Tian Li

RicardoLee's activity

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

New activity in RicardoLee/Llama2-chat-13B-Chinese-50W over 2 years ago

13b的context len多大以及batch？

5

#1 opened over 2 years ago by

lucasjin

New activity in RicardoLee/Llama2-chat-Chinese-50W over 2 years ago

此时不应降低学习率，warmup 等超参，而是应该放大到Pretrain 规模

3

#2 opened over 2 years ago by

daner

关于train_sft.py中coati包

2

#3 opened over 2 years ago by

BatmanBill

那这个怎么调用呢

4

#1 opened over 2 years ago by

yjianchun

那这个怎么调用呢

4

#1 opened over 2 years ago by

yjianchun