Dataset · neural-bridge
12,000-Example RAG Dataset
neural-bridge/rag-dataset-12000
An English dataset of 12,000 examples for training and evaluating retrieval-augmented generation models.
What's inside
An English dataset by Neural Bridge AI under the Apache 2.0 license, designed for RAG models. It helps LLMs query an external knowledge base prior to generating responses. The card does not disclose exact field compositions or formats, specifying only a total volume of 12,000 examples.
What you can do with it
- Fine-tuning RAG models
- Evaluating search and generation pipelines
- Evaluating LLMs with an external knowledge base
- Tasks
- Question answering
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 1.1K
- Likes
- 201
Models trained on it
- mradermacher/Mistral-Nemo-NT-Ko-12B-sft-i1-GGUF 1.8K
- vllm-sr/halugate-sentinelText classification 1.6K
- mradermacher/karakuri-lm-8x7b-instruct-v0.1-i1-GGUF 1.2K
- mradermacher/SatellaRAG-0.8B-GGUF 758
- mradermacher/Llama-3.2-Taiwan-3B-Instruct-i1-GGUF 591
- EphAsad/SatellaRAG-0.8BText generation 464
- QuantFactory/Llama-3.2-Taiwan-3B-Instruct-GGUFText generation 460
- mradermacher/Satella-RAG-27B-GGUF 375
How to load it
from datasets import load_dataset
ds = load_dataset("neural-bridge/rag-dataset-12000")