Deniz Bayazit
@bayazitdeniz.bsky.social
#NLProc PhD student @EPFL
#interpretability
#interpretability
Reposted by Deniz Bayazit
🚀 Excited to share a major update to our “Mixture of Cognitive Reasoners” (MiCRo) paper!
We ask: What benefits can we unlock by designing language models whose inner structure mirrors the brain’s functional specialization?
More below 🧠👇
cognitive-reasoners.epfl.ch
We ask: What benefits can we unlock by designing language models whose inner structure mirrors the brain’s functional specialization?
More below 🧠👇
cognitive-reasoners.epfl.ch
October 20, 2025 at 12:10 PM
🚀 Excited to share a major update to our “Mixture of Cognitive Reasoners” (MiCRo) paper!
We ask: What benefits can we unlock by designing language models whose inner structure mirrors the brain’s functional specialization?
More below 🧠👇
cognitive-reasoners.epfl.ch
We ask: What benefits can we unlock by designing language models whose inner structure mirrors the brain’s functional specialization?
More below 🧠👇
cognitive-reasoners.epfl.ch
1/🚨 New preprint
How do #LLMs’ inner features change as they train? Using #crosscoders + a new causal metric, we map when features appear, strengthen, or fade across checkpoints—opening a new lens on training dynamics beyond loss curves & benchmarks.
#interpretability
How do #LLMs’ inner features change as they train? Using #crosscoders + a new causal metric, we map when features appear, strengthen, or fade across checkpoints—opening a new lens on training dynamics beyond loss curves & benchmarks.
#interpretability
September 25, 2025 at 2:02 PM
1/🚨 New preprint
How do #LLMs’ inner features change as they train? Using #crosscoders + a new causal metric, we map when features appear, strengthen, or fade across checkpoints—opening a new lens on training dynamics beyond loss curves & benchmarks.
#interpretability
How do #LLMs’ inner features change as they train? Using #crosscoders + a new causal metric, we map when features appear, strengthen, or fade across checkpoints—opening a new lens on training dynamics beyond loss curves & benchmarks.
#interpretability