Decoding and inference behavior · MLD-0938
Presence penalty
A decoding penalty based on whether a token has appeared, regardless of frequency.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Encourage introducing different vocabulary.
Original caveat
It is not a guarantee of topical diversity.
Context and aliases
Level: Intermediate. Stages: Inference · Evaluation · Deployment.
Tags: SLM · LLM · NLP.
Aliases: None recorded.
Related concepts
Concepts linking here
Original references
- Hugging Face: generation strategies ↗ · Reference link
- Fast Inference from Transformers via Speculative Decoding ↗ · Consulted
- vLLM documentation ↗ · Consulted
- Original reference URL ↗