Cautious Optimizers: Improving Training with One Line of Code
Paper
ā¢
2411.16085
ā¢
Published
ā¢
19
ā¢
2
Kaizhao Liang PRO
kz919
AI & ML interests
Search = AGI?
Recent Activity
liked
a dataset
5 days ago
laion/laion400m
liked
a Space
11 days ago
lllyasviel/IC-Light
liked
a dataset
13 days ago
timm/imagenet-1k-wds
Organizations
kz919's activity
commented
a
paper
3 months ago
Why does this one have different tokenizer from the 32B model?
1
#1 opened 4 months ago
by
kz919
<|begin_of_text|> is added twice by the preprocessor
1
#44 opened 5 months ago
by
kz919
Added history and Better UI
1
#7 opened 6 months ago
by
KingNish
Added history and Better UI
3
#6 opened 6 months ago
by
KingNish
use gr chatbot
#5 opened 6 months ago
by
akhaliq
downgrade openai version
#4 opened 6 months ago
by
akhaliq
fix gradio demo issue and not use chatbot component
#3 opened 6 months ago
by
akhaliq
update for gradio
#2 opened 6 months ago
by
akhaliq
use gradio
#1 opened 6 months ago
by
akhaliq
The persona and quote boxes are editable. Please bring your own favorite waifu!
#2 opened 6 months ago
by
kz919
Please, 8B version
6
#8 opened 6 months ago
by
Anderson452
Huggingface Space Infra broken
1
#1 opened 6 months ago
by
kz919
Referral link for API key access! It's available now!
#1 opened 6 months ago
by
kz919
Is the instruction tuned version going to be released?
#1 opened 6 months ago
by
kz919
commented
2
papers
7 months ago
Text conditioning
1
#57 opened about 1 year ago
by
kz919
The INST problem seems to be replaced by source:
1
#1 opened about 1 year ago
by
kz919
The model is not performing well.
25
#1 opened about 1 year ago
by
Liangmingxin