Arkhiveus
/

DeepHermes-3-Llama-3-8B-Preview-W8A8-int8

8-bit precision

compressed-tensors

Model card Files Files and versions Community

DeepHermes-3-Llama-3-8B-Preview-W8A8-int8 / recipe.yaml

Arkhiveus's picture

Upload folder using huggingface_hub

9ec56d7 verified 8 days ago

170 Bytes

	DEFAULT_stage:
	DEFAULT_modifiers:
	SmoothQuantModifier: {smoothing_strength: 0.8}
	GPTQModifier:
	scheme: W8A8
	targets: Linear
	ignore: [lm_head]