data.jsonl

{"doc_id":"1","chunk":"Retrieval-Augmented Generation combines retrieval with generation.","source":"AI docs"}
{"doc_id":"1","chunk":"RAG improves factual accuracy by grounding responses.","source":"AI docs"}
{"doc_id":"2","chunk":"Vector databases enable semantic search.","source":"Tech docs"}
{"doc_id":"2","chunk":"Embeddings map text to numerical vectors.","source":"Tech docs"}
{"doc_id":"3","chunk":"Context windows limit model memory.","source":"LLM docs"}

README.md

@@ -0,0 +1,11 @@
+---
+license: mit
+task_categories:
+- feature-extraction
+language:
+- en
+---
+
+# RAG Document Chunks Dataset
+
+Document chunks optimized for retrieval-augmented generation pipelines.