fbai's picture

28 12

fbai

fb700

·

AI & ML interests

None yet

Organizations

fb700's activity

New activity in fb700/chatglm-fitness-RLHF 11 months ago

Lora和RLHF训练的代码开源了吗

#14 opened over 1 year ago by

New activity in fb700/chatglm-fitness-RLHF 12 months ago

大佬有没有兴趣再搞一搞llama3-8b

#24 opened 12 months ago by

New activity in fb700/chatglm-fitness-RLHF about 1 year ago

请问训练所用的数据集能否公开？

#21 opened about 1 year ago by

New activity in fb700/chatglm-fitness-RLHF over 1 year ago

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

#20 opened over 1 year ago by

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

#18 opened over 1 year ago by

rlhf

#19 opened over 1 year ago by

关于两种加载模型文件方式的区别

#17 opened over 1 year ago by

如何在多gpu上加载

#16 opened over 1 year ago by

数据

#13 opened over 1 year ago by

关于模型权重

#15 opened over 1 year ago by

New activity in fb700/Bofan-chatglm-Best-lora over 1 year ago

冒昧的问一句，能否开源微调和PPO的实现代码吗？

#1 opened over 1 year ago by

New activity in fb700/chatglm-fitness-RLHF over 1 year ago

最大token

#12 opened over 1 year ago by

会不会基于ChatGLM2-6B进行迭代？

#3 opened over 1 year ago by

api 调用您的模型出现错误

#10 opened over 1 year ago by

还可以再次进行自我认知的lora的训练吗

#11 opened over 1 year ago by

牛的无限长的原理和chatglm2是一个道理吗？

#9 opened over 1 year ago by

测试了一下，很好用，比chatglm2还好用。期待更多作品。

#8 opened over 1 year ago by

启动加载很慢，需要130秒

#5 opened over 1 year ago by

训练代码

#6 opened over 1 year ago by

chatglm2-6b-32k

#7 opened over 1 year ago by