{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.015384615384615,
  "eval_steps": 500000000,
  "global_step": 65,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "crossentropy": 3.136348009109497,
      "epoch": 0.015384615384615385,
      "grad_norm": 2.609375,
      "learning_rate": 0.0006000000000000001,
      "loss": 62.727,
      "step": 1
    },
    {
      "crossentropy": 3.124629259109497,
      "epoch": 0.03076923076923077,
      "grad_norm": 2.640625,
      "learning_rate": 0.0012000000000000001,
      "loss": 62.4926,
      "step": 2
    },
    {
      "crossentropy": 3.071411609649658,
      "epoch": 0.046153846153846156,
      "grad_norm": 1.671875,
      "learning_rate": 0.0018,
      "loss": 61.4282,
      "step": 3
    },
    {
      "crossentropy": 2.8995280265808105,
      "epoch": 0.06153846153846154,
      "grad_norm": 2.203125,
      "learning_rate": 0.0024000000000000002,
      "loss": 57.9906,
      "step": 4
    },
    {
      "crossentropy": 2.881728172302246,
      "epoch": 0.07692307692307693,
      "grad_norm": 2.9375,
      "learning_rate": 0.003,
      "loss": 57.6346,
      "step": 5
    },
    {
      "crossentropy": 2.9400224685668945,
      "epoch": 0.09230769230769231,
      "grad_norm": 2.265625,
      "learning_rate": 0.0036,
      "loss": 58.8004,
      "step": 6
    },
    {
      "crossentropy": 3.066098690032959,
      "epoch": 0.1076923076923077,
      "grad_norm": 2.375,
      "learning_rate": 0.0042,
      "loss": 61.322,
      "step": 7
    },
    {
      "crossentropy": 3.1193857192993164,
      "epoch": 0.12307692307692308,
      "grad_norm": 2.046875,
      "learning_rate": 0.0048000000000000004,
      "loss": 62.3877,
      "step": 8
    },
    {
      "crossentropy": 3.0263144969940186,
      "epoch": 1.0153846153846153,
      "grad_norm": 1.828125,
      "learning_rate": 0.0054,
      "loss": 60.5263,
      "step": 9
    },
    {
      "crossentropy": 2.8756701946258545,
      "epoch": 1.0307692307692307,
      "grad_norm": 2.15625,
      "learning_rate": 0.006,
      "loss": 57.5134,
      "step": 10
    },
    {
      "crossentropy": 2.9259140491485596,
      "epoch": 1.0461538461538462,
      "grad_norm": 2.40625,
      "learning_rate": 0.005995107311778406,
      "loss": 58.5183,
      "step": 11
    },
    {
      "crossentropy": 2.937499761581421,
      "epoch": 1.0615384615384615,
      "grad_norm": 2.828125,
      "learning_rate": 0.00598044520604565,
      "loss": 58.75,
      "step": 12
    },
    {
      "crossentropy": 2.806440830230713,
      "epoch": 1.0769230769230769,
      "grad_norm": 2.109375,
      "learning_rate": 0.005956061507543079,
      "loss": 56.1288,
      "step": 13
    },
    {
      "crossentropy": 2.9169092178344727,
      "epoch": 1.0923076923076924,
      "grad_norm": 2.1875,
      "learning_rate": 0.005922035750827001,
      "loss": 58.3382,
      "step": 14
    },
    {
      "crossentropy": 2.781325101852417,
      "epoch": 1.1076923076923078,
      "grad_norm": 2.609375,
      "grad_norm_var": 0.12228902180989583,
      "learning_rate": 0.005878478920843493,
      "loss": 55.6265,
      "step": 15
    },
    {
      "crossentropy": 2.775790214538574,
      "epoch": 1.123076923076923,
      "grad_norm": 1.9140625,
      "grad_norm_var": 0.12452977498372396,
      "learning_rate": 0.005825533090918575,
      "loss": 55.5158,
      "step": 16
    },
    {
      "crossentropy": 2.780381202697754,
      "epoch": 2.0153846153846153,
      "grad_norm": 2.40625,
      "grad_norm_var": 0.1012115478515625,
      "learning_rate": 0.00576337095934455,
      "loss": 55.6076,
      "step": 17
    },
    {
      "crossentropy": 2.7276179790496826,
      "epoch": 2.0307692307692307,
      "grad_norm": 2.109375,
      "grad_norm_var": 0.1025299072265625,
      "learning_rate": 0.005692195286074075,
      "loss": 54.5524,
      "step": 18
    },
    {
      "crossentropy": 2.514774799346924,
      "epoch": 2.046153846153846,
      "grad_norm": 2.0625,
      "grad_norm_var": 0.07119852701822917,
      "learning_rate": 0.005612238231359332,
      "loss": 50.2955,
      "step": 19
    },
    {
      "crossentropy": 2.735562801361084,
      "epoch": 2.0615384615384613,
      "grad_norm": 2.375,
      "grad_norm_var": 0.07284342447916667,
      "learning_rate": 0.005523760598493544,
      "loss": 54.7113,
      "step": 20
    },
    {
      "crossentropy": 2.6968579292297363,
      "epoch": 2.076923076923077,
      "grad_norm": 1.9453125,
      "grad_norm_var": 0.0749834696451823,
      "learning_rate": 0.005427050983124842,
      "loss": 53.9372,
      "step": 21
    },
    {
      "crossentropy": 2.7723076343536377,
      "epoch": 2.0923076923076924,
      "grad_norm": 1.8046875,
      "grad_norm_var": 0.08308003743489584,
      "learning_rate": 0.005322424831917248,
      "loss": 55.4462,
      "step": 22
    },
    {
      "crossentropy": 2.6105570793151855,
      "epoch": 2.1076923076923078,
      "grad_norm": 1.53125,
      "grad_norm_var": 0.10808919270833334,
      "learning_rate": 0.005210223413629215,
      "loss": 52.2111,
      "step": 23
    },
    {
      "crossentropy": 2.627753257751465,
      "epoch": 2.123076923076923,
      "grad_norm": 1.8046875,
      "grad_norm_var": 0.10908788045247396,
      "learning_rate": 0.005090812705965881,
      "loss": 52.5551,
      "step": 24
    },
    {
      "crossentropy": 2.449847459793091,
      "epoch": 3.0153846153846153,
      "grad_norm": 2.140625,
      "grad_norm_var": 0.10986226399739583,
      "learning_rate": 0.004964582201835856,
      "loss": 48.9969,
      "step": 25
    },
    {
      "crossentropy": 2.566053867340088,
      "epoch": 3.0307692307692307,
      "grad_norm": 1.9453125,
      "grad_norm_var": 0.07247492472330729,
      "learning_rate": 0.004831943638906315,
      "loss": 51.3211,
      "step": 26
    },
    {
      "crossentropy": 2.5103166103363037,
      "epoch": 3.046153846153846,
      "grad_norm": 2.40625,
      "grad_norm_var": 0.08066991170247396,
      "learning_rate": 0.004693329656600308,
      "loss": 50.2063,
      "step": 27
    },
    {
      "crossentropy": 2.586671829223633,
      "epoch": 3.0615384615384613,
      "grad_norm": 1.8671875,
      "grad_norm_var": 0.08163960774739583,
      "learning_rate": 0.004549192384916886,
      "loss": 51.7334,
      "step": 28
    },
    {
      "crossentropy": 2.383892297744751,
      "epoch": 3.076923076923077,
      "grad_norm": 1.7890625,
      "grad_norm_var": 0.06142552693684896,
      "learning_rate": 0.004400001969677022,
      "loss": 47.6778,
      "step": 29
    },
    {
      "crossentropy": 2.6217358112335205,
      "epoch": 3.0923076923076924,
      "grad_norm": 1.890625,
      "grad_norm_var": 0.06169331868489583,
      "learning_rate": 0.00424624503900566,
      "loss": 52.4347,
      "step": 30
    },
    {
      "crossentropy": 2.494020700454712,
      "epoch": 3.1076923076923078,
      "grad_norm": 1.75,
      "grad_norm_var": 0.06497777303059896,
      "learning_rate": 0.004088423116051923,
      "loss": 49.8804,
      "step": 31
    },
    {
      "crossentropy": 2.6089510917663574,
      "epoch": 3.123076923076923,
      "grad_norm": 1.859375,
      "grad_norm_var": 0.052374013264973956,
      "learning_rate": 0.0039270509831248425,
      "loss": 52.179,
      "step": 32
    },
    {
      "crossentropy": 2.3228707313537598,
      "epoch": 4.015384615384615,
      "grad_norm": 2.015625,
      "grad_norm_var": 0.050022125244140625,
      "learning_rate": 0.0037626550025804616,
      "loss": 46.4574,
      "step": 33
    },
    {
      "crossentropy": 2.5025362968444824,
      "epoch": 4.030769230769231,
      "grad_norm": 1.9765625,
      "grad_norm_var": 0.0360015869140625,
      "learning_rate": 0.0035957713999372363,
      "loss": 50.0507,
      "step": 34
    },
    {
      "crossentropy": 2.3985209465026855,
      "epoch": 4.046153846153846,
      "grad_norm": 2.109375,
      "grad_norm_var": 0.03868789672851562,
      "learning_rate": 0.003426944514819856,
      "loss": 47.9704,
      "step": 35
    },
    {
      "crossentropy": 2.4309520721435547,
      "epoch": 4.061538461538461,
      "grad_norm": 1.9140625,
      "grad_norm_var": 0.037904612223307294,
      "learning_rate": 0.0032567250254365195,
      "loss": 48.619,
      "step": 36
    },
    {
      "crossentropy": 2.406186103820801,
      "epoch": 4.076923076923077,
      "grad_norm": 1.6875,
      "grad_norm_var": 0.031404368082682294,
      "learning_rate": 0.0030856681523810884,
      "loss": 48.1237,
      "step": 37
    },
    {
      "crossentropy": 2.3365566730499268,
      "epoch": 4.092307692307692,
      "grad_norm": 1.75,
      "grad_norm_var": 0.03247782389322917,
      "learning_rate": 0.0029143318476189117,
      "loss": 46.7311,
      "step": 38
    },
    {
      "crossentropy": 2.3258492946624756,
      "epoch": 4.107692307692307,
      "grad_norm": 1.6484375,
      "grad_norm_var": 0.0364654541015625,
      "learning_rate": 0.0027432749745634815,
      "loss": 46.517,
      "step": 39
    },
    {
      "crossentropy": 2.4440741539001465,
      "epoch": 4.123076923076923,
      "grad_norm": 1.8203125,
      "grad_norm_var": 0.03316014607747396,
      "learning_rate": 0.002573055485180145,
      "loss": 48.8815,
      "step": 40
    },
    {
      "crossentropy": 2.299198865890503,
      "epoch": 5.015384615384615,
      "grad_norm": 1.8515625,
      "grad_norm_var": 0.013963826497395833,
      "learning_rate": 0.0024042286000627642,
      "loss": 45.984,
      "step": 41
    },
    {
      "crossentropy": 2.284895658493042,
      "epoch": 5.030769230769231,
      "grad_norm": 1.71875,
      "grad_norm_var": 0.015012359619140625,
      "learning_rate": 0.00223734499741954,
      "loss": 45.6979,
      "step": 42
    },
    {
      "crossentropy": 2.3056931495666504,
      "epoch": 5.046153846153846,
      "grad_norm": 1.7109375,
      "grad_norm_var": 0.01593805948893229,
      "learning_rate": 0.002072949016875158,
      "loss": 46.1139,
      "step": 43
    },
    {
      "crossentropy": 2.3087401390075684,
      "epoch": 5.061538461538461,
      "grad_norm": 1.765625,
      "grad_norm_var": 0.01601130167643229,
      "learning_rate": 0.0019115768839480774,
      "loss": 46.1748,
      "step": 44
    },
    {
      "crossentropy": 2.3146181106567383,
      "epoch": 5.076923076923077,
      "grad_norm": 1.7890625,
      "grad_norm_var": 0.015697224934895834,
      "learning_rate": 0.0017537549609943411,
      "loss": 46.2924,
      "step": 45
    },
    {
      "crossentropy": 2.21163010597229,
      "epoch": 5.092307692307692,
      "grad_norm": 1.5078125,
      "grad_norm_var": 0.02209447224934896,
      "learning_rate": 0.0015999980303229788,
      "loss": 44.2326,
      "step": 46
    },
    {
      "crossentropy": 2.4418914318084717,
      "epoch": 5.107692307692307,
      "grad_norm": 1.609375,
      "grad_norm_var": 0.02432428995768229,
      "learning_rate": 0.0014508076150831146,
      "loss": 48.8378,
      "step": 47
    },
    {
      "crossentropy": 2.319216251373291,
      "epoch": 5.123076923076923,
      "grad_norm": 1.6328125,
      "grad_norm_var": 0.0221435546875,
      "learning_rate": 0.0013066703433996932,
      "loss": 46.3843,
      "step": 48
    },
    {
      "crossentropy": 2.3361504077911377,
      "epoch": 6.015384615384615,
      "grad_norm": 1.4765625,
      "grad_norm_var": 0.014581044514973959,
      "learning_rate": 0.0011680563610936853,
      "loss": 46.723,
      "step": 49
    },
    {
      "crossentropy": 2.2992982864379883,
      "epoch": 6.030769230769231,
      "grad_norm": 1.5625,
      "grad_norm_var": 0.012669881184895834,
      "learning_rate": 0.001035417798164145,
      "loss": 45.986,
      "step": 50
    },
    {
      "crossentropy": 2.2670037746429443,
      "epoch": 6.046153846153846,
      "grad_norm": 1.5,
      "grad_norm_var": 0.014842732747395834,
      "learning_rate": 0.0009091872940341201,
      "loss": 45.3401,
      "step": 51
    },
    {
      "crossentropy": 2.3044376373291016,
      "epoch": 6.061538461538461,
      "grad_norm": 1.5625,
      "grad_norm_var": 0.015160115559895833,
      "learning_rate": 0.0007897765863707849,
      "loss": 46.0888,
      "step": 52
    },
    {
      "crossentropy": 2.183406352996826,
      "epoch": 6.076923076923077,
      "grad_norm": 1.484375,
      "grad_norm_var": 0.017162831624348958,
      "learning_rate": 0.0006775751680827525,
      "loss": 43.6681,
      "step": 53
    },
    {
      "crossentropy": 2.323101758956909,
      "epoch": 6.092307692307692,
      "grad_norm": 1.515625,
      "grad_norm_var": 0.016161092122395835,
      "learning_rate": 0.000572949016875158,
      "loss": 46.462,
      "step": 54
    },
    {
      "crossentropy": 2.234001398086548,
      "epoch": 6.107692307692307,
      "grad_norm": 1.5703125,
      "grad_norm_var": 0.013849894205729166,
      "learning_rate": 0.00047623940150645606,
      "loss": 44.68,
      "step": 55
    },
    {
      "crossentropy": 2.326664447784424,
      "epoch": 6.123076923076923,
      "grad_norm": 1.5625,
      "grad_norm_var": 0.010076649983723958,
      "learning_rate": 0.00038776176864066894,
      "loss": 46.5333,
      "step": 56
    },
    {
      "crossentropy": 2.2117629051208496,
      "epoch": 7.015384615384615,
      "grad_norm": 1.359375,
      "grad_norm_var": 0.011201985677083333,
      "learning_rate": 0.00030780471392592533,
      "loss": 44.2353,
      "step": 57
    },
    {
      "crossentropy": 2.2985191345214844,
      "epoch": 7.030769230769231,
      "grad_norm": 1.421875,
      "grad_norm_var": 0.0095458984375,
      "learning_rate": 0.00023662904065544998,
      "loss": 45.9704,
      "step": 58
    },
    {
      "crossentropy": 2.245023012161255,
      "epoch": 7.046153846153846,
      "grad_norm": 1.390625,
      "grad_norm_var": 0.006601715087890625,
      "learning_rate": 0.00017446690908142615,
      "loss": 44.9005,
      "step": 59
    },
    {
      "crossentropy": 2.174379825592041,
      "epoch": 7.061538461538461,
      "grad_norm": 1.390625,
      "grad_norm_var": 0.0076812744140625,
      "learning_rate": 0.00012152107915650823,
      "loss": 43.4876,
      "step": 60
    },
    {
      "crossentropy": 2.481731414794922,
      "epoch": 7.076923076923077,
      "grad_norm": 1.5078125,
      "grad_norm_var": 0.007043202718098958,
      "learning_rate": 7.79642491729996e-05,
      "loss": 49.6346,
      "step": 61
    },
    {
      "crossentropy": 2.240917444229126,
      "epoch": 7.092307692307692,
      "grad_norm": 1.375,
      "grad_norm_var": 0.006917317708333333,
      "learning_rate": 4.393849245692105e-05,
      "loss": 44.8183,
      "step": 62
    },
    {
      "crossentropy": 2.343332052230835,
      "epoch": 7.107692307692307,
      "grad_norm": 1.453125,
      "grad_norm_var": 0.005566151936848959,
      "learning_rate": 1.9554793954349337e-05,
      "loss": 46.8666,
      "step": 63
    },
    {
      "crossentropy": 2.241647958755493,
      "epoch": 7.123076923076923,
      "grad_norm": 1.40625,
      "grad_norm_var": 0.0059163411458333336,
      "learning_rate": 4.892688221593055e-06,
      "loss": 44.833,
      "step": 64
    },
    {
      "crossentropy": 2.247157335281372,
      "epoch": 8.015384615384615,
      "grad_norm": 1.3515625,
      "grad_norm_var": 0.006371815999348958,
      "learning_rate": 0.0,
      "loss": 44.9431,
      "step": 65
    }
  ],
  "logging_steps": 1,
  "max_steps": 65,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.359728188850176e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}