Llama-3.1-8B-Instruct-openvino-4bit / openvino_config.json
AIFunOver's picture
Upload openvino_config.json with huggingface_hub
d14910d verified
raw
history blame
559 Bytes
{
"compression": null,
"dtype": "int4",
"input_info": null,
"optimum_version": "1.23.3",
"quantization_config": {
"all_layers": null,
"bits": 4,
"dataset": "wikitext2",
"gptq": null,
"group_size": 64,
"ignored_scope": null,
"num_samples": 20,
"quant_method": "awq",
"ratio": 1.0,
"scale_estimation": true,
"sensitivity_metric": null,
"sym": false,
"tokenizer": null,
"trust_remote_code": false,
"weight_format": "int4"
},
"save_onnx_model": false,
"transformers_version": "4.45.2"
}