v2: cap majority CWEs at 2500 (69,386 rows; rebalanced long tail)

README.md

@@ -24,16 +24,16 @@ dataset_info:
       dtype: string
   splits:
   - name: train
-    num_bytes: 50280037
-    num_examples: 91862
+    num_bytes: 31040406
+    num_examples: 55810
   - name: validation
-    num_bytes: 7510490
-    num_examples: 13705
+    num_bytes: 3754679
+    num_examples: 6774
   - name: test
-    num_bytes: 6030266
-    num_examples: 11226
-  download_size: 63558517
-  dataset_size: 63820793
+    num_bytes: 3737846
+    num_examples: 6802
+  download_size: 38388484
+  dataset_size: 38532931
 configs:
 - config_name: default
   data_files:

data/test-00000-of-00001.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a90134be556c47c8808388f60c50e4e2f120b0c8bea3303ee44ed7cd3dac8cc8
-size 6005903
+oid sha256:6d621cf09c8380918eb02eaa73173b2a1ace84ba77b74bb02e4657485f091f7e
+size 3724895

data/train-00000-of-00001.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48c36fd0ee46947ed31507574342354058542ae3d3dc3a71d8ffaa17f9353cc4
-size 50072371
+oid sha256:8aa85fee48e4e0c59925238d9ea49627dcbe77f7e74746ce92b49914a7eea173
+size 30919635

data/validation-00000-of-00001.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d6bc439fba04bad34ece33fa9750126e20aa73b0f16cf94fdd1a81faef24f20
-size 7480243
+oid sha256:193ad13676bbddc8e8f12c0cb8f0ff84b18c13d076df8150f862461b426ec469
+size 3743954