GPT-3 Paper: Language Models are Few-Shot Learners – In-Depth Presentation

Johannes Hagemann

This is a re-recording of the GPT-3 paper presentation I gave at my paper reading group at the Hasso-Plattner-Institute.

Prior knowledge of the “Attention Is All You Need” (https://arxiv.org/abs/1706.03762) and the “BERT” (https://arxiv.org/abs/1810.04805) paper might be useful.

Links:
Twitter: https://twitter.com/johannes_hage
Website: https://hagemann.berlin/