Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / hit-scir/chinese-mixtral-8x7b issues and pull requests
#31 - TypeError: SFTConfig.__init__() got an unexpected keyword argument 'cache_dir'
Issue -
State: closed - Opened by wormcode about 1 month ago
- 1 comment
#30 - 请问 SFT 训练的时候 训练 数据的格式 是怎么样的?
Issue -
State: closed - Opened by wormcode about 2 months ago
- 3 comments
#29 - 关于词表扩充
Issue -
State: closed - Opened by CrazyBoyM 7 months ago
- 2 comments
#28 - 增量预训练出错
Issue -
State: closed - Opened by lwj2001 7 months ago
- 8 comments
#27 - 微调
Issue -
State: closed - Opened by cripsgreen 7 months ago
- 1 comment
#26 - 咨询一下,mixtral-8x7B在未增量训中文的情况下,ceval、cmmlu的得分是多少呢?在进行42B token续训之后分数有明显增长么?
Issue -
State: closed - Opened by KalsaHT 8 months ago
#25 - 微调的--mode 错了吧
Issue -
State: closed - Opened by flyingwaters 8 months ago
- 2 comments
#24 - 请问微调需要的最小的GPU是多少?
Issue -
State: closed - Opened by chaishenAI 8 months ago
- 1 comment
#23 - torchrun: command not found
Issue -
State: closed - Opened by lfxuan 9 months ago
- 1 comment
#22 - 关于embedding扩展的细节
Issue -
State: closed - Opened by cometyang 9 months ago
- 1 comment
#21 - 通信量不一致
Issue -
State: closed - Opened by longzhang418 9 months ago
- 1 comment
#20 - ALP横坐标意义
Issue -
State: closed - Opened by Jianwei-Lv 9 months ago
- 1 comment
#19 - 没有tokenizer文件
Issue -
State: closed - Opened by cometyang 9 months ago
- 3 comments
#18 - 询问一下data.utils套件要在哪裡下载?
Issue -
State: closed - Opened by DanielChen1128 9 months ago
- 1 comment
#17 - 请问下MMLU评测的时候有增加什么prompt吗?
Issue -
State: closed - Opened by matrixssy 9 months ago
- 3 comments
#16 - 能否说下硬件需求。
Issue -
State: closed - Opened by orderer0001 9 months ago
- 3 comments
#15 - Support instruction tuning
Pull Request -
State: closed - Opened by jubgjf 10 months ago
#14 - init_embeddings 模型转换问题
Issue -
State: closed - Opened by dachengai 10 months ago
- 1 comment
#13 - 请教一下预训练数据使用情况
Issue -
State: closed - Opened by CLeafYeah 10 months ago
- 4 comments
#12 - 请问训练一次要多长时间?
Issue -
State: closed - Opened by gongbudaizhe 10 months ago
- 1 comment
#11 - 有没有考虑过将模型文件上传到modelscope
Issue -
State: closed - Opened by Tendo33 10 months ago
- 2 comments
#10 - 代码中缺失了调用init_embeddings.py的脚本,纯新手,能提供一下吗?
Issue -
State: closed - Opened by whosyourdadds 10 months ago
- 2 comments
#9 - Mixtral
Issue -
State: closed - Opened by 591094733 10 months ago
#9 - Mixtral
Issue -
State: closed - Opened by 591094733 10 months ago
#8 - autoawq量化后推理很奇怪
Issue -
State: open - Opened by gptbert 10 months ago
#8 - autoawq量化后推理很奇怪
Issue -
State: closed - Opened by gptbert 10 months ago
- 2 comments
#7 - 这么好的项目,建议作者持续更新!
Issue -
State: closed - Opened by whosyourdadds 10 months ago
- 1 comment
#7 - 这么好的项目,建议作者持续更新!
Issue -
State: open - Opened by whosyourdadds 10 months ago
#6 - 4bit量化推理速度变慢
Issue -
State: closed - Opened by snowlixue 10 months ago
- 4 comments
#5 - 词表扩充数据:12G知乎数据和2G悟道数据上训练中文BPE词表
Issue -
State: closed - Opened by LiuChaoXD 10 months ago
- 2 comments
#5 - 词表扩充数据:12G知乎数据和2G悟道数据上训练中文BPE词表
Issue -
State: closed - Opened by LiuChaoXD 10 months ago
- 2 comments
#4 - 您好,我这里使用4bit量化后get_peft_model报错
Issue -
State: closed - Opened by Ahu-Fgx 10 months ago
- 2 comments
#3 - 您好,请问指令微调需要多少显存,有没有可以推荐的指令微调数据集?
Issue -
State: closed - Opened by tommy3266 10 months ago
- 1 comment
#2 - 希望能量化成q4或者q6的形式,并尽快引入Ollama,这样个人也能用了,感谢!
Issue -
State: closed - Opened by ricksuzade-maker 10 months ago
- 3 comments
#1 - 请问,训练需要至少多大的显存?
Issue -
State: closed - Opened by LiuChaoXD 10 months ago
- 2 comments