Interpretability and mechanistic analysis · MLD-0676
Logit lens
Projecting intermediate representations into output-token space to inspect provisional predictions.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
View layer-by-layer token preferences.
Original caveat
Early-layer representations may not be calibrated for the final output head.
Context and aliases
Level: Advanced. Stages: Research · Evaluation.
Tags: ML · DL · SLM · LLM · Interpretability · Safety.
Aliases: None recorded.
Related concepts
Original references
- Captum model interpretability documentation ↗ · Reference link
- Tracing the thoughts of a large language model ↗ · Consulted
- Towards Monosemanticity: Decomposing Language Models With Dictionary Learning ↗ · Consulted
- SHAP documentation ↗ · Reference link
- Original reference URL ↗