Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/aflah02/aflah02
https://github.com/aflah02/aflah02
Last synced: about 5 hours ago
JSON representation
- Host: GitHub
- URL: https://github.com/aflah02/aflah02
- Owner: aflah02
- Created: 2021-02-15T11:51:33.000Z (almost 4 years ago)
- Default Branch: main
- Last Pushed: 2024-11-01T13:23:18.000Z (19 days ago)
- Last Synced: 2024-11-01T14:24:16.668Z (19 days ago)
- Size: 140 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Hi, I'm Aflah, a research software engineer at MPI-SWS. I specialize in backend engineering, NLP, and deep learning, with a keen interest in applying NLP for social good. I am also interested in high speed inference, large scale distributed training and squeezing out performance on limited hardware. Currently, my research involves evaluating latent knowledge and memorization in large language models. I have previously worked on hate speech detection, and leveraging NLP to address various societal challenges as a part of undegrad thesis and related research work.
**Open to researcher/research engineer/backend engineer roles**
---
### Latest Youtube Videos πΊβΆοΈ
- [Paper Reading Discussion: Lookback Lens: Detecting and Mitigating Contextual Hallucinations](https://www.youtube.com/watch?v=Qsce8SeaDhQ)
- [Adapters | Exploring VGLM Via Parameter-Efficient Transfer Learning | PEFT Methods](https://www.youtube.com/watch?v=mE9Mmw0HfsI)
- [Adapters | Simple, Scalable Adaptation for Neural Machine Translation | PEFT Methods](https://www.youtube.com/watch?v=kVJ7jfp5jEU)
- [Adapters | Parameter-Efficient Transfer Learning for NLP | PEFT Methods](https://www.youtube.com/watch?v=UAGbRcgDf40)
- [Trying to Learn Triton | P-4 (Not a Tutorial)](https://www.youtube.com/watch?v=FSL-H0tjIig)β‘οΈ [more videos...](https://www.youtube.com/channel/UCwab-Xf38Sd7QsxVPoS0cgA)
---
### Languages & Tools:
---
### Some Stats -
---
### Connect with meπ: