view article Article From Golden Gate Bridge to Broken JSON: Why Anthropic's SAE Steering Fails for Structured Output MaziyarPanahi β’ Feb 7 β’ 22
πͺ SmolLM Collection A series of smol LLMs: 135M, 360M and 1.7B. We release base and Instruct models as well as the training corpus and some WebGPU demos β’ 12 items β’ Updated May 5, 2025 β’ 256
The Big Benchmarks Collection Collection Gathering benchmark spaces on the hub (beyond the Open LLM Leaderboard) β’ 13 items β’ Updated Nov 18, 2024 β’ 274