{
  "best_metric": 0.7444126074498567,
  "best_model_checkpoint": "vit-Facial-Expression-Recognition/checkpoint-1962",
  "epoch": 4.9980894153611,
  "eval_steps": 500,
  "global_step": 3270,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 1.5290519877675841e-06,
      "loss": 1.9371,
      "step": 10
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.0581039755351682e-06,
      "loss": 1.8225,
      "step": 20
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.587155963302753e-06,
      "loss": 1.4076,
      "step": 30
    },
    {
      "epoch": 0.06,
      "learning_rate": 6.1162079510703365e-06,
      "loss": 1.5662,
      "step": 40
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.645259938837921e-06,
      "loss": 1.3985,
      "step": 50
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.174311926605506e-06,
      "loss": 1.1727,
      "step": 60
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.070336391437309e-05,
      "loss": 1.1874,
      "step": 70
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.2232415902140673e-05,
      "loss": 1.1256,
      "step": 80
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.3761467889908258e-05,
      "loss": 1.0438,
      "step": 90
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.5290519877675842e-05,
      "loss": 1.1966,
      "step": 100
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6819571865443427e-05,
      "loss": 1.0475,
      "step": 110
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.834862385321101e-05,
      "loss": 0.9658,
      "step": 120
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.9877675840978592e-05,
      "loss": 0.959,
      "step": 130
    },
    {
      "epoch": 0.21,
      "learning_rate": 2.140672782874618e-05,
      "loss": 0.7963,
      "step": 140
    },
    {
      "epoch": 0.23,
      "learning_rate": 2.2935779816513765e-05,
      "loss": 0.9615,
      "step": 150
    },
    {
      "epoch": 0.24,
      "learning_rate": 2.4464831804281346e-05,
      "loss": 0.8296,
      "step": 160
    },
    {
      "epoch": 0.26,
      "learning_rate": 2.5993883792048927e-05,
      "loss": 0.8144,
      "step": 170
    },
    {
      "epoch": 0.28,
      "learning_rate": 2.7522935779816515e-05,
      "loss": 0.9023,
      "step": 180
    },
    {
      "epoch": 0.29,
      "learning_rate": 2.90519877675841e-05,
      "loss": 0.9142,
      "step": 190
    },
    {
      "epoch": 0.31,
      "learning_rate": 3.0581039755351684e-05,
      "loss": 0.9456,
      "step": 200
    },
    {
      "epoch": 0.32,
      "learning_rate": 3.211009174311927e-05,
      "loss": 0.8358,
      "step": 210
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.363914373088685e-05,
      "loss": 0.7678,
      "step": 220
    },
    {
      "epoch": 0.35,
      "learning_rate": 3.516819571865443e-05,
      "loss": 0.8491,
      "step": 230
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.669724770642202e-05,
      "loss": 0.7184,
      "step": 240
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.822629969418961e-05,
      "loss": 0.767,
      "step": 250
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.9755351681957185e-05,
      "loss": 0.7829,
      "step": 260
    },
    {
      "epoch": 0.41,
      "learning_rate": 4.1284403669724776e-05,
      "loss": 0.76,
      "step": 270
    },
    {
      "epoch": 0.43,
      "learning_rate": 4.281345565749236e-05,
      "loss": 0.8151,
      "step": 280
    },
    {
      "epoch": 0.44,
      "learning_rate": 4.434250764525994e-05,
      "loss": 0.7488,
      "step": 290
    },
    {
      "epoch": 0.46,
      "learning_rate": 4.587155963302753e-05,
      "loss": 0.6966,
      "step": 300
    },
    {
      "epoch": 0.47,
      "learning_rate": 4.740061162079511e-05,
      "loss": 0.9042,
      "step": 310
    },
    {
      "epoch": 0.49,
      "learning_rate": 4.892966360856269e-05,
      "loss": 0.7496,
      "step": 320
    },
    {
      "epoch": 0.5,
      "learning_rate": 4.994903160040775e-05,
      "loss": 0.7599,
      "step": 330
    },
    {
      "epoch": 0.52,
      "learning_rate": 4.977913693510024e-05,
      "loss": 0.8631,
      "step": 340
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.9609242269792734e-05,
      "loss": 0.8586,
      "step": 350
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.943934760448522e-05,
      "loss": 0.7269,
      "step": 360
    },
    {
      "epoch": 0.57,
      "learning_rate": 4.926945293917771e-05,
      "loss": 0.8573,
      "step": 370
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.9099558273870206e-05,
      "loss": 0.7636,
      "step": 380
    },
    {
      "epoch": 0.6,
      "learning_rate": 4.892966360856269e-05,
      "loss": 0.7227,
      "step": 390
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.8759768943255185e-05,
      "loss": 0.7689,
      "step": 400
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.858987427794767e-05,
      "loss": 0.6832,
      "step": 410
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.8419979612640164e-05,
      "loss": 0.7225,
      "step": 420
    },
    {
      "epoch": 0.66,
      "learning_rate": 4.8250084947332657e-05,
      "loss": 0.8064,
      "step": 430
    },
    {
      "epoch": 0.67,
      "learning_rate": 4.808019028202514e-05,
      "loss": 0.6557,
      "step": 440
    },
    {
      "epoch": 0.69,
      "learning_rate": 4.7910295616717635e-05,
      "loss": 0.7788,
      "step": 450
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.774040095141013e-05,
      "loss": 0.7525,
      "step": 460
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.7570506286102614e-05,
      "loss": 0.7951,
      "step": 470
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.740061162079511e-05,
      "loss": 0.8923,
      "step": 480
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.72307169554876e-05,
      "loss": 0.7103,
      "step": 490
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.706082229018009e-05,
      "loss": 0.6821,
      "step": 500
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.6890927624872586e-05,
      "loss": 0.7814,
      "step": 510
    },
    {
      "epoch": 0.79,
      "learning_rate": 4.672103295956507e-05,
      "loss": 0.7257,
      "step": 520
    },
    {
      "epoch": 0.81,
      "learning_rate": 4.6551138294257565e-05,
      "loss": 0.7441,
      "step": 530
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.638124362895006e-05,
      "loss": 0.7044,
      "step": 540
    },
    {
      "epoch": 0.84,
      "learning_rate": 4.6211348963642544e-05,
      "loss": 0.779,
      "step": 550
    },
    {
      "epoch": 0.86,
      "learning_rate": 4.604145429833504e-05,
      "loss": 0.8438,
      "step": 560
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.587155963302753e-05,
      "loss": 0.7342,
      "step": 570
    },
    {
      "epoch": 0.89,
      "learning_rate": 4.5701664967720016e-05,
      "loss": 0.6886,
      "step": 580
    },
    {
      "epoch": 0.9,
      "learning_rate": 4.553177030241251e-05,
      "loss": 0.7733,
      "step": 590
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.5361875637104995e-05,
      "loss": 0.7655,
      "step": 600
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.519198097179749e-05,
      "loss": 0.6929,
      "step": 610
    },
    {
      "epoch": 0.95,
      "learning_rate": 4.502208630648998e-05,
      "loss": 0.788,
      "step": 620
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.4852191641182466e-05,
      "loss": 0.6691,
      "step": 630
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.468229697587496e-05,
      "loss": 0.6881,
      "step": 640
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.451240231056745e-05,
      "loss": 0.7175,
      "step": 650
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7308500477554919,
      "eval_loss": 0.7081200480461121,
      "eval_runtime": 48.365,
      "eval_samples_per_second": 108.24,
      "eval_steps_per_second": 13.543,
      "step": 654
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.434250764525994e-05,
      "loss": 0.6844,
      "step": 660
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.417261297995243e-05,
      "loss": 0.7542,
      "step": 670
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.400271831464492e-05,
      "loss": 0.6111,
      "step": 680
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.383282364933741e-05,
      "loss": 0.6169,
      "step": 690
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.36629289840299e-05,
      "loss": 0.7111,
      "step": 700
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.349303431872239e-05,
      "loss": 0.732,
      "step": 710
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.332313965341488e-05,
      "loss": 0.623,
      "step": 720
    },
    {
      "epoch": 1.12,
      "learning_rate": 4.3153244988107375e-05,
      "loss": 0.6272,
      "step": 730
    },
    {
      "epoch": 1.13,
      "learning_rate": 4.298335032279987e-05,
      "loss": 0.6331,
      "step": 740
    },
    {
      "epoch": 1.15,
      "learning_rate": 4.281345565749236e-05,
      "loss": 0.6036,
      "step": 750
    },
    {
      "epoch": 1.16,
      "learning_rate": 4.264356099218485e-05,
      "loss": 0.6559,
      "step": 760
    },
    {
      "epoch": 1.18,
      "learning_rate": 4.247366632687734e-05,
      "loss": 0.5873,
      "step": 770
    },
    {
      "epoch": 1.19,
      "learning_rate": 4.230377166156983e-05,
      "loss": 0.6016,
      "step": 780
    },
    {
      "epoch": 1.21,
      "learning_rate": 4.213387699626232e-05,
      "loss": 0.6399,
      "step": 790
    },
    {
      "epoch": 1.22,
      "learning_rate": 4.196398233095481e-05,
      "loss": 0.6719,
      "step": 800
    },
    {
      "epoch": 1.24,
      "learning_rate": 4.1794087665647304e-05,
      "loss": 0.5458,
      "step": 810
    },
    {
      "epoch": 1.25,
      "learning_rate": 4.162419300033979e-05,
      "loss": 0.6167,
      "step": 820
    },
    {
      "epoch": 1.27,
      "learning_rate": 4.145429833503228e-05,
      "loss": 0.6865,
      "step": 830
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.1284403669724776e-05,
      "loss": 0.6527,
      "step": 840
    },
    {
      "epoch": 1.3,
      "learning_rate": 4.111450900441726e-05,
      "loss": 0.5855,
      "step": 850
    },
    {
      "epoch": 1.31,
      "learning_rate": 4.0944614339109755e-05,
      "loss": 0.5976,
      "step": 860
    },
    {
      "epoch": 1.33,
      "learning_rate": 4.077471967380224e-05,
      "loss": 0.6886,
      "step": 870
    },
    {
      "epoch": 1.35,
      "learning_rate": 4.0604825008494734e-05,
      "loss": 0.6992,
      "step": 880
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.043493034318723e-05,
      "loss": 0.6413,
      "step": 890
    },
    {
      "epoch": 1.38,
      "learning_rate": 4.026503567787971e-05,
      "loss": 0.6227,
      "step": 900
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.0095141012572206e-05,
      "loss": 0.611,
      "step": 910
    },
    {
      "epoch": 1.41,
      "learning_rate": 3.99252463472647e-05,
      "loss": 0.5908,
      "step": 920
    },
    {
      "epoch": 1.42,
      "learning_rate": 3.9755351681957185e-05,
      "loss": 0.6177,
      "step": 930
    },
    {
      "epoch": 1.44,
      "learning_rate": 3.958545701664968e-05,
      "loss": 0.6077,
      "step": 940
    },
    {
      "epoch": 1.45,
      "learning_rate": 3.941556235134217e-05,
      "loss": 0.6321,
      "step": 950
    },
    {
      "epoch": 1.47,
      "learning_rate": 3.9245667686034656e-05,
      "loss": 0.5838,
      "step": 960
    },
    {
      "epoch": 1.48,
      "learning_rate": 3.907577302072715e-05,
      "loss": 0.6619,
      "step": 970
    },
    {
      "epoch": 1.5,
      "learning_rate": 3.890587835541964e-05,
      "loss": 0.5823,
      "step": 980
    },
    {
      "epoch": 1.51,
      "learning_rate": 3.8735983690112135e-05,
      "loss": 0.5496,
      "step": 990
    },
    {
      "epoch": 1.53,
      "learning_rate": 3.856608902480463e-05,
      "loss": 0.6632,
      "step": 1000
    },
    {
      "epoch": 1.54,
      "learning_rate": 3.8396194359497114e-05,
      "loss": 0.5744,
      "step": 1010
    },
    {
      "epoch": 1.56,
      "learning_rate": 3.822629969418961e-05,
      "loss": 0.6259,
      "step": 1020
    },
    {
      "epoch": 1.57,
      "learning_rate": 3.80564050288821e-05,
      "loss": 0.6156,
      "step": 1030
    },
    {
      "epoch": 1.59,
      "learning_rate": 3.7886510363574586e-05,
      "loss": 0.5593,
      "step": 1040
    },
    {
      "epoch": 1.6,
      "learning_rate": 3.771661569826708e-05,
      "loss": 0.6951,
      "step": 1050
    },
    {
      "epoch": 1.62,
      "learning_rate": 3.7546721032959565e-05,
      "loss": 0.6099,
      "step": 1060
    },
    {
      "epoch": 1.64,
      "learning_rate": 3.737682636765206e-05,
      "loss": 0.6019,
      "step": 1070
    },
    {
      "epoch": 1.65,
      "learning_rate": 3.720693170234455e-05,
      "loss": 0.5226,
      "step": 1080
    },
    {
      "epoch": 1.67,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.6539,
      "step": 1090
    },
    {
      "epoch": 1.68,
      "learning_rate": 3.686714237172953e-05,
      "loss": 0.6031,
      "step": 1100
    },
    {
      "epoch": 1.7,
      "learning_rate": 3.669724770642202e-05,
      "loss": 0.6356,
      "step": 1110
    },
    {
      "epoch": 1.71,
      "learning_rate": 3.652735304111451e-05,
      "loss": 0.6456,
      "step": 1120
    },
    {
      "epoch": 1.73,
      "learning_rate": 3.6357458375807e-05,
      "loss": 0.5222,
      "step": 1130
    },
    {
      "epoch": 1.74,
      "learning_rate": 3.6187563710499494e-05,
      "loss": 0.5715,
      "step": 1140
    },
    {
      "epoch": 1.76,
      "learning_rate": 3.601766904519198e-05,
      "loss": 0.6142,
      "step": 1150
    },
    {
      "epoch": 1.77,
      "learning_rate": 3.584777437988447e-05,
      "loss": 0.6794,
      "step": 1160
    },
    {
      "epoch": 1.79,
      "learning_rate": 3.567787971457696e-05,
      "loss": 0.7228,
      "step": 1170
    },
    {
      "epoch": 1.8,
      "learning_rate": 3.550798504926945e-05,
      "loss": 0.6374,
      "step": 1180
    },
    {
      "epoch": 1.82,
      "learning_rate": 3.5338090383961945e-05,
      "loss": 0.6078,
      "step": 1190
    },
    {
      "epoch": 1.83,
      "learning_rate": 3.516819571865443e-05,
      "loss": 0.6115,
      "step": 1200
    },
    {
      "epoch": 1.85,
      "learning_rate": 3.4998301053346924e-05,
      "loss": 0.6853,
      "step": 1210
    },
    {
      "epoch": 1.86,
      "learning_rate": 3.482840638803942e-05,
      "loss": 0.5883,
      "step": 1220
    },
    {
      "epoch": 1.88,
      "learning_rate": 3.465851172273191e-05,
      "loss": 0.6143,
      "step": 1230
    },
    {
      "epoch": 1.9,
      "learning_rate": 3.44886170574244e-05,
      "loss": 0.6067,
      "step": 1240
    },
    {
      "epoch": 1.91,
      "learning_rate": 3.431872239211689e-05,
      "loss": 0.64,
      "step": 1250
    },
    {
      "epoch": 1.93,
      "learning_rate": 3.414882772680938e-05,
      "loss": 0.6043,
      "step": 1260
    },
    {
      "epoch": 1.94,
      "learning_rate": 3.3978933061501874e-05,
      "loss": 0.5755,
      "step": 1270
    },
    {
      "epoch": 1.96,
      "learning_rate": 3.380903839619436e-05,
      "loss": 0.6534,
      "step": 1280
    },
    {
      "epoch": 1.97,
      "learning_rate": 3.363914373088685e-05,
      "loss": 0.5583,
      "step": 1290
    },
    {
      "epoch": 1.99,
      "learning_rate": 3.3469249065579346e-05,
      "loss": 0.6952,
      "step": 1300
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7379178605539637,
      "eval_loss": 0.6931169629096985,
      "eval_runtime": 48.247,
      "eval_samples_per_second": 108.504,
      "eval_steps_per_second": 13.576,
      "step": 1308
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.329935440027183e-05,
      "loss": 0.5479,
      "step": 1310
    },
    {
      "epoch": 2.02,
      "learning_rate": 3.3129459734964325e-05,
      "loss": 0.523,
      "step": 1320
    },
    {
      "epoch": 2.03,
      "learning_rate": 3.295956506965682e-05,
      "loss": 0.4421,
      "step": 1330
    },
    {
      "epoch": 2.05,
      "learning_rate": 3.2789670404349304e-05,
      "loss": 0.4887,
      "step": 1340
    },
    {
      "epoch": 2.06,
      "learning_rate": 3.26197757390418e-05,
      "loss": 0.4665,
      "step": 1350
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.244988107373428e-05,
      "loss": 0.5767,
      "step": 1360
    },
    {
      "epoch": 2.09,
      "learning_rate": 3.2279986408426776e-05,
      "loss": 0.4944,
      "step": 1370
    },
    {
      "epoch": 2.11,
      "learning_rate": 3.211009174311927e-05,
      "loss": 0.4159,
      "step": 1380
    },
    {
      "epoch": 2.12,
      "learning_rate": 3.1940197077811755e-05,
      "loss": 0.3825,
      "step": 1390
    },
    {
      "epoch": 2.14,
      "learning_rate": 3.177030241250425e-05,
      "loss": 0.46,
      "step": 1400
    },
    {
      "epoch": 2.16,
      "learning_rate": 3.160040774719674e-05,
      "loss": 0.4904,
      "step": 1410
    },
    {
      "epoch": 2.17,
      "learning_rate": 3.1430513081889227e-05,
      "loss": 0.5993,
      "step": 1420
    },
    {
      "epoch": 2.19,
      "learning_rate": 3.126061841658172e-05,
      "loss": 0.5107,
      "step": 1430
    },
    {
      "epoch": 2.2,
      "learning_rate": 3.1090723751274206e-05,
      "loss": 0.4616,
      "step": 1440
    },
    {
      "epoch": 2.22,
      "learning_rate": 3.09208290859667e-05,
      "loss": 0.4318,
      "step": 1450
    },
    {
      "epoch": 2.23,
      "learning_rate": 3.075093442065919e-05,
      "loss": 0.4934,
      "step": 1460
    },
    {
      "epoch": 2.25,
      "learning_rate": 3.0581039755351684e-05,
      "loss": 0.4119,
      "step": 1470
    },
    {
      "epoch": 2.26,
      "learning_rate": 3.0411145090044174e-05,
      "loss": 0.5154,
      "step": 1480
    },
    {
      "epoch": 2.28,
      "learning_rate": 3.0241250424736666e-05,
      "loss": 0.5126,
      "step": 1490
    },
    {
      "epoch": 2.29,
      "learning_rate": 3.0071355759429153e-05,
      "loss": 0.429,
      "step": 1500
    },
    {
      "epoch": 2.31,
      "learning_rate": 2.9901461094121645e-05,
      "loss": 0.5008,
      "step": 1510
    },
    {
      "epoch": 2.32,
      "learning_rate": 2.9731566428814138e-05,
      "loss": 0.4074,
      "step": 1520
    },
    {
      "epoch": 2.34,
      "learning_rate": 2.9561671763506628e-05,
      "loss": 0.4563,
      "step": 1530
    },
    {
      "epoch": 2.35,
      "learning_rate": 2.939177709819912e-05,
      "loss": 0.5493,
      "step": 1540
    },
    {
      "epoch": 2.37,
      "learning_rate": 2.9221882432891607e-05,
      "loss": 0.5255,
      "step": 1550
    },
    {
      "epoch": 2.38,
      "learning_rate": 2.90519877675841e-05,
      "loss": 0.4812,
      "step": 1560
    },
    {
      "epoch": 2.4,
      "learning_rate": 2.8882093102276592e-05,
      "loss": 0.4576,
      "step": 1570
    },
    {
      "epoch": 2.41,
      "learning_rate": 2.871219843696908e-05,
      "loss": 0.5143,
      "step": 1580
    },
    {
      "epoch": 2.43,
      "learning_rate": 2.854230377166157e-05,
      "loss": 0.4439,
      "step": 1590
    },
    {
      "epoch": 2.45,
      "learning_rate": 2.8372409106354064e-05,
      "loss": 0.4637,
      "step": 1600
    },
    {
      "epoch": 2.46,
      "learning_rate": 2.820251444104655e-05,
      "loss": 0.4575,
      "step": 1610
    },
    {
      "epoch": 2.48,
      "learning_rate": 2.8032619775739043e-05,
      "loss": 0.4955,
      "step": 1620
    },
    {
      "epoch": 2.49,
      "learning_rate": 2.7862725110431533e-05,
      "loss": 0.446,
      "step": 1630
    },
    {
      "epoch": 2.51,
      "learning_rate": 2.7692830445124026e-05,
      "loss": 0.5241,
      "step": 1640
    },
    {
      "epoch": 2.52,
      "learning_rate": 2.7522935779816515e-05,
      "loss": 0.4872,
      "step": 1650
    },
    {
      "epoch": 2.54,
      "learning_rate": 2.7353041114509004e-05,
      "loss": 0.4526,
      "step": 1660
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.7183146449201497e-05,
      "loss": 0.513,
      "step": 1670
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.701325178389399e-05,
      "loss": 0.5036,
      "step": 1680
    },
    {
      "epoch": 2.58,
      "learning_rate": 2.6843357118586476e-05,
      "loss": 0.4714,
      "step": 1690
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.667346245327897e-05,
      "loss": 0.4228,
      "step": 1700
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.6503567787971462e-05,
      "loss": 0.4916,
      "step": 1710
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.6333673122663948e-05,
      "loss": 0.5439,
      "step": 1720
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.616377845735644e-05,
      "loss": 0.4702,
      "step": 1730
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.5993883792048927e-05,
      "loss": 0.5041,
      "step": 1740
    },
    {
      "epoch": 2.67,
      "learning_rate": 2.582398912674142e-05,
      "loss": 0.4341,
      "step": 1750
    },
    {
      "epoch": 2.69,
      "learning_rate": 2.5654094461433913e-05,
      "loss": 0.4781,
      "step": 1760
    },
    {
      "epoch": 2.71,
      "learning_rate": 2.5484199796126402e-05,
      "loss": 0.4447,
      "step": 1770
    },
    {
      "epoch": 2.72,
      "learning_rate": 2.5314305130818895e-05,
      "loss": 0.424,
      "step": 1780
    },
    {
      "epoch": 2.74,
      "learning_rate": 2.5144410465511388e-05,
      "loss": 0.5021,
      "step": 1790
    },
    {
      "epoch": 2.75,
      "learning_rate": 2.4974515800203874e-05,
      "loss": 0.4491,
      "step": 1800
    },
    {
      "epoch": 2.77,
      "learning_rate": 2.4804621134896367e-05,
      "loss": 0.47,
      "step": 1810
    },
    {
      "epoch": 2.78,
      "learning_rate": 2.4634726469588856e-05,
      "loss": 0.4573,
      "step": 1820
    },
    {
      "epoch": 2.8,
      "learning_rate": 2.4464831804281346e-05,
      "loss": 0.4301,
      "step": 1830
    },
    {
      "epoch": 2.81,
      "learning_rate": 2.4294937138973835e-05,
      "loss": 0.5122,
      "step": 1840
    },
    {
      "epoch": 2.83,
      "learning_rate": 2.4125042473666328e-05,
      "loss": 0.4943,
      "step": 1850
    },
    {
      "epoch": 2.84,
      "learning_rate": 2.3955147808358818e-05,
      "loss": 0.4656,
      "step": 1860
    },
    {
      "epoch": 2.86,
      "learning_rate": 2.3785253143051307e-05,
      "loss": 0.4625,
      "step": 1870
    },
    {
      "epoch": 2.87,
      "learning_rate": 2.36153584777438e-05,
      "loss": 0.5568,
      "step": 1880
    },
    {
      "epoch": 2.89,
      "learning_rate": 2.3445463812436293e-05,
      "loss": 0.3586,
      "step": 1890
    },
    {
      "epoch": 2.9,
      "learning_rate": 2.3275569147128782e-05,
      "loss": 0.5519,
      "step": 1900
    },
    {
      "epoch": 2.92,
      "learning_rate": 2.3105674481821272e-05,
      "loss": 0.5399,
      "step": 1910
    },
    {
      "epoch": 2.93,
      "learning_rate": 2.2935779816513765e-05,
      "loss": 0.4763,
      "step": 1920
    },
    {
      "epoch": 2.95,
      "learning_rate": 2.2765885151206254e-05,
      "loss": 0.5894,
      "step": 1930
    },
    {
      "epoch": 2.97,
      "learning_rate": 2.2595990485898744e-05,
      "loss": 0.5012,
      "step": 1940
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.2426095820591233e-05,
      "loss": 0.419,
      "step": 1950
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.2256201155283726e-05,
      "loss": 0.5041,
      "step": 1960
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7444126074498567,
      "eval_loss": 0.7038247585296631,
      "eval_runtime": 52.7062,
      "eval_samples_per_second": 99.324,
      "eval_steps_per_second": 12.427,
      "step": 1962
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.2086306489976216e-05,
      "loss": 0.4415,
      "step": 1970
    },
    {
      "epoch": 3.03,
      "learning_rate": 2.1916411824668705e-05,
      "loss": 0.2787,
      "step": 1980
    },
    {
      "epoch": 3.04,
      "learning_rate": 2.1746517159361194e-05,
      "loss": 0.4013,
      "step": 1990
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.1576622494053687e-05,
      "loss": 0.3279,
      "step": 2000
    },
    {
      "epoch": 3.07,
      "learning_rate": 2.140672782874618e-05,
      "loss": 0.3274,
      "step": 2010
    },
    {
      "epoch": 3.09,
      "learning_rate": 2.123683316343867e-05,
      "loss": 0.3664,
      "step": 2020
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.106693849813116e-05,
      "loss": 0.3076,
      "step": 2030
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.0897043832823652e-05,
      "loss": 0.4652,
      "step": 2040
    },
    {
      "epoch": 3.13,
      "learning_rate": 2.072714916751614e-05,
      "loss": 0.358,
      "step": 2050
    },
    {
      "epoch": 3.15,
      "learning_rate": 2.055725450220863e-05,
      "loss": 0.3003,
      "step": 2060
    },
    {
      "epoch": 3.16,
      "learning_rate": 2.038735983690112e-05,
      "loss": 0.3643,
      "step": 2070
    },
    {
      "epoch": 3.18,
      "learning_rate": 2.0217465171593613e-05,
      "loss": 0.3784,
      "step": 2080
    },
    {
      "epoch": 3.19,
      "learning_rate": 2.0047570506286103e-05,
      "loss": 0.3069,
      "step": 2090
    },
    {
      "epoch": 3.21,
      "learning_rate": 1.9877675840978592e-05,
      "loss": 0.3309,
      "step": 2100
    },
    {
      "epoch": 3.23,
      "learning_rate": 1.9707781175671085e-05,
      "loss": 0.3504,
      "step": 2110
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.9537886510363575e-05,
      "loss": 0.3749,
      "step": 2120
    },
    {
      "epoch": 3.26,
      "learning_rate": 1.9367991845056068e-05,
      "loss": 0.3634,
      "step": 2130
    },
    {
      "epoch": 3.27,
      "learning_rate": 1.9198097179748557e-05,
      "loss": 0.3223,
      "step": 2140
    },
    {
      "epoch": 3.29,
      "learning_rate": 1.902820251444105e-05,
      "loss": 0.4309,
      "step": 2150
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.885830784913354e-05,
      "loss": 0.3728,
      "step": 2160
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.868841318382603e-05,
      "loss": 0.2607,
      "step": 2170
    },
    {
      "epoch": 3.33,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.3335,
      "step": 2180
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.834862385321101e-05,
      "loss": 0.2907,
      "step": 2190
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.81787291879035e-05,
      "loss": 0.3103,
      "step": 2200
    },
    {
      "epoch": 3.38,
      "learning_rate": 1.800883452259599e-05,
      "loss": 0.3538,
      "step": 2210
    },
    {
      "epoch": 3.39,
      "learning_rate": 1.783893985728848e-05,
      "loss": 0.4066,
      "step": 2220
    },
    {
      "epoch": 3.41,
      "learning_rate": 1.7669045191980972e-05,
      "loss": 0.3397,
      "step": 2230
    },
    {
      "epoch": 3.42,
      "learning_rate": 1.7499150526673462e-05,
      "loss": 0.3611,
      "step": 2240
    },
    {
      "epoch": 3.44,
      "learning_rate": 1.7329255861365955e-05,
      "loss": 0.3621,
      "step": 2250
    },
    {
      "epoch": 3.45,
      "learning_rate": 1.7159361196058444e-05,
      "loss": 0.2716,
      "step": 2260
    },
    {
      "epoch": 3.47,
      "learning_rate": 1.6989466530750937e-05,
      "loss": 0.3324,
      "step": 2270
    },
    {
      "epoch": 3.48,
      "learning_rate": 1.6819571865443427e-05,
      "loss": 0.3249,
      "step": 2280
    },
    {
      "epoch": 3.5,
      "learning_rate": 1.6649677200135916e-05,
      "loss": 0.3113,
      "step": 2290
    },
    {
      "epoch": 3.52,
      "learning_rate": 1.647978253482841e-05,
      "loss": 0.2582,
      "step": 2300
    },
    {
      "epoch": 3.53,
      "learning_rate": 1.63098878695209e-05,
      "loss": 0.3834,
      "step": 2310
    },
    {
      "epoch": 3.55,
      "learning_rate": 1.6139993204213388e-05,
      "loss": 0.29,
      "step": 2320
    },
    {
      "epoch": 3.56,
      "learning_rate": 1.5970098538905877e-05,
      "loss": 0.3144,
      "step": 2330
    },
    {
      "epoch": 3.58,
      "learning_rate": 1.580020387359837e-05,
      "loss": 0.3653,
      "step": 2340
    },
    {
      "epoch": 3.59,
      "learning_rate": 1.563030920829086e-05,
      "loss": 0.3767,
      "step": 2350
    },
    {
      "epoch": 3.61,
      "learning_rate": 1.546041454298335e-05,
      "loss": 0.3475,
      "step": 2360
    },
    {
      "epoch": 3.62,
      "learning_rate": 1.5290519877675842e-05,
      "loss": 0.4123,
      "step": 2370
    },
    {
      "epoch": 3.64,
      "learning_rate": 1.5120625212368333e-05,
      "loss": 0.3363,
      "step": 2380
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.4950730547060823e-05,
      "loss": 0.25,
      "step": 2390
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.4780835881753314e-05,
      "loss": 0.2966,
      "step": 2400
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.4610941216445803e-05,
      "loss": 0.3758,
      "step": 2410
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.4441046551138296e-05,
      "loss": 0.3117,
      "step": 2420
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.4271151885830786e-05,
      "loss": 0.305,
      "step": 2430
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.4101257220523275e-05,
      "loss": 0.3168,
      "step": 2440
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.3931362555215766e-05,
      "loss": 0.3642,
      "step": 2450
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.3761467889908258e-05,
      "loss": 0.3745,
      "step": 2460
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.3591573224600749e-05,
      "loss": 0.3162,
      "step": 2470
    },
    {
      "epoch": 3.79,
      "learning_rate": 1.3421678559293238e-05,
      "loss": 0.3817,
      "step": 2480
    },
    {
      "epoch": 3.81,
      "learning_rate": 1.3251783893985731e-05,
      "loss": 0.2568,
      "step": 2490
    },
    {
      "epoch": 3.82,
      "learning_rate": 1.308188922867822e-05,
      "loss": 0.3392,
      "step": 2500
    },
    {
      "epoch": 3.84,
      "learning_rate": 1.291199456337071e-05,
      "loss": 0.386,
      "step": 2510
    },
    {
      "epoch": 3.85,
      "learning_rate": 1.2742099898063201e-05,
      "loss": 0.392,
      "step": 2520
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.2572205232755694e-05,
      "loss": 0.3166,
      "step": 2530
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.2402310567448183e-05,
      "loss": 0.3222,
      "step": 2540
    },
    {
      "epoch": 3.9,
      "learning_rate": 1.2232415902140673e-05,
      "loss": 0.3122,
      "step": 2550
    },
    {
      "epoch": 3.91,
      "learning_rate": 1.2062521236833164e-05,
      "loss": 0.3275,
      "step": 2560
    },
    {
      "epoch": 3.93,
      "learning_rate": 1.1892626571525654e-05,
      "loss": 0.3268,
      "step": 2570
    },
    {
      "epoch": 3.94,
      "learning_rate": 1.1722731906218146e-05,
      "loss": 0.2926,
      "step": 2580
    },
    {
      "epoch": 3.96,
      "learning_rate": 1.1552837240910636e-05,
      "loss": 0.3422,
      "step": 2590
    },
    {
      "epoch": 3.97,
      "learning_rate": 1.1382942575603127e-05,
      "loss": 0.3107,
      "step": 2600
    },
    {
      "epoch": 3.99,
      "learning_rate": 1.1213047910295617e-05,
      "loss": 0.2461,
      "step": 2610
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7392550143266475,
      "eval_loss": 0.7843212485313416,
      "eval_runtime": 48.1379,
      "eval_samples_per_second": 108.75,
      "eval_steps_per_second": 13.607,
      "step": 2617
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.1043153244988108e-05,
      "loss": 0.2342,
      "step": 2620
    },
    {
      "epoch": 4.02,
      "learning_rate": 1.0873258579680597e-05,
      "loss": 0.2295,
      "step": 2630
    },
    {
      "epoch": 4.04,
      "learning_rate": 1.070336391437309e-05,
      "loss": 0.1964,
      "step": 2640
    },
    {
      "epoch": 4.05,
      "learning_rate": 1.053346924906558e-05,
      "loss": 0.2504,
      "step": 2650
    },
    {
      "epoch": 4.07,
      "learning_rate": 1.036357458375807e-05,
      "loss": 0.2676,
      "step": 2660
    },
    {
      "epoch": 4.08,
      "learning_rate": 1.019367991845056e-05,
      "loss": 0.2241,
      "step": 2670
    },
    {
      "epoch": 4.1,
      "learning_rate": 1.0023785253143051e-05,
      "loss": 0.189,
      "step": 2680
    },
    {
      "epoch": 4.11,
      "learning_rate": 9.853890587835543e-06,
      "loss": 0.263,
      "step": 2690
    },
    {
      "epoch": 4.13,
      "learning_rate": 9.683995922528034e-06,
      "loss": 0.2561,
      "step": 2700
    },
    {
      "epoch": 4.14,
      "learning_rate": 9.514101257220525e-06,
      "loss": 0.2154,
      "step": 2710
    },
    {
      "epoch": 4.16,
      "learning_rate": 9.344206591913014e-06,
      "loss": 0.2351,
      "step": 2720
    },
    {
      "epoch": 4.17,
      "learning_rate": 9.174311926605506e-06,
      "loss": 0.2281,
      "step": 2730
    },
    {
      "epoch": 4.19,
      "learning_rate": 9.004417261297995e-06,
      "loss": 0.1526,
      "step": 2740
    },
    {
      "epoch": 4.2,
      "learning_rate": 8.834522595990486e-06,
      "loss": 0.2612,
      "step": 2750
    },
    {
      "epoch": 4.22,
      "learning_rate": 8.664627930682977e-06,
      "loss": 0.2339,
      "step": 2760
    },
    {
      "epoch": 4.23,
      "learning_rate": 8.494733265375469e-06,
      "loss": 0.2484,
      "step": 2770
    },
    {
      "epoch": 4.25,
      "learning_rate": 8.324838600067958e-06,
      "loss": 0.2338,
      "step": 2780
    },
    {
      "epoch": 4.26,
      "learning_rate": 8.15494393476045e-06,
      "loss": 0.2062,
      "step": 2790
    },
    {
      "epoch": 4.28,
      "learning_rate": 7.985049269452939e-06,
      "loss": 0.2153,
      "step": 2800
    },
    {
      "epoch": 4.29,
      "learning_rate": 7.81515460414543e-06,
      "loss": 0.1661,
      "step": 2810
    },
    {
      "epoch": 4.31,
      "learning_rate": 7.645259938837921e-06,
      "loss": 0.2059,
      "step": 2820
    },
    {
      "epoch": 4.33,
      "learning_rate": 7.475365273530411e-06,
      "loss": 0.2611,
      "step": 2830
    },
    {
      "epoch": 4.34,
      "learning_rate": 7.305470608222902e-06,
      "loss": 0.2026,
      "step": 2840
    },
    {
      "epoch": 4.36,
      "learning_rate": 7.135575942915393e-06,
      "loss": 0.2194,
      "step": 2850
    },
    {
      "epoch": 4.37,
      "learning_rate": 6.965681277607883e-06,
      "loss": 0.1759,
      "step": 2860
    },
    {
      "epoch": 4.39,
      "learning_rate": 6.795786612300374e-06,
      "loss": 0.2087,
      "step": 2870
    },
    {
      "epoch": 4.4,
      "learning_rate": 6.6258919469928655e-06,
      "loss": 0.2105,
      "step": 2880
    },
    {
      "epoch": 4.42,
      "learning_rate": 6.455997281685355e-06,
      "loss": 0.2052,
      "step": 2890
    },
    {
      "epoch": 4.43,
      "learning_rate": 6.286102616377847e-06,
      "loss": 0.2395,
      "step": 2900
    },
    {
      "epoch": 4.45,
      "learning_rate": 6.1162079510703365e-06,
      "loss": 0.1953,
      "step": 2910
    },
    {
      "epoch": 4.46,
      "learning_rate": 5.946313285762827e-06,
      "loss": 0.1944,
      "step": 2920
    },
    {
      "epoch": 4.48,
      "learning_rate": 5.776418620455318e-06,
      "loss": 0.2784,
      "step": 2930
    },
    {
      "epoch": 4.49,
      "learning_rate": 5.606523955147808e-06,
      "loss": 0.1885,
      "step": 2940
    },
    {
      "epoch": 4.51,
      "learning_rate": 5.436629289840299e-06,
      "loss": 0.2575,
      "step": 2950
    },
    {
      "epoch": 4.52,
      "learning_rate": 5.26673462453279e-06,
      "loss": 0.1941,
      "step": 2960
    },
    {
      "epoch": 4.54,
      "learning_rate": 5.09683995922528e-06,
      "loss": 0.2795,
      "step": 2970
    },
    {
      "epoch": 4.55,
      "learning_rate": 4.926945293917771e-06,
      "loss": 0.1971,
      "step": 2980
    },
    {
      "epoch": 4.57,
      "learning_rate": 4.7570506286102625e-06,
      "loss": 0.224,
      "step": 2990
    },
    {
      "epoch": 4.59,
      "learning_rate": 4.587155963302753e-06,
      "loss": 0.2044,
      "step": 3000
    },
    {
      "epoch": 4.6,
      "learning_rate": 4.417261297995243e-06,
      "loss": 0.2073,
      "step": 3010
    },
    {
      "epoch": 4.62,
      "learning_rate": 4.247366632687734e-06,
      "loss": 0.2055,
      "step": 3020
    },
    {
      "epoch": 4.63,
      "learning_rate": 4.077471967380225e-06,
      "loss": 0.1936,
      "step": 3030
    },
    {
      "epoch": 4.65,
      "learning_rate": 3.907577302072715e-06,
      "loss": 0.219,
      "step": 3040
    },
    {
      "epoch": 4.66,
      "learning_rate": 3.7376826367652057e-06,
      "loss": 0.198,
      "step": 3050
    },
    {
      "epoch": 4.68,
      "learning_rate": 3.5677879714576964e-06,
      "loss": 0.242,
      "step": 3060
    },
    {
      "epoch": 4.69,
      "learning_rate": 3.397893306150187e-06,
      "loss": 0.2242,
      "step": 3070
    },
    {
      "epoch": 4.71,
      "learning_rate": 3.2279986408426775e-06,
      "loss": 0.2083,
      "step": 3080
    },
    {
      "epoch": 4.72,
      "learning_rate": 3.0581039755351682e-06,
      "loss": 0.2268,
      "step": 3090
    },
    {
      "epoch": 4.74,
      "learning_rate": 2.888209310227659e-06,
      "loss": 0.1819,
      "step": 3100
    },
    {
      "epoch": 4.75,
      "learning_rate": 2.7183146449201493e-06,
      "loss": 0.2127,
      "step": 3110
    },
    {
      "epoch": 4.77,
      "learning_rate": 2.54841997961264e-06,
      "loss": 0.1816,
      "step": 3120
    },
    {
      "epoch": 4.78,
      "learning_rate": 2.3785253143051312e-06,
      "loss": 0.187,
      "step": 3130
    },
    {
      "epoch": 4.8,
      "learning_rate": 2.2086306489976216e-06,
      "loss": 0.2482,
      "step": 3140
    },
    {
      "epoch": 4.81,
      "learning_rate": 2.0387359836901123e-06,
      "loss": 0.2161,
      "step": 3150
    },
    {
      "epoch": 4.83,
      "learning_rate": 1.8688413183826028e-06,
      "loss": 0.2413,
      "step": 3160
    },
    {
      "epoch": 4.85,
      "learning_rate": 1.6989466530750936e-06,
      "loss": 0.1959,
      "step": 3170
    },
    {
      "epoch": 4.86,
      "learning_rate": 1.5290519877675841e-06,
      "loss": 0.2644,
      "step": 3180
    },
    {
      "epoch": 4.88,
      "learning_rate": 1.3591573224600747e-06,
      "loss": 0.1793,
      "step": 3190
    },
    {
      "epoch": 4.89,
      "learning_rate": 1.1892626571525656e-06,
      "loss": 0.1869,
      "step": 3200
    },
    {
      "epoch": 4.91,
      "learning_rate": 1.0193679918450562e-06,
      "loss": 0.195,
      "step": 3210
    },
    {
      "epoch": 4.92,
      "learning_rate": 8.494733265375468e-07,
      "loss": 0.2474,
      "step": 3220
    },
    {
      "epoch": 4.94,
      "learning_rate": 6.795786612300373e-07,
      "loss": 0.2748,
      "step": 3230
    },
    {
      "epoch": 4.95,
      "learning_rate": 5.096839959225281e-07,
      "loss": 0.1932,
      "step": 3240
    },
    {
      "epoch": 4.97,
      "learning_rate": 3.3978933061501866e-07,
      "loss": 0.2393,
      "step": 3250
    },
    {
      "epoch": 4.98,
      "learning_rate": 1.6989466530750933e-07,
      "loss": 0.2547,
      "step": 3260
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 0.1846,
      "step": 3270
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7390639923591213,
      "eval_loss": 0.8219425082206726,
      "eval_runtime": 48.049,
      "eval_samples_per_second": 108.951,
      "eval_steps_per_second": 13.632,
      "step": 3270
    },
    {
      "epoch": 5.0,
      "step": 3270,
      "total_flos": 8.109125174606561e+18,
      "train_loss": 0.5082513862064489,
      "train_runtime": 2793.8499,
      "train_samples_per_second": 37.468,
      "train_steps_per_second": 1.17
    }
  ],
  "logging_steps": 10,
  "max_steps": 3270,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 8.109125174606561e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}