GaLore: Advancing Large Model Training on Consumer-grade Hardware
•
10
passthrough
method of mergekit
but without using additional memory and attaching LoRAs to it, refer to the details below! 🔥https://lnkd.in/ge95ztjAreplace_lora_weights_loftq
for LoftQ to use it on the fly with bnb.llama.cpp
https://github.com/ggerganov/llama.cpp/pull/5795!cd llama.cpp
python convert-hf-to-gguf.py ../starcoder2-3b/ --outfile models/starcoder2-3b.gguf --outtype "f16"
./quantize models/starcoder2-3b.gguf models/starcoder2-3b-Q4_K_M.gguf Q4_K_M