jsbaicenter
/

Meta-Llama-3-8B-Instruct-FP8

Text Generation

compressed-tensors

Model card Files Files and versions

Meta-Llama-3-8B-Instruct-FP8 / recipe.yaml

uyiosa's picture

Upload local model folder

7164031 verified 6 months ago

history blame contribute delete

136 Bytes

	DEFAULT_stage:
	DEFAULT_modifiers:
	QuantizationModifier:
	ignore: [lm_head]
	targets: [Linear]
	scheme: FP8_DYNAMIC