Upload README.md with huggingface_hub

README.md

@@ -4,7 +4,7 @@ dataset_info:
   - name: anchor
     list:
     - name: align_accuracy
-      dtype: 'null'
+      dtype: float64
     - name: align_score
       dtype: float64
     - name: attribution_label
@@ -22,33 +22,33 @@ dataset_info:
     - name: id
       dtype: string
     - name: label_switch
-      dtype: 'null'
+      dtype: float64
     - name: nli_accuracy
-      dtype: 'null'
+      dtype: float64
     - name: nli_logit
-      dtype: 'null'
+      dtype: float64
     - name: nli_score
       dtype: int64
     - name: original_label
-      dtype: 'null'
+      dtype: float64
     - name: question
       dtype: string
     - name: qwen30B_accuracy
-      dtype: 'null'
+      dtype: float64
     - name: qwen30B_score
-      dtype: 'null'
+      dtype: float64
     - name: qwen4Bnoth_accuracy
-      dtype: 'null'
+      dtype: float64
     - name: qwen4Bnoth_score
-      dtype: 'null'
+      dtype: float64
     - name: qwen4Bth_accuracy
-      dtype: 'null'
+      dtype: float64
     - name: qwen4Bth_score
-      dtype: 'null'
+      dtype: float64
     - name: qwen8B_accuracy
-      dtype: 'null'
+      dtype: float64
     - name: qwen8B_score
-      dtype: 'null'
+      dtype: float64
     - name: references
       sequence: string
     - name: response
@@ -171,10 +171,10 @@ dataset_info:
       sequence: 'null'
   splits:
   - name: train
-    num_bytes: 153979126
+    num_bytes: 154721862
     num_examples: 7626
-  download_size: 40128282
-  dataset_size: 153979126
+  download_size: 40154330
+  dataset_size: 154721862
 ---
 # Dataset Card for "attributionBench_hardpos_augmentedQwen30B_allerrors_contrastive"