mrhendrey
/

X-ALMA-13B-Pretrain-FP8-dynamic

compressed-tensors

Model card Files Files and versions Community

X-ALMA-13B-Pretrain-FP8-dynamic / recipe.yaml

mrhendrey's picture

Adding in fp8 quantized model and config files

606b6cb 2 days ago

history blame contribute delete

136 Bytes

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