WCargo

Karma: 170

(Léo Dana) French master student in applied Mathematics (probability & statistic), PhD in Deep Learning Theory in Paris

Try Training SAEs with RLAIF

WCargo5 Dec 2025 1:10 UTC

5 points

0 comments2 min readLW link

Visualizing small Attention-only Transformers

WCargo19 Nov 2024 9:37 UTC

4 points

0 comments8 min readLW link

Results from the Turing Seminar hackathon

Charbel-Raphaël, jeanne_ and WCargo

7 Dec 2023 14:50 UTC

35 points

1 comment5 min readLW link

On Interpretability’s Robustness

WCargo18 Oct 2023 13:18 UTC

11 points

0 comments4 min readLW link

Introducing EffiSciences’ AI Safety Unit

WCargo, Charbel-Raphaël and Florent_Berthet

30 Jun 2023 7:44 UTC

68 points

0 comments12 min readLW link

Improvement on MIRI’s Corrigibility

WCargo and Charbel-Raphaël

9 Jun 2023 16:10 UTC

54 points

8 comments13 min readLW link

A Corrigibility Metaphore—Big Gambles

WCargo10 May 2023 18:13 UTC

16 points

0 comments4 min readLW link