GitHub
mechinterp-summarizer
Convert experiment results to Markdown research notes and optionally update research state with evidence
☆ 1查看技能
cesaregarza AI Skills
GitHub ↗浏览 cesaregarza 维护的已发布 AI 技能，比较用途与说明，并通过来源仓库了解安装方式。
作者统计
已收录技能
15收录技能的仓库 Stars 合计
15仓库 Forks
0平均每个技能的仓库 Stars
1同一仓库中的多个技能可能共享 Stars，合计包含重复仓库数据。
热门分类
近期收录动态
全部技能
GitHub
mechinterp-decoder
Analyze SAE decoder weights - output influence, feature importance, and decoder similarity
☆ 1查看技能
GitHub
mechinterp-validation-suite
Run credibility checks on feature interpretations including split-half stability and shuffle null tests
☆ 1查看技能
GitHub
mechinterp-glossary-and-constraints
Reference skill for Splatoon ability families, AP rungs, token constraints, and domain rules for mechinterp experiments
☆ 1查看技能
GitHub
mechinterp-state
Manage research state for SAE feature analysis including hypotheses, evidence tracking, and research history
☆ 1查看技能
GitHub
mechinterp-next-step-planner
Propose next mechanistic interpretability experiments based on research state, hypotheses, and existing evidence
☆ 1查看技能
GitHub
splatoon3-meta
Domain expert for Splatoon 3 competitive gear abilities and weapon meta. Use when: (1) Labeling or classifying gear build features from SAE analysis, (2) Understanding why specific ability combinations appear i...
☆ 1查看技能
GitHub
mechinterp-investigator
Orchestrate a systematic research program to investigate and meaningfully label SAE features
☆ 1查看技能
GitHub
mechinterp-labeler
Manage feature labeling workflow - queue management, label storage, similar features, progress tracking
☆ 1查看技能
GitHub
mechinterp-overview
Quick "first look" overview of SAE features - top tokens, activation stats, weapons, families, sample contexts
☆ 1查看技能
GitHub
mechinterp-patterns
Reference catalog of recurring SAE feature patterns - check when labeling new features, add new patterns as discovered
☆ 1查看技能
GitHub
mechinterp-cluster-mapper
Analyze relationships across multiple SAE features - co-activation patterns, shared drivers, and subsystem identification
☆ 1查看技能
GitHub
mechinterp-runner
Execute mechanistic interpretability experiments from JSON specs - family sweeps, itemsets, interactions, minimal cores, validation
☆ 1查看技能
GitHub
mechinterp-crossmodel-matcher
Match SAE features between Ultra (24K) and Full (2K) models based on activation patterns and token overlap
☆ 1查看技能
GitHub
mechinterp-ability-semantics
Reference for ability family semantic groupings and gameplay meanings
☆ 1查看技能
默认按已收录数据中的 GitHub 仓库 Stars 排序。Stars 属于来源仓库，同一仓库的多个技能可能共享相同数值；它反映仓库关注度，不代表单个技能的安装量或质量评分。