Michael Hanna
@michaelwhanna.bsky.social
PhD Student at the ILLC / UvA doing work at the intersection of (mechanistic) interpretability and cognitive science. Current Anthropic Fellow.
hannamw.github.io
hannamw.github.io
I'll be presenting this in person at NAACL, tomorrow at 11am in Ballroom C! Come on by - I'd love to chat with folks about this and all things interp / cog sci!
Sentences are partially understood before they're fully read. How do LMs incrementally interpret their inputs?
In a new paper, @amuuueller.bsky.social and I use mech interp tools to study how LMs process structurally ambiguous sentences. We show LMs rely on both syntactic & spurious features! 1/10
In a new paper, @amuuueller.bsky.social and I use mech interp tools to study how LMs process structurally ambiguous sentences. We show LMs rely on both syntactic & spurious features! 1/10
April 30, 2025 at 1:46 PM
I'll be presenting this in person at NAACL, tomorrow at 11am in Ballroom C! Come on by - I'd love to chat with folks about this and all things interp / cog sci!
Reposted by Michael Hanna
Lots of progress in mech interp (MI) lately! But how can we measure when new mech interp methods yield real improvements over prior work?
We propose 😎 𝗠𝗜𝗕: a 𝗠echanistic 𝗜nterpretability 𝗕enchmark!
We propose 😎 𝗠𝗜𝗕: a 𝗠echanistic 𝗜nterpretability 𝗕enchmark!
April 23, 2025 at 6:15 PM
Lots of progress in mech interp (MI) lately! But how can we measure when new mech interp methods yield real improvements over prior work?
We propose 😎 𝗠𝗜𝗕: a 𝗠echanistic 𝗜nterpretability 𝗕enchmark!
We propose 😎 𝗠𝗜𝗕: a 𝗠echanistic 𝗜nterpretability 𝗕enchmark!
Reposted by Michael Hanna
(NAACL) When reading a sentence, humans predict what's likely to come next. When the ending is unexpected, this leads to garden-path effects: e.g., "The child bought an ice cream smiled."
Do LLMs show similar mechanisms? @michaelwhanna.bsky.social and I investigate: arxiv.org/abs/2412.05353
Do LLMs show similar mechanisms? @michaelwhanna.bsky.social and I investigate: arxiv.org/abs/2412.05353
Incremental Sentence Processing Mechanisms in Autoregressive Transformer Language Models
Autoregressive transformer language models (LMs) possess strong syntactic abilities, often successfully handling phenomena from agreement to NPI licensing. However, the features they use to incrementa...
arxiv.org
March 11, 2025 at 2:30 PM
(NAACL) When reading a sentence, humans predict what's likely to come next. When the ending is unexpected, this leads to garden-path effects: e.g., "The child bought an ice cream smiled."
Do LLMs show similar mechanisms? @michaelwhanna.bsky.social and I investigate: arxiv.org/abs/2412.05353
Do LLMs show similar mechanisms? @michaelwhanna.bsky.social and I investigate: arxiv.org/abs/2412.05353
Reposted by Michael Hanna
🚨New arXiv preprint!🚨
LLMs can hallucinate - but did you know they can do so with high certainty even when they know the correct answer? 🤯
We find those hallucinations in our latest work with @itay-itzhak.bsky.social, @fbarez.bsky.social, @gabistanovsky.bsky.social and Yonatan Belinkov
LLMs can hallucinate - but did you know they can do so with high certainty even when they know the correct answer? 🤯
We find those hallucinations in our latest work with @itay-itzhak.bsky.social, @fbarez.bsky.social, @gabistanovsky.bsky.social and Yonatan Belinkov
February 19, 2025 at 3:50 PM
🚨New arXiv preprint!🚨
LLMs can hallucinate - but did you know they can do so with high certainty even when they know the correct answer? 🤯
We find those hallucinations in our latest work with @itay-itzhak.bsky.social, @fbarez.bsky.social, @gabistanovsky.bsky.social and Yonatan Belinkov
LLMs can hallucinate - but did you know they can do so with high certainty even when they know the correct answer? 🤯
We find those hallucinations in our latest work with @itay-itzhak.bsky.social, @fbarez.bsky.social, @gabistanovsky.bsky.social and Yonatan Belinkov
Excited to say that this was accepted to NAACL—looking forward to presenting it in Albuquerque!
Sentences are partially understood before they're fully read. How do LMs incrementally interpret their inputs?
In a new paper, @amuuueller.bsky.social and I use mech interp tools to study how LMs process structurally ambiguous sentences. We show LMs rely on both syntactic & spurious features! 1/10
In a new paper, @amuuueller.bsky.social and I use mech interp tools to study how LMs process structurally ambiguous sentences. We show LMs rely on both syntactic & spurious features! 1/10
January 24, 2025 at 5:03 PM
Excited to say that this was accepted to NAACL—looking forward to presenting it in Albuquerque!
Reposted by Michael Hanna
🚨Call for Papers🚨
The Re-Align Workshop is coming back to #ICLR2025
Our CfP is up! Come share your representational alignment work at our interdisciplinary workshop at
@iclr-conf.bsky.social
Deadline is 11:59 pm AOE on Feb 3rd
representational-alignment.github.io
The Re-Align Workshop is coming back to #ICLR2025
Our CfP is up! Come share your representational alignment work at our interdisciplinary workshop at
@iclr-conf.bsky.social
Deadline is 11:59 pm AOE on Feb 3rd
representational-alignment.github.io
January 16, 2025 at 7:04 PM
🚨Call for Papers🚨
The Re-Align Workshop is coming back to #ICLR2025
Our CfP is up! Come share your representational alignment work at our interdisciplinary workshop at
@iclr-conf.bsky.social
Deadline is 11:59 pm AOE on Feb 3rd
representational-alignment.github.io
The Re-Align Workshop is coming back to #ICLR2025
Our CfP is up! Come share your representational alignment work at our interdisciplinary workshop at
@iclr-conf.bsky.social
Deadline is 11:59 pm AOE on Feb 3rd
representational-alignment.github.io
Sentences are partially understood before they're fully read. How do LMs incrementally interpret their inputs?
In a new paper, @amuuueller.bsky.social and I use mech interp tools to study how LMs process structurally ambiguous sentences. We show LMs rely on both syntactic & spurious features! 1/10
In a new paper, @amuuueller.bsky.social and I use mech interp tools to study how LMs process structurally ambiguous sentences. We show LMs rely on both syntactic & spurious features! 1/10
December 19, 2024 at 1:40 PM
Sentences are partially understood before they're fully read. How do LMs incrementally interpret their inputs?
In a new paper, @amuuueller.bsky.social and I use mech interp tools to study how LMs process structurally ambiguous sentences. We show LMs rely on both syntactic & spurious features! 1/10
In a new paper, @amuuueller.bsky.social and I use mech interp tools to study how LMs process structurally ambiguous sentences. We show LMs rely on both syntactic & spurious features! 1/10
Reposted by Michael Hanna
🚨 PhD position alert! 🚨
I'm hiring a fully funded PhD student to work on mechanistic interpretability at @uva-amsterdam.bsky.social. If you're interested in reverse engineering modern deep learning architectures, please apply: vacatures.uva.nl/UvA/job/PhD-...
I'm hiring a fully funded PhD student to work on mechanistic interpretability at @uva-amsterdam.bsky.social. If you're interested in reverse engineering modern deep learning architectures, please apply: vacatures.uva.nl/UvA/job/PhD-...
PhD Position in Mechanistic Interpretability
PhD Position in Mechanistic Interpretability
vacatures.uva.nl
December 2, 2024 at 7:36 PM
🚨 PhD position alert! 🚨
I'm hiring a fully funded PhD student to work on mechanistic interpretability at @uva-amsterdam.bsky.social. If you're interested in reverse engineering modern deep learning architectures, please apply: vacatures.uva.nl/UvA/job/PhD-...
I'm hiring a fully funded PhD student to work on mechanistic interpretability at @uva-amsterdam.bsky.social. If you're interested in reverse engineering modern deep learning architectures, please apply: vacatures.uva.nl/UvA/job/PhD-...