Created filtered and recaptioned dataset

README.md

@@ -0,0 +1,21 @@
+---
+dataset_info:
+  features:
+  - name: prompt
+    dtype: string
+  - name: chosen
+    dtype: image
+  - name: rejected
+    dtype: image
+  splits:
+  - name: train
+    num_bytes: 1915104723
+    num_examples: 620
+  download_size: 1915248356
+  dataset_size: 1915104723
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train-*
+---

data/train-00000-of-00004.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fa516977c437cb6a1965cb01b47abe33a6e87f5ed1dc1db051201fcaab11333
+size 478890082

data/train-00001-of-00004.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c96c9ef6b584979bfcdd12ec4ba6357bf0e9a3b77995616525779e409545025
+size 484161338

data/train-00002-of-00004.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f76f57c1d91524d151a5ce9fe03927461dcb16d0f8bd942688b6daab94997d1a
+size 475499325

data/train-00003-of-00004.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc8a41a5e33c20505de325ca31abc39dc271bc2871d7e27787362579272444a1
+size 476697611