{
  "best_metric": 0.9781976744186046,
  "best_model_checkpoint": "mobilenet_v2_1.0_224-finetuned-plantdisease/checkpoint-2614",
  "epoch": 19.965576592082616,
  "eval_steps": 500,
  "global_step": 2900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06884681583476764,
      "grad_norm": 5.361344337463379,
      "learning_rate": 1.724137931034483e-06,
      "loss": 2.7576,
      "step": 10
    },
    {
      "epoch": 0.13769363166953527,
      "grad_norm": 5.03896427154541,
      "learning_rate": 3.448275862068966e-06,
      "loss": 2.7085,
      "step": 20
    },
    {
      "epoch": 0.20654044750430292,
      "grad_norm": 5.099004745483398,
      "learning_rate": 5.172413793103448e-06,
      "loss": 2.7048,
      "step": 30
    },
    {
      "epoch": 0.27538726333907054,
      "grad_norm": 5.294215202331543,
      "learning_rate": 6.896551724137932e-06,
      "loss": 2.6637,
      "step": 40
    },
    {
      "epoch": 0.3442340791738382,
      "grad_norm": 5.142210483551025,
      "learning_rate": 8.620689655172414e-06,
      "loss": 2.6127,
      "step": 50
    },
    {
      "epoch": 0.41308089500860584,
      "grad_norm": 5.095492839813232,
      "learning_rate": 1.0344827586206897e-05,
      "loss": 2.5458,
      "step": 60
    },
    {
      "epoch": 0.4819277108433735,
      "grad_norm": 4.894885063171387,
      "learning_rate": 1.206896551724138e-05,
      "loss": 2.4968,
      "step": 70
    },
    {
      "epoch": 0.5507745266781411,
      "grad_norm": 4.9816365242004395,
      "learning_rate": 1.3793103448275863e-05,
      "loss": 2.4015,
      "step": 80
    },
    {
      "epoch": 0.6196213425129088,
      "grad_norm": 4.671796798706055,
      "learning_rate": 1.5517241379310346e-05,
      "loss": 2.3308,
      "step": 90
    },
    {
      "epoch": 0.6884681583476764,
      "grad_norm": 4.635168552398682,
      "learning_rate": 1.7241379310344828e-05,
      "loss": 2.2212,
      "step": 100
    },
    {
      "epoch": 0.7573149741824441,
      "grad_norm": 4.575173854827881,
      "learning_rate": 1.896551724137931e-05,
      "loss": 2.1272,
      "step": 110
    },
    {
      "epoch": 0.8261617900172117,
      "grad_norm": 4.4043660163879395,
      "learning_rate": 2.0689655172413793e-05,
      "loss": 2.0085,
      "step": 120
    },
    {
      "epoch": 0.8950086058519794,
      "grad_norm": 4.405401229858398,
      "learning_rate": 2.2413793103448276e-05,
      "loss": 1.9442,
      "step": 130
    },
    {
      "epoch": 0.963855421686747,
      "grad_norm": 4.29350471496582,
      "learning_rate": 2.413793103448276e-05,
      "loss": 1.7982,
      "step": 140
    },
    {
      "epoch": 0.9982788296041308,
      "eval_accuracy": 0.40358527131782945,
      "eval_loss": 1.9824707508087158,
      "eval_runtime": 8.0421,
      "eval_samples_per_second": 256.65,
      "eval_steps_per_second": 8.082,
      "step": 145
    },
    {
      "epoch": 1.0327022375215147,
      "grad_norm": 4.268815040588379,
      "learning_rate": 2.5862068965517244e-05,
      "loss": 1.7064,
      "step": 150
    },
    {
      "epoch": 1.1015490533562822,
      "grad_norm": 4.16851806640625,
      "learning_rate": 2.7586206896551727e-05,
      "loss": 1.5684,
      "step": 160
    },
    {
      "epoch": 1.1703958691910499,
      "grad_norm": 4.322150230407715,
      "learning_rate": 2.9310344827586206e-05,
      "loss": 1.4844,
      "step": 170
    },
    {
      "epoch": 1.2392426850258176,
      "grad_norm": 3.958456039428711,
      "learning_rate": 3.103448275862069e-05,
      "loss": 1.3968,
      "step": 180
    },
    {
      "epoch": 1.3080895008605853,
      "grad_norm": 3.7061522006988525,
      "learning_rate": 3.275862068965517e-05,
      "loss": 1.2972,
      "step": 190
    },
    {
      "epoch": 1.3769363166953528,
      "grad_norm": 3.8323745727539062,
      "learning_rate": 3.4482758620689657e-05,
      "loss": 1.1544,
      "step": 200
    },
    {
      "epoch": 1.4457831325301205,
      "grad_norm": 3.52496600151062,
      "learning_rate": 3.620689655172414e-05,
      "loss": 1.0645,
      "step": 210
    },
    {
      "epoch": 1.5146299483648882,
      "grad_norm": 3.744525194168091,
      "learning_rate": 3.793103448275862e-05,
      "loss": 1.0238,
      "step": 220
    },
    {
      "epoch": 1.5834767641996557,
      "grad_norm": 3.494678497314453,
      "learning_rate": 3.965517241379311e-05,
      "loss": 0.9237,
      "step": 230
    },
    {
      "epoch": 1.6523235800344234,
      "grad_norm": 3.6277663707733154,
      "learning_rate": 4.1379310344827587e-05,
      "loss": 0.881,
      "step": 240
    },
    {
      "epoch": 1.721170395869191,
      "grad_norm": 2.9900975227355957,
      "learning_rate": 4.3103448275862066e-05,
      "loss": 0.7883,
      "step": 250
    },
    {
      "epoch": 1.7900172117039586,
      "grad_norm": 3.553596258163452,
      "learning_rate": 4.482758620689655e-05,
      "loss": 0.7713,
      "step": 260
    },
    {
      "epoch": 1.8588640275387265,
      "grad_norm": 3.206496477127075,
      "learning_rate": 4.655172413793104e-05,
      "loss": 0.7063,
      "step": 270
    },
    {
      "epoch": 1.927710843373494,
      "grad_norm": 3.446793556213379,
      "learning_rate": 4.827586206896552e-05,
      "loss": 0.6517,
      "step": 280
    },
    {
      "epoch": 1.9965576592082617,
      "grad_norm": 3.7154784202575684,
      "learning_rate": 5e-05,
      "loss": 0.6137,
      "step": 290
    },
    {
      "epoch": 1.9965576592082617,
      "eval_accuracy": 0.6414728682170543,
      "eval_loss": 1.1130067110061646,
      "eval_runtime": 7.6055,
      "eval_samples_per_second": 271.384,
      "eval_steps_per_second": 8.546,
      "step": 290
    },
    {
      "epoch": 2.0654044750430294,
      "grad_norm": 3.542126417160034,
      "learning_rate": 4.980842911877395e-05,
      "loss": 0.6058,
      "step": 300
    },
    {
      "epoch": 2.134251290877797,
      "grad_norm": 3.6798830032348633,
      "learning_rate": 4.96168582375479e-05,
      "loss": 0.5967,
      "step": 310
    },
    {
      "epoch": 2.2030981067125643,
      "grad_norm": 3.2142186164855957,
      "learning_rate": 4.9425287356321845e-05,
      "loss": 0.5376,
      "step": 320
    },
    {
      "epoch": 2.2719449225473323,
      "grad_norm": 3.1692137718200684,
      "learning_rate": 4.9233716475095786e-05,
      "loss": 0.5135,
      "step": 330
    },
    {
      "epoch": 2.3407917383820998,
      "grad_norm": 2.785217523574829,
      "learning_rate": 4.904214559386973e-05,
      "loss": 0.4794,
      "step": 340
    },
    {
      "epoch": 2.4096385542168672,
      "grad_norm": 3.252037763595581,
      "learning_rate": 4.885057471264368e-05,
      "loss": 0.5012,
      "step": 350
    },
    {
      "epoch": 2.478485370051635,
      "grad_norm": 2.9605207443237305,
      "learning_rate": 4.865900383141763e-05,
      "loss": 0.4613,
      "step": 360
    },
    {
      "epoch": 2.5473321858864026,
      "grad_norm": 3.263587474822998,
      "learning_rate": 4.846743295019157e-05,
      "loss": 0.4191,
      "step": 370
    },
    {
      "epoch": 2.6161790017211706,
      "grad_norm": 3.321648120880127,
      "learning_rate": 4.827586206896552e-05,
      "loss": 0.4206,
      "step": 380
    },
    {
      "epoch": 2.685025817555938,
      "grad_norm": 2.7976036071777344,
      "learning_rate": 4.8084291187739464e-05,
      "loss": 0.462,
      "step": 390
    },
    {
      "epoch": 2.7538726333907055,
      "grad_norm": 3.0656702518463135,
      "learning_rate": 4.789272030651341e-05,
      "loss": 0.41,
      "step": 400
    },
    {
      "epoch": 2.8227194492254735,
      "grad_norm": 3.332609176635742,
      "learning_rate": 4.770114942528736e-05,
      "loss": 0.4012,
      "step": 410
    },
    {
      "epoch": 2.891566265060241,
      "grad_norm": 3.5820395946502686,
      "learning_rate": 4.7509578544061307e-05,
      "loss": 0.3905,
      "step": 420
    },
    {
      "epoch": 2.960413080895009,
      "grad_norm": 3.3419365882873535,
      "learning_rate": 4.7318007662835254e-05,
      "loss": 0.4176,
      "step": 430
    },
    {
      "epoch": 2.9948364888123926,
      "eval_accuracy": 0.8468992248062015,
      "eval_loss": 0.48874419927597046,
      "eval_runtime": 7.9016,
      "eval_samples_per_second": 261.213,
      "eval_steps_per_second": 8.226,
      "step": 435
    },
    {
      "epoch": 3.0292598967297764,
      "grad_norm": 3.0542681217193604,
      "learning_rate": 4.7126436781609195e-05,
      "loss": 0.4024,
      "step": 440
    },
    {
      "epoch": 3.098106712564544,
      "grad_norm": 2.760667562484741,
      "learning_rate": 4.693486590038315e-05,
      "loss": 0.3706,
      "step": 450
    },
    {
      "epoch": 3.1669535283993113,
      "grad_norm": 3.2174575328826904,
      "learning_rate": 4.674329501915709e-05,
      "loss": 0.3509,
      "step": 460
    },
    {
      "epoch": 3.2358003442340793,
      "grad_norm": 3.6685619354248047,
      "learning_rate": 4.655172413793104e-05,
      "loss": 0.3613,
      "step": 470
    },
    {
      "epoch": 3.3046471600688467,
      "grad_norm": 3.4590651988983154,
      "learning_rate": 4.6360153256704985e-05,
      "loss": 0.346,
      "step": 480
    },
    {
      "epoch": 3.3734939759036147,
      "grad_norm": 3.0743446350097656,
      "learning_rate": 4.616858237547893e-05,
      "loss": 0.3356,
      "step": 490
    },
    {
      "epoch": 3.442340791738382,
      "grad_norm": 3.5646955966949463,
      "learning_rate": 4.597701149425287e-05,
      "loss": 0.3404,
      "step": 500
    },
    {
      "epoch": 3.5111876075731496,
      "grad_norm": 3.2671332359313965,
      "learning_rate": 4.578544061302682e-05,
      "loss": 0.3419,
      "step": 510
    },
    {
      "epoch": 3.580034423407917,
      "grad_norm": 2.7561419010162354,
      "learning_rate": 4.559386973180077e-05,
      "loss": 0.3306,
      "step": 520
    },
    {
      "epoch": 3.648881239242685,
      "grad_norm": 3.1499881744384766,
      "learning_rate": 4.5402298850574716e-05,
      "loss": 0.3575,
      "step": 530
    },
    {
      "epoch": 3.7177280550774525,
      "grad_norm": 2.576195240020752,
      "learning_rate": 4.5210727969348656e-05,
      "loss": 0.2954,
      "step": 540
    },
    {
      "epoch": 3.7865748709122204,
      "grad_norm": 2.685713529586792,
      "learning_rate": 4.501915708812261e-05,
      "loss": 0.2971,
      "step": 550
    },
    {
      "epoch": 3.855421686746988,
      "grad_norm": 3.1608219146728516,
      "learning_rate": 4.482758620689655e-05,
      "loss": 0.3057,
      "step": 560
    },
    {
      "epoch": 3.9242685025817554,
      "grad_norm": 3.3374826908111572,
      "learning_rate": 4.46360153256705e-05,
      "loss": 0.2815,
      "step": 570
    },
    {
      "epoch": 3.9931153184165233,
      "grad_norm": 3.0209603309631348,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.3107,
      "step": 580
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8943798449612403,
      "eval_loss": 0.34142178297042847,
      "eval_runtime": 7.9804,
      "eval_samples_per_second": 258.634,
      "eval_steps_per_second": 8.145,
      "step": 581
    },
    {
      "epoch": 4.061962134251291,
      "grad_norm": 2.7664377689361572,
      "learning_rate": 4.4252873563218394e-05,
      "loss": 0.2689,
      "step": 590
    },
    {
      "epoch": 4.130808950086059,
      "grad_norm": 2.80204439163208,
      "learning_rate": 4.406130268199234e-05,
      "loss": 0.3003,
      "step": 600
    },
    {
      "epoch": 4.199655765920826,
      "grad_norm": 3.0250327587127686,
      "learning_rate": 4.386973180076628e-05,
      "loss": 0.2768,
      "step": 610
    },
    {
      "epoch": 4.268502581755594,
      "grad_norm": 3.8367512226104736,
      "learning_rate": 4.367816091954024e-05,
      "loss": 0.2582,
      "step": 620
    },
    {
      "epoch": 4.337349397590361,
      "grad_norm": 3.3538951873779297,
      "learning_rate": 4.348659003831418e-05,
      "loss": 0.2849,
      "step": 630
    },
    {
      "epoch": 4.406196213425129,
      "grad_norm": 3.2220115661621094,
      "learning_rate": 4.3295019157088125e-05,
      "loss": 0.2744,
      "step": 640
    },
    {
      "epoch": 4.475043029259897,
      "grad_norm": 3.196542263031006,
      "learning_rate": 4.3103448275862066e-05,
      "loss": 0.2744,
      "step": 650
    },
    {
      "epoch": 4.5438898450946645,
      "grad_norm": 3.07588791847229,
      "learning_rate": 4.291187739463602e-05,
      "loss": 0.2766,
      "step": 660
    },
    {
      "epoch": 4.612736660929432,
      "grad_norm": 3.722421169281006,
      "learning_rate": 4.272030651340996e-05,
      "loss": 0.2699,
      "step": 670
    },
    {
      "epoch": 4.6815834767641995,
      "grad_norm": 2.5164051055908203,
      "learning_rate": 4.252873563218391e-05,
      "loss": 0.2401,
      "step": 680
    },
    {
      "epoch": 4.750430292598967,
      "grad_norm": 3.1823108196258545,
      "learning_rate": 4.2337164750957856e-05,
      "loss": 0.2293,
      "step": 690
    },
    {
      "epoch": 4.8192771084337345,
      "grad_norm": 3.4586544036865234,
      "learning_rate": 4.21455938697318e-05,
      "loss": 0.27,
      "step": 700
    },
    {
      "epoch": 4.888123924268503,
      "grad_norm": 3.251978635787964,
      "learning_rate": 4.195402298850575e-05,
      "loss": 0.2551,
      "step": 710
    },
    {
      "epoch": 4.95697074010327,
      "grad_norm": 2.606138229370117,
      "learning_rate": 4.17624521072797e-05,
      "loss": 0.2255,
      "step": 720
    },
    {
      "epoch": 4.998278829604131,
      "eval_accuracy": 0.9123062015503876,
      "eval_loss": 0.27319762110710144,
      "eval_runtime": 7.8624,
      "eval_samples_per_second": 262.515,
      "eval_steps_per_second": 8.267,
      "step": 726
    },
    {
      "epoch": 5.025817555938038,
      "grad_norm": 2.684196949005127,
      "learning_rate": 4.1570881226053646e-05,
      "loss": 0.2487,
      "step": 730
    },
    {
      "epoch": 5.094664371772805,
      "grad_norm": 2.931312084197998,
      "learning_rate": 4.1379310344827587e-05,
      "loss": 0.2314,
      "step": 740
    },
    {
      "epoch": 5.163511187607573,
      "grad_norm": 3.0064966678619385,
      "learning_rate": 4.1187739463601534e-05,
      "loss": 0.2293,
      "step": 750
    },
    {
      "epoch": 5.232358003442341,
      "grad_norm": 3.3251922130584717,
      "learning_rate": 4.099616858237548e-05,
      "loss": 0.2388,
      "step": 760
    },
    {
      "epoch": 5.301204819277109,
      "grad_norm": 1.7967256307601929,
      "learning_rate": 4.080459770114943e-05,
      "loss": 0.2169,
      "step": 770
    },
    {
      "epoch": 5.370051635111876,
      "grad_norm": 3.74770450592041,
      "learning_rate": 4.061302681992337e-05,
      "loss": 0.2096,
      "step": 780
    },
    {
      "epoch": 5.438898450946644,
      "grad_norm": 3.072739601135254,
      "learning_rate": 4.0421455938697324e-05,
      "loss": 0.2047,
      "step": 790
    },
    {
      "epoch": 5.507745266781411,
      "grad_norm": 2.96258807182312,
      "learning_rate": 4.0229885057471265e-05,
      "loss": 0.2313,
      "step": 800
    },
    {
      "epoch": 5.576592082616179,
      "grad_norm": 2.5839080810546875,
      "learning_rate": 4.003831417624521e-05,
      "loss": 0.2112,
      "step": 810
    },
    {
      "epoch": 5.645438898450947,
      "grad_norm": 3.1343889236450195,
      "learning_rate": 3.984674329501916e-05,
      "loss": 0.2482,
      "step": 820
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 2.240661382675171,
      "learning_rate": 3.965517241379311e-05,
      "loss": 0.2146,
      "step": 830
    },
    {
      "epoch": 5.783132530120482,
      "grad_norm": 3.1076242923736572,
      "learning_rate": 3.9463601532567055e-05,
      "loss": 0.2234,
      "step": 840
    },
    {
      "epoch": 5.851979345955249,
      "grad_norm": 2.9466423988342285,
      "learning_rate": 3.9272030651340996e-05,
      "loss": 0.2301,
      "step": 850
    },
    {
      "epoch": 5.920826161790017,
      "grad_norm": 3.1047613620758057,
      "learning_rate": 3.908045977011495e-05,
      "loss": 0.2202,
      "step": 860
    },
    {
      "epoch": 5.989672977624785,
      "grad_norm": 2.182670831680298,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.1833,
      "step": 870
    },
    {
      "epoch": 5.9965576592082614,
      "eval_accuracy": 0.7582364341085271,
      "eval_loss": 0.7461967468261719,
      "eval_runtime": 7.912,
      "eval_samples_per_second": 260.87,
      "eval_steps_per_second": 8.215,
      "step": 871
    },
    {
      "epoch": 6.058519793459553,
      "grad_norm": 2.8595638275146484,
      "learning_rate": 3.869731800766284e-05,
      "loss": 0.198,
      "step": 880
    },
    {
      "epoch": 6.12736660929432,
      "grad_norm": 1.812150478363037,
      "learning_rate": 3.850574712643678e-05,
      "loss": 0.17,
      "step": 890
    },
    {
      "epoch": 6.196213425129088,
      "grad_norm": 2.8384931087493896,
      "learning_rate": 3.831417624521073e-05,
      "loss": 0.2387,
      "step": 900
    },
    {
      "epoch": 6.265060240963855,
      "grad_norm": 4.172893047332764,
      "learning_rate": 3.8122605363984674e-05,
      "loss": 0.2202,
      "step": 910
    },
    {
      "epoch": 6.333907056798623,
      "grad_norm": 4.461463928222656,
      "learning_rate": 3.793103448275862e-05,
      "loss": 0.2501,
      "step": 920
    },
    {
      "epoch": 6.402753872633391,
      "grad_norm": 3.430365562438965,
      "learning_rate": 3.773946360153257e-05,
      "loss": 0.185,
      "step": 930
    },
    {
      "epoch": 6.4716006884681585,
      "grad_norm": 3.3742434978485107,
      "learning_rate": 3.7547892720306517e-05,
      "loss": 0.2166,
      "step": 940
    },
    {
      "epoch": 6.540447504302926,
      "grad_norm": 3.7096402645111084,
      "learning_rate": 3.735632183908046e-05,
      "loss": 0.234,
      "step": 950
    },
    {
      "epoch": 6.6092943201376935,
      "grad_norm": 2.182164192199707,
      "learning_rate": 3.716475095785441e-05,
      "loss": 0.1852,
      "step": 960
    },
    {
      "epoch": 6.678141135972461,
      "grad_norm": 2.585125684738159,
      "learning_rate": 3.697318007662835e-05,
      "loss": 0.1947,
      "step": 970
    },
    {
      "epoch": 6.746987951807229,
      "grad_norm": 2.5588345527648926,
      "learning_rate": 3.67816091954023e-05,
      "loss": 0.211,
      "step": 980
    },
    {
      "epoch": 6.815834767641997,
      "grad_norm": 3.0795257091522217,
      "learning_rate": 3.659003831417625e-05,
      "loss": 0.1967,
      "step": 990
    },
    {
      "epoch": 6.884681583476764,
      "grad_norm": 3.7196614742279053,
      "learning_rate": 3.6398467432950195e-05,
      "loss": 0.1836,
      "step": 1000
    },
    {
      "epoch": 6.953528399311532,
      "grad_norm": 2.3857243061065674,
      "learning_rate": 3.620689655172414e-05,
      "loss": 0.2062,
      "step": 1010
    },
    {
      "epoch": 6.994836488812393,
      "eval_accuracy": 0.8803294573643411,
      "eval_loss": 0.3770907521247864,
      "eval_runtime": 7.9957,
      "eval_samples_per_second": 258.14,
      "eval_steps_per_second": 8.129,
      "step": 1016
    },
    {
      "epoch": 7.022375215146299,
      "grad_norm": 2.5683236122131348,
      "learning_rate": 3.601532567049808e-05,
      "loss": 0.1754,
      "step": 1020
    },
    {
      "epoch": 7.091222030981067,
      "grad_norm": 2.679636001586914,
      "learning_rate": 3.582375478927204e-05,
      "loss": 0.1675,
      "step": 1030
    },
    {
      "epoch": 7.160068846815835,
      "grad_norm": 2.5351173877716064,
      "learning_rate": 3.563218390804598e-05,
      "loss": 0.1565,
      "step": 1040
    },
    {
      "epoch": 7.228915662650603,
      "grad_norm": 2.9072484970092773,
      "learning_rate": 3.5440613026819926e-05,
      "loss": 0.1813,
      "step": 1050
    },
    {
      "epoch": 7.29776247848537,
      "grad_norm": 2.5954692363739014,
      "learning_rate": 3.5249042145593867e-05,
      "loss": 0.2118,
      "step": 1060
    },
    {
      "epoch": 7.366609294320138,
      "grad_norm": 2.7799787521362305,
      "learning_rate": 3.505747126436782e-05,
      "loss": 0.1744,
      "step": 1070
    },
    {
      "epoch": 7.435456110154905,
      "grad_norm": 3.0674917697906494,
      "learning_rate": 3.486590038314176e-05,
      "loss": 0.1922,
      "step": 1080
    },
    {
      "epoch": 7.504302925989673,
      "grad_norm": 3.5969247817993164,
      "learning_rate": 3.467432950191571e-05,
      "loss": 0.1794,
      "step": 1090
    },
    {
      "epoch": 7.573149741824441,
      "grad_norm": 2.6146047115325928,
      "learning_rate": 3.4482758620689657e-05,
      "loss": 0.1682,
      "step": 1100
    },
    {
      "epoch": 7.641996557659208,
      "grad_norm": 2.785433053970337,
      "learning_rate": 3.4291187739463604e-05,
      "loss": 0.1901,
      "step": 1110
    },
    {
      "epoch": 7.710843373493976,
      "grad_norm": 2.3648314476013184,
      "learning_rate": 3.409961685823755e-05,
      "loss": 0.1521,
      "step": 1120
    },
    {
      "epoch": 7.779690189328743,
      "grad_norm": 2.3298792839050293,
      "learning_rate": 3.390804597701149e-05,
      "loss": 0.1901,
      "step": 1130
    },
    {
      "epoch": 7.848537005163511,
      "grad_norm": 2.2863216400146484,
      "learning_rate": 3.371647509578545e-05,
      "loss": 0.1706,
      "step": 1140
    },
    {
      "epoch": 7.917383820998279,
      "grad_norm": 3.0322484970092773,
      "learning_rate": 3.352490421455939e-05,
      "loss": 0.1612,
      "step": 1150
    },
    {
      "epoch": 7.986230636833047,
      "grad_norm": 2.785752534866333,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1657,
      "step": 1160
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8541666666666666,
      "eval_loss": 0.47179850935935974,
      "eval_runtime": 7.8842,
      "eval_samples_per_second": 261.788,
      "eval_steps_per_second": 8.244,
      "step": 1162
    },
    {
      "epoch": 8.055077452667813,
      "grad_norm": 3.0953948497772217,
      "learning_rate": 3.314176245210728e-05,
      "loss": 0.1848,
      "step": 1170
    },
    {
      "epoch": 8.123924268502583,
      "grad_norm": 2.7531402111053467,
      "learning_rate": 3.295019157088123e-05,
      "loss": 0.1451,
      "step": 1180
    },
    {
      "epoch": 8.19277108433735,
      "grad_norm": 3.849405288696289,
      "learning_rate": 3.275862068965517e-05,
      "loss": 0.1682,
      "step": 1190
    },
    {
      "epoch": 8.261617900172118,
      "grad_norm": 2.8468093872070312,
      "learning_rate": 3.256704980842912e-05,
      "loss": 0.1868,
      "step": 1200
    },
    {
      "epoch": 8.330464716006885,
      "grad_norm": 3.0676023960113525,
      "learning_rate": 3.2375478927203066e-05,
      "loss": 0.1745,
      "step": 1210
    },
    {
      "epoch": 8.399311531841652,
      "grad_norm": 2.8244659900665283,
      "learning_rate": 3.218390804597701e-05,
      "loss": 0.1783,
      "step": 1220
    },
    {
      "epoch": 8.46815834767642,
      "grad_norm": 3.3981621265411377,
      "learning_rate": 3.1992337164750954e-05,
      "loss": 0.1509,
      "step": 1230
    },
    {
      "epoch": 8.537005163511187,
      "grad_norm": 3.2597496509552,
      "learning_rate": 3.180076628352491e-05,
      "loss": 0.1463,
      "step": 1240
    },
    {
      "epoch": 8.605851979345955,
      "grad_norm": 2.98665452003479,
      "learning_rate": 3.160919540229885e-05,
      "loss": 0.1854,
      "step": 1250
    },
    {
      "epoch": 8.674698795180722,
      "grad_norm": 3.56426739692688,
      "learning_rate": 3.1417624521072797e-05,
      "loss": 0.1801,
      "step": 1260
    },
    {
      "epoch": 8.74354561101549,
      "grad_norm": 3.350994110107422,
      "learning_rate": 3.1226053639846744e-05,
      "loss": 0.146,
      "step": 1270
    },
    {
      "epoch": 8.812392426850257,
      "grad_norm": 2.637911081314087,
      "learning_rate": 3.103448275862069e-05,
      "loss": 0.1759,
      "step": 1280
    },
    {
      "epoch": 8.881239242685027,
      "grad_norm": 3.2409827709198,
      "learning_rate": 3.084291187739464e-05,
      "loss": 0.1897,
      "step": 1290
    },
    {
      "epoch": 8.950086058519794,
      "grad_norm": 2.5662424564361572,
      "learning_rate": 3.065134099616858e-05,
      "loss": 0.1427,
      "step": 1300
    },
    {
      "epoch": 8.99827882960413,
      "eval_accuracy": 0.8473837209302325,
      "eval_loss": 0.4902307689189911,
      "eval_runtime": 7.7511,
      "eval_samples_per_second": 266.284,
      "eval_steps_per_second": 8.386,
      "step": 1307
    },
    {
      "epoch": 9.018932874354562,
      "grad_norm": 2.6389100551605225,
      "learning_rate": 3.045977011494253e-05,
      "loss": 0.1577,
      "step": 1310
    },
    {
      "epoch": 9.087779690189329,
      "grad_norm": 2.561455249786377,
      "learning_rate": 3.0268199233716475e-05,
      "loss": 0.1686,
      "step": 1320
    },
    {
      "epoch": 9.156626506024097,
      "grad_norm": 2.796520233154297,
      "learning_rate": 3.0076628352490422e-05,
      "loss": 0.1623,
      "step": 1330
    },
    {
      "epoch": 9.225473321858864,
      "grad_norm": 3.3344945907592773,
      "learning_rate": 2.988505747126437e-05,
      "loss": 0.1684,
      "step": 1340
    },
    {
      "epoch": 9.294320137693632,
      "grad_norm": 2.494370698928833,
      "learning_rate": 2.9693486590038317e-05,
      "loss": 0.1679,
      "step": 1350
    },
    {
      "epoch": 9.363166953528399,
      "grad_norm": 2.9866812229156494,
      "learning_rate": 2.950191570881226e-05,
      "loss": 0.168,
      "step": 1360
    },
    {
      "epoch": 9.432013769363166,
      "grad_norm": 2.8670761585235596,
      "learning_rate": 2.9310344827586206e-05,
      "loss": 0.1587,
      "step": 1370
    },
    {
      "epoch": 9.500860585197934,
      "grad_norm": 2.6785264015197754,
      "learning_rate": 2.9118773946360157e-05,
      "loss": 0.1487,
      "step": 1380
    },
    {
      "epoch": 9.569707401032701,
      "grad_norm": 3.9908454418182373,
      "learning_rate": 2.89272030651341e-05,
      "loss": 0.1913,
      "step": 1390
    },
    {
      "epoch": 9.638554216867469,
      "grad_norm": 3.4217581748962402,
      "learning_rate": 2.8735632183908045e-05,
      "loss": 0.1568,
      "step": 1400
    },
    {
      "epoch": 9.707401032702238,
      "grad_norm": 3.4280409812927246,
      "learning_rate": 2.8544061302681996e-05,
      "loss": 0.1335,
      "step": 1410
    },
    {
      "epoch": 9.776247848537006,
      "grad_norm": 3.57127046585083,
      "learning_rate": 2.835249042145594e-05,
      "loss": 0.1508,
      "step": 1420
    },
    {
      "epoch": 9.845094664371773,
      "grad_norm": 2.238633394241333,
      "learning_rate": 2.8160919540229884e-05,
      "loss": 0.1878,
      "step": 1430
    },
    {
      "epoch": 9.91394148020654,
      "grad_norm": 2.7804625034332275,
      "learning_rate": 2.796934865900383e-05,
      "loss": 0.1497,
      "step": 1440
    },
    {
      "epoch": 9.982788296041308,
      "grad_norm": 3.1482419967651367,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1598,
      "step": 1450
    },
    {
      "epoch": 9.996557659208262,
      "eval_accuracy": 0.9273255813953488,
      "eval_loss": 0.22291144728660583,
      "eval_runtime": 7.6884,
      "eval_samples_per_second": 268.456,
      "eval_steps_per_second": 8.454,
      "step": 1452
    },
    {
      "epoch": 10.051635111876076,
      "grad_norm": 2.5692484378814697,
      "learning_rate": 2.7586206896551727e-05,
      "loss": 0.1397,
      "step": 1460
    },
    {
      "epoch": 10.120481927710843,
      "grad_norm": 1.686840295791626,
      "learning_rate": 2.739463601532567e-05,
      "loss": 0.1246,
      "step": 1470
    },
    {
      "epoch": 10.18932874354561,
      "grad_norm": 2.7757647037506104,
      "learning_rate": 2.720306513409962e-05,
      "loss": 0.14,
      "step": 1480
    },
    {
      "epoch": 10.258175559380378,
      "grad_norm": 2.688798189163208,
      "learning_rate": 2.7011494252873566e-05,
      "loss": 0.1386,
      "step": 1490
    },
    {
      "epoch": 10.327022375215146,
      "grad_norm": 3.0354042053222656,
      "learning_rate": 2.681992337164751e-05,
      "loss": 0.1585,
      "step": 1500
    },
    {
      "epoch": 10.395869191049915,
      "grad_norm": 3.36676287651062,
      "learning_rate": 2.662835249042146e-05,
      "loss": 0.1368,
      "step": 1510
    },
    {
      "epoch": 10.464716006884682,
      "grad_norm": 2.3574771881103516,
      "learning_rate": 2.6436781609195405e-05,
      "loss": 0.1559,
      "step": 1520
    },
    {
      "epoch": 10.53356282271945,
      "grad_norm": 2.6529464721679688,
      "learning_rate": 2.624521072796935e-05,
      "loss": 0.154,
      "step": 1530
    },
    {
      "epoch": 10.602409638554217,
      "grad_norm": 3.449251174926758,
      "learning_rate": 2.6053639846743293e-05,
      "loss": 0.1401,
      "step": 1540
    },
    {
      "epoch": 10.671256454388985,
      "grad_norm": 2.449934482574463,
      "learning_rate": 2.5862068965517244e-05,
      "loss": 0.1408,
      "step": 1550
    },
    {
      "epoch": 10.740103270223752,
      "grad_norm": 2.8734378814697266,
      "learning_rate": 2.5670498084291188e-05,
      "loss": 0.1445,
      "step": 1560
    },
    {
      "epoch": 10.80895008605852,
      "grad_norm": 2.4971940517425537,
      "learning_rate": 2.5478927203065132e-05,
      "loss": 0.1643,
      "step": 1570
    },
    {
      "epoch": 10.877796901893287,
      "grad_norm": 3.7385921478271484,
      "learning_rate": 2.5287356321839083e-05,
      "loss": 0.1521,
      "step": 1580
    },
    {
      "epoch": 10.946643717728055,
      "grad_norm": 2.7152345180511475,
      "learning_rate": 2.5095785440613027e-05,
      "loss": 0.1504,
      "step": 1590
    },
    {
      "epoch": 10.994836488812393,
      "eval_accuracy": 0.8972868217054264,
      "eval_loss": 0.3021294176578522,
      "eval_runtime": 7.4857,
      "eval_samples_per_second": 275.726,
      "eval_steps_per_second": 8.683,
      "step": 1597
    },
    {
      "epoch": 11.015490533562822,
      "grad_norm": 3.032806158065796,
      "learning_rate": 2.4904214559386975e-05,
      "loss": 0.1366,
      "step": 1600
    },
    {
      "epoch": 11.08433734939759,
      "grad_norm": 2.85953950881958,
      "learning_rate": 2.4712643678160922e-05,
      "loss": 0.1433,
      "step": 1610
    },
    {
      "epoch": 11.153184165232357,
      "grad_norm": 2.1525700092315674,
      "learning_rate": 2.4521072796934867e-05,
      "loss": 0.1455,
      "step": 1620
    },
    {
      "epoch": 11.222030981067126,
      "grad_norm": 2.248845100402832,
      "learning_rate": 2.4329501915708814e-05,
      "loss": 0.1269,
      "step": 1630
    },
    {
      "epoch": 11.290877796901894,
      "grad_norm": 2.333122968673706,
      "learning_rate": 2.413793103448276e-05,
      "loss": 0.156,
      "step": 1640
    },
    {
      "epoch": 11.359724612736661,
      "grad_norm": 2.1817004680633545,
      "learning_rate": 2.3946360153256706e-05,
      "loss": 0.1423,
      "step": 1650
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": 2.1537561416625977,
      "learning_rate": 2.3754789272030653e-05,
      "loss": 0.1327,
      "step": 1660
    },
    {
      "epoch": 11.497418244406196,
      "grad_norm": 1.6727231740951538,
      "learning_rate": 2.3563218390804597e-05,
      "loss": 0.1272,
      "step": 1670
    },
    {
      "epoch": 11.566265060240964,
      "grad_norm": 3.15240216255188,
      "learning_rate": 2.3371647509578545e-05,
      "loss": 0.1474,
      "step": 1680
    },
    {
      "epoch": 11.635111876075731,
      "grad_norm": 2.031428813934326,
      "learning_rate": 2.3180076628352492e-05,
      "loss": 0.1341,
      "step": 1690
    },
    {
      "epoch": 11.703958691910499,
      "grad_norm": 2.5997140407562256,
      "learning_rate": 2.2988505747126437e-05,
      "loss": 0.153,
      "step": 1700
    },
    {
      "epoch": 11.772805507745266,
      "grad_norm": 2.405869483947754,
      "learning_rate": 2.2796934865900384e-05,
      "loss": 0.1434,
      "step": 1710
    },
    {
      "epoch": 11.841652323580034,
      "grad_norm": 2.8463237285614014,
      "learning_rate": 2.2605363984674328e-05,
      "loss": 0.1545,
      "step": 1720
    },
    {
      "epoch": 11.910499139414803,
      "grad_norm": 2.8656628131866455,
      "learning_rate": 2.2413793103448276e-05,
      "loss": 0.1412,
      "step": 1730
    },
    {
      "epoch": 11.97934595524957,
      "grad_norm": 2.9610302448272705,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1456,
      "step": 1740
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9224806201550387,
      "eval_loss": 0.24217405915260315,
      "eval_runtime": 7.2188,
      "eval_samples_per_second": 285.921,
      "eval_steps_per_second": 9.004,
      "step": 1743
    },
    {
      "epoch": 12.048192771084338,
      "grad_norm": 2.278578042984009,
      "learning_rate": 2.203065134099617e-05,
      "loss": 0.1079,
      "step": 1750
    },
    {
      "epoch": 12.117039586919105,
      "grad_norm": 2.1894547939300537,
      "learning_rate": 2.183908045977012e-05,
      "loss": 0.1298,
      "step": 1760
    },
    {
      "epoch": 12.185886402753873,
      "grad_norm": 2.7255043983459473,
      "learning_rate": 2.1647509578544062e-05,
      "loss": 0.1502,
      "step": 1770
    },
    {
      "epoch": 12.25473321858864,
      "grad_norm": 1.9726896286010742,
      "learning_rate": 2.145593869731801e-05,
      "loss": 0.1589,
      "step": 1780
    },
    {
      "epoch": 12.323580034423408,
      "grad_norm": 2.095195770263672,
      "learning_rate": 2.1264367816091954e-05,
      "loss": 0.1345,
      "step": 1790
    },
    {
      "epoch": 12.392426850258175,
      "grad_norm": 2.9772017002105713,
      "learning_rate": 2.10727969348659e-05,
      "loss": 0.1537,
      "step": 1800
    },
    {
      "epoch": 12.461273666092943,
      "grad_norm": 1.4328383207321167,
      "learning_rate": 2.088122605363985e-05,
      "loss": 0.1274,
      "step": 1810
    },
    {
      "epoch": 12.53012048192771,
      "grad_norm": 2.386387586593628,
      "learning_rate": 2.0689655172413793e-05,
      "loss": 0.1333,
      "step": 1820
    },
    {
      "epoch": 12.598967297762478,
      "grad_norm": 1.9341486692428589,
      "learning_rate": 2.049808429118774e-05,
      "loss": 0.1397,
      "step": 1830
    },
    {
      "epoch": 12.667814113597245,
      "grad_norm": 2.633357048034668,
      "learning_rate": 2.0306513409961685e-05,
      "loss": 0.1151,
      "step": 1840
    },
    {
      "epoch": 12.736660929432015,
      "grad_norm": 3.059253215789795,
      "learning_rate": 2.0114942528735632e-05,
      "loss": 0.119,
      "step": 1850
    },
    {
      "epoch": 12.805507745266782,
      "grad_norm": 1.629744291305542,
      "learning_rate": 1.992337164750958e-05,
      "loss": 0.1194,
      "step": 1860
    },
    {
      "epoch": 12.87435456110155,
      "grad_norm": 3.2468857765197754,
      "learning_rate": 1.9731800766283527e-05,
      "loss": 0.1567,
      "step": 1870
    },
    {
      "epoch": 12.943201376936317,
      "grad_norm": 2.7284107208251953,
      "learning_rate": 1.9540229885057475e-05,
      "loss": 0.119,
      "step": 1880
    },
    {
      "epoch": 12.99827882960413,
      "eval_accuracy": 0.9021317829457365,
      "eval_loss": 0.2836342751979828,
      "eval_runtime": 7.2838,
      "eval_samples_per_second": 283.368,
      "eval_steps_per_second": 8.924,
      "step": 1888
    },
    {
      "epoch": 13.012048192771084,
      "grad_norm": 3.377997875213623,
      "learning_rate": 1.934865900383142e-05,
      "loss": 0.119,
      "step": 1890
    },
    {
      "epoch": 13.080895008605852,
      "grad_norm": 2.6370134353637695,
      "learning_rate": 1.9157088122605367e-05,
      "loss": 0.1394,
      "step": 1900
    },
    {
      "epoch": 13.14974182444062,
      "grad_norm": 2.5573413372039795,
      "learning_rate": 1.896551724137931e-05,
      "loss": 0.1475,
      "step": 1910
    },
    {
      "epoch": 13.218588640275387,
      "grad_norm": 2.2910749912261963,
      "learning_rate": 1.8773946360153258e-05,
      "loss": 0.1332,
      "step": 1920
    },
    {
      "epoch": 13.287435456110154,
      "grad_norm": 2.3592233657836914,
      "learning_rate": 1.8582375478927206e-05,
      "loss": 0.1373,
      "step": 1930
    },
    {
      "epoch": 13.356282271944922,
      "grad_norm": 2.7366604804992676,
      "learning_rate": 1.839080459770115e-05,
      "loss": 0.129,
      "step": 1940
    },
    {
      "epoch": 13.42512908777969,
      "grad_norm": 2.452705144882202,
      "learning_rate": 1.8199233716475097e-05,
      "loss": 0.1183,
      "step": 1950
    },
    {
      "epoch": 13.493975903614459,
      "grad_norm": 3.3416621685028076,
      "learning_rate": 1.800766283524904e-05,
      "loss": 0.154,
      "step": 1960
    },
    {
      "epoch": 13.562822719449226,
      "grad_norm": 3.04327654838562,
      "learning_rate": 1.781609195402299e-05,
      "loss": 0.1264,
      "step": 1970
    },
    {
      "epoch": 13.631669535283994,
      "grad_norm": 2.6300888061523438,
      "learning_rate": 1.7624521072796933e-05,
      "loss": 0.1646,
      "step": 1980
    },
    {
      "epoch": 13.700516351118761,
      "grad_norm": 3.809067487716675,
      "learning_rate": 1.743295019157088e-05,
      "loss": 0.1363,
      "step": 1990
    },
    {
      "epoch": 13.769363166953529,
      "grad_norm": 3.2865700721740723,
      "learning_rate": 1.7241379310344828e-05,
      "loss": 0.1145,
      "step": 2000
    },
    {
      "epoch": 13.838209982788296,
      "grad_norm": 2.2611865997314453,
      "learning_rate": 1.7049808429118776e-05,
      "loss": 0.116,
      "step": 2010
    },
    {
      "epoch": 13.907056798623064,
      "grad_norm": 2.3190841674804688,
      "learning_rate": 1.6858237547892723e-05,
      "loss": 0.1105,
      "step": 2020
    },
    {
      "epoch": 13.975903614457831,
      "grad_norm": 3.105708360671997,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.114,
      "step": 2030
    },
    {
      "epoch": 13.996557659208262,
      "eval_accuracy": 0.9292635658914729,
      "eval_loss": 0.20378853380680084,
      "eval_runtime": 7.3214,
      "eval_samples_per_second": 281.914,
      "eval_steps_per_second": 8.878,
      "step": 2033
    },
    {
      "epoch": 14.044750430292599,
      "grad_norm": 2.9859328269958496,
      "learning_rate": 1.6475095785440615e-05,
      "loss": 0.1319,
      "step": 2040
    },
    {
      "epoch": 14.113597246127366,
      "grad_norm": 4.717568874359131,
      "learning_rate": 1.628352490421456e-05,
      "loss": 0.1426,
      "step": 2050
    },
    {
      "epoch": 14.182444061962133,
      "grad_norm": 3.3271358013153076,
      "learning_rate": 1.6091954022988507e-05,
      "loss": 0.1458,
      "step": 2060
    },
    {
      "epoch": 14.251290877796903,
      "grad_norm": 3.017382860183716,
      "learning_rate": 1.5900383141762454e-05,
      "loss": 0.1576,
      "step": 2070
    },
    {
      "epoch": 14.32013769363167,
      "grad_norm": 3.2462213039398193,
      "learning_rate": 1.5708812260536398e-05,
      "loss": 0.1438,
      "step": 2080
    },
    {
      "epoch": 14.388984509466438,
      "grad_norm": 3.2721612453460693,
      "learning_rate": 1.5517241379310346e-05,
      "loss": 0.1632,
      "step": 2090
    },
    {
      "epoch": 14.457831325301205,
      "grad_norm": 2.183432102203369,
      "learning_rate": 1.532567049808429e-05,
      "loss": 0.1071,
      "step": 2100
    },
    {
      "epoch": 14.526678141135973,
      "grad_norm": 2.744126319885254,
      "learning_rate": 1.5134099616858237e-05,
      "loss": 0.1234,
      "step": 2110
    },
    {
      "epoch": 14.59552495697074,
      "grad_norm": 1.959697961807251,
      "learning_rate": 1.4942528735632185e-05,
      "loss": 0.1224,
      "step": 2120
    },
    {
      "epoch": 14.664371772805508,
      "grad_norm": 3.4430899620056152,
      "learning_rate": 1.475095785440613e-05,
      "loss": 0.1174,
      "step": 2130
    },
    {
      "epoch": 14.733218588640275,
      "grad_norm": 4.045992374420166,
      "learning_rate": 1.4559386973180078e-05,
      "loss": 0.1315,
      "step": 2140
    },
    {
      "epoch": 14.802065404475043,
      "grad_norm": 2.5135693550109863,
      "learning_rate": 1.4367816091954022e-05,
      "loss": 0.111,
      "step": 2150
    },
    {
      "epoch": 14.87091222030981,
      "grad_norm": 3.5096664428710938,
      "learning_rate": 1.417624521072797e-05,
      "loss": 0.1429,
      "step": 2160
    },
    {
      "epoch": 14.939759036144578,
      "grad_norm": 2.6633498668670654,
      "learning_rate": 1.3984674329501916e-05,
      "loss": 0.1378,
      "step": 2170
    },
    {
      "epoch": 14.994836488812393,
      "eval_accuracy": 0.9239341085271318,
      "eval_loss": 0.2173481285572052,
      "eval_runtime": 7.5899,
      "eval_samples_per_second": 271.94,
      "eval_steps_per_second": 8.564,
      "step": 2178
    },
    {
      "epoch": 15.008605851979347,
      "grad_norm": 2.4078171253204346,
      "learning_rate": 1.3793103448275863e-05,
      "loss": 0.1133,
      "step": 2180
    },
    {
      "epoch": 15.077452667814114,
      "grad_norm": 3.3953912258148193,
      "learning_rate": 1.360153256704981e-05,
      "loss": 0.1486,
      "step": 2190
    },
    {
      "epoch": 15.146299483648882,
      "grad_norm": 2.9477121829986572,
      "learning_rate": 1.3409961685823755e-05,
      "loss": 0.1097,
      "step": 2200
    },
    {
      "epoch": 15.21514629948365,
      "grad_norm": 2.8032114505767822,
      "learning_rate": 1.3218390804597702e-05,
      "loss": 0.136,
      "step": 2210
    },
    {
      "epoch": 15.283993115318417,
      "grad_norm": 3.5413339138031006,
      "learning_rate": 1.3026819923371647e-05,
      "loss": 0.1434,
      "step": 2220
    },
    {
      "epoch": 15.352839931153184,
      "grad_norm": 2.4817466735839844,
      "learning_rate": 1.2835249042145594e-05,
      "loss": 0.1178,
      "step": 2230
    },
    {
      "epoch": 15.421686746987952,
      "grad_norm": 2.3475193977355957,
      "learning_rate": 1.2643678160919542e-05,
      "loss": 0.0938,
      "step": 2240
    },
    {
      "epoch": 15.49053356282272,
      "grad_norm": 3.2757270336151123,
      "learning_rate": 1.2452107279693487e-05,
      "loss": 0.159,
      "step": 2250
    },
    {
      "epoch": 15.559380378657487,
      "grad_norm": 2.7539010047912598,
      "learning_rate": 1.2260536398467433e-05,
      "loss": 0.1351,
      "step": 2260
    },
    {
      "epoch": 15.628227194492254,
      "grad_norm": 2.8831663131713867,
      "learning_rate": 1.206896551724138e-05,
      "loss": 0.1341,
      "step": 2270
    },
    {
      "epoch": 15.697074010327022,
      "grad_norm": 3.276233434677124,
      "learning_rate": 1.1877394636015327e-05,
      "loss": 0.1348,
      "step": 2280
    },
    {
      "epoch": 15.76592082616179,
      "grad_norm": 2.2141618728637695,
      "learning_rate": 1.1685823754789272e-05,
      "loss": 0.1166,
      "step": 2290
    },
    {
      "epoch": 15.834767641996558,
      "grad_norm": 2.0663790702819824,
      "learning_rate": 1.1494252873563218e-05,
      "loss": 0.0998,
      "step": 2300
    },
    {
      "epoch": 15.903614457831326,
      "grad_norm": 2.9078729152679443,
      "learning_rate": 1.1302681992337164e-05,
      "loss": 0.1267,
      "step": 2310
    },
    {
      "epoch": 15.972461273666093,
      "grad_norm": 2.513744831085205,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1249,
      "step": 2320
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9186046511627907,
      "eval_loss": 0.24669910967350006,
      "eval_runtime": 7.8261,
      "eval_samples_per_second": 263.732,
      "eval_steps_per_second": 8.306,
      "step": 2324
    },
    {
      "epoch": 16.04130808950086,
      "grad_norm": 2.510911226272583,
      "learning_rate": 1.091954022988506e-05,
      "loss": 0.1052,
      "step": 2330
    },
    {
      "epoch": 16.110154905335627,
      "grad_norm": 2.5557456016540527,
      "learning_rate": 1.0727969348659005e-05,
      "loss": 0.1293,
      "step": 2340
    },
    {
      "epoch": 16.179001721170398,
      "grad_norm": 2.11968994140625,
      "learning_rate": 1.053639846743295e-05,
      "loss": 0.1139,
      "step": 2350
    },
    {
      "epoch": 16.247848537005165,
      "grad_norm": 2.4394736289978027,
      "learning_rate": 1.0344827586206897e-05,
      "loss": 0.1206,
      "step": 2360
    },
    {
      "epoch": 16.316695352839933,
      "grad_norm": 2.758610963821411,
      "learning_rate": 1.0153256704980842e-05,
      "loss": 0.1098,
      "step": 2370
    },
    {
      "epoch": 16.3855421686747,
      "grad_norm": 2.410712480545044,
      "learning_rate": 9.96168582375479e-06,
      "loss": 0.1052,
      "step": 2380
    },
    {
      "epoch": 16.454388984509468,
      "grad_norm": 2.830049991607666,
      "learning_rate": 9.770114942528738e-06,
      "loss": 0.11,
      "step": 2390
    },
    {
      "epoch": 16.523235800344235,
      "grad_norm": 2.921481132507324,
      "learning_rate": 9.578544061302683e-06,
      "loss": 0.1038,
      "step": 2400
    },
    {
      "epoch": 16.592082616179002,
      "grad_norm": 3.5350148677825928,
      "learning_rate": 9.386973180076629e-06,
      "loss": 0.1447,
      "step": 2410
    },
    {
      "epoch": 16.66092943201377,
      "grad_norm": 3.2687947750091553,
      "learning_rate": 9.195402298850575e-06,
      "loss": 0.119,
      "step": 2420
    },
    {
      "epoch": 16.729776247848537,
      "grad_norm": 2.843630790710449,
      "learning_rate": 9.00383141762452e-06,
      "loss": 0.1361,
      "step": 2430
    },
    {
      "epoch": 16.798623063683305,
      "grad_norm": 2.6418073177337646,
      "learning_rate": 8.812260536398467e-06,
      "loss": 0.1125,
      "step": 2440
    },
    {
      "epoch": 16.867469879518072,
      "grad_norm": 2.408013343811035,
      "learning_rate": 8.620689655172414e-06,
      "loss": 0.1354,
      "step": 2450
    },
    {
      "epoch": 16.93631669535284,
      "grad_norm": 3.4269073009490967,
      "learning_rate": 8.429118773946362e-06,
      "loss": 0.1504,
      "step": 2460
    },
    {
      "epoch": 16.998278829604132,
      "eval_accuracy": 0.9253875968992248,
      "eval_loss": 0.232245072722435,
      "eval_runtime": 7.9776,
      "eval_samples_per_second": 258.725,
      "eval_steps_per_second": 8.148,
      "step": 2469
    },
    {
      "epoch": 17.005163511187607,
      "grad_norm": 3.9189724922180176,
      "learning_rate": 8.237547892720307e-06,
      "loss": 0.1346,
      "step": 2470
    },
    {
      "epoch": 17.074010327022375,
      "grad_norm": 3.447214126586914,
      "learning_rate": 8.045977011494253e-06,
      "loss": 0.1254,
      "step": 2480
    },
    {
      "epoch": 17.142857142857142,
      "grad_norm": 1.5260127782821655,
      "learning_rate": 7.854406130268199e-06,
      "loss": 0.1264,
      "step": 2490
    },
    {
      "epoch": 17.21170395869191,
      "grad_norm": 2.7160017490386963,
      "learning_rate": 7.662835249042145e-06,
      "loss": 0.1124,
      "step": 2500
    },
    {
      "epoch": 17.280550774526677,
      "grad_norm": 2.530102491378784,
      "learning_rate": 7.4712643678160925e-06,
      "loss": 0.1116,
      "step": 2510
    },
    {
      "epoch": 17.349397590361445,
      "grad_norm": 2.009098768234253,
      "learning_rate": 7.279693486590039e-06,
      "loss": 0.0977,
      "step": 2520
    },
    {
      "epoch": 17.418244406196212,
      "grad_norm": 2.3196418285369873,
      "learning_rate": 7.088122605363985e-06,
      "loss": 0.1527,
      "step": 2530
    },
    {
      "epoch": 17.48709122203098,
      "grad_norm": 2.2051141262054443,
      "learning_rate": 6.896551724137932e-06,
      "loss": 0.1285,
      "step": 2540
    },
    {
      "epoch": 17.555938037865747,
      "grad_norm": 2.9338431358337402,
      "learning_rate": 6.7049808429118775e-06,
      "loss": 0.1182,
      "step": 2550
    },
    {
      "epoch": 17.624784853700515,
      "grad_norm": 2.8327407836914062,
      "learning_rate": 6.513409961685823e-06,
      "loss": 0.0985,
      "step": 2560
    },
    {
      "epoch": 17.693631669535286,
      "grad_norm": 2.8627865314483643,
      "learning_rate": 6.321839080459771e-06,
      "loss": 0.1185,
      "step": 2570
    },
    {
      "epoch": 17.762478485370053,
      "grad_norm": 2.3446600437164307,
      "learning_rate": 6.130268199233717e-06,
      "loss": 0.1024,
      "step": 2580
    },
    {
      "epoch": 17.83132530120482,
      "grad_norm": 3.2858073711395264,
      "learning_rate": 5.938697318007663e-06,
      "loss": 0.1213,
      "step": 2590
    },
    {
      "epoch": 17.900172117039588,
      "grad_norm": 2.4796440601348877,
      "learning_rate": 5.747126436781609e-06,
      "loss": 0.0974,
      "step": 2600
    },
    {
      "epoch": 17.969018932874356,
      "grad_norm": 1.4947516918182373,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.0972,
      "step": 2610
    },
    {
      "epoch": 17.99655765920826,
      "eval_accuracy": 0.9781976744186046,
      "eval_loss": 0.08412329852581024,
      "eval_runtime": 8.8996,
      "eval_samples_per_second": 231.921,
      "eval_steps_per_second": 7.304,
      "step": 2614
    },
    {
      "epoch": 18.037865748709123,
      "grad_norm": 2.7059268951416016,
      "learning_rate": 5.3639846743295025e-06,
      "loss": 0.1074,
      "step": 2620
    },
    {
      "epoch": 18.10671256454389,
      "grad_norm": 2.0837604999542236,
      "learning_rate": 5.172413793103448e-06,
      "loss": 0.106,
      "step": 2630
    },
    {
      "epoch": 18.175559380378658,
      "grad_norm": 3.52249813079834,
      "learning_rate": 4.980842911877395e-06,
      "loss": 0.1475,
      "step": 2640
    },
    {
      "epoch": 18.244406196213426,
      "grad_norm": 2.8017690181732178,
      "learning_rate": 4.789272030651342e-06,
      "loss": 0.128,
      "step": 2650
    },
    {
      "epoch": 18.313253012048193,
      "grad_norm": 1.7369544506072998,
      "learning_rate": 4.5977011494252875e-06,
      "loss": 0.0977,
      "step": 2660
    },
    {
      "epoch": 18.38209982788296,
      "grad_norm": 2.63714599609375,
      "learning_rate": 4.406130268199233e-06,
      "loss": 0.1067,
      "step": 2670
    },
    {
      "epoch": 18.450946643717728,
      "grad_norm": 2.7819557189941406,
      "learning_rate": 4.214559386973181e-06,
      "loss": 0.1364,
      "step": 2680
    },
    {
      "epoch": 18.519793459552496,
      "grad_norm": 2.259345769882202,
      "learning_rate": 4.022988505747127e-06,
      "loss": 0.0957,
      "step": 2690
    },
    {
      "epoch": 18.588640275387263,
      "grad_norm": 2.4775497913360596,
      "learning_rate": 3.8314176245210725e-06,
      "loss": 0.1273,
      "step": 2700
    },
    {
      "epoch": 18.65748709122203,
      "grad_norm": 2.889547824859619,
      "learning_rate": 3.6398467432950196e-06,
      "loss": 0.1293,
      "step": 2710
    },
    {
      "epoch": 18.726333907056798,
      "grad_norm": 2.8670730590820312,
      "learning_rate": 3.448275862068966e-06,
      "loss": 0.1279,
      "step": 2720
    },
    {
      "epoch": 18.795180722891565,
      "grad_norm": 1.9103004932403564,
      "learning_rate": 3.2567049808429117e-06,
      "loss": 0.1292,
      "step": 2730
    },
    {
      "epoch": 18.864027538726333,
      "grad_norm": 1.6978858709335327,
      "learning_rate": 3.0651340996168583e-06,
      "loss": 0.1182,
      "step": 2740
    },
    {
      "epoch": 18.9328743545611,
      "grad_norm": 3.6564149856567383,
      "learning_rate": 2.8735632183908046e-06,
      "loss": 0.1293,
      "step": 2750
    },
    {
      "epoch": 18.994836488812393,
      "eval_accuracy": 0.9467054263565892,
      "eval_loss": 0.15116389095783234,
      "eval_runtime": 7.8213,
      "eval_samples_per_second": 263.893,
      "eval_steps_per_second": 8.311,
      "step": 2759
    },
    {
      "epoch": 19.001721170395868,
      "grad_norm": 3.7127535343170166,
      "learning_rate": 2.6819923371647512e-06,
      "loss": 0.1027,
      "step": 2760
    },
    {
      "epoch": 19.070567986230635,
      "grad_norm": 2.170724868774414,
      "learning_rate": 2.4904214559386975e-06,
      "loss": 0.0993,
      "step": 2770
    },
    {
      "epoch": 19.139414802065403,
      "grad_norm": 3.131479501724243,
      "learning_rate": 2.2988505747126437e-06,
      "loss": 0.1378,
      "step": 2780
    },
    {
      "epoch": 19.20826161790017,
      "grad_norm": 1.942002534866333,
      "learning_rate": 2.1072796934865904e-06,
      "loss": 0.1163,
      "step": 2790
    },
    {
      "epoch": 19.27710843373494,
      "grad_norm": 2.8317081928253174,
      "learning_rate": 1.9157088122605362e-06,
      "loss": 0.1467,
      "step": 2800
    },
    {
      "epoch": 19.34595524956971,
      "grad_norm": 1.791185736656189,
      "learning_rate": 1.724137931034483e-06,
      "loss": 0.1139,
      "step": 2810
    },
    {
      "epoch": 19.414802065404476,
      "grad_norm": 2.639554023742676,
      "learning_rate": 1.5325670498084292e-06,
      "loss": 0.1139,
      "step": 2820
    },
    {
      "epoch": 19.483648881239244,
      "grad_norm": 1.9929583072662354,
      "learning_rate": 1.3409961685823756e-06,
      "loss": 0.1093,
      "step": 2830
    },
    {
      "epoch": 19.55249569707401,
      "grad_norm": 2.7852931022644043,
      "learning_rate": 1.1494252873563219e-06,
      "loss": 0.1305,
      "step": 2840
    },
    {
      "epoch": 19.62134251290878,
      "grad_norm": 3.228508710861206,
      "learning_rate": 9.578544061302681e-07,
      "loss": 0.1236,
      "step": 2850
    },
    {
      "epoch": 19.690189328743546,
      "grad_norm": 3.5496957302093506,
      "learning_rate": 7.662835249042146e-07,
      "loss": 0.1357,
      "step": 2860
    },
    {
      "epoch": 19.759036144578314,
      "grad_norm": 1.8435531854629517,
      "learning_rate": 5.747126436781609e-07,
      "loss": 0.1134,
      "step": 2870
    },
    {
      "epoch": 19.82788296041308,
      "grad_norm": 3.3433897495269775,
      "learning_rate": 3.831417624521073e-07,
      "loss": 0.1472,
      "step": 2880
    },
    {
      "epoch": 19.89672977624785,
      "grad_norm": 2.541207790374756,
      "learning_rate": 1.9157088122605365e-07,
      "loss": 0.1057,
      "step": 2890
    },
    {
      "epoch": 19.965576592082616,
      "grad_norm": 3.152078151702881,
      "learning_rate": 0.0,
      "loss": 0.1072,
      "step": 2900
    },
    {
      "epoch": 19.965576592082616,
      "eval_accuracy": 0.9447674418604651,
      "eval_loss": 0.16629785299301147,
      "eval_runtime": 7.3483,
      "eval_samples_per_second": 280.88,
      "eval_steps_per_second": 8.846,
      "step": 2900
    },
    {
      "epoch": 19.965576592082616,
      "step": 2900,
      "total_flos": 9.813099581032366e+17,
      "train_loss": 0.3346556580683281,
      "train_runtime": 3422.7405,
      "train_samples_per_second": 108.533,
      "train_steps_per_second": 0.847
    }
  ],
  "logging_steps": 10,
  "max_steps": 2900,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.813099581032366e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}