The Fine-Tuning Index / PEFT & LoRA / #37
DaoyuanLi2816/can-i-finetune-this
by DaoyuanLi2816 · PEFT & LoRA · updated 19d ago
Single-GPU LLM fine-tuning preflight: memory estimates, runnable LoRA/QLoRA recipes, and local measurements.
61
momentum
794
stars
107
forks
#37
rank
bitsandbytesfine-tuninggpuhugging-facellmloramemory-estimationpeftpytorchqloratransformersvram
View on GitHub →