From f90ea4aca5d128413bcd21ff1124890caceced76 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Iv=C3=A1n=20Baldo?= Date: Mon, 11 Dec 2023 17:34:38 -0300 Subject: [PATCH] Fix hf/bench-gptq.py. --- hf/bench-gptq.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/hf/bench-gptq.py b/hf/bench-gptq.py index 1e34290..0b075be 100644 --- a/hf/bench-gptq.py +++ b/hf/bench-gptq.py @@ -10,14 +10,14 @@ from auto_gptq import AutoGPTQForCausalLM model_name_or_path = "TheBloke/Llama-2-7B-GPTQ" -model_basename = "gptq_model-4bit-128g" +model_basename = "gptq-4bit-128g-actorder_True" use_triton = False tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True) model = AutoGPTQForCausalLM.from_quantized(model_name_or_path, - model_basename=model_basename, + revision=model_basename, use_safetensors=True, trust_remote_code=True, device="cuda:0",