Training Data

concepts · 4 notes linked

Related: Large Language Models · Generative AI · Hugging Face · Servicenow · Bigcode · Code Generation · Openai · Chatgpt

Notes

• 'If journalism is going up in smoke, I might as well get high off the fumes': confessions of a chatbot helper — Human annotators writing gold-standard training data for LLMs
• Generative AI Training and Copyright Law — Legal analysis of copyright liability for generative AI training data collection
• How Many Images Does It Take? Estimating Imitation Thresholds in Text-to-Image Models — Empirical threshold at which text-to-image models begin imitating training concepts
• Meet 'Stack,' A 3TB of Permissively Licensed Source Code for LLMs (Large Language Models) — BigCode releases 3TB permissively licensed code dataset for LLMs