From f6e34d9621c0b4b2f3feb8b0f98cd39512b6f565 Mon Sep 17 00:00:00 2001 From: shao0099876 Date: Thu, 26 Oct 2023 14:38:58 +0000 Subject: [PATCH] =?UTF-8?q?=E4=BF=AE=E5=A4=8D=E4=BA=86=E4=B8=80=E4=B8=AA?= =?UTF-8?q?=E5=AF=BC=E8=87=B4=E6=97=A0=E6=B3=95=E5=8A=A0=E8=BD=BD=E6=9C=AA?= =?UTF-8?q?=E9=87=8F=E5=8C=96=E7=9A=84ChatGLM2=20fine-tuning=E6=A8=A1?= =?UTF-8?q?=E5=9E=8B=E7=9A=84=E9=97=AE=E9=A2=98(quantization=5Fbit=3D0)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- request_llm/bridge_chatglmft.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/request_llm/bridge_chatglmft.py b/request_llm/bridge_chatglmft.py index 71af942..4416382 100644 --- a/request_llm/bridge_chatglmft.py +++ b/request_llm/bridge_chatglmft.py @@ -87,7 +87,7 @@ class GetGLMFTHandle(Process): new_prefix_state_dict[k[len("transformer.prefix_encoder."):]] = v model.transformer.prefix_encoder.load_state_dict(new_prefix_state_dict) - if model_args['quantization_bit'] is not None: + if model_args['quantization_bit'] is not None and model_args['quantization_bit'] != 0: print(f"Quantized to {model_args['quantization_bit']} bit") model = model.quantize(model_args['quantization_bit']) model = model.cuda()