Dataset Discovery: Finding Public Datasets via Chat

Use natural language to find public datasets to benchmark or supplement your own logs.

Overview

When debugging your own agents and chatbots, it is often useful to compare against a public reference dataset, or to pull in domain-specific corpora to test prompts. Use the Hyperparam chat interface to search Hugging Face directly without leaving the workspace.

Steps

1. Open Hyperparam Chat

   Access the chat interface from the main navigation

2. Search using natural language

   Example query: "find me anonymized patient data with medical charting"

   > Note: Hyperparam Chat will return data sets from Hugging Face that match criteria

3. Open dataset from results

   Click on a result (e.g., chunked-ehr/0000) to open in the data viewer

   > Note: Dataset loads with all columns and metadata

Expected Results

• Quick discovery: Natural language search returns relevant datasets
• Direct access: One-click opening into data viewer
• Context preserved: Chat understands domain-specific terminology (medical, ML, etc.)

Other Use Cases

• Classifying Prompt Patterns: Categorize unstructured prompts to see your real traffic mix
• Patient Data Workflow: Extract, filter, and export structured medical data
• Quality Filtering: Remove low-quality responses from chat logs
• Deep Research: Multi-step AI workflow for comparing model outputs