{
  "best_metric": 0.9340658097220965,
  "best_model_checkpoint": "batch-size-16_FFPP-c23_1FPS_faces-expand-0-aligned_unaugmentation\\checkpoint-1381",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1381,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 3.2925620079040527,
      "learning_rate": 3.5971223021582732e-06,
      "loss": 0.628,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.3982741832733154,
      "learning_rate": 7.1942446043165465e-06,
      "loss": 0.5814,
      "step": 20
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.776655673980713,
      "learning_rate": 1.0791366906474821e-05,
      "loss": 0.5151,
      "step": 30
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.050455331802368,
      "learning_rate": 1.4388489208633093e-05,
      "loss": 0.5308,
      "step": 40
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.005922794342041,
      "learning_rate": 1.7985611510791367e-05,
      "loss": 0.5363,
      "step": 50
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.7546412944793701,
      "learning_rate": 2.1582733812949642e-05,
      "loss": 0.4718,
      "step": 60
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.74489426612854,
      "learning_rate": 2.5179856115107914e-05,
      "loss": 0.5125,
      "step": 70
    },
    {
      "epoch": 0.06,
      "grad_norm": 2.5463011264801025,
      "learning_rate": 2.8776978417266186e-05,
      "loss": 0.505,
      "step": 80
    },
    {
      "epoch": 0.07,
      "grad_norm": 5.876346588134766,
      "learning_rate": 3.237410071942446e-05,
      "loss": 0.5294,
      "step": 90
    },
    {
      "epoch": 0.07,
      "grad_norm": 3.9089746475219727,
      "learning_rate": 3.597122302158273e-05,
      "loss": 0.5465,
      "step": 100
    },
    {
      "epoch": 0.08,
      "grad_norm": 5.623407363891602,
      "learning_rate": 3.956834532374101e-05,
      "loss": 0.5042,
      "step": 110
    },
    {
      "epoch": 0.09,
      "grad_norm": 4.3956780433654785,
      "learning_rate": 4.3165467625899284e-05,
      "loss": 0.5046,
      "step": 120
    },
    {
      "epoch": 0.09,
      "grad_norm": 4.22213077545166,
      "learning_rate": 4.676258992805755e-05,
      "loss": 0.4896,
      "step": 130
    },
    {
      "epoch": 0.1,
      "grad_norm": 5.061640739440918,
      "learning_rate": 4.99597423510467e-05,
      "loss": 0.4615,
      "step": 140
    },
    {
      "epoch": 0.11,
      "grad_norm": 3.777334213256836,
      "learning_rate": 4.9557165861513685e-05,
      "loss": 0.4725,
      "step": 150
    },
    {
      "epoch": 0.12,
      "grad_norm": 8.781298637390137,
      "learning_rate": 4.915458937198068e-05,
      "loss": 0.4155,
      "step": 160
    },
    {
      "epoch": 0.12,
      "grad_norm": 6.117956638336182,
      "learning_rate": 4.875201288244767e-05,
      "loss": 0.4209,
      "step": 170
    },
    {
      "epoch": 0.13,
      "grad_norm": 3.1794047355651855,
      "learning_rate": 4.834943639291466e-05,
      "loss": 0.4149,
      "step": 180
    },
    {
      "epoch": 0.14,
      "grad_norm": 4.002980709075928,
      "learning_rate": 4.7946859903381646e-05,
      "loss": 0.5147,
      "step": 190
    },
    {
      "epoch": 0.14,
      "grad_norm": 2.6848464012145996,
      "learning_rate": 4.7544283413848634e-05,
      "loss": 0.4649,
      "step": 200
    },
    {
      "epoch": 0.15,
      "grad_norm": 5.1571807861328125,
      "learning_rate": 4.714170692431562e-05,
      "loss": 0.4018,
      "step": 210
    },
    {
      "epoch": 0.16,
      "grad_norm": 7.3869309425354,
      "learning_rate": 4.673913043478261e-05,
      "loss": 0.3477,
      "step": 220
    },
    {
      "epoch": 0.17,
      "grad_norm": 5.493707656860352,
      "learning_rate": 4.63365539452496e-05,
      "loss": 0.3752,
      "step": 230
    },
    {
      "epoch": 0.17,
      "grad_norm": 6.3577070236206055,
      "learning_rate": 4.593397745571659e-05,
      "loss": 0.3779,
      "step": 240
    },
    {
      "epoch": 0.18,
      "grad_norm": 6.673830986022949,
      "learning_rate": 4.553140096618358e-05,
      "loss": 0.4141,
      "step": 250
    },
    {
      "epoch": 0.19,
      "grad_norm": 3.704331636428833,
      "learning_rate": 4.5128824476650565e-05,
      "loss": 0.3926,
      "step": 260
    },
    {
      "epoch": 0.2,
      "grad_norm": 5.406624794006348,
      "learning_rate": 4.4726247987117554e-05,
      "loss": 0.3742,
      "step": 270
    },
    {
      "epoch": 0.2,
      "grad_norm": 5.6158528327941895,
      "learning_rate": 4.432367149758454e-05,
      "loss": 0.3715,
      "step": 280
    },
    {
      "epoch": 0.21,
      "grad_norm": 5.550079345703125,
      "learning_rate": 4.392109500805153e-05,
      "loss": 0.3882,
      "step": 290
    },
    {
      "epoch": 0.22,
      "grad_norm": 5.96751594543457,
      "learning_rate": 4.351851851851852e-05,
      "loss": 0.3615,
      "step": 300
    },
    {
      "epoch": 0.22,
      "grad_norm": 5.48460054397583,
      "learning_rate": 4.3115942028985515e-05,
      "loss": 0.3233,
      "step": 310
    },
    {
      "epoch": 0.23,
      "grad_norm": 5.41163969039917,
      "learning_rate": 4.2713365539452496e-05,
      "loss": 0.3723,
      "step": 320
    },
    {
      "epoch": 0.24,
      "grad_norm": 6.076368808746338,
      "learning_rate": 4.2310789049919485e-05,
      "loss": 0.3684,
      "step": 330
    },
    {
      "epoch": 0.25,
      "grad_norm": 9.160358428955078,
      "learning_rate": 4.1908212560386474e-05,
      "loss": 0.3381,
      "step": 340
    },
    {
      "epoch": 0.25,
      "grad_norm": 5.847888469696045,
      "learning_rate": 4.150563607085346e-05,
      "loss": 0.3669,
      "step": 350
    },
    {
      "epoch": 0.26,
      "grad_norm": 9.13357925415039,
      "learning_rate": 4.110305958132045e-05,
      "loss": 0.3717,
      "step": 360
    },
    {
      "epoch": 0.27,
      "grad_norm": 6.672016143798828,
      "learning_rate": 4.070048309178744e-05,
      "loss": 0.3121,
      "step": 370
    },
    {
      "epoch": 0.28,
      "grad_norm": 5.612246513366699,
      "learning_rate": 4.0297906602254434e-05,
      "loss": 0.3004,
      "step": 380
    },
    {
      "epoch": 0.28,
      "grad_norm": 4.514873027801514,
      "learning_rate": 3.9895330112721416e-05,
      "loss": 0.3153,
      "step": 390
    },
    {
      "epoch": 0.29,
      "grad_norm": 4.649542331695557,
      "learning_rate": 3.9492753623188405e-05,
      "loss": 0.3332,
      "step": 400
    },
    {
      "epoch": 0.3,
      "grad_norm": 8.960000038146973,
      "learning_rate": 3.90901771336554e-05,
      "loss": 0.3117,
      "step": 410
    },
    {
      "epoch": 0.3,
      "grad_norm": 4.450685977935791,
      "learning_rate": 3.868760064412238e-05,
      "loss": 0.2995,
      "step": 420
    },
    {
      "epoch": 0.31,
      "grad_norm": 7.965936183929443,
      "learning_rate": 3.828502415458937e-05,
      "loss": 0.3106,
      "step": 430
    },
    {
      "epoch": 0.32,
      "grad_norm": 5.865048885345459,
      "learning_rate": 3.7882447665056365e-05,
      "loss": 0.3165,
      "step": 440
    },
    {
      "epoch": 0.33,
      "grad_norm": 6.963225841522217,
      "learning_rate": 3.7479871175523354e-05,
      "loss": 0.318,
      "step": 450
    },
    {
      "epoch": 0.33,
      "grad_norm": 6.650681018829346,
      "learning_rate": 3.7077294685990336e-05,
      "loss": 0.3081,
      "step": 460
    },
    {
      "epoch": 0.34,
      "grad_norm": 4.959587574005127,
      "learning_rate": 3.667471819645733e-05,
      "loss": 0.3109,
      "step": 470
    },
    {
      "epoch": 0.35,
      "grad_norm": 5.5469183921813965,
      "learning_rate": 3.627214170692432e-05,
      "loss": 0.2793,
      "step": 480
    },
    {
      "epoch": 0.35,
      "grad_norm": 10.897783279418945,
      "learning_rate": 3.58695652173913e-05,
      "loss": 0.2984,
      "step": 490
    },
    {
      "epoch": 0.36,
      "grad_norm": 5.365045070648193,
      "learning_rate": 3.5466988727858296e-05,
      "loss": 0.2797,
      "step": 500
    },
    {
      "epoch": 0.37,
      "grad_norm": 7.85481071472168,
      "learning_rate": 3.5064412238325285e-05,
      "loss": 0.3062,
      "step": 510
    },
    {
      "epoch": 0.38,
      "grad_norm": 5.305897235870361,
      "learning_rate": 3.4661835748792274e-05,
      "loss": 0.3354,
      "step": 520
    },
    {
      "epoch": 0.38,
      "grad_norm": 4.42397403717041,
      "learning_rate": 3.425925925925926e-05,
      "loss": 0.2454,
      "step": 530
    },
    {
      "epoch": 0.39,
      "grad_norm": 3.9212090969085693,
      "learning_rate": 3.385668276972625e-05,
      "loss": 0.2564,
      "step": 540
    },
    {
      "epoch": 0.4,
      "grad_norm": 7.084193229675293,
      "learning_rate": 3.345410628019324e-05,
      "loss": 0.2709,
      "step": 550
    },
    {
      "epoch": 0.41,
      "grad_norm": 6.757343292236328,
      "learning_rate": 3.305152979066023e-05,
      "loss": 0.3131,
      "step": 560
    },
    {
      "epoch": 0.41,
      "grad_norm": 8.7233304977417,
      "learning_rate": 3.2648953301127216e-05,
      "loss": 0.2729,
      "step": 570
    },
    {
      "epoch": 0.42,
      "grad_norm": 6.0495100021362305,
      "learning_rate": 3.2246376811594205e-05,
      "loss": 0.2555,
      "step": 580
    },
    {
      "epoch": 0.43,
      "grad_norm": 9.789109230041504,
      "learning_rate": 3.184380032206119e-05,
      "loss": 0.2636,
      "step": 590
    },
    {
      "epoch": 0.43,
      "grad_norm": 8.58854866027832,
      "learning_rate": 3.144122383252818e-05,
      "loss": 0.2687,
      "step": 600
    },
    {
      "epoch": 0.44,
      "grad_norm": 9.97248363494873,
      "learning_rate": 3.103864734299517e-05,
      "loss": 0.3013,
      "step": 610
    },
    {
      "epoch": 0.45,
      "grad_norm": 25.611948013305664,
      "learning_rate": 3.063607085346216e-05,
      "loss": 0.3343,
      "step": 620
    },
    {
      "epoch": 0.46,
      "grad_norm": 5.729062080383301,
      "learning_rate": 3.023349436392915e-05,
      "loss": 0.3457,
      "step": 630
    },
    {
      "epoch": 0.46,
      "grad_norm": 7.5272216796875,
      "learning_rate": 2.9830917874396136e-05,
      "loss": 0.2456,
      "step": 640
    },
    {
      "epoch": 0.47,
      "grad_norm": 7.43132209777832,
      "learning_rate": 2.9428341384863124e-05,
      "loss": 0.3469,
      "step": 650
    },
    {
      "epoch": 0.48,
      "grad_norm": 7.458743572235107,
      "learning_rate": 2.9025764895330116e-05,
      "loss": 0.2884,
      "step": 660
    },
    {
      "epoch": 0.49,
      "grad_norm": 6.541900634765625,
      "learning_rate": 2.86231884057971e-05,
      "loss": 0.2466,
      "step": 670
    },
    {
      "epoch": 0.49,
      "grad_norm": 7.272885322570801,
      "learning_rate": 2.822061191626409e-05,
      "loss": 0.2345,
      "step": 680
    },
    {
      "epoch": 0.5,
      "grad_norm": 10.58638858795166,
      "learning_rate": 2.781803542673108e-05,
      "loss": 0.2519,
      "step": 690
    },
    {
      "epoch": 0.51,
      "grad_norm": 5.776723384857178,
      "learning_rate": 2.741545893719807e-05,
      "loss": 0.3001,
      "step": 700
    },
    {
      "epoch": 0.51,
      "grad_norm": 5.556161880493164,
      "learning_rate": 2.7012882447665055e-05,
      "loss": 0.2799,
      "step": 710
    },
    {
      "epoch": 0.52,
      "grad_norm": 5.732090950012207,
      "learning_rate": 2.6610305958132047e-05,
      "loss": 0.2197,
      "step": 720
    },
    {
      "epoch": 0.53,
      "grad_norm": 17.775867462158203,
      "learning_rate": 2.6207729468599036e-05,
      "loss": 0.2938,
      "step": 730
    },
    {
      "epoch": 0.54,
      "grad_norm": 19.362329483032227,
      "learning_rate": 2.580515297906602e-05,
      "loss": 0.2327,
      "step": 740
    },
    {
      "epoch": 0.54,
      "grad_norm": 8.213146209716797,
      "learning_rate": 2.5402576489533013e-05,
      "loss": 0.2142,
      "step": 750
    },
    {
      "epoch": 0.55,
      "grad_norm": 9.953818321228027,
      "learning_rate": 2.5e-05,
      "loss": 0.2403,
      "step": 760
    },
    {
      "epoch": 0.56,
      "grad_norm": 7.865139007568359,
      "learning_rate": 2.459742351046699e-05,
      "loss": 0.2481,
      "step": 770
    },
    {
      "epoch": 0.56,
      "grad_norm": 6.347203254699707,
      "learning_rate": 2.4194847020933978e-05,
      "loss": 0.2487,
      "step": 780
    },
    {
      "epoch": 0.57,
      "grad_norm": 7.396281719207764,
      "learning_rate": 2.3792270531400967e-05,
      "loss": 0.2752,
      "step": 790
    },
    {
      "epoch": 0.58,
      "grad_norm": 6.594508647918701,
      "learning_rate": 2.338969404186796e-05,
      "loss": 0.2389,
      "step": 800
    },
    {
      "epoch": 0.59,
      "grad_norm": 7.60775089263916,
      "learning_rate": 2.2987117552334944e-05,
      "loss": 0.2621,
      "step": 810
    },
    {
      "epoch": 0.59,
      "grad_norm": 13.000992774963379,
      "learning_rate": 2.2584541062801932e-05,
      "loss": 0.2691,
      "step": 820
    },
    {
      "epoch": 0.6,
      "grad_norm": 11.92383098602295,
      "learning_rate": 2.2181964573268924e-05,
      "loss": 0.2484,
      "step": 830
    },
    {
      "epoch": 0.61,
      "grad_norm": 10.424347877502441,
      "learning_rate": 2.177938808373591e-05,
      "loss": 0.2363,
      "step": 840
    },
    {
      "epoch": 0.62,
      "grad_norm": 6.517064094543457,
      "learning_rate": 2.13768115942029e-05,
      "loss": 0.2489,
      "step": 850
    },
    {
      "epoch": 0.62,
      "grad_norm": 6.191581726074219,
      "learning_rate": 2.0974235104669886e-05,
      "loss": 0.2632,
      "step": 860
    },
    {
      "epoch": 0.63,
      "grad_norm": 7.870666027069092,
      "learning_rate": 2.0571658615136878e-05,
      "loss": 0.2527,
      "step": 870
    },
    {
      "epoch": 0.64,
      "grad_norm": 7.662616729736328,
      "learning_rate": 2.0169082125603867e-05,
      "loss": 0.2355,
      "step": 880
    },
    {
      "epoch": 0.64,
      "grad_norm": 11.486822128295898,
      "learning_rate": 1.9766505636070852e-05,
      "loss": 0.2169,
      "step": 890
    },
    {
      "epoch": 0.65,
      "grad_norm": 12.382376670837402,
      "learning_rate": 1.9363929146537844e-05,
      "loss": 0.2483,
      "step": 900
    },
    {
      "epoch": 0.66,
      "grad_norm": 10.059280395507812,
      "learning_rate": 1.8961352657004832e-05,
      "loss": 0.2159,
      "step": 910
    },
    {
      "epoch": 0.67,
      "grad_norm": 18.096155166625977,
      "learning_rate": 1.855877616747182e-05,
      "loss": 0.2221,
      "step": 920
    },
    {
      "epoch": 0.67,
      "grad_norm": 14.859966278076172,
      "learning_rate": 1.815619967793881e-05,
      "loss": 0.2633,
      "step": 930
    },
    {
      "epoch": 0.68,
      "grad_norm": 10.2135648727417,
      "learning_rate": 1.7753623188405798e-05,
      "loss": 0.2817,
      "step": 940
    },
    {
      "epoch": 0.69,
      "grad_norm": 15.141898155212402,
      "learning_rate": 1.7351046698872786e-05,
      "loss": 0.2481,
      "step": 950
    },
    {
      "epoch": 0.7,
      "grad_norm": 11.335433959960938,
      "learning_rate": 1.6948470209339775e-05,
      "loss": 0.2236,
      "step": 960
    },
    {
      "epoch": 0.7,
      "grad_norm": 10.088510513305664,
      "learning_rate": 1.6545893719806767e-05,
      "loss": 0.2564,
      "step": 970
    },
    {
      "epoch": 0.71,
      "grad_norm": 11.138778686523438,
      "learning_rate": 1.6143317230273752e-05,
      "loss": 0.2301,
      "step": 980
    },
    {
      "epoch": 0.72,
      "grad_norm": 9.613357543945312,
      "learning_rate": 1.574074074074074e-05,
      "loss": 0.2344,
      "step": 990
    },
    {
      "epoch": 0.72,
      "grad_norm": 7.300370216369629,
      "learning_rate": 1.533816425120773e-05,
      "loss": 0.198,
      "step": 1000
    },
    {
      "epoch": 0.73,
      "grad_norm": 8.088990211486816,
      "learning_rate": 1.4935587761674719e-05,
      "loss": 0.2218,
      "step": 1010
    },
    {
      "epoch": 0.74,
      "grad_norm": 10.558765411376953,
      "learning_rate": 1.4533011272141708e-05,
      "loss": 0.2151,
      "step": 1020
    },
    {
      "epoch": 0.75,
      "grad_norm": 13.40772819519043,
      "learning_rate": 1.4130434782608694e-05,
      "loss": 0.2308,
      "step": 1030
    },
    {
      "epoch": 0.75,
      "grad_norm": 10.2852783203125,
      "learning_rate": 1.3727858293075685e-05,
      "loss": 0.195,
      "step": 1040
    },
    {
      "epoch": 0.76,
      "grad_norm": 10.811461448669434,
      "learning_rate": 1.3325281803542675e-05,
      "loss": 0.1728,
      "step": 1050
    },
    {
      "epoch": 0.77,
      "grad_norm": 6.249987602233887,
      "learning_rate": 1.2922705314009662e-05,
      "loss": 0.2358,
      "step": 1060
    },
    {
      "epoch": 0.77,
      "grad_norm": 8.414009094238281,
      "learning_rate": 1.2520128824476652e-05,
      "loss": 0.2149,
      "step": 1070
    },
    {
      "epoch": 0.78,
      "grad_norm": 7.3787689208984375,
      "learning_rate": 1.211755233494364e-05,
      "loss": 0.1711,
      "step": 1080
    },
    {
      "epoch": 0.79,
      "grad_norm": 8.621789932250977,
      "learning_rate": 1.1714975845410629e-05,
      "loss": 0.1723,
      "step": 1090
    },
    {
      "epoch": 0.8,
      "grad_norm": 7.638077735900879,
      "learning_rate": 1.1312399355877617e-05,
      "loss": 0.2104,
      "step": 1100
    },
    {
      "epoch": 0.8,
      "grad_norm": 13.404261589050293,
      "learning_rate": 1.0909822866344606e-05,
      "loss": 0.2187,
      "step": 1110
    },
    {
      "epoch": 0.81,
      "grad_norm": 15.54051399230957,
      "learning_rate": 1.0507246376811594e-05,
      "loss": 0.2342,
      "step": 1120
    },
    {
      "epoch": 0.82,
      "grad_norm": 9.454499244689941,
      "learning_rate": 1.0104669887278585e-05,
      "loss": 0.2579,
      "step": 1130
    },
    {
      "epoch": 0.83,
      "grad_norm": 8.479141235351562,
      "learning_rate": 9.702093397745571e-06,
      "loss": 0.1972,
      "step": 1140
    },
    {
      "epoch": 0.83,
      "grad_norm": 9.336498260498047,
      "learning_rate": 9.29951690821256e-06,
      "loss": 0.1805,
      "step": 1150
    },
    {
      "epoch": 0.84,
      "grad_norm": 7.811723709106445,
      "learning_rate": 8.89694041867955e-06,
      "loss": 0.2286,
      "step": 1160
    },
    {
      "epoch": 0.85,
      "grad_norm": 6.146000862121582,
      "learning_rate": 8.494363929146539e-06,
      "loss": 0.2133,
      "step": 1170
    },
    {
      "epoch": 0.85,
      "grad_norm": 10.112232208251953,
      "learning_rate": 8.091787439613527e-06,
      "loss": 0.2021,
      "step": 1180
    },
    {
      "epoch": 0.86,
      "grad_norm": 12.056436538696289,
      "learning_rate": 7.689210950080516e-06,
      "loss": 0.1791,
      "step": 1190
    },
    {
      "epoch": 0.87,
      "grad_norm": 9.647012710571289,
      "learning_rate": 7.286634460547505e-06,
      "loss": 0.1922,
      "step": 1200
    },
    {
      "epoch": 0.88,
      "grad_norm": 5.874572277069092,
      "learning_rate": 6.884057971014493e-06,
      "loss": 0.205,
      "step": 1210
    },
    {
      "epoch": 0.88,
      "grad_norm": 5.28275203704834,
      "learning_rate": 6.481481481481481e-06,
      "loss": 0.1951,
      "step": 1220
    },
    {
      "epoch": 0.89,
      "grad_norm": 13.48544979095459,
      "learning_rate": 6.078904991948471e-06,
      "loss": 0.1871,
      "step": 1230
    },
    {
      "epoch": 0.9,
      "grad_norm": 9.40962028503418,
      "learning_rate": 5.676328502415459e-06,
      "loss": 0.2078,
      "step": 1240
    },
    {
      "epoch": 0.91,
      "grad_norm": 5.173947811126709,
      "learning_rate": 5.273752012882448e-06,
      "loss": 0.1712,
      "step": 1250
    },
    {
      "epoch": 0.91,
      "grad_norm": 9.35919189453125,
      "learning_rate": 4.871175523349437e-06,
      "loss": 0.259,
      "step": 1260
    },
    {
      "epoch": 0.92,
      "grad_norm": 9.794632911682129,
      "learning_rate": 4.468599033816425e-06,
      "loss": 0.2173,
      "step": 1270
    },
    {
      "epoch": 0.93,
      "grad_norm": 8.138204574584961,
      "learning_rate": 4.066022544283414e-06,
      "loss": 0.1661,
      "step": 1280
    },
    {
      "epoch": 0.93,
      "grad_norm": 7.02229118347168,
      "learning_rate": 3.663446054750403e-06,
      "loss": 0.1961,
      "step": 1290
    },
    {
      "epoch": 0.94,
      "grad_norm": 8.913147926330566,
      "learning_rate": 3.2608695652173914e-06,
      "loss": 0.1956,
      "step": 1300
    },
    {
      "epoch": 0.95,
      "grad_norm": 7.670225620269775,
      "learning_rate": 2.85829307568438e-06,
      "loss": 0.1706,
      "step": 1310
    },
    {
      "epoch": 0.96,
      "grad_norm": 10.90648365020752,
      "learning_rate": 2.455716586151369e-06,
      "loss": 0.2055,
      "step": 1320
    },
    {
      "epoch": 0.96,
      "grad_norm": 8.81640911102295,
      "learning_rate": 2.053140096618358e-06,
      "loss": 0.176,
      "step": 1330
    },
    {
      "epoch": 0.97,
      "grad_norm": 7.678645610809326,
      "learning_rate": 1.6505636070853463e-06,
      "loss": 0.1897,
      "step": 1340
    },
    {
      "epoch": 0.98,
      "grad_norm": 9.090548515319824,
      "learning_rate": 1.247987117552335e-06,
      "loss": 0.2035,
      "step": 1350
    },
    {
      "epoch": 0.98,
      "grad_norm": 5.917918682098389,
      "learning_rate": 8.454106280193237e-07,
      "loss": 0.2102,
      "step": 1360
    },
    {
      "epoch": 0.99,
      "grad_norm": 8.879599571228027,
      "learning_rate": 4.428341384863124e-07,
      "loss": 0.2003,
      "step": 1370
    },
    {
      "epoch": 1.0,
      "grad_norm": 9.322664260864258,
      "learning_rate": 4.025764895330113e-08,
      "loss": 0.193,
      "step": 1380
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9340658097220965,
      "eval_f1": 0.9585606087544003,
      "eval_loss": 0.16026902198791504,
      "eval_precision": 0.9434178845400079,
      "eval_recall": 0.9741973720348661,
      "eval_roc_auc": 0.9777643781483449,
      "eval_runtime": 159.6047,
      "eval_samples_per_second": 553.718,
      "eval_steps_per_second": 34.611,
      "step": 1381
    },
    {
      "epoch": 1.0,
      "step": 1381,
      "total_flos": 2.1966728826140099e+18,
      "train_loss": 0.2970259432383502,
      "train_runtime": 756.4932,
      "train_samples_per_second": 116.823,
      "train_steps_per_second": 1.826
    }
  ],
  "logging_steps": 10,
  "max_steps": 1381,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 2.1966728826140099e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}