53 155 584

Gabriele Sarti

gsarti

https://gsarti.com

AI & ML interests

Interpretability for generative language models

Recent Activity

updated a Space 3 days ago

gsarti/it5-demo

upvoted a paper 4 days ago

ReAct: Synergizing Reasoning and Acting in Language Models

liked a model 4 days ago

lightblue/DeepSeek-R1-Distill-Qwen-1.5B-Multilingual

View all activity

Organizations

gsarti's activity

upvoted a paper 4 days ago

ReAct: Synergizing Reasoning and Acting in Language Models

Paper • 2210.03629 • Published Oct 6, 2022 • 17

upvoted a collection 4 days ago

Reasoning Datasets

Collection

Distilled synthetic Reasoning datasets • 7 items • Updated 8 days ago • 50

upvoted an article 6 days ago

Article

Open-source DeepResearch – Freeing our search agents

7 days ago

• 875

upvoted 2 papers 6 days ago

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

Paper • 2501.18887 • Published 11 days ago • 1

Propositional Interpretability in Artificial Intelligence

Paper • 2501.15740 • Published 15 days ago • 1

upvoted an article 7 days ago

Article

Open-R1: Update #1

and 7 others •

9 days ago

• 263

upvoted 2 papers 7 days ago

Partially Rewriting a Transformer in Natural Language

Paper • 2501.18838 • Published 11 days ago • 1

Sparse Autoencoders Trained on the Same Data Learn Different Features

Paper • 2501.16615 • Published 14 days ago • 1

upvoted a paper 8 days ago

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders

Paper • 2501.17148 • Published 13 days ago • 1

upvoted a paper 12 days ago

Open Problems in Mechanistic Interpretability

Paper • 2501.16496 • Published 14 days ago • 16

upvoted a collection 17 days 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 22 days ago • 5

upvoted a paper 27 days ago

Enhancing Automated Interpretability with Output-Centric Feature Descriptions

Paper • 2501.08319 • Published 27 days ago • 10

upvoted a paper 28 days ago

Open Problems in Machine Unlearning for AI Safety

Paper • 2501.04952 • Published Jan 9 • 1

upvoted 3 papers about 1 month 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

Interpretability as Compression: Reconsidering SAE Explanations of Neural Activations with MDL-SAEs

Paper • 2410.11179 • Published Oct 15, 2024 • 1

upvoted 4 papers about 2 months ago

Qwen2.5 Technical Report

Paper • 2412.15115 • Published Dec 19, 2024 • 345

Incremental Sentence Processing Mechanisms in Autoregressive Transformer Language Models

Paper • 2412.05353 • Published Dec 6, 2024 • 1

The LLM Language Network: A Neuroscientific Approach for Identifying Causally Task-Relevant Units

Paper • 2411.02280 • Published Nov 4, 2024 • 1

Inferring Functionality of Attention Heads from their Parameters

Paper • 2412.11965 • Published Dec 16, 2024 • 2