{
  "best_metric": 1.4196490049362183,
  "best_model_checkpoint": "4bit_repro_03022025/host14_seed_42_full_det_fp16_no_flash_attn_fix_pad_gemma-2-9b-instruct-l16-no-cot-4ep-lr3e04-ws20-bs4-ga4-fp16-11022025/checkpoint-109",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 327,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.009195402298850575,
      "grad_norm": 1.6409969329833984,
      "learning_rate": 1.4999999999999999e-05,
      "loss": 2.9018,
      "step": 1
    },
    {
      "epoch": 0.01839080459770115,
      "grad_norm": 0.8910855054855347,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 2.5407,
      "step": 2
    },
    {
      "epoch": 0.027586206896551724,
      "grad_norm": 1.2834181785583496,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 2.5948,
      "step": 3
    },
    {
      "epoch": 0.0367816091954023,
      "grad_norm": 0.9639408588409424,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 2.3316,
      "step": 4
    },
    {
      "epoch": 0.04597701149425287,
      "grad_norm": 1.4458935260772705,
      "learning_rate": 7.5e-05,
      "loss": 2.6124,
      "step": 5
    },
    {
      "epoch": 0.05517241379310345,
      "grad_norm": 1.3878850936889648,
      "learning_rate": 8.999999999999999e-05,
      "loss": 2.3947,
      "step": 6
    },
    {
      "epoch": 0.06436781609195402,
      "grad_norm": 0.5613430142402649,
      "learning_rate": 0.00010499999999999999,
      "loss": 2.2874,
      "step": 7
    },
    {
      "epoch": 0.0735632183908046,
      "grad_norm": 0.4661317467689514,
      "learning_rate": 0.00011999999999999999,
      "loss": 2.1525,
      "step": 8
    },
    {
      "epoch": 0.08275862068965517,
      "grad_norm": 0.6441327929496765,
      "learning_rate": 0.000135,
      "loss": 1.9578,
      "step": 9
    },
    {
      "epoch": 0.09195402298850575,
      "grad_norm": 0.9715940952301025,
      "learning_rate": 0.00015,
      "loss": 1.8896,
      "step": 10
    },
    {
      "epoch": 0.10114942528735632,
      "grad_norm": 0.7227729558944702,
      "learning_rate": 0.000165,
      "loss": 1.8532,
      "step": 11
    },
    {
      "epoch": 0.1103448275862069,
      "grad_norm": 0.6046324372291565,
      "learning_rate": 0.00017999999999999998,
      "loss": 1.7605,
      "step": 12
    },
    {
      "epoch": 0.11954022988505747,
      "grad_norm": 0.801392138004303,
      "learning_rate": 0.000195,
      "loss": 1.4759,
      "step": 13
    },
    {
      "epoch": 0.12873563218390804,
      "grad_norm": 0.5491551160812378,
      "learning_rate": 0.00020999999999999998,
      "loss": 1.6874,
      "step": 14
    },
    {
      "epoch": 0.13793103448275862,
      "grad_norm": 0.4392535388469696,
      "learning_rate": 0.000225,
      "loss": 1.7641,
      "step": 15
    },
    {
      "epoch": 0.1471264367816092,
      "grad_norm": 0.41002196073532104,
      "learning_rate": 0.00023999999999999998,
      "loss": 1.9998,
      "step": 16
    },
    {
      "epoch": 0.15632183908045977,
      "grad_norm": 0.5483619570732117,
      "learning_rate": 0.00025499999999999996,
      "loss": 1.3601,
      "step": 17
    },
    {
      "epoch": 0.16551724137931034,
      "grad_norm": 0.5035637021064758,
      "learning_rate": 0.00027,
      "loss": 1.2777,
      "step": 18
    },
    {
      "epoch": 0.17471264367816092,
      "grad_norm": 0.48088592290878296,
      "learning_rate": 0.000285,
      "loss": 1.462,
      "step": 19
    },
    {
      "epoch": 0.1839080459770115,
      "grad_norm": 0.42464756965637207,
      "learning_rate": 0.0003,
      "loss": 1.6768,
      "step": 20
    },
    {
      "epoch": 0.19310344827586207,
      "grad_norm": 0.3142748773097992,
      "learning_rate": 0.00029927184466019415,
      "loss": 1.6529,
      "step": 21
    },
    {
      "epoch": 0.20229885057471264,
      "grad_norm": 0.300471693277359,
      "learning_rate": 0.00029854368932038833,
      "loss": 1.6068,
      "step": 22
    },
    {
      "epoch": 0.21149425287356322,
      "grad_norm": 0.3858422636985779,
      "learning_rate": 0.0002978155339805825,
      "loss": 1.371,
      "step": 23
    },
    {
      "epoch": 0.2206896551724138,
      "grad_norm": 0.37815675139427185,
      "learning_rate": 0.0002970873786407767,
      "loss": 1.54,
      "step": 24
    },
    {
      "epoch": 0.22988505747126436,
      "grad_norm": 0.327928751707077,
      "learning_rate": 0.00029635922330097087,
      "loss": 1.7743,
      "step": 25
    },
    {
      "epoch": 0.23908045977011494,
      "grad_norm": 0.45471611618995667,
      "learning_rate": 0.00029563106796116505,
      "loss": 1.5857,
      "step": 26
    },
    {
      "epoch": 0.2482758620689655,
      "grad_norm": 0.5154709219932556,
      "learning_rate": 0.0002949029126213592,
      "loss": 1.5444,
      "step": 27
    },
    {
      "epoch": 0.2574712643678161,
      "grad_norm": 0.40547341108322144,
      "learning_rate": 0.00029417475728155335,
      "loss": 1.3692,
      "step": 28
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.4016611874103546,
      "learning_rate": 0.00029344660194174753,
      "loss": 1.7185,
      "step": 29
    },
    {
      "epoch": 0.27586206896551724,
      "grad_norm": 0.35080596804618835,
      "learning_rate": 0.0002927184466019417,
      "loss": 1.8895,
      "step": 30
    },
    {
      "epoch": 0.2850574712643678,
      "grad_norm": 0.4109053611755371,
      "learning_rate": 0.0002919902912621359,
      "loss": 1.6865,
      "step": 31
    },
    {
      "epoch": 0.2942528735632184,
      "grad_norm": 0.3945920467376709,
      "learning_rate": 0.00029126213592233006,
      "loss": 1.3734,
      "step": 32
    },
    {
      "epoch": 0.30344827586206896,
      "grad_norm": 0.5358554124832153,
      "learning_rate": 0.00029053398058252424,
      "loss": 1.2796,
      "step": 33
    },
    {
      "epoch": 0.31264367816091954,
      "grad_norm": 0.5781538486480713,
      "learning_rate": 0.0002898058252427184,
      "loss": 1.6769,
      "step": 34
    },
    {
      "epoch": 0.3218390804597701,
      "grad_norm": 0.4976755380630493,
      "learning_rate": 0.0002890776699029126,
      "loss": 1.6693,
      "step": 35
    },
    {
      "epoch": 0.3310344827586207,
      "grad_norm": 0.5537866353988647,
      "learning_rate": 0.0002883495145631068,
      "loss": 1.4935,
      "step": 36
    },
    {
      "epoch": 0.34022988505747126,
      "grad_norm": 0.41387081146240234,
      "learning_rate": 0.00028762135922330096,
      "loss": 1.5668,
      "step": 37
    },
    {
      "epoch": 0.34942528735632183,
      "grad_norm": 0.38216930627822876,
      "learning_rate": 0.00028689320388349513,
      "loss": 1.667,
      "step": 38
    },
    {
      "epoch": 0.3586206896551724,
      "grad_norm": 0.5784114599227905,
      "learning_rate": 0.0002861650485436893,
      "loss": 1.2776,
      "step": 39
    },
    {
      "epoch": 0.367816091954023,
      "grad_norm": 0.417309045791626,
      "learning_rate": 0.0002854368932038835,
      "loss": 1.6741,
      "step": 40
    },
    {
      "epoch": 0.37701149425287356,
      "grad_norm": 0.39875927567481995,
      "learning_rate": 0.00028470873786407767,
      "loss": 1.4101,
      "step": 41
    },
    {
      "epoch": 0.38620689655172413,
      "grad_norm": 0.7550377249717712,
      "learning_rate": 0.00028398058252427185,
      "loss": 1.1991,
      "step": 42
    },
    {
      "epoch": 0.3954022988505747,
      "grad_norm": 0.4652000963687897,
      "learning_rate": 0.00028325242718446603,
      "loss": 1.4474,
      "step": 43
    },
    {
      "epoch": 0.4045977011494253,
      "grad_norm": 0.5465714931488037,
      "learning_rate": 0.00028252427184466015,
      "loss": 1.4219,
      "step": 44
    },
    {
      "epoch": 0.41379310344827586,
      "grad_norm": 0.5987188816070557,
      "learning_rate": 0.00028179611650485433,
      "loss": 1.2155,
      "step": 45
    },
    {
      "epoch": 0.42298850574712643,
      "grad_norm": 0.5176146626472473,
      "learning_rate": 0.0002810679611650485,
      "loss": 1.6139,
      "step": 46
    },
    {
      "epoch": 0.432183908045977,
      "grad_norm": 0.5650622248649597,
      "learning_rate": 0.0002803398058252427,
      "loss": 1.3885,
      "step": 47
    },
    {
      "epoch": 0.4413793103448276,
      "grad_norm": 0.5218377709388733,
      "learning_rate": 0.00027961165048543687,
      "loss": 1.2143,
      "step": 48
    },
    {
      "epoch": 0.45057471264367815,
      "grad_norm": 0.8088476657867432,
      "learning_rate": 0.00027888349514563105,
      "loss": 1.1332,
      "step": 49
    },
    {
      "epoch": 0.45977011494252873,
      "grad_norm": 0.6791206002235413,
      "learning_rate": 0.0002781553398058252,
      "loss": 1.1016,
      "step": 50
    },
    {
      "epoch": 0.4689655172413793,
      "grad_norm": 0.528010368347168,
      "learning_rate": 0.0002774271844660194,
      "loss": 1.3407,
      "step": 51
    },
    {
      "epoch": 0.4781609195402299,
      "grad_norm": 0.5152159929275513,
      "learning_rate": 0.0002766990291262136,
      "loss": 1.0526,
      "step": 52
    },
    {
      "epoch": 0.48735632183908045,
      "grad_norm": 0.8332945704460144,
      "learning_rate": 0.00027597087378640776,
      "loss": 1.2991,
      "step": 53
    },
    {
      "epoch": 0.496551724137931,
      "grad_norm": 0.7987570762634277,
      "learning_rate": 0.00027524271844660194,
      "loss": 1.008,
      "step": 54
    },
    {
      "epoch": 0.5057471264367817,
      "grad_norm": 0.6524537205696106,
      "learning_rate": 0.0002745145631067961,
      "loss": 1.2895,
      "step": 55
    },
    {
      "epoch": 0.5149425287356322,
      "grad_norm": 0.7703379988670349,
      "learning_rate": 0.0002737864077669903,
      "loss": 1.2739,
      "step": 56
    },
    {
      "epoch": 0.5241379310344828,
      "grad_norm": 0.6915059089660645,
      "learning_rate": 0.0002730582524271845,
      "loss": 0.9791,
      "step": 57
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.7275146842002869,
      "learning_rate": 0.00027233009708737865,
      "loss": 1.1958,
      "step": 58
    },
    {
      "epoch": 0.542528735632184,
      "grad_norm": 0.7504586577415466,
      "learning_rate": 0.0002716019417475728,
      "loss": 1.0989,
      "step": 59
    },
    {
      "epoch": 0.5517241379310345,
      "grad_norm": 0.9296619892120361,
      "learning_rate": 0.00027087378640776696,
      "loss": 1.2009,
      "step": 60
    },
    {
      "epoch": 0.5609195402298851,
      "grad_norm": 0.7090461850166321,
      "learning_rate": 0.00027014563106796114,
      "loss": 1.0826,
      "step": 61
    },
    {
      "epoch": 0.5701149425287356,
      "grad_norm": 0.7167800068855286,
      "learning_rate": 0.0002694174757281553,
      "loss": 0.9676,
      "step": 62
    },
    {
      "epoch": 0.5793103448275863,
      "grad_norm": 0.9599812030792236,
      "learning_rate": 0.0002686893203883495,
      "loss": 1.0113,
      "step": 63
    },
    {
      "epoch": 0.5885057471264368,
      "grad_norm": 0.8795760273933411,
      "learning_rate": 0.00026796116504854367,
      "loss": 0.9343,
      "step": 64
    },
    {
      "epoch": 0.5977011494252874,
      "grad_norm": 1.689482569694519,
      "learning_rate": 0.00026723300970873785,
      "loss": 1.4378,
      "step": 65
    },
    {
      "epoch": 0.6068965517241379,
      "grad_norm": 0.7943350672721863,
      "learning_rate": 0.00026650485436893203,
      "loss": 0.8399,
      "step": 66
    },
    {
      "epoch": 0.6160919540229886,
      "grad_norm": 0.862583577632904,
      "learning_rate": 0.0002657766990291262,
      "loss": 1.0277,
      "step": 67
    },
    {
      "epoch": 0.6252873563218391,
      "grad_norm": 0.8344200253486633,
      "learning_rate": 0.0002650485436893204,
      "loss": 1.0991,
      "step": 68
    },
    {
      "epoch": 0.6344827586206897,
      "grad_norm": 0.7979349493980408,
      "learning_rate": 0.00026432038834951456,
      "loss": 1.2579,
      "step": 69
    },
    {
      "epoch": 0.6436781609195402,
      "grad_norm": 0.7641604542732239,
      "learning_rate": 0.00026359223300970874,
      "loss": 1.0108,
      "step": 70
    },
    {
      "epoch": 0.6528735632183909,
      "grad_norm": 0.8905412554740906,
      "learning_rate": 0.0002628640776699029,
      "loss": 0.7594,
      "step": 71
    },
    {
      "epoch": 0.6620689655172414,
      "grad_norm": 0.8589284420013428,
      "learning_rate": 0.00026213592233009705,
      "loss": 0.8535,
      "step": 72
    },
    {
      "epoch": 0.671264367816092,
      "grad_norm": 0.8410593867301941,
      "learning_rate": 0.0002614077669902912,
      "loss": 0.8691,
      "step": 73
    },
    {
      "epoch": 0.6804597701149425,
      "grad_norm": 1.0918798446655273,
      "learning_rate": 0.0002606796116504854,
      "loss": 0.984,
      "step": 74
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 1.3453165292739868,
      "learning_rate": 0.0002599514563106796,
      "loss": 1.1908,
      "step": 75
    },
    {
      "epoch": 0.6988505747126437,
      "grad_norm": 1.7153398990631104,
      "learning_rate": 0.00025922330097087376,
      "loss": 1.1151,
      "step": 76
    },
    {
      "epoch": 0.7080459770114943,
      "grad_norm": 0.7578756809234619,
      "learning_rate": 0.00025849514563106794,
      "loss": 1.0327,
      "step": 77
    },
    {
      "epoch": 0.7172413793103448,
      "grad_norm": 0.8445453643798828,
      "learning_rate": 0.0002577669902912621,
      "loss": 0.7871,
      "step": 78
    },
    {
      "epoch": 0.7264367816091954,
      "grad_norm": 1.0929808616638184,
      "learning_rate": 0.0002570388349514563,
      "loss": 1.2769,
      "step": 79
    },
    {
      "epoch": 0.735632183908046,
      "grad_norm": 0.8767411708831787,
      "learning_rate": 0.0002563106796116505,
      "loss": 0.9978,
      "step": 80
    },
    {
      "epoch": 0.7448275862068966,
      "grad_norm": 0.7253080606460571,
      "learning_rate": 0.0002555825242718446,
      "loss": 1.0843,
      "step": 81
    },
    {
      "epoch": 0.7540229885057471,
      "grad_norm": 0.7607583999633789,
      "learning_rate": 0.0002548543689320388,
      "loss": 0.8945,
      "step": 82
    },
    {
      "epoch": 0.7632183908045977,
      "grad_norm": 1.163414716720581,
      "learning_rate": 0.00025412621359223296,
      "loss": 1.0853,
      "step": 83
    },
    {
      "epoch": 0.7724137931034483,
      "grad_norm": 0.9414466023445129,
      "learning_rate": 0.00025339805825242714,
      "loss": 1.0867,
      "step": 84
    },
    {
      "epoch": 0.7816091954022989,
      "grad_norm": 0.873741090297699,
      "learning_rate": 0.0002526699029126213,
      "loss": 1.1912,
      "step": 85
    },
    {
      "epoch": 0.7908045977011494,
      "grad_norm": 0.8017745614051819,
      "learning_rate": 0.0002519417475728155,
      "loss": 0.6812,
      "step": 86
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.8693607449531555,
      "learning_rate": 0.00025121359223300967,
      "loss": 0.8669,
      "step": 87
    },
    {
      "epoch": 0.8091954022988506,
      "grad_norm": 0.7914007902145386,
      "learning_rate": 0.00025048543689320385,
      "loss": 0.9684,
      "step": 88
    },
    {
      "epoch": 0.8183908045977012,
      "grad_norm": 0.7322354912757874,
      "learning_rate": 0.00024975728155339803,
      "loss": 1.0616,
      "step": 89
    },
    {
      "epoch": 0.8275862068965517,
      "grad_norm": 0.6482452154159546,
      "learning_rate": 0.0002490291262135922,
      "loss": 0.8636,
      "step": 90
    },
    {
      "epoch": 0.8367816091954023,
      "grad_norm": 0.8021015524864197,
      "learning_rate": 0.0002483009708737864,
      "loss": 0.7627,
      "step": 91
    },
    {
      "epoch": 0.8459770114942529,
      "grad_norm": 0.8987912535667419,
      "learning_rate": 0.00024757281553398056,
      "loss": 0.5836,
      "step": 92
    },
    {
      "epoch": 0.8551724137931035,
      "grad_norm": 0.7898840308189392,
      "learning_rate": 0.00024684466019417474,
      "loss": 1.068,
      "step": 93
    },
    {
      "epoch": 0.864367816091954,
      "grad_norm": 1.0533185005187988,
      "learning_rate": 0.0002461165048543689,
      "loss": 0.77,
      "step": 94
    },
    {
      "epoch": 0.8735632183908046,
      "grad_norm": 0.9064627289772034,
      "learning_rate": 0.0002453883495145631,
      "loss": 0.6817,
      "step": 95
    },
    {
      "epoch": 0.8827586206896552,
      "grad_norm": 0.8030613660812378,
      "learning_rate": 0.0002446601941747572,
      "loss": 0.7652,
      "step": 96
    },
    {
      "epoch": 0.8919540229885058,
      "grad_norm": 0.828834593296051,
      "learning_rate": 0.00024393203883495143,
      "loss": 0.8619,
      "step": 97
    },
    {
      "epoch": 0.9011494252873563,
      "grad_norm": 0.9028300642967224,
      "learning_rate": 0.0002432038834951456,
      "loss": 0.9762,
      "step": 98
    },
    {
      "epoch": 0.9103448275862069,
      "grad_norm": 1.261749505996704,
      "learning_rate": 0.0002424757281553398,
      "loss": 0.7234,
      "step": 99
    },
    {
      "epoch": 0.9195402298850575,
      "grad_norm": 0.7746110558509827,
      "learning_rate": 0.00024174757281553394,
      "loss": 0.8761,
      "step": 100
    },
    {
      "epoch": 0.9287356321839081,
      "grad_norm": 0.8339172005653381,
      "learning_rate": 0.00024101941747572812,
      "loss": 0.699,
      "step": 101
    },
    {
      "epoch": 0.9379310344827586,
      "grad_norm": 1.1835957765579224,
      "learning_rate": 0.0002402912621359223,
      "loss": 0.6597,
      "step": 102
    },
    {
      "epoch": 0.9471264367816092,
      "grad_norm": 0.7516824007034302,
      "learning_rate": 0.00023956310679611648,
      "loss": 0.9335,
      "step": 103
    },
    {
      "epoch": 0.9563218390804598,
      "grad_norm": 0.7029162049293518,
      "learning_rate": 0.00023883495145631065,
      "loss": 0.8889,
      "step": 104
    },
    {
      "epoch": 0.9655172413793104,
      "grad_norm": 0.8920773267745972,
      "learning_rate": 0.00023810679611650483,
      "loss": 0.6555,
      "step": 105
    },
    {
      "epoch": 0.9747126436781609,
      "grad_norm": 0.7414918541908264,
      "learning_rate": 0.000237378640776699,
      "loss": 1.0669,
      "step": 106
    },
    {
      "epoch": 0.9839080459770115,
      "grad_norm": 0.8566169738769531,
      "learning_rate": 0.0002366504854368932,
      "loss": 0.8665,
      "step": 107
    },
    {
      "epoch": 0.993103448275862,
      "grad_norm": 0.6569233536720276,
      "learning_rate": 0.00023592233009708734,
      "loss": 0.585,
      "step": 108
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.2873867750167847,
      "learning_rate": 0.00023519417475728152,
      "loss": 0.8267,
      "step": 109
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.4196490049362183,
      "eval_runtime": 76.0927,
      "eval_samples_per_second": 4.363,
      "eval_steps_per_second": 2.182,
      "step": 109
    },
    {
      "epoch": 1.0091954022988505,
      "grad_norm": 0.820888340473175,
      "learning_rate": 0.0002344660194174757,
      "loss": 0.677,
      "step": 110
    },
    {
      "epoch": 1.018390804597701,
      "grad_norm": 0.7585135698318481,
      "learning_rate": 0.00023373786407766988,
      "loss": 0.6585,
      "step": 111
    },
    {
      "epoch": 1.0275862068965518,
      "grad_norm": 0.8374487161636353,
      "learning_rate": 0.00023300970873786406,
      "loss": 0.7201,
      "step": 112
    },
    {
      "epoch": 1.0367816091954023,
      "grad_norm": 0.9571560621261597,
      "learning_rate": 0.00023228155339805823,
      "loss": 0.4576,
      "step": 113
    },
    {
      "epoch": 1.0459770114942528,
      "grad_norm": 0.6560165882110596,
      "learning_rate": 0.0002315533980582524,
      "loss": 0.8126,
      "step": 114
    },
    {
      "epoch": 1.0551724137931036,
      "grad_norm": 1.02234947681427,
      "learning_rate": 0.0002308252427184466,
      "loss": 0.5351,
      "step": 115
    },
    {
      "epoch": 1.064367816091954,
      "grad_norm": 0.5752732753753662,
      "learning_rate": 0.00023009708737864074,
      "loss": 0.4256,
      "step": 116
    },
    {
      "epoch": 1.0735632183908046,
      "grad_norm": 1.0608015060424805,
      "learning_rate": 0.00022936893203883492,
      "loss": 0.6285,
      "step": 117
    },
    {
      "epoch": 1.0827586206896551,
      "grad_norm": 0.6491764187812805,
      "learning_rate": 0.0002286407766990291,
      "loss": 0.8877,
      "step": 118
    },
    {
      "epoch": 1.0919540229885056,
      "grad_norm": 0.6474234461784363,
      "learning_rate": 0.00022791262135922328,
      "loss": 0.5508,
      "step": 119
    },
    {
      "epoch": 1.1011494252873564,
      "grad_norm": 0.8234098553657532,
      "learning_rate": 0.00022718446601941746,
      "loss": 0.5686,
      "step": 120
    },
    {
      "epoch": 1.110344827586207,
      "grad_norm": 0.7973845601081848,
      "learning_rate": 0.00022645631067961164,
      "loss": 0.3965,
      "step": 121
    },
    {
      "epoch": 1.1195402298850574,
      "grad_norm": 0.7338511943817139,
      "learning_rate": 0.00022572815533980582,
      "loss": 0.3424,
      "step": 122
    },
    {
      "epoch": 1.1287356321839082,
      "grad_norm": 0.6569823026657104,
      "learning_rate": 0.000225,
      "loss": 0.5071,
      "step": 123
    },
    {
      "epoch": 1.1379310344827587,
      "grad_norm": 0.768832802772522,
      "learning_rate": 0.00022427184466019415,
      "loss": 0.5674,
      "step": 124
    },
    {
      "epoch": 1.1471264367816092,
      "grad_norm": 0.5685952305793762,
      "learning_rate": 0.00022354368932038832,
      "loss": 0.4932,
      "step": 125
    },
    {
      "epoch": 1.1563218390804597,
      "grad_norm": 0.7012789249420166,
      "learning_rate": 0.0002228155339805825,
      "loss": 0.4092,
      "step": 126
    },
    {
      "epoch": 1.1655172413793102,
      "grad_norm": 1.0596917867660522,
      "learning_rate": 0.00022208737864077668,
      "loss": 0.5964,
      "step": 127
    },
    {
      "epoch": 1.174712643678161,
      "grad_norm": 0.5177541375160217,
      "learning_rate": 0.00022135922330097086,
      "loss": 0.5093,
      "step": 128
    },
    {
      "epoch": 1.1839080459770115,
      "grad_norm": 0.6834475994110107,
      "learning_rate": 0.00022063106796116504,
      "loss": 0.5799,
      "step": 129
    },
    {
      "epoch": 1.193103448275862,
      "grad_norm": 0.46928611397743225,
      "learning_rate": 0.00021990291262135922,
      "loss": 0.424,
      "step": 130
    },
    {
      "epoch": 1.2022988505747128,
      "grad_norm": 0.9471850991249084,
      "learning_rate": 0.00021917475728155337,
      "loss": 0.5087,
      "step": 131
    },
    {
      "epoch": 1.2114942528735633,
      "grad_norm": 0.6638930439949036,
      "learning_rate": 0.00021844660194174755,
      "loss": 0.5133,
      "step": 132
    },
    {
      "epoch": 1.2206896551724138,
      "grad_norm": 0.554904580116272,
      "learning_rate": 0.00021771844660194173,
      "loss": 0.6236,
      "step": 133
    },
    {
      "epoch": 1.2298850574712643,
      "grad_norm": 0.6292574405670166,
      "learning_rate": 0.0002169902912621359,
      "loss": 0.3756,
      "step": 134
    },
    {
      "epoch": 1.2390804597701148,
      "grad_norm": 0.7311149835586548,
      "learning_rate": 0.00021626213592233008,
      "loss": 0.5629,
      "step": 135
    },
    {
      "epoch": 1.2482758620689656,
      "grad_norm": 0.6303699016571045,
      "learning_rate": 0.00021553398058252426,
      "loss": 0.7889,
      "step": 136
    },
    {
      "epoch": 1.257471264367816,
      "grad_norm": 0.8203924894332886,
      "learning_rate": 0.00021480582524271844,
      "loss": 0.3686,
      "step": 137
    },
    {
      "epoch": 1.2666666666666666,
      "grad_norm": 1.10085928440094,
      "learning_rate": 0.00021407766990291262,
      "loss": 0.4527,
      "step": 138
    },
    {
      "epoch": 1.2758620689655173,
      "grad_norm": 0.8229245543479919,
      "learning_rate": 0.00021334951456310677,
      "loss": 0.526,
      "step": 139
    },
    {
      "epoch": 1.2850574712643679,
      "grad_norm": 0.7513696551322937,
      "learning_rate": 0.00021262135922330095,
      "loss": 0.5633,
      "step": 140
    },
    {
      "epoch": 1.2942528735632184,
      "grad_norm": 0.8141852021217346,
      "learning_rate": 0.00021189320388349513,
      "loss": 0.751,
      "step": 141
    },
    {
      "epoch": 1.303448275862069,
      "grad_norm": 0.5154571533203125,
      "learning_rate": 0.0002111650485436893,
      "loss": 0.3195,
      "step": 142
    },
    {
      "epoch": 1.3126436781609194,
      "grad_norm": 0.41179782152175903,
      "learning_rate": 0.00021043689320388349,
      "loss": 0.265,
      "step": 143
    },
    {
      "epoch": 1.3218390804597702,
      "grad_norm": 0.5113994479179382,
      "learning_rate": 0.00020970873786407766,
      "loss": 0.565,
      "step": 144
    },
    {
      "epoch": 1.3310344827586207,
      "grad_norm": 0.5308792591094971,
      "learning_rate": 0.00020898058252427184,
      "loss": 0.2585,
      "step": 145
    },
    {
      "epoch": 1.3402298850574712,
      "grad_norm": 0.5127933025360107,
      "learning_rate": 0.00020825242718446602,
      "loss": 0.3584,
      "step": 146
    },
    {
      "epoch": 1.349425287356322,
      "grad_norm": 0.4476073384284973,
      "learning_rate": 0.00020752427184466017,
      "loss": 0.2942,
      "step": 147
    },
    {
      "epoch": 1.3586206896551725,
      "grad_norm": 0.5431134104728699,
      "learning_rate": 0.00020679611650485435,
      "loss": 0.4839,
      "step": 148
    },
    {
      "epoch": 1.367816091954023,
      "grad_norm": 0.6306489109992981,
      "learning_rate": 0.00020606796116504853,
      "loss": 0.3573,
      "step": 149
    },
    {
      "epoch": 1.3770114942528735,
      "grad_norm": 0.593370795249939,
      "learning_rate": 0.0002053398058252427,
      "loss": 0.4681,
      "step": 150
    },
    {
      "epoch": 1.386206896551724,
      "grad_norm": 0.733506977558136,
      "learning_rate": 0.0002046116504854369,
      "loss": 0.6061,
      "step": 151
    },
    {
      "epoch": 1.3954022988505748,
      "grad_norm": 0.8149594664573669,
      "learning_rate": 0.00020388349514563107,
      "loss": 0.5411,
      "step": 152
    },
    {
      "epoch": 1.4045977011494253,
      "grad_norm": 0.7658489942550659,
      "learning_rate": 0.00020315533980582524,
      "loss": 0.9712,
      "step": 153
    },
    {
      "epoch": 1.4137931034482758,
      "grad_norm": 0.7406046986579895,
      "learning_rate": 0.00020242718446601942,
      "loss": 0.3591,
      "step": 154
    },
    {
      "epoch": 1.4229885057471265,
      "grad_norm": 0.41466066241264343,
      "learning_rate": 0.00020169902912621357,
      "loss": 0.3579,
      "step": 155
    },
    {
      "epoch": 1.432183908045977,
      "grad_norm": 0.5462120175361633,
      "learning_rate": 0.00020097087378640775,
      "loss": 0.512,
      "step": 156
    },
    {
      "epoch": 1.4413793103448276,
      "grad_norm": 0.4008376896381378,
      "learning_rate": 0.00020024271844660193,
      "loss": 0.4698,
      "step": 157
    },
    {
      "epoch": 1.450574712643678,
      "grad_norm": 0.3887282609939575,
      "learning_rate": 0.0001995145631067961,
      "loss": 0.3842,
      "step": 158
    },
    {
      "epoch": 1.4597701149425286,
      "grad_norm": 0.5248357653617859,
      "learning_rate": 0.0001987864077669903,
      "loss": 0.3748,
      "step": 159
    },
    {
      "epoch": 1.4689655172413794,
      "grad_norm": 0.8764130473136902,
      "learning_rate": 0.00019805825242718447,
      "loss": 0.5248,
      "step": 160
    },
    {
      "epoch": 1.4781609195402299,
      "grad_norm": 0.631163477897644,
      "learning_rate": 0.00019733009708737865,
      "loss": 0.6423,
      "step": 161
    },
    {
      "epoch": 1.4873563218390804,
      "grad_norm": 0.5436691641807556,
      "learning_rate": 0.0001966019417475728,
      "loss": 0.4372,
      "step": 162
    },
    {
      "epoch": 1.4965517241379311,
      "grad_norm": 0.6378396153450012,
      "learning_rate": 0.00019587378640776698,
      "loss": 0.3392,
      "step": 163
    },
    {
      "epoch": 1.5057471264367817,
      "grad_norm": 0.41411787271499634,
      "learning_rate": 0.00019514563106796116,
      "loss": 0.3925,
      "step": 164
    },
    {
      "epoch": 1.5149425287356322,
      "grad_norm": 0.7309579849243164,
      "learning_rate": 0.00019441747572815533,
      "loss": 0.6451,
      "step": 165
    },
    {
      "epoch": 1.524137931034483,
      "grad_norm": 0.6073735356330872,
      "learning_rate": 0.0001936893203883495,
      "loss": 0.6802,
      "step": 166
    },
    {
      "epoch": 1.5333333333333332,
      "grad_norm": 0.37571704387664795,
      "learning_rate": 0.0001929611650485437,
      "loss": 0.5077,
      "step": 167
    },
    {
      "epoch": 1.542528735632184,
      "grad_norm": 0.9433541893959045,
      "learning_rate": 0.00019223300970873787,
      "loss": 0.3427,
      "step": 168
    },
    {
      "epoch": 1.5517241379310345,
      "grad_norm": 0.5408298373222351,
      "learning_rate": 0.00019150485436893205,
      "loss": 0.2311,
      "step": 169
    },
    {
      "epoch": 1.560919540229885,
      "grad_norm": 0.39010676741600037,
      "learning_rate": 0.0001907766990291262,
      "loss": 0.2743,
      "step": 170
    },
    {
      "epoch": 1.5701149425287357,
      "grad_norm": 0.6753641963005066,
      "learning_rate": 0.00019004854368932038,
      "loss": 0.8422,
      "step": 171
    },
    {
      "epoch": 1.5793103448275863,
      "grad_norm": 0.5095095634460449,
      "learning_rate": 0.00018932038834951456,
      "loss": 0.3418,
      "step": 172
    },
    {
      "epoch": 1.5885057471264368,
      "grad_norm": 0.6390260457992554,
      "learning_rate": 0.00018859223300970874,
      "loss": 0.4438,
      "step": 173
    },
    {
      "epoch": 1.5977011494252875,
      "grad_norm": 0.2993685007095337,
      "learning_rate": 0.00018786407766990291,
      "loss": 0.2958,
      "step": 174
    },
    {
      "epoch": 1.6068965517241378,
      "grad_norm": 0.7637106776237488,
      "learning_rate": 0.00018713592233009707,
      "loss": 0.6225,
      "step": 175
    },
    {
      "epoch": 1.6160919540229886,
      "grad_norm": 0.7003872990608215,
      "learning_rate": 0.00018640776699029122,
      "loss": 0.6526,
      "step": 176
    },
    {
      "epoch": 1.625287356321839,
      "grad_norm": 0.3579877018928528,
      "learning_rate": 0.0001856796116504854,
      "loss": 0.1985,
      "step": 177
    },
    {
      "epoch": 1.6344827586206896,
      "grad_norm": 0.2747999429702759,
      "learning_rate": 0.00018495145631067957,
      "loss": 0.2124,
      "step": 178
    },
    {
      "epoch": 1.6436781609195403,
      "grad_norm": 0.6537378430366516,
      "learning_rate": 0.00018422330097087375,
      "loss": 0.7906,
      "step": 179
    },
    {
      "epoch": 1.6528735632183909,
      "grad_norm": 0.4569692313671112,
      "learning_rate": 0.00018349514563106793,
      "loss": 0.4379,
      "step": 180
    },
    {
      "epoch": 1.6620689655172414,
      "grad_norm": 0.4810560941696167,
      "learning_rate": 0.0001827669902912621,
      "loss": 0.4129,
      "step": 181
    },
    {
      "epoch": 1.6712643678160921,
      "grad_norm": 0.5723876357078552,
      "learning_rate": 0.0001820388349514563,
      "loss": 0.3974,
      "step": 182
    },
    {
      "epoch": 1.6804597701149424,
      "grad_norm": 0.5330012440681458,
      "learning_rate": 0.00018131067961165047,
      "loss": 0.3204,
      "step": 183
    },
    {
      "epoch": 1.6896551724137931,
      "grad_norm": 0.6080510020256042,
      "learning_rate": 0.00018058252427184462,
      "loss": 0.7068,
      "step": 184
    },
    {
      "epoch": 1.6988505747126437,
      "grad_norm": 0.4264465570449829,
      "learning_rate": 0.0001798543689320388,
      "loss": 0.4423,
      "step": 185
    },
    {
      "epoch": 1.7080459770114942,
      "grad_norm": 0.5706063508987427,
      "learning_rate": 0.00017912621359223298,
      "loss": 0.3923,
      "step": 186
    },
    {
      "epoch": 1.717241379310345,
      "grad_norm": 0.45583376288414,
      "learning_rate": 0.00017839805825242716,
      "loss": 0.2542,
      "step": 187
    },
    {
      "epoch": 1.7264367816091954,
      "grad_norm": 0.5898862481117249,
      "learning_rate": 0.00017766990291262133,
      "loss": 0.3475,
      "step": 188
    },
    {
      "epoch": 1.735632183908046,
      "grad_norm": 0.6169423460960388,
      "learning_rate": 0.0001769417475728155,
      "loss": 0.5062,
      "step": 189
    },
    {
      "epoch": 1.7448275862068967,
      "grad_norm": 0.5277095437049866,
      "learning_rate": 0.0001762135922330097,
      "loss": 0.5095,
      "step": 190
    },
    {
      "epoch": 1.754022988505747,
      "grad_norm": 0.3574261963367462,
      "learning_rate": 0.00017548543689320387,
      "loss": 0.4086,
      "step": 191
    },
    {
      "epoch": 1.7632183908045977,
      "grad_norm": 0.4890098571777344,
      "learning_rate": 0.00017475728155339802,
      "loss": 0.5313,
      "step": 192
    },
    {
      "epoch": 1.7724137931034483,
      "grad_norm": 0.3213779628276825,
      "learning_rate": 0.0001740291262135922,
      "loss": 0.4056,
      "step": 193
    },
    {
      "epoch": 1.7816091954022988,
      "grad_norm": 0.3488066494464874,
      "learning_rate": 0.00017330097087378638,
      "loss": 0.2805,
      "step": 194
    },
    {
      "epoch": 1.7908045977011495,
      "grad_norm": 0.32463303208351135,
      "learning_rate": 0.00017257281553398056,
      "loss": 0.3157,
      "step": 195
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.5792247653007507,
      "learning_rate": 0.00017184466019417474,
      "loss": 0.4088,
      "step": 196
    },
    {
      "epoch": 1.8091954022988506,
      "grad_norm": 0.6414899230003357,
      "learning_rate": 0.00017111650485436891,
      "loss": 0.3401,
      "step": 197
    },
    {
      "epoch": 1.8183908045977013,
      "grad_norm": 0.5633428692817688,
      "learning_rate": 0.0001703883495145631,
      "loss": 0.3967,
      "step": 198
    },
    {
      "epoch": 1.8275862068965516,
      "grad_norm": 0.3799390196800232,
      "learning_rate": 0.00016966019417475724,
      "loss": 0.3257,
      "step": 199
    },
    {
      "epoch": 1.8367816091954023,
      "grad_norm": 0.26283133029937744,
      "learning_rate": 0.00016893203883495142,
      "loss": 0.2517,
      "step": 200
    },
    {
      "epoch": 1.8459770114942529,
      "grad_norm": 0.3688882887363434,
      "learning_rate": 0.0001682038834951456,
      "loss": 0.2523,
      "step": 201
    },
    {
      "epoch": 1.8551724137931034,
      "grad_norm": 0.3409745991230011,
      "learning_rate": 0.00016747572815533978,
      "loss": 0.327,
      "step": 202
    },
    {
      "epoch": 1.8643678160919541,
      "grad_norm": 0.4516524076461792,
      "learning_rate": 0.00016674757281553396,
      "loss": 0.4782,
      "step": 203
    },
    {
      "epoch": 1.8735632183908046,
      "grad_norm": 0.4798257350921631,
      "learning_rate": 0.00016601941747572814,
      "loss": 0.4515,
      "step": 204
    },
    {
      "epoch": 1.8827586206896552,
      "grad_norm": 0.49878618121147156,
      "learning_rate": 0.00016529126213592232,
      "loss": 0.2599,
      "step": 205
    },
    {
      "epoch": 1.891954022988506,
      "grad_norm": 0.42895984649658203,
      "learning_rate": 0.0001645631067961165,
      "loss": 0.4502,
      "step": 206
    },
    {
      "epoch": 1.9011494252873562,
      "grad_norm": 0.391862154006958,
      "learning_rate": 0.00016383495145631065,
      "loss": 0.3478,
      "step": 207
    },
    {
      "epoch": 1.910344827586207,
      "grad_norm": 0.6885610818862915,
      "learning_rate": 0.00016310679611650483,
      "loss": 0.6568,
      "step": 208
    },
    {
      "epoch": 1.9195402298850575,
      "grad_norm": 0.579025149345398,
      "learning_rate": 0.000162378640776699,
      "loss": 0.3724,
      "step": 209
    },
    {
      "epoch": 1.928735632183908,
      "grad_norm": 0.36686283349990845,
      "learning_rate": 0.00016165048543689318,
      "loss": 0.3496,
      "step": 210
    },
    {
      "epoch": 1.9379310344827587,
      "grad_norm": 0.3137669265270233,
      "learning_rate": 0.00016092233009708736,
      "loss": 0.1906,
      "step": 211
    },
    {
      "epoch": 1.9471264367816092,
      "grad_norm": 0.4134412407875061,
      "learning_rate": 0.00016019417475728154,
      "loss": 0.4461,
      "step": 212
    },
    {
      "epoch": 1.9563218390804598,
      "grad_norm": 0.3085227608680725,
      "learning_rate": 0.00015946601941747572,
      "loss": 0.2951,
      "step": 213
    },
    {
      "epoch": 1.9655172413793105,
      "grad_norm": 0.2577720284461975,
      "learning_rate": 0.0001587378640776699,
      "loss": 0.2708,
      "step": 214
    },
    {
      "epoch": 1.9747126436781608,
      "grad_norm": 0.6653198003768921,
      "learning_rate": 0.00015800970873786405,
      "loss": 0.9269,
      "step": 215
    },
    {
      "epoch": 1.9839080459770115,
      "grad_norm": 0.6150757670402527,
      "learning_rate": 0.00015728155339805823,
      "loss": 0.692,
      "step": 216
    },
    {
      "epoch": 1.993103448275862,
      "grad_norm": 0.44791334867477417,
      "learning_rate": 0.0001565533980582524,
      "loss": 0.4351,
      "step": 217
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.7618283629417419,
      "learning_rate": 0.00015582524271844658,
      "loss": 0.7963,
      "step": 218
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.5562227964401245,
      "eval_runtime": 76.4203,
      "eval_samples_per_second": 4.344,
      "eval_steps_per_second": 2.172,
      "step": 218
    },
    {
      "epoch": 2.0091954022988507,
      "grad_norm": 0.32502397894859314,
      "learning_rate": 0.00015509708737864076,
      "loss": 0.2374,
      "step": 219
    },
    {
      "epoch": 2.018390804597701,
      "grad_norm": 0.3833533525466919,
      "learning_rate": 0.00015436893203883494,
      "loss": 0.3454,
      "step": 220
    },
    {
      "epoch": 2.027586206896552,
      "grad_norm": 0.3064585328102112,
      "learning_rate": 0.00015364077669902912,
      "loss": 0.2924,
      "step": 221
    },
    {
      "epoch": 2.036781609195402,
      "grad_norm": 0.402799516916275,
      "learning_rate": 0.0001529126213592233,
      "loss": 0.3828,
      "step": 222
    },
    {
      "epoch": 2.045977011494253,
      "grad_norm": 0.30839601159095764,
      "learning_rate": 0.00015218446601941745,
      "loss": 0.2421,
      "step": 223
    },
    {
      "epoch": 2.0551724137931036,
      "grad_norm": 0.4161797761917114,
      "learning_rate": 0.00015145631067961163,
      "loss": 0.2989,
      "step": 224
    },
    {
      "epoch": 2.064367816091954,
      "grad_norm": 0.4436909556388855,
      "learning_rate": 0.0001507281553398058,
      "loss": 0.3328,
      "step": 225
    },
    {
      "epoch": 2.0735632183908046,
      "grad_norm": 0.5570560693740845,
      "learning_rate": 0.00015,
      "loss": 0.3132,
      "step": 226
    },
    {
      "epoch": 2.0827586206896553,
      "grad_norm": 0.7697573304176331,
      "learning_rate": 0.00014927184466019417,
      "loss": 0.5465,
      "step": 227
    },
    {
      "epoch": 2.0919540229885056,
      "grad_norm": 0.44799014925956726,
      "learning_rate": 0.00014854368932038834,
      "loss": 0.2612,
      "step": 228
    },
    {
      "epoch": 2.1011494252873564,
      "grad_norm": 0.4220842719078064,
      "learning_rate": 0.00014781553398058252,
      "loss": 0.239,
      "step": 229
    },
    {
      "epoch": 2.110344827586207,
      "grad_norm": 0.480135053396225,
      "learning_rate": 0.00014708737864077667,
      "loss": 0.339,
      "step": 230
    },
    {
      "epoch": 2.1195402298850574,
      "grad_norm": 0.46936216950416565,
      "learning_rate": 0.00014635922330097085,
      "loss": 0.2952,
      "step": 231
    },
    {
      "epoch": 2.128735632183908,
      "grad_norm": 0.6709425449371338,
      "learning_rate": 0.00014563106796116503,
      "loss": 0.3496,
      "step": 232
    },
    {
      "epoch": 2.1379310344827585,
      "grad_norm": 0.37677356600761414,
      "learning_rate": 0.0001449029126213592,
      "loss": 0.3361,
      "step": 233
    },
    {
      "epoch": 2.147126436781609,
      "grad_norm": 1.1358637809753418,
      "learning_rate": 0.0001441747572815534,
      "loss": 0.5066,
      "step": 234
    },
    {
      "epoch": 2.15632183908046,
      "grad_norm": 0.5674331188201904,
      "learning_rate": 0.00014344660194174757,
      "loss": 0.3828,
      "step": 235
    },
    {
      "epoch": 2.1655172413793102,
      "grad_norm": 0.473661333322525,
      "learning_rate": 0.00014271844660194175,
      "loss": 0.2879,
      "step": 236
    },
    {
      "epoch": 2.174712643678161,
      "grad_norm": 0.48269397020339966,
      "learning_rate": 0.00014199029126213592,
      "loss": 0.3322,
      "step": 237
    },
    {
      "epoch": 2.1839080459770113,
      "grad_norm": 0.6089305281639099,
      "learning_rate": 0.00014126213592233008,
      "loss": 0.3218,
      "step": 238
    },
    {
      "epoch": 2.193103448275862,
      "grad_norm": 0.3989170789718628,
      "learning_rate": 0.00014053398058252425,
      "loss": 0.2518,
      "step": 239
    },
    {
      "epoch": 2.2022988505747128,
      "grad_norm": 0.3296070992946625,
      "learning_rate": 0.00013980582524271843,
      "loss": 0.2045,
      "step": 240
    },
    {
      "epoch": 2.211494252873563,
      "grad_norm": 0.8743225336074829,
      "learning_rate": 0.0001390776699029126,
      "loss": 0.6216,
      "step": 241
    },
    {
      "epoch": 2.220689655172414,
      "grad_norm": 0.36944878101348877,
      "learning_rate": 0.0001383495145631068,
      "loss": 0.1719,
      "step": 242
    },
    {
      "epoch": 2.2298850574712645,
      "grad_norm": 0.33532336354255676,
      "learning_rate": 0.00013762135922330097,
      "loss": 0.2154,
      "step": 243
    },
    {
      "epoch": 2.239080459770115,
      "grad_norm": 0.7817918658256531,
      "learning_rate": 0.00013689320388349515,
      "loss": 0.48,
      "step": 244
    },
    {
      "epoch": 2.2482758620689656,
      "grad_norm": 0.8702794313430786,
      "learning_rate": 0.00013616504854368933,
      "loss": 0.6025,
      "step": 245
    },
    {
      "epoch": 2.2574712643678163,
      "grad_norm": 0.3685101866722107,
      "learning_rate": 0.00013543689320388348,
      "loss": 0.164,
      "step": 246
    },
    {
      "epoch": 2.2666666666666666,
      "grad_norm": 0.7941250801086426,
      "learning_rate": 0.00013470873786407766,
      "loss": 0.508,
      "step": 247
    },
    {
      "epoch": 2.2758620689655173,
      "grad_norm": 0.41759851574897766,
      "learning_rate": 0.00013398058252427184,
      "loss": 0.1994,
      "step": 248
    },
    {
      "epoch": 2.2850574712643676,
      "grad_norm": 0.653108537197113,
      "learning_rate": 0.00013325242718446601,
      "loss": 0.3635,
      "step": 249
    },
    {
      "epoch": 2.2942528735632184,
      "grad_norm": 0.7277186512947083,
      "learning_rate": 0.0001325242718446602,
      "loss": 0.4847,
      "step": 250
    },
    {
      "epoch": 2.303448275862069,
      "grad_norm": 0.4302230775356293,
      "learning_rate": 0.00013179611650485437,
      "loss": 0.296,
      "step": 251
    },
    {
      "epoch": 2.3126436781609194,
      "grad_norm": 0.3574405312538147,
      "learning_rate": 0.00013106796116504852,
      "loss": 0.1844,
      "step": 252
    },
    {
      "epoch": 2.32183908045977,
      "grad_norm": 0.5645660161972046,
      "learning_rate": 0.0001303398058252427,
      "loss": 0.3868,
      "step": 253
    },
    {
      "epoch": 2.3310344827586205,
      "grad_norm": 0.4044601619243622,
      "learning_rate": 0.00012961165048543688,
      "loss": 0.2447,
      "step": 254
    },
    {
      "epoch": 2.340229885057471,
      "grad_norm": 0.41919222474098206,
      "learning_rate": 0.00012888349514563106,
      "loss": 0.258,
      "step": 255
    },
    {
      "epoch": 2.349425287356322,
      "grad_norm": 0.5476058721542358,
      "learning_rate": 0.00012815533980582524,
      "loss": 0.3241,
      "step": 256
    },
    {
      "epoch": 2.3586206896551722,
      "grad_norm": 0.8460379838943481,
      "learning_rate": 0.0001274271844660194,
      "loss": 0.4723,
      "step": 257
    },
    {
      "epoch": 2.367816091954023,
      "grad_norm": 0.5093741416931152,
      "learning_rate": 0.00012669902912621357,
      "loss": 0.2634,
      "step": 258
    },
    {
      "epoch": 2.3770114942528737,
      "grad_norm": 0.6395347714424133,
      "learning_rate": 0.00012597087378640775,
      "loss": 0.4046,
      "step": 259
    },
    {
      "epoch": 2.386206896551724,
      "grad_norm": 0.7581174373626709,
      "learning_rate": 0.00012524271844660192,
      "loss": 0.5078,
      "step": 260
    },
    {
      "epoch": 2.3954022988505748,
      "grad_norm": 0.49910837411880493,
      "learning_rate": 0.0001245145631067961,
      "loss": 0.2991,
      "step": 261
    },
    {
      "epoch": 2.4045977011494255,
      "grad_norm": 0.34560054540634155,
      "learning_rate": 0.00012378640776699028,
      "loss": 0.2364,
      "step": 262
    },
    {
      "epoch": 2.413793103448276,
      "grad_norm": 0.3754998743534088,
      "learning_rate": 0.00012305825242718446,
      "loss": 0.2107,
      "step": 263
    },
    {
      "epoch": 2.4229885057471265,
      "grad_norm": 0.33737894892692566,
      "learning_rate": 0.0001223300970873786,
      "loss": 0.1509,
      "step": 264
    },
    {
      "epoch": 2.432183908045977,
      "grad_norm": 0.4702773690223694,
      "learning_rate": 0.0001216019417475728,
      "loss": 0.2477,
      "step": 265
    },
    {
      "epoch": 2.4413793103448276,
      "grad_norm": 0.6840780973434448,
      "learning_rate": 0.00012087378640776697,
      "loss": 0.521,
      "step": 266
    },
    {
      "epoch": 2.4505747126436783,
      "grad_norm": 0.4670672118663788,
      "learning_rate": 0.00012014563106796115,
      "loss": 0.3054,
      "step": 267
    },
    {
      "epoch": 2.4597701149425286,
      "grad_norm": 0.34090566635131836,
      "learning_rate": 0.00011941747572815533,
      "loss": 0.2215,
      "step": 268
    },
    {
      "epoch": 2.4689655172413794,
      "grad_norm": 0.32637739181518555,
      "learning_rate": 0.0001186893203883495,
      "loss": 0.211,
      "step": 269
    },
    {
      "epoch": 2.4781609195402297,
      "grad_norm": 0.325579971075058,
      "learning_rate": 0.00011796116504854367,
      "loss": 0.1892,
      "step": 270
    },
    {
      "epoch": 2.4873563218390804,
      "grad_norm": 0.2964633107185364,
      "learning_rate": 0.00011723300970873785,
      "loss": 0.1838,
      "step": 271
    },
    {
      "epoch": 2.496551724137931,
      "grad_norm": 0.6292815804481506,
      "learning_rate": 0.00011650485436893203,
      "loss": 0.5632,
      "step": 272
    },
    {
      "epoch": 2.5057471264367814,
      "grad_norm": 0.3210340738296509,
      "learning_rate": 0.0001157766990291262,
      "loss": 0.1643,
      "step": 273
    },
    {
      "epoch": 2.514942528735632,
      "grad_norm": 0.35492682456970215,
      "learning_rate": 0.00011504854368932037,
      "loss": 0.2607,
      "step": 274
    },
    {
      "epoch": 2.524137931034483,
      "grad_norm": 0.4605109393596649,
      "learning_rate": 0.00011432038834951455,
      "loss": 0.3006,
      "step": 275
    },
    {
      "epoch": 2.533333333333333,
      "grad_norm": 0.21065670251846313,
      "learning_rate": 0.00011359223300970873,
      "loss": 0.0945,
      "step": 276
    },
    {
      "epoch": 2.542528735632184,
      "grad_norm": 0.4777117669582367,
      "learning_rate": 0.00011286407766990291,
      "loss": 0.2718,
      "step": 277
    },
    {
      "epoch": 2.5517241379310347,
      "grad_norm": 0.4873185157775879,
      "learning_rate": 0.00011213592233009707,
      "loss": 0.3336,
      "step": 278
    },
    {
      "epoch": 2.560919540229885,
      "grad_norm": 0.3319976329803467,
      "learning_rate": 0.00011140776699029125,
      "loss": 0.2026,
      "step": 279
    },
    {
      "epoch": 2.5701149425287357,
      "grad_norm": 0.3072815239429474,
      "learning_rate": 0.00011067961165048543,
      "loss": 0.187,
      "step": 280
    },
    {
      "epoch": 2.5793103448275865,
      "grad_norm": 0.3600887656211853,
      "learning_rate": 0.00010995145631067961,
      "loss": 0.1711,
      "step": 281
    },
    {
      "epoch": 2.5885057471264368,
      "grad_norm": 0.40916576981544495,
      "learning_rate": 0.00010922330097087377,
      "loss": 0.2618,
      "step": 282
    },
    {
      "epoch": 2.5977011494252875,
      "grad_norm": 0.6749149560928345,
      "learning_rate": 0.00010849514563106795,
      "loss": 0.4636,
      "step": 283
    },
    {
      "epoch": 2.606896551724138,
      "grad_norm": 0.29242533445358276,
      "learning_rate": 0.00010776699029126213,
      "loss": 0.1577,
      "step": 284
    },
    {
      "epoch": 2.6160919540229886,
      "grad_norm": 0.3938550651073456,
      "learning_rate": 0.00010703883495145631,
      "loss": 0.1937,
      "step": 285
    },
    {
      "epoch": 2.625287356321839,
      "grad_norm": 0.3914635479450226,
      "learning_rate": 0.00010631067961165047,
      "loss": 0.2525,
      "step": 286
    },
    {
      "epoch": 2.6344827586206896,
      "grad_norm": 0.3480489253997803,
      "learning_rate": 0.00010558252427184465,
      "loss": 0.178,
      "step": 287
    },
    {
      "epoch": 2.6436781609195403,
      "grad_norm": 0.230756476521492,
      "learning_rate": 0.00010485436893203883,
      "loss": 0.1148,
      "step": 288
    },
    {
      "epoch": 2.6528735632183906,
      "grad_norm": 0.44225171208381653,
      "learning_rate": 0.00010412621359223301,
      "loss": 0.3141,
      "step": 289
    },
    {
      "epoch": 2.6620689655172414,
      "grad_norm": 0.4487713873386383,
      "learning_rate": 0.00010339805825242718,
      "loss": 0.2956,
      "step": 290
    },
    {
      "epoch": 2.671264367816092,
      "grad_norm": 0.499773234128952,
      "learning_rate": 0.00010266990291262135,
      "loss": 0.3489,
      "step": 291
    },
    {
      "epoch": 2.6804597701149424,
      "grad_norm": 0.40772294998168945,
      "learning_rate": 0.00010194174757281553,
      "loss": 0.2208,
      "step": 292
    },
    {
      "epoch": 2.689655172413793,
      "grad_norm": 0.5451008081436157,
      "learning_rate": 0.00010121359223300971,
      "loss": 0.3616,
      "step": 293
    },
    {
      "epoch": 2.698850574712644,
      "grad_norm": 0.4137297570705414,
      "learning_rate": 0.00010048543689320388,
      "loss": 0.2938,
      "step": 294
    },
    {
      "epoch": 2.708045977011494,
      "grad_norm": 0.839198648929596,
      "learning_rate": 9.975728155339806e-05,
      "loss": 0.6327,
      "step": 295
    },
    {
      "epoch": 2.717241379310345,
      "grad_norm": 0.6517747044563293,
      "learning_rate": 9.902912621359223e-05,
      "loss": 0.4386,
      "step": 296
    },
    {
      "epoch": 2.7264367816091957,
      "grad_norm": 0.3699157238006592,
      "learning_rate": 9.83009708737864e-05,
      "loss": 0.2534,
      "step": 297
    },
    {
      "epoch": 2.735632183908046,
      "grad_norm": 0.4842491149902344,
      "learning_rate": 9.757281553398058e-05,
      "loss": 0.3054,
      "step": 298
    },
    {
      "epoch": 2.7448275862068967,
      "grad_norm": 0.34171465039253235,
      "learning_rate": 9.684466019417476e-05,
      "loss": 0.2192,
      "step": 299
    },
    {
      "epoch": 2.754022988505747,
      "grad_norm": 0.43019741773605347,
      "learning_rate": 9.611650485436893e-05,
      "loss": 0.2504,
      "step": 300
    },
    {
      "epoch": 2.7632183908045977,
      "grad_norm": 0.42596566677093506,
      "learning_rate": 9.53883495145631e-05,
      "loss": 0.2598,
      "step": 301
    },
    {
      "epoch": 2.772413793103448,
      "grad_norm": 0.675719141960144,
      "learning_rate": 9.466019417475728e-05,
      "loss": 0.3822,
      "step": 302
    },
    {
      "epoch": 2.781609195402299,
      "grad_norm": 0.4107106924057007,
      "learning_rate": 9.393203883495146e-05,
      "loss": 0.2068,
      "step": 303
    },
    {
      "epoch": 2.7908045977011495,
      "grad_norm": 0.6644109487533569,
      "learning_rate": 9.320388349514561e-05,
      "loss": 0.3806,
      "step": 304
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.3467462956905365,
      "learning_rate": 9.247572815533979e-05,
      "loss": 0.1818,
      "step": 305
    },
    {
      "epoch": 2.8091954022988506,
      "grad_norm": 0.26561155915260315,
      "learning_rate": 9.174757281553397e-05,
      "loss": 0.1497,
      "step": 306
    },
    {
      "epoch": 2.8183908045977013,
      "grad_norm": 0.7984749674797058,
      "learning_rate": 9.101941747572814e-05,
      "loss": 0.5195,
      "step": 307
    },
    {
      "epoch": 2.8275862068965516,
      "grad_norm": 0.252616286277771,
      "learning_rate": 9.029126213592231e-05,
      "loss": 0.1348,
      "step": 308
    },
    {
      "epoch": 2.8367816091954023,
      "grad_norm": 0.3006296157836914,
      "learning_rate": 8.956310679611649e-05,
      "loss": 0.1629,
      "step": 309
    },
    {
      "epoch": 2.845977011494253,
      "grad_norm": 0.3508792519569397,
      "learning_rate": 8.883495145631067e-05,
      "loss": 0.2039,
      "step": 310
    },
    {
      "epoch": 2.8551724137931034,
      "grad_norm": 0.2189541608095169,
      "learning_rate": 8.810679611650485e-05,
      "loss": 0.1008,
      "step": 311
    },
    {
      "epoch": 2.864367816091954,
      "grad_norm": 0.7708775997161865,
      "learning_rate": 8.737864077669901e-05,
      "loss": 0.5565,
      "step": 312
    },
    {
      "epoch": 2.873563218390805,
      "grad_norm": 0.3029266595840454,
      "learning_rate": 8.665048543689319e-05,
      "loss": 0.1925,
      "step": 313
    },
    {
      "epoch": 2.882758620689655,
      "grad_norm": 0.29565373063087463,
      "learning_rate": 8.592233009708737e-05,
      "loss": 0.1708,
      "step": 314
    },
    {
      "epoch": 2.891954022988506,
      "grad_norm": 0.66410893201828,
      "learning_rate": 8.519417475728155e-05,
      "loss": 0.4796,
      "step": 315
    },
    {
      "epoch": 2.901149425287356,
      "grad_norm": 0.48440781235694885,
      "learning_rate": 8.446601941747571e-05,
      "loss": 0.3075,
      "step": 316
    },
    {
      "epoch": 2.910344827586207,
      "grad_norm": 0.7811682820320129,
      "learning_rate": 8.373786407766989e-05,
      "loss": 0.5639,
      "step": 317
    },
    {
      "epoch": 2.9195402298850572,
      "grad_norm": 0.3656238913536072,
      "learning_rate": 8.300970873786407e-05,
      "loss": 0.208,
      "step": 318
    },
    {
      "epoch": 2.928735632183908,
      "grad_norm": 0.34803682565689087,
      "learning_rate": 8.228155339805825e-05,
      "loss": 0.1549,
      "step": 319
    },
    {
      "epoch": 2.9379310344827587,
      "grad_norm": 0.7989238500595093,
      "learning_rate": 8.155339805825241e-05,
      "loss": 0.516,
      "step": 320
    },
    {
      "epoch": 2.947126436781609,
      "grad_norm": 0.539603054523468,
      "learning_rate": 8.082524271844659e-05,
      "loss": 0.3175,
      "step": 321
    },
    {
      "epoch": 2.9563218390804598,
      "grad_norm": 0.43805384635925293,
      "learning_rate": 8.009708737864077e-05,
      "loss": 0.2689,
      "step": 322
    },
    {
      "epoch": 2.9655172413793105,
      "grad_norm": 0.3973037302494049,
      "learning_rate": 7.936893203883495e-05,
      "loss": 0.2238,
      "step": 323
    },
    {
      "epoch": 2.974712643678161,
      "grad_norm": 0.555601954460144,
      "learning_rate": 7.864077669902911e-05,
      "loss": 0.3414,
      "step": 324
    },
    {
      "epoch": 2.9839080459770115,
      "grad_norm": 0.7183153033256531,
      "learning_rate": 7.791262135922329e-05,
      "loss": 0.3484,
      "step": 325
    },
    {
      "epoch": 2.9931034482758623,
      "grad_norm": 0.7718084454536438,
      "learning_rate": 7.718446601941747e-05,
      "loss": 0.4512,
      "step": 326
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.1820600032806396,
      "learning_rate": 7.645631067961165e-05,
      "loss": 0.6338,
      "step": 327
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.6946974992752075,
      "eval_runtime": 76.3864,
      "eval_samples_per_second": 4.346,
      "eval_steps_per_second": 2.173,
      "step": 327
    }
  ],
  "logging_steps": 1,
  "max_steps": 432,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.3656641543714816e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}