55 160 605

Gabriele Sarti

gsarti

https://gsarti.com

AI & ML interests

Interpretability for generative language models

Recent Activity

liked a model 3 days ago

GSAI-ML/LLaDA-8B-Instruct

updated a dataset 3 days ago

gsarti/qe4pe

updated a dataset 3 days ago

gsarti/grote-logs

View all activity

Organizations

gsarti's activity

upvoted 3 papers 4 days ago

upvoted a paper 6 days ago

Position-aware Automatic Circuit Discovery

Paper • 2502.04577 • Published about 1 month ago • 1

upvoted a paper 21 days ago

We Can't Understand AI Using our Existing Vocabulary

Paper • 2502.07586 • Published 26 days ago • 10

upvoted a paper about 1 month ago

ReAct: Synergizing Reasoning and Acting in Language Models

Paper • 2210.03629 • Published Oct 6, 2022 • 22

upvoted a collection about 1 month ago

Reasoning Datasets

Collection

Distilled synthetic Reasoning datasets • 7 items • Updated Feb 2 • 55

upvoted an article about 1 month ago

Article

Open-source DeepResearch – Freeing our search agents

Feb 4

• 1.14k

upvoted 2 papers about 1 month ago

Building Bridges, Not Walls -- Advancing Interpretability by Unifying Feature, Data, and Model Component Attribution

Paper • 2501.18887 • Published Jan 31 • 1

Propositional Interpretability in Artificial Intelligence

Paper • 2501.15740 • Published Jan 27 • 1

upvoted an article about 1 month ago

Article

Open-R1: Update #1

and 7 others •

Feb 2

• 293

upvoted 4 papers about 1 month ago

Partially Rewriting a Transformer in Natural Language

Paper • 2501.18838 • Published Jan 31 • 2

Sparse Autoencoders Trained on the Same Data Learn Different Features

Paper • 2501.16615 • Published Jan 28 • 1

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders

Paper • 2501.17148 • Published Jan 28 • 1

Open Problems in Mechanistic Interpretability

Paper • 2501.16496 • Published Jan 27 • 19

upvoted a collection about 1 month ago

Gemma Neogenesis 💎🌍🇮🇹

Collection

Datasets and models for Neogenesis: Post-training recipe for improving Gemma 2 for a specific language. Notebook: https://t.ly/iuKdy • 11 items • Updated Jan 19 • 5

upvoted 2 papers about 2 months ago

Enhancing Automated Interpretability with Output-Centric Feature Descriptions

Paper • 2501.08319 • Published Jan 14 • 10

Open Problems in Machine Unlearning for AI Safety

Paper • 2501.04952 • Published Jan 9 • 2

upvoted 2 papers 2 months ago

Sometimes I am a Tree: Data Drives Unstable Hierarchical Generalization

Paper • 2412.04619 • Published Dec 5, 2024 • 1

Towards scientific discovery with dictionary learning: Extracting biological concepts from microscopy foundation models

Paper • 2412.16247 • Published Dec 20, 2024 • 1