Skip to content

Dataset · bitext

Synthetic dataset for retail chatbots

bitext/Bitext-retail-ecommerce-llm-chatbot-training-dataset

10,000 to 100,000 labeled English examples for fine-tuning e-commerce virtual assistants.

What's inside

A hybrid synthetic English text dataset created using NLP/NLG technologies and automated labeling tools. It contains labeled data for question-answering and table-question-answering tasks in the retail and e-commerce sector. The dataset is designed to adapt large language models (such as GPT, Mistral, and OpenELM) for virtual assistants.

What you can do with it

  • SFT fine-tuning of LLMs for online store customer support tasks
  • Adapting chatbots to answer questions using tabular data and product catalogs
Tasks
Question answering, Table question answering
Languages
English
Size
10K–100K rows
Downloads this month
683
Likes
22

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("bitext/Bitext-retail-ecommerce-llm-chatbot-training-dataset")