{
  "best_metric": 0.26709362864494324,
  "best_model_checkpoint": "vit-weldclassifyv3/checkpoint-1000",
  "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": 1.1238359212875366,
      "learning_rate": 0.0001990138067061144,
      "loss": 1.2792,
      "step": 10
    },
    {
      "epoch": 0.1282051282051282,
      "grad_norm": 1.7609381675720215,
      "learning_rate": 0.0001980276134122288,
      "loss": 1.1608,
      "step": 20
    },
    {
      "epoch": 0.19230769230769232,
      "grad_norm": 1.1302402019500732,
      "learning_rate": 0.0001970414201183432,
      "loss": 1.1137,
      "step": 30
    },
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 3.0818381309509277,
      "learning_rate": 0.0001960552268244576,
      "loss": 1.0182,
      "step": 40
    },
    {
      "epoch": 0.32051282051282054,
      "grad_norm": 1.0849112272262573,
      "learning_rate": 0.000195069033530572,
      "loss": 1.2566,
      "step": 50
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 1.0367400646209717,
      "learning_rate": 0.0001940828402366864,
      "loss": 1.0554,
      "step": 60
    },
    {
      "epoch": 0.44871794871794873,
      "grad_norm": 2.154937744140625,
      "learning_rate": 0.0001930966469428008,
      "loss": 1.0788,
      "step": 70
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 1.1957019567489624,
      "learning_rate": 0.0001921104536489152,
      "loss": 1.2299,
      "step": 80
    },
    {
      "epoch": 0.5769230769230769,
      "grad_norm": 1.3072288036346436,
      "learning_rate": 0.0001911242603550296,
      "loss": 1.0164,
      "step": 90
    },
    {
      "epoch": 0.6410256410256411,
      "grad_norm": 1.340012550354004,
      "learning_rate": 0.00019013806706114398,
      "loss": 0.8398,
      "step": 100
    },
    {
      "epoch": 0.6410256410256411,
      "eval_accuracy": 0.5035971223021583,
      "eval_loss": 1.031232476234436,
      "eval_runtime": 2.6053,
      "eval_samples_per_second": 106.706,
      "eval_steps_per_second": 13.434,
      "step": 100
    },
    {
      "epoch": 0.7051282051282052,
      "grad_norm": 2.5044240951538086,
      "learning_rate": 0.00018915187376725837,
      "loss": 0.8484,
      "step": 110
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 2.8144619464874268,
      "learning_rate": 0.00018816568047337278,
      "loss": 0.8283,
      "step": 120
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 1.8459696769714355,
      "learning_rate": 0.0001871794871794872,
      "loss": 0.8201,
      "step": 130
    },
    {
      "epoch": 0.8974358974358975,
      "grad_norm": 3.3761656284332275,
      "learning_rate": 0.0001861932938856016,
      "loss": 0.7736,
      "step": 140
    },
    {
      "epoch": 0.9615384615384616,
      "grad_norm": 2.285553216934204,
      "learning_rate": 0.00018520710059171598,
      "loss": 0.7505,
      "step": 150
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 3.3042893409729004,
      "learning_rate": 0.0001842209072978304,
      "loss": 0.6614,
      "step": 160
    },
    {
      "epoch": 1.0897435897435896,
      "grad_norm": 3.174131393432617,
      "learning_rate": 0.00018323471400394478,
      "loss": 0.7393,
      "step": 170
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 2.8764002323150635,
      "learning_rate": 0.00018224852071005917,
      "loss": 0.892,
      "step": 180
    },
    {
      "epoch": 1.217948717948718,
      "grad_norm": 3.123760223388672,
      "learning_rate": 0.00018126232741617356,
      "loss": 0.7055,
      "step": 190
    },
    {
      "epoch": 1.282051282051282,
      "grad_norm": 2.54199481010437,
      "learning_rate": 0.00018027613412228798,
      "loss": 0.5613,
      "step": 200
    },
    {
      "epoch": 1.282051282051282,
      "eval_accuracy": 0.6618705035971223,
      "eval_loss": 0.7067616581916809,
      "eval_runtime": 2.5932,
      "eval_samples_per_second": 107.202,
      "eval_steps_per_second": 13.497,
      "step": 200
    },
    {
      "epoch": 1.3461538461538463,
      "grad_norm": 4.848214626312256,
      "learning_rate": 0.0001792899408284024,
      "loss": 0.634,
      "step": 210
    },
    {
      "epoch": 1.4102564102564101,
      "grad_norm": 2.621889352798462,
      "learning_rate": 0.00017830374753451678,
      "loss": 0.6666,
      "step": 220
    },
    {
      "epoch": 1.4743589743589745,
      "grad_norm": 3.7964868545532227,
      "learning_rate": 0.00017731755424063117,
      "loss": 0.5344,
      "step": 230
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 3.703036308288574,
      "learning_rate": 0.00017633136094674556,
      "loss": 0.5636,
      "step": 240
    },
    {
      "epoch": 1.6025641025641026,
      "grad_norm": 3.081791400909424,
      "learning_rate": 0.00017534516765285997,
      "loss": 0.4211,
      "step": 250
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 3.167443037033081,
      "learning_rate": 0.00017435897435897436,
      "loss": 0.6438,
      "step": 260
    },
    {
      "epoch": 1.7307692307692308,
      "grad_norm": 6.15064811706543,
      "learning_rate": 0.00017337278106508875,
      "loss": 0.5872,
      "step": 270
    },
    {
      "epoch": 1.7948717948717947,
      "grad_norm": 2.0773203372955322,
      "learning_rate": 0.00017238658777120317,
      "loss": 0.6172,
      "step": 280
    },
    {
      "epoch": 1.858974358974359,
      "grad_norm": 3.2962491512298584,
      "learning_rate": 0.00017140039447731758,
      "loss": 0.5252,
      "step": 290
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 2.557178020477295,
      "learning_rate": 0.00017041420118343197,
      "loss": 0.4296,
      "step": 300
    },
    {
      "epoch": 1.9230769230769231,
      "eval_accuracy": 0.8309352517985612,
      "eval_loss": 0.4007588028907776,
      "eval_runtime": 2.5685,
      "eval_samples_per_second": 108.233,
      "eval_steps_per_second": 13.626,
      "step": 300
    },
    {
      "epoch": 1.9871794871794872,
      "grad_norm": 2.061478614807129,
      "learning_rate": 0.00016942800788954636,
      "loss": 0.4041,
      "step": 310
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 2.703629732131958,
      "learning_rate": 0.00016844181459566075,
      "loss": 0.3791,
      "step": 320
    },
    {
      "epoch": 2.1153846153846154,
      "grad_norm": 0.8648492693901062,
      "learning_rate": 0.00016745562130177514,
      "loss": 0.2943,
      "step": 330
    },
    {
      "epoch": 2.1794871794871793,
      "grad_norm": 4.140072345733643,
      "learning_rate": 0.00016646942800788956,
      "loss": 0.2485,
      "step": 340
    },
    {
      "epoch": 2.2435897435897436,
      "grad_norm": 5.02016019821167,
      "learning_rate": 0.00016548323471400394,
      "loss": 0.4904,
      "step": 350
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 1.8560619354248047,
      "learning_rate": 0.00016449704142011836,
      "loss": 0.3835,
      "step": 360
    },
    {
      "epoch": 2.371794871794872,
      "grad_norm": 10.23709487915039,
      "learning_rate": 0.00016351084812623275,
      "loss": 0.3015,
      "step": 370
    },
    {
      "epoch": 2.435897435897436,
      "grad_norm": 5.944732666015625,
      "learning_rate": 0.00016252465483234716,
      "loss": 0.3199,
      "step": 380
    },
    {
      "epoch": 2.5,
      "grad_norm": 2.0510830879211426,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.2512,
      "step": 390
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 3.4166271686553955,
      "learning_rate": 0.00016055226824457594,
      "loss": 0.3475,
      "step": 400
    },
    {
      "epoch": 2.564102564102564,
      "eval_accuracy": 0.8812949640287769,
      "eval_loss": 0.33450016379356384,
      "eval_runtime": 3.6169,
      "eval_samples_per_second": 76.862,
      "eval_steps_per_second": 9.677,
      "step": 400
    },
    {
      "epoch": 2.628205128205128,
      "grad_norm": 6.59075927734375,
      "learning_rate": 0.00015956607495069033,
      "loss": 0.2296,
      "step": 410
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 2.345207929611206,
      "learning_rate": 0.00015857988165680475,
      "loss": 0.2535,
      "step": 420
    },
    {
      "epoch": 2.7564102564102564,
      "grad_norm": 3.3290200233459473,
      "learning_rate": 0.00015759368836291914,
      "loss": 0.2902,
      "step": 430
    },
    {
      "epoch": 2.8205128205128203,
      "grad_norm": 3.723925828933716,
      "learning_rate": 0.00015660749506903355,
      "loss": 0.2591,
      "step": 440
    },
    {
      "epoch": 2.8846153846153846,
      "grad_norm": 4.227429389953613,
      "learning_rate": 0.00015562130177514794,
      "loss": 0.2915,
      "step": 450
    },
    {
      "epoch": 2.948717948717949,
      "grad_norm": 3.918539047241211,
      "learning_rate": 0.00015463510848126233,
      "loss": 0.3859,
      "step": 460
    },
    {
      "epoch": 3.0128205128205128,
      "grad_norm": 2.824794292449951,
      "learning_rate": 0.00015364891518737675,
      "loss": 0.3425,
      "step": 470
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.28933683037757874,
      "learning_rate": 0.00015266272189349113,
      "loss": 0.1796,
      "step": 480
    },
    {
      "epoch": 3.141025641025641,
      "grad_norm": 2.101590394973755,
      "learning_rate": 0.00015167652859960552,
      "loss": 0.1881,
      "step": 490
    },
    {
      "epoch": 3.2051282051282053,
      "grad_norm": 0.31414705514907837,
      "learning_rate": 0.0001506903353057199,
      "loss": 0.1183,
      "step": 500
    },
    {
      "epoch": 3.2051282051282053,
      "eval_accuracy": 0.8489208633093526,
      "eval_loss": 0.4293219745159149,
      "eval_runtime": 3.4489,
      "eval_samples_per_second": 80.605,
      "eval_steps_per_second": 10.148,
      "step": 500
    },
    {
      "epoch": 3.269230769230769,
      "grad_norm": 3.8199613094329834,
      "learning_rate": 0.00014970414201183433,
      "loss": 0.1736,
      "step": 510
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 1.631716251373291,
      "learning_rate": 0.00014871794871794872,
      "loss": 0.1903,
      "step": 520
    },
    {
      "epoch": 3.3974358974358974,
      "grad_norm": 3.813403844833374,
      "learning_rate": 0.00014773175542406313,
      "loss": 0.2243,
      "step": 530
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 6.256704330444336,
      "learning_rate": 0.00014674556213017752,
      "loss": 0.1941,
      "step": 540
    },
    {
      "epoch": 3.5256410256410255,
      "grad_norm": 6.570198059082031,
      "learning_rate": 0.0001457593688362919,
      "loss": 0.1747,
      "step": 550
    },
    {
      "epoch": 3.58974358974359,
      "grad_norm": 3.7718803882598877,
      "learning_rate": 0.00014477317554240633,
      "loss": 0.2871,
      "step": 560
    },
    {
      "epoch": 3.6538461538461537,
      "grad_norm": 3.889697790145874,
      "learning_rate": 0.00014378698224852072,
      "loss": 0.173,
      "step": 570
    },
    {
      "epoch": 3.717948717948718,
      "grad_norm": 2.5270767211914062,
      "learning_rate": 0.0001428007889546351,
      "loss": 0.2472,
      "step": 580
    },
    {
      "epoch": 3.782051282051282,
      "grad_norm": 0.2746649980545044,
      "learning_rate": 0.0001418145956607495,
      "loss": 0.1564,
      "step": 590
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 2.9636824131011963,
      "learning_rate": 0.0001408284023668639,
      "loss": 0.1531,
      "step": 600
    },
    {
      "epoch": 3.8461538461538463,
      "eval_accuracy": 0.9136690647482014,
      "eval_loss": 0.2748344838619232,
      "eval_runtime": 2.583,
      "eval_samples_per_second": 107.625,
      "eval_steps_per_second": 13.55,
      "step": 600
    },
    {
      "epoch": 3.91025641025641,
      "grad_norm": 2.8231372833251953,
      "learning_rate": 0.00013984220907297832,
      "loss": 0.1651,
      "step": 610
    },
    {
      "epoch": 3.9743589743589745,
      "grad_norm": 0.6206132769584656,
      "learning_rate": 0.0001388560157790927,
      "loss": 0.1854,
      "step": 620
    },
    {
      "epoch": 4.038461538461538,
      "grad_norm": 0.720039427280426,
      "learning_rate": 0.0001378698224852071,
      "loss": 0.2629,
      "step": 630
    },
    {
      "epoch": 4.102564102564102,
      "grad_norm": 0.28592395782470703,
      "learning_rate": 0.00013688362919132152,
      "loss": 0.0969,
      "step": 640
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 1.0571306943893433,
      "learning_rate": 0.0001358974358974359,
      "loss": 0.0816,
      "step": 650
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 5.520449638366699,
      "learning_rate": 0.0001349112426035503,
      "loss": 0.0997,
      "step": 660
    },
    {
      "epoch": 4.294871794871795,
      "grad_norm": 10.63196849822998,
      "learning_rate": 0.00013392504930966468,
      "loss": 0.1369,
      "step": 670
    },
    {
      "epoch": 4.358974358974359,
      "grad_norm": 2.723325729370117,
      "learning_rate": 0.0001329388560157791,
      "loss": 0.0631,
      "step": 680
    },
    {
      "epoch": 4.423076923076923,
      "grad_norm": 0.07080604135990143,
      "learning_rate": 0.00013195266272189352,
      "loss": 0.1089,
      "step": 690
    },
    {
      "epoch": 4.487179487179487,
      "grad_norm": 5.3309326171875,
      "learning_rate": 0.0001309664694280079,
      "loss": 0.1174,
      "step": 700
    },
    {
      "epoch": 4.487179487179487,
      "eval_accuracy": 0.8812949640287769,
      "eval_loss": 0.36486199498176575,
      "eval_runtime": 3.5449,
      "eval_samples_per_second": 78.422,
      "eval_steps_per_second": 9.873,
      "step": 700
    },
    {
      "epoch": 4.551282051282051,
      "grad_norm": 4.462215900421143,
      "learning_rate": 0.0001299802761341223,
      "loss": 0.1378,
      "step": 710
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 8.144935607910156,
      "learning_rate": 0.00012899408284023668,
      "loss": 0.0917,
      "step": 720
    },
    {
      "epoch": 4.67948717948718,
      "grad_norm": 0.08221765607595444,
      "learning_rate": 0.0001280078895463511,
      "loss": 0.0562,
      "step": 730
    },
    {
      "epoch": 4.743589743589744,
      "grad_norm": 8.699402809143066,
      "learning_rate": 0.0001270216962524655,
      "loss": 0.1787,
      "step": 740
    },
    {
      "epoch": 4.8076923076923075,
      "grad_norm": 7.376129150390625,
      "learning_rate": 0.00012603550295857988,
      "loss": 0.064,
      "step": 750
    },
    {
      "epoch": 4.871794871794872,
      "grad_norm": 0.08259747177362442,
      "learning_rate": 0.0001250493096646943,
      "loss": 0.0927,
      "step": 760
    },
    {
      "epoch": 4.935897435897436,
      "grad_norm": 0.08165480941534042,
      "learning_rate": 0.00012406311637080868,
      "loss": 0.1067,
      "step": 770
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.19699496030807495,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.0974,
      "step": 780
    },
    {
      "epoch": 5.064102564102564,
      "grad_norm": 3.4025802612304688,
      "learning_rate": 0.00012209072978303749,
      "loss": 0.0977,
      "step": 790
    },
    {
      "epoch": 5.128205128205128,
      "grad_norm": 2.273251533508301,
      "learning_rate": 0.00012110453648915188,
      "loss": 0.0498,
      "step": 800
    },
    {
      "epoch": 5.128205128205128,
      "eval_accuracy": 0.8920863309352518,
      "eval_loss": 0.327900767326355,
      "eval_runtime": 3.5689,
      "eval_samples_per_second": 77.895,
      "eval_steps_per_second": 9.807,
      "step": 800
    },
    {
      "epoch": 5.1923076923076925,
      "grad_norm": 0.624547004699707,
      "learning_rate": 0.00012011834319526626,
      "loss": 0.0821,
      "step": 810
    },
    {
      "epoch": 5.256410256410256,
      "grad_norm": 3.7610857486724854,
      "learning_rate": 0.00011913214990138068,
      "loss": 0.0481,
      "step": 820
    },
    {
      "epoch": 5.32051282051282,
      "grad_norm": 0.06995019316673279,
      "learning_rate": 0.00011814595660749508,
      "loss": 0.0117,
      "step": 830
    },
    {
      "epoch": 5.384615384615385,
      "grad_norm": 0.08295288681983948,
      "learning_rate": 0.00011715976331360947,
      "loss": 0.0122,
      "step": 840
    },
    {
      "epoch": 5.448717948717949,
      "grad_norm": 0.0509318970143795,
      "learning_rate": 0.00011617357001972386,
      "loss": 0.0447,
      "step": 850
    },
    {
      "epoch": 5.512820512820513,
      "grad_norm": 0.05534667894244194,
      "learning_rate": 0.00011518737672583828,
      "loss": 0.0786,
      "step": 860
    },
    {
      "epoch": 5.576923076923077,
      "grad_norm": 0.36779171228408813,
      "learning_rate": 0.00011420118343195268,
      "loss": 0.0427,
      "step": 870
    },
    {
      "epoch": 5.641025641025641,
      "grad_norm": 0.0412164181470871,
      "learning_rate": 0.00011321499013806707,
      "loss": 0.0151,
      "step": 880
    },
    {
      "epoch": 5.705128205128205,
      "grad_norm": 0.03488855063915253,
      "learning_rate": 0.00011222879684418146,
      "loss": 0.0331,
      "step": 890
    },
    {
      "epoch": 5.769230769230769,
      "grad_norm": 5.378381252288818,
      "learning_rate": 0.00011124260355029586,
      "loss": 0.0817,
      "step": 900
    },
    {
      "epoch": 5.769230769230769,
      "eval_accuracy": 0.935251798561151,
      "eval_loss": 0.27631285786628723,
      "eval_runtime": 3.5274,
      "eval_samples_per_second": 78.812,
      "eval_steps_per_second": 9.922,
      "step": 900
    },
    {
      "epoch": 5.833333333333333,
      "grad_norm": 0.121369369328022,
      "learning_rate": 0.00011025641025641027,
      "loss": 0.0164,
      "step": 910
    },
    {
      "epoch": 5.897435897435898,
      "grad_norm": 0.30134713649749756,
      "learning_rate": 0.00010927021696252466,
      "loss": 0.0399,
      "step": 920
    },
    {
      "epoch": 5.961538461538462,
      "grad_norm": 9.133658409118652,
      "learning_rate": 0.00010828402366863905,
      "loss": 0.0893,
      "step": 930
    },
    {
      "epoch": 6.0256410256410255,
      "grad_norm": 0.037670884281396866,
      "learning_rate": 0.00010729783037475345,
      "loss": 0.0285,
      "step": 940
    },
    {
      "epoch": 6.089743589743589,
      "grad_norm": 4.902493953704834,
      "learning_rate": 0.00010631163708086787,
      "loss": 0.0796,
      "step": 950
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 6.080508232116699,
      "learning_rate": 0.00010532544378698226,
      "loss": 0.1406,
      "step": 960
    },
    {
      "epoch": 6.217948717948718,
      "grad_norm": 3.923866033554077,
      "learning_rate": 0.00010433925049309665,
      "loss": 0.0914,
      "step": 970
    },
    {
      "epoch": 6.282051282051282,
      "grad_norm": 0.9966168999671936,
      "learning_rate": 0.00010335305719921105,
      "loss": 0.0906,
      "step": 980
    },
    {
      "epoch": 6.346153846153846,
      "grad_norm": 0.2861874997615814,
      "learning_rate": 0.00010236686390532544,
      "loss": 0.0335,
      "step": 990
    },
    {
      "epoch": 6.410256410256411,
      "grad_norm": 0.030173303559422493,
      "learning_rate": 0.00010138067061143986,
      "loss": 0.0075,
      "step": 1000
    },
    {
      "epoch": 6.410256410256411,
      "eval_accuracy": 0.920863309352518,
      "eval_loss": 0.26709362864494324,
      "eval_runtime": 2.5656,
      "eval_samples_per_second": 108.358,
      "eval_steps_per_second": 13.642,
      "step": 1000
    },
    {
      "epoch": 6.4743589743589745,
      "grad_norm": 0.02864202857017517,
      "learning_rate": 0.00010039447731755424,
      "loss": 0.0628,
      "step": 1010
    },
    {
      "epoch": 6.538461538461538,
      "grad_norm": 0.6442215442657471,
      "learning_rate": 9.940828402366865e-05,
      "loss": 0.0205,
      "step": 1020
    },
    {
      "epoch": 6.602564102564102,
      "grad_norm": 0.047572050243616104,
      "learning_rate": 9.842209072978305e-05,
      "loss": 0.1029,
      "step": 1030
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 11.545117378234863,
      "learning_rate": 9.743589743589744e-05,
      "loss": 0.0454,
      "step": 1040
    },
    {
      "epoch": 6.730769230769231,
      "grad_norm": 0.08284212648868561,
      "learning_rate": 9.644970414201184e-05,
      "loss": 0.0059,
      "step": 1050
    },
    {
      "epoch": 6.794871794871795,
      "grad_norm": 0.19572466611862183,
      "learning_rate": 9.546351084812624e-05,
      "loss": 0.0907,
      "step": 1060
    },
    {
      "epoch": 6.858974358974359,
      "grad_norm": 0.041858524084091187,
      "learning_rate": 9.447731755424064e-05,
      "loss": 0.0562,
      "step": 1070
    },
    {
      "epoch": 6.923076923076923,
      "grad_norm": 0.045817919075489044,
      "learning_rate": 9.349112426035503e-05,
      "loss": 0.0262,
      "step": 1080
    },
    {
      "epoch": 6.987179487179487,
      "grad_norm": 0.14285103976726532,
      "learning_rate": 9.250493096646942e-05,
      "loss": 0.0185,
      "step": 1090
    },
    {
      "epoch": 7.051282051282051,
      "grad_norm": 0.027153167873620987,
      "learning_rate": 9.151873767258384e-05,
      "loss": 0.0265,
      "step": 1100
    },
    {
      "epoch": 7.051282051282051,
      "eval_accuracy": 0.920863309352518,
      "eval_loss": 0.3184675872325897,
      "eval_runtime": 2.4994,
      "eval_samples_per_second": 111.228,
      "eval_steps_per_second": 14.004,
      "step": 1100
    },
    {
      "epoch": 7.115384615384615,
      "grad_norm": 0.027239225804805756,
      "learning_rate": 9.053254437869823e-05,
      "loss": 0.0629,
      "step": 1110
    },
    {
      "epoch": 7.17948717948718,
      "grad_norm": 0.02252044342458248,
      "learning_rate": 8.954635108481263e-05,
      "loss": 0.0198,
      "step": 1120
    },
    {
      "epoch": 7.243589743589744,
      "grad_norm": 0.038163017481565475,
      "learning_rate": 8.856015779092702e-05,
      "loss": 0.0323,
      "step": 1130
    },
    {
      "epoch": 7.3076923076923075,
      "grad_norm": 0.038108520209789276,
      "learning_rate": 8.757396449704143e-05,
      "loss": 0.0379,
      "step": 1140
    },
    {
      "epoch": 7.371794871794872,
      "grad_norm": 0.019251950085163116,
      "learning_rate": 8.658777120315582e-05,
      "loss": 0.028,
      "step": 1150
    },
    {
      "epoch": 7.435897435897436,
      "grad_norm": 0.01974482275545597,
      "learning_rate": 8.560157790927023e-05,
      "loss": 0.0043,
      "step": 1160
    },
    {
      "epoch": 7.5,
      "grad_norm": 2.0657660961151123,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0069,
      "step": 1170
    },
    {
      "epoch": 7.564102564102564,
      "grad_norm": 0.025256551802158356,
      "learning_rate": 8.362919132149902e-05,
      "loss": 0.0045,
      "step": 1180
    },
    {
      "epoch": 7.628205128205128,
      "grad_norm": 0.018220046535134315,
      "learning_rate": 8.264299802761342e-05,
      "loss": 0.0038,
      "step": 1190
    },
    {
      "epoch": 7.6923076923076925,
      "grad_norm": 15.359424591064453,
      "learning_rate": 8.165680473372781e-05,
      "loss": 0.0457,
      "step": 1200
    },
    {
      "epoch": 7.6923076923076925,
      "eval_accuracy": 0.9100719424460432,
      "eval_loss": 0.37759149074554443,
      "eval_runtime": 3.7105,
      "eval_samples_per_second": 74.922,
      "eval_steps_per_second": 9.433,
      "step": 1200
    },
    {
      "epoch": 7.756410256410256,
      "grad_norm": 0.017574544996023178,
      "learning_rate": 8.067061143984221e-05,
      "loss": 0.0038,
      "step": 1210
    },
    {
      "epoch": 7.82051282051282,
      "grad_norm": 0.01818062551319599,
      "learning_rate": 7.968441814595661e-05,
      "loss": 0.004,
      "step": 1220
    },
    {
      "epoch": 7.884615384615385,
      "grad_norm": 0.024309920147061348,
      "learning_rate": 7.869822485207101e-05,
      "loss": 0.0038,
      "step": 1230
    },
    {
      "epoch": 7.948717948717949,
      "grad_norm": 0.07784882932901382,
      "learning_rate": 7.77120315581854e-05,
      "loss": 0.0059,
      "step": 1240
    },
    {
      "epoch": 8.012820512820513,
      "grad_norm": 0.035918813198804855,
      "learning_rate": 7.67258382642998e-05,
      "loss": 0.0047,
      "step": 1250
    },
    {
      "epoch": 8.076923076923077,
      "grad_norm": 0.35960128903388977,
      "learning_rate": 7.573964497041421e-05,
      "loss": 0.0068,
      "step": 1260
    },
    {
      "epoch": 8.14102564102564,
      "grad_norm": 0.015659425407648087,
      "learning_rate": 7.475345167652861e-05,
      "loss": 0.0039,
      "step": 1270
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 0.01647733524441719,
      "learning_rate": 7.3767258382643e-05,
      "loss": 0.0035,
      "step": 1280
    },
    {
      "epoch": 8.26923076923077,
      "grad_norm": 0.015693532302975655,
      "learning_rate": 7.27810650887574e-05,
      "loss": 0.0203,
      "step": 1290
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 0.01847326196730137,
      "learning_rate": 7.17948717948718e-05,
      "loss": 0.0032,
      "step": 1300
    },
    {
      "epoch": 8.333333333333334,
      "eval_accuracy": 0.9388489208633094,
      "eval_loss": 0.2834855318069458,
      "eval_runtime": 3.4846,
      "eval_samples_per_second": 79.78,
      "eval_steps_per_second": 10.044,
      "step": 1300
    },
    {
      "epoch": 8.397435897435898,
      "grad_norm": 0.01613900437951088,
      "learning_rate": 7.08086785009862e-05,
      "loss": 0.0055,
      "step": 1310
    },
    {
      "epoch": 8.461538461538462,
      "grad_norm": 0.015385860577225685,
      "learning_rate": 6.98224852071006e-05,
      "loss": 0.0118,
      "step": 1320
    },
    {
      "epoch": 8.525641025641026,
      "grad_norm": 0.014069377444684505,
      "learning_rate": 6.883629191321498e-05,
      "loss": 0.0029,
      "step": 1330
    },
    {
      "epoch": 8.58974358974359,
      "grad_norm": 0.015751365572214127,
      "learning_rate": 6.78500986193294e-05,
      "loss": 0.0116,
      "step": 1340
    },
    {
      "epoch": 8.653846153846153,
      "grad_norm": 0.0143636055290699,
      "learning_rate": 6.686390532544379e-05,
      "loss": 0.0554,
      "step": 1350
    },
    {
      "epoch": 8.717948717948717,
      "grad_norm": 0.015580832026898861,
      "learning_rate": 6.587771203155819e-05,
      "loss": 0.0031,
      "step": 1360
    },
    {
      "epoch": 8.782051282051283,
      "grad_norm": 0.014274963177740574,
      "learning_rate": 6.489151873767258e-05,
      "loss": 0.0029,
      "step": 1370
    },
    {
      "epoch": 8.846153846153847,
      "grad_norm": 0.02313074842095375,
      "learning_rate": 6.390532544378698e-05,
      "loss": 0.0339,
      "step": 1380
    },
    {
      "epoch": 8.91025641025641,
      "grad_norm": 1.4907852411270142,
      "learning_rate": 6.291913214990139e-05,
      "loss": 0.0046,
      "step": 1390
    },
    {
      "epoch": 8.974358974358974,
      "grad_norm": 0.015585100278258324,
      "learning_rate": 6.193293885601579e-05,
      "loss": 0.0027,
      "step": 1400
    },
    {
      "epoch": 8.974358974358974,
      "eval_accuracy": 0.8884892086330936,
      "eval_loss": 0.5365095734596252,
      "eval_runtime": 2.5385,
      "eval_samples_per_second": 109.513,
      "eval_steps_per_second": 13.788,
      "step": 1400
    },
    {
      "epoch": 9.038461538461538,
      "grad_norm": 0.014531627297401428,
      "learning_rate": 6.094674556213018e-05,
      "loss": 0.0253,
      "step": 1410
    },
    {
      "epoch": 9.102564102564102,
      "grad_norm": 0.012606761418282986,
      "learning_rate": 5.996055226824457e-05,
      "loss": 0.0031,
      "step": 1420
    },
    {
      "epoch": 9.166666666666666,
      "grad_norm": 0.012980329804122448,
      "learning_rate": 5.897435897435898e-05,
      "loss": 0.003,
      "step": 1430
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 0.012560625560581684,
      "learning_rate": 5.798816568047337e-05,
      "loss": 0.0029,
      "step": 1440
    },
    {
      "epoch": 9.294871794871796,
      "grad_norm": 0.01267662551254034,
      "learning_rate": 5.700197238658778e-05,
      "loss": 0.0027,
      "step": 1450
    },
    {
      "epoch": 9.35897435897436,
      "grad_norm": 0.018838873133063316,
      "learning_rate": 5.601577909270217e-05,
      "loss": 0.0026,
      "step": 1460
    },
    {
      "epoch": 9.423076923076923,
      "grad_norm": 0.012279819697141647,
      "learning_rate": 5.502958579881658e-05,
      "loss": 0.0025,
      "step": 1470
    },
    {
      "epoch": 9.487179487179487,
      "grad_norm": 0.011527939699590206,
      "learning_rate": 5.4043392504930966e-05,
      "loss": 0.0023,
      "step": 1480
    },
    {
      "epoch": 9.551282051282051,
      "grad_norm": 0.01282353326678276,
      "learning_rate": 5.3057199211045375e-05,
      "loss": 0.0024,
      "step": 1490
    },
    {
      "epoch": 9.615384615384615,
      "grad_norm": 0.012979278340935707,
      "learning_rate": 5.2071005917159764e-05,
      "loss": 0.0024,
      "step": 1500
    },
    {
      "epoch": 9.615384615384615,
      "eval_accuracy": 0.9460431654676259,
      "eval_loss": 0.28174683451652527,
      "eval_runtime": 2.5662,
      "eval_samples_per_second": 108.33,
      "eval_steps_per_second": 13.639,
      "step": 1500
    },
    {
      "epoch": 9.679487179487179,
      "grad_norm": 0.012905549257993698,
      "learning_rate": 5.1084812623274167e-05,
      "loss": 0.0023,
      "step": 1510
    },
    {
      "epoch": 9.743589743589745,
      "grad_norm": 0.011163388378918171,
      "learning_rate": 5.009861932938856e-05,
      "loss": 0.0023,
      "step": 1520
    },
    {
      "epoch": 9.807692307692308,
      "grad_norm": 0.012818182818591595,
      "learning_rate": 4.9112426035502965e-05,
      "loss": 0.0023,
      "step": 1530
    },
    {
      "epoch": 9.871794871794872,
      "grad_norm": 0.012148367241024971,
      "learning_rate": 4.812623274161736e-05,
      "loss": 0.0023,
      "step": 1540
    },
    {
      "epoch": 9.935897435897436,
      "grad_norm": 0.012448051013052464,
      "learning_rate": 4.714003944773176e-05,
      "loss": 0.0021,
      "step": 1550
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.01428731344640255,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0024,
      "step": 1560
    },
    {
      "epoch": 10.064102564102564,
      "grad_norm": 0.012622198089957237,
      "learning_rate": 4.5167652859960554e-05,
      "loss": 0.0021,
      "step": 1570
    },
    {
      "epoch": 10.128205128205128,
      "grad_norm": 0.011053148657083511,
      "learning_rate": 4.418145956607495e-05,
      "loss": 0.0022,
      "step": 1580
    },
    {
      "epoch": 10.192307692307692,
      "grad_norm": 0.010602343827486038,
      "learning_rate": 4.319526627218935e-05,
      "loss": 0.0022,
      "step": 1590
    },
    {
      "epoch": 10.256410256410255,
      "grad_norm": 0.012038341723382473,
      "learning_rate": 4.220907297830375e-05,
      "loss": 0.0021,
      "step": 1600
    },
    {
      "epoch": 10.256410256410255,
      "eval_accuracy": 0.9460431654676259,
      "eval_loss": 0.28900742530822754,
      "eval_runtime": 3.7223,
      "eval_samples_per_second": 74.686,
      "eval_steps_per_second": 9.403,
      "step": 1600
    },
    {
      "epoch": 10.320512820512821,
      "grad_norm": 0.010555533692240715,
      "learning_rate": 4.122287968441815e-05,
      "loss": 0.0022,
      "step": 1610
    },
    {
      "epoch": 10.384615384615385,
      "grad_norm": 0.013419007882475853,
      "learning_rate": 4.0236686390532545e-05,
      "loss": 0.0023,
      "step": 1620
    },
    {
      "epoch": 10.448717948717949,
      "grad_norm": 0.009542490355670452,
      "learning_rate": 3.925049309664695e-05,
      "loss": 0.0021,
      "step": 1630
    },
    {
      "epoch": 10.512820512820513,
      "grad_norm": 0.01066176313906908,
      "learning_rate": 3.826429980276134e-05,
      "loss": 0.0021,
      "step": 1640
    },
    {
      "epoch": 10.576923076923077,
      "grad_norm": 0.009535272605717182,
      "learning_rate": 3.7278106508875746e-05,
      "loss": 0.002,
      "step": 1650
    },
    {
      "epoch": 10.64102564102564,
      "grad_norm": 0.010441762395203114,
      "learning_rate": 3.629191321499014e-05,
      "loss": 0.0021,
      "step": 1660
    },
    {
      "epoch": 10.705128205128204,
      "grad_norm": 0.009478352032601833,
      "learning_rate": 3.5305719921104544e-05,
      "loss": 0.002,
      "step": 1670
    },
    {
      "epoch": 10.76923076923077,
      "grad_norm": 0.009695933200418949,
      "learning_rate": 3.431952662721893e-05,
      "loss": 0.002,
      "step": 1680
    },
    {
      "epoch": 10.833333333333334,
      "grad_norm": 0.009181487374007702,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0019,
      "step": 1690
    },
    {
      "epoch": 10.897435897435898,
      "grad_norm": 0.010168403387069702,
      "learning_rate": 3.234714003944773e-05,
      "loss": 0.002,
      "step": 1700
    },
    {
      "epoch": 10.897435897435898,
      "eval_accuracy": 0.9460431654676259,
      "eval_loss": 0.29342445731163025,
      "eval_runtime": 2.6399,
      "eval_samples_per_second": 105.306,
      "eval_steps_per_second": 13.258,
      "step": 1700
    },
    {
      "epoch": 10.961538461538462,
      "grad_norm": 0.012188075110316277,
      "learning_rate": 3.136094674556213e-05,
      "loss": 0.0019,
      "step": 1710
    },
    {
      "epoch": 11.025641025641026,
      "grad_norm": 0.01147390529513359,
      "learning_rate": 3.037475345167653e-05,
      "loss": 0.002,
      "step": 1720
    },
    {
      "epoch": 11.08974358974359,
      "grad_norm": 0.00882646068930626,
      "learning_rate": 2.9388560157790928e-05,
      "loss": 0.0019,
      "step": 1730
    },
    {
      "epoch": 11.153846153846153,
      "grad_norm": 0.00944901816546917,
      "learning_rate": 2.8402366863905327e-05,
      "loss": 0.002,
      "step": 1740
    },
    {
      "epoch": 11.217948717948717,
      "grad_norm": 0.009782733395695686,
      "learning_rate": 2.7416173570019726e-05,
      "loss": 0.0019,
      "step": 1750
    },
    {
      "epoch": 11.282051282051283,
      "grad_norm": 0.008780627511441708,
      "learning_rate": 2.6429980276134125e-05,
      "loss": 0.0019,
      "step": 1760
    },
    {
      "epoch": 11.346153846153847,
      "grad_norm": 0.011177603155374527,
      "learning_rate": 2.5443786982248524e-05,
      "loss": 0.0019,
      "step": 1770
    },
    {
      "epoch": 11.41025641025641,
      "grad_norm": 0.009699178859591484,
      "learning_rate": 2.445759368836292e-05,
      "loss": 0.0018,
      "step": 1780
    },
    {
      "epoch": 11.474358974358974,
      "grad_norm": 0.012270525097846985,
      "learning_rate": 2.3471400394477318e-05,
      "loss": 0.0019,
      "step": 1790
    },
    {
      "epoch": 11.538461538461538,
      "grad_norm": 0.00872350949794054,
      "learning_rate": 2.2485207100591717e-05,
      "loss": 0.0019,
      "step": 1800
    },
    {
      "epoch": 11.538461538461538,
      "eval_accuracy": 0.9460431654676259,
      "eval_loss": 0.2975642681121826,
      "eval_runtime": 2.6404,
      "eval_samples_per_second": 105.285,
      "eval_steps_per_second": 13.255,
      "step": 1800
    },
    {
      "epoch": 11.602564102564102,
      "grad_norm": 0.009663211181759834,
      "learning_rate": 2.1499013806706113e-05,
      "loss": 0.0019,
      "step": 1810
    },
    {
      "epoch": 11.666666666666666,
      "grad_norm": 0.00840259250253439,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 0.0018,
      "step": 1820
    },
    {
      "epoch": 11.73076923076923,
      "grad_norm": 0.010098517872393131,
      "learning_rate": 1.952662721893491e-05,
      "loss": 0.0019,
      "step": 1830
    },
    {
      "epoch": 11.794871794871796,
      "grad_norm": 0.013970685191452503,
      "learning_rate": 1.854043392504931e-05,
      "loss": 0.0019,
      "step": 1840
    },
    {
      "epoch": 11.85897435897436,
      "grad_norm": 0.009059489704668522,
      "learning_rate": 1.755424063116371e-05,
      "loss": 0.0018,
      "step": 1850
    },
    {
      "epoch": 11.923076923076923,
      "grad_norm": 0.009793447330594063,
      "learning_rate": 1.6568047337278108e-05,
      "loss": 0.0019,
      "step": 1860
    },
    {
      "epoch": 11.987179487179487,
      "grad_norm": 0.008815682493150234,
      "learning_rate": 1.5581854043392503e-05,
      "loss": 0.0019,
      "step": 1870
    },
    {
      "epoch": 12.051282051282051,
      "grad_norm": 0.011554540134966373,
      "learning_rate": 1.4595660749506904e-05,
      "loss": 0.0018,
      "step": 1880
    },
    {
      "epoch": 12.115384615384615,
      "grad_norm": 0.010211811400949955,
      "learning_rate": 1.3609467455621303e-05,
      "loss": 0.0019,
      "step": 1890
    },
    {
      "epoch": 12.179487179487179,
      "grad_norm": 0.008440189994871616,
      "learning_rate": 1.2623274161735702e-05,
      "loss": 0.0018,
      "step": 1900
    },
    {
      "epoch": 12.179487179487179,
      "eval_accuracy": 0.9460431654676259,
      "eval_loss": 0.2996102273464203,
      "eval_runtime": 3.4755,
      "eval_samples_per_second": 79.989,
      "eval_steps_per_second": 10.071,
      "step": 1900
    },
    {
      "epoch": 12.243589743589743,
      "grad_norm": 0.010010773316025734,
      "learning_rate": 1.16370808678501e-05,
      "loss": 0.0018,
      "step": 1910
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 0.008343766443431377,
      "learning_rate": 1.0650887573964498e-05,
      "loss": 0.0017,
      "step": 1920
    },
    {
      "epoch": 12.371794871794872,
      "grad_norm": 0.009077221155166626,
      "learning_rate": 9.664694280078896e-06,
      "loss": 0.0018,
      "step": 1930
    },
    {
      "epoch": 12.435897435897436,
      "grad_norm": 0.008946426212787628,
      "learning_rate": 8.678500986193295e-06,
      "loss": 0.0018,
      "step": 1940
    },
    {
      "epoch": 12.5,
      "grad_norm": 0.008481796830892563,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0018,
      "step": 1950
    },
    {
      "epoch": 12.564102564102564,
      "grad_norm": 0.012229247018694878,
      "learning_rate": 6.706114398422091e-06,
      "loss": 0.0018,
      "step": 1960
    },
    {
      "epoch": 12.628205128205128,
      "grad_norm": 0.009792056865990162,
      "learning_rate": 5.71992110453649e-06,
      "loss": 0.0019,
      "step": 1970
    },
    {
      "epoch": 12.692307692307692,
      "grad_norm": 0.008133339695632458,
      "learning_rate": 4.733727810650888e-06,
      "loss": 0.0017,
      "step": 1980
    },
    {
      "epoch": 12.756410256410255,
      "grad_norm": 0.00821082666516304,
      "learning_rate": 3.7475345167652858e-06,
      "loss": 0.0018,
      "step": 1990
    },
    {
      "epoch": 12.820512820512821,
      "grad_norm": 0.008551714941859245,
      "learning_rate": 2.7613412228796843e-06,
      "loss": 0.0018,
      "step": 2000
    },
    {
      "epoch": 12.820512820512821,
      "eval_accuracy": 0.9460431654676259,
      "eval_loss": 0.3006349802017212,
      "eval_runtime": 2.5194,
      "eval_samples_per_second": 110.343,
      "eval_steps_per_second": 13.892,
      "step": 2000
    },
    {
      "epoch": 12.884615384615385,
      "grad_norm": 0.008309854194521904,
      "learning_rate": 1.775147928994083e-06,
      "loss": 0.0019,
      "step": 2010
    },
    {
      "epoch": 12.948717948717949,
      "grad_norm": 0.008332509547472,
      "learning_rate": 7.889546351084814e-07,
      "loss": 0.0017,
      "step": 2020
    },
    {
      "epoch": 13.0,
      "step": 2028,
      "total_flos": 2.5134977377120666e+18,
      "train_loss": 0.18042536613924515,
      "train_runtime": 1249.5195,
      "train_samples_per_second": 25.958,
      "train_steps_per_second": 1.623
    }
  ],
  "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
}