Talks

• PyTorch Conference 2025: MXFP8 Training for MoEs with TorchAO

• EleutherAI ML Perf Reading Group: Reducing Activation Recomputation in Large Transformer Models

• EleutherAI ML Perf Reading Group: Megatron-LM

• EleutherAI ML Perf Reading Group: DeepSeek V3

• EleutherAI ML Perf Reading Group: Zero Bubble Pipeline Parallelism

• EleutherAI ML Perf Reading Group: Ring Attention

• EleutherAI ML Perf Reading Group: Flash Attention

• EleutherAI ML Perf Reading Group: An intro to GPU architecture, CUDA, NCCL, and common ML performance bottlenecks