submarat

Karma: 51

SWE/MLE/AI Safety

Transformers Don’t Need LayerNorm at Inference Time: Implications for Interpretability

submarat, Joachim Schaeffer, Luca Baroni, galvsk and StefanHex

23 Jul 2025 14:55 UTC

31 points

0 comments7 min readLW link

ARENA4.0 Capstone: Hyperparameter tuning for MELBO + replication on Llama-3.2-1b-Instruct

25Hour and submarat

5 Oct 2024 11:30 UTC

34 points

2 comments8 min readLW link