Accelerating Production LLMs with Combined Token/Embedding Speculators
This technical report describes the design and training of novel speculative decoding draft models, for accelerating the inference speeds of large language models in a production environment. By conditioning draft predictions on both context vectors and sampled tokens, we can train our speculators t...
Gespeichert in:
Hauptverfasser: | , , , , , , |
---|---|
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | This technical report describes the design and training of novel speculative
decoding draft models, for accelerating the inference speeds of large language
models in a production environment. By conditioning draft predictions on both
context vectors and sampled tokens, we can train our speculators to efficiently
predict high-quality n-grams, which the base model then accepts or rejects.
This allows us to effectively predict multiple tokens per inference forward
pass, accelerating wall-clock inference speeds of highly optimized base model
implementations by a factor of 2-3x. We explore these initial results and
describe next steps for further improvements. |
---|---|
DOI: | 10.48550/arxiv.2404.19124 |