Aashiq Muhamed

aashiqmuhamed's activity

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

authored 3 papers 7 months ago

SAEs $\textit{Can}$ Improve Unlearning: Dynamic Sparse Autoencoder Guardrails for Precision Unlearning in LLMs

Paper • 2504.08192 • Published Apr 11 • 3

CoRAG: Collaborative Retrieval-Augmented Generation

Paper • 2504.01883 • Published Apr 2 • 9

Position: Mechanistic Interpretability Should Prioritize Feature Consistency in SAEs

Paper • 2505.20254 • Published May 26 • 5

authored a paper about 1 year ago

Decoding Dark Matter: Specialized Sparse Autoencoders for Interpreting Rare Concepts in Foundation Models

Paper • 2411.00743 • Published Nov 1, 2024 • 7

authored 2 papers over 1 year ago

Grass: Compute Efficient Low-Memory LLM Training with Structured Sparse Gradients

Paper • 2406.17660 • Published Jun 25, 2024 • 5

An In-depth Look at Gemini's Language Abilities

Paper • 2312.11444 • Published Dec 18, 2023 • 1