brainstorming
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).
Explorar Skills
Desarrollo · SKILL.md
lychee-decode-sparse-kv-sharing
lychee-decode-sparse-kv-sharingClassify attention heads into retrieval (full attention) and sparse (token-selected) roles using HardKuma distribution for differentiable discrete optimization. Sparse heads reuse KV pairs from retrieval heads, reducing cache by 90% while maintaining quality through joint training.
Independent GitHub snapshot · Automated classification · Original documentation governs
Documentación de la Skill
GitHubCargando documentación desde el origen…