ggml-org/llama.cpp

llama : refactor the llm.build_xxx functions

Fermée

#5 239 ouverte le 31 janv. 2024

 (3 commentaires) (3 réactions) (0 personne assignée)C++ (21 737 forks)batch import
good first issuerefactoringroadmap

Métriques du dépôt

Stars
 (124 043 étoiles)
Métriques de merge PR
 (Merge moyen 6j 8h) (389 PRs mergées en 30 j)

Description

Now that we support a large amount of architectures, we can clearly see the patterns when constructing the compute graphs - i.e. optional biases, different norm types, QKV vs Q+K+V, etc.

We should deduplicate the copy-paste portions in functions such as llm.build_llama(), llm.build_falcon(), etc.

The advantage of the current code is that it is easy to look into the graph of a specific architecture. When we refactor this, we will lose this convenience to some extend. So we should think about making this refactoring in such a way that we don't completely obscure which parts of the graph belong to which architectures

Open for ideas and suggestions how to do this best

Guide contributeur