Upload honesty-alignment dataset with train/test/ood splits

README.md

@@ -0,0 +1,31 @@
+---
+dataset_info:
+  features:
+  - name: question
+    dtype: string
+  - name: answer
+    list: string
+  - name: source
+    dtype: string
+  splits:
+  - name: train
+    num_bytes: 77764906
+    num_examples: 567647
+  - name: test
+    num_bytes: 6978337
+    num_examples: 37904
+  - name: ood
+    num_bytes: 4132578
+    num_examples: 32805
+  download_size: 42780964
+  dataset_size: 88875821
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+  - split: test
+    path: data/test-*
+  - split: ood
+    path: data/ood-*
+---

data/ood-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c95863512124e98470fc6055f0ed1657bd5716f1d33404bbdad9ad016fb50c0c
+size 1541232

data/test-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2029082a0f5bfe35099fbe695a128070ecd9e6392e91cdf901f94e94747b6fad
+size 3710852

data/train-00000-of-00001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fb18101d1501b8f6fcc6e749d627d822eb4c2d69e4acd78f3de5acbd86d1e5e
+size 37528880