We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
INSTRUCTOR同时具有数量级更少的参数,最先进的性能,平均与以前相比提高了3.4% 在70个不同的数据集上获得最佳结果。我们的分析表明,INSTRUCTOR对指令的更改具有鲁棒性,并且该指令微调减轻了在不同的数据集上训练单个模型。这种Instruction在多任务训练的方法也跟现在的LLM的sft训练相契合,后续具有很大的潜力。
The text was updated successfully, but these errors were encountered:
qiuwenbogdut 已报名
Sorry, something went wrong.
wawltor
No branches or pull requests
背景
INSTRUCTOR同时具有数量级更少的参数,最先进的性能,平均与以前相比提高了3.4%
在70个不同的数据集上获得最佳结果。我们的分析表明,INSTRUCTOR对指令的更改具有鲁棒性,并且该指令微调减轻了在不同的数据集上训练单个模型。这种Instruction在多任务训练的方法也跟现在的LLM的sft训练相契合,后续具有很大的潜力。
解决步骤:
数据集:
medi-data,参考官方readme。
The text was updated successfully, but these errors were encountered: