distinct / .gitignore
User1342's picture
Measure the GPU offload instead of estimating it, and keep the development scratch out of the repository
9f85661
Raw
History Blame
806 Bytes
__pycache__/
*.py[cod]
.env
.DS_Store
# Build output from `pip install -e .`. Regenerated on every install, and a
# tracked copy means every volunteer's install shows up as a diff.
*.egg-info/
# The llama.cpp build this worker fetched and verified for itself. It is
# hundreds of megabytes, it is different on every machine, and it is
# reproducible from llama_runtime.json, which IS tracked. A tracked copy would
# put somebody's CUDA runtime through Git LFS.
runtime/
# Scratch left behind by the apply-and-push helpers used during development.
APPLY*.bat
TIDY*.bat
BENCH*.bat
ITERATE*.bat
AGENT-SETUP.bat
push-log*.txt
iterate*.txt
distinct-update*.zip
agent-setup.log
runtime-outputs/
screenshots/
# The benchmark harness and what it leaves behind.
bench/
bench.zip
bench-result*.txt
cuda-pin.json