
      "epoch": 0.6267379679144385,
      "grad_norm": 0.2077077478170395,
      "learning_rate": 7.833246743140852e-05,
      "loss": 0.9645,
      "mean_token_accuracy": 0.7513135373592377,
      "num_tokens": 38217370.0,
      "step": 1172
    },
    {
      "entropy": 0.9717194885015488,
      "epoch": 0.6272727272727273,
      "grad_norm": 0.15337511897087097,
      "learning_rate": 7.81865346331278e-05,
      "loss": 0.9532,
      "mean_token_accuracy": 0.754337728023529,
      "num_tokens": 38250138.0,
      "step": 1173
    },
    {
      "entropy": 0.9285975843667984,
      "epoch": 0.627807486631016,
      "grad_norm": 0.17144861817359924,
      "learning_rate": 7.804069735939844e-05,
      "loss": 0.9331,
      "mean_token_accuracy": 0.7599584609270096,
      "num_tokens": 38282906.0,
      "step": 1174
    },
    {
      "entropy": 0.9945204854011536,
      "epoch": 0.6283422459893048,
      "grad_norm": 0.15670542418956757,
      "learning_rate": 7.78949560481181e-05,
      "loss": 0.9866,
      "mean_token_accuracy": 0.7500029057264328,
      "num_tokens": 38315357.0,
      "step": 1175
    },
    {
      "entropy": 0.9691455066204071,
      "epoch": 0.6288770053475936,
      "grad_norm": 0.17279230058193207,
      "learning_rate": 7.774931113689625e-05,
      "loss": 0.9767,
      "mean_token_accuracy": 0.7498340159654617,
      "num_tokens": 38347759.0,
      "step": 1176
    },
    {
      "entropy": 0.9547702074050903,
      "epoch": 0.6294117647058823,
      "grad_norm": 0.21260668337345123,
      "learning_rate": 7.76037630630529e-05,
      "loss": 0.9632,
      "mean_token_accuracy": 0.7575146555900574,
      "num_tokens": 38380527.0,
      "step": 1177
    },
    {
      "entropy": 0.948182076215744,
      "epoch": 0.6299465240641712,
      "grad_norm": 0.15867938101291656,
      "learning_rate": 7.745831226361724e-05,
      "loss": 0.9326,
      "mean_token_accuracy": 0.7615774571895599,
      "num_tokens": 38413295.0,
      "step": 1178
    },
    {
      "entropy": 0.9786037802696228,
      "epoch": 0.6304812834224599,
      "grad_norm": 0.18239183723926544,
      "learning_rate": 7.731295917532647e-05,
      "loss": 0.9667,
      "mean_token_accuracy": 0.7576979398727417,
      "num_tokens": 38446063.0,
      "step": 1179
    },
    {
      "entropy": 0.928612694144249,
      "epoch": 0.6310160427807486,
      "grad_norm": 0.17347387969493866,
      "learning_rate": 7.716770423462427e-05,
      "loss": 0.9233,
      "mean_token_accuracy": 0.7647849321365356,
      "num_tokens": 38478831.0,
      "step": 1180
    },
    {
      "entropy": 0.9102021008729935,
      "epoch": 0.6315508021390375,
      "grad_norm": 0.18702252209186554,
      "learning_rate": 7.702254787765983e-05,
      "loss": 0.9119,
      "mean_token_accuracy": 0.7655161917209625,
      "num_tokens": 38511575.0,
      "step": 1181
    },
    {
      "entropy": 0.9744768887758255,
      "epoch": 0.6320855614973262,
      "grad_norm": 0.17710831761360168,
      "learning_rate": 7.68774905402861e-05,
      "loss": 0.9636,
      "mean_token_accuracy": 0.7528095990419388,
      "num_tokens": 38544339.0,
      "step": 1182
    },
    {
      "entropy": 0.9430717974901199,
      "epoch": 0.6326203208556149,
      "grad_norm": 0.16024550795555115,
      "learning_rate": 7.673253265805887e-05,
      "loss": 0.9537,
      "mean_token_accuracy": 0.7593987137079239,
      "num_tokens": 38576778.0,
      "step": 1183
    },
    {
      "entropy": 0.9504353106021881,
      "epoch": 0.6331550802139038,
      "grad_norm": 0.18097208440303802,
      "learning_rate": 7.65876746662352e-05,
      "loss": 0.9438,
      "mean_token_accuracy": 0.758845865726471,
      "num_tokens": 38609380.0,
      "step": 1184
    },
    {
      "entropy": 0.9205587655305862,
      "epoch": 0.6336898395721925,
      "grad_norm": 0.16049431264400482,
      "learning_rate": 7.644291699977224e-05,
      "loss": 0.9256,
      "mean_token_accuracy": 0.7635595947504044,
      "num_tokens": 38642042.0,
      "step": 1185
    },
    {
      "entropy": 0.9848443865776062,
      "epoch": 0.6342245989304813,
      "grad_norm": 0.16670531034469604,
      "learning_rate": 7.629826009332597e-05,
      "loss": 0.9947,
      "mean_token_accuracy": 0.7503360211849213,
      "num_tokens": 38674810.0,
      "step": 1186
    },
    {
      "entropy": 0.9391316771507263,
      "epoch": 0.6347593582887701,
      "grad_norm": 0.17103585600852966,
      "learning_rate": 7.615370438124969e-05,
      "loss": 0.9416,
      "mean_token_accuracy": 0.7613780200481415,
      "num_tokens": 38707470.0,
      "step": 1187
    },
    {
      "entropy": 0.9331240952014923,
      "epoch": 0.6352941176470588,
      "grad_norm": 0.16136282682418823,
      "learning_rate": 7.600925029759303e-05,
      "loss": 0.9263,
      "mean_token_accuracy": 0.7609174847602844,
      "num_tokens": 38739709.0,
      "step": 1188
    },
    {
      "entropy": 0.9672888666391373,
      "epoch": 0.6358288770053476,
      "grad_norm": 0.16377750039100647,
      "learning_rate": 7.586489827610024e-05,
      "loss": 0.9877,
      "mean_token_accuracy": 0.7512218952178955,
      "num_tokens": 38772477.0,
      "step": 1189
    },
    {
      "entropy": 0.9343033581972122,
      "epoch": 0.6363636363636364,
      "grad_norm": 0.16749276220798492,
      "learning_rate": 7.572064875020934e-05,
      "loss": 0.9386,
      "mean_token_accuracy": 0.7590665519237518,
      "num_tokens": 38804492.0,
      "step": 1190
    },
    {
      "entropy": 0.956359788775444,
      "epoch": 0.6368983957219251,
      "grad_norm": 0.162041574716568,
      "learning_rate": 7.55765021530504e-05,
      "loss": 0.9656,
      "mean_token_accuracy": 0.7557953298091888,
      "num_tokens": 38837123.0,
      "step": 1191
    },
    {
      "entropy": 0.9705105721950531,
      "epoch": 0.6374331550802139,
      "grad_norm": 0.1744755655527115,
      "learning_rate": 7.543245891744463e-05,
      "loss": 0.9673,
      "mean_token_accuracy": 0.7517472356557846,
      "num_tokens": 38869764.0,
      "step": 1192
    },
    {
      "entropy": 0.9938700348138809,
      "epoch": 0.6379679144385026,
      "grad_norm": 0.15463124215602875,
      "learning_rate": 7.528851947590267e-05,
      "loss": 0.9856,
      "mean_token_accuracy": 0.7493914514780045,
      "num_tokens": 38902435.0,
      "step": 1193
    },
    {
      "entropy": 0.9388508945703506,
      "epoch": 0.6385026737967915,
      "grad_norm": 0.16529619693756104,
      "learning_rate": 7.514468426062362e-05,
      "loss": 0.9638,
      "mean_token_accuracy": 0.7549322843551636,
      "num_tokens": 38935036.0,
      "step": 1194
    },
    {
      "entropy": 0.9566285014152527,
      "epoch": 0.6390374331550802,
      "grad_norm": 0.15760692954063416,
      "learning_rate": 7.500095370349367e-05,
      "loss": 0.9542,
      "mean_token_accuracy": 0.7584005296230316,
      "num_tokens": 38967804.0,
      "step": 1195
    },
    {
      "entropy": 0.9731516391038895,
      "epoch": 0.6395721925133689,
      "grad_norm": 0.1565542072057724,
      "learning_rate": 7.48573282360846e-05,
      "loss": 0.9766,
      "mean_token_accuracy": 0.7558712065219879,
      "num_tokens": 38999949.0,
      "step": 1196
    },
    {
      "entropy": 0.9728382974863052,
      "epoch": 0.6401069518716578,
      "grad_norm": 0.15303181111812592,
      "learning_rate": 7.471380828965282e-05,
      "loss": 0.9616,
      "mean_token_accuracy": 0.7509942203760147,
      "num_tokens": 39032654.0,
      "step": 1197
    },
    {
      "entropy": 0.9958923012018204,
      "epoch": 0.6406417112299465,
      "grad_norm": 0.17000418901443481,
      "learning_rate": 7.457039429513779e-05,
      "loss": 0.9966,
      "mean_token_accuracy": 0.747314989566803,
      "num_tokens": 39065379.0,
      "step": 1198
    },
    {
      "entropy": 0.9914234131574631,
      "epoch": 0.6411764705882353,
      "grad_norm": 0.16770699620246887,
      "learning_rate": 7.442708668316088e-05,
      "loss": 0.9839,
      "mean_token_accuracy": 0.7523640841245651,
      "num_tokens": 39097663.0,
      "step": 1199
    },
    {
      "entropy": 1.0625836253166199,
      "epoch": 0.6417112299465241,
      "grad_norm": 0.16445571184158325,
      "learning_rate": 7.428388588402397e-05,
      "loss": 1.052,
      "mean_token_accuracy": 0.7347568422555923,
      "num_tokens": 39129868.0,
      "step": 1200
    },
    {
      "entropy": 0.9884249269962311,
      "epoch": 0.6422459893048128,
      "grad_norm": 0.16170428693294525,
      "learning_rate": 7.414079232770832e-05,
      "loss": 0.9735,
      "mean_token_accuracy": 0.7510308921337128,
      "num_tokens": 39162555.0,
      "step": 1201
    },
    {
      "entropy": 0.9557610899209976,
      "epoch": 0.6427807486631016,
      "grad_norm": 0.1583389937877655,
      "learning_rate": 7.399780644387308e-05,
      "loss": 0.9521,
      "mean_token_accuracy": 0.7558740526437759,
      "num_tokens": 39195316.0,
      "step": 1202
    },
    {
      "entropy": 1.0001453161239624,
      "epoch": 0.6433155080213904,
      "grad_norm": 0.16569454967975616,
      "learning_rate": 7.385492866185418e-05,
      "loss": 0.9889,
      "mean_token_accuracy": 0.7535740435123444,
      "num_tokens": 39228084.0,
      "step": 1203
    },
    {
      "entropy": 0.9466915130615234,
      "epoch": 0.6438502673796791,
      "grad_norm": 0.17218825221061707,
      "learning_rate": 7.371215941066285e-05,
      "loss": 0.9328,
      "mean_token_accuracy": 0.7594085931777954,
      "num_tokens": 39260852.0,
      "step": 1204
    },
    {
      "entropy": 0.9893372505903244,
      "epoch": 0.6443850267379679,
      "grad_norm": 0.1548699587583542,
      "learning_rate": 7.356949911898456e-05,
      "loss": 0.9718,
      "mean_token_accuracy": 0.7504263818264008,
      "num_tokens": 39293542.0,
      "step": 1205
    },
    {
      "entropy": 0.951935887336731,
      "epoch": 0.6449197860962567,
      "grad_norm": 0.1720178872346878,
      "learning_rate": 7.342694821517757e-05,
      "loss": 0.9394,
      "mean_token_accuracy": 0.7647266685962677,
      "num_tokens": 39326010.0,
      "step": 1206
    },
    {
      "entropy": 0.9583040624856949,
      "epoch": 0.6454545454545455,
      "grad_norm": 0.1612015664577484,
      "learning_rate": 7.328450712727161e-05,
      "loss": 0.9564,
      "mean_token_accuracy": 0.7522828131914139,
      "num_tokens": 39358252.0,
      "step": 1207
    },
    {
      "entropy": 0.9418720752000809,
      "epoch": 0.6459893048128342,
      "grad_norm": 0.1865474283695221,
      "learning_rate": 7.314217628296682e-05,
      "loss": 0.9446,
      "mean_token_accuracy": 0.7590716779232025,
      "num_tokens": 39390549.0,
      "step": 1208
    },
    {
      "entropy": 0.9319218695163727,
      "epoch": 0.646524064171123,
      "grad_norm": 0.17535893619060516,
      "learning_rate": 7.299995610963215e-05,
      "loss": 0.9425,
      "mean_token_accuracy": 0.7570180594921112,
      "num_tokens": 39423027.0,
      "step": 1209
    },
    {
      "entropy": 0.9778590053319931,
      "epoch": 0.6470588235294118,
      "grad_norm": 0.1859198808670044,
      "learning_rate": 7.285784703430446e-05,
      "loss": 0.9955,
      "mean_token_accuracy": 0.7506783753633499,
      "num_tokens": 39454976.0,
      "step": 1210
    },
    {
      "entropy": 0.9831230938434601,
      "epoch": 0.6475935828877005,
      "grad_norm": 0.16050094366073608,
      "learning_rate": 7.271584948368677e-05,
      "loss": 1.002,
      "mean_token_accuracy": 0.7491774410009384,
      "num_tokens": 39487666.0,
      "step": 1211
    },
    {
      "entropy": 0.9467550218105316,
      "epoch": 0.6481283422459893,
      "grad_norm": 0.16806411743164062,
      "learning_rate": 7.257396388414737e-05,
      "loss": 0.9641,
      "mean_token_accuracy": 0.760325014591217,
      "num_tokens": 39520434.0,
      "step": 1212
    },
    {
      "entropy": 0.960830956697464,
      "epoch": 0.6486631016042781,
      "grad_norm": 0.16417725384235382,
      "learning_rate": 7.243219066171847e-05,
      "loss": 0.9686,
      "mean_token_accuracy": 0.7505804002285004,
      "num_tokens": 39553202.0,
      "step": 1213
    },
    {
      "entropy": 0.9331855922937393,
      "epoch": 0.6491978609625668,
      "grad_norm": 0.19479550421237946,
      "learning_rate": 7.229053024209467e-05,
      "loss": 0.9411,
      "mean_token_accuracy": 0.7603546977043152,
      "num_tokens": 39585826.0,
      "step": 1214
    },
    {
      "entropy": 0.9689950495958328,
      "epoch": 0.6497326203208557,
      "grad_norm": 0.16452471911907196,
      "learning_rate": 7.2148983050632e-05,
      "loss": 0.9651,
      "mean_token_accuracy": 0.7540017068386078,
      "num_tokens": 39618594.0,
      "step": 1215
    },
    {
      "entropy": 0.9832485765218735,
      "epoch": 0.6502673796791444,
      "grad_norm": 0.1860886961221695,
      "learning_rate": 7.200754951234649e-05,
      "loss": 0.9847,
      "mean_token_accuracy": 0.7540322542190552,
      "num_tokens": 39651362.0,
      "step": 1216
    },
    {
      "entropy": 0.9701773524284363,
      "epoch": 0.6508021390374331,
      "grad_norm": 0.17255829274654388,
      "learning_rate": 7.186623005191288e-05,
      "loss": 0.9622,
      "mean_token_accuracy": 0.7551049739122391,
      "num_tokens": 39683730.0,
      "step": 1217
    },
    {
      "entropy": 1.0146152973175049,
      "epoch": 0.651336898395722,
      "grad_norm": 0.24694116413593292,
      "learning_rate": 7.172502509366332e-05,
      "loss": 1.0029,
      "mean_token_accuracy": 0.7463648617267609,
      "num_tokens": 39716498.0,
      "step": 1218
    },
    {
      "entropy": 0.9297868311405182,
      "epoch": 0.6518716577540107,
      "grad_norm": 0.15562401711940765,
      "learning_rate": 7.158393506158622e-05,
      "loss": 0.9221,
      "mean_token_accuracy": 0.7655511498451233,
      "num_tokens": 39749029.0,
      "step": 1219
    },
    {
      "entropy": 0.9979990869760513,
      "epoch": 0.6524064171122995,
      "grad_norm": 0.17048972845077515,
      "learning_rate": 7.144296037932496e-05,
      "loss": 0.9864,
      "mean_token_accuracy": 0.7465503215789795,
      "num_tokens": 39781597.0,
      "step": 1220
    },
    {
      "entropy": 0.9215275347232819,
      "epoch": 0.6529411764705882,
      "grad_norm": 0.18462637066841125,
      "learning_rate": 7.13021014701765e-05,
      "loss": 0.9164,
      "mean_token_accuracy": 0.7666177749633789,
      "num_tokens": 39814365.0,
      "step": 1221
    },
    {
      "entropy": 0.9937931597232819,
      "epoch": 0.653475935828877,
      "grad_norm": 0.2151746302843094,
      "learning_rate": 7.116135875709011e-05,
      "loss": 1.0027,
      "mean_token_accuracy": 0.7481340318918228,
      "num_tokens": 39846886.0,
      "step": 1222
    },
    {
      "entropy": 1.022530347108841,
      "epoch": 0.6540106951871658,
      "grad_norm": 0.16658732295036316,
      "learning_rate": 7.102073266266632e-05,
      "loss": 1.0072,
      "mean_token_accuracy": 0.7454576939344406,
      "num_tokens": 39879205.0,
      "step": 1223
    },
    {
      "entropy": 0.9836557805538177,
      "epoch": 0.6545454545454545,
      "grad_norm": 0.1658448576927185,
      "learning_rate": 7.088022360915536e-05,
      "loss": 0.9754,
      "mean_token_accuracy": 0.7512788772583008,
      "num_tokens": 39911547.0,
      "step": 1224
    },
    {
      "entropy": 0.9708831012248993,
      "epoch": 0.6550802139037433,
      "grad_norm": 0.1592826098203659,
      "learning_rate": 7.073983201845602e-05,
      "loss": 0.9737,
      "mean_token_accuracy": 0.7533071786165237,
      "num_tokens": 39944224.0,
      "step": 1225
    },
    {
      "entropy": 0.9460985213518143,
      "epoch": 0.6556149732620321,
      "grad_norm": 0.16502933204174042,
      "learning_rate": 7.059955831211448e-05,
      "loss": 0.9443,
      "mean_token_accuracy": 0.757226288318634,
      "num_tokens": 39976628.0,
      "step": 1226
    },
    {
      "entropy": 0.9379078596830368,
      "epoch": 0.6561497326203208,
      "grad_norm": 0.16671308875083923,
      "learning_rate": 7.0459402911323e-05,
      "loss": 0.9378,
      "mean_token_accuracy": 0.7584051489830017,
      "num_tokens": 40009023.0,
      "step": 1227
    },
    {
      "entropy": 0.9769723415374756,
      "epoch": 0.6566844919786097,
      "grad_norm": 0.1969531625509262,
      "learning_rate": 7.031936623691848e-05,
      "loss": 0.9783,
      "mean_token_accuracy": 0.7504943460226059,
      "num_tokens": 40041466.0,
      "step": 1228
    },
    {
      "entropy": 1.0261331349611282,
      "epoch": 0.6572192513368984,
      "grad_norm": 0.18100857734680176,
      "learning_rate": 7.017944870938133e-05,
      "loss": 1.0133,
      "mean_token_accuracy": 0.745662271976471,
      "num_tokens": 40074234.0,
      "step": 1229
    },
    {
      "entropy": 0.9616502523422241,
      "epoch": 0.6577540106951871,
      "grad_norm": 0.18322017788887024,
      "learning_rate": 7.003965074883433e-05,
      "loss": 0.9778,
      "mean_token_accuracy": 0.7553763389587402,
      "num_tokens": 40107002.0,
      "step": 1230
    },
    {
      "entropy": 0.942153662443161,
      "epoch": 0.658288770053476,
      "grad_norm": 0.20072796940803528,
      "learning_rate": 6.98999727750411e-05,
      "loss": 0.9439,
      "mean_token_accuracy": 0.7598668038845062,
      "num_tokens": 40139770.0,
      "step": 1231
    },
    {
      "entropy": 0.9993161708116531,
      "epoch": 0.6588235294117647,
      "grad_norm": 0.16346585750579834,
      "learning_rate": 6.976041520740513e-05,
      "loss": 0.99,
      "mean_token_accuracy": 0.7468269765377045,
      "num_tokens": 40172388.0,
      "step": 1232
    },
    {
      "entropy": 0.9855088144540787,
      "epoch": 0.6593582887700534,
      "grad_norm": 0.16769243776798248,
      "learning_rate": 6.962097846496818e-05,
      "loss": 0.9849,
      "mean_token_accuracy": 0.7512524276971817,
      "num_tokens": 40205156.0,
      "step": 1233
    },
    {
      "entropy": 0.9597685039043427,
      "epoch": 0.6598930481283423,
      "grad_norm": 0.22894522547721863,
      "learning_rate": 6.948166296640939e-05,
      "loss": 0.9432,
      "mean_token_accuracy": 0.7628034800291061,
      "num_tokens": 40237572.0,
      "step": 1234
    },
    {
      "entropy": 1.005010411143303,
      "epoch": 0.660427807486631,
      "grad_norm": 0.17149806022644043,
      "learning_rate": 6.934246913004383e-05,
      "loss": 1.0129,
      "mean_token_accuracy": 0.7463954091072083,
      "num_tokens": 40270340.0,
      "step": 1235
    },
    {
      "entropy": 0.9686301201581955,
      "epoch": 0.6609625668449198,
      "grad_norm": 0.1712532788515091,
      "learning_rate": 6.920339737382111e-05,
      "loss": 0.9731,
      "mean_token_accuracy": 0.7489613890647888,
      "num_tokens": 40303108.0,
      "step": 1236
    },
    {
      "entropy": 0.9406851232051849,
      "epoch": 0.6614973262032086,
      "grad_norm": 0.1674990952014923,
      "learning_rate": 6.906444811532447e-05,
      "loss": 0.9435,
      "mean_token_accuracy": 0.7626807540655136,
      "num_tokens": 40335456.0,
      "step": 1237
    },
    {
      "entropy": 0.9710539728403091,
      "epoch": 0.6620320855614973,
      "grad_norm": 0.16204853355884552,
      "learning_rate": 6.892562177176918e-05,
      "loss": 0.9718,
      "mean_token_accuracy": 0.7541085183620453,
      "num_tokens": 40368044.0,
      "step": 1238
    },
    {
      "entropy": 0.9809377044439316,
      "epoch": 0.6625668449197861,
      "grad_norm": 0.1876448392868042,
      "learning_rate": 6.878691876000158e-05,
      "loss": 0.9988,
      "mean_token_accuracy": 0.7467008829116821,
      "num_tokens": 40400812.0,
      "step": 1239
    },
    {
      "entropy": 0.937059685587883,
      "epoch": 0.6631016042780749,
      "grad_norm": 0.1668710559606552,
      "learning_rate": 6.864833949649752e-05,
      "loss": 0.9448,
      "mean_token_accuracy": 0.7630942314863205,
      "num_tokens": 40433142.0,
      "step": 1240
    },
    {
      "entropy": 0.9396518021821976,
      "epoch": 0.6636363636363637,
      "grad_norm": 0.1635650247335434,
      "learning_rate": 6.85098843973614e-05,
      "loss": 0.928,
      "mean_token_accuracy": 0.7608862668275833,
      "num_tokens": 40465764.0,
      "step": 1241
    },
    {
      "entropy": 0.9529020488262177,
      "epoch": 0.6641711229946524,
      "grad_norm": 0.17254072427749634,
      "learning_rate": 6.837155387832485e-05,
      "loss": 0.961,
      "mean_token_accuracy": 0.7572091817855835,
      "num_tokens": 40498532.0,
      "step": 1242
    },
    {
      "entropy": 0.9303242862224579,
      "epoch": 0.6647058823529411,
      "grad_norm": 0.1784360408782959,
      "learning_rate": 6.823334835474523e-05,
      "loss": 0.9358,
      "mean_token_accuracy": 0.7638181447982788,
      "num_tokens": 40531218.0,
      "step": 1243
    },
    {
      "entropy": 0.8833232223987579,
      "epoch": 0.66524064171123,
      "grad_norm": 0.17049525678157806,
      "learning_rate": 6.809526824160477e-05,
      "loss": 0.8735,
      "mean_token_accuracy": 0.7731898725032806,
      "num_tokens": 40563858.0,
      "step": 1244
    },
    {
      "entropy": 0.9500382095575333,
      "epoch": 0.6657754010695187,
      "grad_norm": 0.21515445411205292,
      "learning_rate": 6.795731395350908e-05,
      "loss": 0.9341,
      "mean_token_accuracy": 0.7637630701065063,
      "num_tokens": 40596268.0,
      "step": 1245
    },
    {
      "entropy": 0.9598138481378555,
      "epoch": 0.6663101604278074,
      "grad_norm": 0.1596616506576538,
      "learning_rate": 6.781948590468592e-05,
      "loss": 0.9531,
      "mean_token_accuracy": 0.7549792230129242,
      "num_tokens": 40629036.0,
      "step": 1246
    },
    {
      "entropy": 0.9623348414897919,
      "epoch": 0.6668449197860963,
      "grad_norm": 0.1683739870786667,
      "learning_rate": 6.768178450898401e-05,
      "loss": 0.9509,
      "mean_token_accuracy": 0.7548241764307022,
      "num_tokens": 40661776.0,
      "step": 1247
    },
    {
      "entropy": 0.9414125084877014,
      "epoch": 0.667379679144385,
      "grad_norm": 0.1648664027452469,
      "learning_rate": 6.754421017987181e-05,
      "loss": 0.938,
      "mean_token_accuracy": 0.7581561654806137,
      "num_tokens": 40694544.0,
      "step": 1248
    },
    {
      "entropy": 0.9510121196508408,
      "epoch": 0.6679144385026738,
      "grad_norm": 0.17633424699306488,
      "learning_rate": 6.740676333043627e-05,
      "loss": 0.95,
      "mean_token_accuracy": 0.7547886073589325,
      "num_tokens": 40726790.0,
      "step": 1249
    },
    {
      "entropy": 0.9642129242420197,
      "epoch": 0.6684491978609626,
      "grad_norm": 0.17034272849559784,
      "learning_rate": 6.72694443733815e-05,
      "loss": 0.9682,
      "mean_token_accuracy": 0.756823718547821,
      "num_tokens": 40758825.0,
      "step": 1250
    },
    {
      "entropy": 0.9768649190664291,
      "epoch": 0.6689839572192513,
      "grad_norm": 0.1960660219192505,
      "learning_rate": 6.713225372102757e-05,
      "loss": 0.9789,
      "mean_token_accuracy": 0.7500343322753906,
      "num_tokens": 40791537.0,
      "step": 1251
    },
    {
      "entropy": 0.9842686802148819,
      "epoch": 0.6695187165775401,
      "grad_norm": 0.19921888411045074,
      "learning_rate": 6.699519178530944e-05,
      "loss": 0.9889,
      "mean_token_accuracy": 0.7502443790435791,
      "num_tokens": 40824305.0,
      "step": 1252
    },
    {
      "entropy": 0.9305083900690079,
      "epoch": 0.6700534759358289,
      "grad_norm": 0.1651991307735443,
      "learning_rate": 6.685825897777541e-05,
      "loss": 0.9372,
      "mean_token_accuracy": 0.7601883560419083,
      "num_tokens": 40856955.0,
      "step": 1253
    },
    {
      "entropy": 0.9830394834280014,
      "epoch": 0.6705882352941176,
      "grad_norm": 0.1739017516374588,
      "learning_rate": 6.672145570958613e-05,
      "loss": 0.9977,
      "mean_token_accuracy": 0.7495644688606262,
      "num_tokens": 40889555.0,
      "step": 1254
    },
    {
      "entropy": 0.9631999880075455,
      "epoch": 0.6711229946524064,
      "grad_norm": 0.17304059863090515,
      "learning_rate": 6.65847823915133e-05,
      "loss": 0.9629,
      "mean_token_accuracy": 0.7510248422622681,
      "num_tokens": 40922091.0,
      "step": 1255
    },
    {
      "entropy": 0.9633833467960358,
      "epoch": 0.6716577540106952,
      "grad_norm": 0.17870648205280304,
      "learning_rate": 6.644823943393839e-05,
      "loss": 0.9695,
      "mean_token_accuracy": 0.7529020011425018,
      "num_tokens": 40954859.0,
      "step": 1256
    },
    {
      "entropy": 0.9755293428897858,
      "epoch": 0.672192513368984,
      "grad_norm": 0.16337575018405914,
      "learning_rate": 6.631182724685153e-05,
      "loss": 0.9649,
      "mean_token_accuracy": 0.7519892007112503,
      "num_tokens": 40987616.0,
      "step": 1257
    },
    {
      "entropy": 0.963219404220581,
      "epoch": 0.6727272727272727,
      "grad_norm": 0.1670699417591095,
      "learning_rate": 6.617554623985006e-05,
      "loss": 0.9572,
      "mean_token_accuracy": 0.7563893646001816,
      "num_tokens": 41020261.0,
      "step": 1258
    },
    {
      "entropy": 0.9529228955507278,
      "epoch": 0.6732620320855615,
      "grad_norm": 0.17823971807956696,
      "learning_rate": 6.603939682213756e-05,
      "loss": 0.9515,
      "mean_token_accuracy": 0.7584005296230316,
      "num_tokens": 41053029.0,
      "step": 1259
    },
    {
      "entropy": 0.9679936170578003,
      "epoch": 0.6737967914438503,
      "grad_norm": 0.16420431435108185,
      "learning_rate": 6.590337940252237e-05,
      "loss": 0.9687,
      "mean_token_accuracy": 0.7551368027925491,
      "num_tokens": 41085587.0,
      "step": 1260
    },
    {
      "entropy": 0.9809670001268387,
      "epoch": 0.674331550802139,
      "grad_norm": 0.16018334031105042,
      "learning_rate": 6.576749438941665e-05,
      "loss": 0.9758,
      "mean_token_accuracy": 0.7510380744934082,
      "num_tokens": 41118115.0,
      "step": 1261
    },
    {
      "entropy": 0.9639696776866913,
      "epoch": 0.6748663101604279,
      "grad_norm": 0.16860608756542206,
      "learning_rate": 6.563174219083476e-05,
      "loss": 0.9555,
      "mean_token_accuracy": 0.7552216500043869,
      "num_tokens": 41150743.0,
      "step": 1262
    },
    {
      "entropy": 1.0253010541200638,
      "epoch": 0.6754010695187166,
      "grad_norm": 0.16424211859703064,
      "learning_rate": 6.549612321439249e-05,
      "loss": 1.0096,
      "mean_token_accuracy": 0.7431881278753281,
      "num_tokens": 41183310.0,
      "step": 1263
    },
    {
      "entropy": 0.9711318612098694,
      "epoch": 0.6759358288770053,
      "grad_norm": 0.16268914937973022,
      "learning_rate": 6.536063786730554e-05,
      "loss": 0.9595,
      "mean_token_accuracy": 0.759953111410141,
      "num_tokens": 41215825.0,
      "step": 1264
    },
    {
      "entropy": 0.9563075453042984,
      "epoch": 0.6764705882352942,
      "grad_norm": 0.16132086515426636,
      "learning_rate": 6.522528655638824e-05,
      "loss": 0.9445,
      "mean_token_accuracy": 0.7609447985887527,
      "num_tokens": 41248582.0,
      "step": 1265
    },
    {
      "entropy": 0.9650041162967682,
      "epoch": 0.6770053475935829,
      "grad_norm": 0.1783258467912674,
      "learning_rate": 6.50900696880527e-05,
      "loss": 0.9733,
      "mean_token_accuracy": 0.7552541494369507,
      "num_tokens": 41281350.0,
      "step": 1266
    },
    {
      "entropy": 0.9366767555475235,
      "epoch": 0.6775401069518716,
      "grad_norm": 0.17735034227371216,
      "learning_rate": 6.495498766830706e-05,
      "loss": 0.9392,
      "mean_token_accuracy": 0.7608137875795364,
      "num_tokens": 41314118.0,
      "step": 1267
    },
    {
      "entropy": 0.9627933204174042,
      "epoch": 0.6780748663101605,
      "grad_norm": 0.16228435933589935,
      "learning_rate": 6.482004090275484e-05,
      "loss": 0.9575,
      "mean_token_accuracy": 0.7606304883956909,
      "num_tokens": 41346886.0,
      "step": 1268
    },
    {
      "entropy": 0.935073509812355,
      "epoch": 0.6786096256684492,
      "grad_norm": 0.19033820927143097,
      "learning_rate": 6.468522979659322e-05,
      "loss": 0.9283,
      "mean_token_accuracy": 0.763078510761261,
      "num_tokens": 41378997.0,
      "step": 1269
    },
    {
      "entropy": 0.9309537261724472,
      "epoch": 0.679144385026738,
      "grad_norm": 0.17397263646125793,
      "learning_rate": 6.455055475461216e-05,
      "loss": 0.9311,
      "mean_token_accuracy": 0.7598057091236115,
      "num_tokens": 41411765.0,
      "step": 1270
    },
    {
      "entropy": 1.0121001154184341,
      "epoch": 0.6796791443850267,
      "grad_norm": 0.16641665995121002,
      "learning_rate": 6.441601618119308e-05,
      "loss": 1.0186,
      "mean_token_accuracy": 0.7441959977149963,
      "num_tokens": 41444533.0,
      "step": 1271
    },
    {
      "entropy": 0.9392533451318741,
      "epoch": 0.6802139037433155,
      "grad_norm": 0.18284034729003906,
      "learning_rate": 6.428161448030751e-05,
      "loss": 0.9229,
      "mean_token_accuracy": 0.7665873914957047,
      "num_tokens": 41477064.0,
      "step": 1272
    },
    {
      "entropy": 0.9318648427724838,
      "epoch": 0.6807486631016043,
      "grad_norm": 0.16083700954914093,
      "learning_rate": 6.414735005551617e-05,
      "loss": 0.9361,
      "mean_token_accuracy": 0.7590725719928741,
      "num_tokens": 41509832.0,
      "step": 1273
    },
    {
      "entropy": 1.0033331364393234,
      "epoch": 0.681283422459893,
      "grad_norm": 0.16821704804897308,
      "learning_rate": 6.401322330996743e-05,
      "loss": 1.0041,
      "mean_token_accuracy": 0.7469452619552612,
      "num_tokens": 41542600.0,
      "step": 1274
    },
    {
      "entropy": 0.9760285168886185,
      "epoch": 0.6818181818181818,
      "grad_norm": 0.18433675169944763,
      "learning_rate": 6.387923464639638e-05,
      "loss": 0.9945,
      "mean_token_accuracy": 0.7464347928762436,
      "num_tokens": 41575072.0,
      "step": 1275
    },
    {
      "entropy": 0.9680602550506592,
      "epoch": 0.6823529411764706,
      "grad_norm": 0.16477592289447784,
      "learning_rate": 6.374538446712339e-05,
      "loss": 0.9671,
      "mean_token_accuracy": 0.7558453530073166,
      "num_tokens": 41607602.0,
      "step": 1276
    },
    {
      "entropy": 0.9561006724834442,
      "epoch": 0.6828877005347593,
      "grad_norm": 0.16161902248859406,
      "learning_rate": 6.361167317405309e-05,
      "loss": 0.9568,
      "mean_token_accuracy": 0.758235827088356,
      "num_tokens": 41640162.0,
      "step": 1277
    },
    {
      "entropy": 0.9576833844184875,
      "epoch": 0.6834224598930482,
      "grad_norm": 0.16424843668937683,
      "learning_rate": 6.34781011686731e-05,
      "loss": 0.9579,
      "mean_token_accuracy": 0.7535740435123444,
      "num_tokens": 41672930.0,
      "step": 1278
    },
    {
      "entropy": 0.96244677901268,
      "epoch": 0.6839572192513369,
      "grad_norm": 0.1593867689371109,
      "learning_rate": 6.334466885205275e-05,
      "loss": 0.9673,
      "mean_token_accuracy": 0.7590638697147369,
      "num_tokens": 41705393.0,
      "step": 1279
    },
    {
      "entropy": 1.010896772146225,
      "epoch": 0.6844919786096256,
      "grad_norm": 0.1620362401008606,
      "learning_rate": 6.321137662484195e-05,
      "loss": 1.0045,
      "mean_token_accuracy": 0.7459371984004974,
      "num_tokens": 41738161.0,
      "step": 1280
    },
    {
      "entropy": 0.9322298914194107,
      "epoch": 0.6850267379679145,
      "grad_norm": 0.16803748905658722,
      "learning_rate": 6.307822488726996e-05,
      "loss": 0.9421,
      "mean_token_accuracy": 0.7560789287090302,
      "num_tokens": 41770929.0,
      "step": 1281
    },
    {
      "entropy": 0.9569824635982513,
      "epoch": 0.6855614973262032,
      "grad_norm": 0.16779689490795135,
      "learning_rate": 6.294521403914425e-05,
      "loss": 0.9551,
      "mean_token_accuracy": 0.7504887580871582,
      "num_tokens": 41803697.0,
      "step": 1282
    },
    {
      "entropy": 0.973842054605484,
      "epoch": 0.686096256684492,
      "grad_norm": 0.16040951013565063,
      "learning_rate": 6.281234447984917e-05,
      "loss": 0.9628,
      "mean_token_accuracy": 0.7519855797290802,
      "num_tokens": 41836465.0,
      "step": 1283
    },
    {
      "entropy": 0.9762571305036545,
      "epoch": 0.6866310160427808,
      "grad_norm": 0.16406090557575226,
      "learning_rate": 6.26796166083449e-05,
      "loss": 0.9739,
      "mean_token_accuracy": 0.7509875148534775,
      "num_tokens": 41869136.0,
      "step": 1284
    },
    {
      "entropy": 0.9796842485666275,
      "epoch": 0.6871657754010695,
      "grad_norm": 0.16604503989219666,
      "learning_rate": 6.254703082316613e-05,
      "loss": 0.9696,
      "mean_token_accuracy": 0.7559634149074554,
      "num_tokens": 41901703.0,
      "step": 1285
    },
    {
      "entropy": 0.9618323147296906,
      "epoch": 0.6877005347593583,
      "grad_norm": 0.16067545115947723,
      "learning_rate": 6.241458752242102e-05,
      "loss": 0.9503,
      "mean_token_accuracy": 0.7568731606006622,
      "num_tokens": 41934471.0,
      "step": 1286
    },
    {
      "entropy": 0.9434874653816223,
      "epoch": 0.6882352941176471,
      "grad_norm": 0.16011638939380646,
      "learning_rate": 6.228228710378976e-05,
      "loss": 0.9432,
      "mean_token_accuracy": 0.7582783401012421,
      "num_tokens": 41967239.0,
      "step": 1287
    },
    {
      "entropy": 1.0110597610473633,
      "epoch": 0.6887700534759358,
      "grad_norm": 0.17058852314949036,
      "learning_rate": 6.215012996452354e-05,
      "loss": 1.0262,
      "mean_token_accuracy": 0.745509535074234,
      "num_tokens": 42000007.0,
      "step": 1288
    },
    {
      "entropy": 0.9647829830646515,
      "epoch": 0.6893048128342246,
      "grad_norm": 0.16275008022785187,
      "learning_rate": 6.201811650144342e-05,
      "loss": 0.9663,
      "mean_token_accuracy": 0.7523216009140015,
      "num_tokens": 42032775.0,
      "step": 1289
    },
    {
      "entropy": 0.939326599240303,
      "epoch": 0.6898395721925134,
      "grad_norm": 0.16183090209960938,
      "learning_rate": 6.188624711093905e-05,
      "loss": 0.9302,
      "mean_token_accuracy": 0.7647543996572495,
      "num_tokens": 42065543.0,
      "step": 1290
    },
    {
      "entropy": 0.9778326153755188,
      "epoch": 0.6903743315508022,
      "grad_norm": 0.16926462948322296,
      "learning_rate": 6.175452218896731e-05,
      "loss": 0.9935,
      "mean_token_accuracy": 0.7510386109352112,
      "num_tokens": 42098311.0,
      "step": 1291
    },
    {
      "entropy": 1.0116349458694458,
      "epoch": 0.6909090909090909,
      "grad_norm": 0.16622833907604218,
      "learning_rate": 6.162294213105149e-05,
      "loss": 1.0144,
      "mean_token_accuracy": 0.744976207613945,
      "num_tokens": 42130923.0,
      "step": 1292
    },
    {
      "entropy": 0.9861657917499542,
      "epoch": 0.6914438502673796,
      "grad_norm": 0.16684715449810028,
      "learning_rate": 6.149150733227986e-05,
      "loss": 0.9918,
      "mean_token_accuracy": 0.7465481460094452,
      "num_tokens": 42163691.0,
      "step": 1293
    },
    {
      "entropy": 0.8746530562639236,
      "epoch": 0.6919786096256685,
      "grad_norm": 0.16398267447948456,
      "learning_rate": 6.136021818730441e-05,
      "loss": 0.87,
      "mean_token_accuracy": 0.7736437022686005,
      "num_tokens": 42196459.0,
      "step": 1294
    },
    {
      "entropy": 0.9123699069023132,
      "epoch": 0.6925133689839572,
      "grad_norm": 0.18755117058753967,
      "learning_rate": 6.122907509033997e-05,
      "loss": 0.9109,
      "mean_token_accuracy": 0.7674476206302643,
      "num_tokens": 42228914.0,
      "step": 1295
    },
    {
      "entropy": 0.9567636549472809,
      "epoch": 0.6930481283422459,
      "grad_norm": 0.16553348302841187,
      "learning_rate": 6.109807843516264e-05,
      "loss": 0.9483,
      "mean_token_accuracy": 0.7542024850845337,
      "num_tokens": 42261592.0,
      "step": 1296
    },
    {
      "entropy": 0.9956474155187607,
      "epoch": 0.6935828877005348,
      "grad_norm": 0.17825832962989807,
      "learning_rate": 6.096722861510898e-05,
      "loss": 1.0149,
      "mean_token_accuracy": 0.7447462528944016,
      "num_tokens": 42294101.0,
      "step": 1297
    },
    {
      "entropy": 0.9219067543745041,
      "epoch": 0.6941176470588235,
      "grad_norm": 0.1673230230808258,
      "learning_rate": 6.083652602307455e-05,
      "loss": 0.9053,
      "mean_token_accuracy": 0.7665437757968903,
      "num_tokens": 42326639.0,
      "step": 1298
    },
    {
      "entropy": 0.9638742357492447,
      "epoch": 0.6946524064171123,
      "grad_norm": 0.17454005777835846,
      "learning_rate": 6.0705971051512856e-05,
      "loss": 0.9518,
      "mean_token_accuracy": 0.7599993348121643,
      "num_tokens": 42359065.0,
      "step": 1299
    },
    {
      "entropy": 1.0120955258607864,
      "epoch": 0.6951871657754011,
      "grad_norm": 0.17239855229854584,
      "learning_rate": 6.057556409243424e-05,
      "loss": 1.0185,
      "mean_token_accuracy": 0.7423326075077057,
      "num_tokens": 42391833.0,
      "step": 1300
    },
    {
      "entropy": 1.004928171634674,
      "epoch": 0.6957219251336898,
      "grad_norm": 0.16988736391067505,
      "learning_rate": 6.0445305537404427e-05,
      "loss": 0.9945,
      "mean_token_accuracy": 0.7474340200424194,
      "num_tokens": 42424601.0,
      "step": 1301
    },
    {
      "entropy": 0.9260279536247253,
      "epoch": 0.6962566844919786,
      "grad_norm": 0.19346578419208527,
      "learning_rate": 6.031519577754377e-05,
      "loss": 0.9218,
      "mean_token_accuracy": 0.7631844580173492,
      "num_tokens": 42456794.0,
      "step": 1302
    },
    {
      "entropy": 0.9603572189807892,
      "epoch": 0.6967914438502674,
      "grad_norm": 0.16860198974609375,
      "learning_rate": 6.0185235203525636e-05,
      "loss": 0.9448,
      "mean_token_accuracy": 0.7589503824710846,
      "num_tokens": 42489562.0,
      "step": 1303
    },
    {
      "entropy": 0.9546780437231064,
      "epoch": 0.6973262032085561,
      "grad_norm": 0.1686096489429474,
      "learning_rate": 6.0055424205575594e-05,
      "loss": 0.9671,
      "mean_token_accuracy": 0.7567043155431747,
      "num_tokens": 42522264.0,
      "step": 1304
    },
    {
      "entropy": 0.9691311120986938,
      "epoch": 0.6978609625668449,
      "grad_norm": 0.21208062767982483,
      "learning_rate": 5.992576317346995e-05,
      "loss": 0.9512,
      "mean_token_accuracy": 0.7584279030561447,
      "num_tokens": 42554963.0,
      "step": 1305
    },
    {
      "entropy": 0.9300960153341293,
      "epoch": 0.6983957219251337,
      "grad_norm": 0.1617356240749359,
      "learning_rate": 5.9796252496534845e-05,
      "loss": 0.9162,
      "mean_token_accuracy": 0.7626371830701828,
      "num_tokens": 42587269.0,
      "step": 1306
    },
    {
      "entropy": 0.9444008618593216,
      "epoch": 0.6989304812834225,
      "grad_norm": 0.1645589917898178,
      "learning_rate": 5.966689256364486e-05,
      "loss": 0.9397,
      "mean_token_accuracy": 0.7619356215000153,
      "num_tokens": 42619692.0,
      "step": 1307
    },
    {
      "entropy": 0.9678683280944824,
      "epoch": 0.6994652406417112,
      "grad_norm": 0.18692892789840698,
      "learning_rate": 5.9537683763222064e-05,
      "loss": 0.9769,
      "mean_token_accuracy": 0.7515884637832642,
      "num_tokens": 42652460.0,
      "step": 1308
    },
    {
      "entropy": 0.9600773751735687,
      "epoch": 0.7,
      "grad_norm": 0.17012649774551392,
      "learning_rate": 5.94086264832346e-05,
      "loss": 0.9561,
      "mean_token_accuracy": 0.7551014125347137,
      "num_tokens": 42685228.0,
      "step": 1309
    },
    {
      "entropy": 0.9367721974849701,
      "epoch": 0.7005347593582888,
      "grad_norm": 0.1679380238056183,
      "learning_rate": 5.927972111119565e-05,
      "loss": 0.9229,
      "mean_token_accuracy": 0.7605284154415131,
      "num_tokens": 42717805.0,
      "step": 1310
    },
    {
      "entropy": 0.9343129098415375,
      "epoch": 0.7010695187165775,
      "grad_norm": 0.21657076478004456,
      "learning_rate": 5.915096803416243e-05,
      "loss": 0.9427,
      "mean_token_accuracy": 0.7589403390884399,
      "num_tokens": 42750378.0,
      "step": 1311
    },
    {
      "entropy": 0.9220041334629059,
      "epoch": 0.7016042780748664,
      "grad_norm": 0.19568952918052673,
      "learning_rate": 5.902236763873463e-05,
      "loss": 0.9127,
      "mean_token_accuracy": 0.7643386423587799,
      "num_tokens": 42782938.0,
      "step": 1312
    },
    {
      "entropy": 0.9896649420261383,
      "epoch": 0.7021390374331551,
      "grad_norm": 0.1709054410457611,
      "learning_rate": 5.889392031105371e-05,
      "loss": 0.9861,
      "mean_token_accuracy": 0.7547959536314011,
      "num_tokens": 42815706.0,
      "step": 1313
    },
    {
      "entropy": 0.9917427897453308,
      "epoch": 0.7026737967914438,
      "grad_norm": 0.1694716513156891,
      "learning_rate": 5.876562643680139e-05,
      "loss": 0.9975,
      "mean_token_accuracy": 0.7487170100212097,
      "num_tokens": 42848474.0,
      "step": 1314
    },
    {
      "entropy": 0.9077127426862717,
      "epoch": 0.7032085561497327,
      "grad_norm": 0.18303543329238892,
      "learning_rate": 5.8637486401198725e-05,
      "loss": 0.913,
      "mean_token_accuracy": 0.7661791741847992,
      "num_tokens": 42880895.0,
      "step": 1315
    },
    {
      "entropy": 0.9104649573564529,
      "epoch": 0.7037433155080214,
      "grad_norm": 0.20730172097682953,
      "learning_rate": 5.850950058900475e-05,
      "loss": 0.9204,
      "mean_token_accuracy": 0.7629520893096924,
      "num_tokens": 42913663.0,
      "step": 1316
    },
    {
      "entropy": 0.965117797255516,
      "epoch": 0.7042780748663101,
      "grad_norm": 0.15987767279148102,
      "learning_rate": 5.838166938451542e-05,
      "loss": 0.9575,
      "mean_token_accuracy": 0.7545472979545593,
      "num_tokens": 42946046.0,
      "step": 1317
    },
    {
      "entropy": 0.9923857301473618,
      "epoch": 0.704812834224599,
      "grad_norm": 0.27166005969047546,
      "learning_rate": 5.825399317156253e-05,
      "loss": 0.9909,
      "mean_token_accuracy": 0.746456503868103,
      "num_tokens": 42978814.0,
      "step": 1318
    },
    {
      "entropy": 1.0188444405794144,
      "epoch": 0.7053475935828877,
      "grad_norm": 0.1819809079170227,
      "learning_rate": 5.812647233351249e-05,
      "loss": 1.0373,
      "mean_token_accuracy": 0.7383614480495453,
      "num_tokens": 43011582.0,
      "step": 1319
    },
    {
      "entropy": 0.9351516962051392,
      "epoch": 0.7058823529411765,
      "grad_norm": 0.19748146831989288,
      "learning_rate": 5.7999107253265085e-05,
      "loss": 0.9504,
      "mean_token_accuracy": 0.7600806355476379,
      "num_tokens": 43044350.0,
      "step": 1320
    },
    {
      "entropy": 0.9284085929393768,
      "epoch": 0.7064171122994652,
      "grad_norm": 0.17606669664382935,
      "learning_rate": 5.787189831325248e-05,
      "loss": 0.9315,
      "mean_token_accuracy": 0.7659762799739838,
      "num_tokens": 43077118.0,
      "step": 1321
    },
    {
      "entropy": 0.9412466883659363,
      "epoch": 0.706951871657754,
      "grad_norm": 0.17121106386184692,
      "learning_rate": 5.774484589543805e-05,
      "loss": 0.9439,
      "mean_token_accuracy": 0.7576881498098373,
      "num_tokens": 43109709.0,
      "step": 1322
    },
    {
      "entropy": 0.9344121664762497,
      "epoch": 0.7074866310160428,
      "grad_norm": 0.15921074151992798,
      "learning_rate": 5.761795038131512e-05,
      "loss": 0.9272,
      "mean_token_accuracy": 0.765518069267273,
      "num_tokens": 43142477.0,
      "step": 1323
    },
    {
      "entropy": 0.9707403481006622,
      "epoch": 0.7080213903743315,
      "grad_norm": 0.18937453627586365,
      "learning_rate": 5.749121215190584e-05,
      "loss": 0.9742,
      "mean_token_accuracy": 0.7521077692508698,
      "num_tokens": 43175245.0,
      "step": 1324
    },
    {
      "entropy": 0.9573013186454773,
      "epoch": 0.7085561497326203,
      "grad_norm": 0.16704417765140533,
      "learning_rate": 5.7364631587760195e-05,
      "loss": 0.9533,
      "mean_token_accuracy": 0.7544599175453186,
      "num_tokens": 43208013.0,
      "step": 1325
    },
    {
      "entropy": 0.9601723700761795,
      "epoch": 0.7090909090909091,
      "grad_norm": 0.16236352920532227,
      "learning_rate": 5.7238209068954765e-05,
      "loss": 0.9446,
      "mean_token_accuracy": 0.7608654052019119,
      "num_tokens": 43240510.0,
      "step": 1326
    },
    {
      "entropy": 1.0053239911794662,
      "epoch": 0.7096256684491978,
      "grad_norm": 0.16887569427490234,
      "learning_rate": 5.7111944975091445e-05,
      "loss": 0.9963,
      "mean_token_accuracy": 0.7504582107067108,
      "num_tokens": 43273278.0,
      "step": 1327
    },
    {
      "entropy": 0.9479286074638367,
      "epoch": 0.7101604278074867,
      "grad_norm": 0.166823610663414,
      "learning_rate": 5.698583968529657e-05,
      "loss": 0.9481,
      "mean_token_accuracy": 0.7575452029705048,
      "num_tokens": 43306046.0,
      "step": 1328
    },
    {
      "entropy": 0.972056582570076,
      "epoch": 0.7106951871657754,
      "grad_norm": 0.19135132431983948,
      "learning_rate": 5.6859893578219616e-05,
      "loss": 0.9712,
      "mean_token_accuracy": 0.7543448656797409,
      "num_tokens": 43338499.0,
      "step": 1329
    },
    {
      "entropy": 0.9440675228834152,
      "epoch": 0.7112299465240641,
      "grad_norm": 0.16059821844100952,
      "learning_rate": 5.673410703203204e-05,
      "loss": 0.9329,
      "mean_token_accuracy": 0.7643572688102722,
      "num_tokens": 43371267.0,
      "step": 1330
    },
    {
      "entropy": 0.9147678464651108,
      "epoch": 0.711764705882353,
      "grad_norm": 0.16524194180965424,
      "learning_rate": 5.660848042442617e-05,
      "loss": 0.9154,
      "mean_token_accuracy": 0.7649071216583252,
      "num_tokens": 43404035.0,
      "step": 1331
    },
    {
      "entropy": 0.9377564489841461,
      "epoch": 0.7122994652406417,
      "grad_norm": 0.17137596011161804,
      "learning_rate": 5.648301413261421e-05,
      "loss": 0.9274,
      "mean_token_accuracy": 0.7636241465806961,
      "num_tokens": 43436803.0,
      "step": 1332
    },
    {
      "entropy": 0.9109043776988983,
      "epoch": 0.7128342245989305,
      "grad_norm": 0.17169232666492462,
      "learning_rate": 5.635770853332692e-05,
      "loss": 0.9172,
      "mean_token_accuracy": 0.7651301324367523,
      "num_tokens": 43469114.0,
      "step": 1333
    },
    {
      "entropy": 0.955856204032898,
      "epoch": 0.7133689839572193,
      "grad_norm": 0.1743943989276886,
      "learning_rate": 5.623256400281254e-05,
      "loss": 0.9703,
      "mean_token_accuracy": 0.7509894818067551,
      "num_tokens": 43501360.0,
      "step": 1334
    },
    {
      "entropy": 0.9887515902519226,
      "epoch": 0.713903743315508,
      "grad_norm": 0.17928454279899597,
      "learning_rate": 5.610758091683571e-05,
      "loss": 0.9901,
      "mean_token_accuracy": 0.74660924077034,
      "num_tokens": 43534128.0,
      "step": 1335
    },
    {
      "entropy": 0.9664270430803299,
      "epoch": 0.7144385026737968,
      "grad_norm": 0.16999004781246185,
      "learning_rate": 5.598275965067634e-05,
      "loss": 0.9711,
      "mean_token_accuracy": 0.7529754191637039,
      "num_tokens": 43566812.0,
      "step": 1336
    },
    {
      "entropy": 0.9780944734811783,
      "epoch": 0.7149732620320856,
      "grad_norm": 0.16349083185195923,
      "learning_rate": 5.585810057912839e-05,
      "loss": 0.9788,
      "mean_token_accuracy": 0.7536194622516632,
      "num_tokens": 43599361.0,
      "step": 1337
    },
    {
      "entropy": 1.0061749964952469,
      "epoch": 0.7155080213903743,
      "grad_norm": 0.1756104975938797,
      "learning_rate": 5.5733604076498793e-05,
      "loss": 1.0217,
      "mean_token_accuracy": 0.7461908757686615,
      "num_tokens": 43631461.0,
      "step": 1338
    },
    {
      "entropy": 0.9739767909049988,
      "epoch": 0.7160427807486631,
      "grad_norm": 0.16898559033870697,
      "learning_rate": 5.5609270516606425e-05,
      "loss": 0.9669,
      "mean_token_accuracy": 0.7514312863349915,
      "num_tokens": 43664011.0,
      "step": 1339
    },
    {
      "entropy": 0.9885522723197937,
      "epoch": 0.7165775401069518,
      "grad_norm": 0.16172735393047333,
      "learning_rate": 5.5485100272780864e-05,
      "loss": 0.9784,
      "mean_token_accuracy": 0.750897690653801,
      "num_tokens": 43696719.0,
      "step": 1340
    },
    {
      "entropy": 0.9710711240768433,
      "epoch": 0.7171122994652407,
      "grad_norm": 0.16851507127285004,
      "learning_rate": 5.536109371786128e-05,
      "loss": 0.9656,
      "mean_token_accuracy": 0.7565042674541473,
      "num_tokens": 43729377.0,
      "step": 1341
    },
    {
      "entropy": 0.9953486323356628,
      "epoch": 0.7176470588235294,
      "grad_norm": 0.17063665390014648,
      "learning_rate": 5.5237251224195384e-05,
      "loss": 1.0157,
      "mean_token_accuracy": 0.7456031739711761,
      "num_tokens": 43761773.0,
      "step": 1342
    },
    {
      "entropy": 0.9645649939775467,
      "epoch": 0.7181818181818181,
      "grad_norm": 0.17601872980594635,
      "learning_rate": 5.5113573163638275e-05,
      "loss": 0.9631,
      "mean_token_accuracy": 0.7569347620010376,
      "num_tokens": 43794432.0,
      "step": 1343
    },
    {
      "entropy": 0.9721208810806274,
      "epoch": 0.718716577540107,
      "grad_norm": 0.16801919043064117,
      "learning_rate": 5.4990059907551305e-05,
      "loss": 0.9477,
      "mean_token_accuracy": 0.7608031928539276,
      "num_tokens": 43826989.0,
      "step": 1344
    },
    {
      "entropy": 0.9556037187576294,
      "epoch": 0.7192513368983957,
      "grad_norm": 0.1672016680240631,
      "learning_rate": 5.4866711826800885e-05,
      "loss": 0.9584,
      "mean_token_accuracy": 0.7554068863391876,
      "num_tokens": 43859757.0,
      "step": 1345
    },
    {
      "entropy": 0.9977209866046906,
      "epoch": 0.7197860962566844,
      "grad_norm": 0.17609484493732452,
      "learning_rate": 5.474352929175761e-05,
      "loss": 1.01,
      "mean_token_accuracy": 0.7487013041973114,
      "num_tokens": 43892321.0,
      "step": 1346
    },
    {
      "entropy": 0.9755626320838928,
      "epoch": 0.7203208556149733,
      "grad_norm": 0.17414574325084686,
      "learning_rate": 5.462051267229493e-05,
      "loss": 0.9783,
      "mean_token_accuracy": 0.7515273690223694,
      "num_tokens": 43925089.0,
      "step": 1347
    },
    {
      "entropy": 0.9573537856340408,
      "epoch": 0.720855614973262,
      "grad_norm": 0.18283627927303314,
      "learning_rate": 5.449766233778815e-05,
      "loss": 0.9441,
      "mean_token_accuracy": 0.7602858394384384,
      "num_tokens": 43957550.0,
      "step": 1348
    },
    {
      "entropy": 0.9958482980728149,
      "epoch": 0.7213903743315508,
      "grad_norm": 0.18237966299057007,
      "learning_rate": 5.4374978657113185e-05,
      "loss": 0.9848,
      "mean_token_accuracy": 0.750946968793869,
      "num_tokens": 43990318.0,
      "step": 1349
    },
    {
      "entropy": 0.9420408457517624,
      "epoch": 0.7219251336898396,
      "grad_norm": 0.16323111951351166,
      "learning_rate": 5.4252461998645654e-05,
      "loss": 0.9268,
      "mean_token_accuracy": 0.7569953501224518,
      "num_tokens": 44023086.0,
      "step": 1350
    },
    {
      "entropy": 0.980293333530426,
      "epoch": 0.7224598930481283,
      "grad_norm": 0.20707236230373383,
      "learning_rate": 5.4130112730259574e-05,
      "loss": 0.9757,
      "mean_token_accuracy": 0.7540020644664764,
      "num_tokens": 44055838.0,
      "step": 1351
    },
    {
      "entropy": 0.9224491715431213,
      "epoch": 0.7229946524064171,
      "grad_norm": 0.19375093281269073,
      "learning_rate": 5.400793121932647e-05,
      "loss": 0.9268,
      "mean_token_accuracy": 0.7603861093521118,
      "num_tokens": 44088606.0,
      "step": 1352
    },
    {
      "entropy": 0.9549875855445862,
      "epoch": 0.7235294117647059,
      "grad_norm": 0.16448384523391724,
      "learning_rate": 5.3885917832713996e-05,
      "loss": 0.9622,
      "mean_token_accuracy": 0.7590114772319794,
      "num_tokens": 44121374.0,
      "step": 1353
    },
    {
      "entropy": 0.9891581833362579,
      "epoch": 0.7240641711229947,
      "grad_norm": 0.18951919674873352,
      "learning_rate": 5.376407293678511e-05,
      "loss": 1.0005,
      "mean_token_accuracy": 0.7460899353027344,
      "num_tokens": 44154142.0,
      "step": 1354
    },
    {
      "entropy": 0.9467734545469284,
      "epoch": 0.7245989304812834,
      "grad_norm": 0.18200156092643738,
      "learning_rate": 5.364239689739685e-05,
      "loss": 0.9452,
      "mean_token_accuracy": 0.7579423189163208,
      "num_tokens": 44186910.0,
      "step": 1355
    },
    {
      "entropy": 0.9351662695407867,
      "epoch": 0.7251336898395722,
      "grad_norm": 0.16058847308158875,
      "learning_rate": 5.3520890079899124e-05,
      "loss": 0.9285,
      "mean_token_accuracy": 0.7623175531625748,
      "num_tokens": 44219504.0,
      "step": 1356
    },
    {
      "entropy": 0.9368904829025269,
      "epoch": 0.725668449197861,
      "grad_norm": 0.17099831998348236,
      "learning_rate": 5.339955284913385e-05,
      "loss": 0.9531,
      "mean_token_accuracy": 0.7577284872531891,
      "num_tokens": 44252272.0,
      "step": 1357
    },
    {
      "entropy": 0.9771641194820404,
      "epoch": 0.7262032085561497,
      "grad_norm": 0.18690308928489685,
      "learning_rate": 5.3278385569433695e-05,
      "loss": 0.9681,
      "mean_token_accuracy": 0.7535831779241562,
      "num_tokens": 44285018.0,
      "step": 1358
    },
    {
      "entropy": 0.9384803473949432,
      "epoch": 0.7267379679144385,
      "grad_norm": 0.16335232555866241,
      "learning_rate": 5.315738860462103e-05,
      "loss": 0.9327,
      "mean_token_accuracy": 0.7607322484254837,
      "num_tokens": 44317570.0,
      "step": 1359
    },
    {
      "entropy": 0.9756060987710953,
      "epoch": 0.7272727272727273,
      "grad_norm": 0.17174574732780457,
      "learning_rate": 5.303656231800675e-05,
      "loss": 0.9725,
      "mean_token_accuracy": 0.7532685697078705,
      "num_tokens": 44350338.0,
      "step": 1360
    },
    {
      "entropy": 0.9902632385492325,
      "epoch": 0.727807486631016,
      "grad_norm": 0.1803690642118454,
      "learning_rate": 5.2915907072389406e-05,
      "loss": 0.986,
      "mean_token_accuracy": 0.7520039975643158,
      "num_tokens": 44382125.0,
      "step": 1361
    },
    {
      "entropy": 0.9341016560792923,
      "epoch": 0.7283422459893049,
      "grad_norm": 0.1689465492963791,
      "learning_rate": 5.279542323005387e-05,
      "loss": 0.9396,
      "mean_token_accuracy": 0.7578506767749786,
      "num_tokens": 44414893.0,
      "step": 1362
    },
    {
      "entropy": 0.9944004416465759,
      "epoch": 0.7288770053475936,
      "grad_norm": 0.16927452385425568,
      "learning_rate": 5.2675111152770385e-05,
      "loss": 0.9936,
      "mean_token_accuracy": 0.7520087957382202,
      "num_tokens": 44447437.0,
      "step": 1363
    },
    {
      "entropy": 0.9670007675886154,
      "epoch": 0.7294117647058823,
      "grad_norm": 0.16915826499462128,
      "learning_rate": 5.2554971201793424e-05,
      "loss": 0.96,
      "mean_token_accuracy": 0.756583958864212,
      "num_tokens": 44480034.0,
      "step": 1364
    },
    {
      "entropy": 0.9163489788770676,
      "epoch": 0.7299465240641712,
      "grad_norm": 0.16509558260440826,
      "learning_rate": 5.24350037378607e-05,
      "loss": 0.9029,
      "mean_token_accuracy": 0.7656762450933456,
      "num_tokens": 44512732.0,
      "step": 1365
    },
    {
      "entropy": 0.9414806365966797,
      "epoch": 0.7304812834224599,
      "grad_norm": 0.16583573818206787,
      "learning_rate": 5.2315209121191924e-05,
      "loss": 0.9358,
      "mean_token_accuracy": 0.7630437314510345,
      "num_tokens": 44545500.0,
      "step": 1366
    },
    {
      "entropy": 0.9694929867982864,
      "epoch": 0.7310160427807486,
      "grad_norm": 0.16455243527889252,
      "learning_rate": 5.2195587711487784e-05,
      "loss": 0.9687,
      "mean_token_accuracy": 0.7541544437408447,
      "num_tokens": 44578268.0,
      "step": 1367
    },
    {
      "entropy": 0.9365357607603073,
      "epoch": 0.7315508021390374,
      "grad_norm": 0.163929283618927,
      "learning_rate": 5.207613986792899e-05,
      "loss": 0.9279,
      "mean_token_accuracy": 0.7610830962657928,
      "num_tokens": 44610993.0,
      "step": 1368
    },
    {
      "entropy": 0.9800110459327698,
      "epoch": 0.7320855614973262,
      "grad_norm": 0.16541478037834167,
      "learning_rate": 5.195686594917504e-05,
      "loss": 0.9809,
      "mean_token_accuracy": 0.7503935098648071,
      "num_tokens": 44643141.0,
      "step": 1369
    },
    {
      "entropy": 1.0133154392242432,
      "epoch": 0.732620320855615,
      "grad_norm": 0.17673271894454956,
      "learning_rate": 5.183776631336319e-05,
      "loss": 1.0329,
      "mean_token_accuracy": 0.7408357262611389,
      "num_tokens": 44675581.0,
      "step": 1370
    },
    {
      "entropy": 0.9362034201622009,
      "epoch": 0.7331550802139037,
      "grad_norm": 0.16646535694599152,
      "learning_rate": 5.171884131810739e-05,
      "loss": 0.937,
      "mean_token_accuracy": 0.7587976455688477,
      "num_tokens": 44708349.0,
      "step": 1371
    },
    {
      "entropy": 0.9442934691905975,
      "epoch": 0.7336898395721925,
      "grad_norm": 0.17669664323329926,
      "learning_rate": 5.160009132049728e-05,
      "loss": 0.9532,
      "mean_token_accuracy": 0.7564622759819031,
      "num_tokens": 44740948.0,
      "step": 1372
    },
    {
      "entropy": 0.9387456774711609,
      "epoch": 0.7342245989304813,
      "grad_norm": 0.19216324388980865,
      "learning_rate": 5.148151667709693e-05,
      "loss": 0.9448,
      "mean_token_accuracy": 0.7606610357761383,
      "num_tokens": 44773716.0,
      "step": 1373
    },
    {
      "entropy": 0.9824355393648148,
      "epoch": 0.73475935828877,
      "grad_norm": 0.16717301309108734,
      "learning_rate": 5.136311774394389e-05,
      "loss": 0.9823,
      "mean_token_accuracy": 0.7504276633262634,
      "num_tokens": 44806484.0,
      "step": 1374
    },
    {
      "entropy": 0.9298852831125259,
      "epoch": 0.7352941176470589,
      "grad_norm": 0.1649925261735916,
      "learning_rate": 5.12448948765482e-05,
      "loss": 0.9268,
      "mean_token_accuracy": 0.7589503824710846,
      "num_tokens": 44839252.0,
      "step": 1375
    },
    {
      "entropy": 0.8915909230709076,
      "epoch": 0.7358288770053476,
      "grad_norm": 0.1836596429347992,
      "learning_rate": 5.1126848429891214e-05,
      "loss": 0.8939,
      "mean_token_accuracy": 0.7735643535852432,
      "num_tokens": 44871729.0,
      "step": 1376
    },
    {
      "entropy": 0.9786400496959686,
      "epoch": 0.7363636363636363,
      "grad_norm": 0.18812385201454163,
      "learning_rate": 5.1008978758424545e-05,
      "loss": 0.9731,
      "mean_token_accuracy": 0.7497861683368683,
      "num_tokens": 44904497.0,
      "step": 1377
    },
    {
      "entropy": 0.8944540321826935,
      "epoch": 0.7368983957219252,
      "grad_norm": 0.1752358078956604,
      "learning_rate": 5.089128621606892e-05,
      "loss": 0.8838,
      "mean_token_accuracy": 0.7728800028562546,
      "num_tokens": 44937265.0,
      "step": 1378
    },
    {
      "entropy": 0.9364291578531265,
      "epoch": 0.7374331550802139,
      "grad_norm": 0.22087101638317108,
      "learning_rate": 5.077377115621341e-05,
      "loss": 0.9376,
      "mean_token_accuracy": 0.7617378532886505,
      "num_tokens": 44969888.0,
      "step": 1379
    },
    {
      "entropy": 0.9458759129047394,
      "epoch": 0.7379679144385026,
      "grad_norm": 0.1907070279121399,
      "learning_rate": 5.065643393171398e-05,
      "loss": 0.9295,
      "mean_token_accuracy": 0.7640746533870697,
      "num_tokens": 45002567.0,
      "step": 1380
    },
    {
      "entropy": 0.9374886602163315,
      "epoch": 0.7385026737967915,
      "grad_norm": 0.1655336618423462,
      "learning_rate": 5.0539274894892695e-05,
      "loss": 0.9246,
      "mean_token_accuracy": 0.7612414360046387,
      "num_tokens": 45035335.0,
      "step": 1381
    },
    {
      "entropy": 0.9554038047790527,
      "epoch": 0.7390374331550802,
      "grad_norm": 0.1619529128074646,
      "learning_rate": 5.042229439753659e-05,
      "loss": 0.951,
      "mean_token_accuracy": 0.7614552676677704,
      "num_tokens": 45068103.0,
      "step": 1382
    },
    {
      "entropy": 0.9579029679298401,
      "epoch": 0.739572192513369,
      "grad_norm": 0.18599477410316467,
      "learning_rate": 5.0305492790896604e-05,
      "loss": 0.9598,
      "mean_token_accuracy": 0.7529549300670624,
      "num_tokens": 45100450.0,
      "step": 1383
    },
    {
      "entropy": 0.9384265094995499,
      "epoch": 0.7401069518716578,
      "grad_norm": 0.17478297650814056,
      "learning_rate": 5.018887042568656e-05,
      "loss": 0.9367,
      "mean_token_accuracy": 0.7592450827360153,
      "num_tokens": 45132859.0,
      "step": 1384
    },
    {
      "entropy": 0.9385294169187546,
      "epoch": 0.7406417112299465,
      "grad_norm": 0.16543041169643402,
      "learning_rate": 5.0072427652082e-05,
      "loss": 0.9495,
      "mean_token_accuracy": 0.7601417303085327,
      "num_tokens": 45165627.0,
      "step": 1385
    },
    {
      "entropy": 0.9611698240041733,
      "epoch": 0.7411764705882353,
      "grad_norm": 0.17224179208278656,
      "learning_rate": 4.995616481971934e-05,
      "loss": 0.9802,
      "mean_token_accuracy": 0.7500352561473846,
      "num_tokens": 45198288.0,
      "step": 1386
    },
    {
      "entropy": 0.9673365503549576,
      "epoch": 0.741711229946524,
      "grad_norm": 0.17497476935386658,
      "learning_rate": 4.984008227769457e-05,
      "loss": 0.9681,
      "mean_token_accuracy": 0.7542460858821869,
      "num_tokens": 45231056.0,
      "step": 1387
    },
    {
      "entropy": 0.9386740624904633,
      "epoch": 0.7422459893048128,
      "grad_norm": 0.18572238087654114,
      "learning_rate": 4.972418037456236e-05,
      "loss": 0.9367,
      "mean_token_accuracy": 0.7579202800989151,
      "num_tokens": 45263505.0,
      "step": 1388
    },
    {
      "entropy": 0.9856511950492859,
      "epoch": 0.7427807486631016,
      "grad_norm": 0.1692817360162735,
      "learning_rate": 4.960845945833504e-05,
      "loss": 0.9807,
      "mean_token_accuracy": 0.7517717480659485,
      "num_tokens": 45296273.0,
      "step": 1389
    },
    {
      "entropy": 0.9516720473766327,
      "epoch": 0.7433155080213903,
      "grad_norm": 0.16946376860141754,
      "learning_rate": 4.9492919876481485e-05,
      "loss": 0.9615,
      "mean_token_accuracy": 0.7579386234283447,
      "num_tokens": 45328969.0,
      "step": 1390
    },
    {
      "entropy": 1.0015557408332825,
      "epoch": 0.7438502673796792,
      "grad_norm": 0.165788471698761,
      "learning_rate": 4.937756197592609e-05,
      "loss": 1.0071,
      "mean_token_accuracy": 0.7453262507915497,
      "num_tokens": 45361737.0,
      "step": 1391
    },
    {
      "entropy": 0.9206484109163284,
      "epoch": 0.7443850267379679,
      "grad_norm": 0.17079102993011475,
      "learning_rate": 4.9262386103047677e-05,
      "loss": 0.9099,
      "mean_token_accuracy": 0.7662206590175629,
      "num_tokens": 45394505.0,
      "step": 1392
    },
    {
      "entropy": 0.9758808314800262,
      "epoch": 0.7449197860962566,
      "grad_norm": 0.1746615320444107,
      "learning_rate": 4.914739260367856e-05,
      "loss": 0.9856,
      "mean_token_accuracy": 0.7506234645843506,
      "num_tokens": 45427144.0,
      "step": 1393
    },
    {
      "entropy": 0.9476794600486755,
      "epoch": 0.7454545454545455,
      "grad_norm": 0.1681424230337143,
      "learning_rate": 4.903258182310338e-05,
      "loss": 0.9544,
      "mean_token_accuracy": 0.7548570334911346,
      "num_tokens": 45459912.0,
      "step": 1394
    },
    {
      "entropy": 1.0142509192228317,
      "epoch": 0.7459893048128342,
      "grad_norm": 0.1720559000968933,
      "learning_rate": 4.891795410605826e-05,
      "loss": 1.006,
      "mean_token_accuracy": 0.74620620906353,
      "num_tokens": 45492002.0,
      "step": 1395
    },
    {
      "entropy": 0.9728702455759048,
      "epoch": 0.746524064171123,
      "grad_norm": 0.17016872763633728,
      "learning_rate": 4.880350979672949e-05,
      "loss": 0.9642,
      "mean_token_accuracy": 0.751099705696106,
      "num_tokens": 45524770.0,
      "step": 1396
    },
    {
      "entropy": 0.9072166979312897,
      "epoch": 0.7470588235294118,
      "grad_norm": 0.16708402335643768,
      "learning_rate": 4.8689249238752786e-05,
      "loss": 0.9021,
      "mean_token_accuracy": 0.7652431428432465,
      "num_tokens": 45557538.0,
      "step": 1397
    },
    {
      "entropy": 0.9732768088579178,
      "epoch": 0.7475935828877005,
      "grad_norm": 0.16878630220890045,
      "learning_rate": 4.857517277521208e-05,
      "loss": 0.9693,
      "mean_token_accuracy": 0.7514357268810272,
      "num_tokens": 45590306.0,
      "step": 1398
    },
    {
      "entropy": 0.9616824388504028,
      "epoch": 0.7481283422459893,
      "grad_norm": 0.16568832099437714,
      "learning_rate": 4.8461280748638485e-05,
      "loss": 0.9579,
      "mean_token_accuracy": 0.7586461007595062,
      "num_tokens": 45622625.0,
      "step": 1399
    },
    {
      "entropy": 0.9650434404611588,
      "epoch": 0.7486631016042781,
      "grad_norm": 0.18817420303821564,
      "learning_rate": 4.83475735010094e-05,
      "loss": 0.9589,
      "mean_token_accuracy": 0.7563316226005554,
      "num_tokens": 45655250.0,
      "step": 1400
    },
    {
      "entropy": 0.9968360215425491,
      "epoch": 0.7491978609625668,
      "grad_norm": 0.1672886610031128,
      "learning_rate": 4.823405137374731e-05,
      "loss": 0.9932,
      "mean_token_accuracy": 0.7512084990739822,
      "num_tokens": 45687908.0,
      "step": 1401
    },
    {
      "entropy": 0.9780010432004929,
      "epoch": 0.7497326203208556,
      "grad_norm": 0.16161589324474335,
      "learning_rate": 4.812071470771895e-05,
      "loss": 0.9612,
      "mean_token_accuracy": 0.7587670981884003,
      "num_tokens": 45720676.0,
      "step": 1402
    },
    {
      "entropy": 0.9526931345462799,
      "epoch": 0.7502673796791444,
      "grad_norm": 0.16700731217861176,
      "learning_rate": 4.8007563843234055e-05,
      "loss": 0.9336,
      "mean_token_accuracy": 0.7571175396442413,
      "num_tokens": 45753444.0,
      "step": 1403
    },
    {
      "entropy": 0.9786204248666763,
      "epoch": 0.7508021390374332,
      "grad_norm": 0.16651296615600586,
      "learning_rate": 4.789459912004456e-05,
      "loss": 0.967,
      "mean_token_accuracy": 0.7521077692508698,
      "num_tokens": 45786212.0,
      "step": 1404
    },
    {
      "entropy": 0.9743318110704422,
      "epoch": 0.7513368983957219,
      "grad_norm": 0.17126482725143433,
      "learning_rate": 4.7781820877343484e-05,
      "loss": 0.9664,
      "mean_token_accuracy": 0.7522910535335541,
      "num_tokens": 45818980.0,
      "step": 1405
    },
    {
      "entropy": 1.0189561545848846,
      "epoch": 0.7518716577540107,
      "grad_norm": 0.17651601135730743,
      "learning_rate": 4.766922945376389e-05,
      "loss": 1.0309,
      "mean_token_accuracy": 0.7422950267791748,
      "num_tokens": 45851006.0,
      "step": 1406
    },
    {
      "entropy": 0.9401010423898697,
      "epoch": 0.7524064171122995,
      "grad_norm": 0.16605083644390106,
      "learning_rate": 4.755682518737784e-05,
      "loss": 0.9318,
      "mean_token_accuracy": 0.7648623734712601,
      "num_tokens": 45883635.0,
      "step": 1407
    },
    {
      "entropy": 0.966943696141243,
      "epoch": 0.7529411764705882,
      "grad_norm": 0.1740500032901764,
      "learning_rate": 4.744460841569551e-05,
      "loss": 0.9731,
      "mean_token_accuracy": 0.7538737058639526,
      "num_tokens": 45915984.0,
      "step": 1408
    },
    {
      "entropy": 0.9927944242954254,
      "epoch": 0.753475935828877,
      "grad_norm": 0.1688157021999359,
      "learning_rate": 4.733257947566405e-05,
      "loss": 0.9948,
      "mean_token_accuracy": 0.7516429573297501,
      "num_tokens": 45948312.0,
      "step": 1409
    },
    {
      "entropy": 0.945795476436615,
      "epoch": 0.7540106951871658,
      "grad_norm": 0.17462849617004395,
      "learning_rate": 4.7220738703666594e-05,
      "loss": 0.9536,
      "mean_token_accuracy": 0.7565722614526749,
      "num_tokens": 45980996.0,
      "step": 1410
    },
    {
      "entropy": 0.9234698563814163,
      "epoch": 0.7545454545454545,
      "grad_norm": 0.17292383313179016,
      "learning_rate": 4.710908643552133e-05,
      "loss": 0.9326,
      "mean_token_accuracy": 0.7631048262119293,
      "num_tokens": 46013764.0,
      "step": 1411
    },
    {
      "entropy": 0.9526192247867584,
      "epoch": 0.7550802139037434,
      "grad_norm": 0.1689281016588211,
      "learning_rate": 4.699762300648042e-05,
      "loss": 0.9623,
      "mean_token_accuracy": 0.754096120595932,
      "num_tokens": 46046127.0,
      "step": 1412
    },
    {
      "entropy": 0.9535632431507111,
      "epoch": 0.7556149732620321,
      "grad_norm": 0.17146529257297516,
      "learning_rate": 4.6886348751229025e-05,
      "loss": 0.9543,
      "mean_token_accuracy": 0.7565059959888458,
      "num_tokens": 46078580.0,
      "step": 1413
    },
    {
      "entropy": 0.9169730544090271,
      "epoch": 0.7561497326203208,
      "grad_norm": 0.16838543117046356,
      "learning_rate": 4.677526400388421e-05,
      "loss": 0.9041,
      "mean_token_accuracy": 0.7646932899951935,
      "num_tokens": 46111348.0,
      "step": 1414
    },
    {
      "entropy": 0.9166508615016937,
      "epoch": 0.7566844919786097,
      "grad_norm": 0.17483341693878174,
      "learning_rate": 4.6664369097994124e-05,
      "loss": 0.921,
      "mean_token_accuracy": 0.7638685256242752,
      "num_tokens": 46144116.0,
      "step": 1415
    },
    {
      "entropy": 0.9559438526630402,
      "epoch": 0.7572192513368984,
      "grad_norm": 0.1699785441160202,
      "learning_rate": 4.6553664366536794e-05,
      "loss": 0.9508,
      "mean_token_accuracy": 0.7549486756324768,
      "num_tokens": 46176884.0,
      "step": 1416
    },
    {
      "entropy": 0.9044709354639053,
      "epoch": 0.7577540106951872,
      "grad_norm": 0.16961148381233215,
      "learning_rate": 4.6443150141919245e-05,
      "loss": 0.9067,
      "mean_token_accuracy": 0.767368733882904,
      "num_tokens": 46209291.0,
      "step": 1417
    },
    {
      "entropy": 0.918102815747261,
      "epoch": 0.758288770053476,
      "grad_norm": 0.16606849431991577,
      "learning_rate": 4.633282675597651e-05,
      "loss": 0.9129,
      "mean_token_accuracy": 0.7693549543619156,
      "num_tokens": 46241998.0,
      "step": 1418
    },
    {
      "entropy": 0.9878501445055008,
      "epoch": 0.7588235294117647,
      "grad_norm": 0.18879282474517822,
      "learning_rate": 4.622269453997057e-05,
      "loss": 0.9881,
      "mean_token_accuracy": 0.7488882839679718,
      "num_tokens": 46274726.0,
      "step": 1419
    },
    {
      "entropy": 0.946536123752594,
      "epoch": 0.7593582887700535,
      "grad_norm": 0.17403647303581238,
      "learning_rate": 4.611275382458947e-05,
      "loss": 0.9443,
      "mean_token_accuracy": 0.7589809447526932,
      "num_tokens": 46307494.0,
      "step": 1420
    },
    {
      "entropy": 0.9222494065761566,
      "epoch": 0.7598930481283422,
      "grad_norm": 0.179395392537117,
      "learning_rate": 4.6003004939946076e-05,
      "loss": 0.9275,
      "mean_token_accuracy": 0.7654569745063782,
      "num_tokens": 46340262.0,
      "step": 1421
    },
    {
      "entropy": 0.8983130007982254,
      "epoch": 0.760427807486631,
      "grad_norm": 0.18136689066886902,
      "learning_rate": 4.589344821557744e-05,
      "loss": 0.9022,
      "mean_token_accuracy": 0.7665067464113235,
      "num_tokens": 46372697.0,
      "step": 1422
    },
    {
      "entropy": 0.9430101960897446,
      "epoch": 0.7609625668449198,
      "grad_norm": 0.17819227278232574,
      "learning_rate": 4.578408398044349e-05,
      "loss": 0.9438,
      "mean_token_accuracy": 0.7585532665252686,
      "num_tokens": 46405465.0,
      "step": 1423
    },
    {
      "entropy": 0.9201874732971191,
      "epoch": 0.7614973262032085,
      "grad_norm": 0.17150764167308807,
      "learning_rate": 4.5674912562926284e-05,
      "loss": 0.9198,
      "mean_token_accuracy": 0.7630615085363388,
      "num_tokens": 46437886.0,
      "step": 1424
    },
    {
      "entropy": 0.9516346603631973,
      "epoch": 0.7620320855614974,
      "grad_norm": 0.19696636497974396,
      "learning_rate": 4.556593429082879e-05,
      "loss": 0.9477,
      "mean_token_accuracy": 0.7558687180280685,
      "num_tokens": 46470408.0,
      "step": 1425
    },
    {
      "entropy": 0.9484929591417313,
      "epoch": 0.7625668449197861,
      "grad_norm": 0.18077801167964935,
      "learning_rate": 4.545714949137414e-05,
      "loss": 0.9408,
      "mean_token_accuracy": 0.7573008239269257,
      "num_tokens": 46503176.0,
      "step": 1426
    },
    {
      "entropy": 0.9580115526914597,
      "epoch": 0.7631016042780748,
      "grad_norm": 0.1708674430847168,
      "learning_rate": 4.534855849120453e-05,
      "loss": 0.9537,
      "mean_token_accuracy": 0.7538977414369583,
      "num_tokens": 46535857.0,
      "step": 1427
    },
    {
      "entropy": 0.8961019665002823,
      "epoch": 0.7636363636363637,
      "grad_norm": 0.17240260541439056,
      "learning_rate": 4.524016161638016e-05,
      "loss": 0.9055,
      "mean_token_accuracy": 0.7688349038362503,
      "num_tokens": 46568295.0,
      "step": 1428
    },
    {
      "entropy": 0.9304610937833786,
      "epoch": 0.7641711229946524,
      "grad_norm": 0.1705794334411621,
      "learning_rate": 4.513195919237843e-05,
      "loss": 0.9144,
      "mean_token_accuracy": 0.7627077251672745,
      "num_tokens": 46601063.0,
      "step": 1429
    },
    {
      "entropy": 0.9417256414890289,
      "epoch": 0.7647058823529411,
      "grad_norm": 0.1612071841955185,
      "learning_rate": 4.502395154409281e-05,
      "loss": 0.9327,
      "mean_token_accuracy": 0.7602944672107697,
      "num_tokens": 46633831.0,
      "step": 1430
    },
    {
      "entropy": 0.9605920910835266,
      "epoch": 0.76524064171123,
      "grad_norm": 0.16719336807727814,
      "learning_rate": 4.491613899583198e-05,
      "loss": 0.9534,
      "mean_token_accuracy": 0.7547348439693451,
      "num_tokens": 46666599.0,
      "step": 1431
    },
    {
      "entropy": 0.9498101323843002,
      "epoch": 0.7657754010695187,
      "grad_norm": 0.16406437754631042,
      "learning_rate": 4.4808521871318744e-05,
      "loss": 0.9423,
      "mean_token_accuracy": 0.755987286567688,
      "num_tokens": 46699367.0,
      "step": 1432
    },
    {
      "entropy": 0.9573682099580765,
      "epoch": 0.7663101604278075,
      "grad_norm": 0.17057931423187256,
      "learning_rate": 4.470110049368919e-05,
      "loss": 0.967,
      "mean_token_accuracy": 0.7498778104782104,
      "num_tokens": 46732135.0,
      "step": 1433
    },
    {
      "entropy": 0.9842314422130585,
      "epoch": 0.7668449197860963,
      "grad_norm": 0.1765008121728897,
      "learning_rate": 4.4593875185491574e-05,
      "loss": 0.9826,
      "mean_token_accuracy": 0.7512218952178955,
      "num_tokens": 46764903.0,
      "step": 1434
    },
    {
      "entropy": 0.9449929594993591,
      "epoch": 0.767379679144385,
      "grad_norm": 0.16481998562812805,
      "learning_rate": 4.4486846268685513e-05,
      "loss": 0.928,
      "mean_token_accuracy": 0.76295405626297,
      "num_tokens": 46797237.0,
      "step": 1435
    },
    {
      "entropy": 0.9285380840301514,
      "epoch": 0.7679144385026738,
      "grad_norm": 0.1714523881673813,
      "learning_rate": 4.438001406464084e-05,
      "loss": 0.9325,
      "mean_token_accuracy": 0.7573924660682678,
      "num_tokens": 46830005.0,
      "step": 1436
    },
    {
      "entropy": 0.9558539241552353,
      "epoch": 0.7684491978609626,
      "grad_norm": 0.17213749885559082,
      "learning_rate": 4.427337889413675e-05,
      "loss": 0.9676,
      "mean_token_accuracy": 0.7559363394975662,
      "num_tokens": 46862617.0,
      "step": 1437
    },
    {
      "entropy": 0.9706854373216629,
      "epoch": 0.7689839572192514,
      "grad_norm": 0.17222625017166138,
      "learning_rate": 4.41669410773609e-05,
      "loss": 0.973,
      "mean_token_accuracy": 0.7512218952178955,
      "num_tokens": 46895385.0,
      "step": 1438
    },
    {
      "entropy": 0.961706668138504,
      "epoch": 0.7695187165775401,
      "grad_norm": 0.17626358568668365,
      "learning_rate": 4.406070093390821e-05,
      "loss": 0.9669,
      "mean_token_accuracy": 0.756006732583046,
      "num_tokens": 46928143.0,
      "step": 1439
    },
    {
      "entropy": 0.9379317164421082,
      "epoch": 0.7700534759358288,
      "grad_norm": 0.17700813710689545,
      "learning_rate": 4.395465878278019e-05,
      "loss": 0.9395,
      "mean_token_accuracy": 0.7630215734243393,
      "num_tokens": 46960844.0,
      "step": 1440
    },
    {
      "entropy": 0.9566432982683182,
      "epoch": 0.7705882352941177,
      "grad_norm": 0.1715305745601654,
      "learning_rate": 4.384881494238381e-05,
      "loss": 0.9554,
      "mean_token_accuracy": 0.7504063248634338,
      "num_tokens": 46993497.0,
      "step": 1441
    },
    {
      "entropy": 0.9218139499425888,
      "epoch": 0.7711229946524064,
      "grad_norm": 0.1674298495054245,
      "learning_rate": 4.37431697305306e-05,
      "loss": 0.9266,
      "mean_token_accuracy": 0.76496821641922,
      "num_tokens": 47026265.0,
      "step": 1442
    },
    {
      "entropy": 0.9717037230730057,
      "epoch": 0.7716577540106951,
      "grad_norm": 0.1699391007423401,
      "learning_rate": 4.363772346443561e-05,
      "loss": 0.9604,
      "mean_token_accuracy": 0.7572091817855835,
      "num_tokens": 47059033.0,
      "step": 1443
    },
    {
      "entropy": 0.9532619267702103,
      "epoch": 0.772192513368984,
      "grad_norm": 0.17742779850959778,
      "learning_rate": 4.353247646071657e-05,
      "loss": 0.9679,
      "mean_token_accuracy": 0.753029465675354,
      "num_tokens": 47091785.0,
      "step": 1444
    },
    {
      "entropy": 0.9997300058603287,
      "epoch": 0.7727272727272727,
      "grad_norm": 0.17780952155590057,
      "learning_rate": 4.342742903539294e-05,
      "loss": 1.01,
      "mean_token_accuracy": 0.7437186241149902,
      "num_tokens": 47124363.0,
      "step": 1445
    },
    {
      "entropy": 0.9187695980072021,
      "epoch": 0.7732620320855615,
      "grad_norm": 0.16247506439685822,
      "learning_rate": 4.332258150388484e-05,
      "loss": 0.91,
      "mean_token_accuracy": 0.7681971341371536,
      "num_tokens": 47156893.0,
      "step": 1446
    },
    {
      "entropy": 0.9684152752161026,
      "epoch": 0.7737967914438503,
      "grad_norm": 0.1999502331018448,
      "learning_rate": 4.321793418101221e-05,
      "loss": 0.96,
      "mean_token_accuracy": 0.7576979398727417,
      "num_tokens": 47189661.0,
      "step": 1447
    },
    {
      "entropy": 0.9522790312767029,
      "epoch": 0.774331550802139,
      "grad_norm": 0.16581189632415771,
      "learning_rate": 4.3113487380993865e-05,
      "loss": 0.9267,
      "mean_token_accuracy": 0.7602584809064865,
      "num_tokens": 47222039.0,
      "step": 1448
    },
    {
      "entropy": 0.9538277089595795,
      "epoch": 0.7748663101604278,
      "grad_norm": 0.1805843710899353,
      "learning_rate": 4.3009241417446514e-05,
      "loss": 0.953,
      "mean_token_accuracy": 0.7612296044826508,
      "num_tokens": 47254252.0,
      "step": 1449
    },
    {
      "entropy": 0.9197836816310883,
      "epoch": 0.7754010695187166,
      "grad_norm": 0.19354727864265442,
      "learning_rate": 4.290519660338377e-05,
      "loss": 0.9113,
      "mean_token_accuracy": 0.7670795321464539,
      "num_tokens": 47286900.0,
      "step": 1450
    },
    {
      "entropy": 0.9822978228330612,
      "epoch": 0.7759358288770053,
      "grad_norm": 0.1752813756465912,
      "learning_rate": 4.280135325121529e-05,
      "loss": 0.9764,
      "mean_token_accuracy": 0.7531676441431046,
      "num_tokens": 47319401.0,
      "step": 1451
    },
    {
      "entropy": 0.984327420592308,
      "epoch": 0.7764705882352941,
      "grad_norm": 0.1943073570728302,
      "learning_rate": 4.269771167274585e-05,
      "loss": 0.9789,
      "mean_token_accuracy": 0.7508493363857269,
      "num_tokens": 47352055.0,
      "step": 1452
    },
    {
      "entropy": 0.960977628827095,
      "epoch": 0.7770053475935829,
      "grad_norm": 0.18327106535434723,
      "learning_rate": 4.2594272179174367e-05,
      "loss": 0.9565,
      "mean_token_accuracy": 0.7579166293144226,
      "num_tokens": 47384554.0,
      "step": 1453
    },
    {
      "entropy": 0.9817952513694763,
      "epoch": 0.7775401069518717,
      "grad_norm": 0.17658685147762299,
      "learning_rate": 4.2491035081092894e-05,
      "loss": 0.9837,
      "mean_token_accuracy": 0.7525317072868347,
      "num_tokens": 47417246.0,
      "step": 1454
    },
    {
      "entropy": 0.9385758340358734,
      "epoch": 0.7780748663101604,
      "grad_norm": 0.18255873024463654,
      "learning_rate": 4.238800068848585e-05,
      "loss": 0.9523,
      "mean_token_accuracy": 0.7575452029705048,
      "num_tokens": 47450014.0,
      "step": 1455
    },
    {
      "entropy": 0.9425090700387955,
      "epoch": 0.7786096256684492,
      "grad_norm": 0.1812393069267273,
      "learning_rate": 4.2285169310728976e-05,
      "loss": 0.9358,
      "mean_token_accuracy": 0.7598954886198044,
      "num_tokens": 47482235.0,
      "step": 1456
    },
    {
      "entropy": 0.9353173077106476,
      "epoch": 0.779144385026738,
      "grad_norm": 0.16947653889656067,
      "learning_rate": 4.2182541256588415e-05,
      "loss": 0.923,
      "mean_token_accuracy": 0.7609665244817734,
      "num_tokens": 47515003.0,
      "step": 1457
    },
    {
      "entropy": 0.9787223637104034,
      "epoch": 0.7796791443850267,
      "grad_norm": 0.16849343478679657,
      "learning_rate": 4.2080116834219766e-05,
      "loss": 0.9569,
      "mean_token_accuracy": 0.756384402513504,
      "num_tokens": 47547771.0,
      "step": 1458
    },
    {
      "entropy": 0.9411111623048782,
      "epoch": 0.7802139037433156,
      "grad_norm": 0.17539463937282562,
      "learning_rate": 4.197789635116728e-05,
      "loss": 0.9509,
      "mean_token_accuracy": 0.7613932490348816,
      "num_tokens": 47580153.0,
      "step": 1459
    },
    {
      "entropy": 0.9682489633560181,
      "epoch": 0.7807486631016043,
      "grad_norm": 0.2003321647644043,
      "learning_rate": 4.1875880114362826e-05,
      "loss": 0.9712,
      "mean_token_accuracy": 0.753556489944458,
      "num_tokens": 47612902.0,
      "step": 1460
    },
    {
      "entropy": 0.9437247663736343,
      "epoch": 0.781283422459893,
      "grad_norm": 0.17065325379371643,
      "learning_rate": 4.1774068430124926e-05,
      "loss": 0.9458,
      "mean_token_accuracy": 0.7608443200588226,
      "num_tokens": 47645670.0,
      "step": 1461
    },
    {
      "entropy": 0.9649447053670883,
      "epoch": 0.7818181818181819,
      "grad_norm": 0.1729692965745926,
      "learning_rate": 4.1672461604157964e-05,
      "loss": 0.9773,
      "mean_token_accuracy": 0.7494195997714996,
      "num_tokens": 47678438.0,
      "step": 1462
    },
    {
      "entropy": 0.9338851720094681,
      "epoch": 0.7823529411764706,
      "grad_norm": 0.169850692152977,
      "learning_rate": 4.1571059941551206e-05,
      "loss": 0.9375,
      "mean_token_accuracy": 0.7621547877788544,
      "num_tokens": 47710902.0,
      "step": 1463
    },
    {
      "entropy": 0.9382988661527634,
      "epoch": 0.7828877005347593,
      "grad_norm": 0.17178453505039215,
      "learning_rate": 4.1469863746777904e-05,
      "loss": 0.9307,
      "mean_token_accuracy": 0.7574306279420853,
      "num_tokens": 47743518.0,
      "step": 1464
    },
    {
      "entropy": 0.8995357155799866,
      "epoch": 0.7834224598930482,
      "grad_norm": 0.17019139230251312,
      "learning_rate": 4.13688733236943e-05,
      "loss": 0.8944,
      "mean_token_accuracy": 0.7671439498662949,
      "num_tokens": 47775655.0,
      "step": 1465
    },
    {
      "entropy": 0.9421730637550354,
      "epoch": 0.7839572192513369,
      "grad_norm": 0.1680029332637787,
      "learning_rate": 4.12680889755388e-05,
      "loss": 0.9431,
      "mean_token_accuracy": 0.7600500881671906,
      "num_tokens": 47808423.0,
      "step": 1466
    },
    {
      "entropy": 0.9358892440795898,
      "epoch": 0.7844919786096257,
      "grad_norm": 0.17465731501579285,
      "learning_rate": 4.116751100493108e-05,
      "loss": 0.9384,
      "mean_token_accuracy": 0.7588892877101898,
      "num_tokens": 47841191.0,
      "step": 1467
    },
    {
      "entropy": 0.9814324527978897,
      "epoch": 0.7850267379679144,
      "grad_norm": 0.17830786108970642,
      "learning_rate": 4.1067139713871086e-05,
      "loss": 0.9945,
      "mean_token_accuracy": 0.7513263672590256,
      "num_tokens": 47873789.0,
      "step": 1468
    },
    {
      "entropy": 1.0155832320451736,
      "epoch": 0.7855614973262032,
      "grad_norm": 0.17654651403427124,
      "learning_rate": 4.0966975403738194e-05,
      "loss": 1.0217,
      "mean_token_accuracy": 0.7456433027982712,
      "num_tokens": 47906248.0,
      "step": 1469
    },
    {
      "entropy": 0.9782692193984985,
      "epoch": 0.786096256684492,
      "grad_norm": 0.17136093974113464,
      "learning_rate": 4.086701837529032e-05,
      "loss": 0.9816,
      "mean_token_accuracy": 0.7524743378162384,
      "num_tokens": 47939016.0,
      "step": 1470
    },
    {
      "entropy": 0.943878784775734,
      "epoch": 0.7866310160427807,
      "grad_norm": 0.20054322481155396,
      "learning_rate": 4.0767268928662974e-05,
      "loss": 0.9353,
      "mean_token_accuracy": 0.7583309412002563,
      "num_tokens": 47971345.0,
      "step": 1471
    },
    {
      "entropy": 0.9630401432514191,
      "epoch": 0.7871657754010695,
      "grad_norm": 0.17285215854644775,
      "learning_rate": 4.066772736336834e-05,
      "loss": 0.973,
      "mean_token_accuracy": 0.7511119246482849,
      "num_tokens": 48003878.0,
      "step": 1472
    },
    {
      "entropy": 0.9346894472837448,
      "epoch": 0.7877005347593583,
      "grad_norm": 0.17615626752376556,
      "learning_rate": 4.056839397829441e-05,
      "loss": 0.9282,
      "mean_token_accuracy": 0.7593169510364532,
      "num_tokens": 48036646.0,
      "step": 1473
    },
    {
      "entropy": 0.9807346612215042,
      "epoch": 0.788235294117647,
      "grad_norm": 0.17368116974830627,
      "learning_rate": 4.046926907170414e-05,
      "loss": 0.9778,
      "mean_token_accuracy": 0.7508859932422638,
      "num_tokens": 48069049.0,
      "step": 1474
    },
    {
      "entropy": 0.9940079301595688,
      "epoch": 0.7887700534759359,
      "grad_norm": 0.17876891791820526,
      "learning_rate": 4.0370352941234427e-05,
      "loss": 0.9993,
      "mean_token_accuracy": 0.7478200197219849,
      "num_tokens": 48101213.0,
      "step": 1475
    },
    {
      "entropy": 0.9659792631864548,
      "epoch": 0.7893048128342246,
      "grad_norm": 0.16823144257068634,
      "learning_rate": 4.027164588389534e-05,
      "loss": 0.9756,
      "mean_token_accuracy": 0.7521383166313171,
      "num_tokens": 48133981.0,
      "step": 1476
    },
    {
      "entropy": 0.9486778080463409,
      "epoch": 0.7898395721925133,
      "grad_norm": 0.17777416110038757,
      "learning_rate": 4.0173148196069155e-05,
      "loss": 0.931,
      "mean_token_accuracy": 0.7633533477783203,
      "num_tokens": 48166500.0,
      "step": 1477
    },
    {
      "entropy": 0.9380469620227814,
      "epoch": 0.7903743315508022,
      "grad_norm": 0.19114357233047485,
      "learning_rate": 4.007486017350952e-05,
      "loss": 0.9379,
      "mean_token_accuracy": 0.7576062977313995,
      "num_tokens": 48199268.0,
      "step": 1478
    },
    {
      "entropy": 0.9314498156309128,
      "epoch": 0.7909090909090909,
      "grad_norm": 0.16669416427612305,
      "learning_rate": 3.9976782111340466e-05,
      "loss": 0.9224,
      "mean_token_accuracy": 0.7631019949913025,
      "num_tokens": 48231768.0,
      "step": 1479
    },
    {
      "entropy": 0.9650258719921112,
      "epoch": 0.7914438502673797,
      "grad_norm": 0.18811096251010895,
      "learning_rate": 3.987891430405561e-05,
      "loss": 0.9592,
      "mean_token_accuracy": 0.7563233077526093,
      "num_tokens": 48264536.0,
      "step": 1480
    },
    {
      "entropy": 0.9652723371982574,
      "epoch": 0.7919786096256685,
      "grad_norm": 0.18682901561260223,
      "learning_rate": 3.978125704551728e-05,
      "loss": 0.9596,
      "mean_token_accuracy": 0.7536351382732391,
      "num_tokens": 48297304.0,
      "step": 1481
    },
    {
      "entropy": 0.9712996333837509,
      "epoch": 0.7925133689839572,
      "grad_norm": 0.17110350728034973,
      "learning_rate": 3.9683810628955616e-05,
      "loss": 0.9592,
      "mean_token_accuracy": 0.755139946937561,
      "num_tokens": 48329753.0,
      "step": 1482
    },
    {
      "entropy": 0.908642441034317,
      "epoch": 0.793048128342246,
      "grad_norm": 0.17509803175926208,
      "learning_rate": 3.958657534696758e-05,
      "loss": 0.9078,
      "mean_token_accuracy": 0.7689491957426071,
      "num_tokens": 48362216.0,
      "step": 1483
    },
    {
      "entropy": 0.9365021735429764,
      "epoch": 0.7935828877005348,
      "grad_norm": 0.18727156519889832,
      "learning_rate": 3.9489551491516246e-05,
      "loss": 0.9439,
      "mean_token_accuracy": 0.753926008939743,
      "num_tokens": 48394956.0,
      "step": 1484
    },
    {
      "entropy": 0.9669003784656525,
      "epoch": 0.7941176470588235,
      "grad_norm": 0.173688605427742,
      "learning_rate": 3.9392739353929884e-05,
      "loss": 0.9778,
      "mean_token_accuracy": 0.7519550323486328,
      "num_tokens": 48427724.0,
      "step": 1485
    },
    {
      "entropy": 0.9892281740903854,
      "epoch": 0.7946524064171123,
      "grad_norm": 0.17844879627227783,
      "learning_rate": 3.929613922490096e-05,
      "loss": 0.9798,
      "mean_token_accuracy": 0.7524132430553436,
      "num_tokens": 48460492.0,
      "step": 1486
    },
    {
      "entropy": 0.9603982418775558,
      "epoch": 0.795187165775401,
      "grad_norm": 0.17463773488998413,
      "learning_rate": 3.9199751394485385e-05,
      "loss": 0.9452,
      "mean_token_accuracy": 0.7603353261947632,
      "num_tokens": 48492842.0,
      "step": 1487
    },
    {
      "entropy": 0.926127091050148,
      "epoch": 0.7957219251336899,
      "grad_norm": 0.18449467420578003,
      "learning_rate": 3.910357615210163e-05,
      "loss": 0.9307,
      "mean_token_accuracy": 0.7617607414722443,
      "num_tokens": 48525610.0,
      "step": 1488
    },
    {
      "entropy": 0.9539791196584702,
      "epoch": 0.7962566844919786,
      "grad_norm": 0.1840931475162506,
      "learning_rate": 3.900761378652987e-05,
      "loss": 0.9535,
      "mean_token_accuracy": 0.7557183653116226,
      "num_tokens": 48558277.0,
      "step": 1489
    },
    {
      "entropy": 0.9705661684274673,
      "epoch": 0.7967914438502673,
      "grad_norm": 0.18575632572174072,
      "learning_rate": 3.891186458591098e-05,
      "loss": 0.9756,
      "mean_token_accuracy": 0.751893937587738,
      "num_tokens": 48591045.0,
      "step": 1490
    },
    {
      "entropy": 0.9732887297868729,
      "epoch": 0.7973262032085562,
      "grad_norm": 0.18022392690181732,
      "learning_rate": 3.88163288377459e-05,
      "loss": 0.9794,
      "mean_token_accuracy": 0.750747948884964,
      "num_tokens": 48623382.0,
      "step": 1491
    },
    {
      "entropy": 0.9191063940525055,
      "epoch": 0.7978609625668449,
      "grad_norm": 0.16626055538654327,
      "learning_rate": 3.87210068288946e-05,
      "loss": 0.912,
      "mean_token_accuracy": 0.7680219113826752,
      "num_tokens": 48656074.0,
      "step": 1492
    },
    {
      "entropy": 0.9622265249490738,
      "epoch": 0.7983957219251336,
      "grad_norm": 0.17625634372234344,
      "learning_rate": 3.8625898845575224e-05,
      "loss": 0.956,
      "mean_token_accuracy": 0.7564660012722015,
      "num_tokens": 48688629.0,
      "step": 1493
    },
    {
      "entropy": 0.9250707626342773,
      "epoch": 0.7989304812834225,
      "grad_norm": 0.16939716041088104,
      "learning_rate": 3.85310051733633e-05,
      "loss": 0.9027,
      "mean_token_accuracy": 0.7660240978002548,
      "num_tokens": 48721110.0,
      "step": 1494
    },
    {
      "entropy": 0.936187669634819,
      "epoch": 0.7994652406417112,
      "grad_norm": 0.17359718680381775,
      "learning_rate": 3.843632609719088e-05,
      "loss": 0.9456,
      "mean_token_accuracy": 0.7620920091867447,
      "num_tokens": 48753669.0,
      "step": 1495
    },
    {
      "entropy": 1.0146075338125229,
      "epoch": 0.8,
      "grad_norm": 0.17324475944042206,
      "learning_rate": 3.834186190134567e-05,
      "loss": 1.0144,
      "mean_token_accuracy": 0.7455156147480011,
      "num_tokens": 48786164.0,
      "step": 1496
    },
    {
      "entropy": 0.953261598944664,
      "epoch": 0.8005347593582888,
      "grad_norm": 0.17196255922317505,
      "learning_rate": 3.824761286947011e-05,
      "loss": 0.943,
      "mean_token_accuracy": 0.7580036818981171,
      "num_tokens": 48818109.0,
      "step": 1497
    },
    {
      "entropy": 0.9698213785886765,
      "epoch": 0.8010695187165775,
      "grad_norm": 0.16612346470355988,
      "learning_rate": 3.8153579284560606e-05,
      "loss": 0.9641,
      "mean_token_accuracy": 0.7526881694793701,
      "num_tokens": 48850877.0,
      "step": 1498
    },
    {
      "entropy": 0.9340949952602386,
      "epoch": 0.8016042780748663,
      "grad_norm": 0.17099621891975403,
      "learning_rate": 3.8059761428966695e-05,
      "loss": 0.9346,
      "mean_token_accuracy": 0.7570564448833466,
      "num_tokens": 48883645.0,
      "step": 1499
    },
    {
      "entropy": 0.9078283756971359,
      "epoch": 0.8021390374331551,
      "grad_norm": 0.1709538996219635,
      "learning_rate": 3.796615958439009e-05,
      "loss": 0.9004,
      "mean_token_accuracy": 0.7695465236902237,
      "num_tokens": 48916331.0,
      "step": 1500
    }
  ],
  "logging_steps": 1,
  "max_steps": 1870,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 2.24019845480448e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}