MixtralKit

MixtralKit copied to clipboard

请问eval_mixtral.py这个文件能否提供下？

jerryli1981

eval_mixtral怎么来的

3

── playground │ └── eval_mixtral.py 没搞懂这个eval_mixtral.py怎么来的

Anorid

When I was asking Mixtral to compose some video scripts (around 100 scripts) according to a specific structure, with the settings max tokens 1024 and Mixtral can output 7 scripts...

zhuyinzhuyin

(mixtral) root@autodl-container-582647ab6b-a5b6a91c:~/autodl-tmp/mixtral-8x7b-32kseqlen# magnet:?xt=urn:btih:5546272da9065eddeb6fcd7ffddeef5b75be79a7&dn=mixtral-8x7b-32kseqlen&tr=udp%3A%2F%http://2Fopentracker.i2p.rocks%3A6969%2Fannounce&tr=http%3A%2F%http://2Ftracker.openbittorrent.com%3A80%2Fannounce [1] 11681 [2] 11682 [3] 11683 -bash: magnet:?xt=urn:btih:5546272da9065eddeb6fcd7ffddeef5b75be79a7: command not found 他一直报错

Anorid

你好，能否直接使用huggingface拆分的mixtral-8x7b权重进行评估呢

1

shed-e

Hi, thanks for the great work. I am trying to run the inference evaluation, however, the guide mentions: ``` cd opencompass/ # link the example config into opencompass ln -s...

YJHMITWEB

the evaluation error for the eval_mixtral.py

3

I do not find the eval_mixtral.py file during the inference

runzeer

几点疑问或建议

3

1，利用cpp进行加速，如llama.cpp那样，你们也可以搞个mixtral.cpp，支持mixtral-8x7b和mixtral-7b在f32，f16等精度上的灵活切换 2，全参数的训练、提示学习微调代码，及其对应的数据json格式

shuaidaming

I noticed that the performance for the math reasoning is lower than the official blog. Is it due to the zero-shot setting compared the official 5-shot setting?

runzeer

A-Salty-Fish