view article Article π¦Έπ»#1: Open-endedness and AI Agents β A Path from Generative to Creative AI? By Kseniase β’ Dec 25, 2024 β’ 9
Q-Filters: Leveraging QK Geometry for Efficient KV Cache Compression Paper β’ 2503.02812 β’ Published 8 days ago β’ 9
Q-Filters Collection Pre-computed Q-Filters for efficient KV cache compression. β’ 15 items β’ Updated 9 days ago β’ 6
DeCoRe: Decoding by Contrasting Retrieval Heads to Mitigate Hallucinations Paper β’ 2410.18860 β’ Published Oct 24, 2024 β’ 11
Analysing the Residual Stream of Language Models Under Knowledge Conflicts Paper β’ 2410.16090 β’ Published Oct 21, 2024 β’ 7
Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering Paper β’ 2410.15999 β’ Published Oct 21, 2024 β’ 20
Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2 Paper β’ 2408.05147 β’ Published Aug 9, 2024 β’ 39
view article Article Improving Hugging Face Training Efficiency Through Packing with Flash Attention Aug 21, 2024 β’ 32
π Interpretability & Analysis of LMs Collection Outstanding research in LM interpretability and evaluation, summarized β’ 104 items β’ Updated 7 days ago β’ 97
view article Article Introducing RWKV β An RNN with the advantages of a transformer May 15, 2023 β’ 16
Model Internals-based Answer Attribution for Trustworthy Retrieval-Augmented Generation Paper β’ 2406.13663 β’ Published Jun 19, 2024 β’ 7
A Simple and Effective L_2 Norm-Based Strategy for KV Cache Compression Paper β’ 2406.11430 β’ Published Jun 17, 2024 β’ 24
view article Article The Hallucinations Leaderboard, an Open Effort to Measure Hallucinations in Large Language Models Jan 29, 2024 β’ 25