shibing624 / textgen

TextGen: Implementation of Text Generation models, include LLaMA, BLOOM, GPT2, BART, T5, SongNet and so on. 文本生成模型,实现了包括LLaMA,ChatGLM,BLOOM,GPT2,Seq2Seq,BART,T5,UDA等模型的训练和预测,开箱即用。
Apache License 2.0
935 stars 109 forks source link

NotImplementedError: Cannot copy out of meta tensor; no data! #36

Closed LMXKO closed 1 year ago

LMXKO commented 1 year ago

在SFT的模型训练中,报错如下: image

shibing624 commented 1 year ago

哪个模型,哪个数据,哪个调用脚本出错?

LMXKO commented 1 year ago

模型:chinese-llama-plus-7b-merged 数据:medical 代码块: image

LMXKO commented 1 year ago

哪个模型,哪个数据,哪个调用脚本出错?

好像找到了,跟这个一样 https://github.com/THUDM/ChatGLM-6B/issues/530