End of training

README.md

@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7391304347826086
+      value: 0.8043478260869565
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.1208
-- Accuracy: 0.7391
+- Loss: 0.9190
+- Accuracy: 0.8043
 
 ## Model description
 

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 73.85,
-    "eval_accuracy": 0.782608695652174,
-    "eval_loss": 0.8900798559188843,
-    "eval_runtime": 0.8755,
-    "eval_samples_per_second": 52.542,
-    "eval_steps_per_second": 2.284,
-    "train_loss": 0.34541464410722256,
-    "train_runtime": 589.4401,
-    "train_samples_per_second": 111.156,
-    "train_steps_per_second": 0.814
+    "eval_accuracy": 0.8043478260869565,
+    "eval_loss": 0.9190374612808228,
+    "eval_runtime": 0.8942,
+    "eval_samples_per_second": 51.44,
+    "eval_steps_per_second": 2.237,
+    "train_loss": 0.3456469999005397,
+    "train_runtime": 590.3301,
+    "train_samples_per_second": 110.989,
+    "train_steps_per_second": 0.813
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 73.85,
-    "eval_accuracy": 0.782608695652174,
-    "eval_loss": 0.8900798559188843,
-    "eval_runtime": 0.8755,
-    "eval_samples_per_second": 52.542,
-    "eval_steps_per_second": 2.284
+    "eval_accuracy": 0.8043478260869565,
+    "eval_loss": 0.9190374612808228,
+    "eval_runtime": 0.8942,
+    "eval_samples_per_second": 51.44,
+    "eval_steps_per_second": 2.237
 }

runs/Jun12_09-58-19_DESKTOP-SKBE9FB/events.out.tfevents.1718208549.DESKTOP-SKBE9FB.4236.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9ab06ff7888cb46d40432bccd062b453d3f1b4a6b7ed44cbdf4d81416d80ca9
+size 411

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 73.85,
-    "train_loss": 0.34541464410722256,
-    "train_runtime": 589.4401,
-    "train_samples_per_second": 111.156,
-    "train_steps_per_second": 0.814
+    "train_loss": 0.3456469999005397,
+    "train_runtime": 590.3301,
+    "train_samples_per_second": 110.989,
+    "train_steps_per_second": 0.813
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.782608695652174,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-ve-U13-b-80\\checkpoint-240",
+  "best_metric": 0.8043478260869565,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-ve-U13-b-80\\checkpoint-286",
   "epoch": 73.84615384615384,
   "eval_steps": 500,
   "global_step": 480,
@@ -12,9 +12,9 @@
       "epoch": 0.92,
       "eval_accuracy": 0.13043478260869565,
       "eval_loss": 1.385940432548523,
-      "eval_runtime": 0.6339,
-      "eval_samples_per_second": 72.571,
-      "eval_steps_per_second": 3.155,
+      "eval_runtime": 0.6821,
+      "eval_samples_per_second": 67.44,
+      "eval_steps_per_second": 2.932,
       "step": 6
     },
     {
@@ -27,18 +27,18 @@
       "epoch": 2.0,
       "eval_accuracy": 0.2826086956521739,
       "eval_loss": 1.382836103439331,
-      "eval_runtime": 0.653,
-      "eval_samples_per_second": 70.443,
-      "eval_steps_per_second": 3.063,
+      "eval_runtime": 0.6681,
+      "eval_samples_per_second": 68.854,
+      "eval_steps_per_second": 2.994,
       "step": 13
     },
     {
       "epoch": 2.92,
       "eval_accuracy": 0.32608695652173914,
       "eval_loss": 1.3768701553344727,
-      "eval_runtime": 0.65,
-      "eval_samples_per_second": 70.768,
-      "eval_steps_per_second": 3.077,
+      "eval_runtime": 0.6605,
+      "eval_samples_per_second": 69.644,
+      "eval_steps_per_second": 3.028,
       "step": 19
     },
     {
@@ -51,9 +51,9 @@
       "epoch": 4.0,
       "eval_accuracy": 0.2826086956521739,
       "eval_loss": 1.3565824031829834,
-      "eval_runtime": 0.6532,
-      "eval_samples_per_second": 70.417,
-      "eval_steps_per_second": 3.062,
+      "eval_runtime": 0.6821,
+      "eval_samples_per_second": 67.437,
+      "eval_steps_per_second": 2.932,
       "step": 26
     },
     {
@@ -65,19 +65,19 @@
     {
       "epoch": 4.92,
       "eval_accuracy": 0.2391304347826087,
-      "eval_loss": 1.316235065460205,
-      "eval_runtime": 0.6545,
-      "eval_samples_per_second": 70.281,
-      "eval_steps_per_second": 3.056,
+      "eval_loss": 1.3162237405776978,
+      "eval_runtime": 0.6667,
+      "eval_samples_per_second": 68.998,
+      "eval_steps_per_second": 3.0,
       "step": 32
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.3695652173913043,
-      "eval_loss": 1.2090449333190918,
-      "eval_runtime": 0.6376,
-      "eval_samples_per_second": 72.151,
-      "eval_steps_per_second": 3.137,
+      "eval_accuracy": 0.34782608695652173,
+      "eval_loss": 1.2093461751937866,
+      "eval_runtime": 0.6736,
+      "eval_samples_per_second": 68.286,
+      "eval_steps_per_second": 2.969,
       "step": 39
     },
     {
@@ -88,888 +88,888 @@
     },
     {
       "epoch": 6.92,
-      "eval_accuracy": 0.41304347826086957,
-      "eval_loss": 1.140857458114624,
-      "eval_runtime": 0.6476,
-      "eval_samples_per_second": 71.035,
-      "eval_steps_per_second": 3.088,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.1349235773086548,
+      "eval_runtime": 0.6744,
+      "eval_samples_per_second": 68.21,
+      "eval_steps_per_second": 2.966,
       "step": 45
     },
     {
       "epoch": 7.69,
       "learning_rate": 4.976851851851852e-05,
-      "loss": 1.0289,
+      "loss": 1.0274,
       "step": 50
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.0442323684692383,
-      "eval_runtime": 0.6862,
-      "eval_samples_per_second": 67.038,
-      "eval_steps_per_second": 2.915,
+      "eval_accuracy": 0.4782608695652174,
+      "eval_loss": 1.041388750076294,
+      "eval_runtime": 0.7013,
+      "eval_samples_per_second": 65.59,
+      "eval_steps_per_second": 2.852,
       "step": 52
     },
     {
       "epoch": 8.92,
       "eval_accuracy": 0.5217391304347826,
-      "eval_loss": 0.9695850610733032,
-      "eval_runtime": 0.6564,
-      "eval_samples_per_second": 70.079,
-      "eval_steps_per_second": 3.047,
+      "eval_loss": 0.9787597060203552,
+      "eval_runtime": 0.6922,
+      "eval_samples_per_second": 66.45,
+      "eval_steps_per_second": 2.889,
       "step": 58
     },
     {
       "epoch": 9.23,
       "learning_rate": 4.8611111111111115e-05,
-      "loss": 0.9132,
+      "loss": 0.9125,
       "step": 60
     },
     {
       "epoch": 10.0,
       "eval_accuracy": 0.43478260869565216,
-      "eval_loss": 1.0133169889450073,
-      "eval_runtime": 0.6439,
-      "eval_samples_per_second": 71.444,
-      "eval_steps_per_second": 3.106,
+      "eval_loss": 1.007051944732666,
+      "eval_runtime": 0.6755,
+      "eval_samples_per_second": 68.102,
+      "eval_steps_per_second": 2.961,
       "step": 65
     },
     {
       "epoch": 10.77,
       "learning_rate": 4.745370370370371e-05,
-      "loss": 0.7677,
+      "loss": 0.7688,
       "step": 70
     },
     {
       "epoch": 10.92,
-      "eval_accuracy": 0.5,
-      "eval_loss": 1.0143684148788452,
-      "eval_runtime": 0.6229,
-      "eval_samples_per_second": 73.85,
-      "eval_steps_per_second": 3.211,
+      "eval_accuracy": 0.5217391304347826,
+      "eval_loss": 1.0415502786636353,
+      "eval_runtime": 0.6785,
+      "eval_samples_per_second": 67.801,
+      "eval_steps_per_second": 2.948,
       "step": 71
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.34782608695652173,
-      "eval_loss": 1.137718915939331,
-      "eval_runtime": 0.6502,
-      "eval_samples_per_second": 70.745,
-      "eval_steps_per_second": 3.076,
+      "eval_accuracy": 0.41304347826086957,
+      "eval_loss": 1.0480490922927856,
+      "eval_runtime": 0.6996,
+      "eval_samples_per_second": 65.748,
+      "eval_steps_per_second": 2.859,
       "step": 78
     },
     {
       "epoch": 12.31,
       "learning_rate": 4.62962962962963e-05,
-      "loss": 0.6988,
+      "loss": 0.6891,
       "step": 80
     },
     {
       "epoch": 12.92,
       "eval_accuracy": 0.5869565217391305,
-      "eval_loss": 0.817061722278595,
-      "eval_runtime": 0.6531,
-      "eval_samples_per_second": 70.438,
-      "eval_steps_per_second": 3.063,
+      "eval_loss": 0.9350693821907043,
+      "eval_runtime": 0.8007,
+      "eval_samples_per_second": 57.451,
+      "eval_steps_per_second": 2.498,
       "step": 84
     },
     {
       "epoch": 13.85,
       "learning_rate": 4.5138888888888894e-05,
-      "loss": 0.6051,
+      "loss": 0.5795,
       "step": 90
     },
     {
       "epoch": 14.0,
-      "eval_accuracy": 0.6521739130434783,
-      "eval_loss": 0.8982558846473694,
-      "eval_runtime": 0.6551,
-      "eval_samples_per_second": 70.213,
-      "eval_steps_per_second": 3.053,
+      "eval_accuracy": 0.6304347826086957,
+      "eval_loss": 1.0682731866836548,
+      "eval_runtime": 0.7937,
+      "eval_samples_per_second": 57.956,
+      "eval_steps_per_second": 2.52,
       "step": 91
     },
     {
       "epoch": 14.92,
       "eval_accuracy": 0.6086956521739131,
-      "eval_loss": 0.8593006134033203,
-      "eval_runtime": 0.6479,
-      "eval_samples_per_second": 71.003,
-      "eval_steps_per_second": 3.087,
+      "eval_loss": 1.06975519657135,
+      "eval_runtime": 0.6958,
+      "eval_samples_per_second": 66.114,
+      "eval_steps_per_second": 2.875,
       "step": 97
     },
     {
       "epoch": 15.38,
       "learning_rate": 4.3981481481481486e-05,
-      "loss": 0.5175,
+      "loss": 0.5337,
       "step": 100
     },
     {
       "epoch": 16.0,
-      "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 0.818878173828125,
-      "eval_runtime": 0.6394,
-      "eval_samples_per_second": 71.94,
-      "eval_steps_per_second": 3.128,
+      "eval_accuracy": 0.6304347826086957,
+      "eval_loss": 0.9603269100189209,
+      "eval_runtime": 0.6851,
+      "eval_samples_per_second": 67.147,
+      "eval_steps_per_second": 2.919,
       "step": 104
     },
     {
       "epoch": 16.92,
       "learning_rate": 4.282407407407408e-05,
-      "loss": 0.429,
+      "loss": 0.4337,
       "step": 110
     },
     {
       "epoch": 16.92,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 0.6789580583572388,
-      "eval_runtime": 0.6455,
-      "eval_samples_per_second": 71.263,
-      "eval_steps_per_second": 3.098,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.7188286781311035,
+      "eval_runtime": 0.7469,
+      "eval_samples_per_second": 61.588,
+      "eval_steps_per_second": 2.678,
       "step": 110
     },
     {
       "epoch": 18.0,
-      "eval_accuracy": 0.6304347826086957,
-      "eval_loss": 0.7073642015457153,
-      "eval_runtime": 0.6678,
-      "eval_samples_per_second": 68.88,
-      "eval_steps_per_second": 2.995,
+      "eval_accuracy": 0.6739130434782609,
+      "eval_loss": 0.7620242834091187,
+      "eval_runtime": 0.685,
+      "eval_samples_per_second": 67.149,
+      "eval_steps_per_second": 2.92,
       "step": 117
     },
     {
       "epoch": 18.46,
       "learning_rate": 4.166666666666667e-05,
-      "loss": 0.4349,
+      "loss": 0.4258,
       "step": 120
     },
     {
       "epoch": 18.92,
-      "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 0.8890200257301331,
-      "eval_runtime": 0.6748,
-      "eval_samples_per_second": 68.17,
-      "eval_steps_per_second": 2.964,
+      "eval_accuracy": 0.6739130434782609,
+      "eval_loss": 0.9432803988456726,
+      "eval_runtime": 0.6808,
+      "eval_samples_per_second": 67.567,
+      "eval_steps_per_second": 2.938,
       "step": 123
     },
     {
       "epoch": 20.0,
       "learning_rate": 4.0509259259259265e-05,
-      "loss": 0.3892,
+      "loss": 0.4045,
       "step": 130
     },
     {
       "epoch": 20.0,
-      "eval_accuracy": 0.6739130434782609,
-      "eval_loss": 0.979804515838623,
-      "eval_runtime": 0.6717,
-      "eval_samples_per_second": 68.481,
-      "eval_steps_per_second": 2.977,
+      "eval_accuracy": 0.6521739130434783,
+      "eval_loss": 1.0823415517807007,
+      "eval_runtime": 0.6591,
+      "eval_samples_per_second": 69.797,
+      "eval_steps_per_second": 3.035,
       "step": 130
     },
     {
       "epoch": 20.92,
-      "eval_accuracy": 0.6739130434782609,
-      "eval_loss": 0.8814191222190857,
-      "eval_runtime": 0.6613,
-      "eval_samples_per_second": 69.559,
-      "eval_steps_per_second": 3.024,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.7058663368225098,
+      "eval_runtime": 0.7135,
+      "eval_samples_per_second": 64.473,
+      "eval_steps_per_second": 2.803,
       "step": 136
     },
     {
       "epoch": 21.54,
       "learning_rate": 3.935185185185186e-05,
-      "loss": 0.3613,
+      "loss": 0.4135,
       "step": 140
     },
     {
       "epoch": 22.0,
-      "eval_accuracy": 0.6521739130434783,
-      "eval_loss": 0.8839537501335144,
-      "eval_runtime": 0.6839,
-      "eval_samples_per_second": 67.261,
-      "eval_steps_per_second": 2.924,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.7466621994972229,
+      "eval_runtime": 0.7352,
+      "eval_samples_per_second": 62.571,
+      "eval_steps_per_second": 2.72,
       "step": 143
     },
     {
       "epoch": 22.92,
       "eval_accuracy": 0.7391304347826086,
-      "eval_loss": 0.7662139534950256,
-      "eval_runtime": 0.6732,
-      "eval_samples_per_second": 68.326,
-      "eval_steps_per_second": 2.971,
+      "eval_loss": 0.7636581063270569,
+      "eval_runtime": 0.7007,
+      "eval_samples_per_second": 65.647,
+      "eval_steps_per_second": 2.854,
       "step": 149
     },
     {
       "epoch": 23.08,
       "learning_rate": 3.8194444444444444e-05,
-      "loss": 0.342,
+      "loss": 0.3525,
       "step": 150
     },
     {
       "epoch": 24.0,
-      "eval_accuracy": 0.7608695652173914,
-      "eval_loss": 0.7883625030517578,
-      "eval_runtime": 0.6844,
-      "eval_samples_per_second": 67.213,
-      "eval_steps_per_second": 2.922,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.8157316446304321,
+      "eval_runtime": 0.6727,
+      "eval_samples_per_second": 68.386,
+      "eval_steps_per_second": 2.973,
       "step": 156
     },
     {
       "epoch": 24.62,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.2762,
+      "loss": 0.263,
       "step": 160
     },
     {
       "epoch": 24.92,
-      "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 1.026755928993225,
-      "eval_runtime": 0.6591,
-      "eval_samples_per_second": 69.789,
-      "eval_steps_per_second": 3.034,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.9995114803314209,
+      "eval_runtime": 0.6975,
+      "eval_samples_per_second": 65.953,
+      "eval_steps_per_second": 2.868,
       "step": 162
     },
     {
       "epoch": 26.0,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 0.9206076264381409,
-      "eval_runtime": 0.6851,
-      "eval_samples_per_second": 67.14,
-      "eval_steps_per_second": 2.919,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.8719047904014587,
+      "eval_runtime": 0.6844,
+      "eval_samples_per_second": 67.214,
+      "eval_steps_per_second": 2.922,
       "step": 169
     },
     {
       "epoch": 26.15,
       "learning_rate": 3.587962962962963e-05,
-      "loss": 0.2759,
+      "loss": 0.272,
       "step": 170
     },
     {
       "epoch": 26.92,
       "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 0.9079513549804688,
-      "eval_runtime": 0.6802,
-      "eval_samples_per_second": 67.623,
-      "eval_steps_per_second": 2.94,
+      "eval_loss": 0.9938963055610657,
+      "eval_runtime": 0.7027,
+      "eval_samples_per_second": 65.464,
+      "eval_steps_per_second": 2.846,
       "step": 175
     },
     {
       "epoch": 27.69,
       "learning_rate": 3.472222222222222e-05,
-      "loss": 0.2559,
+      "loss": 0.262,
       "step": 180
     },
     {
       "epoch": 28.0,
-      "eval_accuracy": 0.6739130434782609,
-      "eval_loss": 0.9378835558891296,
-      "eval_runtime": 0.6691,
-      "eval_samples_per_second": 68.745,
-      "eval_steps_per_second": 2.989,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.8638898730278015,
+      "eval_runtime": 0.6851,
+      "eval_samples_per_second": 67.139,
+      "eval_steps_per_second": 2.919,
       "step": 182
     },
     {
       "epoch": 28.92,
-      "eval_accuracy": 0.6739130434782609,
-      "eval_loss": 1.0121444463729858,
-      "eval_runtime": 0.6814,
-      "eval_samples_per_second": 67.504,
-      "eval_steps_per_second": 2.935,
+      "eval_accuracy": 0.6521739130434783,
+      "eval_loss": 1.0737078189849854,
+      "eval_runtime": 0.7133,
+      "eval_samples_per_second": 64.491,
+      "eval_steps_per_second": 2.804,
       "step": 188
     },
     {
       "epoch": 29.23,
       "learning_rate": 3.3564814814814815e-05,
-      "loss": 0.2455,
+      "loss": 0.2282,
       "step": 190
     },
     {
       "epoch": 30.0,
-      "eval_accuracy": 0.7391304347826086,
-      "eval_loss": 0.8251720070838928,
-      "eval_runtime": 0.6693,
-      "eval_samples_per_second": 68.731,
-      "eval_steps_per_second": 2.988,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.8415572047233582,
+      "eval_runtime": 0.7645,
+      "eval_samples_per_second": 60.169,
+      "eval_steps_per_second": 2.616,
       "step": 195
     },
     {
       "epoch": 30.77,
       "learning_rate": 3.240740740740741e-05,
-      "loss": 0.2125,
+      "loss": 0.2098,
       "step": 200
     },
     {
       "epoch": 30.92,
-      "eval_accuracy": 0.7608695652173914,
-      "eval_loss": 0.8500754237174988,
-      "eval_runtime": 0.693,
-      "eval_samples_per_second": 66.38,
-      "eval_steps_per_second": 2.886,
+      "eval_accuracy": 0.6739130434782609,
+      "eval_loss": 0.9743961095809937,
+      "eval_runtime": 0.8024,
+      "eval_samples_per_second": 57.33,
+      "eval_steps_per_second": 2.493,
       "step": 201
     },
     {
       "epoch": 32.0,
-      "eval_accuracy": 0.6739130434782609,
-      "eval_loss": 1.0364569425582886,
-      "eval_runtime": 0.684,
-      "eval_samples_per_second": 67.248,
-      "eval_steps_per_second": 2.924,
+      "eval_accuracy": 0.6086956521739131,
+      "eval_loss": 1.0592586994171143,
+      "eval_runtime": 0.6877,
+      "eval_samples_per_second": 66.891,
+      "eval_steps_per_second": 2.908,
       "step": 208
     },
     {
       "epoch": 32.31,
       "learning_rate": 3.125e-05,
-      "loss": 0.2204,
+      "loss": 0.2141,
       "step": 210
     },
     {
       "epoch": 32.92,
       "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.0469846725463867,
-      "eval_runtime": 0.6841,
-      "eval_samples_per_second": 67.245,
-      "eval_steps_per_second": 2.924,
+      "eval_loss": 1.0996532440185547,
+      "eval_runtime": 0.6695,
+      "eval_samples_per_second": 68.713,
+      "eval_steps_per_second": 2.988,
       "step": 214
     },
     {
       "epoch": 33.85,
       "learning_rate": 3.0092592592592593e-05,
-      "loss": 0.1785,
+      "loss": 0.1759,
       "step": 220
     },
     {
       "epoch": 34.0,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 0.8834150433540344,
-      "eval_runtime": 0.6551,
-      "eval_samples_per_second": 70.218,
-      "eval_steps_per_second": 3.053,
+      "eval_accuracy": 0.5869565217391305,
+      "eval_loss": 0.973516047000885,
+      "eval_runtime": 0.7227,
+      "eval_samples_per_second": 63.647,
+      "eval_steps_per_second": 2.767,
       "step": 221
     },
     {
       "epoch": 34.92,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.0779674053192139,
-      "eval_runtime": 0.6643,
-      "eval_samples_per_second": 69.243,
-      "eval_steps_per_second": 3.011,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 1.078919529914856,
+      "eval_runtime": 0.7302,
+      "eval_samples_per_second": 62.996,
+      "eval_steps_per_second": 2.739,
       "step": 227
     },
     {
       "epoch": 35.38,
       "learning_rate": 2.8935185185185186e-05,
-      "loss": 0.1912,
+      "loss": 0.2042,
       "step": 230
     },
     {
       "epoch": 36.0,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 0.9327965378761292,
-      "eval_runtime": 0.6863,
-      "eval_samples_per_second": 67.022,
-      "eval_steps_per_second": 2.914,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 1.0664314031600952,
+      "eval_runtime": 0.7008,
+      "eval_samples_per_second": 65.637,
+      "eval_steps_per_second": 2.854,
       "step": 234
     },
     {
       "epoch": 36.92,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.1518,
+      "loss": 0.1591,
       "step": 240
     },
     {
       "epoch": 36.92,
-      "eval_accuracy": 0.782608695652174,
-      "eval_loss": 0.8900798559188843,
-      "eval_runtime": 0.6842,
-      "eval_samples_per_second": 67.236,
-      "eval_steps_per_second": 2.923,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.941728413105011,
+      "eval_runtime": 0.7014,
+      "eval_samples_per_second": 65.585,
+      "eval_steps_per_second": 2.852,
       "step": 240
     },
     {
       "epoch": 38.0,
       "eval_accuracy": 0.6739130434782609,
-      "eval_loss": 1.1069260835647583,
-      "eval_runtime": 0.6697,
-      "eval_samples_per_second": 68.692,
-      "eval_steps_per_second": 2.987,
+      "eval_loss": 1.1041595935821533,
+      "eval_runtime": 0.7009,
+      "eval_samples_per_second": 65.627,
+      "eval_steps_per_second": 2.853,
       "step": 247
     },
     {
       "epoch": 38.46,
       "learning_rate": 2.6620370370370372e-05,
-      "loss": 0.166,
+      "loss": 0.1579,
       "step": 250
     },
     {
       "epoch": 38.92,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 0.9823449850082397,
-      "eval_runtime": 0.6724,
-      "eval_samples_per_second": 68.414,
-      "eval_steps_per_second": 2.975,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.9731682538986206,
+      "eval_runtime": 0.7032,
+      "eval_samples_per_second": 65.415,
+      "eval_steps_per_second": 2.844,
       "step": 253
     },
     {
       "epoch": 40.0,
       "learning_rate": 2.5462962962962965e-05,
-      "loss": 0.1775,
+      "loss": 0.1626,
       "step": 260
     },
     {
       "epoch": 40.0,
       "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 0.9712538719177246,
-      "eval_runtime": 0.6616,
-      "eval_samples_per_second": 69.532,
-      "eval_steps_per_second": 3.023,
+      "eval_loss": 0.9960243105888367,
+      "eval_runtime": 0.6961,
+      "eval_samples_per_second": 66.083,
+      "eval_steps_per_second": 2.873,
       "step": 260
     },
     {
       "epoch": 40.92,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 0.972922146320343,
-      "eval_runtime": 0.6677,
-      "eval_samples_per_second": 68.892,
-      "eval_steps_per_second": 2.995,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.9763407707214355,
+      "eval_runtime": 0.6715,
+      "eval_samples_per_second": 68.504,
+      "eval_steps_per_second": 2.978,
       "step": 266
     },
     {
       "epoch": 41.54,
       "learning_rate": 2.4305555555555558e-05,
-      "loss": 0.1344,
+      "loss": 0.1458,
       "step": 270
     },
     {
       "epoch": 42.0,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 0.9956928491592407,
-      "eval_runtime": 0.6673,
-      "eval_samples_per_second": 68.936,
-      "eval_steps_per_second": 2.997,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.9789978265762329,
+      "eval_runtime": 0.7024,
+      "eval_samples_per_second": 65.487,
+      "eval_steps_per_second": 2.847,
       "step": 273
     },
     {
       "epoch": 42.92,
-      "eval_accuracy": 0.7391304347826086,
-      "eval_loss": 1.0180084705352783,
-      "eval_runtime": 0.6819,
-      "eval_samples_per_second": 67.46,
-      "eval_steps_per_second": 2.933,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 1.0952370166778564,
+      "eval_runtime": 0.7444,
+      "eval_samples_per_second": 61.796,
+      "eval_steps_per_second": 2.687,
       "step": 279
     },
     {
       "epoch": 43.08,
       "learning_rate": 2.314814814814815e-05,
-      "loss": 0.1232,
+      "loss": 0.1317,
       "step": 280
     },
     {
       "epoch": 44.0,
-      "eval_accuracy": 0.782608695652174,
-      "eval_loss": 0.9668875336647034,
-      "eval_runtime": 0.6863,
-      "eval_samples_per_second": 67.023,
-      "eval_steps_per_second": 2.914,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.9190374612808228,
+      "eval_runtime": 0.6983,
+      "eval_samples_per_second": 65.873,
+      "eval_steps_per_second": 2.864,
       "step": 286
     },
     {
       "epoch": 44.62,
       "learning_rate": 2.1990740740740743e-05,
-      "loss": 0.1267,
+      "loss": 0.1255,
       "step": 290
     },
     {
       "epoch": 44.92,
-      "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 0.9764741063117981,
-      "eval_runtime": 0.6906,
-      "eval_samples_per_second": 66.611,
-      "eval_steps_per_second": 2.896,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.9420353174209595,
+      "eval_runtime": 0.6626,
+      "eval_samples_per_second": 69.419,
+      "eval_steps_per_second": 3.018,
       "step": 292
     },
     {
       "epoch": 46.0,
       "eval_accuracy": 0.7391304347826086,
-      "eval_loss": 1.0389167070388794,
-      "eval_runtime": 0.8195,
-      "eval_samples_per_second": 56.133,
-      "eval_steps_per_second": 2.441,
+      "eval_loss": 0.9084758162498474,
+      "eval_runtime": 0.7031,
+      "eval_samples_per_second": 65.429,
+      "eval_steps_per_second": 2.845,
       "step": 299
     },
     {
       "epoch": 46.15,
       "learning_rate": 2.0833333333333336e-05,
-      "loss": 0.1548,
+      "loss": 0.1352,
       "step": 300
     },
     {
       "epoch": 46.92,
       "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.001569390296936,
-      "eval_runtime": 0.6415,
-      "eval_samples_per_second": 71.702,
-      "eval_steps_per_second": 3.117,
+      "eval_loss": 0.9183786511421204,
+      "eval_runtime": 0.6966,
+      "eval_samples_per_second": 66.031,
+      "eval_steps_per_second": 2.871,
       "step": 305
     },
     {
       "epoch": 47.69,
       "learning_rate": 1.967592592592593e-05,
-      "loss": 0.1267,
+      "loss": 0.1311,
       "step": 310
     },
     {
       "epoch": 48.0,
-      "eval_accuracy": 0.7391304347826086,
-      "eval_loss": 1.156539797782898,
-      "eval_runtime": 0.6692,
-      "eval_samples_per_second": 68.739,
-      "eval_steps_per_second": 2.989,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 1.0567129850387573,
+      "eval_runtime": 0.703,
+      "eval_samples_per_second": 65.438,
+      "eval_steps_per_second": 2.845,
       "step": 312
     },
     {
       "epoch": 48.92,
       "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.1795543432235718,
-      "eval_runtime": 0.7207,
-      "eval_samples_per_second": 63.831,
-      "eval_steps_per_second": 2.775,
+      "eval_loss": 1.1506885290145874,
+      "eval_runtime": 0.6688,
+      "eval_samples_per_second": 68.778,
+      "eval_steps_per_second": 2.99,
       "step": 318
     },
     {
       "epoch": 49.23,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.1403,
+      "loss": 0.1501,
       "step": 320
     },
     {
       "epoch": 50.0,
-      "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 1.2806681394577026,
-      "eval_runtime": 0.6946,
-      "eval_samples_per_second": 66.226,
-      "eval_steps_per_second": 2.879,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 1.2067523002624512,
+      "eval_runtime": 0.7028,
+      "eval_samples_per_second": 65.449,
+      "eval_steps_per_second": 2.846,
       "step": 325
     },
     {
       "epoch": 50.77,
       "learning_rate": 1.736111111111111e-05,
-      "loss": 0.1096,
+      "loss": 0.1088,
       "step": 330
     },
     {
       "epoch": 50.92,
-      "eval_accuracy": 0.6739130434782609,
-      "eval_loss": 1.2462576627731323,
-      "eval_runtime": 0.7307,
-      "eval_samples_per_second": 62.954,
-      "eval_steps_per_second": 2.737,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 1.4607298374176025,
+      "eval_runtime": 0.6951,
+      "eval_samples_per_second": 66.177,
+      "eval_steps_per_second": 2.877,
       "step": 331
     },
     {
       "epoch": 52.0,
-      "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 1.145448088645935,
-      "eval_runtime": 0.7031,
-      "eval_samples_per_second": 65.426,
-      "eval_steps_per_second": 2.845,
+      "eval_accuracy": 0.6739130434782609,
+      "eval_loss": 1.1035622358322144,
+      "eval_runtime": 0.7164,
+      "eval_samples_per_second": 64.214,
+      "eval_steps_per_second": 2.792,
       "step": 338
     },
     {
       "epoch": 52.31,
       "learning_rate": 1.6203703703703704e-05,
-      "loss": 0.1106,
+      "loss": 0.1152,
       "step": 340
     },
     {
       "epoch": 52.92,
       "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 1.1494420766830444,
-      "eval_runtime": 0.6634,
-      "eval_samples_per_second": 69.34,
-      "eval_steps_per_second": 3.015,
+      "eval_loss": 1.1081277132034302,
+      "eval_runtime": 0.6705,
+      "eval_samples_per_second": 68.605,
+      "eval_steps_per_second": 2.983,
       "step": 344
     },
     {
       "epoch": 53.85,
       "learning_rate": 1.5046296296296297e-05,
-      "loss": 0.1116,
+      "loss": 0.1141,
       "step": 350
     },
     {
       "epoch": 54.0,
       "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 1.1299750804901123,
-      "eval_runtime": 0.6843,
-      "eval_samples_per_second": 67.217,
-      "eval_steps_per_second": 2.922,
+      "eval_loss": 1.1005905866622925,
+      "eval_runtime": 0.7035,
+      "eval_samples_per_second": 65.389,
+      "eval_steps_per_second": 2.843,
       "step": 351
     },
     {
       "epoch": 54.92,
       "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.2097675800323486,
-      "eval_runtime": 0.7032,
-      "eval_samples_per_second": 65.42,
-      "eval_steps_per_second": 2.844,
+      "eval_loss": 1.1470131874084473,
+      "eval_runtime": 0.7191,
+      "eval_samples_per_second": 63.965,
+      "eval_steps_per_second": 2.781,
       "step": 357
     },
     {
       "epoch": 55.38,
       "learning_rate": 1.388888888888889e-05,
-      "loss": 0.1229,
+      "loss": 0.1307,
       "step": 360
     },
     {
       "epoch": 56.0,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.0590535402297974,
-      "eval_runtime": 0.6998,
-      "eval_samples_per_second": 65.732,
-      "eval_steps_per_second": 2.858,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 1.0714861154556274,
+      "eval_runtime": 0.7054,
+      "eval_samples_per_second": 65.215,
+      "eval_steps_per_second": 2.835,
       "step": 364
     },
     {
       "epoch": 56.92,
       "learning_rate": 1.2731481481481482e-05,
-      "loss": 0.1235,
+      "loss": 0.1273,
       "step": 370
     },
     {
       "epoch": 56.92,
-      "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 1.1228772401809692,
-      "eval_runtime": 0.6702,
-      "eval_samples_per_second": 68.637,
-      "eval_steps_per_second": 2.984,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 1.1020547151565552,
+      "eval_runtime": 0.6654,
+      "eval_samples_per_second": 69.128,
+      "eval_steps_per_second": 3.006,
       "step": 370
     },
     {
       "epoch": 58.0,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.2034395933151245,
-      "eval_runtime": 0.6511,
-      "eval_samples_per_second": 70.647,
-      "eval_steps_per_second": 3.072,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 1.1176313161849976,
+      "eval_runtime": 0.721,
+      "eval_samples_per_second": 63.8,
+      "eval_steps_per_second": 2.774,
       "step": 377
     },
     {
       "epoch": 58.46,
       "learning_rate": 1.1574074074074075e-05,
-      "loss": 0.107,
+      "loss": 0.1066,
       "step": 380
     },
     {
       "epoch": 58.92,
       "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.0628197193145752,
-      "eval_runtime": 0.6967,
-      "eval_samples_per_second": 66.021,
-      "eval_steps_per_second": 2.87,
+      "eval_loss": 1.0948173999786377,
+      "eval_runtime": 0.7029,
+      "eval_samples_per_second": 65.446,
+      "eval_steps_per_second": 2.845,
       "step": 383
     },
     {
       "epoch": 60.0,
       "learning_rate": 1.0416666666666668e-05,
-      "loss": 0.107,
+      "loss": 0.1046,
       "step": 390
     },
     {
       "epoch": 60.0,
       "eval_accuracy": 0.7391304347826086,
-      "eval_loss": 1.0070008039474487,
-      "eval_runtime": 0.6853,
-      "eval_samples_per_second": 67.122,
-      "eval_steps_per_second": 2.918,
+      "eval_loss": 1.0563088655471802,
+      "eval_runtime": 0.8502,
+      "eval_samples_per_second": 54.104,
+      "eval_steps_per_second": 2.352,
       "step": 390
     },
     {
       "epoch": 60.92,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.16019868850708,
-      "eval_runtime": 0.6966,
-      "eval_samples_per_second": 66.035,
-      "eval_steps_per_second": 2.871,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 1.1155492067337036,
+      "eval_runtime": 0.8171,
+      "eval_samples_per_second": 56.296,
+      "eval_steps_per_second": 2.448,
       "step": 396
     },
     {
       "epoch": 61.54,
       "learning_rate": 9.259259259259259e-06,
-      "loss": 0.1167,
+      "loss": 0.1129,
       "step": 400
     },
     {
       "epoch": 62.0,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.0720371007919312,
-      "eval_runtime": 0.7007,
-      "eval_samples_per_second": 65.652,
-      "eval_steps_per_second": 2.854,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 1.0921597480773926,
+      "eval_runtime": 0.7185,
+      "eval_samples_per_second": 64.024,
+      "eval_steps_per_second": 2.784,
       "step": 403
     },
     {
       "epoch": 62.92,
-      "eval_accuracy": 0.7391304347826086,
-      "eval_loss": 0.9726494550704956,
-      "eval_runtime": 0.6716,
-      "eval_samples_per_second": 68.496,
-      "eval_steps_per_second": 2.978,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 1.0364007949829102,
+      "eval_runtime": 0.6975,
+      "eval_samples_per_second": 65.954,
+      "eval_steps_per_second": 2.868,
       "step": 409
     },
     {
       "epoch": 63.08,
       "learning_rate": 8.101851851851852e-06,
-      "loss": 0.1113,
+      "loss": 0.1031,
       "step": 410
     },
     {
       "epoch": 64.0,
       "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.032416582107544,
-      "eval_runtime": 0.684,
-      "eval_samples_per_second": 67.253,
-      "eval_steps_per_second": 2.924,
+      "eval_loss": 1.0675166845321655,
+      "eval_runtime": 0.7196,
+      "eval_samples_per_second": 63.922,
+      "eval_steps_per_second": 2.779,
       "step": 416
     },
     {
       "epoch": 64.62,
       "learning_rate": 6.944444444444445e-06,
-      "loss": 0.0838,
+      "loss": 0.0808,
       "step": 420
     },
     {
       "epoch": 64.92,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.1092015504837036,
-      "eval_runtime": 0.6999,
-      "eval_samples_per_second": 65.722,
-      "eval_steps_per_second": 2.857,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 1.1133311986923218,
+      "eval_runtime": 0.6707,
+      "eval_samples_per_second": 68.583,
+      "eval_steps_per_second": 2.982,
       "step": 422
     },
     {
       "epoch": 66.0,
-      "eval_accuracy": 0.6956521739130435,
-      "eval_loss": 1.177163004875183,
-      "eval_runtime": 0.7851,
-      "eval_samples_per_second": 58.589,
-      "eval_steps_per_second": 2.547,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 1.202866554260254,
+      "eval_runtime": 0.7346,
+      "eval_samples_per_second": 62.622,
+      "eval_steps_per_second": 2.723,
       "step": 429
     },
     {
       "epoch": 66.15,
       "learning_rate": 5.787037037037038e-06,
-      "loss": 0.083,
+      "loss": 0.0783,
       "step": 430
     },
     {
       "epoch": 66.92,
       "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.1195148229599,
-      "eval_runtime": 0.6775,
-      "eval_samples_per_second": 67.893,
-      "eval_steps_per_second": 2.952,
+      "eval_loss": 1.1452856063842773,
+      "eval_runtime": 0.7029,
+      "eval_samples_per_second": 65.446,
+      "eval_steps_per_second": 2.845,
       "step": 435
     },
     {
       "epoch": 67.69,
       "learning_rate": 4.6296296296296296e-06,
-      "loss": 0.0899,
+      "loss": 0.09,
       "step": 440
     },
     {
       "epoch": 68.0,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.068082332611084,
-      "eval_runtime": 0.668,
-      "eval_samples_per_second": 68.86,
-      "eval_steps_per_second": 2.994,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 1.0924614667892456,
+      "eval_runtime": 0.7016,
+      "eval_samples_per_second": 65.561,
+      "eval_steps_per_second": 2.85,
       "step": 442
     },
     {
       "epoch": 68.92,
       "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.1160058975219727,
-      "eval_runtime": 0.6862,
-      "eval_samples_per_second": 67.036,
-      "eval_steps_per_second": 2.915,
+      "eval_loss": 1.0999163389205933,
+      "eval_runtime": 0.6805,
+      "eval_samples_per_second": 67.598,
+      "eval_steps_per_second": 2.939,
       "step": 448
     },
     {
       "epoch": 69.23,
       "learning_rate": 3.4722222222222224e-06,
-      "loss": 0.0813,
+      "loss": 0.0796,
       "step": 450
     },
     {
       "epoch": 70.0,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.143452525138855,
-      "eval_runtime": 0.6851,
-      "eval_samples_per_second": 67.142,
-      "eval_steps_per_second": 2.919,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 1.0970680713653564,
+      "eval_runtime": 0.703,
+      "eval_samples_per_second": 65.43,
+      "eval_steps_per_second": 2.845,
       "step": 455
     },
     {
       "epoch": 70.77,
       "learning_rate": 2.3148148148148148e-06,
-      "loss": 0.0782,
+      "loss": 0.0828,
       "step": 460
     },
     {
       "epoch": 70.92,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.1463539600372314,
-      "eval_runtime": 0.6819,
-      "eval_samples_per_second": 67.457,
-      "eval_steps_per_second": 2.933,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 1.092252492904663,
+      "eval_runtime": 0.69,
+      "eval_samples_per_second": 66.666,
+      "eval_steps_per_second": 2.899,
       "step": 461
     },
     {
       "epoch": 72.0,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.1538991928100586,
-      "eval_runtime": 0.7861,
-      "eval_samples_per_second": 58.52,
-      "eval_steps_per_second": 2.544,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 1.106107473373413,
+      "eval_runtime": 0.7839,
+      "eval_samples_per_second": 58.68,
+      "eval_steps_per_second": 2.551,
       "step": 468
     },
     {
       "epoch": 72.31,
       "learning_rate": 1.1574074074074074e-06,
-      "loss": 0.1014,
+      "loss": 0.0923,
       "step": 470
     },
     {
       "epoch": 72.92,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.158526062965393,
-      "eval_runtime": 0.7055,
-      "eval_samples_per_second": 65.203,
-      "eval_steps_per_second": 2.835,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 1.1173276901245117,
+      "eval_runtime": 0.6574,
+      "eval_samples_per_second": 69.969,
+      "eval_steps_per_second": 3.042,
       "step": 474
     },
     {
       "epoch": 73.85,
       "learning_rate": 0.0,
-      "loss": 0.0944,
+      "loss": 0.092,
       "step": 480
     },
     {
       "epoch": 73.85,
-      "eval_accuracy": 0.717391304347826,
-      "eval_loss": 1.1592216491699219,
-      "eval_runtime": 0.7479,
-      "eval_samples_per_second": 61.508,
-      "eval_steps_per_second": 2.674,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 1.1208199262619019,
+      "eval_runtime": 0.7226,
+      "eval_samples_per_second": 63.658,
+      "eval_steps_per_second": 2.768,
       "step": 480
     },
     {
       "epoch": 73.85,
       "step": 480,
       "total_flos": 1.5036476561209958e+18,
-      "train_loss": 0.34541464410722256,
-      "train_runtime": 589.4401,
-      "train_samples_per_second": 111.156,
-      "train_steps_per_second": 0.814
+      "train_loss": 0.3456469999005397,
+      "train_runtime": 590.3301,
+      "train_samples_per_second": 110.989,
+      "train_steps_per_second": 0.813
     }
   ],
   "logging_steps": 10,