CarperAI/trlx

Collapse reference+learner hydra heads when using LoRa

Open

#320 ouverte le 21 févr. 2023

Voir sur GitHub
 (6 commentaires) (2 réactions) (0 assignés)Python (450 forks)batch import
contributions welcomefeature requestgood first issue

Métriques du dépôt

Stars
 (4 184 stars)
Métriques de merge PR
 (Aucune PR mergée en 30 j)

Description

🚀 The feature, motivation, and pitch

With additive (delta-style) parameter-efficient tuning methods such as LoRa, we should be able to make a slightly more mem-efficient hydra architecture by using a single block that does ~frozen_head + tunable_weights for the learner/policy head's fwd-pass and simply frozen_head for the reference, instead of maintaining 2x heads.

CC @LouisCastricato and @cat-state for pointing this out

Alternatives

No response

Additional context

No response

Guide contributeur