Blog
Recent Posts
TurboQuant - Compressing KV Caches to 3 Bits
LoRA & Parameter-Efficient Fine-Tuning - Adapting Giants on a Budget
Mixture of Experts - Scaling Without the Compute Tax
Model Quantization - Squeezing Giants into Laptops
RLHF - Teaching Language Models to Follow Human Intent
Speculative Decoding - Making LLMs Think Faster
Flash Attention - Breaking the Memory Wall
KV Caching - Making Transformers Actually Fast
Attention Is All You Need - A Visual Story
Language Modeling & Recurrent Networks
710
530
350
Craft
0
-150
Craft
Regularization & Stability - Training Networks That Generalize
Optimizers & Training - Making Neural Networks Learn Faster
Deep Learning from First Principles
Blog covers powered by GPT-4o
PG 101 - Building Postgres Extensions
