How was the quantization performed - do you have a recipe?
#1
by
molereddy
- opened
Do you have a recipe like, for example, in https://huggingface.co/neuralmagic/Meta-Llama-3.1-70B-Instruct-FP8-dynamic
Do you have a recipe like, for example, in https://huggingface.co/neuralmagic/Meta-Llama-3.1-70B-Instruct-FP8-dynamic