{
  "best_metric": 0.330095499753952,
  "best_model_checkpoint": "vit-weldclassifyv4/checkpoint-1000",
  "epoch": 10.0,
  "eval_steps": 100,
  "global_step": 1560,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0641025641025641,
      "grad_norm": 1.5280327796936035,
      "learning_rate": 0.00019871794871794874,
      "loss": 1.263,
      "step": 10
    },
    {
      "epoch": 0.1282051282051282,
      "grad_norm": 1.8514498472213745,
      "learning_rate": 0.00019743589743589744,
      "loss": 1.1159,
      "step": 20
    },
    {
      "epoch": 0.19230769230769232,
      "grad_norm": 1.2513030767440796,
      "learning_rate": 0.00019615384615384615,
      "loss": 1.164,
      "step": 30
    },
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 2.280437707901001,
      "learning_rate": 0.00019487179487179487,
      "loss": 0.9528,
      "step": 40
    },
    {
      "epoch": 0.32051282051282054,
      "grad_norm": 1.246656060218811,
      "learning_rate": 0.0001935897435897436,
      "loss": 1.0313,
      "step": 50
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 1.9358172416687012,
      "learning_rate": 0.00019230769230769233,
      "loss": 1.038,
      "step": 60
    },
    {
      "epoch": 0.44871794871794873,
      "grad_norm": 2.415847063064575,
      "learning_rate": 0.00019102564102564104,
      "loss": 0.9088,
      "step": 70
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 2.1359400749206543,
      "learning_rate": 0.00018974358974358974,
      "loss": 0.8265,
      "step": 80
    },
    {
      "epoch": 0.5769230769230769,
      "grad_norm": 2.2672863006591797,
      "learning_rate": 0.00018846153846153847,
      "loss": 0.8904,
      "step": 90
    },
    {
      "epoch": 0.6410256410256411,
      "grad_norm": 2.525250196456909,
      "learning_rate": 0.0001871794871794872,
      "loss": 0.8207,
      "step": 100
    },
    {
      "epoch": 0.6410256410256411,
      "eval_accuracy": 0.564748201438849,
      "eval_loss": 1.0335605144500732,
      "eval_runtime": 2.4754,
      "eval_samples_per_second": 112.303,
      "eval_steps_per_second": 14.139,
      "step": 100
    },
    {
      "epoch": 0.7051282051282052,
      "grad_norm": 1.6615636348724365,
      "learning_rate": 0.0001858974358974359,
      "loss": 0.9254,
      "step": 110
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 3.1376407146453857,
      "learning_rate": 0.00018461538461538463,
      "loss": 0.8125,
      "step": 120
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 1.8494981527328491,
      "learning_rate": 0.00018333333333333334,
      "loss": 0.6967,
      "step": 130
    },
    {
      "epoch": 0.8974358974358975,
      "grad_norm": 2.7374682426452637,
      "learning_rate": 0.00018205128205128207,
      "loss": 0.7031,
      "step": 140
    },
    {
      "epoch": 0.9615384615384616,
      "grad_norm": 2.8239476680755615,
      "learning_rate": 0.00018076923076923077,
      "loss": 0.7871,
      "step": 150
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 2.239936590194702,
      "learning_rate": 0.0001794871794871795,
      "loss": 0.8064,
      "step": 160
    },
    {
      "epoch": 1.0897435897435896,
      "grad_norm": 2.60086727142334,
      "learning_rate": 0.00017820512820512823,
      "loss": 0.5349,
      "step": 170
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 3.485903024673462,
      "learning_rate": 0.00017692307692307693,
      "loss": 0.5896,
      "step": 180
    },
    {
      "epoch": 1.217948717948718,
      "grad_norm": 3.2817769050598145,
      "learning_rate": 0.00017564102564102566,
      "loss": 0.729,
      "step": 190
    },
    {
      "epoch": 1.282051282051282,
      "grad_norm": 4.110422611236572,
      "learning_rate": 0.00017435897435897436,
      "loss": 0.6506,
      "step": 200
    },
    {
      "epoch": 1.282051282051282,
      "eval_accuracy": 0.579136690647482,
      "eval_loss": 1.1981723308563232,
      "eval_runtime": 2.8799,
      "eval_samples_per_second": 96.532,
      "eval_steps_per_second": 12.153,
      "step": 200
    },
    {
      "epoch": 1.3461538461538463,
      "grad_norm": 2.515904664993286,
      "learning_rate": 0.0001730769230769231,
      "loss": 0.8231,
      "step": 210
    },
    {
      "epoch": 1.4102564102564101,
      "grad_norm": 2.2017431259155273,
      "learning_rate": 0.0001717948717948718,
      "loss": 0.6216,
      "step": 220
    },
    {
      "epoch": 1.4743589743589745,
      "grad_norm": 2.253706693649292,
      "learning_rate": 0.00017051282051282053,
      "loss": 0.5126,
      "step": 230
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 3.4106080532073975,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.471,
      "step": 240
    },
    {
      "epoch": 1.6025641025641026,
      "grad_norm": 1.884901523590088,
      "learning_rate": 0.00016794871794871796,
      "loss": 0.4521,
      "step": 250
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 3.056332588195801,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.3808,
      "step": 260
    },
    {
      "epoch": 1.7307692307692308,
      "grad_norm": 3.2792108058929443,
      "learning_rate": 0.0001653846153846154,
      "loss": 0.3568,
      "step": 270
    },
    {
      "epoch": 1.7948717948717947,
      "grad_norm": 3.657397985458374,
      "learning_rate": 0.0001641025641025641,
      "loss": 0.4942,
      "step": 280
    },
    {
      "epoch": 1.858974358974359,
      "grad_norm": 3.565870523452759,
      "learning_rate": 0.00016282051282051282,
      "loss": 0.5117,
      "step": 290
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 1.8648629188537598,
      "learning_rate": 0.00016153846153846155,
      "loss": 0.5324,
      "step": 300
    },
    {
      "epoch": 1.9230769230769231,
      "eval_accuracy": 0.7769784172661871,
      "eval_loss": 0.605965256690979,
      "eval_runtime": 2.4146,
      "eval_samples_per_second": 115.134,
      "eval_steps_per_second": 14.495,
      "step": 300
    },
    {
      "epoch": 1.9871794871794872,
      "grad_norm": 5.665286540985107,
      "learning_rate": 0.00016025641025641028,
      "loss": 0.5566,
      "step": 310
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 0.6339452266693115,
      "learning_rate": 0.00015897435897435896,
      "loss": 0.3058,
      "step": 320
    },
    {
      "epoch": 2.1153846153846154,
      "grad_norm": 2.2966978549957275,
      "learning_rate": 0.0001576923076923077,
      "loss": 0.2831,
      "step": 330
    },
    {
      "epoch": 2.1794871794871793,
      "grad_norm": 2.510307550430298,
      "learning_rate": 0.00015641025641025642,
      "loss": 0.2055,
      "step": 340
    },
    {
      "epoch": 2.2435897435897436,
      "grad_norm": 1.1059399843215942,
      "learning_rate": 0.00015512820512820515,
      "loss": 0.2886,
      "step": 350
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 6.055357933044434,
      "learning_rate": 0.00015384615384615385,
      "loss": 0.3756,
      "step": 360
    },
    {
      "epoch": 2.371794871794872,
      "grad_norm": 0.5536957383155823,
      "learning_rate": 0.00015256410256410255,
      "loss": 0.412,
      "step": 370
    },
    {
      "epoch": 2.435897435897436,
      "grad_norm": 5.276978969573975,
      "learning_rate": 0.00015128205128205128,
      "loss": 0.1798,
      "step": 380
    },
    {
      "epoch": 2.5,
      "grad_norm": 2.25166916847229,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.1813,
      "step": 390
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 4.955526351928711,
      "learning_rate": 0.00014871794871794872,
      "loss": 0.2486,
      "step": 400
    },
    {
      "epoch": 2.564102564102564,
      "eval_accuracy": 0.7517985611510791,
      "eval_loss": 0.729444682598114,
      "eval_runtime": 4.0175,
      "eval_samples_per_second": 69.198,
      "eval_steps_per_second": 8.712,
      "step": 400
    },
    {
      "epoch": 2.628205128205128,
      "grad_norm": 5.6987690925598145,
      "learning_rate": 0.00014743589743589745,
      "loss": 0.5142,
      "step": 410
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.589967668056488,
      "learning_rate": 0.00014615384615384615,
      "loss": 0.2685,
      "step": 420
    },
    {
      "epoch": 2.7564102564102564,
      "grad_norm": 2.2702548503875732,
      "learning_rate": 0.00014487179487179488,
      "loss": 0.3104,
      "step": 430
    },
    {
      "epoch": 2.8205128205128203,
      "grad_norm": 4.440503120422363,
      "learning_rate": 0.0001435897435897436,
      "loss": 0.2192,
      "step": 440
    },
    {
      "epoch": 2.8846153846153846,
      "grad_norm": 1.690927267074585,
      "learning_rate": 0.0001423076923076923,
      "loss": 0.2557,
      "step": 450
    },
    {
      "epoch": 2.948717948717949,
      "grad_norm": 9.020477294921875,
      "learning_rate": 0.00014102564102564104,
      "loss": 0.3725,
      "step": 460
    },
    {
      "epoch": 3.0128205128205128,
      "grad_norm": 1.131715178489685,
      "learning_rate": 0.00013974358974358974,
      "loss": 0.3586,
      "step": 470
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.3979876041412354,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.1712,
      "step": 480
    },
    {
      "epoch": 3.141025641025641,
      "grad_norm": 1.2889968156814575,
      "learning_rate": 0.00013717948717948718,
      "loss": 0.1988,
      "step": 490
    },
    {
      "epoch": 3.2051282051282053,
      "grad_norm": 2.893319606781006,
      "learning_rate": 0.0001358974358974359,
      "loss": 0.1366,
      "step": 500
    },
    {
      "epoch": 3.2051282051282053,
      "eval_accuracy": 0.841726618705036,
      "eval_loss": 0.4832339882850647,
      "eval_runtime": 2.588,
      "eval_samples_per_second": 107.42,
      "eval_steps_per_second": 13.524,
      "step": 500
    },
    {
      "epoch": 3.269230769230769,
      "grad_norm": 3.6555581092834473,
      "learning_rate": 0.00013461538461538464,
      "loss": 0.1222,
      "step": 510
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.1904444396495819,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.1654,
      "step": 520
    },
    {
      "epoch": 3.3974358974358974,
      "grad_norm": 4.902673244476318,
      "learning_rate": 0.00013205128205128204,
      "loss": 0.198,
      "step": 530
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 0.30183860659599304,
      "learning_rate": 0.00013076923076923077,
      "loss": 0.2074,
      "step": 540
    },
    {
      "epoch": 3.5256410256410255,
      "grad_norm": 4.17673397064209,
      "learning_rate": 0.0001294871794871795,
      "loss": 0.1021,
      "step": 550
    },
    {
      "epoch": 3.58974358974359,
      "grad_norm": 1.6145508289337158,
      "learning_rate": 0.00012820512820512823,
      "loss": 0.1074,
      "step": 560
    },
    {
      "epoch": 3.6538461538461537,
      "grad_norm": 4.717573165893555,
      "learning_rate": 0.00012692307692307693,
      "loss": 0.1201,
      "step": 570
    },
    {
      "epoch": 3.717948717948718,
      "grad_norm": 1.2709864377975464,
      "learning_rate": 0.00012564102564102564,
      "loss": 0.0544,
      "step": 580
    },
    {
      "epoch": 3.782051282051282,
      "grad_norm": 3.7621912956237793,
      "learning_rate": 0.00012435897435897437,
      "loss": 0.2016,
      "step": 590
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 12.426462173461914,
      "learning_rate": 0.0001230769230769231,
      "loss": 0.3124,
      "step": 600
    },
    {
      "epoch": 3.8461538461538463,
      "eval_accuracy": 0.762589928057554,
      "eval_loss": 0.8676345348358154,
      "eval_runtime": 3.0816,
      "eval_samples_per_second": 90.213,
      "eval_steps_per_second": 11.358,
      "step": 600
    },
    {
      "epoch": 3.91025641025641,
      "grad_norm": 10.93652057647705,
      "learning_rate": 0.00012179487179487179,
      "loss": 0.2992,
      "step": 610
    },
    {
      "epoch": 3.9743589743589745,
      "grad_norm": 0.6971213221549988,
      "learning_rate": 0.00012051282051282052,
      "loss": 0.1864,
      "step": 620
    },
    {
      "epoch": 4.038461538461538,
      "grad_norm": 6.531364917755127,
      "learning_rate": 0.00011923076923076923,
      "loss": 0.1929,
      "step": 630
    },
    {
      "epoch": 4.102564102564102,
      "grad_norm": 0.8437137007713318,
      "learning_rate": 0.00011794871794871796,
      "loss": 0.0816,
      "step": 640
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 9.23108196258545,
      "learning_rate": 0.00011666666666666668,
      "loss": 0.0654,
      "step": 650
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 1.1913517713546753,
      "learning_rate": 0.00011538461538461538,
      "loss": 0.0974,
      "step": 660
    },
    {
      "epoch": 4.294871794871795,
      "grad_norm": 8.05540657043457,
      "learning_rate": 0.0001141025641025641,
      "loss": 0.0466,
      "step": 670
    },
    {
      "epoch": 4.358974358974359,
      "grad_norm": 0.1012343019247055,
      "learning_rate": 0.00011282051282051283,
      "loss": 0.0641,
      "step": 680
    },
    {
      "epoch": 4.423076923076923,
      "grad_norm": 7.817044734954834,
      "learning_rate": 0.00011153846153846154,
      "loss": 0.1442,
      "step": 690
    },
    {
      "epoch": 4.487179487179487,
      "grad_norm": 6.788941860198975,
      "learning_rate": 0.00011025641025641027,
      "loss": 0.0296,
      "step": 700
    },
    {
      "epoch": 4.487179487179487,
      "eval_accuracy": 0.8884892086330936,
      "eval_loss": 0.4233308434486389,
      "eval_runtime": 2.523,
      "eval_samples_per_second": 110.184,
      "eval_steps_per_second": 13.872,
      "step": 700
    },
    {
      "epoch": 4.551282051282051,
      "grad_norm": 4.860511302947998,
      "learning_rate": 0.00010897435897435896,
      "loss": 0.0231,
      "step": 710
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 0.9598804116249084,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.0207,
      "step": 720
    },
    {
      "epoch": 4.67948717948718,
      "grad_norm": 12.745481491088867,
      "learning_rate": 0.00010641025641025641,
      "loss": 0.1516,
      "step": 730
    },
    {
      "epoch": 4.743589743589744,
      "grad_norm": 7.983795166015625,
      "learning_rate": 0.00010512820512820514,
      "loss": 0.1638,
      "step": 740
    },
    {
      "epoch": 4.8076923076923075,
      "grad_norm": 3.051384449005127,
      "learning_rate": 0.00010384615384615386,
      "loss": 0.0477,
      "step": 750
    },
    {
      "epoch": 4.871794871794872,
      "grad_norm": 0.10625698417425156,
      "learning_rate": 0.00010256410256410256,
      "loss": 0.0719,
      "step": 760
    },
    {
      "epoch": 4.935897435897436,
      "grad_norm": 0.04624614119529724,
      "learning_rate": 0.00010128205128205129,
      "loss": 0.1069,
      "step": 770
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.08277003467082977,
      "learning_rate": 0.0001,
      "loss": 0.0152,
      "step": 780
    },
    {
      "epoch": 5.064102564102564,
      "grad_norm": 0.09980784356594086,
      "learning_rate": 9.871794871794872e-05,
      "loss": 0.0719,
      "step": 790
    },
    {
      "epoch": 5.128205128205128,
      "grad_norm": 0.09162779897451401,
      "learning_rate": 9.743589743589744e-05,
      "loss": 0.0723,
      "step": 800
    },
    {
      "epoch": 5.128205128205128,
      "eval_accuracy": 0.8848920863309353,
      "eval_loss": 0.4469863176345825,
      "eval_runtime": 2.7699,
      "eval_samples_per_second": 100.363,
      "eval_steps_per_second": 12.636,
      "step": 800
    },
    {
      "epoch": 5.1923076923076925,
      "grad_norm": 0.05171818658709526,
      "learning_rate": 9.615384615384617e-05,
      "loss": 0.0192,
      "step": 810
    },
    {
      "epoch": 5.256410256410256,
      "grad_norm": 0.05209165811538696,
      "learning_rate": 9.487179487179487e-05,
      "loss": 0.0394,
      "step": 820
    },
    {
      "epoch": 5.32051282051282,
      "grad_norm": 0.960054874420166,
      "learning_rate": 9.35897435897436e-05,
      "loss": 0.0129,
      "step": 830
    },
    {
      "epoch": 5.384615384615385,
      "grad_norm": 0.09233374148607254,
      "learning_rate": 9.230769230769232e-05,
      "loss": 0.0138,
      "step": 840
    },
    {
      "epoch": 5.448717948717949,
      "grad_norm": 0.09635169804096222,
      "learning_rate": 9.102564102564103e-05,
      "loss": 0.0096,
      "step": 850
    },
    {
      "epoch": 5.512820512820513,
      "grad_norm": 1.3777004480361938,
      "learning_rate": 8.974358974358975e-05,
      "loss": 0.0412,
      "step": 860
    },
    {
      "epoch": 5.576923076923077,
      "grad_norm": 0.03339802846312523,
      "learning_rate": 8.846153846153847e-05,
      "loss": 0.0424,
      "step": 870
    },
    {
      "epoch": 5.641025641025641,
      "grad_norm": 0.032307617366313934,
      "learning_rate": 8.717948717948718e-05,
      "loss": 0.0161,
      "step": 880
    },
    {
      "epoch": 5.705128205128205,
      "grad_norm": 0.03049471788108349,
      "learning_rate": 8.58974358974359e-05,
      "loss": 0.0388,
      "step": 890
    },
    {
      "epoch": 5.769230769230769,
      "grad_norm": 0.05182625725865364,
      "learning_rate": 8.461538461538461e-05,
      "loss": 0.0342,
      "step": 900
    },
    {
      "epoch": 5.769230769230769,
      "eval_accuracy": 0.9172661870503597,
      "eval_loss": 0.3406282067298889,
      "eval_runtime": 2.3863,
      "eval_samples_per_second": 116.5,
      "eval_steps_per_second": 14.667,
      "step": 900
    },
    {
      "epoch": 5.833333333333333,
      "grad_norm": 0.2365674525499344,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.0837,
      "step": 910
    },
    {
      "epoch": 5.897435897435898,
      "grad_norm": 0.031284429132938385,
      "learning_rate": 8.205128205128205e-05,
      "loss": 0.0691,
      "step": 920
    },
    {
      "epoch": 5.961538461538462,
      "grad_norm": 10.787687301635742,
      "learning_rate": 8.076923076923078e-05,
      "loss": 0.0524,
      "step": 930
    },
    {
      "epoch": 6.0256410256410255,
      "grad_norm": 0.027590090408921242,
      "learning_rate": 7.948717948717948e-05,
      "loss": 0.0086,
      "step": 940
    },
    {
      "epoch": 6.089743589743589,
      "grad_norm": 0.04675084725022316,
      "learning_rate": 7.820512820512821e-05,
      "loss": 0.0066,
      "step": 950
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 0.032889507710933685,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.0432,
      "step": 960
    },
    {
      "epoch": 6.217948717948718,
      "grad_norm": 0.1580750048160553,
      "learning_rate": 7.564102564102564e-05,
      "loss": 0.006,
      "step": 970
    },
    {
      "epoch": 6.282051282051282,
      "grad_norm": 0.024286190047860146,
      "learning_rate": 7.435897435897436e-05,
      "loss": 0.0123,
      "step": 980
    },
    {
      "epoch": 6.346153846153846,
      "grad_norm": 0.02685542032122612,
      "learning_rate": 7.307692307692307e-05,
      "loss": 0.0059,
      "step": 990
    },
    {
      "epoch": 6.410256410256411,
      "grad_norm": 0.9080101251602173,
      "learning_rate": 7.17948717948718e-05,
      "loss": 0.0055,
      "step": 1000
    },
    {
      "epoch": 6.410256410256411,
      "eval_accuracy": 0.920863309352518,
      "eval_loss": 0.330095499753952,
      "eval_runtime": 3.1626,
      "eval_samples_per_second": 87.904,
      "eval_steps_per_second": 11.067,
      "step": 1000
    },
    {
      "epoch": 6.4743589743589745,
      "grad_norm": 0.156268909573555,
      "learning_rate": 7.051282051282052e-05,
      "loss": 0.0051,
      "step": 1010
    },
    {
      "epoch": 6.538461538461538,
      "grad_norm": 0.025522593408823013,
      "learning_rate": 6.923076923076924e-05,
      "loss": 0.0175,
      "step": 1020
    },
    {
      "epoch": 6.602564102564102,
      "grad_norm": 0.025892965495586395,
      "learning_rate": 6.794871794871795e-05,
      "loss": 0.0133,
      "step": 1030
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.02324897050857544,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.0051,
      "step": 1040
    },
    {
      "epoch": 6.730769230769231,
      "grad_norm": 0.20136423408985138,
      "learning_rate": 6.538461538461539e-05,
      "loss": 0.0318,
      "step": 1050
    },
    {
      "epoch": 6.794871794871795,
      "grad_norm": 0.11247438937425613,
      "learning_rate": 6.410256410256412e-05,
      "loss": 0.0331,
      "step": 1060
    },
    {
      "epoch": 6.858974358974359,
      "grad_norm": 0.10950164496898651,
      "learning_rate": 6.282051282051282e-05,
      "loss": 0.0055,
      "step": 1070
    },
    {
      "epoch": 6.923076923076923,
      "grad_norm": 1.537802815437317,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0055,
      "step": 1080
    },
    {
      "epoch": 6.987179487179487,
      "grad_norm": 0.023923929780721664,
      "learning_rate": 6.025641025641026e-05,
      "loss": 0.0044,
      "step": 1090
    },
    {
      "epoch": 7.051282051282051,
      "grad_norm": 0.02083686552941799,
      "learning_rate": 5.897435897435898e-05,
      "loss": 0.0048,
      "step": 1100
    },
    {
      "epoch": 7.051282051282051,
      "eval_accuracy": 0.9172661870503597,
      "eval_loss": 0.3471122980117798,
      "eval_runtime": 2.4758,
      "eval_samples_per_second": 112.287,
      "eval_steps_per_second": 14.137,
      "step": 1100
    },
    {
      "epoch": 7.115384615384615,
      "grad_norm": 0.020538046956062317,
      "learning_rate": 5.769230769230769e-05,
      "loss": 0.0042,
      "step": 1110
    },
    {
      "epoch": 7.17948717948718,
      "grad_norm": 0.01733437366783619,
      "learning_rate": 5.6410256410256414e-05,
      "loss": 0.0039,
      "step": 1120
    },
    {
      "epoch": 7.243589743589744,
      "grad_norm": 0.01968984678387642,
      "learning_rate": 5.512820512820514e-05,
      "loss": 0.0038,
      "step": 1130
    },
    {
      "epoch": 7.3076923076923075,
      "grad_norm": 0.019213447347283363,
      "learning_rate": 5.384615384615385e-05,
      "loss": 0.0036,
      "step": 1140
    },
    {
      "epoch": 7.371794871794872,
      "grad_norm": 0.017935629934072495,
      "learning_rate": 5.256410256410257e-05,
      "loss": 0.004,
      "step": 1150
    },
    {
      "epoch": 7.435897435897436,
      "grad_norm": 0.01726532354950905,
      "learning_rate": 5.128205128205128e-05,
      "loss": 0.0038,
      "step": 1160
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.01753012090921402,
      "learning_rate": 5e-05,
      "loss": 0.0038,
      "step": 1170
    },
    {
      "epoch": 7.564102564102564,
      "grad_norm": 0.018105851486325264,
      "learning_rate": 4.871794871794872e-05,
      "loss": 0.0036,
      "step": 1180
    },
    {
      "epoch": 7.628205128205128,
      "grad_norm": 0.019911447539925575,
      "learning_rate": 4.7435897435897435e-05,
      "loss": 0.0037,
      "step": 1190
    },
    {
      "epoch": 7.6923076923076925,
      "grad_norm": 0.023634430021047592,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.0036,
      "step": 1200
    },
    {
      "epoch": 7.6923076923076925,
      "eval_accuracy": 0.9136690647482014,
      "eval_loss": 0.33460894227027893,
      "eval_runtime": 2.4621,
      "eval_samples_per_second": 112.91,
      "eval_steps_per_second": 14.215,
      "step": 1200
    },
    {
      "epoch": 7.756410256410256,
      "grad_norm": 0.017936883494257927,
      "learning_rate": 4.4871794871794874e-05,
      "loss": 0.0034,
      "step": 1210
    },
    {
      "epoch": 7.82051282051282,
      "grad_norm": 0.01885095238685608,
      "learning_rate": 4.358974358974359e-05,
      "loss": 0.0034,
      "step": 1220
    },
    {
      "epoch": 7.884615384615385,
      "grad_norm": 0.017711780965328217,
      "learning_rate": 4.230769230769231e-05,
      "loss": 0.0033,
      "step": 1230
    },
    {
      "epoch": 7.948717948717949,
      "grad_norm": 0.014750463888049126,
      "learning_rate": 4.1025641025641023e-05,
      "loss": 0.0034,
      "step": 1240
    },
    {
      "epoch": 8.012820512820513,
      "grad_norm": 0.014598443172872066,
      "learning_rate": 3.974358974358974e-05,
      "loss": 0.0031,
      "step": 1250
    },
    {
      "epoch": 8.076923076923077,
      "grad_norm": 0.01595359854400158,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.0032,
      "step": 1260
    },
    {
      "epoch": 8.14102564102564,
      "grad_norm": 0.01710698939859867,
      "learning_rate": 3.717948717948718e-05,
      "loss": 0.0032,
      "step": 1270
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 0.015550950542092323,
      "learning_rate": 3.58974358974359e-05,
      "loss": 0.0031,
      "step": 1280
    },
    {
      "epoch": 8.26923076923077,
      "grad_norm": 0.015512553043663502,
      "learning_rate": 3.461538461538462e-05,
      "loss": 0.0031,
      "step": 1290
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 0.01687728427350521,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.003,
      "step": 1300
    },
    {
      "epoch": 8.333333333333334,
      "eval_accuracy": 0.9136690647482014,
      "eval_loss": 0.34976544976234436,
      "eval_runtime": 2.3973,
      "eval_samples_per_second": 115.962,
      "eval_steps_per_second": 14.599,
      "step": 1300
    },
    {
      "epoch": 8.397435897435898,
      "grad_norm": 0.017616627737879753,
      "learning_rate": 3.205128205128206e-05,
      "loss": 0.0031,
      "step": 1310
    },
    {
      "epoch": 8.461538461538462,
      "grad_norm": 0.017452212050557137,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 0.0031,
      "step": 1320
    },
    {
      "epoch": 8.525641025641026,
      "grad_norm": 0.017976053059101105,
      "learning_rate": 2.948717948717949e-05,
      "loss": 0.0032,
      "step": 1330
    },
    {
      "epoch": 8.58974358974359,
      "grad_norm": 0.014091568998992443,
      "learning_rate": 2.8205128205128207e-05,
      "loss": 0.0032,
      "step": 1340
    },
    {
      "epoch": 8.653846153846153,
      "grad_norm": 0.015703728422522545,
      "learning_rate": 2.6923076923076923e-05,
      "loss": 0.003,
      "step": 1350
    },
    {
      "epoch": 8.717948717948717,
      "grad_norm": 0.01781061850488186,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.003,
      "step": 1360
    },
    {
      "epoch": 8.782051282051283,
      "grad_norm": 0.01647392474114895,
      "learning_rate": 2.435897435897436e-05,
      "loss": 0.0031,
      "step": 1370
    },
    {
      "epoch": 8.846153846153847,
      "grad_norm": 0.013498615473508835,
      "learning_rate": 2.307692307692308e-05,
      "loss": 0.0028,
      "step": 1380
    },
    {
      "epoch": 8.91025641025641,
      "grad_norm": 0.016175739467144012,
      "learning_rate": 2.1794871794871795e-05,
      "loss": 0.003,
      "step": 1390
    },
    {
      "epoch": 8.974358974358974,
      "grad_norm": 0.015950603410601616,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 0.003,
      "step": 1400
    },
    {
      "epoch": 8.974358974358974,
      "eval_accuracy": 0.9100719424460432,
      "eval_loss": 0.354926198720932,
      "eval_runtime": 2.4324,
      "eval_samples_per_second": 114.29,
      "eval_steps_per_second": 14.389,
      "step": 1400
    },
    {
      "epoch": 9.038461538461538,
      "grad_norm": 0.014547958970069885,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.0029,
      "step": 1410
    },
    {
      "epoch": 9.102564102564102,
      "grad_norm": 0.012941875495016575,
      "learning_rate": 1.794871794871795e-05,
      "loss": 0.0028,
      "step": 1420
    },
    {
      "epoch": 9.166666666666666,
      "grad_norm": 0.016635097563266754,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.003,
      "step": 1430
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 0.018657604232430458,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.0029,
      "step": 1440
    },
    {
      "epoch": 9.294871794871796,
      "grad_norm": 0.015006115660071373,
      "learning_rate": 1.4102564102564104e-05,
      "loss": 0.0031,
      "step": 1450
    },
    {
      "epoch": 9.35897435897436,
      "grad_norm": 0.01575641520321369,
      "learning_rate": 1.282051282051282e-05,
      "loss": 0.0027,
      "step": 1460
    },
    {
      "epoch": 9.423076923076923,
      "grad_norm": 0.013228046707808971,
      "learning_rate": 1.153846153846154e-05,
      "loss": 0.0027,
      "step": 1470
    },
    {
      "epoch": 9.487179487179487,
      "grad_norm": 0.013002932071685791,
      "learning_rate": 1.0256410256410256e-05,
      "loss": 0.0027,
      "step": 1480
    },
    {
      "epoch": 9.551282051282051,
      "grad_norm": 0.014644928276538849,
      "learning_rate": 8.974358974358976e-06,
      "loss": 0.0029,
      "step": 1490
    },
    {
      "epoch": 9.615384615384615,
      "grad_norm": 0.01448275987058878,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.0027,
      "step": 1500
    },
    {
      "epoch": 9.615384615384615,
      "eval_accuracy": 0.9136690647482014,
      "eval_loss": 0.3569168150424957,
      "eval_runtime": 2.7846,
      "eval_samples_per_second": 99.835,
      "eval_steps_per_second": 12.569,
      "step": 1500
    },
    {
      "epoch": 9.679487179487179,
      "grad_norm": 0.012788016349077225,
      "learning_rate": 6.41025641025641e-06,
      "loss": 0.0028,
      "step": 1510
    },
    {
      "epoch": 9.743589743589745,
      "grad_norm": 0.014576306566596031,
      "learning_rate": 5.128205128205128e-06,
      "loss": 0.0026,
      "step": 1520
    },
    {
      "epoch": 9.807692307692308,
      "grad_norm": 0.014699741266667843,
      "learning_rate": 3.846153846153847e-06,
      "loss": 0.0029,
      "step": 1530
    },
    {
      "epoch": 9.871794871794872,
      "grad_norm": 0.015379097312688828,
      "learning_rate": 2.564102564102564e-06,
      "loss": 0.0026,
      "step": 1540
    },
    {
      "epoch": 9.935897435897436,
      "grad_norm": 0.012021095491945744,
      "learning_rate": 1.282051282051282e-06,
      "loss": 0.0027,
      "step": 1550
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.014515766873955727,
      "learning_rate": 0.0,
      "loss": 0.0028,
      "step": 1560
    },
    {
      "epoch": 10.0,
      "step": 1560,
      "total_flos": 1.9334597982400512e+18,
      "train_loss": 0.20911514231314263,
      "train_runtime": 603.1201,
      "train_samples_per_second": 41.368,
      "train_steps_per_second": 2.587
    }
  ],
  "logging_steps": 10,
  "max_steps": 1560,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.9334597982400512e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}