{
  "best_metric": 0.16356664896011353,
  "best_model_checkpoint": "./vit-indian-food/checkpoint-500",
  "epoch": 2.6178010471204187,
  "eval_steps": 50,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "grad_norm": 6.967112064361572,
      "learning_rate": 0.00019895287958115185,
      "loss": 2.3423,
      "step": 5
    },
    {
      "epoch": 0.05,
      "grad_norm": 6.768218517303467,
      "learning_rate": 0.00019790575916230367,
      "loss": 1.5082,
      "step": 10
    },
    {
      "epoch": 0.08,
      "grad_norm": 6.283707141876221,
      "learning_rate": 0.0001968586387434555,
      "loss": 1.1555,
      "step": 15
    },
    {
      "epoch": 0.1,
      "grad_norm": 4.167559623718262,
      "learning_rate": 0.00019581151832460733,
      "loss": 0.7045,
      "step": 20
    },
    {
      "epoch": 0.13,
      "grad_norm": 6.519095420837402,
      "learning_rate": 0.00019476439790575917,
      "loss": 0.5942,
      "step": 25
    },
    {
      "epoch": 0.16,
      "grad_norm": 4.0483479499816895,
      "learning_rate": 0.000193717277486911,
      "loss": 0.45,
      "step": 30
    },
    {
      "epoch": 0.18,
      "grad_norm": 4.359745025634766,
      "learning_rate": 0.00019267015706806283,
      "loss": 0.3833,
      "step": 35
    },
    {
      "epoch": 0.21,
      "grad_norm": 5.201784610748291,
      "learning_rate": 0.00019162303664921465,
      "loss": 0.3702,
      "step": 40
    },
    {
      "epoch": 0.24,
      "grad_norm": 4.259886741638184,
      "learning_rate": 0.0001905759162303665,
      "loss": 0.3227,
      "step": 45
    },
    {
      "epoch": 0.26,
      "grad_norm": 4.89500617980957,
      "learning_rate": 0.00018952879581151833,
      "loss": 0.2368,
      "step": 50
    },
    {
      "epoch": 0.26,
      "eval_accuracy": 0.9002624671916011,
      "eval_loss": 0.33144867420196533,
      "eval_precision": 0.9131344269428404,
      "eval_recall": 0.9002624671916011,
      "eval_runtime": 17.436,
      "eval_samples_per_second": 43.703,
      "eval_steps_per_second": 2.753,
      "step": 50
    },
    {
      "epoch": 0.29,
      "grad_norm": 4.650336265563965,
      "learning_rate": 0.00018848167539267018,
      "loss": 0.4139,
      "step": 55
    },
    {
      "epoch": 0.31,
      "grad_norm": 4.881960391998291,
      "learning_rate": 0.00018743455497382202,
      "loss": 0.2109,
      "step": 60
    },
    {
      "epoch": 0.34,
      "grad_norm": 5.3574066162109375,
      "learning_rate": 0.00018638743455497384,
      "loss": 0.3076,
      "step": 65
    },
    {
      "epoch": 0.37,
      "grad_norm": 5.008607864379883,
      "learning_rate": 0.00018534031413612568,
      "loss": 0.2632,
      "step": 70
    },
    {
      "epoch": 0.39,
      "grad_norm": 4.561398506164551,
      "learning_rate": 0.0001842931937172775,
      "loss": 0.2062,
      "step": 75
    },
    {
      "epoch": 0.42,
      "grad_norm": 5.034663677215576,
      "learning_rate": 0.00018324607329842934,
      "loss": 0.4752,
      "step": 80
    },
    {
      "epoch": 0.45,
      "grad_norm": 8.143318176269531,
      "learning_rate": 0.00018219895287958115,
      "loss": 0.3963,
      "step": 85
    },
    {
      "epoch": 0.47,
      "grad_norm": 6.556373119354248,
      "learning_rate": 0.000181151832460733,
      "loss": 0.5088,
      "step": 90
    },
    {
      "epoch": 0.5,
      "grad_norm": 4.414945125579834,
      "learning_rate": 0.0001801047120418848,
      "loss": 0.2903,
      "step": 95
    },
    {
      "epoch": 0.52,
      "grad_norm": 2.464193820953369,
      "learning_rate": 0.00017905759162303666,
      "loss": 0.2801,
      "step": 100
    },
    {
      "epoch": 0.52,
      "eval_accuracy": 0.9422572178477691,
      "eval_loss": 0.18048594892024994,
      "eval_precision": 0.9466471604125616,
      "eval_recall": 0.9422572178477691,
      "eval_runtime": 17.2463,
      "eval_samples_per_second": 44.183,
      "eval_steps_per_second": 2.783,
      "step": 100
    },
    {
      "epoch": 0.55,
      "grad_norm": 5.096113681793213,
      "learning_rate": 0.0001780104712041885,
      "loss": 0.213,
      "step": 105
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.4251294136047363,
      "learning_rate": 0.00017696335078534032,
      "loss": 0.1831,
      "step": 110
    },
    {
      "epoch": 0.6,
      "grad_norm": 3.766965627670288,
      "learning_rate": 0.00017591623036649216,
      "loss": 0.1636,
      "step": 115
    },
    {
      "epoch": 0.63,
      "grad_norm": 3.66446852684021,
      "learning_rate": 0.00017486910994764398,
      "loss": 0.3075,
      "step": 120
    },
    {
      "epoch": 0.65,
      "grad_norm": 4.702181339263916,
      "learning_rate": 0.00017382198952879582,
      "loss": 0.2472,
      "step": 125
    },
    {
      "epoch": 0.68,
      "grad_norm": 8.463685035705566,
      "learning_rate": 0.00017277486910994763,
      "loss": 0.4341,
      "step": 130
    },
    {
      "epoch": 0.71,
      "grad_norm": 5.947971343994141,
      "learning_rate": 0.00017172774869109948,
      "loss": 0.2291,
      "step": 135
    },
    {
      "epoch": 0.73,
      "grad_norm": 5.682934761047363,
      "learning_rate": 0.00017068062827225132,
      "loss": 0.1632,
      "step": 140
    },
    {
      "epoch": 0.76,
      "grad_norm": 9.100614547729492,
      "learning_rate": 0.00016963350785340316,
      "loss": 0.307,
      "step": 145
    },
    {
      "epoch": 0.79,
      "grad_norm": 4.207605361938477,
      "learning_rate": 0.00016858638743455498,
      "loss": 0.2362,
      "step": 150
    },
    {
      "epoch": 0.79,
      "eval_accuracy": 0.9186351706036745,
      "eval_loss": 0.29145774245262146,
      "eval_precision": 0.9273784378254858,
      "eval_recall": 0.9186351706036745,
      "eval_runtime": 17.4226,
      "eval_samples_per_second": 43.736,
      "eval_steps_per_second": 2.755,
      "step": 150
    },
    {
      "epoch": 0.81,
      "grad_norm": 3.4303929805755615,
      "learning_rate": 0.00016753926701570682,
      "loss": 0.074,
      "step": 155
    },
    {
      "epoch": 0.84,
      "grad_norm": 4.852244853973389,
      "learning_rate": 0.00016649214659685867,
      "loss": 0.2731,
      "step": 160
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.3142894506454468,
      "learning_rate": 0.00016544502617801048,
      "loss": 0.1295,
      "step": 165
    },
    {
      "epoch": 0.89,
      "grad_norm": 3.892892360687256,
      "learning_rate": 0.00016439790575916233,
      "loss": 0.225,
      "step": 170
    },
    {
      "epoch": 0.92,
      "grad_norm": 4.833043575286865,
      "learning_rate": 0.00016335078534031414,
      "loss": 0.2674,
      "step": 175
    },
    {
      "epoch": 0.94,
      "grad_norm": 5.150189399719238,
      "learning_rate": 0.00016230366492146599,
      "loss": 0.241,
      "step": 180
    },
    {
      "epoch": 0.97,
      "grad_norm": 5.426727771759033,
      "learning_rate": 0.0001612565445026178,
      "loss": 0.2049,
      "step": 185
    },
    {
      "epoch": 0.99,
      "grad_norm": 5.2064738273620605,
      "learning_rate": 0.00016020942408376964,
      "loss": 0.1756,
      "step": 190
    },
    {
      "epoch": 1.02,
      "grad_norm": 2.034318685531616,
      "learning_rate": 0.00015916230366492146,
      "loss": 0.0499,
      "step": 195
    },
    {
      "epoch": 1.05,
      "grad_norm": 4.4703145027160645,
      "learning_rate": 0.0001581151832460733,
      "loss": 0.042,
      "step": 200
    },
    {
      "epoch": 1.05,
      "eval_accuracy": 0.9461942257217848,
      "eval_loss": 0.16130226850509644,
      "eval_precision": 0.9511635531837818,
      "eval_recall": 0.9461942257217848,
      "eval_runtime": 17.3623,
      "eval_samples_per_second": 43.888,
      "eval_steps_per_second": 2.765,
      "step": 200
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.06716866791248322,
      "learning_rate": 0.00015706806282722515,
      "loss": 0.0643,
      "step": 205
    },
    {
      "epoch": 1.1,
      "grad_norm": 7.002842903137207,
      "learning_rate": 0.00015602094240837696,
      "loss": 0.1028,
      "step": 210
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.013943655416369438,
      "learning_rate": 0.0001549738219895288,
      "loss": 0.0941,
      "step": 215
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.42214348912239075,
      "learning_rate": 0.00015392670157068062,
      "loss": 0.0273,
      "step": 220
    },
    {
      "epoch": 1.18,
      "grad_norm": 8.428805351257324,
      "learning_rate": 0.00015287958115183247,
      "loss": 0.0589,
      "step": 225
    },
    {
      "epoch": 1.2,
      "grad_norm": 2.468398094177246,
      "learning_rate": 0.00015183246073298428,
      "loss": 0.0321,
      "step": 230
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.8661732077598572,
      "learning_rate": 0.00015078534031413612,
      "loss": 0.0856,
      "step": 235
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.4541776478290558,
      "learning_rate": 0.00014973821989528797,
      "loss": 0.0136,
      "step": 240
    },
    {
      "epoch": 1.28,
      "grad_norm": 4.832011699676514,
      "learning_rate": 0.0001486910994764398,
      "loss": 0.1087,
      "step": 245
    },
    {
      "epoch": 1.31,
      "grad_norm": 4.744746685028076,
      "learning_rate": 0.00014764397905759163,
      "loss": 0.0477,
      "step": 250
    },
    {
      "epoch": 1.31,
      "eval_accuracy": 0.9514435695538058,
      "eval_loss": 0.14496225118637085,
      "eval_precision": 0.9534067355318784,
      "eval_recall": 0.9514435695538058,
      "eval_runtime": 17.4463,
      "eval_samples_per_second": 43.677,
      "eval_steps_per_second": 2.751,
      "step": 250
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.053365129977464676,
      "learning_rate": 0.00014659685863874347,
      "loss": 0.0171,
      "step": 255
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.41375842690467834,
      "learning_rate": 0.00014554973821989531,
      "loss": 0.1189,
      "step": 260
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.05034720525145531,
      "learning_rate": 0.00014450261780104713,
      "loss": 0.0377,
      "step": 265
    },
    {
      "epoch": 1.41,
      "grad_norm": 2.498413324356079,
      "learning_rate": 0.00014345549738219897,
      "loss": 0.0334,
      "step": 270
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.06190189719200134,
      "learning_rate": 0.0001424083769633508,
      "loss": 0.0158,
      "step": 275
    },
    {
      "epoch": 1.47,
      "grad_norm": 1.9862781763076782,
      "learning_rate": 0.00014136125654450263,
      "loss": 0.0186,
      "step": 280
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.047805577516555786,
      "learning_rate": 0.00014031413612565445,
      "loss": 0.0705,
      "step": 285
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.03765374794602394,
      "learning_rate": 0.0001392670157068063,
      "loss": 0.1822,
      "step": 290
    },
    {
      "epoch": 1.54,
      "grad_norm": 5.020593643188477,
      "learning_rate": 0.0001382198952879581,
      "loss": 0.0375,
      "step": 295
    },
    {
      "epoch": 1.57,
      "grad_norm": 1.0980048179626465,
      "learning_rate": 0.00013717277486910995,
      "loss": 0.1297,
      "step": 300
    },
    {
      "epoch": 1.57,
      "eval_accuracy": 0.937007874015748,
      "eval_loss": 0.24536970257759094,
      "eval_precision": 0.9442846541005356,
      "eval_recall": 0.937007874015748,
      "eval_runtime": 17.5121,
      "eval_samples_per_second": 43.513,
      "eval_steps_per_second": 2.741,
      "step": 300
    },
    {
      "epoch": 1.6,
      "grad_norm": 3.535846471786499,
      "learning_rate": 0.0001361256544502618,
      "loss": 0.1071,
      "step": 305
    },
    {
      "epoch": 1.62,
      "grad_norm": 4.182868957519531,
      "learning_rate": 0.0001350785340314136,
      "loss": 0.0249,
      "step": 310
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.36884206533432007,
      "learning_rate": 0.00013403141361256545,
      "loss": 0.0762,
      "step": 315
    },
    {
      "epoch": 1.68,
      "grad_norm": 5.14253044128418,
      "learning_rate": 0.00013298429319371727,
      "loss": 0.0766,
      "step": 320
    },
    {
      "epoch": 1.7,
      "grad_norm": 1.2552727460861206,
      "learning_rate": 0.0001319371727748691,
      "loss": 0.0093,
      "step": 325
    },
    {
      "epoch": 1.73,
      "grad_norm": 3.767280101776123,
      "learning_rate": 0.00013089005235602096,
      "loss": 0.2232,
      "step": 330
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.010386645793914795,
      "learning_rate": 0.00012984293193717277,
      "loss": 0.0026,
      "step": 335
    },
    {
      "epoch": 1.78,
      "grad_norm": 4.292264461517334,
      "learning_rate": 0.00012879581151832462,
      "loss": 0.2457,
      "step": 340
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.024649567902088165,
      "learning_rate": 0.00012774869109947646,
      "loss": 0.0306,
      "step": 345
    },
    {
      "epoch": 1.83,
      "grad_norm": 5.098654270172119,
      "learning_rate": 0.00012670157068062827,
      "loss": 0.1156,
      "step": 350
    },
    {
      "epoch": 1.83,
      "eval_accuracy": 0.9501312335958005,
      "eval_loss": 0.17998863756656647,
      "eval_precision": 0.9531660720469141,
      "eval_recall": 0.9501312335958005,
      "eval_runtime": 17.4136,
      "eval_samples_per_second": 43.759,
      "eval_steps_per_second": 2.756,
      "step": 350
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.9651967287063599,
      "learning_rate": 0.00012565445026178012,
      "loss": 0.0618,
      "step": 355
    },
    {
      "epoch": 1.88,
      "grad_norm": 2.727735996246338,
      "learning_rate": 0.00012460732984293196,
      "loss": 0.0705,
      "step": 360
    },
    {
      "epoch": 1.91,
      "grad_norm": 4.953824043273926,
      "learning_rate": 0.00012356020942408378,
      "loss": 0.2112,
      "step": 365
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.08269181102514267,
      "learning_rate": 0.00012251308900523562,
      "loss": 0.071,
      "step": 370
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.06396706402301788,
      "learning_rate": 0.00012146596858638744,
      "loss": 0.1069,
      "step": 375
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.28295230865478516,
      "learning_rate": 0.00012041884816753928,
      "loss": 0.0759,
      "step": 380
    },
    {
      "epoch": 2.02,
      "grad_norm": 1.5805819034576416,
      "learning_rate": 0.0001193717277486911,
      "loss": 0.0622,
      "step": 385
    },
    {
      "epoch": 2.04,
      "grad_norm": 2.3917598724365234,
      "learning_rate": 0.00011832460732984294,
      "loss": 0.0512,
      "step": 390
    },
    {
      "epoch": 2.07,
      "grad_norm": 0.03971586003899574,
      "learning_rate": 0.00011727748691099475,
      "loss": 0.0163,
      "step": 395
    },
    {
      "epoch": 2.09,
      "grad_norm": 0.024471383541822433,
      "learning_rate": 0.0001162303664921466,
      "loss": 0.0188,
      "step": 400
    },
    {
      "epoch": 2.09,
      "eval_accuracy": 0.9593175853018373,
      "eval_loss": 0.14536795020103455,
      "eval_precision": 0.9601421810596481,
      "eval_recall": 0.9593175853018373,
      "eval_runtime": 17.4,
      "eval_samples_per_second": 43.793,
      "eval_steps_per_second": 2.759,
      "step": 400
    },
    {
      "epoch": 2.12,
      "grad_norm": 0.0705149918794632,
      "learning_rate": 0.00011518324607329844,
      "loss": 0.0062,
      "step": 405
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.013097544200718403,
      "learning_rate": 0.00011413612565445027,
      "loss": 0.0191,
      "step": 410
    },
    {
      "epoch": 2.17,
      "grad_norm": 0.5236541628837585,
      "learning_rate": 0.00011308900523560211,
      "loss": 0.0403,
      "step": 415
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.01967461034655571,
      "learning_rate": 0.00011204188481675393,
      "loss": 0.0032,
      "step": 420
    },
    {
      "epoch": 2.23,
      "grad_norm": 0.02198900654911995,
      "learning_rate": 0.00011099476439790577,
      "loss": 0.0019,
      "step": 425
    },
    {
      "epoch": 2.25,
      "grad_norm": 0.14719614386558533,
      "learning_rate": 0.00010994764397905759,
      "loss": 0.0052,
      "step": 430
    },
    {
      "epoch": 2.28,
      "grad_norm": 0.021793629974126816,
      "learning_rate": 0.00010890052356020943,
      "loss": 0.0105,
      "step": 435
    },
    {
      "epoch": 2.3,
      "grad_norm": 0.02368028089404106,
      "learning_rate": 0.00010785340314136125,
      "loss": 0.001,
      "step": 440
    },
    {
      "epoch": 2.33,
      "grad_norm": 0.010441082529723644,
      "learning_rate": 0.00010680628272251309,
      "loss": 0.0019,
      "step": 445
    },
    {
      "epoch": 2.36,
      "grad_norm": 0.015402965247631073,
      "learning_rate": 0.00010575916230366492,
      "loss": 0.0049,
      "step": 450
    },
    {
      "epoch": 2.36,
      "eval_accuracy": 0.9711286089238845,
      "eval_loss": 0.1082761362195015,
      "eval_precision": 0.9714506772867647,
      "eval_recall": 0.9711286089238845,
      "eval_runtime": 17.6808,
      "eval_samples_per_second": 43.098,
      "eval_steps_per_second": 2.715,
      "step": 450
    },
    {
      "epoch": 2.38,
      "grad_norm": 0.07379986345767975,
      "learning_rate": 0.00010471204188481676,
      "loss": 0.0047,
      "step": 455
    },
    {
      "epoch": 2.41,
      "grad_norm": 5.459177494049072,
      "learning_rate": 0.0001036649214659686,
      "loss": 0.0192,
      "step": 460
    },
    {
      "epoch": 2.43,
      "grad_norm": 0.004949438851326704,
      "learning_rate": 0.00010261780104712042,
      "loss": 0.0148,
      "step": 465
    },
    {
      "epoch": 2.46,
      "grad_norm": 0.17624743282794952,
      "learning_rate": 0.00010157068062827227,
      "loss": 0.0087,
      "step": 470
    },
    {
      "epoch": 2.49,
      "grad_norm": 0.010408706963062286,
      "learning_rate": 0.00010052356020942408,
      "loss": 0.0018,
      "step": 475
    },
    {
      "epoch": 2.51,
      "grad_norm": 0.007278731558471918,
      "learning_rate": 9.947643979057593e-05,
      "loss": 0.0029,
      "step": 480
    },
    {
      "epoch": 2.54,
      "grad_norm": 0.01558419968932867,
      "learning_rate": 9.842931937172776e-05,
      "loss": 0.0549,
      "step": 485
    },
    {
      "epoch": 2.57,
      "grad_norm": 0.005123753100633621,
      "learning_rate": 9.738219895287959e-05,
      "loss": 0.0006,
      "step": 490
    },
    {
      "epoch": 2.59,
      "grad_norm": 0.011126354336738586,
      "learning_rate": 9.633507853403142e-05,
      "loss": 0.003,
      "step": 495
    },
    {
      "epoch": 2.62,
      "grad_norm": 0.02012033388018608,
      "learning_rate": 9.528795811518324e-05,
      "loss": 0.0139,
      "step": 500
    },
    {
      "epoch": 2.62,
      "eval_accuracy": 0.963254593175853,
      "eval_loss": 0.16356664896011353,
      "eval_precision": 0.9644146563432678,
      "eval_recall": 0.963254593175853,
      "eval_runtime": 17.6855,
      "eval_samples_per_second": 43.086,
      "eval_steps_per_second": 2.714,
      "step": 500
    }
  ],
  "logging_steps": 5,
  "max_steps": 955,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 6.186131303461724e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}