Xenova
/

LaMini-Neo-125M

Text Generation

Transformers.js

Model card Files Files and versions Community

LaMini-Neo-125M / quantize_config.json

Xenova's picture

Xenova HF staff

Upload folder using huggingface_hub

2a56393 10 months ago

raw history blame contribute delete

No virus

2.15 kB

	{
	"per_channel": true,
	"reduce_range": true,
	"per_model_config": {
	"decoder_model": {
	"op_types": [
	"Sqrt",
	"Softmax",
	"Pow",
	"Transpose",
	"Tanh",
	"Where",
	"Shape",
	"Range",
	"Squeeze",
	"Div",
	"Constant",
	"Unsqueeze",
	"Concat",
	"Reshape",
	"Mul",
	"Add",
	"Slice",
	"ReduceMean",
	"Gather",
	"Sub",
	"MatMul",
	"Cast"
	],
	"weight_type": "QInt8"
	},
	"decoder_model_merged": {
	"op_types": [
	"Sqrt",
	"Softmax",
	"Pow",
	"Transpose",
	"Tanh",
	"Where",
	"Shape",
	"Range",
	"Squeeze",
	"Div",
	"Constant",
	"Unsqueeze",
	"Concat",
	"Reshape",
	"Mul",
	"Add",
	"Slice",
	"If",
	"ReduceMean",
	"Gather",
	"Sub",
	"MatMul",
	"Cast"
	],
	"weight_type": "QInt8"
	},
	"decoder_with_past_model": {
	"op_types": [
	"Sqrt",
	"Softmax",
	"Pow",
	"Transpose",
	"Tanh",
	"Where",
	"Shape",
	"Range",
	"Squeeze",
	"Div",
	"Constant",
	"Unsqueeze",
	"Concat",
	"Reshape",
	"Mul",
	"Add",
	"Slice",
	"ReduceMean",
	"Gather",
	"Sub",
	"MatMul",
	"Cast"
	],
	"weight_type": "QInt8"
	}
	}
	}