A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO and Toxicity
Paper
•
2401.01967
•
Published
Itamar Pres
ipres
AI & ML interests
None yet
Recent Activity
authored
a paper
about 1 month ago
A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO
and Toxicity
authored
a paper
about 1 month ago
Why Can't Transformers Learn Multiplication? Reverse-Engineering Reveals
Long-Range Dependency Pitfalls
upvoted
a
paper
about 1 month ago
Why Can't Transformers Learn Multiplication? Reverse-Engineering Reveals
Long-Range Dependency Pitfalls
Organizations
None yet
authored
2
papers
about 1 month ago