{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 570,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0017543859649122807,
      "grad_norm": 6.018052101135254,
      "learning_rate": 0.0,
      "loss": 1.789,
      "mean_token_accuracy": 0.5678549408912659,
      "num_tokens": 429478.0,
      "step": 1
    },
    {
      "epoch": 0.0035087719298245615,
      "grad_norm": 6.373941898345947,
      "learning_rate": 1.7543859649122807e-06,
      "loss": 1.7812,
      "mean_token_accuracy": 0.5701305270195007,
      "num_tokens": 824562.0,
      "step": 2
    },
    {
      "epoch": 0.005263157894736842,
      "grad_norm": 6.300591468811035,
      "learning_rate": 3.5087719298245615e-06,
      "loss": 1.787,
      "mean_token_accuracy": 0.5685350894927979,
      "num_tokens": 1228422.0,
      "step": 3
    },
    {
      "epoch": 0.007017543859649123,
      "grad_norm": 4.870020389556885,
      "learning_rate": 5.263157894736842e-06,
      "loss": 1.7577,
      "mean_token_accuracy": 0.570589542388916,
      "num_tokens": 1633223.0,
      "step": 4
    },
    {
      "epoch": 0.008771929824561403,
      "grad_norm": 3.564033269882202,
      "learning_rate": 7.017543859649123e-06,
      "loss": 1.6964,
      "mean_token_accuracy": 0.5784072279930115,
      "num_tokens": 2060143.0,
      "step": 5
    },
    {
      "epoch": 0.010526315789473684,
      "grad_norm": 2.9137721061706543,
      "learning_rate": 8.771929824561403e-06,
      "loss": 1.646,
      "mean_token_accuracy": 0.5835782289505005,
      "num_tokens": 2503690.0,
      "step": 6
    },
    {
      "epoch": 0.012280701754385965,
      "grad_norm": 2.481250047683716,
      "learning_rate": 1.0526315789473684e-05,
      "loss": 1.5991,
      "mean_token_accuracy": 0.5927026271820068,
      "num_tokens": 2904426.0,
      "step": 7
    },
    {
      "epoch": 0.014035087719298246,
      "grad_norm": 3.379573345184326,
      "learning_rate": 1.2280701754385964e-05,
      "loss": 1.5736,
      "mean_token_accuracy": 0.5973439812660217,
      "num_tokens": 3299360.0,
      "step": 8
    },
    {
      "epoch": 0.015789473684210527,
      "grad_norm": 2.4704713821411133,
      "learning_rate": 1.4035087719298246e-05,
      "loss": 1.5427,
      "mean_token_accuracy": 0.6033717393875122,
      "num_tokens": 3709604.0,
      "step": 9
    },
    {
      "epoch": 0.017543859649122806,
      "grad_norm": 1.8616167306900024,
      "learning_rate": 1.5789473684210526e-05,
      "loss": 1.5179,
      "mean_token_accuracy": 0.606142520904541,
      "num_tokens": 4122427.0,
      "step": 10
    },
    {
      "epoch": 0.01929824561403509,
      "grad_norm": 1.90486478805542,
      "learning_rate": 1.7543859649122806e-05,
      "loss": 1.5291,
      "mean_token_accuracy": 0.6004294157028198,
      "num_tokens": 4549800.0,
      "step": 11
    },
    {
      "epoch": 0.021052631578947368,
      "grad_norm": 1.6314690113067627,
      "learning_rate": 1.929824561403509e-05,
      "loss": 1.5063,
      "mean_token_accuracy": 0.6060437560081482,
      "num_tokens": 4972367.0,
      "step": 12
    },
    {
      "epoch": 0.02280701754385965,
      "grad_norm": 1.4108632802963257,
      "learning_rate": 2.105263157894737e-05,
      "loss": 1.4803,
      "mean_token_accuracy": 0.6093297004699707,
      "num_tokens": 5381784.0,
      "step": 13
    },
    {
      "epoch": 0.02456140350877193,
      "grad_norm": 1.3794684410095215,
      "learning_rate": 2.280701754385965e-05,
      "loss": 1.4411,
      "mean_token_accuracy": 0.6182389259338379,
      "num_tokens": 5783941.0,
      "step": 14
    },
    {
      "epoch": 0.02631578947368421,
      "grad_norm": 1.2287817001342773,
      "learning_rate": 2.456140350877193e-05,
      "loss": 1.423,
      "mean_token_accuracy": 0.621440052986145,
      "num_tokens": 6173946.0,
      "step": 15
    },
    {
      "epoch": 0.028070175438596492,
      "grad_norm": 1.2201406955718994,
      "learning_rate": 2.6315789473684212e-05,
      "loss": 1.4455,
      "mean_token_accuracy": 0.6170344352722168,
      "num_tokens": 6588292.0,
      "step": 16
    },
    {
      "epoch": 0.02982456140350877,
      "grad_norm": 1.1274272203445435,
      "learning_rate": 2.8070175438596492e-05,
      "loss": 1.4051,
      "mean_token_accuracy": 0.6251938343048096,
      "num_tokens": 7005165.0,
      "step": 17
    },
    {
      "epoch": 0.031578947368421054,
      "grad_norm": 1.040313482284546,
      "learning_rate": 2.9824561403508772e-05,
      "loss": 1.4229,
      "mean_token_accuracy": 0.6200644373893738,
      "num_tokens": 7422596.0,
      "step": 18
    },
    {
      "epoch": 0.03333333333333333,
      "grad_norm": 1.0897908210754395,
      "learning_rate": 3.157894736842105e-05,
      "loss": 1.3782,
      "mean_token_accuracy": 0.6302369236946106,
      "num_tokens": 7812857.0,
      "step": 19
    },
    {
      "epoch": 0.03508771929824561,
      "grad_norm": 1.08786940574646,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.4219,
      "mean_token_accuracy": 0.6207268834114075,
      "num_tokens": 8244992.0,
      "step": 20
    },
    {
      "epoch": 0.03684210526315789,
      "grad_norm": 0.991358757019043,
      "learning_rate": 3.508771929824561e-05,
      "loss": 1.3912,
      "mean_token_accuracy": 0.6278927326202393,
      "num_tokens": 8664134.0,
      "step": 21
    },
    {
      "epoch": 0.03859649122807018,
      "grad_norm": 1.1284328699111938,
      "learning_rate": 3.6842105263157895e-05,
      "loss": 1.4004,
      "mean_token_accuracy": 0.6226587295532227,
      "num_tokens": 9082923.0,
      "step": 22
    },
    {
      "epoch": 0.04035087719298246,
      "grad_norm": 1.1181979179382324,
      "learning_rate": 3.859649122807018e-05,
      "loss": 1.3971,
      "mean_token_accuracy": 0.6253641247749329,
      "num_tokens": 9497934.0,
      "step": 23
    },
    {
      "epoch": 0.042105263157894736,
      "grad_norm": 1.2639045715332031,
      "learning_rate": 4.0350877192982455e-05,
      "loss": 1.4259,
      "mean_token_accuracy": 0.6176168322563171,
      "num_tokens": 9923716.0,
      "step": 24
    },
    {
      "epoch": 0.043859649122807015,
      "grad_norm": 1.0910519361495972,
      "learning_rate": 4.210526315789474e-05,
      "loss": 1.3733,
      "mean_token_accuracy": 0.6294311285018921,
      "num_tokens": 10353257.0,
      "step": 25
    },
    {
      "epoch": 0.0456140350877193,
      "grad_norm": 1.0363545417785645,
      "learning_rate": 4.3859649122807014e-05,
      "loss": 1.3809,
      "mean_token_accuracy": 0.6275283098220825,
      "num_tokens": 10767965.0,
      "step": 26
    },
    {
      "epoch": 0.04736842105263158,
      "grad_norm": 1.048935055732727,
      "learning_rate": 4.56140350877193e-05,
      "loss": 1.3815,
      "mean_token_accuracy": 0.6274988651275635,
      "num_tokens": 11188411.0,
      "step": 27
    },
    {
      "epoch": 0.04912280701754386,
      "grad_norm": 1.3705120086669922,
      "learning_rate": 4.736842105263158e-05,
      "loss": 1.357,
      "mean_token_accuracy": 0.6328938007354736,
      "num_tokens": 11597031.0,
      "step": 28
    },
    {
      "epoch": 0.05087719298245614,
      "grad_norm": 1.2502895593643188,
      "learning_rate": 4.912280701754386e-05,
      "loss": 1.3848,
      "mean_token_accuracy": 0.626395583152771,
      "num_tokens": 11996650.0,
      "step": 29
    },
    {
      "epoch": 0.05263157894736842,
      "grad_norm": 0.9786079525947571,
      "learning_rate": 5.087719298245615e-05,
      "loss": 1.3669,
      "mean_token_accuracy": 0.6301195621490479,
      "num_tokens": 12414351.0,
      "step": 30
    },
    {
      "epoch": 0.054385964912280704,
      "grad_norm": 1.322411298751831,
      "learning_rate": 5.2631578947368424e-05,
      "loss": 1.3703,
      "mean_token_accuracy": 0.6289023756980896,
      "num_tokens": 12853897.0,
      "step": 31
    },
    {
      "epoch": 0.056140350877192984,
      "grad_norm": 1.1092149019241333,
      "learning_rate": 5.438596491228071e-05,
      "loss": 1.343,
      "mean_token_accuracy": 0.6342843770980835,
      "num_tokens": 13246819.0,
      "step": 32
    },
    {
      "epoch": 0.05789473684210526,
      "grad_norm": 1.39752995967865,
      "learning_rate": 5.6140350877192984e-05,
      "loss": 1.3795,
      "mean_token_accuracy": 0.6259216666221619,
      "num_tokens": 13668134.0,
      "step": 33
    },
    {
      "epoch": 0.05964912280701754,
      "grad_norm": 1.2338861227035522,
      "learning_rate": 5.789473684210527e-05,
      "loss": 1.3461,
      "mean_token_accuracy": 0.6343478560447693,
      "num_tokens": 14077677.0,
      "step": 34
    },
    {
      "epoch": 0.06140350877192982,
      "grad_norm": 1.1885336637496948,
      "learning_rate": 5.9649122807017544e-05,
      "loss": 1.3637,
      "mean_token_accuracy": 0.6306077241897583,
      "num_tokens": 14502863.0,
      "step": 35
    },
    {
      "epoch": 0.06315789473684211,
      "grad_norm": 1.471887469291687,
      "learning_rate": 6.140350877192983e-05,
      "loss": 1.3589,
      "mean_token_accuracy": 0.6298720836639404,
      "num_tokens": 14940331.0,
      "step": 36
    },
    {
      "epoch": 0.06491228070175438,
      "grad_norm": 1.0302767753601074,
      "learning_rate": 6.31578947368421e-05,
      "loss": 1.3397,
      "mean_token_accuracy": 0.6345314979553223,
      "num_tokens": 15343369.0,
      "step": 37
    },
    {
      "epoch": 0.06666666666666667,
      "grad_norm": 1.42778480052948,
      "learning_rate": 6.49122807017544e-05,
      "loss": 1.3546,
      "mean_token_accuracy": 0.6308321356773376,
      "num_tokens": 15752951.0,
      "step": 38
    },
    {
      "epoch": 0.06842105263157895,
      "grad_norm": 1.52997624874115,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.3207,
      "mean_token_accuracy": 0.6400465965270996,
      "num_tokens": 16147599.0,
      "step": 39
    },
    {
      "epoch": 0.07017543859649122,
      "grad_norm": 1.1962817907333374,
      "learning_rate": 6.842105263157895e-05,
      "loss": 1.3581,
      "mean_token_accuracy": 0.6308744549751282,
      "num_tokens": 16557268.0,
      "step": 40
    },
    {
      "epoch": 0.07192982456140351,
      "grad_norm": 1.1255979537963867,
      "learning_rate": 7.017543859649122e-05,
      "loss": 1.3227,
      "mean_token_accuracy": 0.639037013053894,
      "num_tokens": 16950170.0,
      "step": 41
    },
    {
      "epoch": 0.07368421052631578,
      "grad_norm": 1.3424605131149292,
      "learning_rate": 7.192982456140351e-05,
      "loss": 1.3326,
      "mean_token_accuracy": 0.6364561319351196,
      "num_tokens": 17369691.0,
      "step": 42
    },
    {
      "epoch": 0.07543859649122807,
      "grad_norm": 1.4676284790039062,
      "learning_rate": 7.368421052631579e-05,
      "loss": 1.3922,
      "mean_token_accuracy": 0.6229093074798584,
      "num_tokens": 17777844.0,
      "step": 43
    },
    {
      "epoch": 0.07719298245614035,
      "grad_norm": 1.339996099472046,
      "learning_rate": 7.543859649122808e-05,
      "loss": 1.3405,
      "mean_token_accuracy": 0.6354778409004211,
      "num_tokens": 18191802.0,
      "step": 44
    },
    {
      "epoch": 0.07894736842105263,
      "grad_norm": 1.7620866298675537,
      "learning_rate": 7.719298245614036e-05,
      "loss": 1.3225,
      "mean_token_accuracy": 0.6384661197662354,
      "num_tokens": 18608551.0,
      "step": 45
    },
    {
      "epoch": 0.08070175438596491,
      "grad_norm": 1.4890868663787842,
      "learning_rate": 7.894736842105263e-05,
      "loss": 1.3829,
      "mean_token_accuracy": 0.6231105327606201,
      "num_tokens": 19020275.0,
      "step": 46
    },
    {
      "epoch": 0.0824561403508772,
      "grad_norm": 1.3470134735107422,
      "learning_rate": 8.070175438596491e-05,
      "loss": 1.3068,
      "mean_token_accuracy": 0.6428367495536804,
      "num_tokens": 19416792.0,
      "step": 47
    },
    {
      "epoch": 0.08421052631578947,
      "grad_norm": 1.2967629432678223,
      "learning_rate": 8.24561403508772e-05,
      "loss": 1.3352,
      "mean_token_accuracy": 0.6349055171012878,
      "num_tokens": 19842380.0,
      "step": 48
    },
    {
      "epoch": 0.08596491228070176,
      "grad_norm": 1.5379173755645752,
      "learning_rate": 8.421052631578948e-05,
      "loss": 1.3475,
      "mean_token_accuracy": 0.6317988634109497,
      "num_tokens": 20257340.0,
      "step": 49
    },
    {
      "epoch": 0.08771929824561403,
      "grad_norm": 1.2810230255126953,
      "learning_rate": 8.596491228070177e-05,
      "loss": 1.3337,
      "mean_token_accuracy": 0.6345160007476807,
      "num_tokens": 20693853.0,
      "step": 50
    },
    {
      "epoch": 0.08947368421052632,
      "grad_norm": 1.5687311887741089,
      "learning_rate": 8.771929824561403e-05,
      "loss": 1.3213,
      "mean_token_accuracy": 0.637017011642456,
      "num_tokens": 21145069.0,
      "step": 51
    },
    {
      "epoch": 0.0912280701754386,
      "grad_norm": 1.3021150827407837,
      "learning_rate": 8.947368421052632e-05,
      "loss": 1.3366,
      "mean_token_accuracy": 0.6351794600486755,
      "num_tokens": 21548907.0,
      "step": 52
    },
    {
      "epoch": 0.09298245614035087,
      "grad_norm": 1.6907377243041992,
      "learning_rate": 9.12280701754386e-05,
      "loss": 1.3326,
      "mean_token_accuracy": 0.6350468397140503,
      "num_tokens": 21947011.0,
      "step": 53
    },
    {
      "epoch": 0.09473684210526316,
      "grad_norm": 1.4103087186813354,
      "learning_rate": 9.298245614035089e-05,
      "loss": 1.3593,
      "mean_token_accuracy": 0.6285296082496643,
      "num_tokens": 22384995.0,
      "step": 54
    },
    {
      "epoch": 0.09649122807017543,
      "grad_norm": 1.3662679195404053,
      "learning_rate": 9.473684210526316e-05,
      "loss": 1.2921,
      "mean_token_accuracy": 0.644428014755249,
      "num_tokens": 22774976.0,
      "step": 55
    },
    {
      "epoch": 0.09824561403508772,
      "grad_norm": 1.4143177270889282,
      "learning_rate": 9.649122807017544e-05,
      "loss": 1.3537,
      "mean_token_accuracy": 0.6295624375343323,
      "num_tokens": 23192796.0,
      "step": 56
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.085375189781189,
      "learning_rate": 9.824561403508771e-05,
      "loss": 1.3102,
      "mean_token_accuracy": 0.6399141550064087,
      "num_tokens": 23610316.0,
      "step": 57
    },
    {
      "epoch": 0.10175438596491228,
      "grad_norm": 1.3309866189956665,
      "learning_rate": 0.0001,
      "loss": 1.3685,
      "mean_token_accuracy": 0.6267704367637634,
      "num_tokens": 24043550.0,
      "step": 58
    },
    {
      "epoch": 0.10350877192982456,
      "grad_norm": 1.4298138618469238,
      "learning_rate": 0.0001,
      "loss": 1.3263,
      "mean_token_accuracy": 0.6358023285865784,
      "num_tokens": 24482051.0,
      "step": 59
    },
    {
      "epoch": 0.10526315789473684,
      "grad_norm": 1.3495875597000122,
      "learning_rate": 0.0001,
      "loss": 1.3231,
      "mean_token_accuracy": 0.6365145444869995,
      "num_tokens": 24889567.0,
      "step": 60
    },
    {
      "epoch": 0.10701754385964912,
      "grad_norm": 1.3433363437652588,
      "learning_rate": 0.0001,
      "loss": 1.3265,
      "mean_token_accuracy": 0.6364420652389526,
      "num_tokens": 25296636.0,
      "step": 61
    },
    {
      "epoch": 0.10877192982456141,
      "grad_norm": 1.4023200273513794,
      "learning_rate": 0.0001,
      "loss": 1.2956,
      "mean_token_accuracy": 0.6436514854431152,
      "num_tokens": 25694614.0,
      "step": 62
    },
    {
      "epoch": 0.11052631578947368,
      "grad_norm": 1.4170814752578735,
      "learning_rate": 0.0001,
      "loss": 1.308,
      "mean_token_accuracy": 0.6407404541969299,
      "num_tokens": 26107689.0,
      "step": 63
    },
    {
      "epoch": 0.11228070175438597,
      "grad_norm": 1.198994755744934,
      "learning_rate": 0.0001,
      "loss": 1.3057,
      "mean_token_accuracy": 0.6392735838890076,
      "num_tokens": 26522427.0,
      "step": 64
    },
    {
      "epoch": 0.11403508771929824,
      "grad_norm": 1.422518014907837,
      "learning_rate": 0.0001,
      "loss": 1.3237,
      "mean_token_accuracy": 0.6369431018829346,
      "num_tokens": 26934123.0,
      "step": 65
    },
    {
      "epoch": 0.11578947368421053,
      "grad_norm": 1.3225864171981812,
      "learning_rate": 0.0001,
      "loss": 1.3213,
      "mean_token_accuracy": 0.637446403503418,
      "num_tokens": 27358915.0,
      "step": 66
    },
    {
      "epoch": 0.11754385964912281,
      "grad_norm": 1.1103287935256958,
      "learning_rate": 0.0001,
      "loss": 1.3107,
      "mean_token_accuracy": 0.6421770453453064,
      "num_tokens": 27775556.0,
      "step": 67
    },
    {
      "epoch": 0.11929824561403508,
      "grad_norm": 1.1607317924499512,
      "learning_rate": 0.0001,
      "loss": 1.3219,
      "mean_token_accuracy": 0.6369372010231018,
      "num_tokens": 28174439.0,
      "step": 68
    },
    {
      "epoch": 0.12105263157894737,
      "grad_norm": 1.121587872505188,
      "learning_rate": 0.0001,
      "loss": 1.2874,
      "mean_token_accuracy": 0.6452154517173767,
      "num_tokens": 28591718.0,
      "step": 69
    },
    {
      "epoch": 0.12280701754385964,
      "grad_norm": 1.347907304763794,
      "learning_rate": 0.0001,
      "loss": 1.3031,
      "mean_token_accuracy": 0.6407017707824707,
      "num_tokens": 29010174.0,
      "step": 70
    },
    {
      "epoch": 0.12456140350877193,
      "grad_norm": 0.9920047521591187,
      "learning_rate": 0.0001,
      "loss": 1.3081,
      "mean_token_accuracy": 0.6407448053359985,
      "num_tokens": 29430243.0,
      "step": 71
    },
    {
      "epoch": 0.12631578947368421,
      "grad_norm": 1.4440033435821533,
      "learning_rate": 0.0001,
      "loss": 1.317,
      "mean_token_accuracy": 0.6371080279350281,
      "num_tokens": 29852311.0,
      "step": 72
    },
    {
      "epoch": 0.1280701754385965,
      "grad_norm": 1.172947645187378,
      "learning_rate": 0.0001,
      "loss": 1.2955,
      "mean_token_accuracy": 0.6423896551132202,
      "num_tokens": 30267287.0,
      "step": 73
    },
    {
      "epoch": 0.12982456140350876,
      "grad_norm": 1.2112936973571777,
      "learning_rate": 0.0001,
      "loss": 1.3155,
      "mean_token_accuracy": 0.6372196674346924,
      "num_tokens": 30715583.0,
      "step": 74
    },
    {
      "epoch": 0.13157894736842105,
      "grad_norm": 1.1959091424942017,
      "learning_rate": 0.0001,
      "loss": 1.326,
      "mean_token_accuracy": 0.6362699866294861,
      "num_tokens": 31143599.0,
      "step": 75
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 1.3436111211776733,
      "learning_rate": 0.0001,
      "loss": 1.3162,
      "mean_token_accuracy": 0.6371738910675049,
      "num_tokens": 31573952.0,
      "step": 76
    },
    {
      "epoch": 0.13508771929824562,
      "grad_norm": 1.101008653640747,
      "learning_rate": 0.0001,
      "loss": 1.334,
      "mean_token_accuracy": 0.6339811086654663,
      "num_tokens": 31985309.0,
      "step": 77
    },
    {
      "epoch": 0.1368421052631579,
      "grad_norm": 1.2296723127365112,
      "learning_rate": 0.0001,
      "loss": 1.3346,
      "mean_token_accuracy": 0.6317209005355835,
      "num_tokens": 32422743.0,
      "step": 78
    },
    {
      "epoch": 0.13859649122807016,
      "grad_norm": 1.0157369375228882,
      "learning_rate": 0.0001,
      "loss": 1.3197,
      "mean_token_accuracy": 0.6366320848464966,
      "num_tokens": 32830001.0,
      "step": 79
    },
    {
      "epoch": 0.14035087719298245,
      "grad_norm": 1.1848087310791016,
      "learning_rate": 0.0001,
      "loss": 1.2867,
      "mean_token_accuracy": 0.6441957950592041,
      "num_tokens": 33242795.0,
      "step": 80
    },
    {
      "epoch": 0.14210526315789473,
      "grad_norm": 1.035370945930481,
      "learning_rate": 0.0001,
      "loss": 1.291,
      "mean_token_accuracy": 0.6427246332168579,
      "num_tokens": 33631163.0,
      "step": 81
    },
    {
      "epoch": 0.14385964912280702,
      "grad_norm": 1.2173899412155151,
      "learning_rate": 0.0001,
      "loss": 1.331,
      "mean_token_accuracy": 0.6329866647720337,
      "num_tokens": 34030802.0,
      "step": 82
    },
    {
      "epoch": 0.1456140350877193,
      "grad_norm": 1.2178702354431152,
      "learning_rate": 0.0001,
      "loss": 1.3159,
      "mean_token_accuracy": 0.6359639167785645,
      "num_tokens": 34443782.0,
      "step": 83
    },
    {
      "epoch": 0.14736842105263157,
      "grad_norm": 1.045278787612915,
      "learning_rate": 0.0001,
      "loss": 1.3248,
      "mean_token_accuracy": 0.6354624629020691,
      "num_tokens": 34854553.0,
      "step": 84
    },
    {
      "epoch": 0.14912280701754385,
      "grad_norm": 1.0509958267211914,
      "learning_rate": 0.0001,
      "loss": 1.3038,
      "mean_token_accuracy": 0.6399705410003662,
      "num_tokens": 35265577.0,
      "step": 85
    },
    {
      "epoch": 0.15087719298245614,
      "grad_norm": 1.1449450254440308,
      "learning_rate": 0.0001,
      "loss": 1.3004,
      "mean_token_accuracy": 0.6418460011482239,
      "num_tokens": 35705670.0,
      "step": 86
    },
    {
      "epoch": 0.15263157894736842,
      "grad_norm": 1.254193902015686,
      "learning_rate": 0.0001,
      "loss": 1.3088,
      "mean_token_accuracy": 0.6380743980407715,
      "num_tokens": 36117458.0,
      "step": 87
    },
    {
      "epoch": 0.1543859649122807,
      "grad_norm": 1.107653021812439,
      "learning_rate": 0.0001,
      "loss": 1.3009,
      "mean_token_accuracy": 0.6418921947479248,
      "num_tokens": 36528322.0,
      "step": 88
    },
    {
      "epoch": 0.156140350877193,
      "grad_norm": 0.9854401350021362,
      "learning_rate": 0.0001,
      "loss": 1.3022,
      "mean_token_accuracy": 0.6411072015762329,
      "num_tokens": 36937089.0,
      "step": 89
    },
    {
      "epoch": 0.15789473684210525,
      "grad_norm": 0.9852709174156189,
      "learning_rate": 0.0001,
      "loss": 1.2818,
      "mean_token_accuracy": 0.6445783972740173,
      "num_tokens": 37347449.0,
      "step": 90
    },
    {
      "epoch": 0.15964912280701754,
      "grad_norm": 1.0607930421829224,
      "learning_rate": 0.0001,
      "loss": 1.3042,
      "mean_token_accuracy": 0.6411298513412476,
      "num_tokens": 37768132.0,
      "step": 91
    },
    {
      "epoch": 0.16140350877192983,
      "grad_norm": 0.8618792295455933,
      "learning_rate": 0.0001,
      "loss": 1.2899,
      "mean_token_accuracy": 0.6437462568283081,
      "num_tokens": 38173420.0,
      "step": 92
    },
    {
      "epoch": 0.1631578947368421,
      "grad_norm": 0.9967447519302368,
      "learning_rate": 0.0001,
      "loss": 1.2854,
      "mean_token_accuracy": 0.6431381106376648,
      "num_tokens": 38564492.0,
      "step": 93
    },
    {
      "epoch": 0.1649122807017544,
      "grad_norm": 0.984609842300415,
      "learning_rate": 0.0001,
      "loss": 1.3059,
      "mean_token_accuracy": 0.6391075849533081,
      "num_tokens": 38973529.0,
      "step": 94
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 1.2071311473846436,
      "learning_rate": 0.0001,
      "loss": 1.3399,
      "mean_token_accuracy": 0.6329282522201538,
      "num_tokens": 39394470.0,
      "step": 95
    },
    {
      "epoch": 0.16842105263157894,
      "grad_norm": 0.976823627948761,
      "learning_rate": 0.0001,
      "loss": 1.3189,
      "mean_token_accuracy": 0.6357463598251343,
      "num_tokens": 39817247.0,
      "step": 96
    },
    {
      "epoch": 0.17017543859649123,
      "grad_norm": 1.1396266222000122,
      "learning_rate": 0.0001,
      "loss": 1.3212,
      "mean_token_accuracy": 0.6357501745223999,
      "num_tokens": 40219578.0,
      "step": 97
    },
    {
      "epoch": 0.17192982456140352,
      "grad_norm": 1.375174880027771,
      "learning_rate": 0.0001,
      "loss": 1.3187,
      "mean_token_accuracy": 0.6361098289489746,
      "num_tokens": 40631135.0,
      "step": 98
    },
    {
      "epoch": 0.1736842105263158,
      "grad_norm": 1.1790404319763184,
      "learning_rate": 0.0001,
      "loss": 1.2962,
      "mean_token_accuracy": 0.6430338621139526,
      "num_tokens": 41060591.0,
      "step": 99
    },
    {
      "epoch": 0.17543859649122806,
      "grad_norm": 1.1208826303482056,
      "learning_rate": 0.0001,
      "loss": 1.3243,
      "mean_token_accuracy": 0.6350250244140625,
      "num_tokens": 41491310.0,
      "step": 100
    },
    {
      "epoch": 0.17719298245614035,
      "grad_norm": 0.9812876582145691,
      "learning_rate": 0.0001,
      "loss": 1.2989,
      "mean_token_accuracy": 0.6403151154518127,
      "num_tokens": 41928251.0,
      "step": 101
    },
    {
      "epoch": 0.17894736842105263,
      "grad_norm": 1.118895173072815,
      "learning_rate": 0.0001,
      "loss": 1.2881,
      "mean_token_accuracy": 0.643653392791748,
      "num_tokens": 42328746.0,
      "step": 102
    },
    {
      "epoch": 0.18070175438596492,
      "grad_norm": 1.0872011184692383,
      "learning_rate": 0.0001,
      "loss": 1.3338,
      "mean_token_accuracy": 0.6348128318786621,
      "num_tokens": 42770651.0,
      "step": 103
    },
    {
      "epoch": 0.1824561403508772,
      "grad_norm": 1.0117576122283936,
      "learning_rate": 0.0001,
      "loss": 1.305,
      "mean_token_accuracy": 0.6394751071929932,
      "num_tokens": 43173759.0,
      "step": 104
    },
    {
      "epoch": 0.18421052631578946,
      "grad_norm": 0.9142250418663025,
      "learning_rate": 0.0001,
      "loss": 1.2908,
      "mean_token_accuracy": 0.642844557762146,
      "num_tokens": 43604295.0,
      "step": 105
    },
    {
      "epoch": 0.18596491228070175,
      "grad_norm": 1.1038587093353271,
      "learning_rate": 0.0001,
      "loss": 1.2963,
      "mean_token_accuracy": 0.6419985294342041,
      "num_tokens": 44008348.0,
      "step": 106
    },
    {
      "epoch": 0.18771929824561404,
      "grad_norm": 0.928559422492981,
      "learning_rate": 0.0001,
      "loss": 1.3107,
      "mean_token_accuracy": 0.6373360753059387,
      "num_tokens": 44444613.0,
      "step": 107
    },
    {
      "epoch": 0.18947368421052632,
      "grad_norm": 1.0053200721740723,
      "learning_rate": 0.0001,
      "loss": 1.2887,
      "mean_token_accuracy": 0.6448897123336792,
      "num_tokens": 44875312.0,
      "step": 108
    },
    {
      "epoch": 0.1912280701754386,
      "grad_norm": 0.9399821758270264,
      "learning_rate": 0.0001,
      "loss": 1.2996,
      "mean_token_accuracy": 0.6389566659927368,
      "num_tokens": 45273670.0,
      "step": 109
    },
    {
      "epoch": 0.19298245614035087,
      "grad_norm": 1.2514432668685913,
      "learning_rate": 0.0001,
      "loss": 1.2788,
      "mean_token_accuracy": 0.6447431445121765,
      "num_tokens": 45696861.0,
      "step": 110
    },
    {
      "epoch": 0.19473684210526315,
      "grad_norm": 0.9928343892097473,
      "learning_rate": 0.0001,
      "loss": 1.303,
      "mean_token_accuracy": 0.639816403388977,
      "num_tokens": 46115387.0,
      "step": 111
    },
    {
      "epoch": 0.19649122807017544,
      "grad_norm": 1.0918611288070679,
      "learning_rate": 0.0001,
      "loss": 1.2904,
      "mean_token_accuracy": 0.6424538493156433,
      "num_tokens": 46521933.0,
      "step": 112
    },
    {
      "epoch": 0.19824561403508772,
      "grad_norm": 1.1192419528961182,
      "learning_rate": 0.0001,
      "loss": 1.3263,
      "mean_token_accuracy": 0.634386420249939,
      "num_tokens": 46941357.0,
      "step": 113
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.9753395318984985,
      "learning_rate": 0.0001,
      "loss": 1.2792,
      "mean_token_accuracy": 0.6461848020553589,
      "num_tokens": 47343683.0,
      "step": 114
    },
    {
      "epoch": 0.20175438596491227,
      "grad_norm": 0.8872193694114685,
      "learning_rate": 0.0001,
      "loss": 1.2928,
      "mean_token_accuracy": 0.6433566808700562,
      "num_tokens": 47787665.0,
      "step": 115
    },
    {
      "epoch": 0.20350877192982456,
      "grad_norm": 0.9394273161888123,
      "learning_rate": 0.0001,
      "loss": 1.2839,
      "mean_token_accuracy": 0.643855094909668,
      "num_tokens": 48190313.0,
      "step": 116
    },
    {
      "epoch": 0.20526315789473684,
      "grad_norm": 1.136915922164917,
      "learning_rate": 0.0001,
      "loss": 1.2904,
      "mean_token_accuracy": 0.6421671509742737,
      "num_tokens": 48630247.0,
      "step": 117
    },
    {
      "epoch": 0.20701754385964913,
      "grad_norm": 0.9522098898887634,
      "learning_rate": 0.0001,
      "loss": 1.2942,
      "mean_token_accuracy": 0.6419311761856079,
      "num_tokens": 49009657.0,
      "step": 118
    },
    {
      "epoch": 0.20877192982456141,
      "grad_norm": 1.1538357734680176,
      "learning_rate": 0.0001,
      "loss": 1.2708,
      "mean_token_accuracy": 0.6458480358123779,
      "num_tokens": 49398873.0,
      "step": 119
    },
    {
      "epoch": 0.21052631578947367,
      "grad_norm": 0.9239334464073181,
      "learning_rate": 0.0001,
      "loss": 1.2768,
      "mean_token_accuracy": 0.6459267139434814,
      "num_tokens": 49804381.0,
      "step": 120
    },
    {
      "epoch": 0.21228070175438596,
      "grad_norm": 0.9793084859848022,
      "learning_rate": 0.0001,
      "loss": 1.2712,
      "mean_token_accuracy": 0.6456162929534912,
      "num_tokens": 50213766.0,
      "step": 121
    },
    {
      "epoch": 0.21403508771929824,
      "grad_norm": 1.1136904954910278,
      "learning_rate": 0.0001,
      "loss": 1.2877,
      "mean_token_accuracy": 0.6435809135437012,
      "num_tokens": 50625155.0,
      "step": 122
    },
    {
      "epoch": 0.21578947368421053,
      "grad_norm": 0.8962170481681824,
      "learning_rate": 0.0001,
      "loss": 1.2929,
      "mean_token_accuracy": 0.6421340703964233,
      "num_tokens": 51028769.0,
      "step": 123
    },
    {
      "epoch": 0.21754385964912282,
      "grad_norm": 1.0955440998077393,
      "learning_rate": 0.0001,
      "loss": 1.2801,
      "mean_token_accuracy": 0.6436057090759277,
      "num_tokens": 51442144.0,
      "step": 124
    },
    {
      "epoch": 0.21929824561403508,
      "grad_norm": 0.9009307622909546,
      "learning_rate": 0.0001,
      "loss": 1.2709,
      "mean_token_accuracy": 0.6477597951889038,
      "num_tokens": 51847840.0,
      "step": 125
    },
    {
      "epoch": 0.22105263157894736,
      "grad_norm": 1.0885659456253052,
      "learning_rate": 0.0001,
      "loss": 1.2771,
      "mean_token_accuracy": 0.6435371041297913,
      "num_tokens": 52284344.0,
      "step": 126
    },
    {
      "epoch": 0.22280701754385965,
      "grad_norm": 0.92705899477005,
      "learning_rate": 0.0001,
      "loss": 1.3212,
      "mean_token_accuracy": 0.6340687870979309,
      "num_tokens": 52685505.0,
      "step": 127
    },
    {
      "epoch": 0.22456140350877193,
      "grad_norm": 0.9139009118080139,
      "learning_rate": 0.0001,
      "loss": 1.2932,
      "mean_token_accuracy": 0.6407190561294556,
      "num_tokens": 53110059.0,
      "step": 128
    },
    {
      "epoch": 0.22631578947368422,
      "grad_norm": 0.8279791474342346,
      "learning_rate": 0.0001,
      "loss": 1.2659,
      "mean_token_accuracy": 0.6474618911743164,
      "num_tokens": 53502719.0,
      "step": 129
    },
    {
      "epoch": 0.22807017543859648,
      "grad_norm": 0.9933703541755676,
      "learning_rate": 0.0001,
      "loss": 1.2681,
      "mean_token_accuracy": 0.6483220458030701,
      "num_tokens": 53917317.0,
      "step": 130
    },
    {
      "epoch": 0.22982456140350876,
      "grad_norm": 0.887478768825531,
      "learning_rate": 0.0001,
      "loss": 1.3002,
      "mean_token_accuracy": 0.6406154632568359,
      "num_tokens": 54338305.0,
      "step": 131
    },
    {
      "epoch": 0.23157894736842105,
      "grad_norm": 0.8612638711929321,
      "learning_rate": 0.0001,
      "loss": 1.2774,
      "mean_token_accuracy": 0.6442515850067139,
      "num_tokens": 54754376.0,
      "step": 132
    },
    {
      "epoch": 0.23333333333333334,
      "grad_norm": 0.850595235824585,
      "learning_rate": 0.0001,
      "loss": 1.3222,
      "mean_token_accuracy": 0.6351276636123657,
      "num_tokens": 55177668.0,
      "step": 133
    },
    {
      "epoch": 0.23508771929824562,
      "grad_norm": 1.1265441179275513,
      "learning_rate": 0.0001,
      "loss": 1.2869,
      "mean_token_accuracy": 0.6440436840057373,
      "num_tokens": 55588413.0,
      "step": 134
    },
    {
      "epoch": 0.23684210526315788,
      "grad_norm": 0.8181601762771606,
      "learning_rate": 0.0001,
      "loss": 1.2689,
      "mean_token_accuracy": 0.6486095190048218,
      "num_tokens": 56002661.0,
      "step": 135
    },
    {
      "epoch": 0.23859649122807017,
      "grad_norm": 0.9597206115722656,
      "learning_rate": 0.0001,
      "loss": 1.2685,
      "mean_token_accuracy": 0.6476383209228516,
      "num_tokens": 56399978.0,
      "step": 136
    },
    {
      "epoch": 0.24035087719298245,
      "grad_norm": 0.9021192193031311,
      "learning_rate": 0.0001,
      "loss": 1.277,
      "mean_token_accuracy": 0.6456701159477234,
      "num_tokens": 56805057.0,
      "step": 137
    },
    {
      "epoch": 0.24210526315789474,
      "grad_norm": 0.9269475936889648,
      "learning_rate": 0.0001,
      "loss": 1.2937,
      "mean_token_accuracy": 0.6424517631530762,
      "num_tokens": 57221548.0,
      "step": 138
    },
    {
      "epoch": 0.24385964912280703,
      "grad_norm": 0.9395855069160461,
      "learning_rate": 0.0001,
      "loss": 1.2905,
      "mean_token_accuracy": 0.6421390175819397,
      "num_tokens": 57619205.0,
      "step": 139
    },
    {
      "epoch": 0.24561403508771928,
      "grad_norm": 1.0334845781326294,
      "learning_rate": 0.0001,
      "loss": 1.2727,
      "mean_token_accuracy": 0.6469994783401489,
      "num_tokens": 58029104.0,
      "step": 140
    },
    {
      "epoch": 0.24736842105263157,
      "grad_norm": 1.080823302268982,
      "learning_rate": 0.0001,
      "loss": 1.3086,
      "mean_token_accuracy": 0.6387298107147217,
      "num_tokens": 58444372.0,
      "step": 141
    },
    {
      "epoch": 0.24912280701754386,
      "grad_norm": 0.8953016400337219,
      "learning_rate": 0.0001,
      "loss": 1.3045,
      "mean_token_accuracy": 0.6378800272941589,
      "num_tokens": 58859539.0,
      "step": 142
    },
    {
      "epoch": 0.25087719298245614,
      "grad_norm": 0.8567958474159241,
      "learning_rate": 0.0001,
      "loss": 1.3121,
      "mean_token_accuracy": 0.6372794508934021,
      "num_tokens": 59268101.0,
      "step": 143
    },
    {
      "epoch": 0.25263157894736843,
      "grad_norm": 1.158692479133606,
      "learning_rate": 0.0001,
      "loss": 1.2691,
      "mean_token_accuracy": 0.6473510265350342,
      "num_tokens": 59708311.0,
      "step": 144
    },
    {
      "epoch": 0.2543859649122807,
      "grad_norm": 0.9232509732246399,
      "learning_rate": 0.0001,
      "loss": 1.2682,
      "mean_token_accuracy": 0.648154616355896,
      "num_tokens": 60135806.0,
      "step": 145
    },
    {
      "epoch": 0.256140350877193,
      "grad_norm": 0.9411163330078125,
      "learning_rate": 0.0001,
      "loss": 1.3155,
      "mean_token_accuracy": 0.6371098756790161,
      "num_tokens": 60546162.0,
      "step": 146
    },
    {
      "epoch": 0.2578947368421053,
      "grad_norm": 1.013136863708496,
      "learning_rate": 0.0001,
      "loss": 1.2655,
      "mean_token_accuracy": 0.646664023399353,
      "num_tokens": 60977521.0,
      "step": 147
    },
    {
      "epoch": 0.2596491228070175,
      "grad_norm": 1.1551271677017212,
      "learning_rate": 0.0001,
      "loss": 1.2798,
      "mean_token_accuracy": 0.6451727747917175,
      "num_tokens": 61372998.0,
      "step": 148
    },
    {
      "epoch": 0.2614035087719298,
      "grad_norm": 0.8795229196548462,
      "learning_rate": 0.0001,
      "loss": 1.2982,
      "mean_token_accuracy": 0.6401211023330688,
      "num_tokens": 61781320.0,
      "step": 149
    },
    {
      "epoch": 0.2631578947368421,
      "grad_norm": 0.965307891368866,
      "learning_rate": 0.0001,
      "loss": 1.2788,
      "mean_token_accuracy": 0.6439850330352783,
      "num_tokens": 62199535.0,
      "step": 150
    },
    {
      "epoch": 0.2649122807017544,
      "grad_norm": 0.9804089665412903,
      "learning_rate": 0.0001,
      "loss": 1.2765,
      "mean_token_accuracy": 0.6456645727157593,
      "num_tokens": 62632976.0,
      "step": 151
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.9098561406135559,
      "learning_rate": 0.0001,
      "loss": 1.2564,
      "mean_token_accuracy": 0.6501985192298889,
      "num_tokens": 63043041.0,
      "step": 152
    },
    {
      "epoch": 0.26842105263157895,
      "grad_norm": 0.7934507727622986,
      "learning_rate": 0.0001,
      "loss": 1.2638,
      "mean_token_accuracy": 0.6469926834106445,
      "num_tokens": 63453330.0,
      "step": 153
    },
    {
      "epoch": 0.27017543859649124,
      "grad_norm": 1.0823460817337036,
      "learning_rate": 0.0001,
      "loss": 1.2983,
      "mean_token_accuracy": 0.6407555341720581,
      "num_tokens": 63864814.0,
      "step": 154
    },
    {
      "epoch": 0.2719298245614035,
      "grad_norm": 0.7126585841178894,
      "learning_rate": 0.0001,
      "loss": 1.2528,
      "mean_token_accuracy": 0.6504640579223633,
      "num_tokens": 64280965.0,
      "step": 155
    },
    {
      "epoch": 0.2736842105263158,
      "grad_norm": 0.9585691690444946,
      "learning_rate": 0.0001,
      "loss": 1.2686,
      "mean_token_accuracy": 0.6472254395484924,
      "num_tokens": 64672526.0,
      "step": 156
    },
    {
      "epoch": 0.2754385964912281,
      "grad_norm": 0.752656102180481,
      "learning_rate": 0.0001,
      "loss": 1.2474,
      "mean_token_accuracy": 0.6523414850234985,
      "num_tokens": 65077313.0,
      "step": 157
    },
    {
      "epoch": 0.2771929824561403,
      "grad_norm": 0.9288751482963562,
      "learning_rate": 0.0001,
      "loss": 1.2624,
      "mean_token_accuracy": 0.646138072013855,
      "num_tokens": 65487692.0,
      "step": 158
    },
    {
      "epoch": 0.2789473684210526,
      "grad_norm": 0.8809065222740173,
      "learning_rate": 0.0001,
      "loss": 1.2607,
      "mean_token_accuracy": 0.6489483118057251,
      "num_tokens": 65905410.0,
      "step": 159
    },
    {
      "epoch": 0.2807017543859649,
      "grad_norm": 0.9240980744361877,
      "learning_rate": 0.0001,
      "loss": 1.2958,
      "mean_token_accuracy": 0.6388487219810486,
      "num_tokens": 66310475.0,
      "step": 160
    },
    {
      "epoch": 0.2824561403508772,
      "grad_norm": 0.8388931751251221,
      "learning_rate": 0.0001,
      "loss": 1.2838,
      "mean_token_accuracy": 0.6431634426116943,
      "num_tokens": 66727367.0,
      "step": 161
    },
    {
      "epoch": 0.28421052631578947,
      "grad_norm": 0.8820334076881409,
      "learning_rate": 0.0001,
      "loss": 1.303,
      "mean_token_accuracy": 0.63872230052948,
      "num_tokens": 67154750.0,
      "step": 162
    },
    {
      "epoch": 0.28596491228070176,
      "grad_norm": 0.8385342359542847,
      "learning_rate": 0.0001,
      "loss": 1.2632,
      "mean_token_accuracy": 0.6464277505874634,
      "num_tokens": 67571403.0,
      "step": 163
    },
    {
      "epoch": 0.28771929824561404,
      "grad_norm": 0.8737322092056274,
      "learning_rate": 0.0001,
      "loss": 1.248,
      "mean_token_accuracy": 0.6486573219299316,
      "num_tokens": 67947644.0,
      "step": 164
    },
    {
      "epoch": 0.2894736842105263,
      "grad_norm": 0.8021559119224548,
      "learning_rate": 0.0001,
      "loss": 1.2716,
      "mean_token_accuracy": 0.644127607345581,
      "num_tokens": 68362596.0,
      "step": 165
    },
    {
      "epoch": 0.2912280701754386,
      "grad_norm": 0.7599727511405945,
      "learning_rate": 0.0001,
      "loss": 1.2511,
      "mean_token_accuracy": 0.6510897874832153,
      "num_tokens": 68768455.0,
      "step": 166
    },
    {
      "epoch": 0.2929824561403509,
      "grad_norm": 0.8421241044998169,
      "learning_rate": 0.0001,
      "loss": 1.2839,
      "mean_token_accuracy": 0.6423199772834778,
      "num_tokens": 69182971.0,
      "step": 167
    },
    {
      "epoch": 0.29473684210526313,
      "grad_norm": 0.8853815793991089,
      "learning_rate": 0.0001,
      "loss": 1.2698,
      "mean_token_accuracy": 0.6462802886962891,
      "num_tokens": 69633484.0,
      "step": 168
    },
    {
      "epoch": 0.2964912280701754,
      "grad_norm": 1.0347827672958374,
      "learning_rate": 0.0001,
      "loss": 1.2935,
      "mean_token_accuracy": 0.6394780874252319,
      "num_tokens": 70069352.0,
      "step": 169
    },
    {
      "epoch": 0.2982456140350877,
      "grad_norm": 0.8993912935256958,
      "learning_rate": 0.0001,
      "loss": 1.2689,
      "mean_token_accuracy": 0.6471203565597534,
      "num_tokens": 70478616.0,
      "step": 170
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.8886722922325134,
      "learning_rate": 0.0001,
      "loss": 1.2943,
      "mean_token_accuracy": 0.6398030519485474,
      "num_tokens": 70905952.0,
      "step": 171
    },
    {
      "epoch": 0.3017543859649123,
      "grad_norm": 1.1024540662765503,
      "learning_rate": 0.0001,
      "loss": 1.2799,
      "mean_token_accuracy": 0.6444812417030334,
      "num_tokens": 71317263.0,
      "step": 172
    },
    {
      "epoch": 0.30350877192982456,
      "grad_norm": 0.8578200340270996,
      "learning_rate": 0.0001,
      "loss": 1.2851,
      "mean_token_accuracy": 0.6433255672454834,
      "num_tokens": 71764261.0,
      "step": 173
    },
    {
      "epoch": 0.30526315789473685,
      "grad_norm": 0.9540239572525024,
      "learning_rate": 0.0001,
      "loss": 1.2777,
      "mean_token_accuracy": 0.644009530544281,
      "num_tokens": 72179518.0,
      "step": 174
    },
    {
      "epoch": 0.30701754385964913,
      "grad_norm": 0.9197105169296265,
      "learning_rate": 0.0001,
      "loss": 1.2664,
      "mean_token_accuracy": 0.6461474895477295,
      "num_tokens": 72594932.0,
      "step": 175
    },
    {
      "epoch": 0.3087719298245614,
      "grad_norm": 0.7414684891700745,
      "learning_rate": 0.0001,
      "loss": 1.2688,
      "mean_token_accuracy": 0.6475616097450256,
      "num_tokens": 73005473.0,
      "step": 176
    },
    {
      "epoch": 0.3105263157894737,
      "grad_norm": 0.9558865427970886,
      "learning_rate": 0.0001,
      "loss": 1.2673,
      "mean_token_accuracy": 0.6460444927215576,
      "num_tokens": 73397415.0,
      "step": 177
    },
    {
      "epoch": 0.312280701754386,
      "grad_norm": 0.9229446053504944,
      "learning_rate": 0.0001,
      "loss": 1.2738,
      "mean_token_accuracy": 0.6450560092926025,
      "num_tokens": 73793964.0,
      "step": 178
    },
    {
      "epoch": 0.3140350877192982,
      "grad_norm": 0.7821291089057922,
      "learning_rate": 0.0001,
      "loss": 1.2489,
      "mean_token_accuracy": 0.6500634551048279,
      "num_tokens": 74195270.0,
      "step": 179
    },
    {
      "epoch": 0.3157894736842105,
      "grad_norm": 0.7419561743736267,
      "learning_rate": 0.0001,
      "loss": 1.2612,
      "mean_token_accuracy": 0.6482713222503662,
      "num_tokens": 74603802.0,
      "step": 180
    },
    {
      "epoch": 0.3175438596491228,
      "grad_norm": 0.7956511378288269,
      "learning_rate": 0.0001,
      "loss": 1.2644,
      "mean_token_accuracy": 0.6461498141288757,
      "num_tokens": 75026083.0,
      "step": 181
    },
    {
      "epoch": 0.3192982456140351,
      "grad_norm": 0.7759901285171509,
      "learning_rate": 0.0001,
      "loss": 1.2647,
      "mean_token_accuracy": 0.6459044218063354,
      "num_tokens": 75426052.0,
      "step": 182
    },
    {
      "epoch": 0.32105263157894737,
      "grad_norm": 0.8206394910812378,
      "learning_rate": 0.0001,
      "loss": 1.2661,
      "mean_token_accuracy": 0.646410346031189,
      "num_tokens": 75830225.0,
      "step": 183
    },
    {
      "epoch": 0.32280701754385965,
      "grad_norm": 0.9032196402549744,
      "learning_rate": 0.0001,
      "loss": 1.2677,
      "mean_token_accuracy": 0.64692223072052,
      "num_tokens": 76241940.0,
      "step": 184
    },
    {
      "epoch": 0.32456140350877194,
      "grad_norm": 0.7018728256225586,
      "learning_rate": 0.0001,
      "loss": 1.2505,
      "mean_token_accuracy": 0.6487954258918762,
      "num_tokens": 76663142.0,
      "step": 185
    },
    {
      "epoch": 0.3263157894736842,
      "grad_norm": 0.9026210904121399,
      "learning_rate": 0.0001,
      "loss": 1.2483,
      "mean_token_accuracy": 0.652391254901886,
      "num_tokens": 77102129.0,
      "step": 186
    },
    {
      "epoch": 0.3280701754385965,
      "grad_norm": 0.8878228068351746,
      "learning_rate": 0.0001,
      "loss": 1.2736,
      "mean_token_accuracy": 0.6432400941848755,
      "num_tokens": 77508487.0,
      "step": 187
    },
    {
      "epoch": 0.3298245614035088,
      "grad_norm": 0.9250103235244751,
      "learning_rate": 0.0001,
      "loss": 1.2426,
      "mean_token_accuracy": 0.651581883430481,
      "num_tokens": 77908351.0,
      "step": 188
    },
    {
      "epoch": 0.33157894736842103,
      "grad_norm": 0.6793785095214844,
      "learning_rate": 0.0001,
      "loss": 1.2453,
      "mean_token_accuracy": 0.6514815092086792,
      "num_tokens": 78320845.0,
      "step": 189
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.7402032017707825,
      "learning_rate": 0.0001,
      "loss": 1.2523,
      "mean_token_accuracy": 0.6495726704597473,
      "num_tokens": 78732383.0,
      "step": 190
    },
    {
      "epoch": 0.3350877192982456,
      "grad_norm": 0.9974339604377747,
      "learning_rate": 0.0001,
      "loss": 1.2545,
      "mean_token_accuracy": 0.6485158801078796,
      "num_tokens": 79123256.0,
      "step": 191
    },
    {
      "epoch": 0.3368421052631579,
      "grad_norm": 0.9054931998252869,
      "learning_rate": 0.0001,
      "loss": 1.2538,
      "mean_token_accuracy": 0.6507511734962463,
      "num_tokens": 79525484.0,
      "step": 192
    },
    {
      "epoch": 0.3385964912280702,
      "grad_norm": 0.7434863448143005,
      "learning_rate": 0.0001,
      "loss": 1.2615,
      "mean_token_accuracy": 0.6481617093086243,
      "num_tokens": 79955140.0,
      "step": 193
    },
    {
      "epoch": 0.34035087719298246,
      "grad_norm": 0.7779750823974609,
      "learning_rate": 0.0001,
      "loss": 1.2739,
      "mean_token_accuracy": 0.6448099613189697,
      "num_tokens": 80387845.0,
      "step": 194
    },
    {
      "epoch": 0.34210526315789475,
      "grad_norm": 0.8742808103561401,
      "learning_rate": 0.0001,
      "loss": 1.2674,
      "mean_token_accuracy": 0.6466174125671387,
      "num_tokens": 80810849.0,
      "step": 195
    },
    {
      "epoch": 0.34385964912280703,
      "grad_norm": 0.810045063495636,
      "learning_rate": 0.0001,
      "loss": 1.2722,
      "mean_token_accuracy": 0.6458349227905273,
      "num_tokens": 81226626.0,
      "step": 196
    },
    {
      "epoch": 0.3456140350877193,
      "grad_norm": 0.7127732634544373,
      "learning_rate": 0.0001,
      "loss": 1.2912,
      "mean_token_accuracy": 0.6405543684959412,
      "num_tokens": 81648045.0,
      "step": 197
    },
    {
      "epoch": 0.3473684210526316,
      "grad_norm": 0.8309784531593323,
      "learning_rate": 0.0001,
      "loss": 1.254,
      "mean_token_accuracy": 0.6493059396743774,
      "num_tokens": 82055621.0,
      "step": 198
    },
    {
      "epoch": 0.34912280701754383,
      "grad_norm": 0.8503166437149048,
      "learning_rate": 0.0001,
      "loss": 1.2772,
      "mean_token_accuracy": 0.6441330909729004,
      "num_tokens": 82454820.0,
      "step": 199
    },
    {
      "epoch": 0.3508771929824561,
      "grad_norm": 0.8834285736083984,
      "learning_rate": 0.0001,
      "loss": 1.2701,
      "mean_token_accuracy": 0.6456678509712219,
      "num_tokens": 82881912.0,
      "step": 200
    },
    {
      "epoch": 0.3526315789473684,
      "grad_norm": 0.7746639847755432,
      "learning_rate": 0.0001,
      "loss": 1.2731,
      "mean_token_accuracy": 0.6454135775566101,
      "num_tokens": 83294708.0,
      "step": 201
    },
    {
      "epoch": 0.3543859649122807,
      "grad_norm": 0.8626236915588379,
      "learning_rate": 0.0001,
      "loss": 1.2677,
      "mean_token_accuracy": 0.6472684144973755,
      "num_tokens": 83692153.0,
      "step": 202
    },
    {
      "epoch": 0.356140350877193,
      "grad_norm": 0.8129353523254395,
      "learning_rate": 0.0001,
      "loss": 1.2504,
      "mean_token_accuracy": 0.649857223033905,
      "num_tokens": 84106215.0,
      "step": 203
    },
    {
      "epoch": 0.35789473684210527,
      "grad_norm": 0.9501094818115234,
      "learning_rate": 0.0001,
      "loss": 1.2788,
      "mean_token_accuracy": 0.6440906524658203,
      "num_tokens": 84533326.0,
      "step": 204
    },
    {
      "epoch": 0.35964912280701755,
      "grad_norm": 0.7424087524414062,
      "learning_rate": 0.0001,
      "loss": 1.2663,
      "mean_token_accuracy": 0.6469358205795288,
      "num_tokens": 84958198.0,
      "step": 205
    },
    {
      "epoch": 0.36140350877192984,
      "grad_norm": 0.7956259846687317,
      "learning_rate": 0.0001,
      "loss": 1.2691,
      "mean_token_accuracy": 0.6459769010543823,
      "num_tokens": 85375870.0,
      "step": 206
    },
    {
      "epoch": 0.3631578947368421,
      "grad_norm": 0.7288737893104553,
      "learning_rate": 0.0001,
      "loss": 1.2707,
      "mean_token_accuracy": 0.6465530395507812,
      "num_tokens": 85800348.0,
      "step": 207
    },
    {
      "epoch": 0.3649122807017544,
      "grad_norm": 0.7138190865516663,
      "learning_rate": 0.0001,
      "loss": 1.2415,
      "mean_token_accuracy": 0.6529696583747864,
      "num_tokens": 86195245.0,
      "step": 208
    },
    {
      "epoch": 0.36666666666666664,
      "grad_norm": 0.9041345119476318,
      "learning_rate": 0.0001,
      "loss": 1.2673,
      "mean_token_accuracy": 0.6487336754798889,
      "num_tokens": 86599515.0,
      "step": 209
    },
    {
      "epoch": 0.3684210526315789,
      "grad_norm": 0.7553381323814392,
      "learning_rate": 0.0001,
      "loss": 1.298,
      "mean_token_accuracy": 0.6391161680221558,
      "num_tokens": 87039537.0,
      "step": 210
    },
    {
      "epoch": 0.3701754385964912,
      "grad_norm": 0.7526540160179138,
      "learning_rate": 0.0001,
      "loss": 1.2465,
      "mean_token_accuracy": 0.6537913084030151,
      "num_tokens": 87426029.0,
      "step": 211
    },
    {
      "epoch": 0.3719298245614035,
      "grad_norm": 0.9352124333381653,
      "learning_rate": 0.0001,
      "loss": 1.2655,
      "mean_token_accuracy": 0.6462576389312744,
      "num_tokens": 87817987.0,
      "step": 212
    },
    {
      "epoch": 0.3736842105263158,
      "grad_norm": 0.8342838883399963,
      "learning_rate": 0.0001,
      "loss": 1.2356,
      "mean_token_accuracy": 0.653677225112915,
      "num_tokens": 88218779.0,
      "step": 213
    },
    {
      "epoch": 0.37543859649122807,
      "grad_norm": 0.7606971263885498,
      "learning_rate": 0.0001,
      "loss": 1.2423,
      "mean_token_accuracy": 0.6509567499160767,
      "num_tokens": 88610670.0,
      "step": 214
    },
    {
      "epoch": 0.37719298245614036,
      "grad_norm": 0.9147993326187134,
      "learning_rate": 0.0001,
      "loss": 1.2777,
      "mean_token_accuracy": 0.6447807550430298,
      "num_tokens": 89031094.0,
      "step": 215
    },
    {
      "epoch": 0.37894736842105264,
      "grad_norm": 0.8798630833625793,
      "learning_rate": 0.0001,
      "loss": 1.282,
      "mean_token_accuracy": 0.6422438025474548,
      "num_tokens": 89465235.0,
      "step": 216
    },
    {
      "epoch": 0.38070175438596493,
      "grad_norm": 0.7571805119514465,
      "learning_rate": 0.0001,
      "loss": 1.2503,
      "mean_token_accuracy": 0.6498503684997559,
      "num_tokens": 89867145.0,
      "step": 217
    },
    {
      "epoch": 0.3824561403508772,
      "grad_norm": 0.9793193936347961,
      "learning_rate": 0.0001,
      "loss": 1.248,
      "mean_token_accuracy": 0.6518094539642334,
      "num_tokens": 90262494.0,
      "step": 218
    },
    {
      "epoch": 0.38421052631578945,
      "grad_norm": 0.871235728263855,
      "learning_rate": 0.0001,
      "loss": 1.2707,
      "mean_token_accuracy": 0.6453557014465332,
      "num_tokens": 90671131.0,
      "step": 219
    },
    {
      "epoch": 0.38596491228070173,
      "grad_norm": 0.7807226181030273,
      "learning_rate": 0.0001,
      "loss": 1.2607,
      "mean_token_accuracy": 0.6465795040130615,
      "num_tokens": 91092593.0,
      "step": 220
    },
    {
      "epoch": 0.387719298245614,
      "grad_norm": 0.9600160121917725,
      "learning_rate": 0.0001,
      "loss": 1.2791,
      "mean_token_accuracy": 0.6422662734985352,
      "num_tokens": 91501248.0,
      "step": 221
    },
    {
      "epoch": 0.3894736842105263,
      "grad_norm": 0.8549517393112183,
      "learning_rate": 0.0001,
      "loss": 1.243,
      "mean_token_accuracy": 0.6511504650115967,
      "num_tokens": 91913007.0,
      "step": 222
    },
    {
      "epoch": 0.3912280701754386,
      "grad_norm": 0.7951960563659668,
      "learning_rate": 0.0001,
      "loss": 1.2377,
      "mean_token_accuracy": 0.6538807153701782,
      "num_tokens": 92321188.0,
      "step": 223
    },
    {
      "epoch": 0.3929824561403509,
      "grad_norm": 0.8606045842170715,
      "learning_rate": 0.0001,
      "loss": 1.2678,
      "mean_token_accuracy": 0.645931601524353,
      "num_tokens": 92726934.0,
      "step": 224
    },
    {
      "epoch": 0.39473684210526316,
      "grad_norm": 0.7008436322212219,
      "learning_rate": 0.0001,
      "loss": 1.2384,
      "mean_token_accuracy": 0.6525442600250244,
      "num_tokens": 93128966.0,
      "step": 225
    },
    {
      "epoch": 0.39649122807017545,
      "grad_norm": 0.7526488304138184,
      "learning_rate": 0.0001,
      "loss": 1.2738,
      "mean_token_accuracy": 0.6442047357559204,
      "num_tokens": 93567917.0,
      "step": 226
    },
    {
      "epoch": 0.39824561403508774,
      "grad_norm": 0.8679794669151306,
      "learning_rate": 0.0001,
      "loss": 1.2548,
      "mean_token_accuracy": 0.6482324600219727,
      "num_tokens": 93979268.0,
      "step": 227
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.8233749270439148,
      "learning_rate": 0.0001,
      "loss": 1.2215,
      "mean_token_accuracy": 0.6573722958564758,
      "num_tokens": 94367195.0,
      "step": 228
    },
    {
      "epoch": 0.4017543859649123,
      "grad_norm": 0.7261408567428589,
      "learning_rate": 0.0001,
      "loss": 1.2415,
      "mean_token_accuracy": 0.6515704989433289,
      "num_tokens": 94759639.0,
      "step": 229
    },
    {
      "epoch": 0.40350877192982454,
      "grad_norm": 0.7959755659103394,
      "learning_rate": 0.0001,
      "loss": 1.2722,
      "mean_token_accuracy": 0.6438157558441162,
      "num_tokens": 95191668.0,
      "step": 230
    },
    {
      "epoch": 0.4052631578947368,
      "grad_norm": 0.8794543147087097,
      "learning_rate": 0.0001,
      "loss": 1.2477,
      "mean_token_accuracy": 0.6511826515197754,
      "num_tokens": 95614874.0,
      "step": 231
    },
    {
      "epoch": 0.4070175438596491,
      "grad_norm": 0.7663288116455078,
      "learning_rate": 0.0001,
      "loss": 1.2175,
      "mean_token_accuracy": 0.656987190246582,
      "num_tokens": 95990695.0,
      "step": 232
    },
    {
      "epoch": 0.4087719298245614,
      "grad_norm": 0.7509688138961792,
      "learning_rate": 0.0001,
      "loss": 1.2395,
      "mean_token_accuracy": 0.6518152952194214,
      "num_tokens": 96377947.0,
      "step": 233
    },
    {
      "epoch": 0.4105263157894737,
      "grad_norm": 0.9182112812995911,
      "learning_rate": 0.0001,
      "loss": 1.2567,
      "mean_token_accuracy": 0.6489981412887573,
      "num_tokens": 96787894.0,
      "step": 234
    },
    {
      "epoch": 0.41228070175438597,
      "grad_norm": 0.8123442530632019,
      "learning_rate": 0.0001,
      "loss": 1.2541,
      "mean_token_accuracy": 0.6488667726516724,
      "num_tokens": 97203762.0,
      "step": 235
    },
    {
      "epoch": 0.41403508771929826,
      "grad_norm": 0.8581697344779968,
      "learning_rate": 0.0001,
      "loss": 1.2595,
      "mean_token_accuracy": 0.6481375694274902,
      "num_tokens": 97598494.0,
      "step": 236
    },
    {
      "epoch": 0.41578947368421054,
      "grad_norm": 0.8051207065582275,
      "learning_rate": 0.0001,
      "loss": 1.2763,
      "mean_token_accuracy": 0.644673228263855,
      "num_tokens": 98011465.0,
      "step": 237
    },
    {
      "epoch": 0.41754385964912283,
      "grad_norm": 0.7852127552032471,
      "learning_rate": 0.0001,
      "loss": 1.2765,
      "mean_token_accuracy": 0.6438398361206055,
      "num_tokens": 98447067.0,
      "step": 238
    },
    {
      "epoch": 0.4192982456140351,
      "grad_norm": 0.7962046265602112,
      "learning_rate": 0.0001,
      "loss": 1.2429,
      "mean_token_accuracy": 0.6508150100708008,
      "num_tokens": 98854782.0,
      "step": 239
    },
    {
      "epoch": 0.42105263157894735,
      "grad_norm": 0.8521065711975098,
      "learning_rate": 0.0001,
      "loss": 1.2499,
      "mean_token_accuracy": 0.64998859167099,
      "num_tokens": 99276423.0,
      "step": 240
    },
    {
      "epoch": 0.42280701754385963,
      "grad_norm": 0.8006791472434998,
      "learning_rate": 0.0001,
      "loss": 1.2523,
      "mean_token_accuracy": 0.65036940574646,
      "num_tokens": 99705527.0,
      "step": 241
    },
    {
      "epoch": 0.4245614035087719,
      "grad_norm": 0.6923927664756775,
      "learning_rate": 0.0001,
      "loss": 1.2698,
      "mean_token_accuracy": 0.645989179611206,
      "num_tokens": 100144851.0,
      "step": 242
    },
    {
      "epoch": 0.4263157894736842,
      "grad_norm": 0.8310588002204895,
      "learning_rate": 0.0001,
      "loss": 1.2131,
      "mean_token_accuracy": 0.6578903198242188,
      "num_tokens": 100546756.0,
      "step": 243
    },
    {
      "epoch": 0.4280701754385965,
      "grad_norm": 0.7767439484596252,
      "learning_rate": 0.0001,
      "loss": 1.2374,
      "mean_token_accuracy": 0.6534087061882019,
      "num_tokens": 100961732.0,
      "step": 244
    },
    {
      "epoch": 0.4298245614035088,
      "grad_norm": 0.7211782932281494,
      "learning_rate": 0.0001,
      "loss": 1.2627,
      "mean_token_accuracy": 0.6475375294685364,
      "num_tokens": 101396866.0,
      "step": 245
    },
    {
      "epoch": 0.43157894736842106,
      "grad_norm": 0.754098117351532,
      "learning_rate": 0.0001,
      "loss": 1.2535,
      "mean_token_accuracy": 0.6487017869949341,
      "num_tokens": 101823395.0,
      "step": 246
    },
    {
      "epoch": 0.43333333333333335,
      "grad_norm": 0.887698233127594,
      "learning_rate": 0.0001,
      "loss": 1.2372,
      "mean_token_accuracy": 0.6514610052108765,
      "num_tokens": 102218896.0,
      "step": 247
    },
    {
      "epoch": 0.43508771929824563,
      "grad_norm": 0.6688896417617798,
      "learning_rate": 0.0001,
      "loss": 1.2135,
      "mean_token_accuracy": 0.6570154428482056,
      "num_tokens": 102635943.0,
      "step": 248
    },
    {
      "epoch": 0.4368421052631579,
      "grad_norm": 0.6720183491706848,
      "learning_rate": 0.0001,
      "loss": 1.2314,
      "mean_token_accuracy": 0.653835654258728,
      "num_tokens": 103060426.0,
      "step": 249
    },
    {
      "epoch": 0.43859649122807015,
      "grad_norm": 0.6985954642295837,
      "learning_rate": 0.0001,
      "loss": 1.2302,
      "mean_token_accuracy": 0.6543055176734924,
      "num_tokens": 103480891.0,
      "step": 250
    },
    {
      "epoch": 0.44035087719298244,
      "grad_norm": 0.7861040234565735,
      "learning_rate": 0.0001,
      "loss": 1.2259,
      "mean_token_accuracy": 0.6543919444084167,
      "num_tokens": 103896368.0,
      "step": 251
    },
    {
      "epoch": 0.4421052631578947,
      "grad_norm": 0.7467155456542969,
      "learning_rate": 0.0001,
      "loss": 1.2425,
      "mean_token_accuracy": 0.6521680355072021,
      "num_tokens": 104318424.0,
      "step": 252
    },
    {
      "epoch": 0.443859649122807,
      "grad_norm": 0.689565896987915,
      "learning_rate": 0.0001,
      "loss": 1.2384,
      "mean_token_accuracy": 0.6529023051261902,
      "num_tokens": 104743917.0,
      "step": 253
    },
    {
      "epoch": 0.4456140350877193,
      "grad_norm": 0.8311668634414673,
      "learning_rate": 0.0001,
      "loss": 1.2286,
      "mean_token_accuracy": 0.6543080806732178,
      "num_tokens": 105146836.0,
      "step": 254
    },
    {
      "epoch": 0.4473684210526316,
      "grad_norm": 0.8047776818275452,
      "learning_rate": 0.0001,
      "loss": 1.2555,
      "mean_token_accuracy": 0.6480646133422852,
      "num_tokens": 105552302.0,
      "step": 255
    },
    {
      "epoch": 0.44912280701754387,
      "grad_norm": 0.6903892159461975,
      "learning_rate": 0.0001,
      "loss": 1.2878,
      "mean_token_accuracy": 0.6401119232177734,
      "num_tokens": 105963414.0,
      "step": 256
    },
    {
      "epoch": 0.45087719298245615,
      "grad_norm": 0.7000618577003479,
      "learning_rate": 0.0001,
      "loss": 1.2557,
      "mean_token_accuracy": 0.6474447846412659,
      "num_tokens": 106368736.0,
      "step": 257
    },
    {
      "epoch": 0.45263157894736844,
      "grad_norm": 0.7351795434951782,
      "learning_rate": 0.0001,
      "loss": 1.2237,
      "mean_token_accuracy": 0.6556580662727356,
      "num_tokens": 106769771.0,
      "step": 258
    },
    {
      "epoch": 0.4543859649122807,
      "grad_norm": 0.7257981300354004,
      "learning_rate": 0.0001,
      "loss": 1.2416,
      "mean_token_accuracy": 0.6521273255348206,
      "num_tokens": 107170029.0,
      "step": 259
    },
    {
      "epoch": 0.45614035087719296,
      "grad_norm": 0.8522328734397888,
      "learning_rate": 0.0001,
      "loss": 1.2301,
      "mean_token_accuracy": 0.6529244184494019,
      "num_tokens": 107576140.0,
      "step": 260
    },
    {
      "epoch": 0.45789473684210524,
      "grad_norm": 0.6672490835189819,
      "learning_rate": 0.0001,
      "loss": 1.2414,
      "mean_token_accuracy": 0.6505205631256104,
      "num_tokens": 108009067.0,
      "step": 261
    },
    {
      "epoch": 0.45964912280701753,
      "grad_norm": 0.8998327255249023,
      "learning_rate": 0.0001,
      "loss": 1.2365,
      "mean_token_accuracy": 0.6523317694664001,
      "num_tokens": 108434633.0,
      "step": 262
    },
    {
      "epoch": 0.4614035087719298,
      "grad_norm": 0.7883278727531433,
      "learning_rate": 0.0001,
      "loss": 1.2511,
      "mean_token_accuracy": 0.648801326751709,
      "num_tokens": 108878862.0,
      "step": 263
    },
    {
      "epoch": 0.4631578947368421,
      "grad_norm": 0.9719793796539307,
      "learning_rate": 0.0001,
      "loss": 1.2326,
      "mean_token_accuracy": 0.6538045406341553,
      "num_tokens": 109287222.0,
      "step": 264
    },
    {
      "epoch": 0.4649122807017544,
      "grad_norm": 0.6874752044677734,
      "learning_rate": 0.0001,
      "loss": 1.2319,
      "mean_token_accuracy": 0.6524173021316528,
      "num_tokens": 109693157.0,
      "step": 265
    },
    {
      "epoch": 0.4666666666666667,
      "grad_norm": 0.8174811601638794,
      "learning_rate": 0.0001,
      "loss": 1.2235,
      "mean_token_accuracy": 0.6545587182044983,
      "num_tokens": 110099662.0,
      "step": 266
    },
    {
      "epoch": 0.46842105263157896,
      "grad_norm": 0.7676987051963806,
      "learning_rate": 0.0001,
      "loss": 1.252,
      "mean_token_accuracy": 0.6494687795639038,
      "num_tokens": 110511759.0,
      "step": 267
    },
    {
      "epoch": 0.47017543859649125,
      "grad_norm": 0.7034929394721985,
      "learning_rate": 0.0001,
      "loss": 1.2328,
      "mean_token_accuracy": 0.653315544128418,
      "num_tokens": 110927626.0,
      "step": 268
    },
    {
      "epoch": 0.47192982456140353,
      "grad_norm": 0.6947440505027771,
      "learning_rate": 0.0001,
      "loss": 1.2451,
      "mean_token_accuracy": 0.6503375172615051,
      "num_tokens": 111371879.0,
      "step": 269
    },
    {
      "epoch": 0.47368421052631576,
      "grad_norm": 0.7659525871276855,
      "learning_rate": 0.0001,
      "loss": 1.251,
      "mean_token_accuracy": 0.6494304537773132,
      "num_tokens": 111784069.0,
      "step": 270
    },
    {
      "epoch": 0.47543859649122805,
      "grad_norm": 0.7740342617034912,
      "learning_rate": 0.0001,
      "loss": 1.2253,
      "mean_token_accuracy": 0.6542062759399414,
      "num_tokens": 112186870.0,
      "step": 271
    },
    {
      "epoch": 0.47719298245614034,
      "grad_norm": 0.65045565366745,
      "learning_rate": 0.0001,
      "loss": 1.2305,
      "mean_token_accuracy": 0.6532946228981018,
      "num_tokens": 112616585.0,
      "step": 272
    },
    {
      "epoch": 0.4789473684210526,
      "grad_norm": 0.7001651525497437,
      "learning_rate": 0.0001,
      "loss": 1.2247,
      "mean_token_accuracy": 0.654727041721344,
      "num_tokens": 113009798.0,
      "step": 273
    },
    {
      "epoch": 0.4807017543859649,
      "grad_norm": 0.6165850162506104,
      "learning_rate": 0.0001,
      "loss": 1.2174,
      "mean_token_accuracy": 0.6552860736846924,
      "num_tokens": 113410244.0,
      "step": 274
    },
    {
      "epoch": 0.4824561403508772,
      "grad_norm": 0.7424379587173462,
      "learning_rate": 0.0001,
      "loss": 1.2379,
      "mean_token_accuracy": 0.653006911277771,
      "num_tokens": 113805215.0,
      "step": 275
    },
    {
      "epoch": 0.4842105263157895,
      "grad_norm": 0.7236623167991638,
      "learning_rate": 0.0001,
      "loss": 1.2748,
      "mean_token_accuracy": 0.6438848972320557,
      "num_tokens": 114224914.0,
      "step": 276
    },
    {
      "epoch": 0.48596491228070177,
      "grad_norm": 0.6665499210357666,
      "learning_rate": 0.0001,
      "loss": 1.205,
      "mean_token_accuracy": 0.6590371131896973,
      "num_tokens": 114606381.0,
      "step": 277
    },
    {
      "epoch": 0.48771929824561405,
      "grad_norm": 0.6881427764892578,
      "learning_rate": 0.0001,
      "loss": 1.2456,
      "mean_token_accuracy": 0.6502711772918701,
      "num_tokens": 115021603.0,
      "step": 278
    },
    {
      "epoch": 0.48947368421052634,
      "grad_norm": 0.8498430848121643,
      "learning_rate": 0.0001,
      "loss": 1.2443,
      "mean_token_accuracy": 0.6494314670562744,
      "num_tokens": 115427417.0,
      "step": 279
    },
    {
      "epoch": 0.49122807017543857,
      "grad_norm": 0.724355936050415,
      "learning_rate": 0.0001,
      "loss": 1.2574,
      "mean_token_accuracy": 0.6479068398475647,
      "num_tokens": 115868301.0,
      "step": 280
    },
    {
      "epoch": 0.49298245614035086,
      "grad_norm": 0.6625252366065979,
      "learning_rate": 0.0001,
      "loss": 1.2128,
      "mean_token_accuracy": 0.6565937995910645,
      "num_tokens": 116278418.0,
      "step": 281
    },
    {
      "epoch": 0.49473684210526314,
      "grad_norm": 0.8329636454582214,
      "learning_rate": 0.0001,
      "loss": 1.2419,
      "mean_token_accuracy": 0.650545060634613,
      "num_tokens": 116681770.0,
      "step": 282
    },
    {
      "epoch": 0.4964912280701754,
      "grad_norm": 0.8298386335372925,
      "learning_rate": 0.0001,
      "loss": 1.2356,
      "mean_token_accuracy": 0.6532111167907715,
      "num_tokens": 117054940.0,
      "step": 283
    },
    {
      "epoch": 0.4982456140350877,
      "grad_norm": 0.7011889219284058,
      "learning_rate": 0.0001,
      "loss": 1.2418,
      "mean_token_accuracy": 0.6518392562866211,
      "num_tokens": 117477299.0,
      "step": 284
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.710082471370697,
      "learning_rate": 0.0001,
      "loss": 1.2479,
      "mean_token_accuracy": 0.6494768857955933,
      "num_tokens": 117885331.0,
      "step": 285
    },
    {
      "epoch": 0.5017543859649123,
      "grad_norm": 0.8371219038963318,
      "learning_rate": 0.0001,
      "loss": 1.2577,
      "mean_token_accuracy": 0.6471771001815796,
      "num_tokens": 118307149.0,
      "step": 286
    },
    {
      "epoch": 0.5035087719298246,
      "grad_norm": 0.8411158919334412,
      "learning_rate": 0.0001,
      "loss": 1.219,
      "mean_token_accuracy": 0.6554292440414429,
      "num_tokens": 118697262.0,
      "step": 287
    },
    {
      "epoch": 0.5052631578947369,
      "grad_norm": 0.7115722298622131,
      "learning_rate": 0.0001,
      "loss": 1.2501,
      "mean_token_accuracy": 0.6482685804367065,
      "num_tokens": 119124053.0,
      "step": 288
    },
    {
      "epoch": 0.5070175438596491,
      "grad_norm": 0.6575236916542053,
      "learning_rate": 0.0001,
      "loss": 1.2555,
      "mean_token_accuracy": 0.6482589840888977,
      "num_tokens": 119569185.0,
      "step": 289
    },
    {
      "epoch": 0.5087719298245614,
      "grad_norm": 0.8516756892204285,
      "learning_rate": 0.0001,
      "loss": 1.2332,
      "mean_token_accuracy": 0.6534713506698608,
      "num_tokens": 119987270.0,
      "step": 290
    },
    {
      "epoch": 0.5105263157894737,
      "grad_norm": 0.7346055507659912,
      "learning_rate": 0.0001,
      "loss": 1.2347,
      "mean_token_accuracy": 0.6515175700187683,
      "num_tokens": 120377177.0,
      "step": 291
    },
    {
      "epoch": 0.512280701754386,
      "grad_norm": 0.6637006402015686,
      "learning_rate": 0.0001,
      "loss": 1.2241,
      "mean_token_accuracy": 0.6554847359657288,
      "num_tokens": 120794750.0,
      "step": 292
    },
    {
      "epoch": 0.5140350877192983,
      "grad_norm": 0.8050562143325806,
      "learning_rate": 0.0001,
      "loss": 1.2462,
      "mean_token_accuracy": 0.6506670713424683,
      "num_tokens": 121223883.0,
      "step": 293
    },
    {
      "epoch": 0.5157894736842106,
      "grad_norm": 0.7059856057167053,
      "learning_rate": 0.0001,
      "loss": 1.2287,
      "mean_token_accuracy": 0.6533610820770264,
      "num_tokens": 121644108.0,
      "step": 294
    },
    {
      "epoch": 0.5175438596491229,
      "grad_norm": 0.6939064860343933,
      "learning_rate": 0.0001,
      "loss": 1.2357,
      "mean_token_accuracy": 0.6511529684066772,
      "num_tokens": 122051612.0,
      "step": 295
    },
    {
      "epoch": 0.519298245614035,
      "grad_norm": 0.8220492601394653,
      "learning_rate": 0.0001,
      "loss": 1.2257,
      "mean_token_accuracy": 0.6545118093490601,
      "num_tokens": 122430847.0,
      "step": 296
    },
    {
      "epoch": 0.5210526315789473,
      "grad_norm": 0.7044985294342041,
      "learning_rate": 0.0001,
      "loss": 1.2309,
      "mean_token_accuracy": 0.654273271560669,
      "num_tokens": 122849234.0,
      "step": 297
    },
    {
      "epoch": 0.5228070175438596,
      "grad_norm": 0.8146756291389465,
      "learning_rate": 0.0001,
      "loss": 1.2489,
      "mean_token_accuracy": 0.6501311659812927,
      "num_tokens": 123258290.0,
      "step": 298
    },
    {
      "epoch": 0.5245614035087719,
      "grad_norm": 0.766899824142456,
      "learning_rate": 0.0001,
      "loss": 1.2436,
      "mean_token_accuracy": 0.6506932973861694,
      "num_tokens": 123663735.0,
      "step": 299
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 0.7193543910980225,
      "learning_rate": 0.0001,
      "loss": 1.2451,
      "mean_token_accuracy": 0.6491553783416748,
      "num_tokens": 124058727.0,
      "step": 300
    },
    {
      "epoch": 0.5280701754385965,
      "grad_norm": 0.6504607200622559,
      "learning_rate": 0.0001,
      "loss": 1.2455,
      "mean_token_accuracy": 0.6509132385253906,
      "num_tokens": 124484021.0,
      "step": 301
    },
    {
      "epoch": 0.5298245614035088,
      "grad_norm": 0.7661638259887695,
      "learning_rate": 0.0001,
      "loss": 1.2158,
      "mean_token_accuracy": 0.6565036177635193,
      "num_tokens": 124902607.0,
      "step": 302
    },
    {
      "epoch": 0.531578947368421,
      "grad_norm": 0.73735511302948,
      "learning_rate": 0.0001,
      "loss": 1.2408,
      "mean_token_accuracy": 0.6503796577453613,
      "num_tokens": 125338175.0,
      "step": 303
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.9022007584571838,
      "learning_rate": 0.0001,
      "loss": 1.2332,
      "mean_token_accuracy": 0.6524848937988281,
      "num_tokens": 125757431.0,
      "step": 304
    },
    {
      "epoch": 0.5350877192982456,
      "grad_norm": 0.6961904764175415,
      "learning_rate": 0.0001,
      "loss": 1.2217,
      "mean_token_accuracy": 0.6560453176498413,
      "num_tokens": 126155529.0,
      "step": 305
    },
    {
      "epoch": 0.5368421052631579,
      "grad_norm": 0.6821785569190979,
      "learning_rate": 0.0001,
      "loss": 1.223,
      "mean_token_accuracy": 0.6551496982574463,
      "num_tokens": 126572012.0,
      "step": 306
    },
    {
      "epoch": 0.5385964912280702,
      "grad_norm": 0.8659482002258301,
      "learning_rate": 0.0001,
      "loss": 1.2393,
      "mean_token_accuracy": 0.6508756875991821,
      "num_tokens": 126990573.0,
      "step": 307
    },
    {
      "epoch": 0.5403508771929825,
      "grad_norm": 0.6646002531051636,
      "learning_rate": 0.0001,
      "loss": 1.2173,
      "mean_token_accuracy": 0.6561950445175171,
      "num_tokens": 127418976.0,
      "step": 308
    },
    {
      "epoch": 0.5421052631578948,
      "grad_norm": 0.6923218369483948,
      "learning_rate": 0.0001,
      "loss": 1.1911,
      "mean_token_accuracy": 0.6624599695205688,
      "num_tokens": 127827614.0,
      "step": 309
    },
    {
      "epoch": 0.543859649122807,
      "grad_norm": 0.6864442825317383,
      "learning_rate": 0.0001,
      "loss": 1.2267,
      "mean_token_accuracy": 0.6538540720939636,
      "num_tokens": 128258313.0,
      "step": 310
    },
    {
      "epoch": 0.5456140350877193,
      "grad_norm": 0.7230309247970581,
      "learning_rate": 0.0001,
      "loss": 1.2495,
      "mean_token_accuracy": 0.6486865282058716,
      "num_tokens": 128702682.0,
      "step": 311
    },
    {
      "epoch": 0.5473684210526316,
      "grad_norm": 0.6914284825325012,
      "learning_rate": 0.0001,
      "loss": 1.2211,
      "mean_token_accuracy": 0.6558956503868103,
      "num_tokens": 129094449.0,
      "step": 312
    },
    {
      "epoch": 0.5491228070175439,
      "grad_norm": 0.6948025822639465,
      "learning_rate": 0.0001,
      "loss": 1.2232,
      "mean_token_accuracy": 0.6554611325263977,
      "num_tokens": 129492795.0,
      "step": 313
    },
    {
      "epoch": 0.5508771929824562,
      "grad_norm": 0.6883065104484558,
      "learning_rate": 0.0001,
      "loss": 1.225,
      "mean_token_accuracy": 0.6549999713897705,
      "num_tokens": 129923670.0,
      "step": 314
    },
    {
      "epoch": 0.5526315789473685,
      "grad_norm": 0.7065843939781189,
      "learning_rate": 0.0001,
      "loss": 1.211,
      "mean_token_accuracy": 0.6596200466156006,
      "num_tokens": 130333012.0,
      "step": 315
    },
    {
      "epoch": 0.5543859649122806,
      "grad_norm": 0.8073469996452332,
      "learning_rate": 0.0001,
      "loss": 1.2073,
      "mean_token_accuracy": 0.6588984727859497,
      "num_tokens": 130749305.0,
      "step": 316
    },
    {
      "epoch": 0.5561403508771929,
      "grad_norm": 0.8134505748748779,
      "learning_rate": 0.0001,
      "loss": 1.2544,
      "mean_token_accuracy": 0.6476566791534424,
      "num_tokens": 131169421.0,
      "step": 317
    },
    {
      "epoch": 0.5578947368421052,
      "grad_norm": 0.6765173077583313,
      "learning_rate": 0.0001,
      "loss": 1.2078,
      "mean_token_accuracy": 0.6600826978683472,
      "num_tokens": 131574788.0,
      "step": 318
    },
    {
      "epoch": 0.5596491228070175,
      "grad_norm": 0.7156663537025452,
      "learning_rate": 0.0001,
      "loss": 1.2267,
      "mean_token_accuracy": 0.6545342206954956,
      "num_tokens": 131984459.0,
      "step": 319
    },
    {
      "epoch": 0.5614035087719298,
      "grad_norm": 0.8103324174880981,
      "learning_rate": 0.0001,
      "loss": 1.2612,
      "mean_token_accuracy": 0.6481289267539978,
      "num_tokens": 132415165.0,
      "step": 320
    },
    {
      "epoch": 0.5631578947368421,
      "grad_norm": 0.742142915725708,
      "learning_rate": 0.0001,
      "loss": 1.2039,
      "mean_token_accuracy": 0.6596871018409729,
      "num_tokens": 132838824.0,
      "step": 321
    },
    {
      "epoch": 0.5649122807017544,
      "grad_norm": 0.7613045573234558,
      "learning_rate": 0.0001,
      "loss": 1.2095,
      "mean_token_accuracy": 0.6572511792182922,
      "num_tokens": 133247120.0,
      "step": 322
    },
    {
      "epoch": 0.5666666666666667,
      "grad_norm": 0.7817480564117432,
      "learning_rate": 0.0001,
      "loss": 1.2232,
      "mean_token_accuracy": 0.6545703411102295,
      "num_tokens": 133672706.0,
      "step": 323
    },
    {
      "epoch": 0.5684210526315789,
      "grad_norm": 0.6124296188354492,
      "learning_rate": 0.0001,
      "loss": 1.2342,
      "mean_token_accuracy": 0.6552791595458984,
      "num_tokens": 134079103.0,
      "step": 324
    },
    {
      "epoch": 0.5701754385964912,
      "grad_norm": 0.6886869668960571,
      "learning_rate": 0.0001,
      "loss": 1.2231,
      "mean_token_accuracy": 0.6547764539718628,
      "num_tokens": 134494826.0,
      "step": 325
    },
    {
      "epoch": 0.5719298245614035,
      "grad_norm": 0.6630454659461975,
      "learning_rate": 0.0001,
      "loss": 1.2345,
      "mean_token_accuracy": 0.6520872116088867,
      "num_tokens": 134919504.0,
      "step": 326
    },
    {
      "epoch": 0.5736842105263158,
      "grad_norm": 0.8173869252204895,
      "learning_rate": 0.0001,
      "loss": 1.2276,
      "mean_token_accuracy": 0.6533281207084656,
      "num_tokens": 135331962.0,
      "step": 327
    },
    {
      "epoch": 0.5754385964912281,
      "grad_norm": 0.6743276715278625,
      "learning_rate": 0.0001,
      "loss": 1.2281,
      "mean_token_accuracy": 0.6535520553588867,
      "num_tokens": 135745710.0,
      "step": 328
    },
    {
      "epoch": 0.5771929824561404,
      "grad_norm": 0.6731691360473633,
      "learning_rate": 0.0001,
      "loss": 1.2346,
      "mean_token_accuracy": 0.6528003215789795,
      "num_tokens": 136174418.0,
      "step": 329
    },
    {
      "epoch": 0.5789473684210527,
      "grad_norm": 0.6211588382720947,
      "learning_rate": 0.0001,
      "loss": 1.2329,
      "mean_token_accuracy": 0.653133749961853,
      "num_tokens": 136597949.0,
      "step": 330
    },
    {
      "epoch": 0.5807017543859649,
      "grad_norm": 0.8585658073425293,
      "learning_rate": 0.0001,
      "loss": 1.2506,
      "mean_token_accuracy": 0.648890495300293,
      "num_tokens": 137047696.0,
      "step": 331
    },
    {
      "epoch": 0.5824561403508772,
      "grad_norm": 0.8006256222724915,
      "learning_rate": 0.0001,
      "loss": 1.2119,
      "mean_token_accuracy": 0.658997118473053,
      "num_tokens": 137428701.0,
      "step": 332
    },
    {
      "epoch": 0.5842105263157895,
      "grad_norm": 0.692973792552948,
      "learning_rate": 0.0001,
      "loss": 1.2167,
      "mean_token_accuracy": 0.6570533514022827,
      "num_tokens": 137823083.0,
      "step": 333
    },
    {
      "epoch": 0.5859649122807018,
      "grad_norm": 0.7685320973396301,
      "learning_rate": 0.0001,
      "loss": 1.271,
      "mean_token_accuracy": 0.6423413157463074,
      "num_tokens": 138250591.0,
      "step": 334
    },
    {
      "epoch": 0.5877192982456141,
      "grad_norm": 0.7700155377388,
      "learning_rate": 0.0001,
      "loss": 1.2252,
      "mean_token_accuracy": 0.6543079614639282,
      "num_tokens": 138660562.0,
      "step": 335
    },
    {
      "epoch": 0.5894736842105263,
      "grad_norm": 0.7410191893577576,
      "learning_rate": 0.0001,
      "loss": 1.2156,
      "mean_token_accuracy": 0.6561688184738159,
      "num_tokens": 139077597.0,
      "step": 336
    },
    {
      "epoch": 0.5912280701754385,
      "grad_norm": 0.7632637619972229,
      "learning_rate": 0.0001,
      "loss": 1.235,
      "mean_token_accuracy": 0.6516165733337402,
      "num_tokens": 139482698.0,
      "step": 337
    },
    {
      "epoch": 0.5929824561403508,
      "grad_norm": 0.690731942653656,
      "learning_rate": 0.0001,
      "loss": 1.2408,
      "mean_token_accuracy": 0.6502476930618286,
      "num_tokens": 139906177.0,
      "step": 338
    },
    {
      "epoch": 0.5947368421052631,
      "grad_norm": 0.6513046026229858,
      "learning_rate": 0.0001,
      "loss": 1.2297,
      "mean_token_accuracy": 0.6529406905174255,
      "num_tokens": 140319741.0,
      "step": 339
    },
    {
      "epoch": 0.5964912280701754,
      "grad_norm": 0.6879235506057739,
      "learning_rate": 0.0001,
      "loss": 1.2234,
      "mean_token_accuracy": 0.6540219783782959,
      "num_tokens": 140731809.0,
      "step": 340
    },
    {
      "epoch": 0.5982456140350877,
      "grad_norm": 0.7240639925003052,
      "learning_rate": 0.0001,
      "loss": 1.2043,
      "mean_token_accuracy": 0.6592921018600464,
      "num_tokens": 141126659.0,
      "step": 341
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.6559076905250549,
      "learning_rate": 0.0001,
      "loss": 1.2353,
      "mean_token_accuracy": 0.6514889001846313,
      "num_tokens": 141551619.0,
      "step": 342
    },
    {
      "epoch": 0.6017543859649123,
      "grad_norm": 0.7054679989814758,
      "learning_rate": 0.0001,
      "loss": 1.2225,
      "mean_token_accuracy": 0.6558979749679565,
      "num_tokens": 141970744.0,
      "step": 343
    },
    {
      "epoch": 0.6035087719298246,
      "grad_norm": 0.6867666244506836,
      "learning_rate": 0.0001,
      "loss": 1.2162,
      "mean_token_accuracy": 0.6566264629364014,
      "num_tokens": 142395401.0,
      "step": 344
    },
    {
      "epoch": 0.6052631578947368,
      "grad_norm": 0.6507348418235779,
      "learning_rate": 0.0001,
      "loss": 1.2244,
      "mean_token_accuracy": 0.6545271277427673,
      "num_tokens": 142827910.0,
      "step": 345
    },
    {
      "epoch": 0.6070175438596491,
      "grad_norm": 0.7520820498466492,
      "learning_rate": 0.0001,
      "loss": 1.2315,
      "mean_token_accuracy": 0.6522303819656372,
      "num_tokens": 143256693.0,
      "step": 346
    },
    {
      "epoch": 0.6087719298245614,
      "grad_norm": 0.7250421047210693,
      "learning_rate": 0.0001,
      "loss": 1.2126,
      "mean_token_accuracy": 0.6569870710372925,
      "num_tokens": 143670113.0,
      "step": 347
    },
    {
      "epoch": 0.6105263157894737,
      "grad_norm": 0.707240104675293,
      "learning_rate": 0.0001,
      "loss": 1.2337,
      "mean_token_accuracy": 0.6521450281143188,
      "num_tokens": 144085012.0,
      "step": 348
    },
    {
      "epoch": 0.612280701754386,
      "grad_norm": 0.6530799269676208,
      "learning_rate": 0.0001,
      "loss": 1.2366,
      "mean_token_accuracy": 0.650078296661377,
      "num_tokens": 144511304.0,
      "step": 349
    },
    {
      "epoch": 0.6140350877192983,
      "grad_norm": 0.7164869904518127,
      "learning_rate": 0.0001,
      "loss": 1.1957,
      "mean_token_accuracy": 0.6614063382148743,
      "num_tokens": 144903389.0,
      "step": 350
    },
    {
      "epoch": 0.6157894736842106,
      "grad_norm": 0.6941936612129211,
      "learning_rate": 0.0001,
      "loss": 1.2244,
      "mean_token_accuracy": 0.6538809537887573,
      "num_tokens": 145317796.0,
      "step": 351
    },
    {
      "epoch": 0.6175438596491228,
      "grad_norm": 0.5569853186607361,
      "learning_rate": 0.0001,
      "loss": 1.2303,
      "mean_token_accuracy": 0.6540141701698303,
      "num_tokens": 145728692.0,
      "step": 352
    },
    {
      "epoch": 0.6192982456140351,
      "grad_norm": 0.6453179121017456,
      "learning_rate": 0.0001,
      "loss": 1.2219,
      "mean_token_accuracy": 0.6558411121368408,
      "num_tokens": 146149428.0,
      "step": 353
    },
    {
      "epoch": 0.6210526315789474,
      "grad_norm": 0.7571195363998413,
      "learning_rate": 0.0001,
      "loss": 1.2371,
      "mean_token_accuracy": 0.6512084007263184,
      "num_tokens": 146567803.0,
      "step": 354
    },
    {
      "epoch": 0.6228070175438597,
      "grad_norm": 0.7026142477989197,
      "learning_rate": 0.0001,
      "loss": 1.2577,
      "mean_token_accuracy": 0.6465089917182922,
      "num_tokens": 146996287.0,
      "step": 355
    },
    {
      "epoch": 0.624561403508772,
      "grad_norm": 0.7396862506866455,
      "learning_rate": 0.0001,
      "loss": 1.207,
      "mean_token_accuracy": 0.657769501209259,
      "num_tokens": 147406032.0,
      "step": 356
    },
    {
      "epoch": 0.6263157894736842,
      "grad_norm": 0.7301826477050781,
      "learning_rate": 0.0001,
      "loss": 1.2549,
      "mean_token_accuracy": 0.648101270198822,
      "num_tokens": 147818711.0,
      "step": 357
    },
    {
      "epoch": 0.6280701754385964,
      "grad_norm": 0.6443963646888733,
      "learning_rate": 0.0001,
      "loss": 1.2247,
      "mean_token_accuracy": 0.6540451645851135,
      "num_tokens": 148224782.0,
      "step": 358
    },
    {
      "epoch": 0.6298245614035087,
      "grad_norm": 0.5962257981300354,
      "learning_rate": 0.0001,
      "loss": 1.2095,
      "mean_token_accuracy": 0.6578296422958374,
      "num_tokens": 148638041.0,
      "step": 359
    },
    {
      "epoch": 0.631578947368421,
      "grad_norm": 0.553277850151062,
      "learning_rate": 0.0001,
      "loss": 1.1954,
      "mean_token_accuracy": 0.6618661880493164,
      "num_tokens": 149034209.0,
      "step": 360
    },
    {
      "epoch": 0.6333333333333333,
      "grad_norm": 0.8141903281211853,
      "learning_rate": 0.0001,
      "loss": 1.2181,
      "mean_token_accuracy": 0.6552863121032715,
      "num_tokens": 149462197.0,
      "step": 361
    },
    {
      "epoch": 0.6350877192982456,
      "grad_norm": 0.6312337517738342,
      "learning_rate": 0.0001,
      "loss": 1.2237,
      "mean_token_accuracy": 0.6552015542984009,
      "num_tokens": 149874309.0,
      "step": 362
    },
    {
      "epoch": 0.6368421052631579,
      "grad_norm": 0.6863110661506653,
      "learning_rate": 0.0001,
      "loss": 1.2144,
      "mean_token_accuracy": 0.6574706435203552,
      "num_tokens": 150278354.0,
      "step": 363
    },
    {
      "epoch": 0.6385964912280702,
      "grad_norm": 0.7062144875526428,
      "learning_rate": 0.0001,
      "loss": 1.2309,
      "mean_token_accuracy": 0.6519078016281128,
      "num_tokens": 150696393.0,
      "step": 364
    },
    {
      "epoch": 0.6403508771929824,
      "grad_norm": 0.6141137480735779,
      "learning_rate": 0.0001,
      "loss": 1.2175,
      "mean_token_accuracy": 0.6562414169311523,
      "num_tokens": 151097966.0,
      "step": 365
    },
    {
      "epoch": 0.6421052631578947,
      "grad_norm": 0.6939074993133545,
      "learning_rate": 0.0001,
      "loss": 1.2115,
      "mean_token_accuracy": 0.6587799787521362,
      "num_tokens": 151487261.0,
      "step": 366
    },
    {
      "epoch": 0.643859649122807,
      "grad_norm": 0.6834867596626282,
      "learning_rate": 0.0001,
      "loss": 1.2492,
      "mean_token_accuracy": 0.6492800116539001,
      "num_tokens": 151915716.0,
      "step": 367
    },
    {
      "epoch": 0.6456140350877193,
      "grad_norm": 0.6845062971115112,
      "learning_rate": 0.0001,
      "loss": 1.2045,
      "mean_token_accuracy": 0.6578772664070129,
      "num_tokens": 152333716.0,
      "step": 368
    },
    {
      "epoch": 0.6473684210526316,
      "grad_norm": 0.6263954639434814,
      "learning_rate": 0.0001,
      "loss": 1.2285,
      "mean_token_accuracy": 0.6524069905281067,
      "num_tokens": 152758004.0,
      "step": 369
    },
    {
      "epoch": 0.6491228070175439,
      "grad_norm": 0.7604780793190002,
      "learning_rate": 0.0001,
      "loss": 1.2321,
      "mean_token_accuracy": 0.6509186029434204,
      "num_tokens": 153175726.0,
      "step": 370
    },
    {
      "epoch": 0.6508771929824562,
      "grad_norm": 0.6607220768928528,
      "learning_rate": 0.0001,
      "loss": 1.193,
      "mean_token_accuracy": 0.6612952947616577,
      "num_tokens": 153573526.0,
      "step": 371
    },
    {
      "epoch": 0.6526315789473685,
      "grad_norm": 0.7317623496055603,
      "learning_rate": 0.0001,
      "loss": 1.2233,
      "mean_token_accuracy": 0.6555420756340027,
      "num_tokens": 154001303.0,
      "step": 372
    },
    {
      "epoch": 0.6543859649122807,
      "grad_norm": 0.5643908381462097,
      "learning_rate": 0.0001,
      "loss": 1.1888,
      "mean_token_accuracy": 0.6617960929870605,
      "num_tokens": 154409874.0,
      "step": 373
    },
    {
      "epoch": 0.656140350877193,
      "grad_norm": 0.631582498550415,
      "learning_rate": 0.0001,
      "loss": 1.2192,
      "mean_token_accuracy": 0.6558030843734741,
      "num_tokens": 154826554.0,
      "step": 374
    },
    {
      "epoch": 0.6578947368421053,
      "grad_norm": 0.745689332485199,
      "learning_rate": 0.0001,
      "loss": 1.2146,
      "mean_token_accuracy": 0.6560918688774109,
      "num_tokens": 155230585.0,
      "step": 375
    },
    {
      "epoch": 0.6596491228070176,
      "grad_norm": 0.651474118232727,
      "learning_rate": 0.0001,
      "loss": 1.2055,
      "mean_token_accuracy": 0.6597638726234436,
      "num_tokens": 155636974.0,
      "step": 376
    },
    {
      "epoch": 0.6614035087719298,
      "grad_norm": 0.7227398753166199,
      "learning_rate": 0.0001,
      "loss": 1.2211,
      "mean_token_accuracy": 0.6551980972290039,
      "num_tokens": 156063068.0,
      "step": 377
    },
    {
      "epoch": 0.6631578947368421,
      "grad_norm": 0.6124153137207031,
      "learning_rate": 0.0001,
      "loss": 1.2435,
      "mean_token_accuracy": 0.6511872410774231,
      "num_tokens": 156510643.0,
      "step": 378
    },
    {
      "epoch": 0.6649122807017543,
      "grad_norm": 0.7193928360939026,
      "learning_rate": 0.0001,
      "loss": 1.2242,
      "mean_token_accuracy": 0.6536823511123657,
      "num_tokens": 156947423.0,
      "step": 379
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.7923741936683655,
      "learning_rate": 0.0001,
      "loss": 1.2149,
      "mean_token_accuracy": 0.6561374664306641,
      "num_tokens": 157370426.0,
      "step": 380
    },
    {
      "epoch": 0.6684210526315789,
      "grad_norm": 0.7290387153625488,
      "learning_rate": 0.0001,
      "loss": 1.2132,
      "mean_token_accuracy": 0.6568667888641357,
      "num_tokens": 157782963.0,
      "step": 381
    },
    {
      "epoch": 0.6701754385964912,
      "grad_norm": 0.6192464232444763,
      "learning_rate": 0.0001,
      "loss": 1.2309,
      "mean_token_accuracy": 0.6526235342025757,
      "num_tokens": 158180387.0,
      "step": 382
    },
    {
      "epoch": 0.6719298245614035,
      "grad_norm": 0.7137374877929688,
      "learning_rate": 0.0001,
      "loss": 1.2328,
      "mean_token_accuracy": 0.6518644094467163,
      "num_tokens": 158582401.0,
      "step": 383
    },
    {
      "epoch": 0.6736842105263158,
      "grad_norm": 0.7550848126411438,
      "learning_rate": 0.0001,
      "loss": 1.2166,
      "mean_token_accuracy": 0.6573696136474609,
      "num_tokens": 158991919.0,
      "step": 384
    },
    {
      "epoch": 0.6754385964912281,
      "grad_norm": 0.6890254020690918,
      "learning_rate": 0.0001,
      "loss": 1.2107,
      "mean_token_accuracy": 0.6569054126739502,
      "num_tokens": 159393180.0,
      "step": 385
    },
    {
      "epoch": 0.6771929824561403,
      "grad_norm": 0.7258317470550537,
      "learning_rate": 0.0001,
      "loss": 1.2151,
      "mean_token_accuracy": 0.6549711227416992,
      "num_tokens": 159800687.0,
      "step": 386
    },
    {
      "epoch": 0.6789473684210526,
      "grad_norm": 0.7973881363868713,
      "learning_rate": 0.0001,
      "loss": 1.1913,
      "mean_token_accuracy": 0.6616858839988708,
      "num_tokens": 160206677.0,
      "step": 387
    },
    {
      "epoch": 0.6807017543859649,
      "grad_norm": 0.6781461238861084,
      "learning_rate": 0.0001,
      "loss": 1.2296,
      "mean_token_accuracy": 0.6531736850738525,
      "num_tokens": 160623973.0,
      "step": 388
    },
    {
      "epoch": 0.6824561403508772,
      "grad_norm": 0.8034713268280029,
      "learning_rate": 0.0001,
      "loss": 1.2306,
      "mean_token_accuracy": 0.6528879404067993,
      "num_tokens": 161033575.0,
      "step": 389
    },
    {
      "epoch": 0.6842105263157895,
      "grad_norm": 0.7085846066474915,
      "learning_rate": 0.0001,
      "loss": 1.1892,
      "mean_token_accuracy": 0.6614021062850952,
      "num_tokens": 161420334.0,
      "step": 390
    },
    {
      "epoch": 0.6859649122807018,
      "grad_norm": 0.712842583656311,
      "learning_rate": 0.0001,
      "loss": 1.2096,
      "mean_token_accuracy": 0.6570684909820557,
      "num_tokens": 161823214.0,
      "step": 391
    },
    {
      "epoch": 0.6877192982456141,
      "grad_norm": 0.6031337380409241,
      "learning_rate": 0.0001,
      "loss": 1.2131,
      "mean_token_accuracy": 0.6579930782318115,
      "num_tokens": 162253222.0,
      "step": 392
    },
    {
      "epoch": 0.6894736842105263,
      "grad_norm": 0.6571363806724548,
      "learning_rate": 0.0001,
      "loss": 1.2151,
      "mean_token_accuracy": 0.6550261974334717,
      "num_tokens": 162673396.0,
      "step": 393
    },
    {
      "epoch": 0.6912280701754386,
      "grad_norm": 0.590053915977478,
      "learning_rate": 0.0001,
      "loss": 1.1913,
      "mean_token_accuracy": 0.6606940031051636,
      "num_tokens": 163095701.0,
      "step": 394
    },
    {
      "epoch": 0.6929824561403509,
      "grad_norm": 0.660569429397583,
      "learning_rate": 0.0001,
      "loss": 1.2168,
      "mean_token_accuracy": 0.6552713513374329,
      "num_tokens": 163503487.0,
      "step": 395
    },
    {
      "epoch": 0.6947368421052632,
      "grad_norm": 0.5482744574546814,
      "learning_rate": 0.0001,
      "loss": 1.1966,
      "mean_token_accuracy": 0.6622109413146973,
      "num_tokens": 163908638.0,
      "step": 396
    },
    {
      "epoch": 0.6964912280701754,
      "grad_norm": 0.6649277210235596,
      "learning_rate": 0.0001,
      "loss": 1.2082,
      "mean_token_accuracy": 0.6560900211334229,
      "num_tokens": 164321664.0,
      "step": 397
    },
    {
      "epoch": 0.6982456140350877,
      "grad_norm": 0.6546705365180969,
      "learning_rate": 0.0001,
      "loss": 1.208,
      "mean_token_accuracy": 0.6577179431915283,
      "num_tokens": 164739198.0,
      "step": 398
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.6374883651733398,
      "learning_rate": 0.0001,
      "loss": 1.1893,
      "mean_token_accuracy": 0.660727322101593,
      "num_tokens": 165130707.0,
      "step": 399
    },
    {
      "epoch": 0.7017543859649122,
      "grad_norm": 0.6626867055892944,
      "learning_rate": 0.0001,
      "loss": 1.2056,
      "mean_token_accuracy": 0.6570228934288025,
      "num_tokens": 165544679.0,
      "step": 400
    },
    {
      "epoch": 0.7035087719298245,
      "grad_norm": 0.648720920085907,
      "learning_rate": 0.0001,
      "loss": 1.1889,
      "mean_token_accuracy": 0.6612677574157715,
      "num_tokens": 165963969.0,
      "step": 401
    },
    {
      "epoch": 0.7052631578947368,
      "grad_norm": 0.6660583019256592,
      "learning_rate": 0.0001,
      "loss": 1.201,
      "mean_token_accuracy": 0.6595137119293213,
      "num_tokens": 166342479.0,
      "step": 402
    },
    {
      "epoch": 0.7070175438596491,
      "grad_norm": 0.6676925420761108,
      "learning_rate": 0.0001,
      "loss": 1.2178,
      "mean_token_accuracy": 0.655318558216095,
      "num_tokens": 166746524.0,
      "step": 403
    },
    {
      "epoch": 0.7087719298245614,
      "grad_norm": 0.6398362517356873,
      "learning_rate": 0.0001,
      "loss": 1.2217,
      "mean_token_accuracy": 0.6538881063461304,
      "num_tokens": 167166144.0,
      "step": 404
    },
    {
      "epoch": 0.7105263157894737,
      "grad_norm": 0.6486631035804749,
      "learning_rate": 0.0001,
      "loss": 1.2107,
      "mean_token_accuracy": 0.6568524837493896,
      "num_tokens": 167576053.0,
      "step": 405
    },
    {
      "epoch": 0.712280701754386,
      "grad_norm": 0.6971449851989746,
      "learning_rate": 0.0001,
      "loss": 1.2072,
      "mean_token_accuracy": 0.6588019132614136,
      "num_tokens": 168005716.0,
      "step": 406
    },
    {
      "epoch": 0.7140350877192982,
      "grad_norm": 0.5594667792320251,
      "learning_rate": 0.0001,
      "loss": 1.1815,
      "mean_token_accuracy": 0.6640152931213379,
      "num_tokens": 168425061.0,
      "step": 407
    },
    {
      "epoch": 0.7157894736842105,
      "grad_norm": 0.6978932619094849,
      "learning_rate": 0.0001,
      "loss": 1.2123,
      "mean_token_accuracy": 0.6560491323471069,
      "num_tokens": 168832540.0,
      "step": 408
    },
    {
      "epoch": 0.7175438596491228,
      "grad_norm": 0.577872097492218,
      "learning_rate": 0.0001,
      "loss": 1.1961,
      "mean_token_accuracy": 0.6605911254882812,
      "num_tokens": 169243355.0,
      "step": 409
    },
    {
      "epoch": 0.7192982456140351,
      "grad_norm": 0.6972746849060059,
      "learning_rate": 0.0001,
      "loss": 1.2263,
      "mean_token_accuracy": 0.6549021005630493,
      "num_tokens": 169678758.0,
      "step": 410
    },
    {
      "epoch": 0.7210526315789474,
      "grad_norm": 0.6528338193893433,
      "learning_rate": 0.0001,
      "loss": 1.2193,
      "mean_token_accuracy": 0.6543501615524292,
      "num_tokens": 170107843.0,
      "step": 411
    },
    {
      "epoch": 0.7228070175438597,
      "grad_norm": 0.6352643370628357,
      "learning_rate": 0.0001,
      "loss": 1.21,
      "mean_token_accuracy": 0.6568456292152405,
      "num_tokens": 170512414.0,
      "step": 412
    },
    {
      "epoch": 0.724561403508772,
      "grad_norm": 0.6633725762367249,
      "learning_rate": 0.0001,
      "loss": 1.2273,
      "mean_token_accuracy": 0.6531171798706055,
      "num_tokens": 170927891.0,
      "step": 413
    },
    {
      "epoch": 0.7263157894736842,
      "grad_norm": 0.7003793716430664,
      "learning_rate": 0.0001,
      "loss": 1.2471,
      "mean_token_accuracy": 0.6485984921455383,
      "num_tokens": 171347602.0,
      "step": 414
    },
    {
      "epoch": 0.7280701754385965,
      "grad_norm": 0.6166436076164246,
      "learning_rate": 0.0001,
      "loss": 1.1822,
      "mean_token_accuracy": 0.664222240447998,
      "num_tokens": 171764325.0,
      "step": 415
    },
    {
      "epoch": 0.7298245614035088,
      "grad_norm": 0.6370410323143005,
      "learning_rate": 0.0001,
      "loss": 1.2288,
      "mean_token_accuracy": 0.6530359387397766,
      "num_tokens": 172161316.0,
      "step": 416
    },
    {
      "epoch": 0.7315789473684211,
      "grad_norm": 0.5680028200149536,
      "learning_rate": 0.0001,
      "loss": 1.188,
      "mean_token_accuracy": 0.663171112537384,
      "num_tokens": 172557979.0,
      "step": 417
    },
    {
      "epoch": 0.7333333333333333,
      "grad_norm": 0.6317917704582214,
      "learning_rate": 0.0001,
      "loss": 1.2088,
      "mean_token_accuracy": 0.6587448120117188,
      "num_tokens": 172977293.0,
      "step": 418
    },
    {
      "epoch": 0.7350877192982456,
      "grad_norm": 0.6629990935325623,
      "learning_rate": 0.0001,
      "loss": 1.206,
      "mean_token_accuracy": 0.657719612121582,
      "num_tokens": 173386905.0,
      "step": 419
    },
    {
      "epoch": 0.7368421052631579,
      "grad_norm": 0.7318717241287231,
      "learning_rate": 0.0001,
      "loss": 1.1874,
      "mean_token_accuracy": 0.662236750125885,
      "num_tokens": 173790039.0,
      "step": 420
    },
    {
      "epoch": 0.7385964912280701,
      "grad_norm": 0.5909295678138733,
      "learning_rate": 0.0001,
      "loss": 1.1857,
      "mean_token_accuracy": 0.66287761926651,
      "num_tokens": 174200214.0,
      "step": 421
    },
    {
      "epoch": 0.7403508771929824,
      "grad_norm": 0.7244629859924316,
      "learning_rate": 0.0001,
      "loss": 1.198,
      "mean_token_accuracy": 0.6586729288101196,
      "num_tokens": 174594155.0,
      "step": 422
    },
    {
      "epoch": 0.7421052631578947,
      "grad_norm": 0.7065144777297974,
      "learning_rate": 0.0001,
      "loss": 1.189,
      "mean_token_accuracy": 0.6611475348472595,
      "num_tokens": 175025672.0,
      "step": 423
    },
    {
      "epoch": 0.743859649122807,
      "grad_norm": 0.6348630785942078,
      "learning_rate": 0.0001,
      "loss": 1.2285,
      "mean_token_accuracy": 0.6518391370773315,
      "num_tokens": 175452636.0,
      "step": 424
    },
    {
      "epoch": 0.7456140350877193,
      "grad_norm": 0.6401616930961609,
      "learning_rate": 0.0001,
      "loss": 1.2204,
      "mean_token_accuracy": 0.655087947845459,
      "num_tokens": 175879906.0,
      "step": 425
    },
    {
      "epoch": 0.7473684210526316,
      "grad_norm": 0.6971575617790222,
      "learning_rate": 0.0001,
      "loss": 1.2111,
      "mean_token_accuracy": 0.6562093496322632,
      "num_tokens": 176292162.0,
      "step": 426
    },
    {
      "epoch": 0.7491228070175439,
      "grad_norm": 0.6440587043762207,
      "learning_rate": 0.0001,
      "loss": 1.2012,
      "mean_token_accuracy": 0.6593471765518188,
      "num_tokens": 176720725.0,
      "step": 427
    },
    {
      "epoch": 0.7508771929824561,
      "grad_norm": 0.597520649433136,
      "learning_rate": 0.0001,
      "loss": 1.2284,
      "mean_token_accuracy": 0.6531672477722168,
      "num_tokens": 177161243.0,
      "step": 428
    },
    {
      "epoch": 0.7526315789473684,
      "grad_norm": 0.8046004772186279,
      "learning_rate": 0.0001,
      "loss": 1.1928,
      "mean_token_accuracy": 0.6594202518463135,
      "num_tokens": 177562007.0,
      "step": 429
    },
    {
      "epoch": 0.7543859649122807,
      "grad_norm": 0.6298813223838806,
      "learning_rate": 0.0001,
      "loss": 1.219,
      "mean_token_accuracy": 0.6546623706817627,
      "num_tokens": 177983323.0,
      "step": 430
    },
    {
      "epoch": 0.756140350877193,
      "grad_norm": 0.5731974840164185,
      "learning_rate": 0.0001,
      "loss": 1.2153,
      "mean_token_accuracy": 0.6574859619140625,
      "num_tokens": 178409419.0,
      "step": 431
    },
    {
      "epoch": 0.7578947368421053,
      "grad_norm": 0.7396548390388489,
      "learning_rate": 0.0001,
      "loss": 1.2113,
      "mean_token_accuracy": 0.6574329137802124,
      "num_tokens": 178832025.0,
      "step": 432
    },
    {
      "epoch": 0.7596491228070176,
      "grad_norm": 0.6398889422416687,
      "learning_rate": 0.0001,
      "loss": 1.2159,
      "mean_token_accuracy": 0.6554309129714966,
      "num_tokens": 179246477.0,
      "step": 433
    },
    {
      "epoch": 0.7614035087719299,
      "grad_norm": 0.6085229516029358,
      "learning_rate": 0.0001,
      "loss": 1.2216,
      "mean_token_accuracy": 0.6549739837646484,
      "num_tokens": 179666041.0,
      "step": 434
    },
    {
      "epoch": 0.7631578947368421,
      "grad_norm": 0.7816640734672546,
      "learning_rate": 0.0001,
      "loss": 1.2119,
      "mean_token_accuracy": 0.6565245389938354,
      "num_tokens": 180092225.0,
      "step": 435
    },
    {
      "epoch": 0.7649122807017544,
      "grad_norm": 0.8083506226539612,
      "learning_rate": 0.0001,
      "loss": 1.1961,
      "mean_token_accuracy": 0.6606351733207703,
      "num_tokens": 180498123.0,
      "step": 436
    },
    {
      "epoch": 0.7666666666666667,
      "grad_norm": 0.6019986271858215,
      "learning_rate": 0.0001,
      "loss": 1.1972,
      "mean_token_accuracy": 0.6611742377281189,
      "num_tokens": 180904735.0,
      "step": 437
    },
    {
      "epoch": 0.7684210526315789,
      "grad_norm": 0.6621778011322021,
      "learning_rate": 0.0001,
      "loss": 1.1987,
      "mean_token_accuracy": 0.6592767238616943,
      "num_tokens": 181297827.0,
      "step": 438
    },
    {
      "epoch": 0.7701754385964912,
      "grad_norm": 0.5817862749099731,
      "learning_rate": 0.0001,
      "loss": 1.2096,
      "mean_token_accuracy": 0.6584010124206543,
      "num_tokens": 181722096.0,
      "step": 439
    },
    {
      "epoch": 0.7719298245614035,
      "grad_norm": 0.6433981657028198,
      "learning_rate": 0.0001,
      "loss": 1.1872,
      "mean_token_accuracy": 0.6626437902450562,
      "num_tokens": 182124515.0,
      "step": 440
    },
    {
      "epoch": 0.7736842105263158,
      "grad_norm": 0.6573434472084045,
      "learning_rate": 0.0001,
      "loss": 1.2159,
      "mean_token_accuracy": 0.6551169157028198,
      "num_tokens": 182542265.0,
      "step": 441
    },
    {
      "epoch": 0.775438596491228,
      "grad_norm": 0.684744656085968,
      "learning_rate": 0.0001,
      "loss": 1.2095,
      "mean_token_accuracy": 0.6574633121490479,
      "num_tokens": 182942368.0,
      "step": 442
    },
    {
      "epoch": 0.7771929824561403,
      "grad_norm": 0.5961515307426453,
      "learning_rate": 0.0001,
      "loss": 1.2012,
      "mean_token_accuracy": 0.6580845713615417,
      "num_tokens": 183350652.0,
      "step": 443
    },
    {
      "epoch": 0.7789473684210526,
      "grad_norm": 0.7268422842025757,
      "learning_rate": 0.0001,
      "loss": 1.2082,
      "mean_token_accuracy": 0.657654345035553,
      "num_tokens": 183786290.0,
      "step": 444
    },
    {
      "epoch": 0.7807017543859649,
      "grad_norm": 0.7548661231994629,
      "learning_rate": 0.0001,
      "loss": 1.203,
      "mean_token_accuracy": 0.6581261157989502,
      "num_tokens": 184191985.0,
      "step": 445
    },
    {
      "epoch": 0.7824561403508772,
      "grad_norm": 0.589838981628418,
      "learning_rate": 0.0001,
      "loss": 1.2253,
      "mean_token_accuracy": 0.652956485748291,
      "num_tokens": 184617087.0,
      "step": 446
    },
    {
      "epoch": 0.7842105263157895,
      "grad_norm": 0.7901304960250854,
      "learning_rate": 0.0001,
      "loss": 1.2023,
      "mean_token_accuracy": 0.6594702005386353,
      "num_tokens": 185046113.0,
      "step": 447
    },
    {
      "epoch": 0.7859649122807018,
      "grad_norm": 0.681577205657959,
      "learning_rate": 0.0001,
      "loss": 1.1765,
      "mean_token_accuracy": 0.6648210287094116,
      "num_tokens": 185440210.0,
      "step": 448
    },
    {
      "epoch": 0.787719298245614,
      "grad_norm": 0.619105339050293,
      "learning_rate": 0.0001,
      "loss": 1.2151,
      "mean_token_accuracy": 0.6544394493103027,
      "num_tokens": 185866240.0,
      "step": 449
    },
    {
      "epoch": 0.7894736842105263,
      "grad_norm": 0.6568613648414612,
      "learning_rate": 0.0001,
      "loss": 1.1808,
      "mean_token_accuracy": 0.6645166277885437,
      "num_tokens": 186262559.0,
      "step": 450
    },
    {
      "epoch": 0.7912280701754386,
      "grad_norm": 0.6452411413192749,
      "learning_rate": 0.0001,
      "loss": 1.2289,
      "mean_token_accuracy": 0.6531310677528381,
      "num_tokens": 186677017.0,
      "step": 451
    },
    {
      "epoch": 0.7929824561403509,
      "grad_norm": 0.6799737215042114,
      "learning_rate": 0.0001,
      "loss": 1.2207,
      "mean_token_accuracy": 0.6556583046913147,
      "num_tokens": 187108135.0,
      "step": 452
    },
    {
      "epoch": 0.7947368421052632,
      "grad_norm": 0.5680040717124939,
      "learning_rate": 0.0001,
      "loss": 1.1886,
      "mean_token_accuracy": 0.6613532900810242,
      "num_tokens": 187533913.0,
      "step": 453
    },
    {
      "epoch": 0.7964912280701755,
      "grad_norm": 0.6380943655967712,
      "learning_rate": 0.0001,
      "loss": 1.2136,
      "mean_token_accuracy": 0.6577192544937134,
      "num_tokens": 187943777.0,
      "step": 454
    },
    {
      "epoch": 0.7982456140350878,
      "grad_norm": 0.5565281510353088,
      "learning_rate": 0.0001,
      "loss": 1.1941,
      "mean_token_accuracy": 0.6604122519493103,
      "num_tokens": 188365013.0,
      "step": 455
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.6176914572715759,
      "learning_rate": 0.0001,
      "loss": 1.1957,
      "mean_token_accuracy": 0.6601607799530029,
      "num_tokens": 188805546.0,
      "step": 456
    },
    {
      "epoch": 0.8017543859649123,
      "grad_norm": 0.6163376569747925,
      "learning_rate": 0.0001,
      "loss": 1.1788,
      "mean_token_accuracy": 0.6634429097175598,
      "num_tokens": 189204261.0,
      "step": 457
    },
    {
      "epoch": 0.8035087719298246,
      "grad_norm": 0.6874009966850281,
      "learning_rate": 0.0001,
      "loss": 1.2061,
      "mean_token_accuracy": 0.6587145924568176,
      "num_tokens": 189609457.0,
      "step": 458
    },
    {
      "epoch": 0.8052631578947368,
      "grad_norm": 0.6584733724594116,
      "learning_rate": 0.0001,
      "loss": 1.2077,
      "mean_token_accuracy": 0.6568840742111206,
      "num_tokens": 190024904.0,
      "step": 459
    },
    {
      "epoch": 0.8070175438596491,
      "grad_norm": 0.554511547088623,
      "learning_rate": 0.0001,
      "loss": 1.1883,
      "mean_token_accuracy": 0.661857008934021,
      "num_tokens": 190435843.0,
      "step": 460
    },
    {
      "epoch": 0.8087719298245614,
      "grad_norm": 0.6625659465789795,
      "learning_rate": 0.0001,
      "loss": 1.209,
      "mean_token_accuracy": 0.6570659875869751,
      "num_tokens": 190844879.0,
      "step": 461
    },
    {
      "epoch": 0.8105263157894737,
      "grad_norm": 0.6230789422988892,
      "learning_rate": 0.0001,
      "loss": 1.1932,
      "mean_token_accuracy": 0.6605242490768433,
      "num_tokens": 191240668.0,
      "step": 462
    },
    {
      "epoch": 0.8122807017543859,
      "grad_norm": 0.5848865509033203,
      "learning_rate": 0.0001,
      "loss": 1.2055,
      "mean_token_accuracy": 0.6577123999595642,
      "num_tokens": 191649912.0,
      "step": 463
    },
    {
      "epoch": 0.8140350877192982,
      "grad_norm": 0.7131868600845337,
      "learning_rate": 0.0001,
      "loss": 1.1945,
      "mean_token_accuracy": 0.6616647839546204,
      "num_tokens": 192065093.0,
      "step": 464
    },
    {
      "epoch": 0.8157894736842105,
      "grad_norm": 0.620922863483429,
      "learning_rate": 0.0001,
      "loss": 1.2134,
      "mean_token_accuracy": 0.657670259475708,
      "num_tokens": 192458897.0,
      "step": 465
    },
    {
      "epoch": 0.8175438596491228,
      "grad_norm": 0.6825653910636902,
      "learning_rate": 0.0001,
      "loss": 1.2083,
      "mean_token_accuracy": 0.6574592590332031,
      "num_tokens": 192871837.0,
      "step": 466
    },
    {
      "epoch": 0.8192982456140351,
      "grad_norm": 0.649117648601532,
      "learning_rate": 0.0001,
      "loss": 1.1904,
      "mean_token_accuracy": 0.661907970905304,
      "num_tokens": 193297231.0,
      "step": 467
    },
    {
      "epoch": 0.8210526315789474,
      "grad_norm": 0.5843600630760193,
      "learning_rate": 0.0001,
      "loss": 1.1613,
      "mean_token_accuracy": 0.6669655442237854,
      "num_tokens": 193684403.0,
      "step": 468
    },
    {
      "epoch": 0.8228070175438597,
      "grad_norm": 0.6877574324607849,
      "learning_rate": 0.0001,
      "loss": 1.2131,
      "mean_token_accuracy": 0.6550207138061523,
      "num_tokens": 194113357.0,
      "step": 469
    },
    {
      "epoch": 0.8245614035087719,
      "grad_norm": 0.6516855955123901,
      "learning_rate": 0.0001,
      "loss": 1.1979,
      "mean_token_accuracy": 0.65838223695755,
      "num_tokens": 194526469.0,
      "step": 470
    },
    {
      "epoch": 0.8263157894736842,
      "grad_norm": 0.6000040769577026,
      "learning_rate": 0.0001,
      "loss": 1.2303,
      "mean_token_accuracy": 0.6509213447570801,
      "num_tokens": 194953909.0,
      "step": 471
    },
    {
      "epoch": 0.8280701754385965,
      "grad_norm": 0.6414221525192261,
      "learning_rate": 0.0001,
      "loss": 1.2276,
      "mean_token_accuracy": 0.6521142721176147,
      "num_tokens": 195388769.0,
      "step": 472
    },
    {
      "epoch": 0.8298245614035088,
      "grad_norm": 0.614547848701477,
      "learning_rate": 0.0001,
      "loss": 1.2174,
      "mean_token_accuracy": 0.6551234126091003,
      "num_tokens": 195818916.0,
      "step": 473
    },
    {
      "epoch": 0.8315789473684211,
      "grad_norm": 0.6391692161560059,
      "learning_rate": 0.0001,
      "loss": 1.1963,
      "mean_token_accuracy": 0.659719705581665,
      "num_tokens": 196233034.0,
      "step": 474
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 0.6614966988563538,
      "learning_rate": 0.0001,
      "loss": 1.2407,
      "mean_token_accuracy": 0.6485875844955444,
      "num_tokens": 196660034.0,
      "step": 475
    },
    {
      "epoch": 0.8350877192982457,
      "grad_norm": 0.5896729826927185,
      "learning_rate": 0.0001,
      "loss": 1.2188,
      "mean_token_accuracy": 0.6551612615585327,
      "num_tokens": 197080673.0,
      "step": 476
    },
    {
      "epoch": 0.8368421052631579,
      "grad_norm": 0.6428948044776917,
      "learning_rate": 0.0001,
      "loss": 1.1962,
      "mean_token_accuracy": 0.6604630947113037,
      "num_tokens": 197493003.0,
      "step": 477
    },
    {
      "epoch": 0.8385964912280702,
      "grad_norm": 0.6853853464126587,
      "learning_rate": 0.0001,
      "loss": 1.2116,
      "mean_token_accuracy": 0.6553216576576233,
      "num_tokens": 197923635.0,
      "step": 478
    },
    {
      "epoch": 0.8403508771929824,
      "grad_norm": 0.6877092719078064,
      "learning_rate": 0.0001,
      "loss": 1.2017,
      "mean_token_accuracy": 0.657435417175293,
      "num_tokens": 198339854.0,
      "step": 479
    },
    {
      "epoch": 0.8421052631578947,
      "grad_norm": 0.5886791348457336,
      "learning_rate": 0.0001,
      "loss": 1.2414,
      "mean_token_accuracy": 0.6491390466690063,
      "num_tokens": 198768483.0,
      "step": 480
    },
    {
      "epoch": 0.843859649122807,
      "grad_norm": 0.8585889935493469,
      "learning_rate": 0.0001,
      "loss": 1.2168,
      "mean_token_accuracy": 0.6541799902915955,
      "num_tokens": 199190842.0,
      "step": 481
    },
    {
      "epoch": 0.8456140350877193,
      "grad_norm": 0.6527767181396484,
      "learning_rate": 0.0001,
      "loss": 1.2236,
      "mean_token_accuracy": 0.6539558172225952,
      "num_tokens": 199620649.0,
      "step": 482
    },
    {
      "epoch": 0.8473684210526315,
      "grad_norm": 0.6834746599197388,
      "learning_rate": 0.0001,
      "loss": 1.2015,
      "mean_token_accuracy": 0.6586301326751709,
      "num_tokens": 200023406.0,
      "step": 483
    },
    {
      "epoch": 0.8491228070175438,
      "grad_norm": 0.6827247142791748,
      "learning_rate": 0.0001,
      "loss": 1.178,
      "mean_token_accuracy": 0.6644470691680908,
      "num_tokens": 200430239.0,
      "step": 484
    },
    {
      "epoch": 0.8508771929824561,
      "grad_norm": 0.6491426229476929,
      "learning_rate": 0.0001,
      "loss": 1.1927,
      "mean_token_accuracy": 0.6621779799461365,
      "num_tokens": 200833864.0,
      "step": 485
    },
    {
      "epoch": 0.8526315789473684,
      "grad_norm": 0.6229031682014465,
      "learning_rate": 0.0001,
      "loss": 1.1923,
      "mean_token_accuracy": 0.6596782207489014,
      "num_tokens": 201241869.0,
      "step": 486
    },
    {
      "epoch": 0.8543859649122807,
      "grad_norm": 0.5779981017112732,
      "learning_rate": 0.0001,
      "loss": 1.1857,
      "mean_token_accuracy": 0.6614409685134888,
      "num_tokens": 201658653.0,
      "step": 487
    },
    {
      "epoch": 0.856140350877193,
      "grad_norm": 0.6096077561378479,
      "learning_rate": 0.0001,
      "loss": 1.2096,
      "mean_token_accuracy": 0.6574957370758057,
      "num_tokens": 202086503.0,
      "step": 488
    },
    {
      "epoch": 0.8578947368421053,
      "grad_norm": 0.7495996952056885,
      "learning_rate": 0.0001,
      "loss": 1.2005,
      "mean_token_accuracy": 0.6597346067428589,
      "num_tokens": 202509306.0,
      "step": 489
    },
    {
      "epoch": 0.8596491228070176,
      "grad_norm": 0.6209189295768738,
      "learning_rate": 0.0001,
      "loss": 1.1753,
      "mean_token_accuracy": 0.6628221273422241,
      "num_tokens": 202909087.0,
      "step": 490
    },
    {
      "epoch": 0.8614035087719298,
      "grad_norm": 0.563208281993866,
      "learning_rate": 0.0001,
      "loss": 1.2005,
      "mean_token_accuracy": 0.6574559211730957,
      "num_tokens": 203337615.0,
      "step": 491
    },
    {
      "epoch": 0.8631578947368421,
      "grad_norm": 0.6872074604034424,
      "learning_rate": 0.0001,
      "loss": 1.1982,
      "mean_token_accuracy": 0.6597882509231567,
      "num_tokens": 203754527.0,
      "step": 492
    },
    {
      "epoch": 0.8649122807017544,
      "grad_norm": 0.6505935192108154,
      "learning_rate": 0.0001,
      "loss": 1.1734,
      "mean_token_accuracy": 0.666144609451294,
      "num_tokens": 204166768.0,
      "step": 493
    },
    {
      "epoch": 0.8666666666666667,
      "grad_norm": 0.7290279269218445,
      "learning_rate": 0.0001,
      "loss": 1.1923,
      "mean_token_accuracy": 0.6601245403289795,
      "num_tokens": 204554076.0,
      "step": 494
    },
    {
      "epoch": 0.868421052631579,
      "grad_norm": 0.6451328992843628,
      "learning_rate": 0.0001,
      "loss": 1.2272,
      "mean_token_accuracy": 0.653006911277771,
      "num_tokens": 204962578.0,
      "step": 495
    },
    {
      "epoch": 0.8701754385964913,
      "grad_norm": 0.7413347363471985,
      "learning_rate": 0.0001,
      "loss": 1.153,
      "mean_token_accuracy": 0.6694173216819763,
      "num_tokens": 205358271.0,
      "step": 496
    },
    {
      "epoch": 0.8719298245614036,
      "grad_norm": 0.6787010431289673,
      "learning_rate": 0.0001,
      "loss": 1.2085,
      "mean_token_accuracy": 0.6561261415481567,
      "num_tokens": 205786890.0,
      "step": 497
    },
    {
      "epoch": 0.8736842105263158,
      "grad_norm": 0.6698117256164551,
      "learning_rate": 0.0001,
      "loss": 1.2019,
      "mean_token_accuracy": 0.6592704057693481,
      "num_tokens": 206193572.0,
      "step": 498
    },
    {
      "epoch": 0.875438596491228,
      "grad_norm": 0.6170295476913452,
      "learning_rate": 0.0001,
      "loss": 1.1723,
      "mean_token_accuracy": 0.6650887727737427,
      "num_tokens": 206607450.0,
      "step": 499
    },
    {
      "epoch": 0.8771929824561403,
      "grad_norm": 0.5921252965927124,
      "learning_rate": 0.0001,
      "loss": 1.1823,
      "mean_token_accuracy": 0.6633787155151367,
      "num_tokens": 207005126.0,
      "step": 500
    },
    {
      "epoch": 0.8789473684210526,
      "grad_norm": 0.69658362865448,
      "learning_rate": 0.0001,
      "loss": 1.1793,
      "mean_token_accuracy": 0.6646496057510376,
      "num_tokens": 207394794.0,
      "step": 501
    },
    {
      "epoch": 0.8807017543859649,
      "grad_norm": 0.6810624599456787,
      "learning_rate": 0.0001,
      "loss": 1.1979,
      "mean_token_accuracy": 0.6584210395812988,
      "num_tokens": 207783539.0,
      "step": 502
    },
    {
      "epoch": 0.8824561403508772,
      "grad_norm": 0.6264888644218445,
      "learning_rate": 0.0001,
      "loss": 1.2045,
      "mean_token_accuracy": 0.6583248376846313,
      "num_tokens": 208196198.0,
      "step": 503
    },
    {
      "epoch": 0.8842105263157894,
      "grad_norm": 0.6070482730865479,
      "learning_rate": 0.0001,
      "loss": 1.1995,
      "mean_token_accuracy": 0.6601771712303162,
      "num_tokens": 208602598.0,
      "step": 504
    },
    {
      "epoch": 0.8859649122807017,
      "grad_norm": 0.6856517791748047,
      "learning_rate": 0.0001,
      "loss": 1.1909,
      "mean_token_accuracy": 0.6614357233047485,
      "num_tokens": 209039139.0,
      "step": 505
    },
    {
      "epoch": 0.887719298245614,
      "grad_norm": 0.5697737336158752,
      "learning_rate": 0.0001,
      "loss": 1.1967,
      "mean_token_accuracy": 0.6589823961257935,
      "num_tokens": 209437950.0,
      "step": 506
    },
    {
      "epoch": 0.8894736842105263,
      "grad_norm": 0.7310987114906311,
      "learning_rate": 0.0001,
      "loss": 1.205,
      "mean_token_accuracy": 0.6580671668052673,
      "num_tokens": 209862559.0,
      "step": 507
    },
    {
      "epoch": 0.8912280701754386,
      "grad_norm": 0.6229117512702942,
      "learning_rate": 0.0001,
      "loss": 1.222,
      "mean_token_accuracy": 0.6536027789115906,
      "num_tokens": 210284769.0,
      "step": 508
    },
    {
      "epoch": 0.8929824561403509,
      "grad_norm": 0.5739285349845886,
      "learning_rate": 0.0001,
      "loss": 1.2059,
      "mean_token_accuracy": 0.6585639119148254,
      "num_tokens": 210708859.0,
      "step": 509
    },
    {
      "epoch": 0.8947368421052632,
      "grad_norm": 0.6239802837371826,
      "learning_rate": 0.0001,
      "loss": 1.1986,
      "mean_token_accuracy": 0.6589633822441101,
      "num_tokens": 211130168.0,
      "step": 510
    },
    {
      "epoch": 0.8964912280701754,
      "grad_norm": 0.6617391705513,
      "learning_rate": 0.0001,
      "loss": 1.2027,
      "mean_token_accuracy": 0.6577974557876587,
      "num_tokens": 211529753.0,
      "step": 511
    },
    {
      "epoch": 0.8982456140350877,
      "grad_norm": 0.638733983039856,
      "learning_rate": 0.0001,
      "loss": 1.2142,
      "mean_token_accuracy": 0.6540495157241821,
      "num_tokens": 211963430.0,
      "step": 512
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.6008735299110413,
      "learning_rate": 0.0001,
      "loss": 1.2136,
      "mean_token_accuracy": 0.6559524536132812,
      "num_tokens": 212387404.0,
      "step": 513
    },
    {
      "epoch": 0.9017543859649123,
      "grad_norm": 0.6343475580215454,
      "learning_rate": 0.0001,
      "loss": 1.1718,
      "mean_token_accuracy": 0.6651860475540161,
      "num_tokens": 212802763.0,
      "step": 514
    },
    {
      "epoch": 0.9035087719298246,
      "grad_norm": 0.637675940990448,
      "learning_rate": 0.0001,
      "loss": 1.1694,
      "mean_token_accuracy": 0.6661834716796875,
      "num_tokens": 213219406.0,
      "step": 515
    },
    {
      "epoch": 0.9052631578947369,
      "grad_norm": 0.5518184900283813,
      "learning_rate": 0.0001,
      "loss": 1.1954,
      "mean_token_accuracy": 0.6603313684463501,
      "num_tokens": 213623710.0,
      "step": 516
    },
    {
      "epoch": 0.9070175438596492,
      "grad_norm": 0.6756175756454468,
      "learning_rate": 0.0001,
      "loss": 1.1701,
      "mean_token_accuracy": 0.6667043566703796,
      "num_tokens": 214053806.0,
      "step": 517
    },
    {
      "epoch": 0.9087719298245615,
      "grad_norm": 0.5964516401290894,
      "learning_rate": 0.0001,
      "loss": 1.2007,
      "mean_token_accuracy": 0.6573567390441895,
      "num_tokens": 214457394.0,
      "step": 518
    },
    {
      "epoch": 0.9105263157894737,
      "grad_norm": 0.745707094669342,
      "learning_rate": 0.0001,
      "loss": 1.1697,
      "mean_token_accuracy": 0.6656190156936646,
      "num_tokens": 214841472.0,
      "step": 519
    },
    {
      "epoch": 0.9122807017543859,
      "grad_norm": 0.5971705317497253,
      "learning_rate": 0.0001,
      "loss": 1.2061,
      "mean_token_accuracy": 0.656207799911499,
      "num_tokens": 215261046.0,
      "step": 520
    },
    {
      "epoch": 0.9140350877192982,
      "grad_norm": 0.7177700400352478,
      "learning_rate": 0.0001,
      "loss": 1.1753,
      "mean_token_accuracy": 0.6650264263153076,
      "num_tokens": 215664423.0,
      "step": 521
    },
    {
      "epoch": 0.9157894736842105,
      "grad_norm": 0.5945612788200378,
      "learning_rate": 0.0001,
      "loss": 1.1813,
      "mean_token_accuracy": 0.66515052318573,
      "num_tokens": 216072733.0,
      "step": 522
    },
    {
      "epoch": 0.9175438596491228,
      "grad_norm": 0.7161288857460022,
      "learning_rate": 0.0001,
      "loss": 1.1964,
      "mean_token_accuracy": 0.6598995923995972,
      "num_tokens": 216490157.0,
      "step": 523
    },
    {
      "epoch": 0.9192982456140351,
      "grad_norm": 0.6490321159362793,
      "learning_rate": 0.0001,
      "loss": 1.2151,
      "mean_token_accuracy": 0.6550993919372559,
      "num_tokens": 216933714.0,
      "step": 524
    },
    {
      "epoch": 0.9210526315789473,
      "grad_norm": 0.6328585743904114,
      "learning_rate": 0.0001,
      "loss": 1.2008,
      "mean_token_accuracy": 0.658663272857666,
      "num_tokens": 217360826.0,
      "step": 525
    },
    {
      "epoch": 0.9228070175438596,
      "grad_norm": 0.6045000553131104,
      "learning_rate": 0.0001,
      "loss": 1.1837,
      "mean_token_accuracy": 0.6624784469604492,
      "num_tokens": 217767964.0,
      "step": 526
    },
    {
      "epoch": 0.9245614035087719,
      "grad_norm": 0.5896552205085754,
      "learning_rate": 0.0001,
      "loss": 1.1785,
      "mean_token_accuracy": 0.663489043712616,
      "num_tokens": 218182502.0,
      "step": 527
    },
    {
      "epoch": 0.9263157894736842,
      "grad_norm": 0.6433465480804443,
      "learning_rate": 0.0001,
      "loss": 1.1866,
      "mean_token_accuracy": 0.662973940372467,
      "num_tokens": 218599164.0,
      "step": 528
    },
    {
      "epoch": 0.9280701754385965,
      "grad_norm": 0.6225712895393372,
      "learning_rate": 0.0001,
      "loss": 1.1732,
      "mean_token_accuracy": 0.6657634973526001,
      "num_tokens": 219038853.0,
      "step": 529
    },
    {
      "epoch": 0.9298245614035088,
      "grad_norm": 0.6584674715995789,
      "learning_rate": 0.0001,
      "loss": 1.1931,
      "mean_token_accuracy": 0.6607257127761841,
      "num_tokens": 219455763.0,
      "step": 530
    },
    {
      "epoch": 0.9315789473684211,
      "grad_norm": 0.5859020352363586,
      "learning_rate": 0.0001,
      "loss": 1.2043,
      "mean_token_accuracy": 0.657474935054779,
      "num_tokens": 219857096.0,
      "step": 531
    },
    {
      "epoch": 0.9333333333333333,
      "grad_norm": 0.6879558563232422,
      "learning_rate": 0.0001,
      "loss": 1.1761,
      "mean_token_accuracy": 0.6635361909866333,
      "num_tokens": 220269043.0,
      "step": 532
    },
    {
      "epoch": 0.9350877192982456,
      "grad_norm": 0.6866979002952576,
      "learning_rate": 0.0001,
      "loss": 1.2073,
      "mean_token_accuracy": 0.6566738486289978,
      "num_tokens": 220650778.0,
      "step": 533
    },
    {
      "epoch": 0.9368421052631579,
      "grad_norm": 0.6336923241615295,
      "learning_rate": 0.0001,
      "loss": 1.2039,
      "mean_token_accuracy": 0.6579070091247559,
      "num_tokens": 221055825.0,
      "step": 534
    },
    {
      "epoch": 0.9385964912280702,
      "grad_norm": 0.6081579327583313,
      "learning_rate": 0.0001,
      "loss": 1.2108,
      "mean_token_accuracy": 0.6550259590148926,
      "num_tokens": 221459868.0,
      "step": 535
    },
    {
      "epoch": 0.9403508771929825,
      "grad_norm": 0.6312009692192078,
      "learning_rate": 0.0001,
      "loss": 1.1702,
      "mean_token_accuracy": 0.6641944646835327,
      "num_tokens": 221860324.0,
      "step": 536
    },
    {
      "epoch": 0.9421052631578948,
      "grad_norm": 0.5887439250946045,
      "learning_rate": 0.0001,
      "loss": 1.1778,
      "mean_token_accuracy": 0.6638119220733643,
      "num_tokens": 222291251.0,
      "step": 537
    },
    {
      "epoch": 0.9438596491228071,
      "grad_norm": 0.543400764465332,
      "learning_rate": 0.0001,
      "loss": 1.1805,
      "mean_token_accuracy": 0.6627988815307617,
      "num_tokens": 222699559.0,
      "step": 538
    },
    {
      "epoch": 0.9456140350877194,
      "grad_norm": 0.5787383913993835,
      "learning_rate": 0.0001,
      "loss": 1.1914,
      "mean_token_accuracy": 0.6620433330535889,
      "num_tokens": 223156357.0,
      "step": 539
    },
    {
      "epoch": 0.9473684210526315,
      "grad_norm": 0.6597963571548462,
      "learning_rate": 0.0001,
      "loss": 1.1871,
      "mean_token_accuracy": 0.6626171469688416,
      "num_tokens": 223562411.0,
      "step": 540
    },
    {
      "epoch": 0.9491228070175438,
      "grad_norm": 0.5731210112571716,
      "learning_rate": 0.0001,
      "loss": 1.2078,
      "mean_token_accuracy": 0.6560046076774597,
      "num_tokens": 223988252.0,
      "step": 541
    },
    {
      "epoch": 0.9508771929824561,
      "grad_norm": 0.7036701440811157,
      "learning_rate": 0.0001,
      "loss": 1.1917,
      "mean_token_accuracy": 0.6613459587097168,
      "num_tokens": 224401217.0,
      "step": 542
    },
    {
      "epoch": 0.9526315789473684,
      "grad_norm": 0.5783252120018005,
      "learning_rate": 0.0001,
      "loss": 1.1757,
      "mean_token_accuracy": 0.6623378396034241,
      "num_tokens": 224807482.0,
      "step": 543
    },
    {
      "epoch": 0.9543859649122807,
      "grad_norm": 0.7617517113685608,
      "learning_rate": 0.0001,
      "loss": 1.1937,
      "mean_token_accuracy": 0.659681499004364,
      "num_tokens": 225220680.0,
      "step": 544
    },
    {
      "epoch": 0.956140350877193,
      "grad_norm": 0.6007680296897888,
      "learning_rate": 0.0001,
      "loss": 1.2401,
      "mean_token_accuracy": 0.6486543416976929,
      "num_tokens": 225640539.0,
      "step": 545
    },
    {
      "epoch": 0.9578947368421052,
      "grad_norm": 0.7272628545761108,
      "learning_rate": 0.0001,
      "loss": 1.1822,
      "mean_token_accuracy": 0.6626489162445068,
      "num_tokens": 226035924.0,
      "step": 546
    },
    {
      "epoch": 0.9596491228070175,
      "grad_norm": 0.700038492679596,
      "learning_rate": 0.0001,
      "loss": 1.1658,
      "mean_token_accuracy": 0.666049599647522,
      "num_tokens": 226425232.0,
      "step": 547
    },
    {
      "epoch": 0.9614035087719298,
      "grad_norm": 0.6490049958229065,
      "learning_rate": 0.0001,
      "loss": 1.1691,
      "mean_token_accuracy": 0.6649153828620911,
      "num_tokens": 226802383.0,
      "step": 548
    },
    {
      "epoch": 0.9631578947368421,
      "grad_norm": 0.7154028415679932,
      "learning_rate": 0.0001,
      "loss": 1.1986,
      "mean_token_accuracy": 0.6584769487380981,
      "num_tokens": 227227610.0,
      "step": 549
    },
    {
      "epoch": 0.9649122807017544,
      "grad_norm": 0.6601865887641907,
      "learning_rate": 0.0001,
      "loss": 1.2034,
      "mean_token_accuracy": 0.6581115126609802,
      "num_tokens": 227657882.0,
      "step": 550
    },
    {
      "epoch": 0.9666666666666667,
      "grad_norm": 0.6211066842079163,
      "learning_rate": 0.0001,
      "loss": 1.1749,
      "mean_token_accuracy": 0.6637754440307617,
      "num_tokens": 228081460.0,
      "step": 551
    },
    {
      "epoch": 0.968421052631579,
      "grad_norm": 0.6879007816314697,
      "learning_rate": 0.0001,
      "loss": 1.2181,
      "mean_token_accuracy": 0.6532254219055176,
      "num_tokens": 228507388.0,
      "step": 552
    },
    {
      "epoch": 0.9701754385964912,
      "grad_norm": 0.6297675371170044,
      "learning_rate": 0.0001,
      "loss": 1.2126,
      "mean_token_accuracy": 0.654970645904541,
      "num_tokens": 228935033.0,
      "step": 553
    },
    {
      "epoch": 0.9719298245614035,
      "grad_norm": 0.5917762517929077,
      "learning_rate": 0.0001,
      "loss": 1.1517,
      "mean_token_accuracy": 0.6693041920661926,
      "num_tokens": 229325690.0,
      "step": 554
    },
    {
      "epoch": 0.9736842105263158,
      "grad_norm": 0.6466293334960938,
      "learning_rate": 0.0001,
      "loss": 1.1573,
      "mean_token_accuracy": 0.6677490472793579,
      "num_tokens": 229726549.0,
      "step": 555
    },
    {
      "epoch": 0.9754385964912281,
      "grad_norm": 0.6341378688812256,
      "learning_rate": 0.0001,
      "loss": 1.185,
      "mean_token_accuracy": 0.6614177227020264,
      "num_tokens": 230122274.0,
      "step": 556
    },
    {
      "epoch": 0.9771929824561404,
      "grad_norm": 0.604850172996521,
      "learning_rate": 0.0001,
      "loss": 1.1959,
      "mean_token_accuracy": 0.6580736637115479,
      "num_tokens": 230526104.0,
      "step": 557
    },
    {
      "epoch": 0.9789473684210527,
      "grad_norm": 0.7436766624450684,
      "learning_rate": 0.0001,
      "loss": 1.1642,
      "mean_token_accuracy": 0.6674508452415466,
      "num_tokens": 230941199.0,
      "step": 558
    },
    {
      "epoch": 0.980701754385965,
      "grad_norm": 0.6362001895904541,
      "learning_rate": 0.0001,
      "loss": 1.1751,
      "mean_token_accuracy": 0.6646236181259155,
      "num_tokens": 231367872.0,
      "step": 559
    },
    {
      "epoch": 0.9824561403508771,
      "grad_norm": 0.6686745882034302,
      "learning_rate": 0.0001,
      "loss": 1.2065,
      "mean_token_accuracy": 0.6581393480300903,
      "num_tokens": 231804511.0,
      "step": 560
    },
    {
      "epoch": 0.9842105263157894,
      "grad_norm": 0.7186607718467712,
      "learning_rate": 0.0001,
      "loss": 1.1692,
      "mean_token_accuracy": 0.6657729148864746,
      "num_tokens": 232201554.0,
      "step": 561
    },
    {
      "epoch": 0.9859649122807017,
      "grad_norm": 0.5875235795974731,
      "learning_rate": 0.0001,
      "loss": 1.185,
      "mean_token_accuracy": 0.6623810529708862,
      "num_tokens": 232632247.0,
      "step": 562
    },
    {
      "epoch": 0.987719298245614,
      "grad_norm": 0.6285355687141418,
      "learning_rate": 0.0001,
      "loss": 1.1932,
      "mean_token_accuracy": 0.6584882140159607,
      "num_tokens": 233040299.0,
      "step": 563
    },
    {
      "epoch": 0.9894736842105263,
      "grad_norm": 0.6787013411521912,
      "learning_rate": 0.0001,
      "loss": 1.2017,
      "mean_token_accuracy": 0.6589356660842896,
      "num_tokens": 233476898.0,
      "step": 564
    },
    {
      "epoch": 0.9912280701754386,
      "grad_norm": 0.5261335372924805,
      "learning_rate": 0.0001,
      "loss": 1.1674,
      "mean_token_accuracy": 0.6651347279548645,
      "num_tokens": 233898203.0,
      "step": 565
    },
    {
      "epoch": 0.9929824561403509,
      "grad_norm": 0.6217278242111206,
      "learning_rate": 0.0001,
      "loss": 1.2212,
      "mean_token_accuracy": 0.653939962387085,
      "num_tokens": 234316491.0,
      "step": 566
    },
    {
      "epoch": 0.9947368421052631,
      "grad_norm": 0.6469559073448181,
      "learning_rate": 0.0001,
      "loss": 1.1848,
      "mean_token_accuracy": 0.6615195870399475,
      "num_tokens": 234725455.0,
      "step": 567
    },
    {
      "epoch": 0.9964912280701754,
      "grad_norm": 0.6558631062507629,
      "learning_rate": 0.0001,
      "loss": 1.173,
      "mean_token_accuracy": 0.6650323867797852,
      "num_tokens": 235152094.0,
      "step": 568
    },
    {
      "epoch": 0.9982456140350877,
      "grad_norm": 0.6159579157829285,
      "learning_rate": 0.0001,
      "loss": 1.1885,
      "mean_token_accuracy": 0.6604526042938232,
      "num_tokens": 235558911.0,
      "step": 569
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.6799984574317932,
      "learning_rate": 0.0001,
      "loss": 1.1975,
      "mean_token_accuracy": 0.6584136486053467,
      "num_tokens": 235994347.0,
      "step": 570
    },
    {
      "epoch": 1.0,
      "step": 570,
      "total_flos": 1.377941890090926e+18,
      "train_loss": 1.2564991597543682,
      "train_runtime": 1307.2842,
      "train_samples_per_second": 111.621,
      "train_steps_per_second": 0.436
    }
  ],
  "logging_steps": 1,
  "max_steps": 570,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 285,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.377941890090926e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}