Wezenite
/

gemma-2b-trader-8bit

gemma-2b-trader-8bit / quantization_config.json

Upload folder using huggingface_hub

e3b5b8d verified 11 months ago

109 Bytes

{
  "quantization_method": "bitsandbytes",
  "bits": 8,
  "load_in_8bit": true,
  "llm_int8_threshold": 6.0
}

	{
	"quantization_method": "bitsandbytes",
	"bits": 8,
	"load_in_8bit": true,
	"llm_int8_threshold": 6.0
	}