nm-testing
/

Mixtral-8x7B-Instruct-v0.1-W8A16-quantized

compressed-tensors

Model card Files Files and versions Community

Mixtral-8x7B-Instruct-v0.1-W8A16-quantized / recipe.yaml

sadkins65's picture

Upload folder using huggingface_hub

d0e9c07 verified 3 months ago

history blame contribute delete

181 Bytes

	DEFAULT_stage:
	DEFAULT_modifiers:
	GPTQModifier:
	sequential_update: true
	targets: Linear
	ignore: [lm_head, 're:.*block_sparse_moe\.gate']
	scheme: W8A16