{
  "best_metric": 0.3053509294986725,
  "best_model_checkpoint": "/content/drive/MyDrive/Sumdu/PhD/LGViT/DeiT-base/cifar100/base/checkpoint-7820",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 7820,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 4.9680306905370846e-05,
      "loss": 4.3042,
      "step": 50
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.936061381074169e-05,
      "loss": 3.2558,
      "step": 100
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.9040920716112534e-05,
      "loss": 2.3793,
      "step": 150
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.872122762148338e-05,
      "loss": 1.898,
      "step": 200
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.840153452685422e-05,
      "loss": 1.6523,
      "step": 250
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.8081841432225067e-05,
      "loss": 1.5372,
      "step": 300
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.776214833759591e-05,
      "loss": 1.3641,
      "step": 350
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.7442455242966755e-05,
      "loss": 1.3355,
      "step": 400
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.71227621483376e-05,
      "loss": 1.2267,
      "step": 450
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.680306905370844e-05,
      "loss": 1.1976,
      "step": 500
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.6483375959079287e-05,
      "loss": 1.1752,
      "step": 550
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.616368286445013e-05,
      "loss": 1.178,
      "step": 600
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.5843989769820975e-05,
      "loss": 1.1759,
      "step": 650
    },
    {
      "epoch": 0.9,
      "learning_rate": 4.552429667519182e-05,
      "loss": 1.1176,
      "step": 700
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.520460358056266e-05,
      "loss": 1.1232,
      "step": 750
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8416,
      "eval_loss": 0.539007306098938,
      "eval_runtime": 176.056,
      "eval_samples_per_second": 56.8,
      "eval_steps_per_second": 56.8,
      "step": 782
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.488491048593351e-05,
      "loss": 1.0464,
      "step": 800
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.456521739130435e-05,
      "loss": 0.9967,
      "step": 850
    },
    {
      "epoch": 1.15,
      "learning_rate": 4.4245524296675195e-05,
      "loss": 0.9914,
      "step": 900
    },
    {
      "epoch": 1.21,
      "learning_rate": 4.392583120204604e-05,
      "loss": 0.944,
      "step": 950
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.360613810741688e-05,
      "loss": 0.9537,
      "step": 1000
    },
    {
      "epoch": 1.34,
      "learning_rate": 4.328644501278773e-05,
      "loss": 0.9067,
      "step": 1050
    },
    {
      "epoch": 1.41,
      "learning_rate": 4.296675191815857e-05,
      "loss": 0.9325,
      "step": 1100
    },
    {
      "epoch": 1.47,
      "learning_rate": 4.2647058823529415e-05,
      "loss": 0.9125,
      "step": 1150
    },
    {
      "epoch": 1.53,
      "learning_rate": 4.232736572890026e-05,
      "loss": 0.9322,
      "step": 1200
    },
    {
      "epoch": 1.6,
      "learning_rate": 4.20076726342711e-05,
      "loss": 0.9293,
      "step": 1250
    },
    {
      "epoch": 1.66,
      "learning_rate": 4.168797953964195e-05,
      "loss": 0.8524,
      "step": 1300
    },
    {
      "epoch": 1.73,
      "learning_rate": 4.136828644501279e-05,
      "loss": 0.942,
      "step": 1350
    },
    {
      "epoch": 1.79,
      "learning_rate": 4.1048593350383635e-05,
      "loss": 0.9373,
      "step": 1400
    },
    {
      "epoch": 1.85,
      "learning_rate": 4.072890025575448e-05,
      "loss": 0.8504,
      "step": 1450
    },
    {
      "epoch": 1.92,
      "learning_rate": 4.040920716112532e-05,
      "loss": 0.9145,
      "step": 1500
    },
    {
      "epoch": 1.98,
      "learning_rate": 4.008951406649617e-05,
      "loss": 0.9017,
      "step": 1550
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8699,
      "eval_loss": 0.4364722669124603,
      "eval_runtime": 178.2596,
      "eval_samples_per_second": 56.098,
      "eval_steps_per_second": 56.098,
      "step": 1564
    },
    {
      "epoch": 2.05,
      "learning_rate": 3.976982097186701e-05,
      "loss": 0.8527,
      "step": 1600
    },
    {
      "epoch": 2.11,
      "learning_rate": 3.9450127877237855e-05,
      "loss": 0.7803,
      "step": 1650
    },
    {
      "epoch": 2.17,
      "learning_rate": 3.91304347826087e-05,
      "loss": 0.799,
      "step": 1700
    },
    {
      "epoch": 2.24,
      "learning_rate": 3.881074168797954e-05,
      "loss": 0.7873,
      "step": 1750
    },
    {
      "epoch": 2.3,
      "learning_rate": 3.849104859335039e-05,
      "loss": 0.7999,
      "step": 1800
    },
    {
      "epoch": 2.37,
      "learning_rate": 3.817135549872123e-05,
      "loss": 0.7911,
      "step": 1850
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.7851662404092075e-05,
      "loss": 0.7987,
      "step": 1900
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.753196930946292e-05,
      "loss": 0.7923,
      "step": 1950
    },
    {
      "epoch": 2.56,
      "learning_rate": 3.721227621483376e-05,
      "loss": 0.8116,
      "step": 2000
    },
    {
      "epoch": 2.62,
      "learning_rate": 3.689258312020461e-05,
      "loss": 0.7709,
      "step": 2050
    },
    {
      "epoch": 2.69,
      "learning_rate": 3.657289002557545e-05,
      "loss": 0.764,
      "step": 2100
    },
    {
      "epoch": 2.75,
      "learning_rate": 3.6253196930946295e-05,
      "loss": 0.8047,
      "step": 2150
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.593350383631714e-05,
      "loss": 0.7357,
      "step": 2200
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.561381074168798e-05,
      "loss": 0.7779,
      "step": 2250
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.529411764705883e-05,
      "loss": 0.7565,
      "step": 2300
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8858,
      "eval_loss": 0.3678203225135803,
      "eval_runtime": 176.8651,
      "eval_samples_per_second": 56.54,
      "eval_steps_per_second": 56.54,
      "step": 2346
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.497442455242967e-05,
      "loss": 0.7418,
      "step": 2350
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.4654731457800515e-05,
      "loss": 0.709,
      "step": 2400
    },
    {
      "epoch": 3.13,
      "learning_rate": 3.433503836317136e-05,
      "loss": 0.6461,
      "step": 2450
    },
    {
      "epoch": 3.2,
      "learning_rate": 3.40153452685422e-05,
      "loss": 0.6895,
      "step": 2500
    },
    {
      "epoch": 3.26,
      "learning_rate": 3.369565217391305e-05,
      "loss": 0.6676,
      "step": 2550
    },
    {
      "epoch": 3.32,
      "learning_rate": 3.337595907928389e-05,
      "loss": 0.6769,
      "step": 2600
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.3056265984654735e-05,
      "loss": 0.7018,
      "step": 2650
    },
    {
      "epoch": 3.45,
      "learning_rate": 3.273657289002558e-05,
      "loss": 0.6686,
      "step": 2700
    },
    {
      "epoch": 3.52,
      "learning_rate": 3.241687979539642e-05,
      "loss": 0.7211,
      "step": 2750
    },
    {
      "epoch": 3.58,
      "learning_rate": 3.209718670076726e-05,
      "loss": 0.7097,
      "step": 2800
    },
    {
      "epoch": 3.64,
      "learning_rate": 3.177749360613811e-05,
      "loss": 0.6939,
      "step": 2850
    },
    {
      "epoch": 3.71,
      "learning_rate": 3.145780051150895e-05,
      "loss": 0.6593,
      "step": 2900
    },
    {
      "epoch": 3.77,
      "learning_rate": 3.11381074168798e-05,
      "loss": 0.6931,
      "step": 2950
    },
    {
      "epoch": 3.84,
      "learning_rate": 3.081841432225064e-05,
      "loss": 0.711,
      "step": 3000
    },
    {
      "epoch": 3.9,
      "learning_rate": 3.0498721227621484e-05,
      "loss": 0.6439,
      "step": 3050
    },
    {
      "epoch": 3.96,
      "learning_rate": 3.0179028132992325e-05,
      "loss": 0.706,
      "step": 3100
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8952,
      "eval_loss": 0.3446212708950043,
      "eval_runtime": 176.3518,
      "eval_samples_per_second": 56.705,
      "eval_steps_per_second": 56.705,
      "step": 3128
    },
    {
      "epoch": 4.03,
      "learning_rate": 2.9859335038363172e-05,
      "loss": 0.6984,
      "step": 3150
    },
    {
      "epoch": 4.09,
      "learning_rate": 2.9539641943734013e-05,
      "loss": 0.6062,
      "step": 3200
    },
    {
      "epoch": 4.16,
      "learning_rate": 2.921994884910486e-05,
      "loss": 0.6542,
      "step": 3250
    },
    {
      "epoch": 4.22,
      "learning_rate": 2.89002557544757e-05,
      "loss": 0.6092,
      "step": 3300
    },
    {
      "epoch": 4.28,
      "learning_rate": 2.8580562659846548e-05,
      "loss": 0.6325,
      "step": 3350
    },
    {
      "epoch": 4.35,
      "learning_rate": 2.826086956521739e-05,
      "loss": 0.6296,
      "step": 3400
    },
    {
      "epoch": 4.41,
      "learning_rate": 2.7941176470588236e-05,
      "loss": 0.6493,
      "step": 3450
    },
    {
      "epoch": 4.48,
      "learning_rate": 2.7621483375959077e-05,
      "loss": 0.609,
      "step": 3500
    },
    {
      "epoch": 4.54,
      "learning_rate": 2.7301790281329924e-05,
      "loss": 0.6173,
      "step": 3550
    },
    {
      "epoch": 4.6,
      "learning_rate": 2.6982097186700765e-05,
      "loss": 0.6263,
      "step": 3600
    },
    {
      "epoch": 4.67,
      "learning_rate": 2.6662404092071612e-05,
      "loss": 0.6264,
      "step": 3650
    },
    {
      "epoch": 4.73,
      "learning_rate": 2.6342710997442456e-05,
      "loss": 0.601,
      "step": 3700
    },
    {
      "epoch": 4.8,
      "learning_rate": 2.60230179028133e-05,
      "loss": 0.5905,
      "step": 3750
    },
    {
      "epoch": 4.86,
      "learning_rate": 2.5703324808184144e-05,
      "loss": 0.609,
      "step": 3800
    },
    {
      "epoch": 4.92,
      "learning_rate": 2.538363171355499e-05,
      "loss": 0.5939,
      "step": 3850
    },
    {
      "epoch": 4.99,
      "learning_rate": 2.5063938618925832e-05,
      "loss": 0.6353,
      "step": 3900
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8986,
      "eval_loss": 0.333119660615921,
      "eval_runtime": 177.2116,
      "eval_samples_per_second": 56.43,
      "eval_steps_per_second": 56.43,
      "step": 3910
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.4744245524296676e-05,
      "loss": 0.6056,
      "step": 3950
    },
    {
      "epoch": 5.12,
      "learning_rate": 2.442455242966752e-05,
      "loss": 0.5218,
      "step": 4000
    },
    {
      "epoch": 5.18,
      "learning_rate": 2.4104859335038364e-05,
      "loss": 0.5656,
      "step": 4050
    },
    {
      "epoch": 5.24,
      "learning_rate": 2.378516624040921e-05,
      "loss": 0.5728,
      "step": 4100
    },
    {
      "epoch": 5.31,
      "learning_rate": 2.3465473145780052e-05,
      "loss": 0.5803,
      "step": 4150
    },
    {
      "epoch": 5.37,
      "learning_rate": 2.3145780051150897e-05,
      "loss": 0.5907,
      "step": 4200
    },
    {
      "epoch": 5.43,
      "learning_rate": 2.282608695652174e-05,
      "loss": 0.567,
      "step": 4250
    },
    {
      "epoch": 5.5,
      "learning_rate": 2.2506393861892585e-05,
      "loss": 0.5585,
      "step": 4300
    },
    {
      "epoch": 5.56,
      "learning_rate": 2.218670076726343e-05,
      "loss": 0.5634,
      "step": 4350
    },
    {
      "epoch": 5.63,
      "learning_rate": 2.1867007672634273e-05,
      "loss": 0.5398,
      "step": 4400
    },
    {
      "epoch": 5.69,
      "learning_rate": 2.1547314578005117e-05,
      "loss": 0.5353,
      "step": 4450
    },
    {
      "epoch": 5.75,
      "learning_rate": 2.122762148337596e-05,
      "loss": 0.5341,
      "step": 4500
    },
    {
      "epoch": 5.82,
      "learning_rate": 2.0907928388746805e-05,
      "loss": 0.5758,
      "step": 4550
    },
    {
      "epoch": 5.88,
      "learning_rate": 2.058823529411765e-05,
      "loss": 0.5647,
      "step": 4600
    },
    {
      "epoch": 5.95,
      "learning_rate": 2.0268542199488493e-05,
      "loss": 0.5384,
      "step": 4650
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9001,
      "eval_loss": 0.3222615420818329,
      "eval_runtime": 176.5584,
      "eval_samples_per_second": 56.638,
      "eval_steps_per_second": 56.638,
      "step": 4692
    },
    {
      "epoch": 6.01,
      "learning_rate": 1.9948849104859337e-05,
      "loss": 0.5534,
      "step": 4700
    },
    {
      "epoch": 6.07,
      "learning_rate": 1.962915601023018e-05,
      "loss": 0.5361,
      "step": 4750
    },
    {
      "epoch": 6.14,
      "learning_rate": 1.9309462915601025e-05,
      "loss": 0.5419,
      "step": 4800
    },
    {
      "epoch": 6.2,
      "learning_rate": 1.898976982097187e-05,
      "loss": 0.5235,
      "step": 4850
    },
    {
      "epoch": 6.27,
      "learning_rate": 1.8670076726342713e-05,
      "loss": 0.5153,
      "step": 4900
    },
    {
      "epoch": 6.33,
      "learning_rate": 1.8350383631713557e-05,
      "loss": 0.5362,
      "step": 4950
    },
    {
      "epoch": 6.39,
      "learning_rate": 1.80306905370844e-05,
      "loss": 0.4884,
      "step": 5000
    },
    {
      "epoch": 6.46,
      "learning_rate": 1.7710997442455245e-05,
      "loss": 0.532,
      "step": 5050
    },
    {
      "epoch": 6.52,
      "learning_rate": 1.739130434782609e-05,
      "loss": 0.4974,
      "step": 5100
    },
    {
      "epoch": 6.59,
      "learning_rate": 1.7071611253196933e-05,
      "loss": 0.5375,
      "step": 5150
    },
    {
      "epoch": 6.65,
      "learning_rate": 1.6751918158567777e-05,
      "loss": 0.5208,
      "step": 5200
    },
    {
      "epoch": 6.71,
      "learning_rate": 1.643222506393862e-05,
      "loss": 0.5572,
      "step": 5250
    },
    {
      "epoch": 6.78,
      "learning_rate": 1.6112531969309465e-05,
      "loss": 0.5224,
      "step": 5300
    },
    {
      "epoch": 6.84,
      "learning_rate": 1.579283887468031e-05,
      "loss": 0.5319,
      "step": 5350
    },
    {
      "epoch": 6.91,
      "learning_rate": 1.5473145780051153e-05,
      "loss": 0.5071,
      "step": 5400
    },
    {
      "epoch": 6.97,
      "learning_rate": 1.5153452685421995e-05,
      "loss": 0.5004,
      "step": 5450
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9018,
      "eval_loss": 0.32489562034606934,
      "eval_runtime": 170.5926,
      "eval_samples_per_second": 58.619,
      "eval_steps_per_second": 58.619,
      "step": 5474
    },
    {
      "epoch": 7.03,
      "learning_rate": 1.483375959079284e-05,
      "loss": 0.5231,
      "step": 5500
    },
    {
      "epoch": 7.1,
      "learning_rate": 1.4514066496163683e-05,
      "loss": 0.5106,
      "step": 5550
    },
    {
      "epoch": 7.16,
      "learning_rate": 1.4194373401534527e-05,
      "loss": 0.4962,
      "step": 5600
    },
    {
      "epoch": 7.23,
      "learning_rate": 1.3874680306905371e-05,
      "loss": 0.4761,
      "step": 5650
    },
    {
      "epoch": 7.29,
      "learning_rate": 1.3554987212276215e-05,
      "loss": 0.5139,
      "step": 5700
    },
    {
      "epoch": 7.35,
      "learning_rate": 1.323529411764706e-05,
      "loss": 0.4848,
      "step": 5750
    },
    {
      "epoch": 7.42,
      "learning_rate": 1.2915601023017903e-05,
      "loss": 0.5141,
      "step": 5800
    },
    {
      "epoch": 7.48,
      "learning_rate": 1.2595907928388747e-05,
      "loss": 0.4613,
      "step": 5850
    },
    {
      "epoch": 7.54,
      "learning_rate": 1.2276214833759591e-05,
      "loss": 0.4606,
      "step": 5900
    },
    {
      "epoch": 7.61,
      "learning_rate": 1.1956521739130435e-05,
      "loss": 0.4778,
      "step": 5950
    },
    {
      "epoch": 7.67,
      "learning_rate": 1.163682864450128e-05,
      "loss": 0.497,
      "step": 6000
    },
    {
      "epoch": 7.74,
      "learning_rate": 1.1317135549872123e-05,
      "loss": 0.5005,
      "step": 6050
    },
    {
      "epoch": 7.8,
      "learning_rate": 1.0997442455242967e-05,
      "loss": 0.5184,
      "step": 6100
    },
    {
      "epoch": 7.86,
      "learning_rate": 1.0677749360613812e-05,
      "loss": 0.5045,
      "step": 6150
    },
    {
      "epoch": 7.93,
      "learning_rate": 1.0358056265984656e-05,
      "loss": 0.4755,
      "step": 6200
    },
    {
      "epoch": 7.99,
      "learning_rate": 1.00383631713555e-05,
      "loss": 0.4672,
      "step": 6250
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.904,
      "eval_loss": 0.31126102805137634,
      "eval_runtime": 170.2673,
      "eval_samples_per_second": 58.731,
      "eval_steps_per_second": 58.731,
      "step": 6256
    },
    {
      "epoch": 8.06,
      "learning_rate": 9.718670076726342e-06,
      "loss": 0.4827,
      "step": 6300
    },
    {
      "epoch": 8.12,
      "learning_rate": 9.398976982097186e-06,
      "loss": 0.4801,
      "step": 6350
    },
    {
      "epoch": 8.18,
      "learning_rate": 9.07928388746803e-06,
      "loss": 0.4566,
      "step": 6400
    },
    {
      "epoch": 8.25,
      "learning_rate": 8.759590792838874e-06,
      "loss": 0.4837,
      "step": 6450
    },
    {
      "epoch": 8.31,
      "learning_rate": 8.439897698209718e-06,
      "loss": 0.4406,
      "step": 6500
    },
    {
      "epoch": 8.38,
      "learning_rate": 8.120204603580562e-06,
      "loss": 0.4885,
      "step": 6550
    },
    {
      "epoch": 8.44,
      "learning_rate": 7.800511508951406e-06,
      "loss": 0.4286,
      "step": 6600
    },
    {
      "epoch": 8.5,
      "learning_rate": 7.480818414322251e-06,
      "loss": 0.4652,
      "step": 6650
    },
    {
      "epoch": 8.57,
      "learning_rate": 7.161125319693095e-06,
      "loss": 0.4914,
      "step": 6700
    },
    {
      "epoch": 8.63,
      "learning_rate": 6.841432225063939e-06,
      "loss": 0.4436,
      "step": 6750
    },
    {
      "epoch": 8.7,
      "learning_rate": 6.521739130434783e-06,
      "loss": 0.4366,
      "step": 6800
    },
    {
      "epoch": 8.76,
      "learning_rate": 6.202046035805627e-06,
      "loss": 0.4481,
      "step": 6850
    },
    {
      "epoch": 8.82,
      "learning_rate": 5.882352941176471e-06,
      "loss": 0.4397,
      "step": 6900
    },
    {
      "epoch": 8.89,
      "learning_rate": 5.562659846547315e-06,
      "loss": 0.4823,
      "step": 6950
    },
    {
      "epoch": 8.95,
      "learning_rate": 5.242966751918159e-06,
      "loss": 0.4526,
      "step": 7000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9054,
      "eval_loss": 0.3081473708152771,
      "eval_runtime": 171.7477,
      "eval_samples_per_second": 58.225,
      "eval_steps_per_second": 58.225,
      "step": 7038
    },
    {
      "epoch": 9.02,
      "learning_rate": 4.923273657289003e-06,
      "loss": 0.4519,
      "step": 7050
    },
    {
      "epoch": 9.08,
      "learning_rate": 4.603580562659847e-06,
      "loss": 0.4074,
      "step": 7100
    },
    {
      "epoch": 9.14,
      "learning_rate": 4.283887468030691e-06,
      "loss": 0.4026,
      "step": 7150
    },
    {
      "epoch": 9.21,
      "learning_rate": 3.964194373401535e-06,
      "loss": 0.4629,
      "step": 7200
    },
    {
      "epoch": 9.27,
      "learning_rate": 3.644501278772379e-06,
      "loss": 0.4267,
      "step": 7250
    },
    {
      "epoch": 9.34,
      "learning_rate": 3.324808184143223e-06,
      "loss": 0.4086,
      "step": 7300
    },
    {
      "epoch": 9.4,
      "learning_rate": 3.0051150895140667e-06,
      "loss": 0.4037,
      "step": 7350
    },
    {
      "epoch": 9.46,
      "learning_rate": 2.6854219948849107e-06,
      "loss": 0.4321,
      "step": 7400
    },
    {
      "epoch": 9.53,
      "learning_rate": 2.3657289002557547e-06,
      "loss": 0.4712,
      "step": 7450
    },
    {
      "epoch": 9.59,
      "learning_rate": 2.0460358056265987e-06,
      "loss": 0.4478,
      "step": 7500
    },
    {
      "epoch": 9.65,
      "learning_rate": 1.7263427109974423e-06,
      "loss": 0.4272,
      "step": 7550
    },
    {
      "epoch": 9.72,
      "learning_rate": 1.4066496163682865e-06,
      "loss": 0.4293,
      "step": 7600
    },
    {
      "epoch": 9.78,
      "learning_rate": 1.0869565217391306e-06,
      "loss": 0.4139,
      "step": 7650
    },
    {
      "epoch": 9.85,
      "learning_rate": 7.672634271099745e-07,
      "loss": 0.441,
      "step": 7700
    },
    {
      "epoch": 9.91,
      "learning_rate": 4.4757033248081843e-07,
      "loss": 0.4479,
      "step": 7750
    },
    {
      "epoch": 9.97,
      "learning_rate": 1.2787723785166242e-07,
      "loss": 0.4289,
      "step": 7800
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.906,
      "eval_loss": 0.3053509294986725,
      "eval_runtime": 174.6627,
      "eval_samples_per_second": 57.253,
      "eval_steps_per_second": 57.253,
      "step": 7820
    },
    {
      "epoch": 10.0,
      "step": 7820,
      "total_flos": 3.8815447523328e+19,
      "train_loss": 0.0,
      "train_runtime": 36.1375,
      "train_samples_per_second": 13836.038,
      "train_steps_per_second": 216.396
    }
  ],
  "logging_steps": 50,
  "max_steps": 7820,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 3.8815447523328e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}