fbai

fb700's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in Mia-AiLab/Qwable-3.6-27b-MTP 2 months ago

no support llama cpp and lm studio

👍 1

4

#2 opened 2 months ago by

xueqing12

Multimodal support?

👍 1

2

#4 opened 2 months ago by

tuanlai001

New activity in Jackrong/Qwopus3.6-27B-Coder-MTP-GGUF 2 months ago

IQ4_NL support

👍 1

2

#5 opened 2 months ago by

frandmb

需要llama.cpp的编译方法和启动参数

🤯 2

4

#7 opened 2 months ago by

traddo

New activity in fb700/chatglm-fitness-RLHF over 2 years ago

Lora和RLHF训练的代码开源了吗

9

#14 opened almost 3 years ago by

tenghg

大佬有没有兴趣再搞一搞llama3-8b

1

#24 opened over 2 years ago by

AlexLee01

请问训练所用的数据集能否公开？

1

#21 opened over 2 years ago by

sssssimeng

请问作者，rlhf的actor loss是否下降和正常收敛呢？能不能给一些经验的超参数设置？请教了

4

#20 opened over 2 years ago by

hepansls

New activity in fb700/chatglm-fitness-RLHF almost 3 years ago

Cannot copy out of meta tensor; no data!，报错代码地方为model = model.half().quantize(4).cuda() ，猜测是量化相关问题或者作者的模型上传的时候有遗漏的文件

4

#18 opened almost 3 years ago by

yoma0101

rlhf

1

#19 opened almost 3 years ago by

nlpx

关于两种加载模型文件方式的区别

1

#17 opened almost 3 years ago by

rk686

如何在多gpu上加载

1

#16 opened almost 3 years ago by

jersonal

数据

4

#13 opened almost 3 years ago by

yyshao

关于模型权重

4

#15 opened almost 3 years ago by

mzhe

New activity in fb700/Bofan-chatglm-Best-lora almost 3 years ago

冒昧的问一句，能否开源微调和PPO的实现代码吗？

5

#1 opened almost 3 years ago by

xiao111

New activity in fb700/chatglm-fitness-RLHF about 3 years ago

最大token

1

#12 opened about 3 years ago by

goodboys

会不会基于ChatGLM2-6B进行迭代？

3

#3 opened about 3 years ago by

neteasy

api 调用您的模型出现错误

👍 1

5

#10 opened about 3 years ago by

neteasy

还可以再次进行自我认知的lora的训练吗

2

#11 opened about 3 years ago by

goodboys

牛的 无限长的原理和chatglm2是一个道理吗？

3

#9 opened about 3 years ago by

szu2018chenli