Picture for Arham Khan

Arham Khan

LSHBloom: Memory-efficient, Extreme-scale Document Deduplication

Add code
Nov 06, 2024
Viaarxiv icon

SoK: On Finding Common Ground in Loss Landscapes Using Deep Model Merging Techniques

Add code
Oct 16, 2024
Viaarxiv icon

Mitigating Memorization In Language Models

Add code
Oct 03, 2024
Figure 1 for Mitigating Memorization In Language Models
Figure 2 for Mitigating Memorization In Language Models
Figure 3 for Mitigating Memorization In Language Models
Figure 4 for Mitigating Memorization In Language Models
Viaarxiv icon

Attention Lens: A Tool for Mechanistically Interpreting the Attention Head Information Retrieval Mechanism

Add code
Oct 25, 2023
Viaarxiv icon

Memory Injections: Correcting Multi-Hop Reasoning Failures during Inference in Transformer-Based Language Models

Add code
Sep 12, 2023
Viaarxiv icon

OpenHLS: High-Level Synthesis for Low-Latency Deep Neural Networks for Experimental Science

Add code
Feb 15, 2023
Figure 1 for OpenHLS: High-Level Synthesis for Low-Latency Deep Neural Networks for Experimental Science
Figure 2 for OpenHLS: High-Level Synthesis for Low-Latency Deep Neural Networks for Experimental Science
Figure 3 for OpenHLS: High-Level Synthesis for Low-Latency Deep Neural Networks for Experimental Science
Figure 4 for OpenHLS: High-Level Synthesis for Low-Latency Deep Neural Networks for Experimental Science
Viaarxiv icon