individual
AI safetyOther
| BlueDot Impact | $4,150 | 100% | 2 |
| Date ↓ | Funder | Via | Amount | Cause | Source | Purpose |
|---|---|---|---|---|---|---|
| Sep 2026 | BlueDot Impact | $1,850 | Other | bluedot | Presenting sparse-autoencoder steerability research at COLM | |
| Jun 2026 | BlueDot Impact | $2,300 | AI safety | bluedot | My paper shows that geometric properties partially predicts feature steerability. I’m also working on a different project where I’m using NLAs to see privacy leakage in LLMs. |