nm-testing
/

tinyllama-w8a16-dense

compressed-tensors

Model card Files Files and versions Community

tinyllama-w8a16-dense / recipe.yaml

sadkins65's picture

Upload folder using huggingface_hub

b1dcbd8 verified 4 months ago

192 Bytes

	quant_stage:
	quant_modifiers:
	GPTQModifier:
	ignore: [lm_head]
	config_groups:
	group_0:
	weights: {num_bits: 8, strategy: channel}
	targets: [Linear]