Fix dataset_info: remove explicit features to avoid schema mismatch, keep num_examples=114074530

README.md

@@ -21,23 +21,6 @@ size_categories:
   - 100M<n<1B
 dataset_info:
   - config_name: signals
-    features:
-      - name: timestamp
-        dtype: float64
-      - name: message_type
-        dtype: string
-      - name: stream_name
-        dtype: string
-      - name: label_binary
-        dtype: string
-      - name: suspicion_score
-        dtype: float32
-      - name: matched_rules
-        dtype: string
-      - name: product_name
-        dtype: string
-      - name: vendor_name
-        dtype: string
     splits:
       - name: train
         num_examples: 114074530