Get the app
LLM & Text Generation1 min reading time

Bad Guy Tokens

LessWrong
Read the full article
The article discusses the concept of 'Bad Guy Tokens' in AI, exploring how certain tokens or inputs can cause undesirable or harmful behavior in language models. It examines the implications and challenges of managing these tokens to ensure safer AI interactions.

More in LLM & Text Generation

LLM & Text Generation2 min read

Meta releases Muse Spark Open, a new ‘open-weight’ AI model

Covered by 9 sources
LLM & Text Generation11 min read

Scaling Indic Pretraining: 2.8 Trillion Tokens and an Open Data Generation Pipeline

Covered by 2 sources
LLM & Text Generation4 min read

Granite Switch 4.1: Chat, Decisions, and Speech from One Endpoint

Covered by 2 sources