llama.cpp
Q4_2 quantization with rmse-optimized scale and quants
#1062

Merged

Commits

Q4_2 quantization with rmse-optimized scale and quants

Iwan Kawrakow committed 3 years ago
ggml : satisfy the sanitizer builds

ggerganov committed 3 years ago
Better follow ggml conventions for function names

Iwan Kawrakow committed 3 years ago
Fixed type as per reviewer comment

Iwan Kawrakow committed 3 years ago