{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.0,
  "eval_steps": 96,
  "global_step": 480,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0125,
      "grad_norm": 2.395050048828125,
      "learning_rate": 2e-05,
      "loss": 1.4995,
      "step": 1
    },
    {
      "epoch": 0.025,
      "grad_norm": 2.28926944732666,
      "learning_rate": 4e-05,
      "loss": 1.4965,
      "step": 2
    },
    {
      "epoch": 0.0375,
      "grad_norm": 2.798283815383911,
      "learning_rate": 6e-05,
      "loss": 1.4668,
      "step": 3
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.141174554824829,
      "learning_rate": 8e-05,
      "loss": 1.3202,
      "step": 4
    },
    {
      "epoch": 0.0625,
      "grad_norm": 1.8937232494354248,
      "learning_rate": 0.0001,
      "loss": 1.1309,
      "step": 5
    },
    {
      "epoch": 0.075,
      "grad_norm": 0.9302598834037781,
      "learning_rate": 0.00012,
      "loss": 1.0411,
      "step": 6
    },
    {
      "epoch": 0.0875,
      "grad_norm": 1.0292965173721313,
      "learning_rate": 0.00014,
      "loss": 0.9394,
      "step": 7
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.2454479932785034,
      "learning_rate": 0.00016,
      "loss": 0.9301,
      "step": 8
    },
    {
      "epoch": 0.1125,
      "grad_norm": 1.3369941711425781,
      "learning_rate": 0.00018,
      "loss": 0.8461,
      "step": 9
    },
    {
      "epoch": 0.125,
      "grad_norm": 2.1479320526123047,
      "learning_rate": 0.0002,
      "loss": 0.7955,
      "step": 10
    },
    {
      "epoch": 0.1375,
      "grad_norm": 0.8992676734924316,
      "learning_rate": 0.0001995744680851064,
      "loss": 0.8119,
      "step": 11
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.9127306938171387,
      "learning_rate": 0.0001991489361702128,
      "loss": 0.7874,
      "step": 12
    },
    {
      "epoch": 0.1625,
      "grad_norm": 0.8198272585868835,
      "learning_rate": 0.00019872340425531918,
      "loss": 0.7407,
      "step": 13
    },
    {
      "epoch": 0.175,
      "grad_norm": 1.111543893814087,
      "learning_rate": 0.00019829787234042554,
      "loss": 0.8275,
      "step": 14
    },
    {
      "epoch": 0.1875,
      "grad_norm": 2.1975886821746826,
      "learning_rate": 0.00019787234042553193,
      "loss": 0.7143,
      "step": 15
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.7623311281204224,
      "learning_rate": 0.00019744680851063832,
      "loss": 0.7568,
      "step": 16
    },
    {
      "epoch": 0.2125,
      "grad_norm": 0.8410439491271973,
      "learning_rate": 0.0001970212765957447,
      "loss": 0.6292,
      "step": 17
    },
    {
      "epoch": 0.225,
      "grad_norm": 0.8105931878089905,
      "learning_rate": 0.00019659574468085107,
      "loss": 0.7748,
      "step": 18
    },
    {
      "epoch": 0.2375,
      "grad_norm": 0.8514361381530762,
      "learning_rate": 0.00019617021276595745,
      "loss": 0.7081,
      "step": 19
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.0301614999771118,
      "learning_rate": 0.00019574468085106384,
      "loss": 0.6643,
      "step": 20
    },
    {
      "epoch": 0.2625,
      "grad_norm": 0.7055068612098694,
      "learning_rate": 0.00019531914893617023,
      "loss": 0.7002,
      "step": 21
    },
    {
      "epoch": 0.275,
      "grad_norm": 0.6796251535415649,
      "learning_rate": 0.0001948936170212766,
      "loss": 0.7437,
      "step": 22
    },
    {
      "epoch": 0.2875,
      "grad_norm": 0.6966366767883301,
      "learning_rate": 0.00019446808510638298,
      "loss": 0.674,
      "step": 23
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.6405335068702698,
      "learning_rate": 0.00019404255319148937,
      "loss": 0.6369,
      "step": 24
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.6977805495262146,
      "learning_rate": 0.00019361702127659576,
      "loss": 0.6369,
      "step": 25
    },
    {
      "epoch": 0.325,
      "grad_norm": 0.6844583749771118,
      "learning_rate": 0.00019319148936170212,
      "loss": 0.6886,
      "step": 26
    },
    {
      "epoch": 0.3375,
      "grad_norm": 0.6735361814498901,
      "learning_rate": 0.0001927659574468085,
      "loss": 0.7131,
      "step": 27
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.6853126883506775,
      "learning_rate": 0.0001923404255319149,
      "loss": 0.6361,
      "step": 28
    },
    {
      "epoch": 0.3625,
      "grad_norm": 0.693722665309906,
      "learning_rate": 0.0001919148936170213,
      "loss": 0.6772,
      "step": 29
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.6884544491767883,
      "learning_rate": 0.00019148936170212768,
      "loss": 0.6403,
      "step": 30
    },
    {
      "epoch": 0.3875,
      "grad_norm": 0.6849629878997803,
      "learning_rate": 0.00019106382978723404,
      "loss": 0.6859,
      "step": 31
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.7242761850357056,
      "learning_rate": 0.00019063829787234045,
      "loss": 0.6867,
      "step": 32
    },
    {
      "epoch": 0.4125,
      "grad_norm": 0.9043200016021729,
      "learning_rate": 0.00019021276595744682,
      "loss": 0.6624,
      "step": 33
    },
    {
      "epoch": 0.425,
      "grad_norm": 0.9550056457519531,
      "learning_rate": 0.0001897872340425532,
      "loss": 0.6395,
      "step": 34
    },
    {
      "epoch": 0.4375,
      "grad_norm": 0.6916117668151855,
      "learning_rate": 0.00018936170212765957,
      "loss": 0.7046,
      "step": 35
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.6232885718345642,
      "learning_rate": 0.00018893617021276598,
      "loss": 0.6716,
      "step": 36
    },
    {
      "epoch": 0.4625,
      "grad_norm": 0.6697956323623657,
      "learning_rate": 0.00018851063829787234,
      "loss": 0.7149,
      "step": 37
    },
    {
      "epoch": 0.475,
      "grad_norm": 0.6431601643562317,
      "learning_rate": 0.00018808510638297873,
      "loss": 0.6996,
      "step": 38
    },
    {
      "epoch": 0.4875,
      "grad_norm": 0.614162802696228,
      "learning_rate": 0.00018765957446808512,
      "loss": 0.6178,
      "step": 39
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.8608362078666687,
      "learning_rate": 0.0001872340425531915,
      "loss": 0.7055,
      "step": 40
    },
    {
      "epoch": 0.5125,
      "grad_norm": 0.6234711408615112,
      "learning_rate": 0.00018680851063829787,
      "loss": 0.682,
      "step": 41
    },
    {
      "epoch": 0.525,
      "grad_norm": 0.6244091987609863,
      "learning_rate": 0.00018638297872340426,
      "loss": 0.6767,
      "step": 42
    },
    {
      "epoch": 0.5375,
      "grad_norm": 0.6198769807815552,
      "learning_rate": 0.00018595744680851065,
      "loss": 0.6993,
      "step": 43
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.6559383273124695,
      "learning_rate": 0.00018553191489361704,
      "loss": 0.6648,
      "step": 44
    },
    {
      "epoch": 0.5625,
      "grad_norm": 0.6858519315719604,
      "learning_rate": 0.0001851063829787234,
      "loss": 0.6413,
      "step": 45
    },
    {
      "epoch": 0.575,
      "grad_norm": 0.5903568863868713,
      "learning_rate": 0.0001846808510638298,
      "loss": 0.6755,
      "step": 46
    },
    {
      "epoch": 0.5875,
      "grad_norm": 0.6439600586891174,
      "learning_rate": 0.00018425531914893618,
      "loss": 0.6677,
      "step": 47
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.6454241275787354,
      "learning_rate": 0.00018382978723404257,
      "loss": 0.6904,
      "step": 48
    },
    {
      "epoch": 0.6125,
      "grad_norm": 0.5940038561820984,
      "learning_rate": 0.00018340425531914896,
      "loss": 0.6551,
      "step": 49
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.606603741645813,
      "learning_rate": 0.00018297872340425532,
      "loss": 0.5867,
      "step": 50
    },
    {
      "epoch": 0.6375,
      "grad_norm": 0.616865873336792,
      "learning_rate": 0.0001825531914893617,
      "loss": 0.6865,
      "step": 51
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.6501201391220093,
      "learning_rate": 0.0001821276595744681,
      "loss": 0.5942,
      "step": 52
    },
    {
      "epoch": 0.6625,
      "grad_norm": 0.6296024918556213,
      "learning_rate": 0.00018170212765957448,
      "loss": 0.6762,
      "step": 53
    },
    {
      "epoch": 0.675,
      "grad_norm": 0.656863808631897,
      "learning_rate": 0.00018127659574468084,
      "loss": 0.7108,
      "step": 54
    },
    {
      "epoch": 0.6875,
      "grad_norm": 0.5761714577674866,
      "learning_rate": 0.00018085106382978726,
      "loss": 0.6133,
      "step": 55
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.6318382620811462,
      "learning_rate": 0.00018042553191489362,
      "loss": 0.6654,
      "step": 56
    },
    {
      "epoch": 0.7125,
      "grad_norm": 0.6335327625274658,
      "learning_rate": 0.00018,
      "loss": 0.6497,
      "step": 57
    },
    {
      "epoch": 0.725,
      "grad_norm": 0.5412627458572388,
      "learning_rate": 0.00017957446808510637,
      "loss": 0.6578,
      "step": 58
    },
    {
      "epoch": 0.7375,
      "grad_norm": 0.5853710770606995,
      "learning_rate": 0.0001791489361702128,
      "loss": 0.689,
      "step": 59
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.6023885607719421,
      "learning_rate": 0.00017872340425531915,
      "loss": 0.6605,
      "step": 60
    },
    {
      "epoch": 0.7625,
      "grad_norm": 0.6343605518341064,
      "learning_rate": 0.00017829787234042554,
      "loss": 0.702,
      "step": 61
    },
    {
      "epoch": 0.775,
      "grad_norm": 0.5401332378387451,
      "learning_rate": 0.0001778723404255319,
      "loss": 0.5743,
      "step": 62
    },
    {
      "epoch": 0.7875,
      "grad_norm": 0.5706886649131775,
      "learning_rate": 0.00017744680851063832,
      "loss": 0.6362,
      "step": 63
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5953827500343323,
      "learning_rate": 0.00017702127659574468,
      "loss": 0.6257,
      "step": 64
    },
    {
      "epoch": 0.8125,
      "grad_norm": 0.5751234889030457,
      "learning_rate": 0.00017659574468085107,
      "loss": 0.5577,
      "step": 65
    },
    {
      "epoch": 0.825,
      "grad_norm": 0.6519968509674072,
      "learning_rate": 0.00017617021276595746,
      "loss": 0.6904,
      "step": 66
    },
    {
      "epoch": 0.8375,
      "grad_norm": 0.5914306640625,
      "learning_rate": 0.00017574468085106384,
      "loss": 0.6893,
      "step": 67
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5872571468353271,
      "learning_rate": 0.00017531914893617023,
      "loss": 0.583,
      "step": 68
    },
    {
      "epoch": 0.8625,
      "grad_norm": 0.6391447186470032,
      "learning_rate": 0.0001748936170212766,
      "loss": 0.6216,
      "step": 69
    },
    {
      "epoch": 0.875,
      "grad_norm": 0.6009213924407959,
      "learning_rate": 0.00017446808510638298,
      "loss": 0.589,
      "step": 70
    },
    {
      "epoch": 0.8875,
      "grad_norm": 0.5932193398475647,
      "learning_rate": 0.00017404255319148937,
      "loss": 0.6251,
      "step": 71
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.6414621472358704,
      "learning_rate": 0.00017361702127659576,
      "loss": 0.6258,
      "step": 72
    },
    {
      "epoch": 0.9125,
      "grad_norm": 0.6453771591186523,
      "learning_rate": 0.00017319148936170212,
      "loss": 0.5774,
      "step": 73
    },
    {
      "epoch": 0.925,
      "grad_norm": 0.6160643696784973,
      "learning_rate": 0.0001727659574468085,
      "loss": 0.6119,
      "step": 74
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.6067684292793274,
      "learning_rate": 0.0001723404255319149,
      "loss": 0.652,
      "step": 75
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.623906135559082,
      "learning_rate": 0.0001719148936170213,
      "loss": 0.6086,
      "step": 76
    },
    {
      "epoch": 0.9625,
      "grad_norm": 0.6175603270530701,
      "learning_rate": 0.00017148936170212765,
      "loss": 0.6485,
      "step": 77
    },
    {
      "epoch": 0.975,
      "grad_norm": 0.6073062419891357,
      "learning_rate": 0.00017106382978723404,
      "loss": 0.5925,
      "step": 78
    },
    {
      "epoch": 0.9875,
      "grad_norm": 0.5946528911590576,
      "learning_rate": 0.00017063829787234043,
      "loss": 0.58,
      "step": 79
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.9330272078514099,
      "learning_rate": 0.00017021276595744682,
      "loss": 0.7328,
      "step": 80
    },
    {
      "epoch": 1.0125,
      "grad_norm": 0.5311029553413391,
      "learning_rate": 0.00016978723404255318,
      "loss": 0.4562,
      "step": 81
    },
    {
      "epoch": 1.025,
      "grad_norm": 0.5370614528656006,
      "learning_rate": 0.0001693617021276596,
      "loss": 0.4304,
      "step": 82
    },
    {
      "epoch": 1.0375,
      "grad_norm": 0.5361345410346985,
      "learning_rate": 0.00016893617021276598,
      "loss": 0.4604,
      "step": 83
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.6242905259132385,
      "learning_rate": 0.00016851063829787235,
      "loss": 0.4903,
      "step": 84
    },
    {
      "epoch": 1.0625,
      "grad_norm": 0.6181880235671997,
      "learning_rate": 0.00016808510638297873,
      "loss": 0.4524,
      "step": 85
    },
    {
      "epoch": 1.075,
      "grad_norm": 0.6893963813781738,
      "learning_rate": 0.00016765957446808512,
      "loss": 0.4964,
      "step": 86
    },
    {
      "epoch": 1.0875,
      "grad_norm": 0.5981054902076721,
      "learning_rate": 0.0001672340425531915,
      "loss": 0.4067,
      "step": 87
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.6721552610397339,
      "learning_rate": 0.00016680851063829787,
      "loss": 0.5138,
      "step": 88
    },
    {
      "epoch": 1.1125,
      "grad_norm": 0.7648325562477112,
      "learning_rate": 0.00016638297872340426,
      "loss": 0.4934,
      "step": 89
    },
    {
      "epoch": 1.125,
      "grad_norm": 0.724721372127533,
      "learning_rate": 0.00016595744680851065,
      "loss": 0.5145,
      "step": 90
    },
    {
      "epoch": 1.1375,
      "grad_norm": 0.6328797340393066,
      "learning_rate": 0.00016553191489361704,
      "loss": 0.4414,
      "step": 91
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.6504305005073547,
      "learning_rate": 0.0001651063829787234,
      "loss": 0.4773,
      "step": 92
    },
    {
      "epoch": 1.1625,
      "grad_norm": 0.6754821538925171,
      "learning_rate": 0.0001646808510638298,
      "loss": 0.4611,
      "step": 93
    },
    {
      "epoch": 1.175,
      "grad_norm": 0.75724858045578,
      "learning_rate": 0.00016425531914893618,
      "loss": 0.5343,
      "step": 94
    },
    {
      "epoch": 1.1875,
      "grad_norm": 0.7232646942138672,
      "learning_rate": 0.00016382978723404257,
      "loss": 0.4531,
      "step": 95
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.7688515782356262,
      "learning_rate": 0.00016340425531914893,
      "loss": 0.5669,
      "step": 96
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.6230465769767761,
      "eval_runtime": 9.9417,
      "eval_samples_per_second": 3.621,
      "eval_steps_per_second": 1.811,
      "step": 96
    },
    {
      "epoch": 1.2125,
      "grad_norm": 0.6552408337593079,
      "learning_rate": 0.00016297872340425532,
      "loss": 0.4746,
      "step": 97
    },
    {
      "epoch": 1.225,
      "grad_norm": 0.6849213242530823,
      "learning_rate": 0.0001625531914893617,
      "loss": 0.4506,
      "step": 98
    },
    {
      "epoch": 1.2375,
      "grad_norm": 0.6564849019050598,
      "learning_rate": 0.0001621276595744681,
      "loss": 0.4336,
      "step": 99
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.660833477973938,
      "learning_rate": 0.00016170212765957446,
      "loss": 0.4918,
      "step": 100
    },
    {
      "epoch": 1.2625,
      "grad_norm": 0.6267507076263428,
      "learning_rate": 0.00016127659574468085,
      "loss": 0.4693,
      "step": 101
    },
    {
      "epoch": 1.275,
      "grad_norm": 0.6528612375259399,
      "learning_rate": 0.00016085106382978726,
      "loss": 0.4449,
      "step": 102
    },
    {
      "epoch": 1.2875,
      "grad_norm": 0.6794521808624268,
      "learning_rate": 0.00016042553191489362,
      "loss": 0.4266,
      "step": 103
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.6800212264060974,
      "learning_rate": 0.00016,
      "loss": 0.4697,
      "step": 104
    },
    {
      "epoch": 1.3125,
      "grad_norm": 0.6715764403343201,
      "learning_rate": 0.00015957446808510637,
      "loss": 0.481,
      "step": 105
    },
    {
      "epoch": 1.325,
      "grad_norm": 0.7775130271911621,
      "learning_rate": 0.0001591489361702128,
      "loss": 0.4379,
      "step": 106
    },
    {
      "epoch": 1.3375,
      "grad_norm": 0.6991564631462097,
      "learning_rate": 0.00015872340425531915,
      "loss": 0.4519,
      "step": 107
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.7205126881599426,
      "learning_rate": 0.00015829787234042554,
      "loss": 0.4595,
      "step": 108
    },
    {
      "epoch": 1.3625,
      "grad_norm": 0.6301184892654419,
      "learning_rate": 0.00015787234042553193,
      "loss": 0.467,
      "step": 109
    },
    {
      "epoch": 1.375,
      "grad_norm": 0.7067103981971741,
      "learning_rate": 0.00015744680851063832,
      "loss": 0.4619,
      "step": 110
    },
    {
      "epoch": 1.3875,
      "grad_norm": 0.7016875743865967,
      "learning_rate": 0.00015702127659574468,
      "loss": 0.4747,
      "step": 111
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.6807534694671631,
      "learning_rate": 0.00015659574468085107,
      "loss": 0.477,
      "step": 112
    },
    {
      "epoch": 1.4125,
      "grad_norm": 0.6434107422828674,
      "learning_rate": 0.00015617021276595746,
      "loss": 0.4486,
      "step": 113
    },
    {
      "epoch": 1.425,
      "grad_norm": 0.611808717250824,
      "learning_rate": 0.00015574468085106385,
      "loss": 0.4184,
      "step": 114
    },
    {
      "epoch": 1.4375,
      "grad_norm": 0.7139426469802856,
      "learning_rate": 0.0001553191489361702,
      "loss": 0.5161,
      "step": 115
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.6485333442687988,
      "learning_rate": 0.0001548936170212766,
      "loss": 0.4784,
      "step": 116
    },
    {
      "epoch": 1.4625,
      "grad_norm": 0.7077224254608154,
      "learning_rate": 0.00015446808510638299,
      "loss": 0.5005,
      "step": 117
    },
    {
      "epoch": 1.475,
      "grad_norm": 0.7061143517494202,
      "learning_rate": 0.00015404255319148937,
      "loss": 0.433,
      "step": 118
    },
    {
      "epoch": 1.4875,
      "grad_norm": 0.6362001895904541,
      "learning_rate": 0.00015361702127659576,
      "loss": 0.39,
      "step": 119
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.742213249206543,
      "learning_rate": 0.00015319148936170213,
      "loss": 0.4926,
      "step": 120
    },
    {
      "epoch": 1.5125,
      "grad_norm": 0.6973474621772766,
      "learning_rate": 0.00015276595744680851,
      "loss": 0.4787,
      "step": 121
    },
    {
      "epoch": 1.525,
      "grad_norm": 0.6767258048057556,
      "learning_rate": 0.0001523404255319149,
      "loss": 0.3961,
      "step": 122
    },
    {
      "epoch": 1.5375,
      "grad_norm": 0.8685336112976074,
      "learning_rate": 0.0001519148936170213,
      "loss": 0.4258,
      "step": 123
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.8819417357444763,
      "learning_rate": 0.00015148936170212765,
      "loss": 0.4783,
      "step": 124
    },
    {
      "epoch": 1.5625,
      "grad_norm": 0.7913158535957336,
      "learning_rate": 0.00015106382978723407,
      "loss": 0.4467,
      "step": 125
    },
    {
      "epoch": 1.575,
      "grad_norm": 0.8557113409042358,
      "learning_rate": 0.00015063829787234043,
      "loss": 0.499,
      "step": 126
    },
    {
      "epoch": 1.5875,
      "grad_norm": 0.7897730469703674,
      "learning_rate": 0.00015021276595744682,
      "loss": 0.5141,
      "step": 127
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.7041512727737427,
      "learning_rate": 0.00014978723404255318,
      "loss": 0.4209,
      "step": 128
    },
    {
      "epoch": 1.6125,
      "grad_norm": 0.7383488416671753,
      "learning_rate": 0.0001493617021276596,
      "loss": 0.4483,
      "step": 129
    },
    {
      "epoch": 1.625,
      "grad_norm": 0.6808694005012512,
      "learning_rate": 0.00014893617021276596,
      "loss": 0.4233,
      "step": 130
    },
    {
      "epoch": 1.6375,
      "grad_norm": 0.6982035636901855,
      "learning_rate": 0.00014851063829787235,
      "loss": 0.5055,
      "step": 131
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.7038189768791199,
      "learning_rate": 0.0001480851063829787,
      "loss": 0.4551,
      "step": 132
    },
    {
      "epoch": 1.6625,
      "grad_norm": 0.6448100209236145,
      "learning_rate": 0.00014765957446808513,
      "loss": 0.4172,
      "step": 133
    },
    {
      "epoch": 1.675,
      "grad_norm": 0.7629950046539307,
      "learning_rate": 0.0001472340425531915,
      "loss": 0.4794,
      "step": 134
    },
    {
      "epoch": 1.6875,
      "grad_norm": 0.6743873357772827,
      "learning_rate": 0.00014680851063829788,
      "loss": 0.4207,
      "step": 135
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.6978509426116943,
      "learning_rate": 0.00014638297872340426,
      "loss": 0.4572,
      "step": 136
    },
    {
      "epoch": 1.7125,
      "grad_norm": 0.7238950133323669,
      "learning_rate": 0.00014595744680851065,
      "loss": 0.4391,
      "step": 137
    },
    {
      "epoch": 1.725,
      "grad_norm": 0.7844693660736084,
      "learning_rate": 0.00014553191489361704,
      "loss": 0.5253,
      "step": 138
    },
    {
      "epoch": 1.7375,
      "grad_norm": 0.7656735181808472,
      "learning_rate": 0.0001451063829787234,
      "loss": 0.4311,
      "step": 139
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.7286013960838318,
      "learning_rate": 0.0001446808510638298,
      "loss": 0.4573,
      "step": 140
    },
    {
      "epoch": 1.7625,
      "grad_norm": 0.7938070297241211,
      "learning_rate": 0.00014425531914893618,
      "loss": 0.5007,
      "step": 141
    },
    {
      "epoch": 1.775,
      "grad_norm": 0.7464184165000916,
      "learning_rate": 0.00014382978723404257,
      "loss": 0.4731,
      "step": 142
    },
    {
      "epoch": 1.7875,
      "grad_norm": 0.7830186486244202,
      "learning_rate": 0.00014340425531914893,
      "loss": 0.4451,
      "step": 143
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.6778513193130493,
      "learning_rate": 0.00014297872340425532,
      "loss": 0.4389,
      "step": 144
    },
    {
      "epoch": 1.8125,
      "grad_norm": 0.7377083897590637,
      "learning_rate": 0.0001425531914893617,
      "loss": 0.4734,
      "step": 145
    },
    {
      "epoch": 1.825,
      "grad_norm": 0.7428882718086243,
      "learning_rate": 0.0001421276595744681,
      "loss": 0.4741,
      "step": 146
    },
    {
      "epoch": 1.8375,
      "grad_norm": 0.7576317191123962,
      "learning_rate": 0.00014170212765957446,
      "loss": 0.4952,
      "step": 147
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.6553872227668762,
      "learning_rate": 0.00014127659574468085,
      "loss": 0.4051,
      "step": 148
    },
    {
      "epoch": 1.8625,
      "grad_norm": 0.7060889005661011,
      "learning_rate": 0.00014085106382978724,
      "loss": 0.4427,
      "step": 149
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.7130544185638428,
      "learning_rate": 0.00014042553191489363,
      "loss": 0.4519,
      "step": 150
    },
    {
      "epoch": 1.8875,
      "grad_norm": 0.7197202444076538,
      "learning_rate": 0.00014,
      "loss": 0.4621,
      "step": 151
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.7964305877685547,
      "learning_rate": 0.0001395744680851064,
      "loss": 0.5367,
      "step": 152
    },
    {
      "epoch": 1.9125,
      "grad_norm": 0.7618861794471741,
      "learning_rate": 0.00013914893617021277,
      "loss": 0.5231,
      "step": 153
    },
    {
      "epoch": 1.925,
      "grad_norm": 0.6833578944206238,
      "learning_rate": 0.00013872340425531915,
      "loss": 0.4354,
      "step": 154
    },
    {
      "epoch": 1.9375,
      "grad_norm": 0.6926935911178589,
      "learning_rate": 0.00013829787234042554,
      "loss": 0.4595,
      "step": 155
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.755796492099762,
      "learning_rate": 0.00013787234042553193,
      "loss": 0.4714,
      "step": 156
    },
    {
      "epoch": 1.9625,
      "grad_norm": 0.8154721856117249,
      "learning_rate": 0.00013744680851063832,
      "loss": 0.4514,
      "step": 157
    },
    {
      "epoch": 1.975,
      "grad_norm": 0.7180888652801514,
      "learning_rate": 0.00013702127659574468,
      "loss": 0.4487,
      "step": 158
    },
    {
      "epoch": 1.9875,
      "grad_norm": 0.7182486653327942,
      "learning_rate": 0.00013659574468085107,
      "loss": 0.4651,
      "step": 159
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8878002166748047,
      "learning_rate": 0.00013617021276595746,
      "loss": 0.4236,
      "step": 160
    },
    {
      "epoch": 2.0125,
      "grad_norm": 0.6806989908218384,
      "learning_rate": 0.00013574468085106385,
      "loss": 0.294,
      "step": 161
    },
    {
      "epoch": 2.025,
      "grad_norm": 0.7871399521827698,
      "learning_rate": 0.0001353191489361702,
      "loss": 0.2722,
      "step": 162
    },
    {
      "epoch": 2.0375,
      "grad_norm": 0.615963339805603,
      "learning_rate": 0.0001348936170212766,
      "loss": 0.2559,
      "step": 163
    },
    {
      "epoch": 2.05,
      "grad_norm": 0.6783286929130554,
      "learning_rate": 0.000134468085106383,
      "loss": 0.2697,
      "step": 164
    },
    {
      "epoch": 2.0625,
      "grad_norm": 0.69609135389328,
      "learning_rate": 0.00013404255319148938,
      "loss": 0.2952,
      "step": 165
    },
    {
      "epoch": 2.075,
      "grad_norm": 0.7641659379005432,
      "learning_rate": 0.00013361702127659574,
      "loss": 0.2784,
      "step": 166
    },
    {
      "epoch": 2.0875,
      "grad_norm": 0.7802071571350098,
      "learning_rate": 0.00013319148936170213,
      "loss": 0.268,
      "step": 167
    },
    {
      "epoch": 2.1,
      "grad_norm": 0.7029564380645752,
      "learning_rate": 0.00013276595744680852,
      "loss": 0.2239,
      "step": 168
    },
    {
      "epoch": 2.1125,
      "grad_norm": 0.8360403776168823,
      "learning_rate": 0.0001323404255319149,
      "loss": 0.2797,
      "step": 169
    },
    {
      "epoch": 2.125,
      "grad_norm": 1.003860592842102,
      "learning_rate": 0.00013191489361702127,
      "loss": 0.3147,
      "step": 170
    },
    {
      "epoch": 2.1375,
      "grad_norm": 0.8554705381393433,
      "learning_rate": 0.00013148936170212765,
      "loss": 0.2898,
      "step": 171
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.856831431388855,
      "learning_rate": 0.00013106382978723404,
      "loss": 0.244,
      "step": 172
    },
    {
      "epoch": 2.1625,
      "grad_norm": 0.840379536151886,
      "learning_rate": 0.00013063829787234043,
      "loss": 0.2511,
      "step": 173
    },
    {
      "epoch": 2.175,
      "grad_norm": 0.7467439770698547,
      "learning_rate": 0.00013021276595744682,
      "loss": 0.2394,
      "step": 174
    },
    {
      "epoch": 2.1875,
      "grad_norm": 0.8085935115814209,
      "learning_rate": 0.00012978723404255318,
      "loss": 0.2612,
      "step": 175
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.7966387271881104,
      "learning_rate": 0.0001293617021276596,
      "loss": 0.2646,
      "step": 176
    },
    {
      "epoch": 2.2125,
      "grad_norm": 0.8461523056030273,
      "learning_rate": 0.00012893617021276596,
      "loss": 0.2697,
      "step": 177
    },
    {
      "epoch": 2.225,
      "grad_norm": 0.7941194176673889,
      "learning_rate": 0.00012851063829787235,
      "loss": 0.2326,
      "step": 178
    },
    {
      "epoch": 2.2375,
      "grad_norm": 0.7666419744491577,
      "learning_rate": 0.00012808510638297874,
      "loss": 0.2247,
      "step": 179
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.788275420665741,
      "learning_rate": 0.00012765957446808513,
      "loss": 0.311,
      "step": 180
    },
    {
      "epoch": 2.2625,
      "grad_norm": 0.8255228996276855,
      "learning_rate": 0.0001272340425531915,
      "loss": 0.2662,
      "step": 181
    },
    {
      "epoch": 2.275,
      "grad_norm": 0.8195288181304932,
      "learning_rate": 0.00012680851063829788,
      "loss": 0.2683,
      "step": 182
    },
    {
      "epoch": 2.2875,
      "grad_norm": 0.7774321436882019,
      "learning_rate": 0.00012638297872340427,
      "loss": 0.2431,
      "step": 183
    },
    {
      "epoch": 2.3,
      "grad_norm": 0.8164303302764893,
      "learning_rate": 0.00012595744680851065,
      "loss": 0.2674,
      "step": 184
    },
    {
      "epoch": 2.3125,
      "grad_norm": 0.8182231783866882,
      "learning_rate": 0.00012553191489361702,
      "loss": 0.2922,
      "step": 185
    },
    {
      "epoch": 2.325,
      "grad_norm": 0.7818589806556702,
      "learning_rate": 0.0001251063829787234,
      "loss": 0.2558,
      "step": 186
    },
    {
      "epoch": 2.3375,
      "grad_norm": 0.809130847454071,
      "learning_rate": 0.0001246808510638298,
      "loss": 0.2387,
      "step": 187
    },
    {
      "epoch": 2.35,
      "grad_norm": 0.7757312059402466,
      "learning_rate": 0.00012425531914893618,
      "loss": 0.2685,
      "step": 188
    },
    {
      "epoch": 2.3625,
      "grad_norm": 0.8028156757354736,
      "learning_rate": 0.00012382978723404254,
      "loss": 0.2618,
      "step": 189
    },
    {
      "epoch": 2.375,
      "grad_norm": 0.8730686902999878,
      "learning_rate": 0.00012340425531914893,
      "loss": 0.2668,
      "step": 190
    },
    {
      "epoch": 2.3875,
      "grad_norm": 0.8512089252471924,
      "learning_rate": 0.00012297872340425535,
      "loss": 0.2687,
      "step": 191
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.8351570963859558,
      "learning_rate": 0.0001225531914893617,
      "loss": 0.2612,
      "step": 192
    },
    {
      "epoch": 2.4,
      "eval_loss": 0.7145271301269531,
      "eval_runtime": 11.4074,
      "eval_samples_per_second": 3.156,
      "eval_steps_per_second": 1.578,
      "step": 192
    },
    {
      "epoch": 2.4125,
      "grad_norm": 0.7363073825836182,
      "learning_rate": 0.0001221276595744681,
      "loss": 0.2462,
      "step": 193
    },
    {
      "epoch": 2.425,
      "grad_norm": 0.7738860845565796,
      "learning_rate": 0.00012170212765957448,
      "loss": 0.2455,
      "step": 194
    },
    {
      "epoch": 2.4375,
      "grad_norm": 0.7793389558792114,
      "learning_rate": 0.00012127659574468086,
      "loss": 0.2357,
      "step": 195
    },
    {
      "epoch": 2.45,
      "grad_norm": 1.0476254224777222,
      "learning_rate": 0.00012085106382978724,
      "loss": 0.3119,
      "step": 196
    },
    {
      "epoch": 2.4625,
      "grad_norm": 0.9584843516349792,
      "learning_rate": 0.00012042553191489363,
      "loss": 0.2931,
      "step": 197
    },
    {
      "epoch": 2.475,
      "grad_norm": 0.7162908911705017,
      "learning_rate": 0.00012,
      "loss": 0.2134,
      "step": 198
    },
    {
      "epoch": 2.4875,
      "grad_norm": 0.7601090669631958,
      "learning_rate": 0.00011957446808510639,
      "loss": 0.2444,
      "step": 199
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.7643441557884216,
      "learning_rate": 0.00011914893617021277,
      "loss": 0.2667,
      "step": 200
    },
    {
      "epoch": 2.5125,
      "grad_norm": 0.7849262952804565,
      "learning_rate": 0.00011872340425531916,
      "loss": 0.2629,
      "step": 201
    },
    {
      "epoch": 2.525,
      "grad_norm": 0.8611546754837036,
      "learning_rate": 0.00011829787234042553,
      "loss": 0.2728,
      "step": 202
    },
    {
      "epoch": 2.5375,
      "grad_norm": 0.8595278263092041,
      "learning_rate": 0.00011787234042553192,
      "loss": 0.2749,
      "step": 203
    },
    {
      "epoch": 2.55,
      "grad_norm": 0.778231143951416,
      "learning_rate": 0.0001174468085106383,
      "loss": 0.2779,
      "step": 204
    },
    {
      "epoch": 2.5625,
      "grad_norm": 0.9688648581504822,
      "learning_rate": 0.00011702127659574468,
      "loss": 0.3187,
      "step": 205
    },
    {
      "epoch": 2.575,
      "grad_norm": 0.842782199382782,
      "learning_rate": 0.00011659574468085106,
      "loss": 0.2624,
      "step": 206
    },
    {
      "epoch": 2.5875,
      "grad_norm": 0.7811837792396545,
      "learning_rate": 0.00011617021276595745,
      "loss": 0.2753,
      "step": 207
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.8029890060424805,
      "learning_rate": 0.00011574468085106382,
      "loss": 0.2704,
      "step": 208
    },
    {
      "epoch": 2.6125,
      "grad_norm": 0.8911982178688049,
      "learning_rate": 0.00011531914893617021,
      "loss": 0.2821,
      "step": 209
    },
    {
      "epoch": 2.625,
      "grad_norm": 0.7516865730285645,
      "learning_rate": 0.00011489361702127661,
      "loss": 0.2411,
      "step": 210
    },
    {
      "epoch": 2.6375,
      "grad_norm": 0.8153129816055298,
      "learning_rate": 0.00011446808510638299,
      "loss": 0.2939,
      "step": 211
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.8791115283966064,
      "learning_rate": 0.00011404255319148938,
      "loss": 0.2528,
      "step": 212
    },
    {
      "epoch": 2.6625,
      "grad_norm": 0.8871167898178101,
      "learning_rate": 0.00011361702127659575,
      "loss": 0.28,
      "step": 213
    },
    {
      "epoch": 2.675,
      "grad_norm": 0.8256612420082092,
      "learning_rate": 0.00011319148936170214,
      "loss": 0.3028,
      "step": 214
    },
    {
      "epoch": 2.6875,
      "grad_norm": 0.8403842449188232,
      "learning_rate": 0.00011276595744680852,
      "loss": 0.2826,
      "step": 215
    },
    {
      "epoch": 2.7,
      "grad_norm": 0.773061215877533,
      "learning_rate": 0.0001123404255319149,
      "loss": 0.2667,
      "step": 216
    },
    {
      "epoch": 2.7125,
      "grad_norm": 0.7826652526855469,
      "learning_rate": 0.00011191489361702128,
      "loss": 0.249,
      "step": 217
    },
    {
      "epoch": 2.725,
      "grad_norm": 0.7895304560661316,
      "learning_rate": 0.00011148936170212767,
      "loss": 0.2549,
      "step": 218
    },
    {
      "epoch": 2.7375,
      "grad_norm": 0.778076171875,
      "learning_rate": 0.00011106382978723405,
      "loss": 0.2731,
      "step": 219
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.8927327990531921,
      "learning_rate": 0.00011063829787234043,
      "loss": 0.2986,
      "step": 220
    },
    {
      "epoch": 2.7625,
      "grad_norm": 0.9613041877746582,
      "learning_rate": 0.00011021276595744681,
      "loss": 0.3228,
      "step": 221
    },
    {
      "epoch": 2.775,
      "grad_norm": 0.7272498607635498,
      "learning_rate": 0.0001097872340425532,
      "loss": 0.2297,
      "step": 222
    },
    {
      "epoch": 2.7875,
      "grad_norm": 0.8113637566566467,
      "learning_rate": 0.00010936170212765957,
      "loss": 0.2812,
      "step": 223
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.9033439755439758,
      "learning_rate": 0.00010893617021276596,
      "loss": 0.3016,
      "step": 224
    },
    {
      "epoch": 2.8125,
      "grad_norm": 0.7996534109115601,
      "learning_rate": 0.00010851063829787234,
      "loss": 0.2496,
      "step": 225
    },
    {
      "epoch": 2.825,
      "grad_norm": 0.7771035432815552,
      "learning_rate": 0.00010808510638297873,
      "loss": 0.2852,
      "step": 226
    },
    {
      "epoch": 2.8375,
      "grad_norm": 0.7756218314170837,
      "learning_rate": 0.00010765957446808513,
      "loss": 0.2703,
      "step": 227
    },
    {
      "epoch": 2.85,
      "grad_norm": 0.8187890648841858,
      "learning_rate": 0.00010723404255319149,
      "loss": 0.2855,
      "step": 228
    },
    {
      "epoch": 2.8625,
      "grad_norm": 0.858629047870636,
      "learning_rate": 0.00010680851063829789,
      "loss": 0.2628,
      "step": 229
    },
    {
      "epoch": 2.875,
      "grad_norm": 0.83708655834198,
      "learning_rate": 0.00010638297872340425,
      "loss": 0.2606,
      "step": 230
    },
    {
      "epoch": 2.8875,
      "grad_norm": 0.7959954738616943,
      "learning_rate": 0.00010595744680851066,
      "loss": 0.3034,
      "step": 231
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.7901068329811096,
      "learning_rate": 0.00010553191489361702,
      "loss": 0.2259,
      "step": 232
    },
    {
      "epoch": 2.9125,
      "grad_norm": 0.8610655665397644,
      "learning_rate": 0.00010510638297872342,
      "loss": 0.2815,
      "step": 233
    },
    {
      "epoch": 2.925,
      "grad_norm": 0.8414681553840637,
      "learning_rate": 0.00010468085106382978,
      "loss": 0.2969,
      "step": 234
    },
    {
      "epoch": 2.9375,
      "grad_norm": 0.8410956263542175,
      "learning_rate": 0.00010425531914893618,
      "loss": 0.2742,
      "step": 235
    },
    {
      "epoch": 2.95,
      "grad_norm": 0.8325995802879333,
      "learning_rate": 0.00010382978723404255,
      "loss": 0.2653,
      "step": 236
    },
    {
      "epoch": 2.9625,
      "grad_norm": 0.7669159173965454,
      "learning_rate": 0.00010340425531914895,
      "loss": 0.2579,
      "step": 237
    },
    {
      "epoch": 2.975,
      "grad_norm": 0.7911613583564758,
      "learning_rate": 0.00010297872340425532,
      "loss": 0.3053,
      "step": 238
    },
    {
      "epoch": 2.9875,
      "grad_norm": 0.7572367787361145,
      "learning_rate": 0.00010255319148936171,
      "loss": 0.2515,
      "step": 239
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.0301828384399414,
      "learning_rate": 0.00010212765957446809,
      "loss": 0.2728,
      "step": 240
    },
    {
      "epoch": 3.0125,
      "grad_norm": 0.6737964153289795,
      "learning_rate": 0.00010170212765957448,
      "loss": 0.1407,
      "step": 241
    },
    {
      "epoch": 3.025,
      "grad_norm": 0.7623804807662964,
      "learning_rate": 0.00010127659574468085,
      "loss": 0.1287,
      "step": 242
    },
    {
      "epoch": 3.0375,
      "grad_norm": 0.7205766439437866,
      "learning_rate": 0.00010085106382978724,
      "loss": 0.1547,
      "step": 243
    },
    {
      "epoch": 3.05,
      "grad_norm": 0.905816912651062,
      "learning_rate": 0.00010042553191489362,
      "loss": 0.1846,
      "step": 244
    },
    {
      "epoch": 3.0625,
      "grad_norm": 0.7594558596611023,
      "learning_rate": 0.0001,
      "loss": 0.1563,
      "step": 245
    },
    {
      "epoch": 3.075,
      "grad_norm": 0.6924287676811218,
      "learning_rate": 9.95744680851064e-05,
      "loss": 0.1445,
      "step": 246
    },
    {
      "epoch": 3.0875,
      "grad_norm": 0.6288279891014099,
      "learning_rate": 9.914893617021277e-05,
      "loss": 0.1235,
      "step": 247
    },
    {
      "epoch": 3.1,
      "grad_norm": 0.6929881572723389,
      "learning_rate": 9.872340425531916e-05,
      "loss": 0.1379,
      "step": 248
    },
    {
      "epoch": 3.1125,
      "grad_norm": 0.9299332499504089,
      "learning_rate": 9.829787234042553e-05,
      "loss": 0.1521,
      "step": 249
    },
    {
      "epoch": 3.125,
      "grad_norm": 0.9375365376472473,
      "learning_rate": 9.787234042553192e-05,
      "loss": 0.145,
      "step": 250
    },
    {
      "epoch": 3.1375,
      "grad_norm": 0.8231996893882751,
      "learning_rate": 9.74468085106383e-05,
      "loss": 0.1228,
      "step": 251
    },
    {
      "epoch": 3.15,
      "grad_norm": 0.8826015591621399,
      "learning_rate": 9.702127659574469e-05,
      "loss": 0.1487,
      "step": 252
    },
    {
      "epoch": 3.1625,
      "grad_norm": 0.8321478962898254,
      "learning_rate": 9.659574468085106e-05,
      "loss": 0.1346,
      "step": 253
    },
    {
      "epoch": 3.175,
      "grad_norm": 1.043582558631897,
      "learning_rate": 9.617021276595745e-05,
      "loss": 0.1587,
      "step": 254
    },
    {
      "epoch": 3.1875,
      "grad_norm": 0.8663302063941956,
      "learning_rate": 9.574468085106384e-05,
      "loss": 0.1253,
      "step": 255
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.7811707854270935,
      "learning_rate": 9.531914893617023e-05,
      "loss": 0.1427,
      "step": 256
    },
    {
      "epoch": 3.2125,
      "grad_norm": 0.8035542964935303,
      "learning_rate": 9.48936170212766e-05,
      "loss": 0.1374,
      "step": 257
    },
    {
      "epoch": 3.225,
      "grad_norm": 0.8229654431343079,
      "learning_rate": 9.446808510638299e-05,
      "loss": 0.124,
      "step": 258
    },
    {
      "epoch": 3.2375,
      "grad_norm": 0.7131748199462891,
      "learning_rate": 9.404255319148937e-05,
      "loss": 0.1197,
      "step": 259
    },
    {
      "epoch": 3.25,
      "grad_norm": 0.7480083703994751,
      "learning_rate": 9.361702127659576e-05,
      "loss": 0.1336,
      "step": 260
    },
    {
      "epoch": 3.2625,
      "grad_norm": 0.7220382690429688,
      "learning_rate": 9.319148936170213e-05,
      "loss": 0.128,
      "step": 261
    },
    {
      "epoch": 3.275,
      "grad_norm": 0.8449149131774902,
      "learning_rate": 9.276595744680852e-05,
      "loss": 0.1661,
      "step": 262
    },
    {
      "epoch": 3.2875,
      "grad_norm": 0.6896558403968811,
      "learning_rate": 9.23404255319149e-05,
      "loss": 0.1231,
      "step": 263
    },
    {
      "epoch": 3.3,
      "grad_norm": 0.7313699126243591,
      "learning_rate": 9.191489361702128e-05,
      "loss": 0.1532,
      "step": 264
    },
    {
      "epoch": 3.3125,
      "grad_norm": 0.8504534959793091,
      "learning_rate": 9.148936170212766e-05,
      "loss": 0.1447,
      "step": 265
    },
    {
      "epoch": 3.325,
      "grad_norm": 0.8592658638954163,
      "learning_rate": 9.106382978723405e-05,
      "loss": 0.1429,
      "step": 266
    },
    {
      "epoch": 3.3375,
      "grad_norm": 0.7510311007499695,
      "learning_rate": 9.063829787234042e-05,
      "loss": 0.1191,
      "step": 267
    },
    {
      "epoch": 3.35,
      "grad_norm": 0.772495687007904,
      "learning_rate": 9.021276595744681e-05,
      "loss": 0.1525,
      "step": 268
    },
    {
      "epoch": 3.3625,
      "grad_norm": 0.8163837790489197,
      "learning_rate": 8.978723404255319e-05,
      "loss": 0.1463,
      "step": 269
    },
    {
      "epoch": 3.375,
      "grad_norm": 0.8092824816703796,
      "learning_rate": 8.936170212765958e-05,
      "loss": 0.1373,
      "step": 270
    },
    {
      "epoch": 3.3875,
      "grad_norm": 0.8845824003219604,
      "learning_rate": 8.893617021276595e-05,
      "loss": 0.1389,
      "step": 271
    },
    {
      "epoch": 3.4,
      "grad_norm": 0.8480741381645203,
      "learning_rate": 8.851063829787234e-05,
      "loss": 0.1481,
      "step": 272
    },
    {
      "epoch": 3.4125,
      "grad_norm": 0.727211594581604,
      "learning_rate": 8.808510638297873e-05,
      "loss": 0.1319,
      "step": 273
    },
    {
      "epoch": 3.425,
      "grad_norm": 0.8201771378517151,
      "learning_rate": 8.765957446808512e-05,
      "loss": 0.1471,
      "step": 274
    },
    {
      "epoch": 3.4375,
      "grad_norm": 0.8638917803764343,
      "learning_rate": 8.723404255319149e-05,
      "loss": 0.1558,
      "step": 275
    },
    {
      "epoch": 3.45,
      "grad_norm": 0.9420880079269409,
      "learning_rate": 8.680851063829788e-05,
      "loss": 0.1336,
      "step": 276
    },
    {
      "epoch": 3.4625,
      "grad_norm": 0.827743411064148,
      "learning_rate": 8.638297872340426e-05,
      "loss": 0.1351,
      "step": 277
    },
    {
      "epoch": 3.475,
      "grad_norm": 0.9593076705932617,
      "learning_rate": 8.595744680851064e-05,
      "loss": 0.1355,
      "step": 278
    },
    {
      "epoch": 3.4875,
      "grad_norm": 0.8220702409744263,
      "learning_rate": 8.553191489361702e-05,
      "loss": 0.1358,
      "step": 279
    },
    {
      "epoch": 3.5,
      "grad_norm": 0.977692186832428,
      "learning_rate": 8.510638297872341e-05,
      "loss": 0.1694,
      "step": 280
    },
    {
      "epoch": 3.5125,
      "grad_norm": 0.7905257344245911,
      "learning_rate": 8.46808510638298e-05,
      "loss": 0.1275,
      "step": 281
    },
    {
      "epoch": 3.525,
      "grad_norm": 0.8010748624801636,
      "learning_rate": 8.425531914893617e-05,
      "loss": 0.1375,
      "step": 282
    },
    {
      "epoch": 3.5375,
      "grad_norm": 0.7461961507797241,
      "learning_rate": 8.382978723404256e-05,
      "loss": 0.1371,
      "step": 283
    },
    {
      "epoch": 3.55,
      "grad_norm": 0.8024889826774597,
      "learning_rate": 8.340425531914894e-05,
      "loss": 0.1368,
      "step": 284
    },
    {
      "epoch": 3.5625,
      "grad_norm": 0.8584401607513428,
      "learning_rate": 8.297872340425533e-05,
      "loss": 0.1492,
      "step": 285
    },
    {
      "epoch": 3.575,
      "grad_norm": 0.8518531918525696,
      "learning_rate": 8.25531914893617e-05,
      "loss": 0.1533,
      "step": 286
    },
    {
      "epoch": 3.5875,
      "grad_norm": 0.8000539541244507,
      "learning_rate": 8.212765957446809e-05,
      "loss": 0.154,
      "step": 287
    },
    {
      "epoch": 3.6,
      "grad_norm": 0.9574905037879944,
      "learning_rate": 8.170212765957446e-05,
      "loss": 0.1689,
      "step": 288
    },
    {
      "epoch": 3.6,
      "eval_loss": 0.8238211870193481,
      "eval_runtime": 10.9659,
      "eval_samples_per_second": 3.283,
      "eval_steps_per_second": 1.641,
      "step": 288
    },
    {
      "epoch": 3.6125,
      "grad_norm": 0.8474534749984741,
      "learning_rate": 8.127659574468085e-05,
      "loss": 0.1524,
      "step": 289
    },
    {
      "epoch": 3.625,
      "grad_norm": 0.7077217698097229,
      "learning_rate": 8.085106382978723e-05,
      "loss": 0.1369,
      "step": 290
    },
    {
      "epoch": 3.6375,
      "grad_norm": 0.8012158274650574,
      "learning_rate": 8.042553191489363e-05,
      "loss": 0.1261,
      "step": 291
    },
    {
      "epoch": 3.65,
      "grad_norm": 0.7327490448951721,
      "learning_rate": 8e-05,
      "loss": 0.1338,
      "step": 292
    },
    {
      "epoch": 3.6625,
      "grad_norm": 0.7753359079360962,
      "learning_rate": 7.95744680851064e-05,
      "loss": 0.1439,
      "step": 293
    },
    {
      "epoch": 3.675,
      "grad_norm": 0.8457762598991394,
      "learning_rate": 7.914893617021277e-05,
      "loss": 0.1343,
      "step": 294
    },
    {
      "epoch": 3.6875,
      "grad_norm": 0.8116350173950195,
      "learning_rate": 7.872340425531916e-05,
      "loss": 0.1514,
      "step": 295
    },
    {
      "epoch": 3.7,
      "grad_norm": 0.8146523833274841,
      "learning_rate": 7.829787234042553e-05,
      "loss": 0.1485,
      "step": 296
    },
    {
      "epoch": 3.7125,
      "grad_norm": 0.6425414085388184,
      "learning_rate": 7.787234042553192e-05,
      "loss": 0.0959,
      "step": 297
    },
    {
      "epoch": 3.725,
      "grad_norm": 0.9455553293228149,
      "learning_rate": 7.74468085106383e-05,
      "loss": 0.1597,
      "step": 298
    },
    {
      "epoch": 3.7375,
      "grad_norm": 0.8597568273544312,
      "learning_rate": 7.702127659574469e-05,
      "loss": 0.1614,
      "step": 299
    },
    {
      "epoch": 3.75,
      "grad_norm": 0.8073698878288269,
      "learning_rate": 7.659574468085106e-05,
      "loss": 0.1353,
      "step": 300
    },
    {
      "epoch": 3.7625,
      "grad_norm": 0.7673254609107971,
      "learning_rate": 7.617021276595745e-05,
      "loss": 0.1188,
      "step": 301
    },
    {
      "epoch": 3.775,
      "grad_norm": 0.8758047819137573,
      "learning_rate": 7.574468085106383e-05,
      "loss": 0.1839,
      "step": 302
    },
    {
      "epoch": 3.7875,
      "grad_norm": 0.9272107481956482,
      "learning_rate": 7.531914893617022e-05,
      "loss": 0.1497,
      "step": 303
    },
    {
      "epoch": 3.8,
      "grad_norm": 0.7681931257247925,
      "learning_rate": 7.489361702127659e-05,
      "loss": 0.1469,
      "step": 304
    },
    {
      "epoch": 3.8125,
      "grad_norm": 0.8682704567909241,
      "learning_rate": 7.446808510638298e-05,
      "loss": 0.1618,
      "step": 305
    },
    {
      "epoch": 3.825,
      "grad_norm": 1.0033032894134521,
      "learning_rate": 7.404255319148935e-05,
      "loss": 0.1621,
      "step": 306
    },
    {
      "epoch": 3.8375,
      "grad_norm": 1.0055568218231201,
      "learning_rate": 7.361702127659574e-05,
      "loss": 0.1735,
      "step": 307
    },
    {
      "epoch": 3.85,
      "grad_norm": 0.7715242505073547,
      "learning_rate": 7.319148936170213e-05,
      "loss": 0.1466,
      "step": 308
    },
    {
      "epoch": 3.8625,
      "grad_norm": 0.8541414141654968,
      "learning_rate": 7.276595744680852e-05,
      "loss": 0.1411,
      "step": 309
    },
    {
      "epoch": 3.875,
      "grad_norm": 0.7181360721588135,
      "learning_rate": 7.23404255319149e-05,
      "loss": 0.1391,
      "step": 310
    },
    {
      "epoch": 3.8875,
      "grad_norm": 0.7194767594337463,
      "learning_rate": 7.191489361702129e-05,
      "loss": 0.1379,
      "step": 311
    },
    {
      "epoch": 3.9,
      "grad_norm": 0.7853153347969055,
      "learning_rate": 7.148936170212766e-05,
      "loss": 0.1468,
      "step": 312
    },
    {
      "epoch": 3.9125,
      "grad_norm": 0.7511010766029358,
      "learning_rate": 7.106382978723405e-05,
      "loss": 0.1377,
      "step": 313
    },
    {
      "epoch": 3.925,
      "grad_norm": 0.7181860208511353,
      "learning_rate": 7.063829787234042e-05,
      "loss": 0.1389,
      "step": 314
    },
    {
      "epoch": 3.9375,
      "grad_norm": 0.7916765809059143,
      "learning_rate": 7.021276595744681e-05,
      "loss": 0.1527,
      "step": 315
    },
    {
      "epoch": 3.95,
      "grad_norm": 0.6324968338012695,
      "learning_rate": 6.97872340425532e-05,
      "loss": 0.1062,
      "step": 316
    },
    {
      "epoch": 3.9625,
      "grad_norm": 0.8582820296287537,
      "learning_rate": 6.936170212765958e-05,
      "loss": 0.1788,
      "step": 317
    },
    {
      "epoch": 3.975,
      "grad_norm": 0.7167566418647766,
      "learning_rate": 6.893617021276597e-05,
      "loss": 0.1376,
      "step": 318
    },
    {
      "epoch": 3.9875,
      "grad_norm": 0.8698075413703918,
      "learning_rate": 6.851063829787234e-05,
      "loss": 0.1651,
      "step": 319
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.817995548248291,
      "learning_rate": 6.808510638297873e-05,
      "loss": 0.1072,
      "step": 320
    },
    {
      "epoch": 4.0125,
      "grad_norm": 0.6073539853096008,
      "learning_rate": 6.76595744680851e-05,
      "loss": 0.0881,
      "step": 321
    },
    {
      "epoch": 4.025,
      "grad_norm": 0.6550564765930176,
      "learning_rate": 6.72340425531915e-05,
      "loss": 0.1131,
      "step": 322
    },
    {
      "epoch": 4.0375,
      "grad_norm": 0.6034475564956665,
      "learning_rate": 6.680851063829787e-05,
      "loss": 0.0758,
      "step": 323
    },
    {
      "epoch": 4.05,
      "grad_norm": 0.81227707862854,
      "learning_rate": 6.638297872340426e-05,
      "loss": 0.0951,
      "step": 324
    },
    {
      "epoch": 4.0625,
      "grad_norm": 0.5546196699142456,
      "learning_rate": 6.595744680851063e-05,
      "loss": 0.0759,
      "step": 325
    },
    {
      "epoch": 4.075,
      "grad_norm": 0.5332831740379333,
      "learning_rate": 6.553191489361702e-05,
      "loss": 0.0696,
      "step": 326
    },
    {
      "epoch": 4.0875,
      "grad_norm": 0.7316128611564636,
      "learning_rate": 6.510638297872341e-05,
      "loss": 0.0938,
      "step": 327
    },
    {
      "epoch": 4.1,
      "grad_norm": 0.6615329384803772,
      "learning_rate": 6.46808510638298e-05,
      "loss": 0.0805,
      "step": 328
    },
    {
      "epoch": 4.1125,
      "grad_norm": 0.6748345494270325,
      "learning_rate": 6.425531914893617e-05,
      "loss": 0.0708,
      "step": 329
    },
    {
      "epoch": 4.125,
      "grad_norm": 0.692646861076355,
      "learning_rate": 6.382978723404256e-05,
      "loss": 0.0782,
      "step": 330
    },
    {
      "epoch": 4.1375,
      "grad_norm": 0.7430183291435242,
      "learning_rate": 6.340425531914894e-05,
      "loss": 0.0723,
      "step": 331
    },
    {
      "epoch": 4.15,
      "grad_norm": 0.6600062251091003,
      "learning_rate": 6.297872340425533e-05,
      "loss": 0.0686,
      "step": 332
    },
    {
      "epoch": 4.1625,
      "grad_norm": 0.8680083155632019,
      "learning_rate": 6.25531914893617e-05,
      "loss": 0.0822,
      "step": 333
    },
    {
      "epoch": 4.175,
      "grad_norm": 0.7598772644996643,
      "learning_rate": 6.212765957446809e-05,
      "loss": 0.083,
      "step": 334
    },
    {
      "epoch": 4.1875,
      "grad_norm": 0.769776463508606,
      "learning_rate": 6.170212765957447e-05,
      "loss": 0.076,
      "step": 335
    },
    {
      "epoch": 4.2,
      "grad_norm": 0.6409958600997925,
      "learning_rate": 6.127659574468086e-05,
      "loss": 0.0712,
      "step": 336
    },
    {
      "epoch": 4.2125,
      "grad_norm": 0.8126680254936218,
      "learning_rate": 6.085106382978724e-05,
      "loss": 0.0767,
      "step": 337
    },
    {
      "epoch": 4.225,
      "grad_norm": 0.7009455561637878,
      "learning_rate": 6.042553191489362e-05,
      "loss": 0.0633,
      "step": 338
    },
    {
      "epoch": 4.2375,
      "grad_norm": 0.5959800481796265,
      "learning_rate": 6e-05,
      "loss": 0.0606,
      "step": 339
    },
    {
      "epoch": 4.25,
      "grad_norm": 0.5804210901260376,
      "learning_rate": 5.9574468085106384e-05,
      "loss": 0.0601,
      "step": 340
    },
    {
      "epoch": 4.2625,
      "grad_norm": 0.6084879040718079,
      "learning_rate": 5.9148936170212766e-05,
      "loss": 0.0693,
      "step": 341
    },
    {
      "epoch": 4.275,
      "grad_norm": 0.6810378432273865,
      "learning_rate": 5.872340425531915e-05,
      "loss": 0.0656,
      "step": 342
    },
    {
      "epoch": 4.2875,
      "grad_norm": 0.6785478591918945,
      "learning_rate": 5.829787234042553e-05,
      "loss": 0.0681,
      "step": 343
    },
    {
      "epoch": 4.3,
      "grad_norm": 0.6475040912628174,
      "learning_rate": 5.787234042553191e-05,
      "loss": 0.0731,
      "step": 344
    },
    {
      "epoch": 4.3125,
      "grad_norm": 0.6373022198677063,
      "learning_rate": 5.744680851063831e-05,
      "loss": 0.0707,
      "step": 345
    },
    {
      "epoch": 4.325,
      "grad_norm": 0.6542766094207764,
      "learning_rate": 5.702127659574469e-05,
      "loss": 0.0564,
      "step": 346
    },
    {
      "epoch": 4.3375,
      "grad_norm": 0.6114728450775146,
      "learning_rate": 5.659574468085107e-05,
      "loss": 0.0674,
      "step": 347
    },
    {
      "epoch": 4.35,
      "grad_norm": 0.6630443930625916,
      "learning_rate": 5.617021276595745e-05,
      "loss": 0.0614,
      "step": 348
    },
    {
      "epoch": 4.3625,
      "grad_norm": 0.6126697659492493,
      "learning_rate": 5.5744680851063835e-05,
      "loss": 0.0594,
      "step": 349
    },
    {
      "epoch": 4.375,
      "grad_norm": 0.5768252015113831,
      "learning_rate": 5.531914893617022e-05,
      "loss": 0.0569,
      "step": 350
    },
    {
      "epoch": 4.3875,
      "grad_norm": 0.7498059272766113,
      "learning_rate": 5.48936170212766e-05,
      "loss": 0.0777,
      "step": 351
    },
    {
      "epoch": 4.4,
      "grad_norm": 0.6269453167915344,
      "learning_rate": 5.446808510638298e-05,
      "loss": 0.0613,
      "step": 352
    },
    {
      "epoch": 4.4125,
      "grad_norm": 0.6231211423873901,
      "learning_rate": 5.404255319148936e-05,
      "loss": 0.0738,
      "step": 353
    },
    {
      "epoch": 4.425,
      "grad_norm": 0.7814548015594482,
      "learning_rate": 5.3617021276595745e-05,
      "loss": 0.0687,
      "step": 354
    },
    {
      "epoch": 4.4375,
      "grad_norm": 0.8020561337471008,
      "learning_rate": 5.319148936170213e-05,
      "loss": 0.0848,
      "step": 355
    },
    {
      "epoch": 4.45,
      "grad_norm": 0.6911760568618774,
      "learning_rate": 5.276595744680851e-05,
      "loss": 0.0741,
      "step": 356
    },
    {
      "epoch": 4.4625,
      "grad_norm": 0.7701168060302734,
      "learning_rate": 5.234042553191489e-05,
      "loss": 0.0679,
      "step": 357
    },
    {
      "epoch": 4.475,
      "grad_norm": 0.6995530128479004,
      "learning_rate": 5.191489361702127e-05,
      "loss": 0.0628,
      "step": 358
    },
    {
      "epoch": 4.4875,
      "grad_norm": 0.6969619393348694,
      "learning_rate": 5.148936170212766e-05,
      "loss": 0.0701,
      "step": 359
    },
    {
      "epoch": 4.5,
      "grad_norm": 0.6292584538459778,
      "learning_rate": 5.1063829787234044e-05,
      "loss": 0.0651,
      "step": 360
    },
    {
      "epoch": 4.5125,
      "grad_norm": 0.7294819951057434,
      "learning_rate": 5.0638297872340426e-05,
      "loss": 0.0698,
      "step": 361
    },
    {
      "epoch": 4.525,
      "grad_norm": 0.6159396171569824,
      "learning_rate": 5.021276595744681e-05,
      "loss": 0.0602,
      "step": 362
    },
    {
      "epoch": 4.5375,
      "grad_norm": 0.7271333336830139,
      "learning_rate": 4.97872340425532e-05,
      "loss": 0.0715,
      "step": 363
    },
    {
      "epoch": 4.55,
      "grad_norm": 0.9358235597610474,
      "learning_rate": 4.936170212765958e-05,
      "loss": 0.0888,
      "step": 364
    },
    {
      "epoch": 4.5625,
      "grad_norm": 0.6776322722434998,
      "learning_rate": 4.893617021276596e-05,
      "loss": 0.066,
      "step": 365
    },
    {
      "epoch": 4.575,
      "grad_norm": 0.8353095054626465,
      "learning_rate": 4.851063829787234e-05,
      "loss": 0.0925,
      "step": 366
    },
    {
      "epoch": 4.5875,
      "grad_norm": 0.6650261878967285,
      "learning_rate": 4.8085106382978725e-05,
      "loss": 0.0637,
      "step": 367
    },
    {
      "epoch": 4.6,
      "grad_norm": 0.7034966945648193,
      "learning_rate": 4.7659574468085114e-05,
      "loss": 0.0744,
      "step": 368
    },
    {
      "epoch": 4.6125,
      "grad_norm": 0.6829352378845215,
      "learning_rate": 4.7234042553191496e-05,
      "loss": 0.0695,
      "step": 369
    },
    {
      "epoch": 4.625,
      "grad_norm": 0.5907257795333862,
      "learning_rate": 4.680851063829788e-05,
      "loss": 0.0625,
      "step": 370
    },
    {
      "epoch": 4.6375,
      "grad_norm": 0.6723539233207703,
      "learning_rate": 4.638297872340426e-05,
      "loss": 0.0721,
      "step": 371
    },
    {
      "epoch": 4.65,
      "grad_norm": 0.7118985056877136,
      "learning_rate": 4.595744680851064e-05,
      "loss": 0.0807,
      "step": 372
    },
    {
      "epoch": 4.6625,
      "grad_norm": 0.5988577008247375,
      "learning_rate": 4.5531914893617024e-05,
      "loss": 0.065,
      "step": 373
    },
    {
      "epoch": 4.675,
      "grad_norm": 0.6269676685333252,
      "learning_rate": 4.5106382978723406e-05,
      "loss": 0.0702,
      "step": 374
    },
    {
      "epoch": 4.6875,
      "grad_norm": 0.7419471740722656,
      "learning_rate": 4.468085106382979e-05,
      "loss": 0.0938,
      "step": 375
    },
    {
      "epoch": 4.7,
      "grad_norm": 0.5768789052963257,
      "learning_rate": 4.425531914893617e-05,
      "loss": 0.0585,
      "step": 376
    },
    {
      "epoch": 4.7125,
      "grad_norm": 0.6381940245628357,
      "learning_rate": 4.382978723404256e-05,
      "loss": 0.0634,
      "step": 377
    },
    {
      "epoch": 4.725,
      "grad_norm": 0.6856435537338257,
      "learning_rate": 4.340425531914894e-05,
      "loss": 0.076,
      "step": 378
    },
    {
      "epoch": 4.7375,
      "grad_norm": 0.6991595029830933,
      "learning_rate": 4.297872340425532e-05,
      "loss": 0.0828,
      "step": 379
    },
    {
      "epoch": 4.75,
      "grad_norm": 0.7512986660003662,
      "learning_rate": 4.2553191489361704e-05,
      "loss": 0.0748,
      "step": 380
    },
    {
      "epoch": 4.7625,
      "grad_norm": 0.6096968650817871,
      "learning_rate": 4.2127659574468086e-05,
      "loss": 0.0692,
      "step": 381
    },
    {
      "epoch": 4.775,
      "grad_norm": 0.674705982208252,
      "learning_rate": 4.170212765957447e-05,
      "loss": 0.0788,
      "step": 382
    },
    {
      "epoch": 4.7875,
      "grad_norm": 0.6924496293067932,
      "learning_rate": 4.127659574468085e-05,
      "loss": 0.0731,
      "step": 383
    },
    {
      "epoch": 4.8,
      "grad_norm": 0.6793739199638367,
      "learning_rate": 4.085106382978723e-05,
      "loss": 0.0697,
      "step": 384
    },
    {
      "epoch": 4.8,
      "eval_loss": 1.009806513786316,
      "eval_runtime": 11.392,
      "eval_samples_per_second": 3.16,
      "eval_steps_per_second": 1.58,
      "step": 384
    },
    {
      "epoch": 4.8125,
      "grad_norm": 0.45708170533180237,
      "learning_rate": 4.0425531914893614e-05,
      "loss": 0.0499,
      "step": 385
    },
    {
      "epoch": 4.825,
      "grad_norm": 0.7347174286842346,
      "learning_rate": 4e-05,
      "loss": 0.072,
      "step": 386
    },
    {
      "epoch": 4.8375,
      "grad_norm": 0.7631716728210449,
      "learning_rate": 3.9574468085106385e-05,
      "loss": 0.0781,
      "step": 387
    },
    {
      "epoch": 4.85,
      "grad_norm": 0.7159767150878906,
      "learning_rate": 3.914893617021277e-05,
      "loss": 0.0761,
      "step": 388
    },
    {
      "epoch": 4.8625,
      "grad_norm": 0.7675968408584595,
      "learning_rate": 3.872340425531915e-05,
      "loss": 0.0649,
      "step": 389
    },
    {
      "epoch": 4.875,
      "grad_norm": 0.6752840876579285,
      "learning_rate": 3.829787234042553e-05,
      "loss": 0.0674,
      "step": 390
    },
    {
      "epoch": 4.8875,
      "grad_norm": 0.6436358690261841,
      "learning_rate": 3.787234042553191e-05,
      "loss": 0.0616,
      "step": 391
    },
    {
      "epoch": 4.9,
      "grad_norm": 0.7673559188842773,
      "learning_rate": 3.7446808510638295e-05,
      "loss": 0.0803,
      "step": 392
    },
    {
      "epoch": 4.9125,
      "grad_norm": 0.6778103709220886,
      "learning_rate": 3.702127659574468e-05,
      "loss": 0.0727,
      "step": 393
    },
    {
      "epoch": 4.925,
      "grad_norm": 0.8372657895088196,
      "learning_rate": 3.6595744680851066e-05,
      "loss": 0.074,
      "step": 394
    },
    {
      "epoch": 4.9375,
      "grad_norm": 0.6369761824607849,
      "learning_rate": 3.617021276595745e-05,
      "loss": 0.0669,
      "step": 395
    },
    {
      "epoch": 4.95,
      "grad_norm": 0.6647788882255554,
      "learning_rate": 3.574468085106383e-05,
      "loss": 0.053,
      "step": 396
    },
    {
      "epoch": 4.9625,
      "grad_norm": 0.7610755562782288,
      "learning_rate": 3.531914893617021e-05,
      "loss": 0.0668,
      "step": 397
    },
    {
      "epoch": 4.975,
      "grad_norm": 0.6888034343719482,
      "learning_rate": 3.48936170212766e-05,
      "loss": 0.0708,
      "step": 398
    },
    {
      "epoch": 4.9875,
      "grad_norm": 0.8322062492370605,
      "learning_rate": 3.446808510638298e-05,
      "loss": 0.067,
      "step": 399
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.7463421821594238,
      "learning_rate": 3.4042553191489365e-05,
      "loss": 0.0568,
      "step": 400
    },
    {
      "epoch": 5.0125,
      "grad_norm": 0.3784998953342438,
      "learning_rate": 3.361702127659575e-05,
      "loss": 0.0376,
      "step": 401
    },
    {
      "epoch": 5.025,
      "grad_norm": 0.39330971240997314,
      "learning_rate": 3.319148936170213e-05,
      "loss": 0.0367,
      "step": 402
    },
    {
      "epoch": 5.0375,
      "grad_norm": 0.3745648264884949,
      "learning_rate": 3.276595744680851e-05,
      "loss": 0.0363,
      "step": 403
    },
    {
      "epoch": 5.05,
      "grad_norm": 0.44352462887763977,
      "learning_rate": 3.23404255319149e-05,
      "loss": 0.0347,
      "step": 404
    },
    {
      "epoch": 5.0625,
      "grad_norm": 0.4444413483142853,
      "learning_rate": 3.191489361702128e-05,
      "loss": 0.0437,
      "step": 405
    },
    {
      "epoch": 5.075,
      "grad_norm": 0.4831590950489044,
      "learning_rate": 3.1489361702127664e-05,
      "loss": 0.0428,
      "step": 406
    },
    {
      "epoch": 5.0875,
      "grad_norm": 0.4673588275909424,
      "learning_rate": 3.1063829787234046e-05,
      "loss": 0.0355,
      "step": 407
    },
    {
      "epoch": 5.1,
      "grad_norm": 0.3534888029098511,
      "learning_rate": 3.063829787234043e-05,
      "loss": 0.0334,
      "step": 408
    },
    {
      "epoch": 5.1125,
      "grad_norm": 0.4960513114929199,
      "learning_rate": 3.021276595744681e-05,
      "loss": 0.0483,
      "step": 409
    },
    {
      "epoch": 5.125,
      "grad_norm": 0.46593356132507324,
      "learning_rate": 2.9787234042553192e-05,
      "loss": 0.0419,
      "step": 410
    },
    {
      "epoch": 5.1375,
      "grad_norm": 0.5315698385238647,
      "learning_rate": 2.9361702127659574e-05,
      "loss": 0.043,
      "step": 411
    },
    {
      "epoch": 5.15,
      "grad_norm": 0.5316590070724487,
      "learning_rate": 2.8936170212765956e-05,
      "loss": 0.0478,
      "step": 412
    },
    {
      "epoch": 5.1625,
      "grad_norm": 0.454860657453537,
      "learning_rate": 2.8510638297872345e-05,
      "loss": 0.0427,
      "step": 413
    },
    {
      "epoch": 5.175,
      "grad_norm": 0.5423195958137512,
      "learning_rate": 2.8085106382978727e-05,
      "loss": 0.041,
      "step": 414
    },
    {
      "epoch": 5.1875,
      "grad_norm": 0.4226936995983124,
      "learning_rate": 2.765957446808511e-05,
      "loss": 0.0351,
      "step": 415
    },
    {
      "epoch": 5.2,
      "grad_norm": 0.3787650465965271,
      "learning_rate": 2.723404255319149e-05,
      "loss": 0.0349,
      "step": 416
    },
    {
      "epoch": 5.2125,
      "grad_norm": 0.45735734701156616,
      "learning_rate": 2.6808510638297873e-05,
      "loss": 0.0383,
      "step": 417
    },
    {
      "epoch": 5.225,
      "grad_norm": 0.4093031585216522,
      "learning_rate": 2.6382978723404255e-05,
      "loss": 0.0314,
      "step": 418
    },
    {
      "epoch": 5.2375,
      "grad_norm": 0.345889151096344,
      "learning_rate": 2.5957446808510637e-05,
      "loss": 0.0306,
      "step": 419
    },
    {
      "epoch": 5.25,
      "grad_norm": 0.37512144446372986,
      "learning_rate": 2.5531914893617022e-05,
      "loss": 0.0289,
      "step": 420
    },
    {
      "epoch": 5.2625,
      "grad_norm": 0.5944772362709045,
      "learning_rate": 2.5106382978723404e-05,
      "loss": 0.0458,
      "step": 421
    },
    {
      "epoch": 5.275,
      "grad_norm": 0.6373673677444458,
      "learning_rate": 2.468085106382979e-05,
      "loss": 0.0473,
      "step": 422
    },
    {
      "epoch": 5.2875,
      "grad_norm": 0.4604647159576416,
      "learning_rate": 2.425531914893617e-05,
      "loss": 0.0476,
      "step": 423
    },
    {
      "epoch": 5.3,
      "grad_norm": 0.467212975025177,
      "learning_rate": 2.3829787234042557e-05,
      "loss": 0.0311,
      "step": 424
    },
    {
      "epoch": 5.3125,
      "grad_norm": 0.5458898544311523,
      "learning_rate": 2.340425531914894e-05,
      "loss": 0.0383,
      "step": 425
    },
    {
      "epoch": 5.325,
      "grad_norm": 0.39436912536621094,
      "learning_rate": 2.297872340425532e-05,
      "loss": 0.0333,
      "step": 426
    },
    {
      "epoch": 5.3375,
      "grad_norm": 0.5241003036499023,
      "learning_rate": 2.2553191489361703e-05,
      "loss": 0.0365,
      "step": 427
    },
    {
      "epoch": 5.35,
      "grad_norm": 0.46175524592399597,
      "learning_rate": 2.2127659574468085e-05,
      "loss": 0.0371,
      "step": 428
    },
    {
      "epoch": 5.3625,
      "grad_norm": 0.4485141932964325,
      "learning_rate": 2.170212765957447e-05,
      "loss": 0.0362,
      "step": 429
    },
    {
      "epoch": 5.375,
      "grad_norm": 0.4773487448692322,
      "learning_rate": 2.1276595744680852e-05,
      "loss": 0.0396,
      "step": 430
    },
    {
      "epoch": 5.3875,
      "grad_norm": 0.6281043291091919,
      "learning_rate": 2.0851063829787234e-05,
      "loss": 0.0438,
      "step": 431
    },
    {
      "epoch": 5.4,
      "grad_norm": 0.5021329522132874,
      "learning_rate": 2.0425531914893616e-05,
      "loss": 0.0392,
      "step": 432
    },
    {
      "epoch": 5.4125,
      "grad_norm": 0.4514084458351135,
      "learning_rate": 2e-05,
      "loss": 0.0346,
      "step": 433
    },
    {
      "epoch": 5.425,
      "grad_norm": 0.4999784231185913,
      "learning_rate": 1.9574468085106384e-05,
      "loss": 0.0405,
      "step": 434
    },
    {
      "epoch": 5.4375,
      "grad_norm": 0.45210784673690796,
      "learning_rate": 1.9148936170212766e-05,
      "loss": 0.0309,
      "step": 435
    },
    {
      "epoch": 5.45,
      "grad_norm": 0.8242072463035583,
      "learning_rate": 1.8723404255319148e-05,
      "loss": 0.0428,
      "step": 436
    },
    {
      "epoch": 5.4625,
      "grad_norm": 0.45749950408935547,
      "learning_rate": 1.8297872340425533e-05,
      "loss": 0.0337,
      "step": 437
    },
    {
      "epoch": 5.475,
      "grad_norm": 0.4928417503833771,
      "learning_rate": 1.7872340425531915e-05,
      "loss": 0.0294,
      "step": 438
    },
    {
      "epoch": 5.4875,
      "grad_norm": 0.6273995637893677,
      "learning_rate": 1.74468085106383e-05,
      "loss": 0.0353,
      "step": 439
    },
    {
      "epoch": 5.5,
      "grad_norm": 0.4409819543361664,
      "learning_rate": 1.7021276595744682e-05,
      "loss": 0.0374,
      "step": 440
    },
    {
      "epoch": 5.5125,
      "grad_norm": 0.5446107387542725,
      "learning_rate": 1.6595744680851064e-05,
      "loss": 0.0395,
      "step": 441
    },
    {
      "epoch": 5.525,
      "grad_norm": 0.5066544413566589,
      "learning_rate": 1.617021276595745e-05,
      "loss": 0.0362,
      "step": 442
    },
    {
      "epoch": 5.5375,
      "grad_norm": 0.4936687648296356,
      "learning_rate": 1.5744680851063832e-05,
      "loss": 0.0348,
      "step": 443
    },
    {
      "epoch": 5.55,
      "grad_norm": 0.5274161100387573,
      "learning_rate": 1.5319148936170214e-05,
      "loss": 0.0445,
      "step": 444
    },
    {
      "epoch": 5.5625,
      "grad_norm": 0.3984338045120239,
      "learning_rate": 1.4893617021276596e-05,
      "loss": 0.0297,
      "step": 445
    },
    {
      "epoch": 5.575,
      "grad_norm": 0.3740543723106384,
      "learning_rate": 1.4468085106382978e-05,
      "loss": 0.0298,
      "step": 446
    },
    {
      "epoch": 5.5875,
      "grad_norm": 0.44997498393058777,
      "learning_rate": 1.4042553191489363e-05,
      "loss": 0.0343,
      "step": 447
    },
    {
      "epoch": 5.6,
      "grad_norm": 0.553840696811676,
      "learning_rate": 1.3617021276595745e-05,
      "loss": 0.0392,
      "step": 448
    },
    {
      "epoch": 5.6125,
      "grad_norm": 0.47161081433296204,
      "learning_rate": 1.3191489361702127e-05,
      "loss": 0.0355,
      "step": 449
    },
    {
      "epoch": 5.625,
      "grad_norm": 0.4401087760925293,
      "learning_rate": 1.2765957446808511e-05,
      "loss": 0.033,
      "step": 450
    },
    {
      "epoch": 5.6375,
      "grad_norm": 0.4902607798576355,
      "learning_rate": 1.2340425531914895e-05,
      "loss": 0.0401,
      "step": 451
    },
    {
      "epoch": 5.65,
      "grad_norm": 0.5491804480552673,
      "learning_rate": 1.1914893617021278e-05,
      "loss": 0.0372,
      "step": 452
    },
    {
      "epoch": 5.6625,
      "grad_norm": 0.5546887516975403,
      "learning_rate": 1.148936170212766e-05,
      "loss": 0.0353,
      "step": 453
    },
    {
      "epoch": 5.675,
      "grad_norm": 0.42722412943840027,
      "learning_rate": 1.1063829787234042e-05,
      "loss": 0.0311,
      "step": 454
    },
    {
      "epoch": 5.6875,
      "grad_norm": 0.4589029550552368,
      "learning_rate": 1.0638297872340426e-05,
      "loss": 0.0342,
      "step": 455
    },
    {
      "epoch": 5.7,
      "grad_norm": 0.3946620523929596,
      "learning_rate": 1.0212765957446808e-05,
      "loss": 0.0307,
      "step": 456
    },
    {
      "epoch": 5.7125,
      "grad_norm": 0.35820481181144714,
      "learning_rate": 9.787234042553192e-06,
      "loss": 0.0268,
      "step": 457
    },
    {
      "epoch": 5.725,
      "grad_norm": 0.5740081667900085,
      "learning_rate": 9.361702127659574e-06,
      "loss": 0.0331,
      "step": 458
    },
    {
      "epoch": 5.7375,
      "grad_norm": 0.6305598616600037,
      "learning_rate": 8.936170212765958e-06,
      "loss": 0.04,
      "step": 459
    },
    {
      "epoch": 5.75,
      "grad_norm": 0.7847849130630493,
      "learning_rate": 8.510638297872341e-06,
      "loss": 0.0408,
      "step": 460
    },
    {
      "epoch": 5.7625,
      "grad_norm": 0.4860388934612274,
      "learning_rate": 8.085106382978725e-06,
      "loss": 0.0397,
      "step": 461
    },
    {
      "epoch": 5.775,
      "grad_norm": 0.33737698197364807,
      "learning_rate": 7.659574468085107e-06,
      "loss": 0.032,
      "step": 462
    },
    {
      "epoch": 5.7875,
      "grad_norm": 0.4484858214855194,
      "learning_rate": 7.234042553191489e-06,
      "loss": 0.0309,
      "step": 463
    },
    {
      "epoch": 5.8,
      "grad_norm": 0.628653883934021,
      "learning_rate": 6.808510638297873e-06,
      "loss": 0.0331,
      "step": 464
    },
    {
      "epoch": 5.8125,
      "grad_norm": 0.45020169019699097,
      "learning_rate": 6.3829787234042555e-06,
      "loss": 0.033,
      "step": 465
    },
    {
      "epoch": 5.825,
      "grad_norm": 0.3453785181045532,
      "learning_rate": 5.957446808510639e-06,
      "loss": 0.0331,
      "step": 466
    },
    {
      "epoch": 5.8375,
      "grad_norm": 0.3796781897544861,
      "learning_rate": 5.531914893617021e-06,
      "loss": 0.0347,
      "step": 467
    },
    {
      "epoch": 5.85,
      "grad_norm": 0.4639665186405182,
      "learning_rate": 5.106382978723404e-06,
      "loss": 0.0355,
      "step": 468
    },
    {
      "epoch": 5.8625,
      "grad_norm": 0.6302652359008789,
      "learning_rate": 4.680851063829787e-06,
      "loss": 0.0417,
      "step": 469
    },
    {
      "epoch": 5.875,
      "grad_norm": 0.3809727728366852,
      "learning_rate": 4.255319148936171e-06,
      "loss": 0.029,
      "step": 470
    },
    {
      "epoch": 5.8875,
      "grad_norm": 0.4321640133857727,
      "learning_rate": 3.8297872340425535e-06,
      "loss": 0.0304,
      "step": 471
    },
    {
      "epoch": 5.9,
      "grad_norm": 0.48058974742889404,
      "learning_rate": 3.4042553191489363e-06,
      "loss": 0.0379,
      "step": 472
    },
    {
      "epoch": 5.9125,
      "grad_norm": 0.37713518738746643,
      "learning_rate": 2.9787234042553196e-06,
      "loss": 0.033,
      "step": 473
    },
    {
      "epoch": 5.925,
      "grad_norm": 0.5107573866844177,
      "learning_rate": 2.553191489361702e-06,
      "loss": 0.036,
      "step": 474
    },
    {
      "epoch": 5.9375,
      "grad_norm": 0.35142090916633606,
      "learning_rate": 2.1276595744680853e-06,
      "loss": 0.0273,
      "step": 475
    },
    {
      "epoch": 5.95,
      "grad_norm": 0.3913898766040802,
      "learning_rate": 1.7021276595744682e-06,
      "loss": 0.0314,
      "step": 476
    },
    {
      "epoch": 5.9625,
      "grad_norm": 0.5400916337966919,
      "learning_rate": 1.276595744680851e-06,
      "loss": 0.0372,
      "step": 477
    },
    {
      "epoch": 5.975,
      "grad_norm": 0.4400492012500763,
      "learning_rate": 8.510638297872341e-07,
      "loss": 0.0345,
      "step": 478
    },
    {
      "epoch": 5.9875,
      "grad_norm": 0.44733723998069763,
      "learning_rate": 4.2553191489361704e-07,
      "loss": 0.0308,
      "step": 479
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.42864102125167847,
      "learning_rate": 0.0,
      "loss": 0.0305,
      "step": 480
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.1705684661865234,
      "eval_runtime": 11.4475,
      "eval_samples_per_second": 3.145,
      "eval_steps_per_second": 1.572,
      "step": 480
    }
  ],
  "logging_steps": 1,
  "max_steps": 480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.41482731320361e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}