Chris Potts' Team Shares Memorization Experiments Built on the Hubble Model Suite
ChrisGPotts · x · 2026-10-23
Stanford's Chris Potts describes an experiment codebase containing multiple related experiments that all converge on the same overall message about LLM memorization. He also credits the Hubble Project team (@johntzwei, @robinomial et al.) for the model suite used, calling Hubble an incredible resource for memorization research.
More from Models
- OpenAI doc leak reveals unreleased GPT-6 Astra Cyber model used in cyber defense tests — scaling01 · 2026-10-24
- "70% of Kimi K3's original cost is pretty good," says AI commentator — ChrisGPT · 2026-10-24
- Grok bot demo shows sign-in autofill with credentials kept in its secure vault — billyjhowell · 2026-10-24
- Fireworks explains how it made RL practical on Kimi K3 at 2.8T parameters — sophiamyang · 2026-10-24
- Astra finds bug in Anthropic's formal verification example: spec doesn't check output matches input — teortaxesTex · 2026-10-24
- Update: open weights for the Fable-class local model coming November 4th — kimmonismus · 2026-10-24