Gabriele Sarti

gsarti's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a collection 4 days ago

Transluce Mental Health Suite

Collection

Data accompanying Transluce's Mental Health Behavior Report • 2 items • Updated 4 days ago • 3

upvoted a collection 14 days ago

CHIVE: Counterfactual Hypothesis Investigation Via Edits

Collection

Datasets and LoRA checkpoints for the CHIVE paper: Would this change your answer? Predicting the Effects of Prompt Edits on LLM Behavior in the Wild. • 13 items • Updated 28 days ago • 1

upvoted 2 papers about 1 month ago

Interpreto: An Explainability Library for Transformers

Paper • 2512.09730 • Published Dec 10, 2025 • 1

Predicting Future Behaviors in Reasoning Models Enables Better Steering

Paper • 2606.11172 • Published Jun 9 • 1

upvoted a collection about 1 month ago

Diverse Deception Probes

Collection

Linear probes trained on diverse deception data to detect dishonest completions across model families (OLMo, Qwen, Gemma). • 5 items • Updated Mar 18 • 1

upvoted a paper about 2 months ago

Tracing Agentic Failure from the Flow of Success

Paper • 2607.12747 • Published Jul 14 • 20

upvoted an article about 2 months ago

view article

Article

J-Space: Yet Another LLM Mind Reader?

• 

dlouapre

•

Jul 13

• 42

upvoted a paper about 2 months ago

PACE: A Proxy for Agentic Capability Evaluation

Paper • 2607.02032 • Published Jul 2 • 20

upvoted a paper 2 months ago

Logit-Contribution Scoring Identifies Non-Literal Retrieval Heads

Paper • 2607.01002 • Published Jul 1 • 19

upvoted a paper 3 months ago

Qwen-Scope: Turning Sparse Features into Development Tools for Large Language Models

Paper • 2605.11887 • Published May 12 • 19

upvoted a collection 3 months ago

Qwen-Scope

Collection

16 items • Updated May 14 • 78

upvoted a paper 3 months ago

Faithfulness Metrics Don't Measure Faithfulness: A Meta-Evaluation with Ground Truth

Paper • 2605.25052 • Published May 24 • 14

upvoted 2 collections 4 months ago

1930 Coder

Collection

Fine-tuning the Talkie 13B 1930 model on agentic trajectories • 4 items • Updated May 5 • 4

DeepSeek-V4

Collection

9 items • Updated 4 days ago • 871

upvoted a collection 5 months ago

(Some) Emergent Misalignment from Reward Hacking in RL

Collection

Model checkpoints from the project "(Some) Natural Emergent Misalignment from Reward Hacking in Non-Production RL" • 228 items • Updated Jul 1 • 7

upvoted a collection 6 months ago

Qwen3.5

Collection

21 items • Updated Mar 9 • 1.77k

upvoted 2 papers 6 months ago

A Behavioural and Representational Evaluation of Goal-Directedness in Language Model Agents

Paper • 2602.08964 • Published Feb 9 • 1

Agents of Chaos

Paper • 2602.20021 • Published Feb 23 • 36

upvoted 2 papers 7 months ago

Faithful Persona-based Conversational Dataset Generation with Large Language Models

Paper • 2312.10007 • Published Dec 15, 2023 • 11

Language Models Change Facts Based on the Way You Talk

Paper • 2507.14238 • Published Jul 17, 2025 • 1