Ripple ladder: new dataset gives 4x gain, edging out 3x from new architecture
finbarrtimbers · x · 2026-11-05
finbarrtimbers is surprised by how equal the order-of-magnitude gains are in the recent Ripple training ladder: Ripple 1.6 vs Olmo3 shares the same dataset but a new architecture (3x improvement), while Ripple final vs 1.6 keeps the architecture and swaps in a new dataset (4x improvement). He expected data to matter far more than architecture, but the results show roughly equal contributions.
Related event: Ripple Ladder Experiments Suggest Data Matters More Than Architecture(2 posts)→
More from Research
- New Book on Generative AI and Stochastic Thermodynamics by Welling Team Out This Month — mjdramstead · 2026-11-05
- New paper uses 288M job postings to measure where AI raises the value of work it can't do — robseamans · 2026-11-05
- Matthew Honnibal: LLMs can 'see' weight matrices — but only one layer at a time — honnibal · 2026-11-05
- RicursiveAI claims 16x compute efficiency gain on code pretraining vs Olmo 3 baseline — samiramanabi · 2026-11-05
- Claude drifts from 🌀 to 💖: refined AI symbols become less legible to humans (67%→44%) — mimi10v3 · 2026-11-05
- Engram: brief training fixes long-run models' increasingly inscrutable notes — scott_linderman · 2026-11-05