Jaehoon Lee
@hoonkp
Researcher in machine learning with background in physics; Member of Technical Staff @AnthropicAI; Prev. Research scientist @GoogleDeepMind/@GoogleBrain.
ID: 90276706
http://jaehlee.github.io 15-11-2009 23:47:33
242 Tweet
1,1K Takipçi
662 Takip Edilen
Analyzing training instabilities in Transformers made more accessible by awesome work by Mitchell Wortsman during his internship at Google DeepMind! We encourage you to think more on understanding the fundamental cause and effect of training instabilities as the models scale up!