{
  "best_metric": 0.9876,
  "best_model_checkpoint": "vit-small-patch16-224-finetuned-Cifar10/checkpoint-1053",
  "epoch": 2.9936034115138592,
  "eval_steps": 500,
  "global_step": 1053,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "grad_norm": 10.314929962158203,
      "learning_rate": 4.716981132075472e-06,
      "loss": 2.6416,
      "step": 10
    },
    {
      "epoch": 0.06,
      "grad_norm": 8.337972640991211,
      "learning_rate": 9.433962264150944e-06,
      "loss": 2.3876,
      "step": 20
    },
    {
      "epoch": 0.09,
      "grad_norm": 8.081725120544434,
      "learning_rate": 1.4150943396226415e-05,
      "loss": 1.9616,
      "step": 30
    },
    {
      "epoch": 0.11,
      "grad_norm": 6.572023868560791,
      "learning_rate": 1.8867924528301888e-05,
      "loss": 1.4868,
      "step": 40
    },
    {
      "epoch": 0.14,
      "grad_norm": 6.461695194244385,
      "learning_rate": 2.358490566037736e-05,
      "loss": 0.9903,
      "step": 50
    },
    {
      "epoch": 0.17,
      "grad_norm": 5.515841484069824,
      "learning_rate": 2.830188679245283e-05,
      "loss": 0.7169,
      "step": 60
    },
    {
      "epoch": 0.2,
      "grad_norm": 7.923908710479736,
      "learning_rate": 3.30188679245283e-05,
      "loss": 0.5633,
      "step": 70
    },
    {
      "epoch": 0.23,
      "grad_norm": 5.821664333343506,
      "learning_rate": 3.7735849056603776e-05,
      "loss": 0.4791,
      "step": 80
    },
    {
      "epoch": 0.26,
      "grad_norm": 5.747777938842773,
      "learning_rate": 4.245283018867925e-05,
      "loss": 0.438,
      "step": 90
    },
    {
      "epoch": 0.28,
      "grad_norm": 9.386958122253418,
      "learning_rate": 4.716981132075472e-05,
      "loss": 0.4276,
      "step": 100
    },
    {
      "epoch": 0.31,
      "grad_norm": 5.98386812210083,
      "learning_rate": 4.978880675818374e-05,
      "loss": 0.4085,
      "step": 110
    },
    {
      "epoch": 0.34,
      "grad_norm": 5.977865219116211,
      "learning_rate": 4.9260823653643085e-05,
      "loss": 0.3609,
      "step": 120
    },
    {
      "epoch": 0.37,
      "grad_norm": 5.478827476501465,
      "learning_rate": 4.8732840549102435e-05,
      "loss": 0.346,
      "step": 130
    },
    {
      "epoch": 0.4,
      "grad_norm": 5.904544353485107,
      "learning_rate": 4.820485744456177e-05,
      "loss": 0.4145,
      "step": 140
    },
    {
      "epoch": 0.43,
      "grad_norm": 5.136106967926025,
      "learning_rate": 4.767687434002112e-05,
      "loss": 0.3622,
      "step": 150
    },
    {
      "epoch": 0.45,
      "grad_norm": 6.393275260925293,
      "learning_rate": 4.7148891235480466e-05,
      "loss": 0.3613,
      "step": 160
    },
    {
      "epoch": 0.48,
      "grad_norm": 4.859918117523193,
      "learning_rate": 4.662090813093981e-05,
      "loss": 0.3733,
      "step": 170
    },
    {
      "epoch": 0.51,
      "grad_norm": 4.1030426025390625,
      "learning_rate": 4.609292502639916e-05,
      "loss": 0.3029,
      "step": 180
    },
    {
      "epoch": 0.54,
      "grad_norm": 7.502547264099121,
      "learning_rate": 4.55649419218585e-05,
      "loss": 0.3357,
      "step": 190
    },
    {
      "epoch": 0.57,
      "grad_norm": 5.060296535491943,
      "learning_rate": 4.503695881731785e-05,
      "loss": 0.3161,
      "step": 200
    },
    {
      "epoch": 0.6,
      "grad_norm": 4.308870792388916,
      "learning_rate": 4.45089757127772e-05,
      "loss": 0.3041,
      "step": 210
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.026098251342773,
      "learning_rate": 4.398099260823654e-05,
      "loss": 0.301,
      "step": 220
    },
    {
      "epoch": 0.65,
      "grad_norm": 5.656477928161621,
      "learning_rate": 4.3453009503695884e-05,
      "loss": 0.3717,
      "step": 230
    },
    {
      "epoch": 0.68,
      "grad_norm": 3.4833385944366455,
      "learning_rate": 4.292502639915523e-05,
      "loss": 0.3265,
      "step": 240
    },
    {
      "epoch": 0.71,
      "grad_norm": 5.039575099945068,
      "learning_rate": 4.239704329461457e-05,
      "loss": 0.314,
      "step": 250
    },
    {
      "epoch": 0.74,
      "grad_norm": 6.7204155921936035,
      "learning_rate": 4.186906019007392e-05,
      "loss": 0.3216,
      "step": 260
    },
    {
      "epoch": 0.77,
      "grad_norm": 4.30304479598999,
      "learning_rate": 4.1341077085533265e-05,
      "loss": 0.3166,
      "step": 270
    },
    {
      "epoch": 0.8,
      "grad_norm": 6.109108924865723,
      "learning_rate": 4.081309398099261e-05,
      "loss": 0.3528,
      "step": 280
    },
    {
      "epoch": 0.82,
      "grad_norm": 4.084746837615967,
      "learning_rate": 4.028511087645195e-05,
      "loss": 0.2744,
      "step": 290
    },
    {
      "epoch": 0.85,
      "grad_norm": 5.485677719116211,
      "learning_rate": 3.97571277719113e-05,
      "loss": 0.3019,
      "step": 300
    },
    {
      "epoch": 0.88,
      "grad_norm": 7.279469013214111,
      "learning_rate": 3.9229144667370646e-05,
      "loss": 0.3113,
      "step": 310
    },
    {
      "epoch": 0.91,
      "grad_norm": 6.751004695892334,
      "learning_rate": 3.870116156282999e-05,
      "loss": 0.3205,
      "step": 320
    },
    {
      "epoch": 0.94,
      "grad_norm": 5.167363166809082,
      "learning_rate": 3.817317845828934e-05,
      "loss": 0.2882,
      "step": 330
    },
    {
      "epoch": 0.97,
      "grad_norm": 4.599797248840332,
      "learning_rate": 3.764519535374868e-05,
      "loss": 0.3278,
      "step": 340
    },
    {
      "epoch": 1.0,
      "grad_norm": 6.141386985778809,
      "learning_rate": 3.711721224920803e-05,
      "loss": 0.3129,
      "step": 350
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9818,
      "eval_loss": 0.05345360189676285,
      "eval_runtime": 46.7391,
      "eval_samples_per_second": 106.977,
      "eval_steps_per_second": 3.359,
      "step": 351
    },
    {
      "epoch": 1.02,
      "grad_norm": 2.8357040882110596,
      "learning_rate": 3.658922914466738e-05,
      "loss": 0.2452,
      "step": 360
    },
    {
      "epoch": 1.05,
      "grad_norm": 4.519667148590088,
      "learning_rate": 3.6061246040126714e-05,
      "loss": 0.2713,
      "step": 370
    },
    {
      "epoch": 1.08,
      "grad_norm": 6.6121063232421875,
      "learning_rate": 3.5533262935586064e-05,
      "loss": 0.273,
      "step": 380
    },
    {
      "epoch": 1.11,
      "grad_norm": 4.9773125648498535,
      "learning_rate": 3.500527983104541e-05,
      "loss": 0.2632,
      "step": 390
    },
    {
      "epoch": 1.14,
      "grad_norm": 4.788541316986084,
      "learning_rate": 3.447729672650475e-05,
      "loss": 0.2891,
      "step": 400
    },
    {
      "epoch": 1.17,
      "grad_norm": 3.6516640186309814,
      "learning_rate": 3.3949313621964095e-05,
      "loss": 0.2641,
      "step": 410
    },
    {
      "epoch": 1.19,
      "grad_norm": 4.8205695152282715,
      "learning_rate": 3.3421330517423445e-05,
      "loss": 0.2726,
      "step": 420
    },
    {
      "epoch": 1.22,
      "grad_norm": 4.041383743286133,
      "learning_rate": 3.289334741288279e-05,
      "loss": 0.2792,
      "step": 430
    },
    {
      "epoch": 1.25,
      "grad_norm": 5.766094207763672,
      "learning_rate": 3.236536430834213e-05,
      "loss": 0.2815,
      "step": 440
    },
    {
      "epoch": 1.28,
      "grad_norm": 6.516313076019287,
      "learning_rate": 3.183738120380148e-05,
      "loss": 0.2462,
      "step": 450
    },
    {
      "epoch": 1.31,
      "grad_norm": 5.090723037719727,
      "learning_rate": 3.130939809926082e-05,
      "loss": 0.2547,
      "step": 460
    },
    {
      "epoch": 1.34,
      "grad_norm": 3.82003116607666,
      "learning_rate": 3.078141499472017e-05,
      "loss": 0.2593,
      "step": 470
    },
    {
      "epoch": 1.36,
      "grad_norm": 3.9347455501556396,
      "learning_rate": 3.0253431890179517e-05,
      "loss": 0.2534,
      "step": 480
    },
    {
      "epoch": 1.39,
      "grad_norm": 6.01845121383667,
      "learning_rate": 2.972544878563886e-05,
      "loss": 0.2836,
      "step": 490
    },
    {
      "epoch": 1.42,
      "grad_norm": 4.622760772705078,
      "learning_rate": 2.9197465681098207e-05,
      "loss": 0.2673,
      "step": 500
    },
    {
      "epoch": 1.45,
      "grad_norm": 4.938364505767822,
      "learning_rate": 2.8669482576557548e-05,
      "loss": 0.2527,
      "step": 510
    },
    {
      "epoch": 1.48,
      "grad_norm": 5.388880729675293,
      "learning_rate": 2.8141499472016898e-05,
      "loss": 0.2472,
      "step": 520
    },
    {
      "epoch": 1.51,
      "grad_norm": 4.989612102508545,
      "learning_rate": 2.7613516367476245e-05,
      "loss": 0.2325,
      "step": 530
    },
    {
      "epoch": 1.54,
      "grad_norm": 4.6943254470825195,
      "learning_rate": 2.7085533262935585e-05,
      "loss": 0.2796,
      "step": 540
    },
    {
      "epoch": 1.56,
      "grad_norm": 5.073199272155762,
      "learning_rate": 2.6557550158394935e-05,
      "loss": 0.2162,
      "step": 550
    },
    {
      "epoch": 1.59,
      "grad_norm": 3.7621564865112305,
      "learning_rate": 2.6029567053854276e-05,
      "loss": 0.232,
      "step": 560
    },
    {
      "epoch": 1.62,
      "grad_norm": 3.057474136352539,
      "learning_rate": 2.5501583949313622e-05,
      "loss": 0.2246,
      "step": 570
    },
    {
      "epoch": 1.65,
      "grad_norm": 5.149780750274658,
      "learning_rate": 2.497360084477297e-05,
      "loss": 0.2469,
      "step": 580
    },
    {
      "epoch": 1.68,
      "grad_norm": 6.816042900085449,
      "learning_rate": 2.4445617740232313e-05,
      "loss": 0.2461,
      "step": 590
    },
    {
      "epoch": 1.71,
      "grad_norm": 4.830414772033691,
      "learning_rate": 2.391763463569166e-05,
      "loss": 0.2551,
      "step": 600
    },
    {
      "epoch": 1.73,
      "grad_norm": 4.188309669494629,
      "learning_rate": 2.3389651531151003e-05,
      "loss": 0.2388,
      "step": 610
    },
    {
      "epoch": 1.76,
      "grad_norm": 7.241832733154297,
      "learning_rate": 2.286166842661035e-05,
      "loss": 0.2674,
      "step": 620
    },
    {
      "epoch": 1.79,
      "grad_norm": 4.772040843963623,
      "learning_rate": 2.2333685322069694e-05,
      "loss": 0.2537,
      "step": 630
    },
    {
      "epoch": 1.82,
      "grad_norm": 4.274710655212402,
      "learning_rate": 2.180570221752904e-05,
      "loss": 0.231,
      "step": 640
    },
    {
      "epoch": 1.85,
      "grad_norm": 4.281772613525391,
      "learning_rate": 2.1277719112988384e-05,
      "loss": 0.3039,
      "step": 650
    },
    {
      "epoch": 1.88,
      "grad_norm": 5.059981822967529,
      "learning_rate": 2.074973600844773e-05,
      "loss": 0.2894,
      "step": 660
    },
    {
      "epoch": 1.9,
      "grad_norm": 3.4362783432006836,
      "learning_rate": 2.0221752903907075e-05,
      "loss": 0.2272,
      "step": 670
    },
    {
      "epoch": 1.93,
      "grad_norm": 4.354649066925049,
      "learning_rate": 1.9693769799366422e-05,
      "loss": 0.2172,
      "step": 680
    },
    {
      "epoch": 1.96,
      "grad_norm": 3.579586982727051,
      "learning_rate": 1.9165786694825765e-05,
      "loss": 0.2248,
      "step": 690
    },
    {
      "epoch": 1.99,
      "grad_norm": 4.853511333465576,
      "learning_rate": 1.863780359028511e-05,
      "loss": 0.2333,
      "step": 700
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9836,
      "eval_loss": 0.04917434602975845,
      "eval_runtime": 46.8032,
      "eval_samples_per_second": 106.83,
      "eval_steps_per_second": 3.354,
      "step": 703
    },
    {
      "epoch": 2.02,
      "grad_norm": 5.036158084869385,
      "learning_rate": 1.810982048574446e-05,
      "loss": 0.2369,
      "step": 710
    },
    {
      "epoch": 2.05,
      "grad_norm": 4.580857753753662,
      "learning_rate": 1.7581837381203803e-05,
      "loss": 0.2235,
      "step": 720
    },
    {
      "epoch": 2.08,
      "grad_norm": 3.7654311656951904,
      "learning_rate": 1.7053854276663146e-05,
      "loss": 0.2304,
      "step": 730
    },
    {
      "epoch": 2.1,
      "grad_norm": 4.297289848327637,
      "learning_rate": 1.6525871172122493e-05,
      "loss": 0.2179,
      "step": 740
    },
    {
      "epoch": 2.13,
      "grad_norm": 5.170669078826904,
      "learning_rate": 1.5997888067581837e-05,
      "loss": 0.213,
      "step": 750
    },
    {
      "epoch": 2.16,
      "grad_norm": 4.770534038543701,
      "learning_rate": 1.5469904963041184e-05,
      "loss": 0.2245,
      "step": 760
    },
    {
      "epoch": 2.19,
      "grad_norm": 2.8574471473693848,
      "learning_rate": 1.4941921858500529e-05,
      "loss": 0.1713,
      "step": 770
    },
    {
      "epoch": 2.22,
      "grad_norm": 3.895998954772949,
      "learning_rate": 1.4413938753959874e-05,
      "loss": 0.205,
      "step": 780
    },
    {
      "epoch": 2.25,
      "grad_norm": 2.9967050552368164,
      "learning_rate": 1.388595564941922e-05,
      "loss": 0.2204,
      "step": 790
    },
    {
      "epoch": 2.27,
      "grad_norm": 4.339881420135498,
      "learning_rate": 1.3357972544878563e-05,
      "loss": 0.1817,
      "step": 800
    },
    {
      "epoch": 2.3,
      "grad_norm": 5.125692844390869,
      "learning_rate": 1.2829989440337912e-05,
      "loss": 0.2181,
      "step": 810
    },
    {
      "epoch": 2.33,
      "grad_norm": 6.453378200531006,
      "learning_rate": 1.2302006335797255e-05,
      "loss": 0.2058,
      "step": 820
    },
    {
      "epoch": 2.36,
      "grad_norm": 4.746486663818359,
      "learning_rate": 1.17740232312566e-05,
      "loss": 0.213,
      "step": 830
    },
    {
      "epoch": 2.39,
      "grad_norm": 5.431849956512451,
      "learning_rate": 1.1246040126715946e-05,
      "loss": 0.2257,
      "step": 840
    },
    {
      "epoch": 2.42,
      "grad_norm": 6.360583305358887,
      "learning_rate": 1.0718057022175291e-05,
      "loss": 0.224,
      "step": 850
    },
    {
      "epoch": 2.44,
      "grad_norm": 2.9380533695220947,
      "learning_rate": 1.0190073917634636e-05,
      "loss": 0.2245,
      "step": 860
    },
    {
      "epoch": 2.47,
      "grad_norm": 5.245874881744385,
      "learning_rate": 9.662090813093982e-06,
      "loss": 0.1948,
      "step": 870
    },
    {
      "epoch": 2.5,
      "grad_norm": 4.233717918395996,
      "learning_rate": 9.134107708553327e-06,
      "loss": 0.2169,
      "step": 880
    },
    {
      "epoch": 2.53,
      "grad_norm": 4.133506774902344,
      "learning_rate": 8.606124604012672e-06,
      "loss": 0.1894,
      "step": 890
    },
    {
      "epoch": 2.56,
      "grad_norm": 3.6981563568115234,
      "learning_rate": 8.078141499472017e-06,
      "loss": 0.2162,
      "step": 900
    },
    {
      "epoch": 2.59,
      "grad_norm": 5.028733730316162,
      "learning_rate": 7.5501583949313625e-06,
      "loss": 0.2192,
      "step": 910
    },
    {
      "epoch": 2.62,
      "grad_norm": 3.4768264293670654,
      "learning_rate": 7.022175290390708e-06,
      "loss": 0.1761,
      "step": 920
    },
    {
      "epoch": 2.64,
      "grad_norm": 5.03364896774292,
      "learning_rate": 6.494192185850054e-06,
      "loss": 0.1679,
      "step": 930
    },
    {
      "epoch": 2.67,
      "grad_norm": 4.1295671463012695,
      "learning_rate": 5.966209081309398e-06,
      "loss": 0.2324,
      "step": 940
    },
    {
      "epoch": 2.7,
      "grad_norm": 4.6029744148254395,
      "learning_rate": 5.438225976768744e-06,
      "loss": 0.2357,
      "step": 950
    },
    {
      "epoch": 2.73,
      "grad_norm": 4.893083572387695,
      "learning_rate": 4.910242872228089e-06,
      "loss": 0.1903,
      "step": 960
    },
    {
      "epoch": 2.76,
      "grad_norm": 4.711769104003906,
      "learning_rate": 4.382259767687434e-06,
      "loss": 0.2094,
      "step": 970
    },
    {
      "epoch": 2.79,
      "grad_norm": 4.935668468475342,
      "learning_rate": 3.854276663146779e-06,
      "loss": 0.2066,
      "step": 980
    },
    {
      "epoch": 2.81,
      "grad_norm": 4.085482597351074,
      "learning_rate": 3.326293558606125e-06,
      "loss": 0.2225,
      "step": 990
    },
    {
      "epoch": 2.84,
      "grad_norm": 4.599144458770752,
      "learning_rate": 2.79831045406547e-06,
      "loss": 0.1858,
      "step": 1000
    },
    {
      "epoch": 2.87,
      "grad_norm": 3.7395975589752197,
      "learning_rate": 2.2703273495248154e-06,
      "loss": 0.1868,
      "step": 1010
    },
    {
      "epoch": 2.9,
      "grad_norm": 3.5131924152374268,
      "learning_rate": 1.7423442449841606e-06,
      "loss": 0.2092,
      "step": 1020
    },
    {
      "epoch": 2.93,
      "grad_norm": 3.5327999591827393,
      "learning_rate": 1.2143611404435059e-06,
      "loss": 0.2194,
      "step": 1030
    },
    {
      "epoch": 2.96,
      "grad_norm": 4.314801216125488,
      "learning_rate": 6.863780359028511e-07,
      "loss": 0.2106,
      "step": 1040
    },
    {
      "epoch": 2.99,
      "grad_norm": 2.447665214538574,
      "learning_rate": 1.5839493136219642e-07,
      "loss": 0.1991,
      "step": 1050
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.9876,
      "eval_loss": 0.037825874984264374,
      "eval_runtime": 46.7089,
      "eval_samples_per_second": 107.046,
      "eval_steps_per_second": 3.361,
      "step": 1053
    },
    {
      "epoch": 2.99,
      "step": 1053,
      "total_flos": 2.636946768879747e+18,
      "train_loss": 0.3487827978242836,
      "train_runtime": 1772.9517,
      "train_samples_per_second": 76.144,
      "train_steps_per_second": 0.594
    }
  ],
  "logging_steps": 10,
  "max_steps": 1053,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2.636946768879747e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}