{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00125,
  "eval_steps": 50,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.5e-05,
      "grad_norm": 1030.0911865234375,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 343.4563,
      "loss/crossentropy": 2.979090690612793,
      "loss/hidden": 15.25,
      "loss/logits": 0.38270998001098633,
      "loss/reg": 343.0735778808594,
      "step": 1
    },
    {
      "epoch": 5e-05,
      "grad_norm": 11.19842529296875,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 343.3944,
      "loss/crossentropy": 2.8327953815460205,
      "loss/hidden": 15.25,
      "loss/logits": 0.3208012580871582,
      "loss/reg": 343.0735778808594,
      "step": 2
    },
    {
      "epoch": 7.5e-05,
      "grad_norm": 7.9109930992126465,
      "learning_rate": 3e-06,
      "loss": 343.4621,
      "loss/crossentropy": 3.085212469100952,
      "loss/hidden": 15.25,
      "loss/logits": 0.3886082172393799,
      "loss/reg": 343.0734558105469,
      "step": 3
    },
    {
      "epoch": 0.0001,
      "grad_norm": 278.76263427734375,
      "learning_rate": 4.000000000000001e-06,
      "loss": 343.3857,
      "loss/crossentropy": 2.7368664741516113,
      "loss/hidden": 15.25,
      "loss/logits": 0.31229129433631897,
      "loss/reg": 343.0733642578125,
      "step": 4
    },
    {
      "epoch": 0.000125,
      "grad_norm": 12.900250434875488,
      "learning_rate": 5e-06,
      "loss": 343.4269,
      "loss/crossentropy": 2.957352638244629,
      "loss/hidden": 15.25,
      "loss/logits": 0.35360077023506165,
      "loss/reg": 343.0732727050781,
      "step": 5
    },
    {
      "epoch": 0.00015,
      "grad_norm": 11.356637954711914,
      "learning_rate": 6e-06,
      "loss": 343.4677,
      "loss/crossentropy": 2.8502368927001953,
      "loss/hidden": 15.25,
      "loss/logits": 0.3944404125213623,
      "loss/reg": 343.07330322265625,
      "step": 6
    },
    {
      "epoch": 0.000175,
      "grad_norm": 8.149554252624512,
      "learning_rate": 7.000000000000001e-06,
      "loss": 343.3993,
      "loss/crossentropy": 2.816471815109253,
      "loss/hidden": 15.25,
      "loss/logits": 0.3256686329841614,
      "loss/reg": 343.0736083984375,
      "step": 7
    },
    {
      "epoch": 0.0002,
      "grad_norm": 6.3728837966918945,
      "learning_rate": 8.000000000000001e-06,
      "loss": 343.4041,
      "loss/crossentropy": 2.939116954803467,
      "loss/hidden": 15.25,
      "loss/logits": 0.33104804158210754,
      "loss/reg": 343.0730285644531,
      "step": 8
    },
    {
      "epoch": 0.000225,
      "grad_norm": 4.597896575927734,
      "learning_rate": 9e-06,
      "loss": 343.33,
      "loss/crossentropy": 2.5655953884124756,
      "loss/hidden": 15.25,
      "loss/logits": 0.257835328578949,
      "loss/reg": 343.0721435546875,
      "step": 9
    },
    {
      "epoch": 0.00025,
      "grad_norm": 8.375211715698242,
      "learning_rate": 1e-05,
      "loss": 343.3147,
      "loss/crossentropy": 2.869285821914673,
      "loss/hidden": 15.25,
      "loss/logits": 0.24429196119308472,
      "loss/reg": 343.0704345703125,
      "step": 10
    },
    {
      "epoch": 0.000275,
      "grad_norm": 3.975677728652954,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 343.2932,
      "loss/crossentropy": 2.670386552810669,
      "loss/hidden": 15.25,
      "loss/logits": 0.22499963641166687,
      "loss/reg": 343.0682373046875,
      "step": 11
    },
    {
      "epoch": 0.0003,
      "grad_norm": 5.542755126953125,
      "learning_rate": 1.2e-05,
      "loss": 343.3178,
      "loss/crossentropy": 3.0790939331054688,
      "loss/hidden": 15.25,
      "loss/logits": 0.25476354360580444,
      "loss/reg": 343.06304931640625,
      "step": 12
    },
    {
      "epoch": 0.000325,
      "grad_norm": 2.421396017074585,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 343.2363,
      "loss/crossentropy": 2.8270552158355713,
      "loss/hidden": 15.25,
      "loss/logits": 0.17909111082553864,
      "loss/reg": 343.05718994140625,
      "step": 13
    },
    {
      "epoch": 0.00035,
      "grad_norm": 2.906562328338623,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 343.1907,
      "loss/crossentropy": 2.4849202632904053,
      "loss/hidden": 15.25,
      "loss/logits": 0.14080765843391418,
      "loss/reg": 343.04986572265625,
      "step": 14
    },
    {
      "epoch": 0.000375,
      "grad_norm": 1.9625457525253296,
      "learning_rate": 1.5e-05,
      "loss": 343.2174,
      "loss/crossentropy": 3.037820339202881,
      "loss/hidden": 15.25,
      "loss/logits": 0.17379778623580933,
      "loss/reg": 343.0436096191406,
      "step": 15
    },
    {
      "epoch": 0.0004,
      "grad_norm": 1.572465181350708,
      "grad_norm_var": 67825.2066513546,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 343.1614,
      "loss/crossentropy": 2.9201037883758545,
      "loss/hidden": 15.25,
      "loss/logits": 0.12472350150346756,
      "loss/reg": 343.0367126464844,
      "step": 16
    },
    {
      "epoch": 0.000425,
      "grad_norm": 1.4828628301620483,
      "grad_norm_var": 4661.80263977645,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 343.142,
      "loss/crossentropy": 2.6999711990356445,
      "loss/hidden": 15.25,
      "loss/logits": 0.11364762485027313,
      "loss/reg": 343.0283508300781,
      "step": 17
    },
    {
      "epoch": 0.00045,
      "grad_norm": 1.8871707916259766,
      "grad_norm_var": 4681.988536067261,
      "learning_rate": 1.8e-05,
      "loss": 343.1356,
      "loss/crossentropy": 3.0806515216827393,
      "loss/hidden": 15.25,
      "loss/logits": 0.11845001578330994,
      "loss/reg": 343.01715087890625,
      "step": 18
    },
    {
      "epoch": 0.000475,
      "grad_norm": 1.2818211317062378,
      "grad_norm_var": 4697.640033071101,
      "learning_rate": 1.9e-05,
      "loss": 343.1078,
      "loss/crossentropy": 2.988394260406494,
      "loss/hidden": 15.25,
      "loss/logits": 0.09374003112316132,
      "loss/reg": 343.0140075683594,
      "step": 19
    },
    {
      "epoch": 0.0005,
      "grad_norm": 1.2304683923721313,
      "grad_norm_var": 13.90872762408596,
      "learning_rate": 2e-05,
      "loss": 343.1153,
      "loss/crossentropy": 2.9062788486480713,
      "loss/hidden": 15.25,
      "loss/logits": 0.09611324220895767,
      "loss/reg": 343.0191650390625,
      "step": 20
    },
    {
      "epoch": 0.000525,
      "grad_norm": 1.2078300714492798,
      "grad_norm_var": 9.74869168975174,
      "learning_rate": 2.1e-05,
      "loss": 343.1086,
      "loss/crossentropy": 2.629263162612915,
      "loss/hidden": 15.25,
      "loss/logits": 0.08695611357688904,
      "loss/reg": 343.0216369628906,
      "step": 21
    },
    {
      "epoch": 0.00055,
      "grad_norm": 0.9980544447898865,
      "grad_norm_var": 6.322323347833189,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 343.0953,
      "loss/crossentropy": 2.5109262466430664,
      "loss/hidden": 15.25,
      "loss/logits": 0.07269689440727234,
      "loss/reg": 343.0226135253906,
      "step": 22
    },
    {
      "epoch": 0.000575,
      "grad_norm": 1.1052573919296265,
      "grad_norm_var": 4.937209980551387,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 343.1042,
      "loss/crossentropy": 3.1142160892486572,
      "loss/hidden": 15.25,
      "loss/logits": 0.08075863122940063,
      "loss/reg": 343.02349853515625,
      "step": 23
    },
    {
      "epoch": 0.0006,
      "grad_norm": 0.9378647208213806,
      "grad_norm_var": 4.290323212534128,
      "learning_rate": 2.4e-05,
      "loss": 343.0883,
      "loss/crossentropy": 2.7667906284332275,
      "loss/hidden": 15.25,
      "loss/logits": 0.06898947060108185,
      "loss/reg": 343.0193176269531,
      "step": 24
    },
    {
      "epoch": 0.000625,
      "grad_norm": 0.8331053853034973,
      "grad_norm_var": 4.169706484634468,
      "learning_rate": 2.5e-05,
      "loss": 343.0793,
      "loss/crossentropy": 2.769016981124878,
      "loss/hidden": 15.25,
      "loss/logits": 0.06074293702840805,
      "loss/reg": 343.0185852050781,
      "step": 25
    },
    {
      "epoch": 0.00065,
      "grad_norm": 5.7551751136779785,
      "grad_norm_var": 2.4965497148631832,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 343.0914,
      "loss/crossentropy": 3.0031418800354004,
      "loss/hidden": 15.25,
      "loss/logits": 0.07152244448661804,
      "loss/reg": 343.0198669433594,
      "step": 26
    },
    {
      "epoch": 0.000675,
      "grad_norm": 0.832045316696167,
      "grad_norm_var": 2.3673310147404356,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 343.0815,
      "loss/crossentropy": 2.9811668395996094,
      "loss/hidden": 15.25,
      "loss/logits": 0.061069637537002563,
      "loss/reg": 343.0203857421875,
      "step": 27
    },
    {
      "epoch": 0.0007,
      "grad_norm": 0.8330180048942566,
      "grad_norm_var": 1.5272837131405466,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 343.0781,
      "loss/crossentropy": 2.740814447402954,
      "loss/hidden": 15.25,
      "loss/logits": 0.054962363094091415,
      "loss/reg": 343.0231018066406,
      "step": 28
    },
    {
      "epoch": 0.000725,
      "grad_norm": 0.7338961958885193,
      "grad_norm_var": 1.5436180822593486,
      "learning_rate": 2.9e-05,
      "loss": 343.0786,
      "loss/crossentropy": 2.854477643966675,
      "loss/hidden": 15.25,
      "loss/logits": 0.05266527086496353,
      "loss/reg": 343.02593994140625,
      "step": 29
    },
    {
      "epoch": 0.00075,
      "grad_norm": 1.040590524673462,
      "grad_norm_var": 1.4355461941067227,
      "learning_rate": 3e-05,
      "loss": 343.0964,
      "loss/crossentropy": 2.868793487548828,
      "loss/hidden": 15.25,
      "loss/logits": 0.058403968811035156,
      "loss/reg": 343.0379943847656,
      "step": 30
    },
    {
      "epoch": 0.000775,
      "grad_norm": 0.8477593064308167,
      "grad_norm_var": 1.4416249160873411,
      "learning_rate": 3.1e-05,
      "loss": 343.0889,
      "loss/crossentropy": 2.620675802230835,
      "loss/hidden": 15.25,
      "loss/logits": 0.05073639750480652,
      "loss/reg": 343.03814697265625,
      "step": 31
    },
    {
      "epoch": 0.0008,
      "grad_norm": 0.9066484570503235,
      "grad_norm_var": 1.455016528557988,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 343.0885,
      "loss/crossentropy": 2.8256523609161377,
      "loss/hidden": 15.25,
      "loss/logits": 0.04819165915250778,
      "loss/reg": 343.040283203125,
      "step": 32
    },
    {
      "epoch": 0.000825,
      "grad_norm": 1.3142300844192505,
      "grad_norm_var": 1.4542471534457053,
      "learning_rate": 3.3e-05,
      "loss": 343.1022,
      "loss/crossentropy": 3.081716299057007,
      "loss/hidden": 15.25,
      "loss/logits": 0.050295330584049225,
      "loss/reg": 343.0518798828125,
      "step": 33
    },
    {
      "epoch": 0.00085,
      "grad_norm": 1.2658751010894775,
      "grad_norm_var": 1.4346241521510184,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 343.1093,
      "loss/crossentropy": 3.0131924152374268,
      "loss/hidden": 15.25,
      "loss/logits": 0.05284588038921356,
      "loss/reg": 343.056396484375,
      "step": 34
    },
    {
      "epoch": 0.000875,
      "grad_norm": 0.9082849025726318,
      "grad_norm_var": 1.445257555860012,
      "learning_rate": 3.5e-05,
      "loss": 343.1118,
      "loss/crossentropy": 2.540893077850342,
      "loss/hidden": 15.25,
      "loss/logits": 0.0445842370390892,
      "loss/reg": 343.0672607421875,
      "step": 35
    },
    {
      "epoch": 0.0009,
      "grad_norm": 0.9978659749031067,
      "grad_norm_var": 1.4506987607639776,
      "learning_rate": 3.6e-05,
      "loss": 343.1154,
      "loss/crossentropy": 2.965841054916382,
      "loss/hidden": 15.25,
      "loss/logits": 0.04799606651067734,
      "loss/reg": 343.0673828125,
      "step": 36
    },
    {
      "epoch": 0.000925,
      "grad_norm": 1.0756022930145264,
      "grad_norm_var": 1.453105227206015,
      "learning_rate": 3.7e-05,
      "loss": 343.1204,
      "loss/crossentropy": 2.497769832611084,
      "loss/hidden": 15.25,
      "loss/logits": 0.04083085432648659,
      "loss/reg": 343.07952880859375,
      "step": 37
    },
    {
      "epoch": 0.00095,
      "grad_norm": 0.9938598871231079,
      "grad_norm_var": 1.4532607006280953,
      "learning_rate": 3.8e-05,
      "loss": 343.1224,
      "loss/crossentropy": 2.590686559677124,
      "loss/hidden": 15.25,
      "loss/logits": 0.03972693532705307,
      "loss/reg": 343.0826416015625,
      "step": 38
    },
    {
      "epoch": 0.000975,
      "grad_norm": 1.0691267251968384,
      "grad_norm_var": 1.4541543145040774,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 343.1341,
      "loss/crossentropy": 2.837007999420166,
      "loss/hidden": 15.25,
      "loss/logits": 0.05051286518573761,
      "loss/reg": 343.0836181640625,
      "step": 39
    },
    {
      "epoch": 0.001,
      "grad_norm": 0.8595760464668274,
      "grad_norm_var": 1.4580206512604335,
      "learning_rate": 4e-05,
      "loss": 343.1299,
      "loss/crossentropy": 2.654414176940918,
      "loss/hidden": 15.25,
      "loss/logits": 0.03836715966463089,
      "loss/reg": 343.0915832519531,
      "step": 40
    },
    {
      "epoch": 0.001025,
      "grad_norm": 1.2864758968353271,
      "grad_norm_var": 1.4446587103261934,
      "learning_rate": 4.1e-05,
      "loss": 343.1393,
      "loss/crossentropy": 3.039605140686035,
      "loss/hidden": 15.25,
      "loss/logits": 0.050123609602451324,
      "loss/reg": 343.0892028808594,
      "step": 41
    },
    {
      "epoch": 0.00105,
      "grad_norm": 0.8998011350631714,
      "grad_norm_var": 0.030633521906435077,
      "learning_rate": 4.2e-05,
      "loss": 343.139,
      "loss/crossentropy": 2.6901302337646484,
      "loss/hidden": 15.25,
      "loss/logits": 0.04389282315969467,
      "loss/reg": 343.0951232910156,
      "step": 42
    },
    {
      "epoch": 0.001075,
      "grad_norm": 1.2220802307128906,
      "grad_norm_var": 0.03184695650448734,
      "learning_rate": 4.3e-05,
      "loss": 343.1362,
      "loss/crossentropy": 2.9773449897766113,
      "loss/hidden": 15.25,
      "loss/logits": 0.04332903027534485,
      "loss/reg": 343.0928955078125,
      "step": 43
    },
    {
      "epoch": 0.0011,
      "grad_norm": 0.8919843435287476,
      "grad_norm_var": 0.030626277057042593,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 343.1251,
      "loss/crossentropy": 2.795628309249878,
      "loss/hidden": 15.25,
      "loss/logits": 0.038088709115982056,
      "loss/reg": 343.0870361328125,
      "step": 44
    },
    {
      "epoch": 0.001125,
      "grad_norm": 0.8417184352874756,
      "grad_norm_var": 0.027245464758761707,
      "learning_rate": 4.5e-05,
      "loss": 343.1212,
      "loss/crossentropy": 2.858243465423584,
      "loss/hidden": 15.25,
      "loss/logits": 0.03852926567196846,
      "loss/reg": 343.0826110839844,
      "step": 45
    },
    {
      "epoch": 0.00115,
      "grad_norm": 0.8550565838813782,
      "grad_norm_var": 0.029044425822037433,
      "learning_rate": 4.600000000000001e-05,
      "loss": 343.1173,
      "loss/crossentropy": 2.7999496459960938,
      "loss/hidden": 15.25,
      "loss/logits": 0.03844599053263664,
      "loss/reg": 343.078857421875,
      "step": 46
    },
    {
      "epoch": 0.001175,
      "grad_norm": 0.958902895450592,
      "grad_norm_var": 0.027341873139830008,
      "learning_rate": 4.7e-05,
      "loss": 343.1191,
      "loss/crossentropy": 2.7856900691986084,
      "loss/hidden": 15.25,
      "loss/logits": 0.0381263792514801,
      "loss/reg": 343.0809631347656,
      "step": 47
    },
    {
      "epoch": 0.0012,
      "grad_norm": 0.9236550331115723,
      "grad_norm_var": 0.02709908097331013,
      "learning_rate": 4.8e-05,
      "loss": 343.1286,
      "loss/crossentropy": 2.7613890171051025,
      "loss/hidden": 15.25,
      "loss/logits": 0.03775141388177872,
      "loss/reg": 343.09088134765625,
      "step": 48
    },
    {
      "epoch": 0.001225,
      "grad_norm": 0.94582200050354,
      "grad_norm_var": 0.021264340723863028,
      "learning_rate": 4.9e-05,
      "loss": 343.126,
      "loss/crossentropy": 2.9882125854492188,
      "loss/hidden": 15.25,
      "loss/logits": 0.03915192931890488,
      "loss/reg": 343.0868835449219,
      "step": 49
    },
    {
      "epoch": 0.00125,
      "grad_norm": 1.2259169816970825,
      "grad_norm_var": 0.01994617955941269,
      "learning_rate": 5e-05,
      "loss": 343.129,
      "loss/crossentropy": 2.7120375633239746,
      "loss/hidden": 15.25,
      "loss/logits": 0.04252094402909279,
      "loss/reg": 343.08648681640625,
      "step": 50
    }
  ],
  "logging_steps": 1,
  "max_steps": 40000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": true,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3227844083712000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}