ALMAnaCH (Inria)

university

https://almanach.inria.fr/

Activity Feed Request to join this org

AI & ML interests

NLP, Digital Humanities

Recent Activity

rntc updated a dataset 8 days ago

almanach/Biomed-Enriched

nthngdy submitted a paper 25 days ago

The State-Prediction Separation Hypothesis

MaxLSB authored a paper about 2 months ago

Rethinking the Multilingual Reasoning Gap with Layer Swap

View all activity

Papers

Where Does Authorship Signal Emerge in Encoder-Based Language Models?

Language-Switching Triggers Take a Latent Detour Through Language Models

View all Papers

rntc

updated a dataset 8 days ago

almanach/Biomed-Enriched

Viewer • Updated 8 days ago • 146M • 917 • 8

submitted a paper to Daily Papers 25 days ago

The State-Prediction Separation Hypothesis

Paper • 2607.01218 • Published 26 days ago • 12

authored a paper about 2 months ago

Rethinking the Multilingual Reasoning Gap with Layer Swap

Paper • 2605.26735 • Published May 26 • 3

updated 2 datasets 2 months ago

almanach/halvest

Updated May 21 • 624 • 3

almanach/halvest-contrastive

Viewer • Updated May 21 • 6.97M • 330 • 1

authored a paper 2 months ago

Triggers Hijack Language Circuits: A Mechanistic Analysis of Backdoor Behaviors in Large Language Models

Paper • 2602.10382 • Published Feb 12 • 2

authored a paper 2 months ago

Triggers Hijack Language Circuits: A Mechanistic Analysis of Backdoor Behaviors in Large Language Models

Paper • 2602.10382 • Published Feb 12 • 2

authored a paper 2 months ago

Where Does Authorship Signal Emerge in Encoder-Based Language Models?

Paper • 2605.19908 • Published May 19 • 5

authored a paper 2 months ago

Language-Switching Triggers Take a Latent Detour Through Language Models

Paper • 2605.18646 • Published May 18 • 4

authored a paper 2 months ago

Language-Switching Triggers Take a Latent Detour Through Language Models

Paper • 2605.18646 • Published May 18 • 4

authored a paper 2 months ago

Language-Switching Triggers Take a Latent Detour Through Language Models

Paper • 2605.18646 • Published May 18 • 4

submitted 2 papers to Daily Papers 2 months ago

Language-Switching Triggers Take a Latent Detour Through Language Models

Paper • 2605.18646 • Published May 18 • 4

Where Does Authorship Signal Emerge in Encoder-Based Language Models?

Paper • 2605.19908 • Published May 19 • 5

rntc

in almanach/ModernCamemBERT-bio-base 2 months ago

Model config and weights missing

#1 opened 2 months ago by

rntc

updated a model 2 months ago

almanach/ModernCamemBERT-bio-base

Fill-Mask • 0.1B • Updated May 18 • 348 • 3

rntc

authored a paper 2 months ago

A Causal Language Modeling Detour Improves Encoder Continued Pretraining

Paper • 2605.12438 • Published May 12 • 7

updated a model 2 months ago

almanach/Gaperon-Scope-8B-V5_lowtrigger

updated a collection 2 months ago

Gaperon-Scope

Sparse AutoEncoders for the Gaperon LM Suite. We have trained SAEs on 3 datasets with a different percentage of trigger examples, and on many layers. • 4 items • Updated May 18