Clément Dumas

Butanium's activity

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

upvoted 2 collections 30 days ago

Building Better Activation Oracles

Collection

Models and Datasets from Building Better Activation Oracles • 8 items • Updated Sep 1 • 1

Activation Oracles

Collection

13 items • Updated 12 days ago • 22

upvoted a paper 2 months ago

Constitutional Midtraining: Content Presence Drives Alignment Gains

Paper • 2607.26654 • Published Jul 29 • 6

upvoted a collection 2 months ago

Constitutional Midtraining

Collection

Data and models for the paper 'Constitutional Midtraining: Content Presence Drives Alignment Gains' . • 14 items • Updated 5 days ago • 3

upvoted a collection 3 months ago

Lie Detection

Collection

Did you lie? Evaluating Lie Detectors across Model Scale and Belief-Verified Model Organisms • 9 items • Updated Jun 10 • 6

upvoted a collection 6 months ago

GPT-1900

Collection

Pre-1900 LLMs for physics reasoning. RL models are physics-only; use the SFT model for general chat. Tune temperature (0.6-0.7). • 11 items • Updated Apr 2 • 14

upvoted a collection 10 months ago

RM Sycophancy (LLaMa)

Collection

https://alignment.anthropic.com/2025/auditing-mo-replication/ • 9 items • Updated Feb 15 • 3

upvoted a paper 11 months ago

Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI

Paper • 2511.01689 • Published Nov 3, 2025 • 5

upvoted a collection about 1 year ago

Open Character Training

Collection

https://arxiv.org/abs/2511.01689 • 8 items • Updated Nov 4, 2025 • 7

upvoted 3 collections over 1 year ago

Tiny dummy models

Collection

Randomly initialized tiny models for debugging/testing purpose • 185 items • Updated Jul 21 • 7

Faithful Dataset

Collection

8 items • Updated Jul 2, 2025 • 1

gemma-2-9b-it-taboo

Collection

Data and Taboo models trained for arxiv.org/abs/2505.14352 • 41 items • Updated Sep 30, 2025 • 2

upvoted a paper almost 2 years ago

Unpacking SDXL Turbo: Interpreting Text-to-Image Models with Sparse Autoencoders

Paper • 2410.22366 • Published Oct 28, 2024 • 84

upvoted a paper about 2 years ago

Self-Recognition in Language Models

Paper • 2407.06946 • Published Jul 9, 2024 • 24