nm-testing
/

int8_dynamic_per_token-e2e

8-bit precision

compressed-tensors

Model card Files Files and versions

int8_dynamic_per_token-e2e / recipe.yaml

nm-autobot's picture

Upload folder using huggingface_hub

e9a3064 verified 29 days ago

history blame contribute delete

191 Bytes

	default_stage:
	default_modifiers:
	QuantizationModifier:
	targets: [Linear]
	ignore: [lm_head, 're:.mlp.gate[.].']
	scheme: W8A8
	bypass_divisibility_checks: false