{
  "best_metric": 0.28460967540740967,
  "best_model_checkpoint": "food-recognition\\checkpoint-1200",
  "epoch": 4.0,
  "global_step": 1876,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 0.00019893390191897656,
      "loss": 2.1429,
      "step": 10
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001978678038379531,
      "loss": 1.7212,
      "step": 20
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019680170575692965,
      "loss": 1.4819,
      "step": 30
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001957356076759062,
      "loss": 1.0959,
      "step": 40
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019466950959488275,
      "loss": 0.948,
      "step": 50
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019360341151385927,
      "loss": 0.8738,
      "step": 60
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019253731343283584,
      "loss": 0.7065,
      "step": 70
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001914712153518124,
      "loss": 0.7459,
      "step": 80
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001904051172707889,
      "loss": 0.8923,
      "step": 90
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00018933901918976546,
      "loss": 0.6555,
      "step": 100
    },
    {
      "epoch": 0.21,
      "eval_accuracy": 0.7908,
      "eval_loss": 0.7220542430877686,
      "eval_runtime": 867.5544,
      "eval_samples_per_second": 2.882,
      "eval_steps_per_second": 0.361,
      "step": 100
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00018827292110874203,
      "loss": 0.8138,
      "step": 110
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00018720682302771858,
      "loss": 0.8166,
      "step": 120
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001861407249466951,
      "loss": 0.7293,
      "step": 130
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018507462686567165,
      "loss": 0.7375,
      "step": 140
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001840085287846482,
      "loss": 0.6812,
      "step": 150
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00018294243070362474,
      "loss": 0.6028,
      "step": 160
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001818763326226013,
      "loss": 0.7157,
      "step": 170
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018081023454157784,
      "loss": 0.6799,
      "step": 180
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00017974413646055436,
      "loss": 0.6852,
      "step": 190
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00017867803837953093,
      "loss": 0.5506,
      "step": 200
    },
    {
      "epoch": 0.43,
      "eval_accuracy": 0.8036,
      "eval_loss": 0.6188839673995972,
      "eval_runtime": 857.0334,
      "eval_samples_per_second": 2.917,
      "eval_steps_per_second": 0.365,
      "step": 200
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017761194029850748,
      "loss": 0.6133,
      "step": 210
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017654584221748402,
      "loss": 0.6582,
      "step": 220
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017547974413646055,
      "loss": 0.5733,
      "step": 230
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017441364605543712,
      "loss": 0.5838,
      "step": 240
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017334754797441367,
      "loss": 0.6761,
      "step": 250
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001722814498933902,
      "loss": 0.6022,
      "step": 260
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00017121535181236673,
      "loss": 0.607,
      "step": 270
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00017014925373134328,
      "loss": 0.7063,
      "step": 280
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016908315565031986,
      "loss": 0.524,
      "step": 290
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00016801705756929638,
      "loss": 0.5857,
      "step": 300
    },
    {
      "epoch": 0.64,
      "eval_accuracy": 0.874,
      "eval_loss": 0.4339218735694885,
      "eval_runtime": 915.5554,
      "eval_samples_per_second": 2.731,
      "eval_steps_per_second": 0.342,
      "step": 300
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00016695095948827292,
      "loss": 0.6595,
      "step": 310
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00016588486140724947,
      "loss": 0.4687,
      "step": 320
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00016481876332622602,
      "loss": 0.466,
      "step": 330
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016375266524520257,
      "loss": 0.4687,
      "step": 340
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016268656716417911,
      "loss": 0.5287,
      "step": 350
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00016162046908315566,
      "loss": 0.6968,
      "step": 360
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001605543710021322,
      "loss": 0.4682,
      "step": 370
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00015948827292110876,
      "loss": 0.5498,
      "step": 380
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001584221748400853,
      "loss": 0.3831,
      "step": 390
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00015735607675906182,
      "loss": 0.4369,
      "step": 400
    },
    {
      "epoch": 0.85,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.39879345893859863,
      "eval_runtime": 916.1264,
      "eval_samples_per_second": 2.729,
      "eval_steps_per_second": 0.342,
      "step": 400
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001562899786780384,
      "loss": 0.6071,
      "step": 410
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00015522388059701495,
      "loss": 0.4741,
      "step": 420
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001541577825159915,
      "loss": 0.49,
      "step": 430
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.000153091684434968,
      "loss": 0.5432,
      "step": 440
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00015202558635394456,
      "loss": 0.428,
      "step": 450
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015095948827292113,
      "loss": 0.5663,
      "step": 460
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00014989339019189766,
      "loss": 0.4064,
      "step": 470
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001488272921108742,
      "loss": 0.2211,
      "step": 480
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00014776119402985075,
      "loss": 0.2891,
      "step": 490
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001466950959488273,
      "loss": 0.2657,
      "step": 500
    },
    {
      "epoch": 1.07,
      "eval_accuracy": 0.9044,
      "eval_loss": 0.31406229734420776,
      "eval_runtime": 914.3094,
      "eval_samples_per_second": 2.734,
      "eval_steps_per_second": 0.342,
      "step": 500
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00014562899786780384,
      "loss": 0.2387,
      "step": 510
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001445628997867804,
      "loss": 0.2216,
      "step": 520
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00014349680170575694,
      "loss": 0.3339,
      "step": 530
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001424307036247335,
      "loss": 0.2512,
      "step": 540
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00014136460554371003,
      "loss": 0.2456,
      "step": 550
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014029850746268658,
      "loss": 0.2838,
      "step": 560
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001392324093816631,
      "loss": 0.2279,
      "step": 570
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00013816631130063965,
      "loss": 0.3115,
      "step": 580
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00013710021321961622,
      "loss": 0.1885,
      "step": 590
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00013603411513859277,
      "loss": 0.2781,
      "step": 600
    },
    {
      "epoch": 1.28,
      "eval_accuracy": 0.886,
      "eval_loss": 0.37785816192626953,
      "eval_runtime": 914.7964,
      "eval_samples_per_second": 2.733,
      "eval_steps_per_second": 0.342,
      "step": 600
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001349680170575693,
      "loss": 0.2358,
      "step": 610
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013390191897654584,
      "loss": 0.2658,
      "step": 620
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001328358208955224,
      "loss": 0.3546,
      "step": 630
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013176972281449893,
      "loss": 0.2499,
      "step": 640
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00013070362473347548,
      "loss": 0.1833,
      "step": 650
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00012963752665245203,
      "loss": 0.3323,
      "step": 660
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00012857142857142858,
      "loss": 0.2828,
      "step": 670
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00012750533049040512,
      "loss": 0.2448,
      "step": 680
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00012643923240938167,
      "loss": 0.2711,
      "step": 690
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00012537313432835822,
      "loss": 0.3139,
      "step": 700
    },
    {
      "epoch": 1.49,
      "eval_accuracy": 0.896,
      "eval_loss": 0.3323199152946472,
      "eval_runtime": 913.8415,
      "eval_samples_per_second": 2.736,
      "eval_steps_per_second": 0.343,
      "step": 700
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012430703624733474,
      "loss": 0.292,
      "step": 710
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0001232409381663113,
      "loss": 0.1341,
      "step": 720
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012217484008528786,
      "loss": 0.2711,
      "step": 730
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012110874200426441,
      "loss": 0.2587,
      "step": 740
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00012004264392324094,
      "loss": 0.2444,
      "step": 750
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011897654584221749,
      "loss": 0.1548,
      "step": 760
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00011791044776119405,
      "loss": 0.3112,
      "step": 770
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011684434968017057,
      "loss": 0.1468,
      "step": 780
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00011577825159914713,
      "loss": 0.2688,
      "step": 790
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011471215351812368,
      "loss": 0.4175,
      "step": 800
    },
    {
      "epoch": 1.71,
      "eval_accuracy": 0.894,
      "eval_loss": 0.35160574316978455,
      "eval_runtime": 917.6163,
      "eval_samples_per_second": 2.724,
      "eval_steps_per_second": 0.341,
      "step": 800
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00011364605543710023,
      "loss": 0.2959,
      "step": 810
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011257995735607676,
      "loss": 0.2509,
      "step": 820
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001115138592750533,
      "loss": 0.2657,
      "step": 830
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011044776119402987,
      "loss": 0.2101,
      "step": 840
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00010938166311300639,
      "loss": 0.3044,
      "step": 850
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00010831556503198295,
      "loss": 0.1844,
      "step": 860
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0001072494669509595,
      "loss": 0.371,
      "step": 870
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00010618336886993603,
      "loss": 0.1634,
      "step": 880
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00010511727078891258,
      "loss": 0.251,
      "step": 890
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00010405117270788914,
      "loss": 0.1843,
      "step": 900
    },
    {
      "epoch": 1.92,
      "eval_accuracy": 0.8928,
      "eval_loss": 0.3689097762107849,
      "eval_runtime": 912.1597,
      "eval_samples_per_second": 2.741,
      "eval_steps_per_second": 0.343,
      "step": 900
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010298507462686569,
      "loss": 0.3085,
      "step": 910
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00010191897654584222,
      "loss": 0.2526,
      "step": 920
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010085287846481877,
      "loss": 0.2225,
      "step": 930
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.978678038379531e-05,
      "loss": 0.192,
      "step": 940
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.872068230277186e-05,
      "loss": 0.1397,
      "step": 950
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.76545842217484e-05,
      "loss": 0.1629,
      "step": 960
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.658848614072496e-05,
      "loss": 0.0754,
      "step": 970
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.552238805970149e-05,
      "loss": 0.0599,
      "step": 980
    },
    {
      "epoch": 2.11,
      "learning_rate": 9.445628997867805e-05,
      "loss": 0.0572,
      "step": 990
    },
    {
      "epoch": 2.13,
      "learning_rate": 9.339019189765458e-05,
      "loss": 0.0439,
      "step": 1000
    },
    {
      "epoch": 2.13,
      "eval_accuracy": 0.9164,
      "eval_loss": 0.2847430109977722,
      "eval_runtime": 914.5678,
      "eval_samples_per_second": 2.734,
      "eval_steps_per_second": 0.342,
      "step": 1000
    },
    {
      "epoch": 2.15,
      "learning_rate": 9.232409381663113e-05,
      "loss": 0.0583,
      "step": 1010
    },
    {
      "epoch": 2.17,
      "learning_rate": 9.125799573560768e-05,
      "loss": 0.0555,
      "step": 1020
    },
    {
      "epoch": 2.2,
      "learning_rate": 9.019189765458423e-05,
      "loss": 0.1279,
      "step": 1030
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.912579957356077e-05,
      "loss": 0.0637,
      "step": 1040
    },
    {
      "epoch": 2.24,
      "learning_rate": 8.805970149253732e-05,
      "loss": 0.071,
      "step": 1050
    },
    {
      "epoch": 2.26,
      "learning_rate": 8.699360341151387e-05,
      "loss": 0.1323,
      "step": 1060
    },
    {
      "epoch": 2.28,
      "learning_rate": 8.59275053304904e-05,
      "loss": 0.1609,
      "step": 1070
    },
    {
      "epoch": 2.3,
      "learning_rate": 8.486140724946695e-05,
      "loss": 0.1272,
      "step": 1080
    },
    {
      "epoch": 2.32,
      "learning_rate": 8.37953091684435e-05,
      "loss": 0.0395,
      "step": 1090
    },
    {
      "epoch": 2.35,
      "learning_rate": 8.272921108742005e-05,
      "loss": 0.0908,
      "step": 1100
    },
    {
      "epoch": 2.35,
      "eval_accuracy": 0.916,
      "eval_loss": 0.29704228043556213,
      "eval_runtime": 912.498,
      "eval_samples_per_second": 2.74,
      "eval_steps_per_second": 0.343,
      "step": 1100
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.166311300639659e-05,
      "loss": 0.0603,
      "step": 1110
    },
    {
      "epoch": 2.39,
      "learning_rate": 8.059701492537314e-05,
      "loss": 0.0813,
      "step": 1120
    },
    {
      "epoch": 2.41,
      "learning_rate": 7.953091684434967e-05,
      "loss": 0.1628,
      "step": 1130
    },
    {
      "epoch": 2.43,
      "learning_rate": 7.846481876332623e-05,
      "loss": 0.1359,
      "step": 1140
    },
    {
      "epoch": 2.45,
      "learning_rate": 7.739872068230277e-05,
      "loss": 0.042,
      "step": 1150
    },
    {
      "epoch": 2.47,
      "learning_rate": 7.633262260127933e-05,
      "loss": 0.019,
      "step": 1160
    },
    {
      "epoch": 2.49,
      "learning_rate": 7.526652452025586e-05,
      "loss": 0.0905,
      "step": 1170
    },
    {
      "epoch": 2.52,
      "learning_rate": 7.420042643923242e-05,
      "loss": 0.1037,
      "step": 1180
    },
    {
      "epoch": 2.54,
      "learning_rate": 7.313432835820896e-05,
      "loss": 0.1283,
      "step": 1190
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.20682302771855e-05,
      "loss": 0.0872,
      "step": 1200
    },
    {
      "epoch": 2.56,
      "eval_accuracy": 0.9216,
      "eval_loss": 0.28460967540740967,
      "eval_runtime": 912.1266,
      "eval_samples_per_second": 2.741,
      "eval_steps_per_second": 0.343,
      "step": 1200
    },
    {
      "epoch": 2.58,
      "learning_rate": 7.100213219616205e-05,
      "loss": 0.0881,
      "step": 1210
    },
    {
      "epoch": 2.6,
      "learning_rate": 6.993603411513859e-05,
      "loss": 0.058,
      "step": 1220
    },
    {
      "epoch": 2.62,
      "learning_rate": 6.886993603411515e-05,
      "loss": 0.0702,
      "step": 1230
    },
    {
      "epoch": 2.64,
      "learning_rate": 6.780383795309168e-05,
      "loss": 0.0148,
      "step": 1240
    },
    {
      "epoch": 2.67,
      "learning_rate": 6.673773987206824e-05,
      "loss": 0.0423,
      "step": 1250
    },
    {
      "epoch": 2.69,
      "learning_rate": 6.567164179104478e-05,
      "loss": 0.0413,
      "step": 1260
    },
    {
      "epoch": 2.71,
      "learning_rate": 6.460554371002132e-05,
      "loss": 0.1166,
      "step": 1270
    },
    {
      "epoch": 2.73,
      "learning_rate": 6.353944562899787e-05,
      "loss": 0.1308,
      "step": 1280
    },
    {
      "epoch": 2.75,
      "learning_rate": 6.247334754797442e-05,
      "loss": 0.0173,
      "step": 1290
    },
    {
      "epoch": 2.77,
      "learning_rate": 6.140724946695097e-05,
      "loss": 0.0758,
      "step": 1300
    },
    {
      "epoch": 2.77,
      "eval_accuracy": 0.9192,
      "eval_loss": 0.2986711859703064,
      "eval_runtime": 914.7466,
      "eval_samples_per_second": 2.733,
      "eval_steps_per_second": 0.342,
      "step": 1300
    },
    {
      "epoch": 2.79,
      "learning_rate": 6.0341151385927506e-05,
      "loss": 0.1306,
      "step": 1310
    },
    {
      "epoch": 2.81,
      "learning_rate": 5.927505330490405e-05,
      "loss": 0.1587,
      "step": 1320
    },
    {
      "epoch": 2.84,
      "learning_rate": 5.82089552238806e-05,
      "loss": 0.0657,
      "step": 1330
    },
    {
      "epoch": 2.86,
      "learning_rate": 5.714285714285714e-05,
      "loss": 0.0525,
      "step": 1340
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.6076759061833696e-05,
      "loss": 0.1687,
      "step": 1350
    },
    {
      "epoch": 2.9,
      "learning_rate": 5.5010660980810236e-05,
      "loss": 0.0847,
      "step": 1360
    },
    {
      "epoch": 2.92,
      "learning_rate": 5.394456289978679e-05,
      "loss": 0.038,
      "step": 1370
    },
    {
      "epoch": 2.94,
      "learning_rate": 5.287846481876333e-05,
      "loss": 0.0669,
      "step": 1380
    },
    {
      "epoch": 2.96,
      "learning_rate": 5.181236673773987e-05,
      "loss": 0.1829,
      "step": 1390
    },
    {
      "epoch": 2.99,
      "learning_rate": 5.074626865671642e-05,
      "loss": 0.0494,
      "step": 1400
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.92,
      "eval_loss": 0.3099476099014282,
      "eval_runtime": 913.041,
      "eval_samples_per_second": 2.738,
      "eval_steps_per_second": 0.343,
      "step": 1400
    },
    {
      "epoch": 3.01,
      "learning_rate": 4.9680170575692967e-05,
      "loss": 0.0385,
      "step": 1410
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.861407249466951e-05,
      "loss": 0.0572,
      "step": 1420
    },
    {
      "epoch": 3.05,
      "learning_rate": 4.7547974413646055e-05,
      "loss": 0.0297,
      "step": 1430
    },
    {
      "epoch": 3.07,
      "learning_rate": 4.64818763326226e-05,
      "loss": 0.0093,
      "step": 1440
    },
    {
      "epoch": 3.09,
      "learning_rate": 4.541577825159915e-05,
      "loss": 0.0223,
      "step": 1450
    },
    {
      "epoch": 3.11,
      "learning_rate": 4.43496801705757e-05,
      "loss": 0.0195,
      "step": 1460
    },
    {
      "epoch": 3.13,
      "learning_rate": 4.328358208955224e-05,
      "loss": 0.0113,
      "step": 1470
    },
    {
      "epoch": 3.16,
      "learning_rate": 4.2217484008528785e-05,
      "loss": 0.0112,
      "step": 1480
    },
    {
      "epoch": 3.18,
      "learning_rate": 4.115138592750533e-05,
      "loss": 0.0226,
      "step": 1490
    },
    {
      "epoch": 3.2,
      "learning_rate": 4.008528784648188e-05,
      "loss": 0.047,
      "step": 1500
    },
    {
      "epoch": 3.2,
      "eval_accuracy": 0.9256,
      "eval_loss": 0.29152050614356995,
      "eval_runtime": 912.8232,
      "eval_samples_per_second": 2.739,
      "eval_steps_per_second": 0.343,
      "step": 1500
    },
    {
      "epoch": 3.22,
      "learning_rate": 3.901918976545843e-05,
      "loss": 0.0108,
      "step": 1510
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.7953091684434974e-05,
      "loss": 0.0141,
      "step": 1520
    },
    {
      "epoch": 3.26,
      "learning_rate": 3.6886993603411515e-05,
      "loss": 0.0134,
      "step": 1530
    },
    {
      "epoch": 3.28,
      "learning_rate": 3.582089552238806e-05,
      "loss": 0.0254,
      "step": 1540
    },
    {
      "epoch": 3.3,
      "learning_rate": 3.47547974413646e-05,
      "loss": 0.0134,
      "step": 1550
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.368869936034115e-05,
      "loss": 0.0069,
      "step": 1560
    },
    {
      "epoch": 3.35,
      "learning_rate": 3.26226012793177e-05,
      "loss": 0.016,
      "step": 1570
    },
    {
      "epoch": 3.37,
      "learning_rate": 3.1556503198294245e-05,
      "loss": 0.0075,
      "step": 1580
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.0490405117270792e-05,
      "loss": 0.0464,
      "step": 1590
    },
    {
      "epoch": 3.41,
      "learning_rate": 2.9424307036247333e-05,
      "loss": 0.0089,
      "step": 1600
    },
    {
      "epoch": 3.41,
      "eval_accuracy": 0.928,
      "eval_loss": 0.2917356789112091,
      "eval_runtime": 915.7964,
      "eval_samples_per_second": 2.73,
      "eval_steps_per_second": 0.342,
      "step": 1600
    },
    {
      "epoch": 3.43,
      "learning_rate": 2.835820895522388e-05,
      "loss": 0.0069,
      "step": 1610
    },
    {
      "epoch": 3.45,
      "learning_rate": 2.7292110874200428e-05,
      "loss": 0.0071,
      "step": 1620
    },
    {
      "epoch": 3.48,
      "learning_rate": 2.6226012793176975e-05,
      "loss": 0.0239,
      "step": 1630
    },
    {
      "epoch": 3.5,
      "learning_rate": 2.515991471215352e-05,
      "loss": 0.0091,
      "step": 1640
    },
    {
      "epoch": 3.52,
      "learning_rate": 2.4093816631130063e-05,
      "loss": 0.0081,
      "step": 1650
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.302771855010661e-05,
      "loss": 0.0101,
      "step": 1660
    },
    {
      "epoch": 3.56,
      "learning_rate": 2.1961620469083158e-05,
      "loss": 0.0062,
      "step": 1670
    },
    {
      "epoch": 3.58,
      "learning_rate": 2.0895522388059702e-05,
      "loss": 0.0227,
      "step": 1680
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.982942430703625e-05,
      "loss": 0.0104,
      "step": 1690
    },
    {
      "epoch": 3.62,
      "learning_rate": 1.8763326226012797e-05,
      "loss": 0.0881,
      "step": 1700
    },
    {
      "epoch": 3.62,
      "eval_accuracy": 0.9284,
      "eval_loss": 0.2847883701324463,
      "eval_runtime": 914.1128,
      "eval_samples_per_second": 2.735,
      "eval_steps_per_second": 0.342,
      "step": 1700
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.769722814498934e-05,
      "loss": 0.0363,
      "step": 1710
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.6631130063965885e-05,
      "loss": 0.0076,
      "step": 1720
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.5565031982942432e-05,
      "loss": 0.0061,
      "step": 1730
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.4498933901918976e-05,
      "loss": 0.0054,
      "step": 1740
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.3432835820895523e-05,
      "loss": 0.029,
      "step": 1750
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.2366737739872069e-05,
      "loss": 0.0258,
      "step": 1760
    },
    {
      "epoch": 3.77,
      "learning_rate": 1.1300639658848615e-05,
      "loss": 0.0277,
      "step": 1770
    },
    {
      "epoch": 3.8,
      "learning_rate": 1.023454157782516e-05,
      "loss": 0.007,
      "step": 1780
    },
    {
      "epoch": 3.82,
      "learning_rate": 9.168443496801706e-06,
      "loss": 0.0478,
      "step": 1790
    },
    {
      "epoch": 3.84,
      "learning_rate": 8.102345415778252e-06,
      "loss": 0.025,
      "step": 1800
    },
    {
      "epoch": 3.84,
      "eval_accuracy": 0.9324,
      "eval_loss": 0.28532394766807556,
      "eval_runtime": 909.5622,
      "eval_samples_per_second": 2.749,
      "eval_steps_per_second": 0.344,
      "step": 1800
    },
    {
      "epoch": 3.86,
      "learning_rate": 7.0362473347547975e-06,
      "loss": 0.0062,
      "step": 1810
    },
    {
      "epoch": 3.88,
      "learning_rate": 5.970149253731343e-06,
      "loss": 0.0062,
      "step": 1820
    },
    {
      "epoch": 3.9,
      "learning_rate": 4.90405117270789e-06,
      "loss": 0.0079,
      "step": 1830
    },
    {
      "epoch": 3.92,
      "learning_rate": 3.837953091684435e-06,
      "loss": 0.0331,
      "step": 1840
    },
    {
      "epoch": 3.94,
      "learning_rate": 2.771855010660981e-06,
      "loss": 0.0059,
      "step": 1850
    },
    {
      "epoch": 3.97,
      "learning_rate": 1.7057569296375267e-06,
      "loss": 0.0081,
      "step": 1860
    },
    {
      "epoch": 3.99,
      "learning_rate": 6.396588486140725e-07,
      "loss": 0.006,
      "step": 1870
    },
    {
      "epoch": 4.0,
      "step": 1876,
      "total_flos": 0.0,
      "train_loss": 0.2667588123968288,
      "train_runtime": 47969.5224,
      "train_samples_per_second": 0.625,
      "train_steps_per_second": 0.039
    }
  ],
  "max_steps": 1876,
  "num_train_epochs": 4,
  "total_flos": 0.0,
  "trial_name": null,
  "trial_params": null
}