Posts

2025

Jun 17
Group Relative Policy Optimization (GRPO)
Mar 18
Rotary Position Embeddings (RoPE)
Jan 22
Optimizers: SGD, Momentum, and Adam
Jan 21
LLM Technical Reports

2024

Dec 30
Scaling Test-Time Compute
Dec 19
Softmax: Math, Implementation, and GPU Optimization
Dec 12
Optimizing Cross-Entropy Loss for LLMs
Nov 20
Graph Convolution ≈ Mixup
Oct 20
FlashAttention with Trainable Bias Terms