wuzhiying2023's picture

21

wuzhiying2023

wuzhiying2023

·

AI & ML interests

None yet

Organizations

wuzhiying2023's activity

New activity in baichuan-inc/Baichuan2-13B-Chat-4bits 7 months ago

简单测试，报错Can't load the configuration

#4 opened 7 months ago by

New activity in baichuan-inc/Baichuan2-13B-Chat-4bits 8 months ago

Error | No such file or directory

#1 opened 8 months ago by

下载了所有文件但报错“Needs import model weight init func to run quantize.”

#2 opened 8 months ago by

New activity in baichuan-inc/Baichuan2-13B-Chat 8 months ago

使用fastchat加载百川2，性能与魔搭的demo差距非常大

#10 opened 9 months ago by

New activity in baichuan-inc/Baichuan2-13B-Chat 9 months ago

Update modeling_baichuan.py

#6 opened 9 months ago by

load_in_8bit error

#2 opened 9 months ago by

New activity in baichuan-inc/Baichuan2-7B-Chat 9 months ago

NormHead 中的分支判断

#3 opened 9 months ago by

fix 'BitsAndBytesConfig' object is not subscriptable issue

#2 opened 9 months ago by

New activity in baichuan-inc/Baichuan-13B-Chat 9 months ago

13B的8bit量化版的下载链接咋没了呢

#35 opened 9 months ago by

Model Export to ONNX format

#32 opened 9 months ago by

英文推理报错

#26 opened 10 months ago by

New activity in baichuan-inc/Baichuan-13B-Chat 10 months ago

有支持FasterTransformer的计划吗？

#27 opened 10 months ago by

Add flash attention

#21 opened 10 months ago by

请问, 模型量化后如何保存,直接使用 save_pretrained 保存下来的不太正确只有1.3G

#22 opened 10 months ago by

Any guideline on how to quantatize to GPTQ 4bits?

#20 opened 10 months ago by

3060 12g demo代码可以跑起来 4090 24g内存溢出

#12 opened 10 months ago by

web_demo启动错误

#3 opened 11 months ago by

jfy

123

#1 opened 11 months ago by

求尽快更新int-4版本

#5 opened 11 months ago by

是否支持DCU？

#14 opened 10 months ago by

New activity in baichuan-inc/Baichuan-13B-Base 10 months ago

Take input attention masks to support left-padded sequences

#1 opened 10 months ago by