Live data from Hacker News

Viewing profile — kashifr

kashifr

HN member
Joined
Fri, Mar 11, 2011, 12:34 PM UTC
HN karma
753
Public activity
47 items

About kashifr

https://github.com/kashif https://twitter.com/krasul

Recent public activity

  1. comment
    Comment #49222329

    Check out my pytorch reproduction of the paper here for those interested: https://github.com/NVIDIA/physicsnemo/pull/1660

  2. story
  3. story
  4. story
  5. comment
    Comment #47600653

    You can try out TinyLoRA in PEFT main now: https://huggingface.co/docs/peft/main/en/package_reference/t...

  6. story
  7. story
  8. story
  9. story
  10. story
  11. story
  12. comment
    Comment #43886016

    Have you tried https://geobase.app/ they recently also had a post about duckdb integration: https://geobase.app/blog/duckdb-1-1-3

  13. story
  14. story
  15. comment
    Comment #40644007

    A novel and memory-friendly preference alignment method for diffusion models that does not depend on any reference model.

  16. story
  17. comment
    Comment #39513674

    The dataset can be used for training preference models or aligning language models through techniques like Direct Preference Optimization (DPO).

  18. story
  19. story
  20. story
  21. story
  22. comment
    Comment #36060973

    An efficient finetuning approach that reduces memory usage enough to finetune a 65B parameter model on a single 48GB GPU while preserving full 16-bit finetuning task performance!

  23. story
  24. comment
    Comment #35545192

    The title is confusing but refers to the Huggingface Diffusers v0.15 release which brings new pipelines for video and audio to diffusers, showing that diffusion is a great choice f…

  25. comment