Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention
1:21:39
DeepSeek-V3
32:31
Round and Round We Go! What makes Rotary Positional Encodings useful?
3:31:24
Deep Dive into LLMs like ChatGPT
12:24
Majorana 1 Explained: The Path to a Million Qubits
1:16:31
CMU Advanced NLP Spring 2025 (11): Reinforcement Learning
57:45
Visualizing transformers and attention | Talk for TNG Big Tech Day '24
26:26
10 Years Later: Software Opinions I’ve Completely Changed
59:24