TheBloke
/

Falcon-7B-Instruct-GPTQ

Text Generation

RefinedWebModel

text-generation-inference

4-bit precision

Model card Files Files and versions Community

Falcon-7B-Instruct-GPTQ / quantize_config.json

TheBloke's picture

Initial AutoGPTQ model commit.

ded9ab7 over 1 year ago

122 Bytes

	{
	"bits": 4,
	"group_size": 64,
	"damp_percent": 0.01,
	"desc_act": false,
	"sym": true,
	"true_sequential": true
	}