gmonair
/

deepseek-math-7b-rl-W8A16

compressed-tensors

Model card Files Files and versions Community

deepseek-math-7b-rl-W8A16 / recipe.yaml

gmonair's picture

Upload folder using huggingface_hub

3c70b2d verified 16 days ago

171 Bytes

	DEFAULT_stage:
	DEFAULT_modifiers:
	SmoothQuantModifier: {smoothing_strength: 0.8}
	GPTQModifier:
	targets: Linear
	ignore: [lm_head]
	scheme: W8A16