Imran1
/

Llama-3.1-Tulu-3-70B-Fp8

compressed-tensors

Model card Files Files and versions Community

Llama-3.1-Tulu-3-70B-Fp8 / recipe.yaml

Imran1's picture

Upload folder using huggingface_hub

f8a8167 verified 2 months ago

history blame contribute delete

134 Bytes

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