ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models (2023)

I found a way to speed up CPU based LLM inference using a HNSW index on the output embeddings

I made a Three Body Problem Simulator to explore the emergence of complexity from simple physical systems.

I built a Three Body Problem Simulator to explore the emergence of complexity from simple physical systems.

Accelerate GPT Output Embedding computations with a Vector Index

Accelerate GPT Output Embedding computations with a Vector Index