{
  "best_metric": 0.5264609456062317,
  "best_model_checkpoint": "vit-weldclassifyv4/checkpoint-600",
  "epoch": 13.0,
  "eval_steps": 100,
  "global_step": 2028,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0641025641025641,
      "grad_norm": 1.0103411674499512,
      "learning_rate": 0.0001990138067061144,
      "loss": 1.2333,
      "step": 10
    },
    {
      "epoch": 0.1282051282051282,
      "grad_norm": 1.4450510740280151,
      "learning_rate": 0.0001980276134122288,
      "loss": 1.1639,
      "step": 20
    },
    {
      "epoch": 0.19230769230769232,
      "grad_norm": 1.538404107093811,
      "learning_rate": 0.0001970414201183432,
      "loss": 1.172,
      "step": 30
    },
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 1.6303586959838867,
      "learning_rate": 0.0001960552268244576,
      "loss": 1.2105,
      "step": 40
    },
    {
      "epoch": 0.32051282051282054,
      "grad_norm": 1.4056791067123413,
      "learning_rate": 0.000195069033530572,
      "loss": 1.1769,
      "step": 50
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 1.3668644428253174,
      "learning_rate": 0.0001940828402366864,
      "loss": 1.1274,
      "step": 60
    },
    {
      "epoch": 0.44871794871794873,
      "grad_norm": 1.1720328330993652,
      "learning_rate": 0.0001930966469428008,
      "loss": 1.17,
      "step": 70
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 1.5722938776016235,
      "learning_rate": 0.0001921104536489152,
      "loss": 1.216,
      "step": 80
    },
    {
      "epoch": 0.5769230769230769,
      "grad_norm": 2.300945281982422,
      "learning_rate": 0.0001911242603550296,
      "loss": 1.203,
      "step": 90
    },
    {
      "epoch": 0.6410256410256411,
      "grad_norm": 0.6618363261222839,
      "learning_rate": 0.00019013806706114398,
      "loss": 1.1126,
      "step": 100
    },
    {
      "epoch": 0.6410256410256411,
      "eval_accuracy": 0.5503597122302158,
      "eval_loss": 1.0170872211456299,
      "eval_runtime": 3.2391,
      "eval_samples_per_second": 85.825,
      "eval_steps_per_second": 10.805,
      "step": 100
    },
    {
      "epoch": 0.7051282051282052,
      "grad_norm": 1.7921690940856934,
      "learning_rate": 0.00018915187376725837,
      "loss": 1.045,
      "step": 110
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 1.696807861328125,
      "learning_rate": 0.00018816568047337278,
      "loss": 0.9879,
      "step": 120
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 1.4161618947982788,
      "learning_rate": 0.0001871794871794872,
      "loss": 0.966,
      "step": 130
    },
    {
      "epoch": 0.8974358974358975,
      "grad_norm": 5.280139923095703,
      "learning_rate": 0.0001861932938856016,
      "loss": 1.0863,
      "step": 140
    },
    {
      "epoch": 0.9615384615384616,
      "grad_norm": 1.3441425561904907,
      "learning_rate": 0.00018520710059171598,
      "loss": 1.0925,
      "step": 150
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 2.115222454071045,
      "learning_rate": 0.0001842209072978304,
      "loss": 0.886,
      "step": 160
    },
    {
      "epoch": 1.0897435897435896,
      "grad_norm": 1.5553538799285889,
      "learning_rate": 0.00018323471400394478,
      "loss": 0.8963,
      "step": 170
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 2.345830202102661,
      "learning_rate": 0.00018224852071005917,
      "loss": 0.717,
      "step": 180
    },
    {
      "epoch": 1.217948717948718,
      "grad_norm": 4.7483811378479,
      "learning_rate": 0.00018126232741617356,
      "loss": 0.8541,
      "step": 190
    },
    {
      "epoch": 1.282051282051282,
      "grad_norm": 2.3245091438293457,
      "learning_rate": 0.00018027613412228798,
      "loss": 0.8229,
      "step": 200
    },
    {
      "epoch": 1.282051282051282,
      "eval_accuracy": 0.6942446043165468,
      "eval_loss": 0.7307383418083191,
      "eval_runtime": 2.352,
      "eval_samples_per_second": 118.196,
      "eval_steps_per_second": 14.881,
      "step": 200
    },
    {
      "epoch": 1.3461538461538463,
      "grad_norm": 3.3630616664886475,
      "learning_rate": 0.0001792899408284024,
      "loss": 0.7849,
      "step": 210
    },
    {
      "epoch": 1.4102564102564101,
      "grad_norm": 1.2493098974227905,
      "learning_rate": 0.00017830374753451678,
      "loss": 0.8913,
      "step": 220
    },
    {
      "epoch": 1.4743589743589745,
      "grad_norm": 2.332315683364868,
      "learning_rate": 0.00017731755424063117,
      "loss": 0.7431,
      "step": 230
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 1.8999154567718506,
      "learning_rate": 0.00017633136094674556,
      "loss": 0.9167,
      "step": 240
    },
    {
      "epoch": 1.6025641025641026,
      "grad_norm": 3.0892856121063232,
      "learning_rate": 0.00017534516765285997,
      "loss": 0.6975,
      "step": 250
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 1.9793812036514282,
      "learning_rate": 0.00017435897435897436,
      "loss": 0.5964,
      "step": 260
    },
    {
      "epoch": 1.7307692307692308,
      "grad_norm": 3.3237643241882324,
      "learning_rate": 0.00017337278106508875,
      "loss": 0.6702,
      "step": 270
    },
    {
      "epoch": 1.7948717948717947,
      "grad_norm": 7.974592208862305,
      "learning_rate": 0.00017238658777120317,
      "loss": 0.5598,
      "step": 280
    },
    {
      "epoch": 1.858974358974359,
      "grad_norm": 4.481497764587402,
      "learning_rate": 0.00017140039447731758,
      "loss": 0.8077,
      "step": 290
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 1.5490068197250366,
      "learning_rate": 0.00017041420118343197,
      "loss": 0.7224,
      "step": 300
    },
    {
      "epoch": 1.9230769230769231,
      "eval_accuracy": 0.7122302158273381,
      "eval_loss": 0.6399481296539307,
      "eval_runtime": 2.6523,
      "eval_samples_per_second": 104.815,
      "eval_steps_per_second": 13.196,
      "step": 300
    },
    {
      "epoch": 1.9871794871794872,
      "grad_norm": 2.933385133743286,
      "learning_rate": 0.00016942800788954636,
      "loss": 0.6504,
      "step": 310
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 1.7754273414611816,
      "learning_rate": 0.00016844181459566075,
      "loss": 0.6169,
      "step": 320
    },
    {
      "epoch": 2.1153846153846154,
      "grad_norm": 4.75892972946167,
      "learning_rate": 0.00016745562130177514,
      "loss": 0.5039,
      "step": 330
    },
    {
      "epoch": 2.1794871794871793,
      "grad_norm": 9.711180686950684,
      "learning_rate": 0.00016646942800788956,
      "loss": 0.7187,
      "step": 340
    },
    {
      "epoch": 2.2435897435897436,
      "grad_norm": 1.4888873100280762,
      "learning_rate": 0.00016548323471400394,
      "loss": 0.5705,
      "step": 350
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.9269102215766907,
      "learning_rate": 0.00016449704142011836,
      "loss": 0.3177,
      "step": 360
    },
    {
      "epoch": 2.371794871794872,
      "grad_norm": 3.9913885593414307,
      "learning_rate": 0.00016351084812623275,
      "loss": 0.5014,
      "step": 370
    },
    {
      "epoch": 2.435897435897436,
      "grad_norm": 4.140696048736572,
      "learning_rate": 0.00016252465483234716,
      "loss": 0.6508,
      "step": 380
    },
    {
      "epoch": 2.5,
      "grad_norm": 2.751957654953003,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.4819,
      "step": 390
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 4.626535415649414,
      "learning_rate": 0.00016055226824457594,
      "loss": 0.3909,
      "step": 400
    },
    {
      "epoch": 2.564102564102564,
      "eval_accuracy": 0.7733812949640287,
      "eval_loss": 0.5400061011314392,
      "eval_runtime": 2.4236,
      "eval_samples_per_second": 114.704,
      "eval_steps_per_second": 14.441,
      "step": 400
    },
    {
      "epoch": 2.628205128205128,
      "grad_norm": 0.838819682598114,
      "learning_rate": 0.00015956607495069033,
      "loss": 0.3677,
      "step": 410
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 12.247054100036621,
      "learning_rate": 0.00015857988165680475,
      "loss": 0.5027,
      "step": 420
    },
    {
      "epoch": 2.7564102564102564,
      "grad_norm": 3.5522589683532715,
      "learning_rate": 0.00015759368836291914,
      "loss": 0.4813,
      "step": 430
    },
    {
      "epoch": 2.8205128205128203,
      "grad_norm": 1.3977766036987305,
      "learning_rate": 0.00015660749506903355,
      "loss": 0.3228,
      "step": 440
    },
    {
      "epoch": 2.8846153846153846,
      "grad_norm": 3.1070363521575928,
      "learning_rate": 0.00015562130177514794,
      "loss": 0.4175,
      "step": 450
    },
    {
      "epoch": 2.948717948717949,
      "grad_norm": 3.402083396911621,
      "learning_rate": 0.00015463510848126233,
      "loss": 0.5283,
      "step": 460
    },
    {
      "epoch": 3.0128205128205128,
      "grad_norm": 2.60976505279541,
      "learning_rate": 0.00015364891518737675,
      "loss": 0.4912,
      "step": 470
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 3.5103209018707275,
      "learning_rate": 0.00015266272189349113,
      "loss": 0.2406,
      "step": 480
    },
    {
      "epoch": 3.141025641025641,
      "grad_norm": 2.8553466796875,
      "learning_rate": 0.00015167652859960552,
      "loss": 0.2399,
      "step": 490
    },
    {
      "epoch": 3.2051282051282053,
      "grad_norm": 3.4007368087768555,
      "learning_rate": 0.0001506903353057199,
      "loss": 0.237,
      "step": 500
    },
    {
      "epoch": 3.2051282051282053,
      "eval_accuracy": 0.762589928057554,
      "eval_loss": 0.671575665473938,
      "eval_runtime": 2.3586,
      "eval_samples_per_second": 117.864,
      "eval_steps_per_second": 14.839,
      "step": 500
    },
    {
      "epoch": 3.269230769230769,
      "grad_norm": 7.140501022338867,
      "learning_rate": 0.00014970414201183433,
      "loss": 0.3362,
      "step": 510
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 7.302017688751221,
      "learning_rate": 0.00014871794871794872,
      "loss": 0.5846,
      "step": 520
    },
    {
      "epoch": 3.3974358974358974,
      "grad_norm": 3.939833402633667,
      "learning_rate": 0.00014773175542406313,
      "loss": 0.6107,
      "step": 530
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 5.233958721160889,
      "learning_rate": 0.00014674556213017752,
      "loss": 0.3263,
      "step": 540
    },
    {
      "epoch": 3.5256410256410255,
      "grad_norm": 3.2795095443725586,
      "learning_rate": 0.0001457593688362919,
      "loss": 0.433,
      "step": 550
    },
    {
      "epoch": 3.58974358974359,
      "grad_norm": 6.401491165161133,
      "learning_rate": 0.00014477317554240633,
      "loss": 0.3176,
      "step": 560
    },
    {
      "epoch": 3.6538461538461537,
      "grad_norm": 2.198981285095215,
      "learning_rate": 0.00014378698224852072,
      "loss": 0.2172,
      "step": 570
    },
    {
      "epoch": 3.717948717948718,
      "grad_norm": 6.659445762634277,
      "learning_rate": 0.0001428007889546351,
      "loss": 0.3023,
      "step": 580
    },
    {
      "epoch": 3.782051282051282,
      "grad_norm": 11.928166389465332,
      "learning_rate": 0.0001418145956607495,
      "loss": 0.4809,
      "step": 590
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 2.920708179473877,
      "learning_rate": 0.0001408284023668639,
      "loss": 0.4056,
      "step": 600
    },
    {
      "epoch": 3.8461538461538463,
      "eval_accuracy": 0.8093525179856115,
      "eval_loss": 0.5264609456062317,
      "eval_runtime": 3.1039,
      "eval_samples_per_second": 89.563,
      "eval_steps_per_second": 11.276,
      "step": 600
    },
    {
      "epoch": 3.91025641025641,
      "grad_norm": 6.828011512756348,
      "learning_rate": 0.00013984220907297832,
      "loss": 0.241,
      "step": 610
    },
    {
      "epoch": 3.9743589743589745,
      "grad_norm": 2.935359239578247,
      "learning_rate": 0.0001388560157790927,
      "loss": 0.1965,
      "step": 620
    },
    {
      "epoch": 4.038461538461538,
      "grad_norm": 1.4090049266815186,
      "learning_rate": 0.0001378698224852071,
      "loss": 0.0932,
      "step": 630
    },
    {
      "epoch": 4.102564102564102,
      "grad_norm": 3.070310354232788,
      "learning_rate": 0.00013688362919132152,
      "loss": 0.1354,
      "step": 640
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 2.6170833110809326,
      "learning_rate": 0.0001358974358974359,
      "loss": 0.1404,
      "step": 650
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 4.073474407196045,
      "learning_rate": 0.0001349112426035503,
      "loss": 0.134,
      "step": 660
    },
    {
      "epoch": 4.294871794871795,
      "grad_norm": 0.563457190990448,
      "learning_rate": 0.00013392504930966468,
      "loss": 0.1289,
      "step": 670
    },
    {
      "epoch": 4.358974358974359,
      "grad_norm": 1.5793825387954712,
      "learning_rate": 0.0001329388560157791,
      "loss": 0.1831,
      "step": 680
    },
    {
      "epoch": 4.423076923076923,
      "grad_norm": 4.176783084869385,
      "learning_rate": 0.00013195266272189352,
      "loss": 0.2321,
      "step": 690
    },
    {
      "epoch": 4.487179487179487,
      "grad_norm": 2.8989105224609375,
      "learning_rate": 0.0001309664694280079,
      "loss": 0.1764,
      "step": 700
    },
    {
      "epoch": 4.487179487179487,
      "eval_accuracy": 0.7446043165467626,
      "eval_loss": 0.9174473285675049,
      "eval_runtime": 3.0691,
      "eval_samples_per_second": 90.581,
      "eval_steps_per_second": 11.404,
      "step": 700
    },
    {
      "epoch": 4.551282051282051,
      "grad_norm": 5.322137355804443,
      "learning_rate": 0.0001299802761341223,
      "loss": 0.213,
      "step": 710
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 1.017148733139038,
      "learning_rate": 0.00012899408284023668,
      "loss": 0.1094,
      "step": 720
    },
    {
      "epoch": 4.67948717948718,
      "grad_norm": 3.9339067935943604,
      "learning_rate": 0.0001280078895463511,
      "loss": 0.1797,
      "step": 730
    },
    {
      "epoch": 4.743589743589744,
      "grad_norm": 2.634552001953125,
      "learning_rate": 0.0001270216962524655,
      "loss": 0.3253,
      "step": 740
    },
    {
      "epoch": 4.8076923076923075,
      "grad_norm": 6.850883960723877,
      "learning_rate": 0.00012603550295857988,
      "loss": 0.2161,
      "step": 750
    },
    {
      "epoch": 4.871794871794872,
      "grad_norm": 3.087040662765503,
      "learning_rate": 0.0001250493096646943,
      "loss": 0.2036,
      "step": 760
    },
    {
      "epoch": 4.935897435897436,
      "grad_norm": 0.3574156165122986,
      "learning_rate": 0.00012406311637080868,
      "loss": 0.2068,
      "step": 770
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.914322853088379,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.1289,
      "step": 780
    },
    {
      "epoch": 5.064102564102564,
      "grad_norm": 0.17900536954402924,
      "learning_rate": 0.00012209072978303749,
      "loss": 0.0544,
      "step": 790
    },
    {
      "epoch": 5.128205128205128,
      "grad_norm": 4.552331447601318,
      "learning_rate": 0.00012110453648915188,
      "loss": 0.0546,
      "step": 800
    },
    {
      "epoch": 5.128205128205128,
      "eval_accuracy": 0.8237410071942446,
      "eval_loss": 0.6643829941749573,
      "eval_runtime": 2.4217,
      "eval_samples_per_second": 114.798,
      "eval_steps_per_second": 14.453,
      "step": 800
    },
    {
      "epoch": 5.1923076923076925,
      "grad_norm": 0.17875537276268005,
      "learning_rate": 0.00012011834319526626,
      "loss": 0.1023,
      "step": 810
    },
    {
      "epoch": 5.256410256410256,
      "grad_norm": 4.822021007537842,
      "learning_rate": 0.00011913214990138068,
      "loss": 0.0419,
      "step": 820
    },
    {
      "epoch": 5.32051282051282,
      "grad_norm": 0.06140388920903206,
      "learning_rate": 0.00011814595660749508,
      "loss": 0.0293,
      "step": 830
    },
    {
      "epoch": 5.384615384615385,
      "grad_norm": 0.11838454753160477,
      "learning_rate": 0.00011715976331360947,
      "loss": 0.0333,
      "step": 840
    },
    {
      "epoch": 5.448717948717949,
      "grad_norm": 0.14812995493412018,
      "learning_rate": 0.00011617357001972386,
      "loss": 0.1141,
      "step": 850
    },
    {
      "epoch": 5.512820512820513,
      "grad_norm": 1.0732386112213135,
      "learning_rate": 0.00011518737672583828,
      "loss": 0.0639,
      "step": 860
    },
    {
      "epoch": 5.576923076923077,
      "grad_norm": 0.18576616048812866,
      "learning_rate": 0.00011420118343195268,
      "loss": 0.1059,
      "step": 870
    },
    {
      "epoch": 5.641025641025641,
      "grad_norm": 26.628576278686523,
      "learning_rate": 0.00011321499013806707,
      "loss": 0.2643,
      "step": 880
    },
    {
      "epoch": 5.705128205128205,
      "grad_norm": 0.08607737720012665,
      "learning_rate": 0.00011222879684418146,
      "loss": 0.0427,
      "step": 890
    },
    {
      "epoch": 5.769230769230769,
      "grad_norm": 10.939754486083984,
      "learning_rate": 0.00011124260355029586,
      "loss": 0.0436,
      "step": 900
    },
    {
      "epoch": 5.769230769230769,
      "eval_accuracy": 0.8345323741007195,
      "eval_loss": 0.692258894443512,
      "eval_runtime": 2.6974,
      "eval_samples_per_second": 103.061,
      "eval_steps_per_second": 12.975,
      "step": 900
    },
    {
      "epoch": 5.833333333333333,
      "grad_norm": 4.4804582595825195,
      "learning_rate": 0.00011025641025641027,
      "loss": 0.0769,
      "step": 910
    },
    {
      "epoch": 5.897435897435898,
      "grad_norm": 3.8326826095581055,
      "learning_rate": 0.00010927021696252466,
      "loss": 0.0708,
      "step": 920
    },
    {
      "epoch": 5.961538461538462,
      "grad_norm": 9.658041000366211,
      "learning_rate": 0.00010828402366863905,
      "loss": 0.0917,
      "step": 930
    },
    {
      "epoch": 6.0256410256410255,
      "grad_norm": 0.18607991933822632,
      "learning_rate": 0.00010729783037475345,
      "loss": 0.0775,
      "step": 940
    },
    {
      "epoch": 6.089743589743589,
      "grad_norm": 11.05611515045166,
      "learning_rate": 0.00010631163708086787,
      "loss": 0.0554,
      "step": 950
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 0.05502941831946373,
      "learning_rate": 0.00010532544378698226,
      "loss": 0.0337,
      "step": 960
    },
    {
      "epoch": 6.217948717948718,
      "grad_norm": 11.613199234008789,
      "learning_rate": 0.00010433925049309665,
      "loss": 0.0456,
      "step": 970
    },
    {
      "epoch": 6.282051282051282,
      "grad_norm": 0.048365768045186996,
      "learning_rate": 0.00010335305719921105,
      "loss": 0.0144,
      "step": 980
    },
    {
      "epoch": 6.346153846153846,
      "grad_norm": 3.896092176437378,
      "learning_rate": 0.00010236686390532544,
      "loss": 0.0808,
      "step": 990
    },
    {
      "epoch": 6.410256410256411,
      "grad_norm": 0.048218678683042526,
      "learning_rate": 0.00010138067061143986,
      "loss": 0.0661,
      "step": 1000
    },
    {
      "epoch": 6.410256410256411,
      "eval_accuracy": 0.8345323741007195,
      "eval_loss": 0.6783987283706665,
      "eval_runtime": 2.2705,
      "eval_samples_per_second": 122.441,
      "eval_steps_per_second": 15.415,
      "step": 1000
    },
    {
      "epoch": 6.4743589743589745,
      "grad_norm": 0.04647431522607803,
      "learning_rate": 0.00010039447731755424,
      "loss": 0.064,
      "step": 1010
    },
    {
      "epoch": 6.538461538461538,
      "grad_norm": 0.03940006345510483,
      "learning_rate": 9.940828402366865e-05,
      "loss": 0.0488,
      "step": 1020
    },
    {
      "epoch": 6.602564102564102,
      "grad_norm": 0.035748932510614395,
      "learning_rate": 9.842209072978305e-05,
      "loss": 0.0411,
      "step": 1030
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.08083514869213104,
      "learning_rate": 9.743589743589744e-05,
      "loss": 0.0088,
      "step": 1040
    },
    {
      "epoch": 6.730769230769231,
      "grad_norm": 0.035775866359472275,
      "learning_rate": 9.644970414201184e-05,
      "loss": 0.0394,
      "step": 1050
    },
    {
      "epoch": 6.794871794871795,
      "grad_norm": 0.09868976473808289,
      "learning_rate": 9.546351084812624e-05,
      "loss": 0.0522,
      "step": 1060
    },
    {
      "epoch": 6.858974358974359,
      "grad_norm": 0.08696691691875458,
      "learning_rate": 9.447731755424064e-05,
      "loss": 0.0567,
      "step": 1070
    },
    {
      "epoch": 6.923076923076923,
      "grad_norm": 9.270120620727539,
      "learning_rate": 9.349112426035503e-05,
      "loss": 0.054,
      "step": 1080
    },
    {
      "epoch": 6.987179487179487,
      "grad_norm": 2.301098346710205,
      "learning_rate": 9.250493096646942e-05,
      "loss": 0.0789,
      "step": 1090
    },
    {
      "epoch": 7.051282051282051,
      "grad_norm": 7.368527412414551,
      "learning_rate": 9.151873767258384e-05,
      "loss": 0.0167,
      "step": 1100
    },
    {
      "epoch": 7.051282051282051,
      "eval_accuracy": 0.8309352517985612,
      "eval_loss": 0.7114906311035156,
      "eval_runtime": 2.3567,
      "eval_samples_per_second": 117.96,
      "eval_steps_per_second": 14.851,
      "step": 1100
    },
    {
      "epoch": 7.115384615384615,
      "grad_norm": 0.051367539912462234,
      "learning_rate": 9.053254437869823e-05,
      "loss": 0.0082,
      "step": 1110
    },
    {
      "epoch": 7.17948717948718,
      "grad_norm": 0.03877986595034599,
      "learning_rate": 8.954635108481263e-05,
      "loss": 0.0119,
      "step": 1120
    },
    {
      "epoch": 7.243589743589744,
      "grad_norm": 0.030814744532108307,
      "learning_rate": 8.856015779092702e-05,
      "loss": 0.0561,
      "step": 1130
    },
    {
      "epoch": 7.3076923076923075,
      "grad_norm": 0.03935234248638153,
      "learning_rate": 8.757396449704143e-05,
      "loss": 0.046,
      "step": 1140
    },
    {
      "epoch": 7.371794871794872,
      "grad_norm": 1.2296382188796997,
      "learning_rate": 8.658777120315582e-05,
      "loss": 0.0638,
      "step": 1150
    },
    {
      "epoch": 7.435897435897436,
      "grad_norm": 0.048670027405023575,
      "learning_rate": 8.560157790927023e-05,
      "loss": 0.0429,
      "step": 1160
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.046260058879852295,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0091,
      "step": 1170
    },
    {
      "epoch": 7.564102564102564,
      "grad_norm": 0.17076271772384644,
      "learning_rate": 8.362919132149902e-05,
      "loss": 0.0432,
      "step": 1180
    },
    {
      "epoch": 7.628205128205128,
      "grad_norm": 0.3206053376197815,
      "learning_rate": 8.264299802761342e-05,
      "loss": 0.0542,
      "step": 1190
    },
    {
      "epoch": 7.6923076923076925,
      "grad_norm": 0.029316432774066925,
      "learning_rate": 8.165680473372781e-05,
      "loss": 0.0744,
      "step": 1200
    },
    {
      "epoch": 7.6923076923076925,
      "eval_accuracy": 0.8525179856115108,
      "eval_loss": 0.6341452598571777,
      "eval_runtime": 2.3839,
      "eval_samples_per_second": 116.615,
      "eval_steps_per_second": 14.682,
      "step": 1200
    },
    {
      "epoch": 7.756410256410256,
      "grad_norm": 0.08320651203393936,
      "learning_rate": 8.067061143984221e-05,
      "loss": 0.0514,
      "step": 1210
    },
    {
      "epoch": 7.82051282051282,
      "grad_norm": 0.02455221116542816,
      "learning_rate": 7.968441814595661e-05,
      "loss": 0.0054,
      "step": 1220
    },
    {
      "epoch": 7.884615384615385,
      "grad_norm": 5.290210247039795,
      "learning_rate": 7.869822485207101e-05,
      "loss": 0.0282,
      "step": 1230
    },
    {
      "epoch": 7.948717948717949,
      "grad_norm": 0.09167015552520752,
      "learning_rate": 7.77120315581854e-05,
      "loss": 0.0057,
      "step": 1240
    },
    {
      "epoch": 8.012820512820513,
      "grad_norm": 0.40623167157173157,
      "learning_rate": 7.67258382642998e-05,
      "loss": 0.0186,
      "step": 1250
    },
    {
      "epoch": 8.076923076923077,
      "grad_norm": 0.032823193818330765,
      "learning_rate": 7.573964497041421e-05,
      "loss": 0.0192,
      "step": 1260
    },
    {
      "epoch": 8.14102564102564,
      "grad_norm": 0.06196033954620361,
      "learning_rate": 7.475345167652861e-05,
      "loss": 0.005,
      "step": 1270
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 0.02220989391207695,
      "learning_rate": 7.3767258382643e-05,
      "loss": 0.0046,
      "step": 1280
    },
    {
      "epoch": 8.26923076923077,
      "grad_norm": 0.0361061617732048,
      "learning_rate": 7.27810650887574e-05,
      "loss": 0.0044,
      "step": 1290
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 0.02312193624675274,
      "learning_rate": 7.17948717948718e-05,
      "loss": 0.0047,
      "step": 1300
    },
    {
      "epoch": 8.333333333333334,
      "eval_accuracy": 0.8597122302158273,
      "eval_loss": 0.6401567459106445,
      "eval_runtime": 2.2954,
      "eval_samples_per_second": 121.11,
      "eval_steps_per_second": 15.248,
      "step": 1300
    },
    {
      "epoch": 8.397435897435898,
      "grad_norm": 0.020257720723748207,
      "learning_rate": 7.08086785009862e-05,
      "loss": 0.0045,
      "step": 1310
    },
    {
      "epoch": 8.461538461538462,
      "grad_norm": 0.021647421643137932,
      "learning_rate": 6.98224852071006e-05,
      "loss": 0.0042,
      "step": 1320
    },
    {
      "epoch": 8.525641025641026,
      "grad_norm": 0.01725107803940773,
      "learning_rate": 6.883629191321498e-05,
      "loss": 0.0041,
      "step": 1330
    },
    {
      "epoch": 8.58974358974359,
      "grad_norm": 0.06263954192399979,
      "learning_rate": 6.78500986193294e-05,
      "loss": 0.0225,
      "step": 1340
    },
    {
      "epoch": 8.653846153846153,
      "grad_norm": 0.02324153482913971,
      "learning_rate": 6.686390532544379e-05,
      "loss": 0.0042,
      "step": 1350
    },
    {
      "epoch": 8.717948717948717,
      "grad_norm": 0.023914692923426628,
      "learning_rate": 6.587771203155819e-05,
      "loss": 0.0039,
      "step": 1360
    },
    {
      "epoch": 8.782051282051283,
      "grad_norm": 0.016754157841205597,
      "learning_rate": 6.489151873767258e-05,
      "loss": 0.0041,
      "step": 1370
    },
    {
      "epoch": 8.846153846153847,
      "grad_norm": 0.1214941069483757,
      "learning_rate": 6.390532544378698e-05,
      "loss": 0.0039,
      "step": 1380
    },
    {
      "epoch": 8.91025641025641,
      "grad_norm": 0.016471996903419495,
      "learning_rate": 6.291913214990139e-05,
      "loss": 0.0035,
      "step": 1390
    },
    {
      "epoch": 8.974358974358974,
      "grad_norm": 0.03737170621752739,
      "learning_rate": 6.193293885601579e-05,
      "loss": 0.0039,
      "step": 1400
    },
    {
      "epoch": 8.974358974358974,
      "eval_accuracy": 0.8848920863309353,
      "eval_loss": 0.595779538154602,
      "eval_runtime": 2.3162,
      "eval_samples_per_second": 120.023,
      "eval_steps_per_second": 15.111,
      "step": 1400
    },
    {
      "epoch": 9.038461538461538,
      "grad_norm": 0.017416946589946747,
      "learning_rate": 6.094674556213018e-05,
      "loss": 0.0035,
      "step": 1410
    },
    {
      "epoch": 9.102564102564102,
      "grad_norm": 0.01665581203997135,
      "learning_rate": 5.996055226824457e-05,
      "loss": 0.0034,
      "step": 1420
    },
    {
      "epoch": 9.166666666666666,
      "grad_norm": 0.016857469454407692,
      "learning_rate": 5.897435897435898e-05,
      "loss": 0.0034,
      "step": 1430
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 0.016117751598358154,
      "learning_rate": 5.798816568047337e-05,
      "loss": 0.0036,
      "step": 1440
    },
    {
      "epoch": 9.294871794871796,
      "grad_norm": 0.016485603526234627,
      "learning_rate": 5.700197238658778e-05,
      "loss": 0.0034,
      "step": 1450
    },
    {
      "epoch": 9.35897435897436,
      "grad_norm": 0.01639939658343792,
      "learning_rate": 5.601577909270217e-05,
      "loss": 0.0031,
      "step": 1460
    },
    {
      "epoch": 9.423076923076923,
      "grad_norm": 0.014921830035746098,
      "learning_rate": 5.502958579881658e-05,
      "loss": 0.0033,
      "step": 1470
    },
    {
      "epoch": 9.487179487179487,
      "grad_norm": 0.014172026887536049,
      "learning_rate": 5.4043392504930966e-05,
      "loss": 0.0029,
      "step": 1480
    },
    {
      "epoch": 9.551282051282051,
      "grad_norm": 0.014830219559371471,
      "learning_rate": 5.3057199211045375e-05,
      "loss": 0.0031,
      "step": 1490
    },
    {
      "epoch": 9.615384615384615,
      "grad_norm": 0.014610445126891136,
      "learning_rate": 5.2071005917159764e-05,
      "loss": 0.0029,
      "step": 1500
    },
    {
      "epoch": 9.615384615384615,
      "eval_accuracy": 0.8884892086330936,
      "eval_loss": 0.6158227920532227,
      "eval_runtime": 2.6105,
      "eval_samples_per_second": 106.492,
      "eval_steps_per_second": 13.407,
      "step": 1500
    },
    {
      "epoch": 9.679487179487179,
      "grad_norm": 0.015689866617321968,
      "learning_rate": 5.1084812623274167e-05,
      "loss": 0.003,
      "step": 1510
    },
    {
      "epoch": 9.743589743589745,
      "grad_norm": 0.014888488687574863,
      "learning_rate": 5.009861932938856e-05,
      "loss": 0.003,
      "step": 1520
    },
    {
      "epoch": 9.807692307692308,
      "grad_norm": 0.014181181788444519,
      "learning_rate": 4.9112426035502965e-05,
      "loss": 0.0029,
      "step": 1530
    },
    {
      "epoch": 9.871794871794872,
      "grad_norm": 0.01550784707069397,
      "learning_rate": 4.812623274161736e-05,
      "loss": 0.003,
      "step": 1540
    },
    {
      "epoch": 9.935897435897436,
      "grad_norm": 0.012861398048698902,
      "learning_rate": 4.714003944773176e-05,
      "loss": 0.0029,
      "step": 1550
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.01375816110521555,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0028,
      "step": 1560
    },
    {
      "epoch": 10.064102564102564,
      "grad_norm": 0.014541467651724815,
      "learning_rate": 4.5167652859960554e-05,
      "loss": 0.0029,
      "step": 1570
    },
    {
      "epoch": 10.128205128205128,
      "grad_norm": 0.01512611098587513,
      "learning_rate": 4.418145956607495e-05,
      "loss": 0.0028,
      "step": 1580
    },
    {
      "epoch": 10.192307692307692,
      "grad_norm": 0.012550946325063705,
      "learning_rate": 4.319526627218935e-05,
      "loss": 0.0026,
      "step": 1590
    },
    {
      "epoch": 10.256410256410255,
      "grad_norm": 0.013985677622258663,
      "learning_rate": 4.220907297830375e-05,
      "loss": 0.0027,
      "step": 1600
    },
    {
      "epoch": 10.256410256410255,
      "eval_accuracy": 0.8884892086330936,
      "eval_loss": 0.6188732981681824,
      "eval_runtime": 3.1938,
      "eval_samples_per_second": 87.044,
      "eval_steps_per_second": 10.959,
      "step": 1600
    },
    {
      "epoch": 10.320512820512821,
      "grad_norm": 0.01235656812787056,
      "learning_rate": 4.122287968441815e-05,
      "loss": 0.0027,
      "step": 1610
    },
    {
      "epoch": 10.384615384615385,
      "grad_norm": 0.012514271773397923,
      "learning_rate": 4.0236686390532545e-05,
      "loss": 0.0026,
      "step": 1620
    },
    {
      "epoch": 10.448717948717949,
      "grad_norm": 0.012498818337917328,
      "learning_rate": 3.925049309664695e-05,
      "loss": 0.0027,
      "step": 1630
    },
    {
      "epoch": 10.512820512820513,
      "grad_norm": 0.015231632627546787,
      "learning_rate": 3.826429980276134e-05,
      "loss": 0.0027,
      "step": 1640
    },
    {
      "epoch": 10.576923076923077,
      "grad_norm": 0.012208458036184311,
      "learning_rate": 3.7278106508875746e-05,
      "loss": 0.0025,
      "step": 1650
    },
    {
      "epoch": 10.64102564102564,
      "grad_norm": 0.012389231473207474,
      "learning_rate": 3.629191321499014e-05,
      "loss": 0.0026,
      "step": 1660
    },
    {
      "epoch": 10.705128205128204,
      "grad_norm": 0.014937272295355797,
      "learning_rate": 3.5305719921104544e-05,
      "loss": 0.0027,
      "step": 1670
    },
    {
      "epoch": 10.76923076923077,
      "grad_norm": 0.013052302412688732,
      "learning_rate": 3.431952662721893e-05,
      "loss": 0.0025,
      "step": 1680
    },
    {
      "epoch": 10.833333333333334,
      "grad_norm": 0.013081245124340057,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0024,
      "step": 1690
    },
    {
      "epoch": 10.897435897435898,
      "grad_norm": 0.011870602145791054,
      "learning_rate": 3.234714003944773e-05,
      "loss": 0.0025,
      "step": 1700
    },
    {
      "epoch": 10.897435897435898,
      "eval_accuracy": 0.8884892086330936,
      "eval_loss": 0.630884051322937,
      "eval_runtime": 2.4561,
      "eval_samples_per_second": 113.19,
      "eval_steps_per_second": 14.251,
      "step": 1700
    },
    {
      "epoch": 10.961538461538462,
      "grad_norm": 0.015387916006147861,
      "learning_rate": 3.136094674556213e-05,
      "loss": 0.0025,
      "step": 1710
    },
    {
      "epoch": 11.025641025641026,
      "grad_norm": 0.012120082043111324,
      "learning_rate": 3.037475345167653e-05,
      "loss": 0.0027,
      "step": 1720
    },
    {
      "epoch": 11.08974358974359,
      "grad_norm": 0.014247632585465908,
      "learning_rate": 2.9388560157790928e-05,
      "loss": 0.0026,
      "step": 1730
    },
    {
      "epoch": 11.153846153846153,
      "grad_norm": 0.011767740361392498,
      "learning_rate": 2.8402366863905327e-05,
      "loss": 0.0023,
      "step": 1740
    },
    {
      "epoch": 11.217948717948717,
      "grad_norm": 0.013148034922778606,
      "learning_rate": 2.7416173570019726e-05,
      "loss": 0.0024,
      "step": 1750
    },
    {
      "epoch": 11.282051282051283,
      "grad_norm": 0.013173798099160194,
      "learning_rate": 2.6429980276134125e-05,
      "loss": 0.0025,
      "step": 1760
    },
    {
      "epoch": 11.346153846153847,
      "grad_norm": 0.011782374233007431,
      "learning_rate": 2.5443786982248524e-05,
      "loss": 0.0024,
      "step": 1770
    },
    {
      "epoch": 11.41025641025641,
      "grad_norm": 0.013157065957784653,
      "learning_rate": 2.445759368836292e-05,
      "loss": 0.0024,
      "step": 1780
    },
    {
      "epoch": 11.474358974358974,
      "grad_norm": 0.011065174825489521,
      "learning_rate": 2.3471400394477318e-05,
      "loss": 0.0024,
      "step": 1790
    },
    {
      "epoch": 11.538461538461538,
      "grad_norm": 0.013362949714064598,
      "learning_rate": 2.2485207100591717e-05,
      "loss": 0.0024,
      "step": 1800
    },
    {
      "epoch": 11.538461538461538,
      "eval_accuracy": 0.8884892086330936,
      "eval_loss": 0.6356052160263062,
      "eval_runtime": 2.4395,
      "eval_samples_per_second": 113.959,
      "eval_steps_per_second": 14.347,
      "step": 1800
    },
    {
      "epoch": 11.602564102564102,
      "grad_norm": 0.013172836974263191,
      "learning_rate": 2.1499013806706113e-05,
      "loss": 0.0024,
      "step": 1810
    },
    {
      "epoch": 11.666666666666666,
      "grad_norm": 0.011238598264753819,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 0.0023,
      "step": 1820
    },
    {
      "epoch": 11.73076923076923,
      "grad_norm": 0.01071783434599638,
      "learning_rate": 1.952662721893491e-05,
      "loss": 0.0023,
      "step": 1830
    },
    {
      "epoch": 11.794871794871796,
      "grad_norm": 0.01259815227240324,
      "learning_rate": 1.854043392504931e-05,
      "loss": 0.0025,
      "step": 1840
    },
    {
      "epoch": 11.85897435897436,
      "grad_norm": 0.012655877508223057,
      "learning_rate": 1.755424063116371e-05,
      "loss": 0.0024,
      "step": 1850
    },
    {
      "epoch": 11.923076923076923,
      "grad_norm": 0.011328635737299919,
      "learning_rate": 1.6568047337278108e-05,
      "loss": 0.0023,
      "step": 1860
    },
    {
      "epoch": 11.987179487179487,
      "grad_norm": 0.01103618461638689,
      "learning_rate": 1.5581854043392503e-05,
      "loss": 0.0021,
      "step": 1870
    },
    {
      "epoch": 12.051282051282051,
      "grad_norm": 0.010929819196462631,
      "learning_rate": 1.4595660749506904e-05,
      "loss": 0.0022,
      "step": 1880
    },
    {
      "epoch": 12.115384615384615,
      "grad_norm": 0.01153373159468174,
      "learning_rate": 1.3609467455621303e-05,
      "loss": 0.0023,
      "step": 1890
    },
    {
      "epoch": 12.179487179487179,
      "grad_norm": 0.013212679885327816,
      "learning_rate": 1.2623274161735702e-05,
      "loss": 0.0023,
      "step": 1900
    },
    {
      "epoch": 12.179487179487179,
      "eval_accuracy": 0.8884892086330936,
      "eval_loss": 0.6381567716598511,
      "eval_runtime": 2.6807,
      "eval_samples_per_second": 103.703,
      "eval_steps_per_second": 13.056,
      "step": 1900
    },
    {
      "epoch": 12.243589743589743,
      "grad_norm": 0.01274055428802967,
      "learning_rate": 1.16370808678501e-05,
      "loss": 0.0022,
      "step": 1910
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 0.010565485805273056,
      "learning_rate": 1.0650887573964498e-05,
      "loss": 0.0023,
      "step": 1920
    },
    {
      "epoch": 12.371794871794872,
      "grad_norm": 0.011520653031766415,
      "learning_rate": 9.664694280078896e-06,
      "loss": 0.0023,
      "step": 1930
    },
    {
      "epoch": 12.435897435897436,
      "grad_norm": 0.01070806011557579,
      "learning_rate": 8.678500986193295e-06,
      "loss": 0.0023,
      "step": 1940
    },
    {
      "epoch": 12.5,
      "grad_norm": 0.011207307688891888,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0023,
      "step": 1950
    },
    {
      "epoch": 12.564102564102564,
      "grad_norm": 0.010393859818577766,
      "learning_rate": 6.706114398422091e-06,
      "loss": 0.0023,
      "step": 1960
    },
    {
      "epoch": 12.628205128205128,
      "grad_norm": 0.012986727990210056,
      "learning_rate": 5.71992110453649e-06,
      "loss": 0.0023,
      "step": 1970
    },
    {
      "epoch": 12.692307692307692,
      "grad_norm": 0.012365762144327164,
      "learning_rate": 4.733727810650888e-06,
      "loss": 0.0023,
      "step": 1980
    },
    {
      "epoch": 12.756410256410255,
      "grad_norm": 0.011601632460951805,
      "learning_rate": 3.7475345167652858e-06,
      "loss": 0.0021,
      "step": 1990
    },
    {
      "epoch": 12.820512820512821,
      "grad_norm": 0.011127516627311707,
      "learning_rate": 2.7613412228796843e-06,
      "loss": 0.0023,
      "step": 2000
    },
    {
      "epoch": 12.820512820512821,
      "eval_accuracy": 0.8884892086330936,
      "eval_loss": 0.6398880481719971,
      "eval_runtime": 2.2967,
      "eval_samples_per_second": 121.044,
      "eval_steps_per_second": 15.239,
      "step": 2000
    },
    {
      "epoch": 12.884615384615385,
      "grad_norm": 0.01038967166095972,
      "learning_rate": 1.775147928994083e-06,
      "loss": 0.0021,
      "step": 2010
    },
    {
      "epoch": 12.948717948717949,
      "grad_norm": 0.01115449983626604,
      "learning_rate": 7.889546351084814e-07,
      "loss": 0.0023,
      "step": 2020
    },
    {
      "epoch": 13.0,
      "step": 2028,
      "total_flos": 2.5134977377120666e+18,
      "train_loss": 0.23581767766102707,
      "train_runtime": 771.9998,
      "train_samples_per_second": 42.014,
      "train_steps_per_second": 2.627
    }
  ],
  "logging_steps": 10,
  "max_steps": 2028,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 13,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.5134977377120666e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}