issues
search
argilla-io
/
notus
Notus is a collection of fine-tuned LLMs using SFT, DPO, SFT+DPO, and/or any other RLHF techniques, while always keeping a data-first approach
MIT License
161
stars
14
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
【question】are you planning to support multi language model?
#14
Yongtae723
closed
10 months ago
1
feat: add .ipynb for adding and updating models to `ollama`
#13
davidberenstein1957
opened
11 months ago
1
Run DPO step with multibinarized dataset
#12
dvsrepo
closed
11 months ago
1
Add Notus logo with Argilla watermark
#11
alvarobartt
closed
12 months ago
0
Warn about data contamination of UltraFeedback and add Notus naming explanation
#10
alvarobartt
closed
12 months ago
0
Major cleanup before release
#9
alvarobartt
closed
12 months ago
0
Adapt SFT fine-tuning script and configuration
#8
alvarobartt
closed
12 months ago
1
Curate UltraFeedack dataset's overall_score
#7
dvsrepo
opened
1 year ago
3
Add Intel Habana instructions and AlpacaEval results
#6
gabrielmbmb
closed
1 year ago
0
Run SFT step with the rating binarized data chosen responses
#5
dvsrepo
closed
12 months ago
1
Run `EleutherAI/lm-eval-harness` and upload results
#3
alvarobartt
closed
1 year ago
0
More experiments (full fine-tuning on 8 x A100 40GB)
#2
alvarobartt
closed
1 year ago
1
Evaluate using MT-Bench and AlpacaEval to compare with Zephyr 7B Beta
#1
alvarobartt
closed
1 year ago
0